Part 2: Abnormalities of the Kidney and Urinary Tract

APPROACH TO THE CHILD WITH HEMATURIA

Hematuria is a common sign of urinary tract disease but also occurs in otherwise healthy children. Hematuria can present as either discolored red, brown, or tea-colored urine, or it can present as yellow urine with a positive dipstick for blood. Clues to the diagnosis may be obtained from a history focused on whether the hematuria is painless, intermittent, or persistent, or microscopic or gross. A family history of hematuria or renal disease is important. Brown or tea-colored urine is common in glomerulonephritis, whereas red or obviously bloody urine suggests postglomerular bleeding. Glomerulonephritis is commonly associated with other abnormalities of the urine such as proteinuria and cellular casts as well as hypertension, edema, and reduced renal function. Table 464-1 provides a simple mnemonic for the causes of hematuria. A diagnostic approach and differential diagnosis for discolored urine and microscopic hematuria are shown in Figure 464-1.

Microscopic hematuria, defined as the presence of a positive urine dipstick for blood and more than 5 red blood cells (RBCs) per high-power field (hpf) on centrifuged urine, occurs in approximately 1% of school-age children on at least 1 urine sample. About 0.5% will continue to have hematuria on 2 of 3 samples, and a third have hematuria on 3 samples. Fewer than one-third of patients diagnosed with hematuria demonstrate hematuria 1 year later. Only a very small number of children with microscopic hematuria had significant renal or urologic disease. Microscopic hematuria can occur following vigorous exercise; thus, it is common in school-age children and, in the majority of cases, is benign.

Hypercalciuria, defined as more than 4 mg of urinary calcium/kg/d, has been found in significant numbers of children with microscopic hematuria. This association is even stronger if there is a family history of stone disease or the occurrence of gross hematuria. The calcium-to-creatinine (Ca/Cr) ratio can be used to screen for hypercalciuria; for children or adolescents, a urinary Ca/Cr ratio greater than 0.2 is abnormal.

The value of urinary tract imaging in children with microscopic hematuria is controversial. The yield of routine ultrasound is low, and reported findings are frequently of little clinical significance. When microscopic hematuria persists for several months, decisions on the need for urinary tract imaging are made based on other concerns. Voiding cystourethrograms and cystoscopy are rarely helpful and are not indicated for the routine evaluation of children with either microscopic or gross hematuria.

Gross hematuria is much less common than microscopic hematuria and warrants a more thorough investigation, as outlined in Figure 464-1. Trauma should be considered, especially if the urine contains clots. If trauma is a likely cause, the urinary tract should be promptly imaged by computed tomography scan. Similarly, if a palpable flank mass is noted, a renal tumor must be considered. Brown or tea-colored urine, especially with cellular casts and protein, suggests glomerulonephritis. The presence of dysuria or fever suggests hemorrhagic cystitis, which can be caused by both bacterial and viral agents. Sickle cell hemoglobinopathies are associated with recurrent gross hematuria that at times can be quite severe. Nephrolithiasis is associated with abdominal pain and nausea. Hypercalciuria, even in the absence of a demonstrable stone, is a common cause of gross hematuria. Renal cysts, especially those in autosomal dominant polycystic kidney disease (ADPKD), occasionally rupture and cause gross hematuria. Coagulopathies are an extremely rare cause of gross hematuria and are almost always associated with bleeding elsewhere. Rare causes include obstructive uropathy, tumors, and vascular malformations.

APPROACH TO THE CHILD WITH PROTEINURIA

Most children excrete urine free of protein when evaluated by the semiquantitative dipstick method. Proteinuria is not an uncommon finding in otherwise healthy children. An approach to proteinuria is shown in Figure 464-2. If the proteinuria is associated with other signs of renal disease, such as edema, hematuria, hypertension, growth failure, or elevated serum creatinine, then a more extensive evaluation is indicated. Proteinuria can be transient, orthostatic, or fixed.

Transient proteinuria, defined as disappearance of proteinuria on several occasions after a positive test, accounts for 80% of patients with isolated proteinuria and occurs in conjunction with fever, exercise, cold exposure, heat stress, and emotional distress. In an otherwise healthy child, transient proteinuria is not associated with renal disease.

Orthostatic proteinuria, defined as proteinuria only in an upright position, is the next most frequent cause and occurs commonly in adolescents. Evaluation for orthostatic proteinuria is accomplished by having the patient empty the bladder before going to bed at night and checking the first urine passed on waking. A second sample is obtained later in the day, after the patient has been upright for several hours. Urine protein is evaluated by dipstick, urine protein/creatinine ratio, or quantitative collection. If the recumbent urine is negative for protein and the upright urine has protein, the patient is considered to have orthostatic proteinuria. If dipsticks are used, this determination must be repeated over 5 to 7 days because of a 22% to 54% false-negative rate for protein. A more precise method is to measure the urine protein/creatinine ratio in the recumbent and upright urine samples or to quantify protein in each position. In orthostatic proteinuria, the recumbent ratio will be less than 0.2, but the upright sample will have an increased ratio, or the recumbent sample will have less than 100 mg protein/12 h and the upright sample will have 300 to 900 mg/12 h. If the total quantitative protein excretion exceeds 1.0 g/d, the diagnosis of orthostatic proteinuria is unlikely. Orthostatic proteinuria is usually transient, remitting after some time, but it can be persistent. Long-term studies in adults have shown orthostatic proteinuria to have a favorable outcome and not be associated with renal disease; however, strict criteria, including normal renal function, blood pressure, and serum competent levels, must be used to exclude patients with other renal diseases. Patients with orthostatic proteinuria should be seen at least yearly for several years to be certain that this is not the harbinger of more serious renal disease.

Fixed proteinuria is persistent and occurs in both upright and recumbent positions. It requires more thorough evaluation (see Fig. 464-2). Protein excretion should be quantified in a timed collection, renal function should be assessed using serum creatinine, and a serum C3 complement level should be obtained. If the serum creatinine and C3 complement are normal and the protein excretion rate is greater than 40 mg/m² per hour (< 3 g/d), then the patient may be followed every 6 to 12 months. If these results are abnormal, then more extensive testing should be done, possibly including a renal biopsy.

APPROACH TO THE CHILD WITH HEMATURIA AND PROTEINURIA

The presence of both hematuria and proteinuria strongly suggests a diagnosis of more serious renal disease than either condition alone, especially if other abnormalities such as cellular casts, edema, hypertension, and azotemia are present. The approach to evaluation is shown in Figure 464-3. Serum creatinine, C3 complement, streptococcal antibody titers (usually ASO and DNase B), and possibly an antinuclear antibody titer should be obtained. The urine protein excretion rate should be evaluated, and if edema is present, serum albumin should be measured. A renal biopsy may be indicated, especially if the diagnosis of poststreptococcal glomerulonephritis is excluded. The decision to perform a renal biopsy is influenced by the severity of the proteinuria, the presence of other signs and symptoms, and the inability to establish a specific diagnosis. Patients who have a combined nephritic (hematuria, hypertension, azotemia) and nephrotic (proteinuria, low albumin, edema) syndrome are likely to have a systemic disease (systemic lupus erythematosus, Henoch-Schonlein purpura, vasculitis) or primary glomerulonephritis. In children with proteinuria, hematuria, or both, a low C3 complement level suggests 1 of 4 diagnoses: (1) postinfectious glomerulonephritis, (2) systemic lupus erythematosus nephritis, (3) membranoproliferative glomerulonephritis, or (4) nephritis associated with chronic infection.

EDEMA

The approach to evaluating edema is shown in Figure 464-4. Renal disease is the most common cause, especially when persistent. Congestive heart failure may cause edema in children, and allergic disorders occasionally cause localized edema that is usually transient and often associated with other signs such as urticaria. Hypoalbuminemia from decreased hepatic albumin synthesis or, rarely, from a protein-losing enteropathy will cause generalized edema. A careful history, physical examination, and selected laboratory tests (such as a serum albumin and a urinalysis) will usually elucidate the cause.

Edema in the child with renal disease is most evident in places where tissue resistance is low, such as the periorbital area, abdomen, scrotum, and labia, or where the hydrostatic pressure is high, such as the lower extremities. Early in renal disease, the only place edema may be noticeable is in the periorbital tissue following several hours in the recumbent position. Edema from acute glomerulonephritis is usually caused by a decreased glomerular filtration rate (GFR) with enhanced tubular reabsorption of sodium and water. In patients with nephrotic syndrome, marked proteinuria combined with hypoalbuminemia (< 2.5 g/dL) leads to decreased plasma oncotic pressure, decreased sodium and water excretion, and edema. With severe or long-standing renal disease, the GFR can be reduced so much that the urine output is less than the fluid intake, even with a high fractional excretion of sodium, and edema occurs.

PYURIA

Pyuria is defined as more than 5 white blood cells (WBCs)/hpf on centrifuged urine. The most common cause of pyuria is a urinary tract infection (UTI). In UTI, pyuria may be associated with other signs of infection including fever and dysuria. The definitive diagnosis of a UTI is made by growth of a single organism in cultured urine of greater that 100,000 colonies/mL urine. Pyuria may also be the result of a viral, mycobacterial, or fungal infection of the urinary tract; pelvic inflammatory disease; or peritonitis. Pyuria also occurs as a consequence of fever and dehydration. Systemic lupus erythematosus and interstitial nephritis are unusual causes of pyuria. Interstitial nephritis is often due to a drug reaction, and pyuria is often associated with the presence of eosinophils in the urine (sometimes with eosinophilia) or may be associated with uveitis, which is then known as tubular interstitial nephritis uveitis syndrome.

The first step in evaluating children with pyuria is to culture the urine for bacteria. If the urine culture is negative, then several additional urinalyses should be obtained to determine if pyuria is persistent. Transient pyuria in an otherwise healthy child requires no further evaluation. Persistent pyuria should be investigated, and the approach to investigation is based on the history, physical examination, and presence of other abnormal elements in the urinalysis such as eosinophils, casts, and proteinuria.

URINATION FREQUENCY AND URGENCY

Children with polyuria void large volumes frequently and have polydipsia, whereas children with urinary frequency void small volumes often. Normal children between the ages of 5 and 12 years void about 4 to 8 times a day. Frequency refers to a pattern of increased number of voids each day. Some children with frequency will urinate small amounts several times an hour. UTIs are the most common cause, especially if accompanied by dysuria and fever. Frequency may persist for weeks after resolution of the infection. Other causes include urethral irritation from vulvovaginitis in girls or meatitis in boys, trauma, foreign body, or bubble baths. Hypercalciuria has also been reported to cause frequency and dysuria. In a number of children, especially younger children, no discernible cause can be found, and the problem resolves with time. On occasion, anticholinergic agents, such as oxybutynin, are useful in reducing the number of voidings. Usually, little evaluation is necessary beyond a history, physical examination, urinalysis, and urine culture. Imaging studies of the bladder and urethra are rarely indicated.

OLIGURIA

Oliguria denotes a very low urine flow, and anuria means virtually no urine flow. Oliguria is common in children and is usually related to volume contraction rather than to a renal abnormality. Oliguria is defined as a urine flow rate less than 0.5 mL/kg/h in children or less than 1 mL/kg/h in infants. Eight percent of healthy newborns do not void in the first 24 hours of life, but 98% do so by 48 hours. The average urine output in the first 2 days of life is 20 mL/d and rises to 200 mL by the end of the second week of life. Most children with oliguria do not have renal disease, but the low urine flow is a renal response to dehydration from other nonrenal sources of volume and electrolyte loss such as diarrhea or sweat (especially in cystic fibrosis) and increased insensible losses. A careful history and physical examination can usually identify the presence of dehydration as the cause for the oliguria. Laboratory clues include an elevated blood urea nitrogen compared to serum creatinine and concentrated urine. With severe dehydration, there may also be a small amount of blood and protein in the urine along with granular, but not cellular, casts. With the development of renal parenchymal injury from dehydration and hypotension (acute tubular necrosis), the kidney often loses the ability to concentrate and retain sodium, so the urine specific gravity is fixed at 1.010 to 1.012, the urine/serum creatinine ratio is less than 10, and the fraction excretion of sodium is greater than 3%.

Oliguria from acute glomerulonephritis is associated with a high specific gravity and low fractional excretion of sodium (< 1%). The presence of gross hematuria, heavy proteinuria, and cellular casts distinguishes between dehydration and acute glomerulonephritis. Disorders of the renal vasculature and obstruction to urine flow should also be considered as a cause of oliguria. A renal ultrasound with Doppler assessment of renal arterial and venous flow is useful to assess renal blood flow. The approach to the child with oliguria is presented in Figure 464-5. Anuria is uncommon and should raise the suspicion of obstruction.

The first step is to evaluate for dehydration. If volume contraction is present (based on physical exam findings and vital signs), then a trial of volume expansion (20 mL/kg of normal saline) with careful monitoring is administered to determine if increased urine flow can be established. If dehydration is not present or the patient has edema, then other causes for oliguria should be considered, such as acute glomerulonephritis, nephrotoxic injury from medications, or acute obstruction. Congenital absence of the kidneys and vascular occlusions are important causes in the neonate.

ENURESIS

Enuresis is defined as the inappropriate discharge of urine in a child who has reached the age at which bladder control would be expected. However, bladder control is a maturational event, and what constitutes “normal” is ambiguous. The average age of successful toilet training is 2.4 years, with a range of 0.75 to 5 years. Approximately 75% of 3-year-olds and 85% to 90% of 5-year-olds are dry at night. Between the ages of 5 and 9 years, 10% to 15% of children experiencing bed-wetting will achieve dryness each year. Thus, most children will become dry with time.

Enuresis may be nocturnal (bed-wetting) or diurnal (daytime wetting) or both. Nocturnal enuresis is much more common than diurnal enuresis. Approximately 10% of children with nocturnal enuresis will have daytime wetting as well, whereas 50% of children with daytime wetting will have nocturnal enuresis. Primary enuresis refers to children who have never achieved dryness for an extended period of time and is more common than secondary enuresis and less likely to be associated with disease. Secondary enuresis refers to the appearance of enuresis in a child who has been dry for at least 6 months.

Nocturnal Enuresis

This common problem occurs in 15% of 5-year-olds, 7% of 10-year-olds, and 1% of adults. The majority of children with this problem do not have either organic or functional disease. However, bed-wetting does cause significant psychosocial problems in older children. Nocturnal enuresis is commonly a familial disorder, as 74% of enuretic males and 58% of enuretic females have at least 1 parent with a history of enuresis. Under the age of 12 years, it is twice as common in males as females, but over the age of 12 years, this difference disappears. Primary nocturnal enuresis is thought to occur because of (1) small bladder capacity, (2) excessive urine formation during sleep, (3) a problem with sleep arousal, or (4) uninhibited detrusor contractions during deep sleep. Children with nocturnal enuresis may have a smaller functional bladder capacity such that they feel the need to void with smaller volumes than age-matched nonenuretic children; however, measurements of bladder capacity under anesthesia show no difference between enuretic and nonenuretic children. Some children with nocturnal enuresis do not have the normally expected increase in antidiuretic hormone levels at night and therefore do not decrease urine formation at night. Sleep studies have shown that nocturnal enuresis occurs randomly throughout the night and not in a particular sleep stage. Although most families report that their enuretic child is the most difficult to arouse from sleep, sleep studies do not consistently support this. Finally, to remain dry throughout the night, the child must be able to centrally inhibit bladder contractions while asleep, just as he must while awake to remain dry during the day. As this is a maturational event, it is not surprising that there is a wide range of ages when this is achieved. Rare causes of enuresis are nocturnal seizures and sleep apnea.

The evaluation of children with nocturnal enuresis begins with a complete history that includes determining the number of wet nights per week, the longest period of dry nights, fluid intake in the late afternoon and evening, voiding pattern or difficulties, stool history and history of soiling, UTI symptoms, family history of nocturnal enuresis, and family or behavioral problems. The appearance of secondary enuresis should alert the physician to a possible organic cause such as UTI, diabetes mellitus, or diabetes insipidus or to a recent psychologically traumatic event in the child’s life. A careful physical examination should also be performed with attention paid to the presence of a full bladder, an abnormal urinary stream, lumbar spine or sacral abnormalities, and neurologic examination of the lower extremities. A urinalysis on a first-morning void should be obtained from all children with enuresis to exclude a urine concentrating problem (normal elevation in specific gravity), renal disease (hematuria and/or proteinuria), diabetes mellitus (absence of glycosuria), and UTI (WBCs, a positive reaction for nitrate, or leukocyte esterase). A simple measurement of bladder capacity may also be useful (normal bladder capacity in ounces is 2 + the age in years).

A variety of treatment approaches may be helpful, but it is important to recognize that most children become dry with time no matter what approach is used. Educating the child and the family is the cornerstone of initial therapy. Parents must be made aware that nocturnal enuresis is not under the child’s voluntary control, so punishment is counterproductive. Constipation should be addressed as well as this may be contributing to the enuresis. Parents should be counseled to limit fluid intake in the late afternoon and evening. A trial of awakening the child during the night to void can be tried initially but is often difficult for parents to maintain.

Bed-wetting alarms are useful nonpharmacologic options that work by conditioning. The alarm goes off and awakens the child during voiding; the child gets out of bed and finishes voiding in the toilet or holds urine until later. They are quite effective, with an initial cure rate of approximately 70% and only a 10% to 15% relapse rate. They do require motivated patients and parents who are willing to learn how to use the devices properly. Use for 3 to 4 months is usually required to achieve dryness. Initially, many children will need to be awakened when the alarm goes off.

Desmopressin (DDAVP) is quite effective in reducing the number of wet nights in most children and allowing many to be completely dry. The initial dose is 0.2 mg taken at bedtime; if necessary, the dose can be titrated to 0.6 mg. It is important that fluid be restricted in the evening. However, the relapse rate exceeds 80% when treatment is stopped. Gradually decreasing the dose may decrease the relapse rate. It can be used intermittently to allow enuretic children to participate in sleepovers and camp.

Imipramine has been used extensively for enuresis and combines an anticholinergic effect for increasing functional bladder capacity with a noradrenergic effect to decrease detrusor contractions. The starting dose is 25 mg at night with an increase to a maximum of 50 mg in children less than 12 years of age and 75 mg in adolescents. The cure rate is between 10% and 60%, with a high relapse rate. The relapse rate can be improved if administered for 3 to 4 months and then gradually tapered over 3 to 4 weeks. Imipramine does have the potential for serious toxicity and therefore should be used after other methods have failed and only with appropriate warnings regarding risks of overdose and other complications.

Daytime Incontinence/Diurnal Enuresis

Diurnal wetting problems arise as a consequence of the transition from an infantile voiding pattern to an adult pattern. In the infant, the bladder fills and reflexively empties when a set volume is reached. Adults are able to cortically inhibit these reflex bladder contractions until there is a convenient time and place to void. Problems in making this transition in bladder control lead to most daytime enuresis. Organic causes of daytime wetting include bladder outlet obstruction, neurogenic bladder, ectopic ureter, and UTIs. Vaginal reflux of urine that seeps out upon rising from the toilet is a common physiologic cause in girls. Giggle incontinence (the sudden involuntary emptying of the bladder with giggling) is a rare cause of daytime wetting that predominantly occurs in girls and is often associated with a family history. Sudden fright will cause involuntary voiding in young children. Other associated behavioral issues that may establish a daytime wetting pattern include constant anxiety, such as the loss of a parent, marital discord, or an abusive home environment. Some children, usually boys, are resistive to toilet training and have a history of toilet avoidance that prevents achievement of daytime dryness. Some children either do not recognize the sensation of a full bladder or ignore it and continue playing. Some children appear to have an unstable bladder, detrusor instability, and persistence of an infantile bladder that manifests as having difficulty cortically inhibiting the spontaneous bladder contractions that occur at lesser bladder volumes. These children withhold voiding by tightening their external sphincter. They frequently are fidgety and use external compression to avoid wetting their pants, such as pressing the heel against the perineum (girls) or grabbing the penis (boys). This behavior can interfere with normal relaxation of the perineum with voiding and can lead to incomplete bladder emptying.

UTIs may initiate these problems, which can persist following treatment. Severe problems with perineal relaxation result in the Hinman syndrome (nonneurogenic neurogenic bladder) and renal damage. A staccato pattern of voiding is a clue to this problem. Constipation is often associated with voiding problems, and its correction often leads to less wetting or resolution. It is unclear whether the constipation directly causes the wetting or may be a reflection of problems with perineal relaxation that is necessary for complete emptying of both the bowel and the bladder.

Only a limited evaluation is generally indicated for children with daytime incontinence. The history should focus on the voiding diary, toilet training, problems with urinary stream in boys, symptoms of UTI or history of a UTI, presence of constipation/encopresis, and the family’s response to wetting accidents. The physical examination should evaluate for a distended bladder and possible spinal defects. The urinary stream in boys should be observed. It is more useful to listen to the sound of voiding in girls; normally, there is a smooth rise and fall in the force, but a forceful, intermittent stream (staccato voider) or abrupt cessation of voiding suggests problems with relaxation of the perineum. A urinalysis and possibly a urine culture are the only laboratory tests that are usually useful. Radiographic studies are indicated only if there are problems with the urinary stream in the male, presence or history of a UTI, constant dampness from a possible ectopic ureter, or physical signs suggestive of a neurogenic bladder. Uroflowmetry and urodynamic studies are reserved for children resistant to conservative measures.

SUGGESTED READINGS

INTRODUCTION

Renal malformations account for 30% to 50% of end-stage renal disease in children. Approximately one-half of such cases are associated with lower urinary tract abnormalities. A spectrum of phenotypes exists and has been termed CAKUT (congenital abnormalities of the kidney and urinary tract). This chapter focuses on the epidemiology of these abnormalities and their etiologies, clinical manifestations, and aspects of clinical management.

CLASSIFICATION AND DEFINITION

Kidney malformations are classified at the clinical level by gross and microscopic anatomic features. One commonly used classification system is as follows:

• Renal agenesis—congenital absence of the kidney and ureter

• Simple renal hypoplasia—small kidney with a reduced number of nephrons and normal renal architecture

• Renal dysplasia—presence of malformed kidney tissue elements

• Renal dysplasia/hypoplasia (hypodysplasia)—small, dysplastic kidney

Characteristic microscopic features of the dysplastic kidney (Fig. 465-1) include abnormal differentiation of mesenchymal and epithelial elements; a decreased number of nephrons; loss of corticomedullary differentiation; and the presence of dysplastic elements, including cartilage and bone. Dysplastic kidneys range in size from large distended kidneys with multiple large cysts to small kidneys, with or without cysts. A small dysplastic kidney without macroscopic cysts is often classified clinically as a hypoplastic/dysplastic kidney, since pathologic examination, which provides a means to distinguish between simple hypoplasia and dysplasia, is not commonly performed during life. The multicystic dysplastic kidney (MCDK) is an extreme form of renal dysplasia.

EPIDEMIOLOGY

The majority of renal malformations are detected during the antenatal period by fetal ultrasound and account for 20% to 30% of all anomalies detected. The incidence of renal and urinary tract malformations is 0.3 to 1.6 per 1000 live-born and stillborn infants. Renal agenesis and dysplasia are 1.3- to 1.9-fold more common in males than in females. Fifty percent of individuals with a kidney malformation have a lower urinary tract anomaly. Of these, 25% have vesicoureteral reflux, 11% have ureteropelvic junction obstruction, and 11% are affected by ureterovesical junction obstruction. Renal malformations, other than mild antenatal pelviectasis, are associated with malformation of nonrenal tissues in about 30% of cases. There are over 100 syndromes associated with renal and urinary tract malformations. The most frequent syndromes are cited in Table 465-1. Bilateral renal agenesis occurs in 1 in 3000 to 10,000 births, while unilateral renal agenesis is estimated to have a prevalence of 1 in 1000. The incidence of unilateral dysplasia is 1 in 3000 to 5000 births (1 in 3640 for MCDK) compared to 1 in 10,000 for bilateral dysplasia. Nine percent of first-degree relatives of patients with bilateral renal agenesis or bilateral renal dysgenesis have some type of renal malformation. The incidence of renal ectopia is 1 in 1000 autopsies, but the clinical recognition is estimated to be only 1 in 10,000 patients. Males and females are equally affected. Renal ectopia is bilateral in 10% of cases; when unilateral, there is a slight predilection for the left side. The incidence of fusion anomalies is estimated to be about 1 in 600 infants.

GENETIC CAUSES

Over 30 genes have been identified in children with syndromic forms of renal malformation (Table 465-2). For a complete list of syndromes featuring renal malformations, the reader is referred to McKusick’s Online Mendelian Inheritance in Man (OMIM, https://www.omim.org/). Emerging evidence suggests that sporadic cases are associated with genetic mutations as well. Below, examples of genetic mutations associated with syndromic and nonsyndromic forms are discussed.

Branchio-Oto-Renal Syndrome

Branchio-oto-renal (BOR) syndrome, defined by the association of branchial, otic, and renal anomalies, is an autosomal dominant disorder with incomplete penetrance and variable expressivity (OMIM no. 113650). However, individual patients may have only 1 or 2 of these major features in association with other minor features such as external ear anomalies, preauricular tags, or other facial abnormalities (eg, branchio-otic syndrome, OMIM no. 120502). Renal malformations include unilateral or bilateral renal agenesis; hypodysplasia; and malformation of the lower urinary tract, including vesicoureteral reflux, pyeloureteral obstruction, and ureteral duplication. Different renal malformations can be observed in the same family. The genes EYA1 and SIX1 are mutated in BOR syndrome. More than 50 different mutations in EYA1 have been identified. Variability of the phenotype in individuals with the same mutation limits accurate prediction of the type and severity of the renal malformation based on genetic testing. Heterozygous mutations in EYA1 or SIX1 have been associated with mutations in the GDNF and RET genes in a small number of affected kindreds. A SIX1 mutation has been found in association with a PAX2 mutation in 2 siblings with severe renal hypodysplasia that led to renal insufficiency during childhood and optic nerve coloboma. Interestingly, their father, who harbored only the PAX2 mutation, had a mild renal malformation with late-onset renal insufficiency and no ocular defect.

Renal-Coloboma Syndrome

Renal-coloboma syndrome (RCS), an autosomal dominant disorder, is characterized by renal hypoplasia, vesicoureteric reflux, and optic nerve coloboma. A wide range of renal and ocular malformations are observed in RCS. Renal hypoplasia and vesicoureteric reflux are most frequent, but multicystic dysplasia and pelviureteric junction obstruction have also been described. RCS is caused by mutations in PAX2, which encodes a transcription factor that belongs to the paired-box family of homeotic genes. Among the more than 30 mutations reported, the vast majority of the mutations are located in exons 2 and 3, which encode the DNA-binding domain. Despite this clustering of mutations, the RCS phenotype is highly variable even in patients harboring the same PAX2 mutation.

Renal Cyst and Diabetes Syndrome

Renal cyst and diabetes syndrome (RCAD) is defined by the presence of renal hypodysplasia with cysts and diabetes mellitus. Occasional features include genital malformations. Diabetes mellitus, which is present in approximately 60% of all the cases reported, usually occurs before 25 years of age and is often associated with pancreatic atrophy. The presence of cysts is the most consistent feature of the renal phenotype, leading to the name renal cysts and diabetes syndrome. The cysts are usually cortical, bilateral, and small. Mutations in TCF2 encoding the transcription factor HNF1b are present in affected patients. Mutations in TCF2 are being increasingly identified in patients with renal hypoplasia and dysplasia, multicystic dysplastic kidneys, renal agenesis, horseshoe kidneys, pelviureteric junction obstruction, and clubbing and tiny diverticula of the calyces. When histology is available, the most specific finding is the presence of cortical glomerular cysts with dilatation of the Bowman spaces (glomerulocystic dysplasia). Other nonspecific lesions such as cystic renal dysplasia, interstitial fibrosis, or oligomeganephronia have also been reported. Antenatal presentation with enlarged hyperechoic kidneys is a common finding. More than 50 mutations have been reported in TCF2. Most are located in the first 4 exons that encode the DNA-binding domain. In more than one-third of the cases, the gene is entirely deleted. There is no correlation between the type of mutation and the phenotype. Importantly, deletions frequently arise de novo in the proband. As observed in other syndromes, phenotypic variability can be observed between families and in family members with the same mutation.

Townes-Brock Syndrome

Townes-Brock syndrome (TBS) is an autosomal dominant malformation syndrome characterized by imperforate anus; preaxial polydactyly; or triphalangeal thumbs, external ear defects, sensorineural hearing loss, and, less frequently, kidney, urogenital, and heart malformations. REAR syndrome (renal-ear-anal-radial) has also been a term used to describe this condition. The gene mutated in human TBS is SALL1, a member of the Spalt family that is required for the normal development of the limbs, nervous system, kidney, and heart.

Renal Tubular Dysgenesis

Renal tubular dysgenesis (RTD) is a severe perinatal disorder characterized by the absence or paucity of differentiated proximal tubules, early severe oligohydramnios, and perinatal death usually due to pulmonary hypoplasia and skull ossification defects. It may be acquired during fetal development or inherited as an autosomal recessive disease. Autosomal recessive renal tubular dysgenesis is genetically heterogeneous and linked to mutations in the genes that encode components of the renin-angiotensin system. This condition also occurs in fetuses exposed in utero to angiotensin-converting enzyme (ACE) inhibitors or angiotensin II receptor antagonists.

Genetic Causes of Isolated (Nonsyndromic) Renal Malformation

In the majority of children with renal malformation, neither a syndrome nor a Mendelian pattern of inheritance is obvious. Yet familial clustering, observed in more than 20% of CAKUT patients, supports a genetic contribution to pathogenic mechanisms. It has become apparent that mutations in genes that control renal development in experimental animals can be identified in many such patients. Initial studies focused on examination of genes known to be mutated in syndromic cases of CAKUT. For example, analysis of a cohort of 100 patients with renal hypodysplasia and renal insufficiency demonstrated that 16% had mutations in 1 gene encoding for a transcription factor. The majority of mutations were identified in TCF2 (especially in the subset with kidney cysts) and PAX2. EYA1 and SALL1 mutations were found in single cases. Some of the mutations that were identified in these genes were de novo, which explains the sporadic appearance of renal malformations. Careful analysis of patients with TCF2 and PAX2 mutations revealed the presence of extrarenal symptoms in only half; this supports previous reports that TCF2 and PAX2 mutations can be responsible for isolated renal tract anomalies or at least CAKUT malformations with minimal extrarenal features. This study demonstrates that subtle extrarenal symptoms in syndromal renal malformations can easily be missed.

Other studies have focused more broadly on genes that play critical roles in renal development in mice but are not necessarily associated with syndromic forms of CAKUT. As an example, mutations in RET, which encodes a tyrosine kinase cell surface receptor, have been detected in patients with renal malformations. These mutations (Y791F and S649L) have also been detected in individuals with medullary thyroid carcinoma (MTC). None of the patients or their carrier relatives had clinical evidence of MTC at the time of the study. A low penetrance of renal malformation was suggested by minimal or no renal phenotypes in most carrier relatives.

More recently, targeted next-generation sequencing has been applied to the analysis of sporadic forms of CAKUT. A study of 208 genes, selected from studies on familial, syndromic, and experimental models of CAKUT, in a cohort of 453 CAKUT patients identified 148 candidate gene variants in 82 different genes in 151 patients. The vast majority of variants were heterozygous in nature. The frequency of mutations in genes previously implicated in CAKUT was less than 20% of patients studied. The pathogenicity of the majority of variants detected remains to be proven. Thus, this study demonstrates the breadth of genetic mechanisms that contribute to CAKUT.

MANAGEMENT

The widespread use and sensitivity of fetal ultrasound have increased the prenatal diagnosis of renal malformations. The sensitivity of prenatal ultrasound screening for renal malformations is about 82%, and the average time at which these malformations are detected is 23 weeks of gestation. In general, urinary tract malformations detected antenatally are isolated and present as mild hydronephrosis with no therapeutic consequences. Parents should be reassured. In contrast, bilateral forms of renal agenesis, severe dysgenesis, bilateral ureteric obstruction, or obstruction of the bladder outlet or the urethra can cause severe oligohydramnios as early as 18 weeks. Because amniotic fluid is critical to lung development, oligohydramnios as early as the second trimester can result in lung hypoplasia, a potentially fatal disorder. The oligohydramnios sequence, termed Potter syndrome, in its most severe form consists of a typical facial appearance characterized by pseudoepicanthus; recessed chin; posteriorly rotated, flattened ears and flattened nose; decreased fetal movement; clubfoot and clubhand; hip dislocation; joint contractures; and pulmonary hypoplasia. Severe renal malformations are usually the cause of oligohydramnios, which gives rise to the facial and limb deformities and impairs normal lung development. However, nonurinary causes such as amniotic fluid leakage and placental pathology could provoke oligohydramnios with the same consequences.

The renal prognosis can be evaluated antenatally, with poor outcome likely when there is severe oligohydramnios and small, hyperechogenic kidneys. Antenatal diagnosis and assessment of the renal prognosis are important for considering early termination in cases of fatal or eventually severe renal disease and to prepare parents and medical staff for the likelihood of neonatal renal insufficiency. Other organ malformations should be sought carefully, and if detected, a karyotype should be done.

The clinical presentation of renal malformation in the postnatal period depends on the amount of functioning renal mass, the presence of bilateral urinary tract obstruction, and the occurrence of urinary tract infection. Bilateral renal agenesis or severe dysplasia is likely to present soon after birth with decreased renal function. This may be accompanied by oliguria. Alternatively, patients may present with a flank mass or an asymptomatic abnormality detected by renal imaging.

A detailed history and careful physical examination should be carried out on all infants with an antenatally detected renal malformation. An early renal ultrasound (within 24 hours of life) is recommended for newborns with a history of oligohydramnios, progressive antenatal hydronephrosis, distended bladder on antenatal sonograms, and bilateral severe hydroureteronephrosis. In male infants, a distended bladder and bilateral hydroureteronephrosis may be secondary to posterior urethral valves, a condition that requires immediate renal imaging and clinical intervention. In general, unilateral anomalies do not require urgent investigation after birth. Rather, renal ultrasound is recommended after the first 72 hours of life, because urine output gradually increases over the first 24 to 48 hours of life as renal plasma flow and glomerular filtration rate increase. Thus, the degree of urinary tract dilatation can be underestimated if performed during this period of transition.

Genetic testing in children with renal malformations should be preceded by a thorough clinical evaluation for extrarenal symptoms, including eye, ear, and metabolic anomalies. The presence of nonrenal anomalies increases the likelihood of detecting a specific genetic abnormality (Table 465-3). In addition, mutations in genes that are usually associated with syndromes can occur in patients with isolated renal malformations.

CLINICAL APPROACH TO SPECIFIC MALFORMATIONS

Unilateral Renal Agenesis

A diagnosis of unilateral renal agenesis depends on the certainty that a second kidney does not exist in the pelvis or some other ectopic location. Since absence of 1 kidney induces compensatory hypertrophy in the existing kidney, the presence of a large kidney on 1 side suggests the possibility of unilateral renal agenesis. This condition is associated with contralateral urinary tract abnormalities, including ureteropelvic junction obstruction and vesicoureteral reflux in 20% to 40% of cases. Thus, imaging of the contralateral side by ultrasonography is indicated. If hydronephrosis and/or hydroureter is present, a voiding cystourethrogram may be indicated to identify vesicoureteral reflux. Managing affected patients involves determining the functional status of the contralateral kidney. A normal contralateral kidney is associated with an excellent long-term functional outcome. Since some studies have revealed that a substantial proportion of patients will develop proteinuria and hypertension in the long term, affected individuals should have their blood pressure measured and their urine periodically tested for protein throughout life.

Renal Hypoplasia

Unless associated with other malformations, renal hypoplasia can be asymptomatic. Unilateral hypoplasia is often discovered as an incidental finding during an abdominal sonogram or other imaging study. In contrast, patients with bilateral renal hypoplasia are at risk for decreased renal function and chronic kidney disease.

Renal Dysplasia

The dysplastic kidney is generally smaller than normal. However, cystic elements can contribute to large kidney size, the most extreme example being the multicystic dysplastic kidney. During the antenatal period, unilateral disease is likely to be discovered as an incidental finding. This may also be the case for bilateral renal dysplasia unless it is associated with oligohydramnios. After birth, bilateral renal dysplasia may limit glomerular filtration, causing renal failure that is usually progressive. Postnatal ultrasonography of the dysplastic kidney is characterized by increased echogenicity, loss of corticomedullary differentiation, and cortical cysts. Renal dysplasia is strongly associated with dilatation of the upper and lower urinary tract and with vesicoureteric reflux and posterior urethral valves. Accordingly, imaging of the lower urinary tract should be performed to determine whether these abnormalities are present. The approach to imaging is based on the findings of an ultrasound examination. Dilatation of the upper and/or lower urinary tract may serve as an indication to proceed with further imaging tests such as a voiding cystourethrogram (VCUG).

Multicystic Dysplastic Kidney

MCDK appears on ultrasonography as a large cystic nonreniform mass in the renal fossa; by palpation, MCDK appears as a flank mass. The MCDK is nonfunctional, a condition that can be demonstrated by imaging with MAG3 or DTPA radionuclide scanning. The MCDK is usually unilateral. If bilateral, it is fatal. Complications of MCDK include hypertension (0.01–0.1%), Wilms tumor, and renal cell carcinoma. However, the incidence of these disorders is extremely low. In 25% of cases, the contralateral urinary tract is abnormal. Contralateral abnormalities can include rotational or positional anomalies, renal hypoplasia, vesicoureteric efflux, and ureteropelvic junction obstruction. Contralateral ureteropelvic junction obstruction occurs in 5% to 10% of cases.

Gradual reduction in renal size and eventual resolution of the mass of the MCDK is common. At 2 years, an involution in size has been noted by ultrasound in up to 60% of affected kidneys. Complete disappearance of the MCDK can occur in a minority of patients (3–4%) by the time of birth and in 20% to 25% of patients by 2 years. Increase in the size of MCDK can be seen in some cases. The contralateral kidney shows compensatory hypertrophy by ultrasound evaluation.

Managing patients with MCDK has shifted from routine nephrectomy to observation and medical therapy. Because of the risk of associated anomalies in the contralateral kidney, the possibility of vesicoureteral reflux should be evaluated, particularly in patients with hydronephrosis and/or hydroureter, and blood pressure should be measured. Renal ultrasound is generally recommended at an interval of 3 months for the first year of life and then every 6 months up to involution of the mass, or at least up to 5 years. Compensatory hypertrophy of the contralateral kidney is expected and should be monitored by renal ultrasound.

Renal Ectopia

Normally, the kidneys lie on either side of the spine in the lumbar region and are located in the retroperitoneal renal fossae. Rapid caudal growth during embryogenesis results in migration of the developing kidney from the pelvis to the retroperitoneal renal fossa. With ascension comes a 90-degree rotation from a horizontal to a vertical position, with the renal hilum finally directed medially. Migration and rotation are complete by 8 weeks of gestation.

Simple congenital ectopy refers to a low-lying kidney that failed to ascend normally. An ectopic kidney that lies over the pelvic brim or in the pelvis is termed a pelvic kidney. Less commonly, the kidney may lie on the contralateral side of the body, a state that is termed crossed ectopy without fusion. Patients may be asymptomatic or symptomatic. An ectopic kidney is frequently discovered as an incidental finding during routine antenatal sonography. Symptomatic presentation occurs with urinary tract infections. On examination, an abdominal mass may be palpable. Other presenting features include hematuria, incontinence, renal insufficiency, and hypertension. Renal ectopia is highly associated with urologic abnormalities. Vesicoureteral reflux is the most common, occurring in 20% of crossed renal ectopia cases and 30% of simple renal ectopia cases. In bilateral simple renal ectopia, the incidence is as high as 70%. Other associated urologic abnormalities include contralateral renal dysplasia (4%), cryptorchidism (5%), and hypospadias (5%). Reduced renal function in the ectopic kidney is commonly observed by radionuclide scan. Female genital anomalies such as agenesis of the uterus and vagina or unicornuate uterus have also been associated with ectopic kidneys. Other anomalies described include adrenal, cardiac, and skeletal anomalies. Clinical assessment should therefore include a careful physical examination for other anomalies. Renal ultrasonography will help with diagnosis, defining the underlying anatomy, and will provide a basis for further imaging studies such as a VCUG. A dimercaptosuccinic acid (DMSA) scan is recommended as a tool to assess differential renal function.

Renal Fusion

Renal fusion is defined as the fusion of 2 kidneys. The most common fusion anomaly is the horseshoe kidney, in which fusion occurs at 1 pole of each kidney, usually the lower one. The fused kidney may be located in the midline (symmetric horseshoe kidney) or laterally (asymmetric horseshoe kidney). In a crossed fused ectopic kidney, the kidney from 1 side has crossed the midline to fuse with the kidney on the other side. Fusion is thought to occur before the kidneys ascend from the pelvis to their normal dorsolumbar position. This is usually between the fourth and ninth week of gestation. As a result, fusion anomalies seldom assume the high position of normal kidneys. The blood supply may be abnormal, such as from the iliac arteries. Abnormal rotation is also associated with early fusion of the developing kidneys.

The pelvis of each kidney lies anteriorly; therefore, the ureter traverses over the isthmus of a horseshoe kidney or over the anterior surface of the fused kidney. Ureteric compression may occur due to external compression by a traversing aberrant artery. The majority of patients may be asymptomatic. Some, however, develop obstruction that presents with loin pain and hematuria and may be associated with urinary tract infections due to urinary stasis or vesicoureteric reflux. Renal calculi may occur in up to 20% of cases. Other associated urologic anomalies include ureteral duplication, ectopic ureter, and retrocaval ureter. Genital anomalies such as bicornuate or septate uterus, hypospadias, and undescended testes have also been described. Associated nonrenal anomalies involve the gastrointestinal tract (anorectal malformations such as imperforate anus, malrotation, and Meckel diverticulum), the central nervous system (neural tube defects), and the skeleton (rib defects, clubfoot, or congenital hip dislocation). Investigations should begin with static imaging (renal ultrasound), the results of which provide a basis for further decision regarding the use of functional imaging including a DMSA scan and a VCUG.

SUGGESTED READINGS

INTRODUCTION

Virtually all renal cystic illnesses are monogenic diseases (Table 466-1). A recent unifying theory of their pathophysiology suggests that all gene products (“cystoproteins”) that are mutated in cystic kidney diseases localize to primary cilia, basal bodies, or centrosomes. Primary cilia are antenna-like cellular organelles produced by virtually every cell type in the body. They are important for perceiving extracellular cues, including photosensation, mechanosensation, osmosensation, and olfactory sensation. Cilia are assembled from basal bodies, which represent 1 of the 2 centrosomes. Centrosomes and basal bodies contain some of the same protein complexes that are part of the mitotic spindle in mitosis. These protein complexes are crucial for planar cell polarity, or the orientation of epithelial cells in 3-dimensional space. Disruption of their function leads to cyst development and to extrarenal defects that have been summarized under the term ciliopathies. In general, it seems that the pathogenesis of ciliopathies is based on an inability of epithelial cells to sense or process extracellular cues.

DIAGNOSIS

Cystic disorders of the kidney are among the most common causes for end-stage kidney disease (ESKD) in children. Since most renal cystic diseases are monogenic disorders, pedigree analysis to determine the pattern of inheritance should be emphasized when obtaining the history. In autosomal dominant diseases, a genetic defect in only 1 of the 2 alleles of the disease gene leads to the disease phenotype. Typically, there are affected individuals in more than 1 generation. Father-to-son transmission may be present (as opposed to X-linked diseases), and the disease course is usually milder and has later onset than in recessive variants of the disease. In autosomal recessive diseases, defects in both alleles of a gene (on both parental chromosomes) are necessary for the disease to occur. Typically, there are affected individuals in only 1 generation, the disease course is usually more severe than in a dominant variant of the condition, and onset occurs already in childhood or adolescence. Knowing the mode of inheritance is an important guide for differential diagnosis, for genetic counseling, and for obtaining a molecular genetic diagnosis. In some diseases, defects in different genes localized at different chromosomal loci can give rise to identical or very similar disease phenotypes in different patients. This phenomenon is called genetic locus heterogeneity and is exemplified by autosomal dominant polycystic kidney disease (ADPKD) types 1 and 2.

In most renal cystic disorders, the responsible gene has been identified, with the number of monogenic genes causing cystic kidney diseases exceeding 100. Thus, a definitive diagnosis may be made by whole-exome sequencing, which may identify the disease-causing mutation in affected patients. (Updates on laboratories that perform molecular genetic diagnoses can be found at http://www.genetests.org/ and www.renalgenes.org.) Molecular genetic diagnosis should be initiated ideally within the context of genetic counseling by a certified genetic counselor, because there may be some difficult ethical issues to consider.

Of note, molecular genetic diagnostics in ADPKD and medullary cystic kidney disease type 1 (MCKD1) cannot be achieved by whole-exome sequencing alone and may require specialized genetic amplification protocols.

MANAGEMENT

Renal cystic disorders eventually lead to ESKD. No treatment is available that changes disease progression or treats the underlying cause of these disorders. Therefore, therapy is symptomatic and is aimed at complications that might accelerate progression. This includes controlling blood pressure in patients with ADPKD to decelerate progression into renal failure. In the phase of compensated chronic renal insufficiency, symptomatic treatment for anemia, acidosis, growth retardation, and renal osteodystrophy is most important (see Chapter 473). Eventually, renal replacement therapy by dialysis and transplantation becomes necessary (see Chapter 474).

MULTICYSTIC DYSPLASTIC KIDNEY (MCDK)

MCDK, also termed cystic dysplasia, is a developmental defect in kidney organogenesis that is discussed in Chapter 466.

AUTOSOMAL RECESSIVE POLYCYSTIC KIDNEY DISEASE

ARPKD occurs in 1 in 6000 to 40,000 live births. In a study of 66 boys and 49 girls with ARPKD, the diagnosis was made prenatally in 11%, in the neonatal period in 41%, in infancy in 23%, and later in 25% (Fig. 466-1). Uniformly, there is bilateral kidney enlargement as a result of transformation of collecting ducts into fusiform cysts. Among extrarenal organ manifestations, congenital hepatic fibrosis is the most important one. ARPKD is always associated with hepatic fibrosis, which may not be clinically evident initially, because renal failure frequently develops earlier than clinical sequelae of hepatic fibrosis. Over time, cystic dilation of the biliary tree (Caroli disease) develops, followed by portal hypertension with a risk of gastrointestinal bleeding from esophageal varices. Liver cell function is initially normal until cirrhosis develops. Rarely, cysts are present in the pancreas or spleen. A few patients will not manifest the renal components of their disease and will have reasonably well-preserved renal function but will present with symptoms related to congenital hepatic fibrosis and portal hypertension.

A single recessive gene, PKHD1, has been identified by positional cloning as causing ARPDK. Because ARPKD is a recessive disease, an a priori risk of 25% exists for a sibling of an affected individual. ARPKD carries a high lethality in the perinatal period if 2 truncating mutations are present in the PKHD1 gene. If at least 1 missense mutation is present, end-stage renal disease develops later in childhood or adolescence.

DIAGNOSIS

The diagnosis of ARPKD rests on clinical signs and symptoms, renal and hepatic imaging, and molecular genetic diagnostics. Prenatal findings in ARPKD include oligohydramnios and a renal sonographic echotexture known as salt-and-pepper appearance. Enlargement of fetal kidneys may not occur until after 24 to 28 weeks of gestation, making early diagnosis difficult. Oligohydramnios from intrauterine renal failure is associated with insufficient lung development and the typical “Potter facies,” which is characterized by low-set ears, a flat nose, and a retracted chin. In these severe prenatal cases, abnormal kidney growth can already be detected by ultrasound in the second half of pregnancy. A definitive diagnosis of ARPKD is theoretically possible by direct mutation analysis of the PKHD1 gene, with an approximate 80% detection rate. Mutation analysis of the PKHD1 gene should be initiated with the guidance of a genetic counselor, for both pre- and postnatal testing. For the diagnosis of ARPKD, a negative renal ultrasound in the parents is obligatory in order to distinguish the disease from early-onset autosomal dominant polycystic kidney disease (ADPKD).

Initial features that lead to the diagnosis of ARPKD in infants and children include a palpable abdominal mass, hypertension, urinary tract infections, an affected sibling, liver involvement, and growth failure. Bilaterally enlarged hyperechogenic kidneys maintain a reniform shape even when grossly enlarged. On ultrasound, kidneys continue to exhibit a salt-and-pepper appearance until later stages, when distinct, rounded cysts appear. In ARPKD, the kidneys can be grossly enlarged so that palpable abdominal masses are common, as are hypertension and urinary tract infections. On hepatic ultrasound, the liver is hyperechogenic, and later, visible biliary cysts may develop. Renal and hepatic changes can also be confirmed by magnetic resonance imaging, where the radial arrangement of fusiformly dilated collecting ducts can be noted. Kidney biopsy plays a decreasing role in the diagnosis due to molecular genetic diagnostics. On renal histology, there is fusiform dilation and cyst development, involving 10% to 90% of collecting ducts, depending on severity and age of onset. Liver histology exhibits portal and interlobular fibrosis as well as biliary duct hyperplasia. Liver biopsy is useful to assess the degree of liver failure in later stages.

DIFFERENTIAL DIAGNOSIS

Other cystic diseases that occur in infancy and early childhood need to be considered (see Fig. 466-1). Specifically, medullary cystic kidney disease, infantile nephronophthisis, and Meckel-Gruber syndrome can be confused with ARPKD (see Table 466-1). Multicystic dysplastic kidney can be distinguished from ARPKD by reduced kidney size and loss of the reniform shape. ADPKD may occur rarely in infants if large deletions are present in the PKD1 gene. Meckel-Gruber syndrome, in addition to renal agenesis and dysgenesis, shows a wide variety of extrarenal organ involvement, including occipital encephalocele, cleft palate, polydactyly, and ocular anomalies, and usually leads to perinatal lethality. A rare but important differential diagnosis is infantile nephronophthisis, which can be diagnosed by mutation analysis of the NPHP2/INVS gene (www.renalgenes.org). Bilateral Wilms tumor is rare and can be distinguished from ARPKD by a highly irregular parenchymal pattern on imaging techniques. Bardet-Biedl syndrome can clearly be distinguished from ARPKD by the presence of retinitis pigmentosa and polydactyly, together with obesity, diabetes mellitus, mental retardation, and hypogonadism.

COMPLICATIONS

Chronic renal failure with anemia, growth retardation, and osteodystrophy is common. There is a high risk of urinary tract infection. Urinary outflow obstruction from renal cysts may also occur. Hypertension is frequent, and nephrolithiasis occurs in about 15% of cases. Bleeding from esophageal varices and hypersplenism can result from liver fibrosis. Ascending cholangitis occurs from infections of hepatic cysts. Fertility- and pregnancy-related problems are also common.

TREATMENT

The severe hypertension that occurs in young infants with ARPKD requires aggressive management to optimally preserve renal function. Early detection and aggressive treatment of urinary tract infections are also important. Managing portal hypertension frequently becomes a major problem. In end-stage renal disease, renal replacement therapy by dialysis and transplantation becomes necessary. Depending on whether liver failure has developed, combined kidney and liver transplantation or secondary liver transplantation after initial kidney transplantation may be warranted.

PROGNOSIS

Children presenting perinatally have a 30% to 50% mortality rate in the first month of life. Those who survive through the first month have a 1-year survival rate of 85% and a 10-year survival rate of 82%. Chronic renal failure is first detected at a mean age of 4 years. First renal replacement therapy is necessary by 5 years of age in 86% of cases. Three-quarters of cases develop systemic hypertension. Sequelae of congenital hepatic fibrosis and portal hypertension develop in 45% of patients and are related to age.

AUTOSOMAL DOMINANT POLYCYSTIC KIDNEY DISEASE

ADPKD is the most frequent lethal disease of autosomal dominant inheritance in the United States and Europe, with a prevalence of 1:1000 in the general population. About 10% to 15% of all patients in adult chronic dialysis programs have ADPKD. ESKD usually develops between age 60 and 70 years. Clinically, ADPKD2 is very similar to ADPKD1 but follows a somewhat milder course toward renal failure and does not present in childhood. ADPKD usually manifests in late adulthood but can lead to clinical symptoms in older children and young adults, including abdominal pain, palpable abdominal mass, hematuria, urinary tract infections, hypertension, and abdominal or inguinal hernias. Cerebral aneurysms and diverticulosis of the gut may develop in adults with ADPKD.

The PKD1 gene is localized on chromosome 16p13.1 and codes for polycystin-1, a protein with a large transmembrane domain and a complex extracellular domain. Polycystin-1 interacts with the gene product (polycystin-2) of the PKD2 gene. Although a germline mutation segregates in affected families in an autosomal dominant fashion, the disease mechanism, by which renal cysts develop gradually over several decades, is due to the loss of heterozygosity (LOH) of the second allele in individual cells, consistent with a “second hit mechanism.” In rare instances, ADPKD can occur in infancy if patients carry large deletions, which affect the PKD1 gene and the neighboring TSC2 gene, in way of a “contiguous gene deletion syndrome.” Among patients with tuberous sclerosis, apparently only individuals carrying such extensive deletions develop large bilateral renal cysts and exhibit infantile onset of ADPKD.

DIAGNOSIS

In children who are known to be genetically at risk due to ADPKD in a parent, the presence of even a single cyst in normal-sized kidneys is highly predictive for the development of ADPKD. Hepatic, pancreatic, or ovarian cysts are rarely detected before puberty. Molecular genetic diagnostics of PKD1 can be initiated after genetic counseling. Early genetic diagnosis may be valuable to monitor for complications such as hypertension, urinary tract infection, and cerebral aneurysms and to evaluate living related kidney transplant donors.

TREATMENT

Although in ADPKD chronic renal failure develops at a median age of approximately 65 years, at-risk children and young adults should be followed annually for hematuria and hypertension, and at greater intervals by renal ultrasound, especially during pregnancy. Aggressive treatment of hypertension or urinary tract infection is indicated, because both pose considerable risk for accelerated progression into ESKD. In families with a history of cerebral aneurysms, or if symptoms warrant, cranial computed tomography (CT) or magnetic resonance imaging (MRI) should be followed, but routine screening is not usually undertaken. Phase II trials have been initiated for some experimental treatment approaches in adult patients with cystic kidney diseases.

NEPHRONOPHTHISIS AND MEDULLARY CYSTIC KIDNEY DISEASE

Diseases of the nephronophthisis (NPHP)/medullary cystic kidney disease (MCKD) complex are renal cystic diseases that share a virtually identical renal histology, characterized by thickening and disintegration of the tubular basement membrane, interstitial lymphocytic infiltrations with fibrosis, and distal tubular atrophy with cysts (Fig. 466-2A). Over time, chronic sclerosing tubulointerstitial nephropathy develops. In contrast to polycystic kidney disease, cysts occur primarily at the corticomedullary junction of the kidneys, and kidney size remains normal or is slightly diminished (Fig. 466-2B). Although histology is similar in all forms of the disease, the 2 different disease groups of recessive NPHP and dominant MCKD can be clearly distinguished by age of onset or pattern of inheritance.

NEPHRONOPHTHISIS

Autosomal recessive NPHP leads to ESKD at a median age of 13 years. In NPHP, a history of polyuria, polydipsia, anemia, and growth retardation is common. Patients often describe nighttime fluid intake starting at about 6 years of age. Partially due to excessive salt loss, these children do not develop the usually typical symptoms of chronic renal disease (hematuria, proteinuria, hypertension, or edema). Therefore, the diagnosis is usually made late in the course of the disease. Renal ultrasound is an important aid in diagnosing NPHP (Fig. 466-2B). The ultrasonographic characteristics of NPHP are normal or slightly reduced kidney size; increased echogenicity; a lack of corticomedullary differentiation; and, beyond the age of 9 years, cysts at the corticomedullary border of the kidneys. A history of similar disease in preceding generations suggests a diagnosis of dominant MCKD, whereas a lack of family history suggests recessive NPHP. Today whole-exome sequencing following informed consent will reveal a causative mutation in 1 of about 100 NPHP-causing genes in 50% to 70% of cases.

Extrarenal manifestations may occur in all forms of recessive NPHP, so ophthalmologic examination and brain MRI are warranted. Retinitis pigmentosa is associated with NPHP in Senior-Løken syndrome and cerebellar vermis aplasia and coloboma of the eye in Joubert syndrome. The inability to perform horizontal eye movements (ocular motor apraxia type Cogan) was found in infants and toddlers with homozygous deletions in the NPHP1 gene. An association of NPHP with hepatic fibrosis or cone-shaped epiphyses has also been described. In addition, truncating mutations in NPHP3, NPHP6, and NPHP8 can lead to Meckel-Gruber syndrome. If obesity, diabetes mellitus, infertility, or polydactyly is present, Bardet-Biedl syndrome should be considered and may be confirmed by mutation analysis. In addition, Jeune syndrome (asphyxiating thoracic dysplasia) and Ellis-van Creveld syndrome can be part of the differential diagnosis.

Mutations in 9 different recessive genes have been identified as causing NPHP. These are NPHP1, NPHP2/inversin, NPHP3, NPHP4, NPHP5, NPHP6/CEP290, NPHP7/GLIS2, NPHP8/RPGRIP1L, and NPHP9/NEK8, defining NPHP types 1 through 9, respectively (see Table 466-1). Gene identification generated new insights into disease mechanisms of NPHP, as described above and shown in Figure 466-3. Gene identification has made definite molecular genetic diagnostics possible for approximately 30% of cases. Deletions of the NPHP1 gene on both parental chromosomes account for approximately 21% of all NPHP cases, whereas the other genes, NPHP2 to NPHP9 (see Table 466-1), contribute about 1% each. The causative genes are still unknown in about 70% of cases. Following genetic counseling, mutation analysis will first be performed in the most frequent form, NPHP1. Mutation analysis will unequivocally confirm the diagnosis of NPHP in approximately 30% of children with NPHP. If changes on renal ultrasound are seen before the age of 4 years, mutation analysis is warranted in NPHP2/INVS.

MEDULLARY CYSTIC KIDNEY DISEASE

MCKD is characterized by autosomal dominant inheritance and adult-onset of ESKD. The extrarenal involvement seen in NPHP is absent. However, MCKD can be associated with hyperuricemia and gouty arthritis in most affected individuals. A gene locus (MCKD1) for MCKD type 1 has been mapped to chromosome 1q21, but the gene is unknown. Mutations of the UMOD gene encoding uromodulin/Tamm-Horsfall protein were detected as the cause of MCKD type 2. Histology of MCKD is indistinguishable from that seen in NPHP (see Fig. 467-2A). Definitive molecular genetic diagnostics is possible by UMOD mutation analysis. Mutations in UMOD may also cause the clinical picture of glomerulocystic kidney disease (see the following section). MCKD1 diagnosis requires a specialized sequencing approach due to the complex nature of MUC1 mutations.

GLOMERULOCYSTIC KIDNEY DISEASE

The term glomerulocystic kidney disease (GCKD) describes a heterogeneous group of diseases characterized by multiple small cortical cysts that result from cystic dilation of the Bowman space and the initial proximal convoluted tubule. The lack of further tubular involvement differentiates GCKD from other cystic diseases, in which cysts arise from tubular dilation. Although this histologic picture is found in families with ARPKD and in several rare diseases, including orofacial-digital syndrome type I, a distinct disease entity called familial hypoplastic glomerulocystic disease exists which follows autosomal dominant inheritance.

The typical presentation of non–autosomal dominant GCKD is an infant with an abdominal mass and renal insufficiency. Hypertension is common at presentation. Hepatic cysts have been described. Some patients present only as adults with hypertension, flank pain, and hematuria. The degree of renal dysfunction is variable. Children with familial hypoplastic GCKD (autosomal dominant) have chronic renal failure early in life, but some have stable courses without progression to end-stage renal disease. Renal ultrasound shows small kidneys and an abnormality of medullary pyramids with collecting systems and absent papillae. Dominant mutations of HNF1B/TCF2 have been described as the cause of GCKD in association with maturity-onset diabetes of the young type 5 (MODY5). In addition, mutations of UMOD are known to cause isolated GCKD. This may be understood within the ciliopathy theory of cystic kidney diseases, as UMOD is under the transcriptional control of HNF1B/TCF2.

OTHER RENAL CYSTIC DISORDERS

MECKEL-GRUBER SYNDROME (MKS)

MKS is an autosomal recessive inherited disorder featuring developmental defects of different organs, including the kidney, and usually leads to perinatal death. The diagnosis is based on early onset; renal ultrasonography; and the presence of cystic dysplasia of the kidney, hepatic fibrosis, occipital encephalocele, and postaxial polydactyly. Renal or liver biopsy sometimes has a diagnostic role. Prognosis is guarded in cases with onset of renal insufficiency in the neonatal period. In the differential diagnosis of MKS, ARPKD, Joubert syndrome, and carbohydrate-deficient glycoprotein disease have to be considered. Two causative genes (MKS1 and MKS3/TMEM67) have been identified, and 1 gene locus has been mapped (MKS2). In addition, mutations in NPHP3, NPHP6, and NPHP8 can lead to MKS if 2 truncating mutations are present, whereas the disease variant is milder in the presence of at least 1 missense mutation, showing the features of Joubert syndrome or Senior-Løken syndrome.

BARDET-BIEDL SYNDROME (BBS)

BBS is an autosomal recessive disease characterized by obesity, retinitis pigmentosa, infertility, polydactyly, mental retardation, and the occurrence of cystic dysplasia of the kidney or an NPHP-like histology. Twelve different causative recessive genes (BBS1–BBS12) have been identified by positional cloning. Direct molecular genetic diagnosis of BBS is possible in certain cases. Regarding its pathophysiology, BBS is also considered a “ciliopathy,” in which multiple organ involvement can be explained through defects in the function of primary cilia in multiple organs (see Fig. 466-3).

MEDULLARY SPONGE KIDNEY

Medullary sponge kidney is a rare congenital cystic disorder in which there is ectasia of cortical ducts within the inner medulla of the kidney, resulting in a spongelike appearance. Calcifications, which develop in these locations, are visible on plain x-ray films. There is a strong propensity to renal calculi and urinary tract infections. Recently, familial cases have been described within the familial ureteral abnormalities syndrome, which raises the question of whether there might exist a monogenic form of this disorder. The characteristic pyelogram with linear striations from dilated collecting ducts and enlarged calices is usually diagnostic of the disease. Generally there is no role for sonography, CT, or arteriography. The prognosis is good, because in the absence of complications, renal function is normal, and there is no increased morbidity or mortality resulting from the disease.

SUGGESTED READINGS

INTRODUCTION

Acute kidney injury (AKI) is “an abrupt (within 48 hours) reduction in kidney function defined as an absolute increase in serum creatinine by > 0.3 mg/dL or a relative increase of > 50% from baseline, or oliguria of < 0.5 mL/kg/h for > 6 hours.” This definition has replaced the old term acute renal failure, classically hallmarked by a sudden and rapid decline in glomerular filtration rate (GFR), leading to accumulation of nitrogenous wastes such as blood urea nitrogen (BUN) and creatinine. This change has been the result of a deliberate campaign by experts in the field to underscore the association of even subtle disturbances in renal function with adverse outcomes, while emphasizing that AKI is a continuum on a spectrum from subclinical renal injury to full-blown complete failure with total loss of kidney function. Even small increases in serum creatinine are now recognized as contributing to poor outcomes. In adults, an increase in serum creatinine of only 0.3 mg/dL was associated with increased mortality rates, even when controlling for significant patient comorbidity. Similar results were noted in pediatric patients with acute decompensated heart failure, in whom a 0.3-mg/dL or greater rise in serum creatinine demonstrated a 7-fold increased risk of mortality. These findings highlight the need for a focus on earlier detection and treatment of AKI, well before the serum creatinine begins to rise and other nonrenal complications ensue. With improved understanding about incomplete recovery from AKI episodes leading to chronic kidney disease (CKD) and the independent mortality and morbidity risk that AKI poses in hospitalized adults and children, AKI is now seen as a public health threat. Unfortunately, there are no satisfactory therapeutic options.

During the last decade, the definition of AKI has been standardized by the proposition and subsequent validation of consensus criteria. Currently, there are 3 standardized definitions for AKI that have been modified for and validated in pediatric patients, pRIFLE (Pediatric Risk, Injury, Failure, Loss, End-Stage Renal Disease), Acute Kidney Injury Network (AKIN), and Kidney Disease: Improving Global Outcomes (KDIGO) in chronological order (Table 467-1). All of these definitions rely on relative changes in serum creatinine from baseline and varying degrees of oliguria. Based on these definitions, we now know that AKI is a common finding in critically ill children, seen in as many as 30% of all children admitted to intensive care units (ICUs) and 50% of patients with shock, and increases risk of death by approximately 15-fold independent of comorbidities. Furthermore, mortality and renal morbidity rates increase with increasing AKI severity.

CLASSIFICATION

Oliguria, defined as a urine output of < 0.5 mL/kg/h, is the most readily available tool for assessment of renal function at the bedside. However, oliguria occurs in only about half of cases. AKI can be classified as oligoanuric versus nonoliguric, depending on the amount of urine production. Duration of oliguria to diagnose AKI is controversial, ranging from 6 to 24 hours depending on the definition used.

From an etiologic perspective, AKI can result from ischemic, toxic, or inflammatory causes. Mechanistically, AKI can be classified as (1) functional (prerenal) AKI, due to a functional response of structurally normal kidneys to hypoperfusion; (2) intrinsic AKI, due to structural damage to the kidneys from prolonged ischemia, nephrotoxins, sepsis, or intrinsic renal disease; and (3) postrenal AKI, due to obstruction of the urinary tract.

Another classification that can be applied is the setting in which AKI is seen: community-acquired versus hospital-acquired AKI. Community-acquired AKI tends to be either prerenal/hypoperfusion related or secondary to primary renal diseases such as poststreptococcal glomerulonephritis or hemolytic-uremic syndrome. Hospital-acquired AKI is usually multifactorial in etiology, happening in “at-risk” individuals who often are critically ill or have other serious comorbidities. Nephrotoxin-mediated AKI can be seen in both settings. There seems to be a threshold effect to nephrotoxins, with patients receiving 3 or more nephrotoxic medications having increased risk of developing an AKI.

Functional AKI (prerenal AKI) usually is rapidly reversed by restoration of renal perfusion, but early treatment is essential in order to prevent the progression to intrinsic AKI. Once established, there is no effective treatment for AKI, and the clinician can provide only supportive care with medical management and dialysis. In fact, timely treatment of septic shock with adequate fluid and antibiotic therapy has been shown to decrease AKI. Although the worst outcomes are encountered in patients requiring dialysis, even mild degrees of AKI, such as occurs with only a small increase in serum creatinine, are predictive of an increase in mortality and morbidity rates, irrespective of the underlying cause. During the last decade, the cellular and molecular tools of modern science have provided critical new insights into the pathogenetic mechanisms and early diagnosis of AKI. Improved understanding of the pathophysiology and early diagnosis enabled by new biomarkers will facilitate novel strategies that target the newly discovered pathways to prevent and treat AKI in the near future.

EPIDEMIOLOGY

The global incidence of pediatric AKI is on the rise, although the precise rate remains unknown. There is a component of increased recognition that has paralleled the attempts to streamline clinical definitions and staging. Nonetheless, pediatric standardized definitions have been used in only a minority of publications, and all date after 2010. A recent meta-analysis including more than 150 studies and more than 3 million subjects revealed that 1 in 5 adults and 1 in 3 children acquired AKI during hospitalization. The incidence of AKI requiring renal replacement therapy (RRT) has also increased significantly during the last decade.

Etiology depends on the clinical setting and the geographical location. Prior to the 1990s, pediatric studies reported primary renal causes such as hemolytic-uremic syndrome and other infections as the most prevalent causes leading to AKI. Since the introduction of standardized definitions, numerous reports have identified a dramatic shift in the epidemiology of AKI. It is now commonly seen in the context of multiple-organ failure and is frequently a hospital-acquired illness, with the most common causes being renal ischemia, nephrotoxin use, congenital heart disease, bone marrow transplantation, and sepsis. Resource-limited settings of Africa and tropical Asia report community-acquired AKI secondary to gastroenteritis, septicemia, acute glomerulonephritis, or falciparum malaria. Hemolytic-uremic syndrome and leptospirosis are common occurrences in Latin America, and AKI due to snakebites often is encountered in rural Asia. In a report focusing on outcomes of pediatric AKI in sub-Saharan Africa, the most common etiologies for AKI were infection, glomerular disease, nephrotoxins, and volume depletion. In particular, malaria emerged as a major cause when AKI was seen in the setting of a specific disease. In addition, intoxications such as hair dye (paraphenyldiamine) and solvent (diethyl glycerol) are also common causes of AKI in Africa.

Nosocomial pediatric AKI rates are escalating at an alarming rate, more than 9-fold from the 1980s through the 2000s. Frequently, the causes are multifactorial, impacting patients already at risk due to underlying chronic diseases or comorbid conditions. A review of an administrative database in which more than 2.5 million admissions were screened for coding-based AKI diagnoses revealed an incidence of 3.9 cases per 1000 admissions, with an age difference; the highest event rate of 6.6 per 1000 admissions was seen in those age 15 to 18 years old. Patients with AKI had a 25-fold increase in mortality rates (15% vs 0.6%). The incidence of the most severe forms of AKI, defined by dialysis requirement, ranges from 1% to 2% of all critically ill children. When less strict criteria for AKI are used, such as doubling of serum creatinine, the incidence rises to 21% to 25% of children in ICUs. In a recent single-center study from Canada exploring risk factors for AKI in critically ill children, AKI defined by RIFLE or KDIGO criteria developed in 24% of approximately 4000 patients over the course of 3 years. The incidence jumps to greater than 80% in critically ill children who receive invasive mechanical ventilation or at least 1 vasoactive medication. Unfortunately, despite more than 80% of practicing intensivists being familiar with standardized criteria, fewer than 50% of them apply these definitions and staging in clinical practice, leading to underrecognition and delayed treatment of AKI. AKI also complicates as many as 24% of neonatal ICU admissions and is present in 60% of neonates with severe asphyxia. A recent international multicenter study of critically ill children found a daily prevalence of AKI that increases from 15% to 20% within the first week of pediatric ICU (PICU) admission, with an overall incidence of 27%. Approximately 40% of patients with sepsis develop AKI, with 20% requiring RRT. In children undergoing cardiopulmonary bypass, the incidence of AKI is in the range of 30% to 50%, depending on the definition used. In children receiving stem cell transplants, the incidence of AKI (defined by doubling of serum creatinine) is approximately 20%, increasing to more than 80% in the first 100 days after transplantation.

The severity of illness and other organ involvement are the most important risk factors associated with development of AKI in critically ill children. In a single-center study from Canada exploring risk factors for AKI in the PICU, circulatory disorders, shock, extracorporeal membrane oxygenation, respiratory dysfunction, and nephrotoxin use were associated with increased risk of developing AKI. The single greatest risk factor for development of ICU-acquired AKI was the use of nephrotoxic medications. However, the first creatinine in the first 24 hours of PICU admission was used as the baseline creatinine, and as we have learned from earlier studies using standardized definitions of AKI, 30% of patients already have AKI on admission to the PICU. Therefore, AKI was likely underdiagnosed in this cohort. In another single-center report that predates the current standard definitions, with AKI defined as what would most closely resemble KDIGO stage II when standard definitions are applied, age older than 12 years, thrombocytopenia, hypoxemia, hypotension, and coagulopathy were significant risk factors for developing AKI. In another single-center study of more than 3000 PICU patients, there were associations among age, weight, vasopressor requirement, diuretic drip use, mechanical ventilation, and illness severity score with AKI occurrence. In a series reported in Taiwan, treatment with vasopressors, mechanical ventilation, and a diagnosis of malignancy were independent predictors of mortality in pediatric patients with AKI. In neonates, low birth weight, perinatal asphyxia, low gestational age, congenital diaphragmatic hernia and ECMO, bronchopulmonary dysplasia, and maternal exposure to certain medications such as angiotensin-converting enzyme inhibitors are associated with increased rates of AKI. In addition, patent ductus arteriosus and treatment with indomethacin or ibuprofen are associated with higher incidence of AKI, as are postnatal exposure to other nephrotoxic agents such as gentamicin and vancomycin.

DIAGNOSIS

The current clinical diagnosis of AKI relies on absolute increases in serum creatinine compared to baseline levels. Serum creatinine is a by-product of muscle metabolism and, thus, is impacted by muscle mass. Normal levels of serum creatinine vary widely in the pediatric population due to changes with age and body size, rendering a single cutoff threshold across all age groups impossible. In addition, serum creatinine is inversely related to changes in total body water. In other words, patients who are volume depleted (such as in dehydration or low serum albumin states like nephrotic syndrome) can have elevations in serum creatinine that are simply markers of contracted fluid status and resolve with appropriate fluid therapy and are not reflective of structural damage in the kidneys. Conversely, in patients with fluid accumulation from any reason, serum creatinine will be lower, leading to potential underdiagnosis of AKI. In addition, there is evidence that endogenous creatinine production is reduced in critical illness. Serum creatinine is an excellent marker of GFR in steady state but is a late marker of acute changes in GFR. So despite being very useful in CKD where GFR is more stable, in acutely changing conditions of developing AKI, GFR needs to decrease by approximately 50% before a change in serum creatinine is seen. Therefore, serum creatinine elevation lags behind onset of AKI by approximately 24 to 72 hours, rendering it a late and suboptimal biomarker. Current definitions including pRIFLE and KDIGO still rely on changes in serum creatinine and, therefore, can miss early AKI. In addition, current definitions cannot reliably differentiate intrinsic AKI from functional AKI, rendering new diagnostic tools necessary.

An alternative to serum creatinine is serum cystatin C, a cysteine protease inhibitor present in all nucleated cells, freely filtered by the glomerulus and completely taken up and metabolized by intact proximal tubule. Cystatin C is proposed as a more accurate marker of GFR because it is not impacted by hydration status, muscle mass, and age (children > 12 months old have the same normal levels as adults). However, thyroid function, steroid use, and certain inflammatory conditions may impact serum cystatin C levels. Regardless, cystatin C is another valuable marker that can be used in certain conditions, such as malnutrition, muscle wasting, or fluid overload. In critically ill children, serum cystatin C predicts elevation in serum creatinine 24 hours in advance.

Both serum creatinine and serum cystatin C are functional biomarkers. Structural tubular damage precedes any fluctuations in serum creatinine in AKI. Advances in basic research illuminating the pathogenesis of AKI and leading to successful therapeutic interventions in animal models have not translated successfully to humans. A major reason for this is the lack of early markers for AKI, akin to troponins in acute myocardial disease and, hence, an unacceptable delay in initiating therapy. Animal studies have shown that while AKI can be prevented or treated by several maneuvers, these measures must be instituted very early after the insult. The lack of early biomarkers of acute renal injury in humans has hitherto crippled our ability to launch potentially effective therapies in a timely manner. Several potential candidates have been developed and tested as early biomarkers of acute renal injury (Table 467-2). Examples include neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), and interleukin (IL)-18. There are now numerous multicenter studies of these biomarkers, alone or in combination, in adults with different risk factors and conditions, with varying performance. The pediatric biomarker studies are largely limited to homogenous populations such as postcardiac surgical populations or children with primary kidney diseases (renal transplant, nephrotic syndrome, and glomerulonephritis); most studies are single-center with small sample sizes. Multicenter studies are in children undergoing cardiopulmonary bypass, where the timing of the renal injury is well known, lending to easier identification of sampling windows. Urinary NGAL and IL-18 predicted AKI severity in critically ill children; however, the urine samples were obtained 24 to 72 hours after admission to the ICU, preventing study of these biomarkers as early AKI detection tools. In a pilot study in 232 children presenting to the emergency department with milder disease (as demonstrated by only 1% of the cohort being admitted to the ICU), KIM-1, NGAL, and β₂-microglobulin correlated with mild AKI (pRIFLE). In 288 children undergoing cardiopulmonary bypass, postoperative plasma cystatin C predicted length of ICU stay. NGAL and IL-18 predicted AKI in 20 children undergoing cardiopulmonary bypass. In 48 premature babies, serial KIM-1 indicated renal injury. In 214 children with severe sepsis, the biomarkers plasma NGAL, MMP-8, and Ela-2 predicted severe AKI in risk-stratified patients. In 49 heterogeneous PICU patients, NGAL predicted AKI, and cystatin C performed the same as creatinine.

Subclinical AKI, characterized by normal (or baseline) serum creatinine but elevations in structural damage biomarkers, is a new term introduced with the aid of this recent translational research. Adult patients seen in the emergency room and admitted to the hospital with elevated NGAL and KIM-1 levels despite baseline serum creatinine had adverse outcome (dialysis or mortality) rates comparable to patients with AKI diagnosed with creatinine elevations. Similar results have been reported in other studies. Children have different comorbidities than adults, and biomarker levels are proposed to vary with age, rendering the precise application of cutoffs generated in adult studies difficult. However, integration of structural biomarkers into clinical definitions driven by functional markers will likely further refine AKI diagnosis and staging.

A newer generation of biomarkers, called cell cycle arrest biomarkers, has been studied for early diagnosis of AKI. When cells are “stressed,” they respond by becoming senescent and arrest the cell cycle in G₁ phase. Tissue inhibitor of metalloproteinase-2 (TIMP-2) and insulin-like growth factor binding protein 7 (IGFBP7) are both inducers of the G₁ phase of cell cycle and can be detected in the urine 12 hours before elevation in serum creatinine occurs. In 133 heterogenous critically ill children, TIMP-2 and IGFBP7 predicted AKI and adverse outcomes. There are some conflicting data regarding impact of common comorbidities on the cell cycle arrest biomarkers. It has been proposed that decreasing urinary levels of these biomarkers might represent tissue healing and signal recovery from AKI.

One of the major drawbacks of the newer biomarkers is the fact that their peak is evanescent. In addition, because AKI is “painless” and symptomatic only in the late stages, efforts have been directed to focus biomarker testing in the highest risk group. Combining structural damage markers with clinical stratification has been proposed as a useful tool to increase pretest probability and increase meaningful utilization. With peaks occurring at different times after initial insult, it is likely that a collection of strategically selected proteins may provide the hitherto elusive “AKI panel” for the early and rapid diagnosis of AKI.

PATHOPHYSIOLOGY

Functional (Prerenal) AKI

Prerenal or functional AKI, also referred to as prerenal azotemia in the past, is an appropriate functional response of structurally normal kidneys to hypoperfusion. The oliguria in this situation is frequently an appropriate renal response for preserving intravascular volume; however, in certain instances such as congestive heart failure, it could be maladaptive. Table 467-3 lists the common causes of prerenal AKI. A reduction in circulatory volume—actual or effective—evokes a systemic response aimed at normalizing intravascular volume at the expense of GFR. Baroreceptor-mediated activation of the sympathetic nervous system and renin-angiotensin axis results in afferent renal vasoconstriction and the consequent reduction in GFR (Fig. 467-1). In response to renal hypoperfusion, several intrarenal autoregulatory mechanisms help maintain GFR (Fig. 467-2). Prolonged reduction in renal perfusion pressure (eg, due to dehydration or systemic hypotension) can overwhelm these compensatory mechanisms, and intrinsic AKI can set in.

The most effective compensatory mechanism involves the intrarenal generation of vasodilatory prostaglandins that dilate the afferent arterioles. Nonsteroidal anti-inflammatory drugs (NSAIDs) inhibit this response and can precipitate AKI, especially in the presence of decreased circulatory volume. Younger children who are febrile and then become dehydrated are at particular risk for developing AKI when treated with NSAIDs. In fact, despite adherence to recommended dosing guidelines in a majority of patients, NSAIDs were responsible for 3% of de novo AKI in hospitalized children. A second mechanism results from a differential in the constrictor effect of angiotensin II on the efferent versus afferent arteriole. Angiotensin II constricts both the afferent and efferent arterioles, but this effect is more marked in the efferent arteriole, leading to increased hydrostatic pressure across the glomerulus and maintenance of GFR. Angiotensin-converting enzyme inhibitor (ACEi) therapy can interfere with this compensatory mechanism. Because the autoregulation of GFR is more tenuously balanced in newborns, ACEis have been associated with precipitous declines in GFR in infants. If patients taking ACEis become dehydrated, GFR will decline and AKI can develop. A third mechanism, termed myogenic autoregulation, refers to the unique ability of the afferent arterioles to rapidly vasodilate in response to a decrease in lateral stretch caused by hypoperfusion. Immunosuppressant medications commonly used in patients after kidney transplantation (eg, cyclosporine) can interfere with the myogenic mechanism.

Intrinsic AKI

Intrinsic AKI usually is caused by prolonged ischemia, nephrotoxins, or sepsis (see Table 467-3). In clinical practice, especially in critically and/or chronically ill children, it is frequently multifactorial, often occurring concomitantly with overlapping pathogenic mechanisms. The most severe histopathologic finding in experimental models of intrinsic AKI is acute tubular necrosis (ATN), which led to the terms intrinsic AKI and ATN to be used interchangeably. However, ATN is a misnomer because histopathology of AKI specimens is surprisingly bland in human AKI and frank tubular cell necrosis is rarely found. Instead, there is effacement and loss of proximal tubule brush border, patchy loss of tubule cells, focal proximal tubular dilatation and distal tubular casts, and areas of cellular regeneration. Necrotic cell death is restricted to the highly susceptible outer medullary regions (the S3 segment of the proximal tubule and medullary thick ascending limb of Henle’s loop), which are relatively hypoxic despite high metabolic demand of the region. Apoptosis in both distal and proximal tubules plays a pivotal role in both ischemic and nephrotoxic forms of human AKI. In addition, peritubular capillaries in the outer medulla have been shown to display a striking vascular congestion and leukocyte accumulation. Animal models of AKI usually have a single mechanism of injury and do not recapitulate human AKI well, especially in the setting of multiple comorbidities such as cancer, immune-suppressed status, infections, and hemodynamic perturbations. However, they have greatly enhanced our understanding of the principles of evolving AKI pathophysiology over the years. The mechanisms underlying the morphologic changes observed in AKI are detailed below.

There is evidence for hemodynamic alterations in intrinsic AKI. Total renal blood flow is reduced to approximately 50% of normal due to persistent intense renal vasoconstriction, as shown in Fig. 467-3. Furthermore, there are regional alterations in renal blood flow, with marked congestion of the outer medullary region, where oxygen tensions are normally low. This region contains tubular segments with high energy requirements, namely the S3 segment of the proximal tubule and medullary thick ascending limb. The postischemic congestion worsens the relative hypoxia, leading to prolonged injury and necrotic cell death in these segments. Cellular mechanisms underlying these hemodynamic alterations relate primarily to endothelial damage, leading to a local imbalance of vasoactive substances, including enhanced release of the vasoconstrictor endothelin and reduced release of vasodilatory endothelium-derived nitric oxide. Endothelin receptor antagonists ameliorate AKI in animal models, but human data are lacking. However, these hemodynamic abnormalities do not account for the profound loss of renal function, and several human trials of vasodilators such as dopamine have failed to demonstrate improvement in GFR in established AKI despite augmentation of total renal blood flow. In fact, endothelin-receptor antagonists led to a paradoxical worsening of renal injury in human radiocontrast-induced AKI.

Proposed alterations in tubular dynamics in established AKI include obstruction, back-leak, and activation of tubuloglomerular feedback. Their interplay is illustrated in Fig. 467-3. The consistent finding of proximal tubular dilatation and distal tubular casts in human AKI are indicative of obstruction to tubular fluid flow. Tamm-Horsfall protein, which is normally secreted by the proximal tubule as a monomer, is converted into a gel-like polymer in the sodium-rich intraluminal environment of the distal tubule, leading to cast formation with incorporation of desquamated cells and shed brush border. It is unlikely that obstruction alone can account for the profound dysfunction in clinical AKI because human studies using forced diuresis with furosemide or mannitol do not show an impact on the survival and renal recovery rate of patients with established AKI. Similarly, although movement of the glomerular filtrate back into the circulation has been shown to occur, it accounts for only a very minor component of the decrease in GFR in human AKI.

The role for activating tubuloglomerular feedback is controversial. The increased delivery of sodium chloride to the macula densa due to abnormalities in the ischemic proximal tubule would be expected to induce afferent arteriolar constriction via A1 adenosine receptor (A1AR) activation and thereby decrease GFR. However, a knockout of the A1AR resulted in a paradoxical worsening of ischemic renal injury in mice, whereas exogenous activation of A1AR was protective. Thus, tubuloglomerular feedback activation following ischemic injury may represent a beneficial phenomenon that limits wasteful delivery of ions and solutes to the damaged tubules. It also might be protective by decreasing energy consumption in the metabolically active proximal tubule that is already under stress, thereby decreasing oxygen demand. This phenomenon might be more important in the outer medulla, which is relatively hypoxic, receiving less of the total renal blood flow, and has a high oxygen extraction ratio even in baseline healthy conditions.

In injured kidney tubule cells, a rapid and profound reduction in the content of intracellular adenosine triphosphate (ATP) occurs, which triggers several metabolic events (Fig. 467-4). Oxygen deprivation leads to a rapid degradation of ATP to adenosine monophosphate (AMP) and to hypoxanthine. These metabolites are freely diffusible from tubule cells, and their depletion prevents resynthesis of ATP during reperfusion. Although provision of exogenous adenine nucleotides or thyroxine (which stimulates mitochondrial ATP regeneration) can ameliorate the cellular injury in animal models of ischemic AKI, it has proven ineffective in established human AKI. An increase in free intracellular calcium has also been documented following AKI, but its role has remained controversial; calcium channel blockers may provide some protection from renal injury in the transplant setting, but evidence for their efficacy in other forms of AKI is lacking.

Reactive oxygen species play a substantial role in the pathogenesis of AKI. During reperfusion, hydrogen peroxide and superoxide are generated in tubule cells. In the presence of iron, hydrogen peroxide forms the highly reactive hydroxyl radical. Concomitantly, ischemia induces nitric oxide synthase in tubule cells, and the nitric oxide generated interacts with superoxide to form peroxynitrite, which results in cell damage via oxidant injury and protein nitrosylation. Several scavengers of reactive oxygen molecules protect against AKI in animals, but human studies have been inconclusive. Also, the iron scavenger deferoxamine ameliorates ischemia-reperfusion injury in animal models, but the associated toxicity precludes its clinical use in human AKI. Iron chelation seems to be protective against AKI; human apotransferrin, an iron-binding protein, protects against renal ischemia-reperfusion injury in animals by abrogating renal superoxide formation. NGAL, an endogenous iron-transporting protein, is a highly inducible protein in the kidney following early ischemic and nephrotoxic injury and provides significant functional protection in animal models of AKI when exogenously administered. As it is quickly upregulated after ischemia-reperfusion injury and shed in urine, it has emerged as an important biomarker for early diagnosis of AKI, especially in clinical scenarios with ischemic insult such as cardiopulmonary bypass.

The tubule cell’s biologic response to acute injury is multifaceted and includes loss of cell polarity and brush borders, cell death, dedifferentiation of viable cells, proliferation, and restitution of a normal epithelium (Fig. 467-5). There is a rapid disruption of the apical actin cytoskeleton and redistribution of actin from the apical domain and microvilli into the cytoplasm resulting in loss of brush border membranes, which contribute to cast formation and obstruction. Acute injury also leads to the early disruption of at least 2 basolaterally polarized proteins, namely Na/K-ATPase and integrins. The adapter protein ankyrin normally tethers Na/K-ATPase to the spectrin-based cytoskeleton at the basolateral domain, providing the driving force for the normal reabsorption of salt and water by the proximal tubule. Injury leads to a reversible cytoplasmic accumulation of Na/K-ATPase, ankyrin, and spectrin in viable cells.

Postulated mechanisms leading to loss of Na/K-ATPase polarity include disruption of the cortical actin cytoskeleton and cleavage of spectrin by ischemia-induced activation of proteases such as calpain. The physiologic consequence is the impaired proximal tubular sodium reabsorption that is characteristic of intrinsic AKI. The alpha-1 integrins usually are polarized to the basal domain, where they mediate cell-substratum adhesions. Acute injury leads to a redistribution of integrins to the apical membrane, causing viable cells to detach from the basement membrane. In addition, there is abnormal adhesion between these exfoliated cells within the tubular lumen, mediated by an interaction between apical integrin and the Arg-Gly-Asp (RGD) motif of integrin receptors. Administration of synthetic RGD compounds attenuates tubular obstruction and renal impairment in animal models of ischemic AKI.

Tubular cell death occurs via several pathophysiologic mechanisms in AKI. Necrosis is characterized by loss of membrane integrity, cytoplasmic swelling, nuclear pyknosis, cellular fragmentation, and an inflammatory response. Apoptosis is characterized by cytoplasmic and nuclear shrinkage, DNA fragmentation, and breakdown of the cell into apoptotic bodies that are rapidly removed by phagocytosis. These 2 forms of cell death frequently coexist and are considered to be 2 extremes of a continuum. After kidney injury, the mode of tubule cell death depends primarily on the severity of the insult and the resistance of the cell type. Necrosis occurs after more severe injury and in the more susceptible proximal tubules, whereas apoptosis predominates after less severe injury and especially in the ischemia-resistant distal nephron segments. Apoptosis is an important mechanism of early tubule cell death in both animal and human models of AKI. Considerable attention has recently been directed toward unraveling the molecular pathways involved in renal tubule cell apoptosis (Fig. 467-6). Intrinsic, extrinsic, and regulatory factors all appear to be activated during AKI. Inhibition of apoptosis holds promise in AKI.

Mitochondrial injury is widespread in all types of AKI and typically precedes clinical onset. Autophagy pathways seem particularly important in development of AKI. Metabolically active proximal tubular cells are very rich in mitochondria. Mitochondria have a turnover cycle characterized by ongoing biogenesis and mitophagy with fusion and fission. AKI is thought to be an increased fission state; thus, mitophagy (autophagy of mitochondria) is needed to clear up the fragmented organelles when fission goes awry. In addition, injured mitochondria are a large source of reactive oxygen species. Blocking mitochondrial reactive oxygen species generation or mitochondria-targeted antioxidants might be a therapeutic target in AKI.

Autophagy, or “self-eating,” is a highly conserved, tightly regulated ontologic process in which the cell removes damaged organelles and subcellular structures. Damage-associated molecular patterns (DAMPs) are recognized by the innate immunity’s pattern recognition receptors such as Toll-like receptor 4 in the tubules (parenchyma and interstitium). Sterile inflammation is an important contributor to the pathophysiology of AKI. When autophagy is impaired, DAMPs persist in the local milieu, propagating damage. Pharmacologic inhibitors of autophagy (eg, chloroquine) predispose to AKI in experimental models. Autophagy seems to be beneficial in some models of AKI as inhibition of autophagy potentiates renal injury. Autophagy in the proximal tubule was demonstrated to be protective against cisplatin-induced AKI, and knockout mice that have autophagy inhibited are susceptible to cisplatin-induced AKI; it was partially reversible by treating with rapamycin (a known inducer of autophagy). However, the timing seems crucial, as prolonged autophagy is also detrimental.

Renal tubule cells possess a remarkable ability to regenerate and proliferate after acute injury. Morphologically, repair is heralded by the appearance of dedifferentiated tubule cells. In the next phase, the cells upregulate genes that encode for a variety of growth factors such as IGF-1 and undergo marked proliferation. In the final phase, cells undergo redifferentiation until the normal epithelium is restored. Thus, during recovery from injury, renal tubule cells recapitulate processes similar to those during normal kidney development. Understanding the molecular mechanisms of repair may provide clues toward accelerating recovery from AKI. The potential use of progenitor cells and stem cells as therapeutic strategies in AKI is currently in the initial stages of investigation. Controversy exists as to whether cells carrying out repairs were resident cells or migrated from elsewhere in response to chemical signals and growth factors. Current evidence suggests that segment restricted progenitor cells contribute to renal healing in adults, although unlike fetal cells, stem cells are not fully turned on in adults. Emerging data suggest that all differentiated proximal tubule cells possess the ability to dedifferentiate, proliferate, and redifferentiate. The kidney’s ability to repair itself is limited, and after several insults, maladaptive repair will lead to fibrosis and scarring. Certain cytokines and growth factors such as transforming growth factor-β play an important role in scarring because they stimulate growth factors, formation of fibronectin, and deposition of collagen.

Additionally, bone marrow–derived stem cells migrate to the kidney in response to various endothelium and matrix-derived adhesion signals and chemokines and become mesenchymal stem cells (MSCs). These cells do not undergo differentiation to populate the tubular epithelium but rather orchestrate the dedifferentiation, proliferation, differentiation, and organization of tubular epithelial cells through paracrine signals, producing anti-inflammatory cytokines and growth factors. One mechanism that might allow MSCs to modulate the behavior of cells in the vicinity is transfer of proteins and chemokines through microvesicles. MSC-derived extracellular microvesicles containing miRNA and mRNA have reduced AKI under experimental conditions.

Inflammation plays an important role in the pathogenesis of AKI. DAMPs such as high mobility group box 1 (HMGB-1), histones, heat shock protein, and fibronectin are recognized by pattern recognition receptors such as Toll-like receptors of the innate immunity system, leading to activation of macrophages and release of chemokines. The major components of this response include endothelial injury, leukocyte recruitment, and production of inflammatory mediators by tubule cells (Fig. 467-7). The endothelial injury is manifested by endothelial swelling, narrowing of the blood vessels, abnormal retrograde blood flow upon reperfusion, and the induction of adhesion molecules such as ICAM-1 and P-selectin that promote endothelial-leukocyte interactions. Although ablation of the ICAM-1 gene and pretreatment with ICAM-1 antibody rendered mice resistant to ischemic AKI, human trials with anti-ICAM-1 antibody did not prevent AKI in cadaveric transplant recipients.

The injured proximal tubular epithelium can generate several mediators that potentiate the inflammatory response. They include cytokines such as tumor necrosis factor-α, IL-6, IL-1, and chemotactic cytokines (MCP-1, IL-8, RANTES). Recent studies have shown that the plasma levels of the proinflammatory cytokines IL-6 and IL-8 predict mortality in adults with AKI. Resident renal macrophages can be both proinflammatory (M1) and reparative (M2). Macrophage colony-stimulating factor channels macrophages through to the M1 pathway, suggesting that cerebrospinal fluid might be playing a role in which macrophage phenotype will be expressed. Depleting dendritic cells protects against ischemic AKI but worsens cisplatin AKI. Strategies that modulate the inflammatory response may provide significant beneficial effects in human AKI.

Inflammation in the kidney could potentially have distant effects with spillover of chemokines and cytokines into the bloodstream or travel of activated immune cells elsewhere. Localized damage in AKI triggers systemic inflammation. Ischemic kidney releases numerous cytokines and chemokines, including endothelin-1 and HMGB-1, into the circulation. Uric acid is a prototypical alarmin, released into circulation after ischemia reperfusion, and it potentiates inflammation. Uric acid interacts with TLR-2 receptors forming Weibel-Palade bodies that release von Willebrand factor, IL-8, angiopoetin-2, eotaxin, and endothelin-1. Contents of the Weibel-Palade bodies either trigger distant inflammation or promote fibrosis. Once HMGB-1 is released into the circulation, it acts as a DAMP. Preventing translocation of HMGB-1 from nucleus to cytoplasm ameliorates AKI in animal models. Released ATP is also a DAMP, attracting phagocytes and activating inflammasomes such as NLRP3. Alkaline phosphatase has been shown to be protective in septic AKI animal models secondary to dephosphorylation and inactivation of lipopolysaccharide and ATP and regeneration of adenosine. Clinical trials exploring use of alkaline phosphatase in human septic AKI are ongoing. Extracellular adenosine is thought to be protective against oxidant injury in the PT cells by acting as an anti-inflammatory and tissue-protective signaling molecule perhaps through A(2A) receptors.

Postrenal AKI

Postrenal AKI is a result of bilateral obstruction of the outflow tracts and is uncommon beyond the neonatal period. Postrenal AKI is usually reversed by relief of the obstruction but is accompanied by a very significant postobstructive diuresis. The common causes of postrenal AKI are listed in Table 467-3.

CLINICAL FEATURES AND DIAGNOSTIC EVALUATION

AKI is largely asymptomatic in early stages and most commonly presents with a progressive accumulation of nitrogenous wastes and fluid overload once it is well established. The evaluation of patients requires a complete history, physical examination, laboratory testing, renal imaging, and, when indicated, a kidney biopsy. It is important to recognize that early in the course of AKI, changes in BUN and creatinine do not provide an accurate reflection of renal function. Despite the characteristically precipitous reduction in GFR that occurs in AKI, the BUN usually rises by only about 20 mg/dL per day and the serum creatinine by 1 to 2 mg/dL per day, until the values reach a steady state that more accurately represents the actual degree of renal dysfunction. This relationship is important from several perspectives: (1) it distinguishes AKI from CKD, as the BUN and creatinine in CKD do not progressively increase but display a steady high value; (2) if serial laboratory values are available, the timing of the original insult that leads to the development of AKI can be estimated; and (3) the BUN and serum creatinine are delayed and unreliable biomarkers of AKI. In the case of BUN, an increase can be encountered in the absence of AKI, especially in patients undergoing steroid therapy, total parenteral nutrition, excessive catabolism, and bleeding within the gastrointestinal tract. Impact of extrarenal conditions on variations in serum creatinine was discussed earlier. In addition, due to the normal maturation in GFR that occurs with age, adult values are not reached until 2 years of age in children. Maternal serum creatinine impacts measurement in neonates for the first week of life. Given the variations in body habitus, muscle mass, and fluid status that are typical in the pediatric population, a single serum creatinine threshold does not exist, rendering the reported serum creatinine normal range too wide to be clinically applicable. In addition, even in normal healthy subjects, baseline GFR is a highly individualized value, with many patients exhibiting values higher than the accepted normal of 100 to 120 mL/min/1.73 m², rendering comparisons to patients’ own baselines tricky when previous serum creatinine levels that predate acute illness are not available.

In patients with suspected AKI, the diagnostic evaluation is directed toward identifying the underlying cause, distinguishing between prerenal and intrinsic AKI, and discriminating between AKI and CKD.

Identifying the Underlying Cause

A detailed history and physical examination directed toward the salient aspects listed in Table 467-4 are the springboard for AKI workup. In particular, detailed medication history, including over-the-counter medications, should be carefully reviewed. A careful urinalysis will yield many clues to the etiology of AKI in most of cases and should always be performed. Typically, the urine in functional AKI contains only a few hyaline and fine granular casts, with little protein or blood. In contrast, proteinuria and hematuria are prominent in intrinsic AKI. Heme-positive urine in the absence of red blood cells (RBCs) in the sediment suggests hemolysis or rhabdomyolysis. Urine microscopy reveals dysmorphic RBCs in primary glomerular disease and reveals white blood cells (WBCs) in pyelonephritis or interstitial nephritis. The 2 latter conditions might also demonstrate white cell clumps and eosinophils in the urine, respectively. Broad brown granular casts are typically encountered in ischemic or toxic ATN, whereas RBC casts are characteristic of acute glomerulonephritis, and WBC casts indicate interstitial nephritis or pyelonephritis. A renal ultrasound, although not very specific, is nonetheless useful to rule out obstruction and also to evaluate the size and echotexture of the kidneys where the degree of disease chronicity could be in question.

Distinguishing Between Prerenal and Intrinsic Aki

Differentiating between functional AKI and intrinsic AKI has important clinical applications, as management of one is quite different from that of the other. Functional AKI is a sodium-avid state with maximal reabsorption of solutes and water by the intact proximal tubule; on the other hand, the tubular cell damage typical of intrinsic AKI results in impaired reabsorptive capacity of the proximal tubule. Urinary indices based on this principle are shown in Table 467-5. The fractional excretion of sodium (FENa) is reportedly the most accurate in making this distinction. It is calculated from measured concentrations of sodium (Na) and creatinine (Cr) in the urine (U) and plasma (P), as follows:

In the pediatric age group, functional AKI is typically characterized by an FENa < 1%, whereas it is > 3% in intrinsic AKI. In neonates, particularly premature babies, as urine concentration capacity is limited, a higher threshold of < 2.5% for prerenal is used. Interpretation of these indices must take into account any abnormal presence of substances in the urine such as protein, glucose, mannitol, contrast agents, and diuretic therapy. By the same principle of increased solute reabsorption, the BUN/creatinine ratio in the serum is markedly elevated (> 20) in prerenal AKI.

As most diuretics are natriuretics and change renal sodium excretion, calculation of FENa in patients treated with diuretics (such as a patient with congestive heart failure who is on chronic diuretic treatment) is unreliable. Fractional excretion of urea, calculated on the same principle, can be used in patients who have received diuretics. A value of 35% or less is considered supportive of prerenal AKI. NGAL is a promising biomarker that can differentiate between functional and intrinsic AKI.

Distinguishing Between AKI and CKD

A kidney and bladder ultrasound is a sensitive, noninvasive modality that can help differentiate between AKI and CKD and that can rule out a postrenal etiology. Typically, the kidneys in AKI are normal or enlarged, with increased echogenicity, whereas those in CKD are frequently small and shrunken. Other distinguishing features are shown in Table 467-5.

COMPLICATIONS

Common complications of AKI result from disruption of normal electrolyte and fluid homeostasis and are listed in Table 467-6. Hyponatremia is a common laboratory finding and usually is dilutional (secondary to fluid retention and administration of hypotonic fluids) but can be due to sodium depletion or associated hyperglycemia. Hypernatremia in AKI usually is a result of excessive sodium administration (inappropriate fluid therapy or overzealous sodium bicarbonate administration). Hyperkalemia is caused by the reduction in GFR and tubular secretion, increased catabolism, and metabolic acidosis (each 0.1-unit reduction in arterial pH raises serum potassium by 0.3 mEq/L). Hyperkalemia is most pronounced in patients with excessive endogenous production (rhabdomyolysis, hemolysis, and tumor lysis syndrome). Symptoms are nonspecific and may include malaise, nausea, and muscle weakness. Electrocardiogram changes should be looked for in all patients suspected of having hyperkalemia. A high anion gap metabolic acidosis is a common finding and is secondary to the impaired renal excretion of acid and the impaired reabsorption and regeneration of bicarbonate. Acidosis is most severe in shock, sepsis, or impaired respiratory compensation. Hypocalcemia in AKI is caused by increased serum phosphate and impaired renal conversion of vitamin D to its active form. Hypocalcemia is most pronounced in patients with rhabdomyolysis. Metabolic acidosis increases the fraction of ionized calcium. Therefore, overzealous administration of bicarbonate therapy can decrease the concentration of ionized calcium and precipitate symptoms of hypocalcemia, including tetany, seizures, and cardiac arrhythmias. Hyperphosphatemia is primarily caused by impaired renal excretion and can aggravate the hypocalcemia. During recovery from AKI, the vigorous diuretic phase may be accompanied by significant volume depletion, hypernatremia, hypokalemia, and hypophosphatemia.

The genesis of nonrenal complications is complex, and AKI per se is a major risk factor for their development. The accumulation of unidentified “uremic toxins” is postulated to play a major role in the cardiac, pulmonary, gastrointestinal, neuropsychiatric, and infectious disturbances. AKI is also clearly associated with an increase in mortality rate, irrespective of the underlying cause.

Inflammation in the kidney could potentially have distant effects with spillover of chemokines and cytokines into the bloodstream or travel of activated immune cells elsewhere. Animal models of AKI have improved our understanding of extrarenal complications and potential routes of organ cross-talk. Pulmonary capillary permeability is increased in AKI, and salt and water channels are dysregulated with decreased transport of fluid from the alveoli, leading to decreased pulmonary compliance and worsening oxygenation. Although sepsis is a common cause of AKI, it seems that the relationship is bidirectional, and patients with AKI are more likely to develop infectious complications. The exact reason behind this observation is not currently known, but AKI is known to interfere with T-cell trafficking and neutrophil function. Cardiorenal syndrome is a term coined to underscore the dynamic and reciprocal relationship between cardiac and renal dysfunction. While chronic venous congestion in CHF as well as impaired perfusion from systolic heart failure cause AKI, AKI also leads to decreased systolic performance. Blood-brain permeability is increased in AKI.

Fluid overload has come under the spotlight during the last decade, especially in niche populations such as children with multiple organ failure on RRT. Timely and adequate resuscitation is essential for adequate resolution of organ injury. Restrictive fluid resuscitation is associated with increased AKI in burn victims. However, there is a strong association between accumulation of excessive fluids and adverse outcomes in both pediatric and adult patients. This association seems to impact respiratory morbidity in particular. AKI is a frequent occurrence in patients with acute respiratory distress syndrome and increases duration of ventilation. In pediatric patients on extracorporeal membrane oxygenation, the degree of fluid overload at the start of RRT was associated with mortality; however, reversal of fluid overload did not abolish increased risk of mortality, suggesting that the relationships among fluid accumulation, AKI, and outcomes are complex. Whether fluid accumulation represents “occult” AKI caused by dilutional effects of excess body water on serum creatinine, a surrogate for higher severity of illness not captured by current index scores, or an independent risk factor is unknown, mostly due to the retrospective and observational nature of most of the data available. What is clear is that intravenous fluids are medications that require attention to dosage identical to all other medications.

Abdominal compartment syndrome is both a potential complication and cause of AKI. Increased intra-abdominal pressure, due to space-occupying lesions, trauma, diuretic refractory ascites, or overzealous fluid resuscitation with accompanying third-spacing characteristic of systemic inflammation, surpasses splanchnic venous pressure and leads to decreased perfusion pressure and organ dysfunction. The clinician should have a low index of suspicion to detect the condition in subtle cases. Physical examination typically will reveal a tense abdomen. Bladder-pressure measurement as a surrogate for intra-abdominal pressure is diagnostic, and treatment is relief of pressure usually by drainage of free fluid. In rare cases, decompressive laparotomy with delayed abdominal closure might be required.

TREATMENT

Fluids, Electrolytes, and Nutrition

Functional AKI associated with intravascular volume depletion requires prompt and vigorous fluid resuscitation with normal saline (20 mL/kg over 30–60 minutes, repeated 2–3 times if necessary). If onset of micturition with urine output does not occur in less than 4 hours, the bladder should be catheterized to confirm the anuria. In severe sepsis or septic shock, early protocolized treatment that facilitates timely fluid resuscitation and administration of antibiotics is beneficial and has been associated with decreased AKI. In functional AKI with total-body fluid accumulation and selective renal/splanchnic hypoperfusion, such as congestive heart failure, treatment for AKI involves decongestion with diuretics. Although hepatorenal syndrome also is an effective hypovolemia state, due to peculiar splanchnic vasodilation with maldistribution of circulating volume that accompanies the disorder, treatment is a bit more complex and involves removal of diuretics, agents that induce splanchnic vasoconstriction (such as terlipressin or norepinephrine), colloid loading, and careful reintroduction of diuretics. In all cases of AKI, potassium is contraindicated until urine flow is well established and the serum potassium begins to normalize. Once oliguric AKI sets in, conservative fluid management is indicated. Fluid intake should be limited to equal insensible water loss plus replacement of output. If patients are fluid-overloaded, then urine output should not be replaced, and diuretics are indicated. If urine flow cannot be achieved with diuretic challenge, RRT will be required.

AKI is a catabolic process, and children with AKI are at high risk of underfeeding due to a combination of fluid restriction and false beliefs about need to restrict protein intake in hopes of avoiding RRT. Aggressive nutritional support is crucial to enhance the recovery process. Adequate calories to account for maintenance requirements and supplemental calories to combat excessive catabolism should be administered by oral, enteral, or parenteral routes as appropriate. Intake of proteins should not be restricted. If adequate nutrition cannot be achieved because of fluid restriction, early institution of RRT should be strongly considered.

Medications

Nephrotoxic agents should be avoided in AKI, as they may worsen the injury and delay recovery of function. The dose of all medications should be adjusted based on residual renal function. Importantly, patients with AKI in the early phase of a rising creatinine should be assumed to have a GFR of less than 10 mL/min, regardless of the absolute value of the serum creatinine, because this value may not reflect actual renal function. NSAIDs have emerged as an important contributor to hospital-acquired AKI in children. In patients at risk, such as dehydrated patients with fever, infants younger than 6 months of age, or critically ill patients, especially if they are exposed to other nephrotoxic medications or have hemodynamic instability, NSAIDs are relatively contraindicated and should be avoided. In addition, chronic use of certain medications can lead to interstitial nephritis. In a single-center study from Canada exploring risk factors for AKI, the single greatest risk factor in ICU was use of nephrotoxic medications for development of ICU-acquired AKI in critically ill children. Fenoldopam, natriuretic peptides (eg, nesiritide), N-acetylcysteine, bicarbonate infusions, and saline loading outside of rehydration have not led to consistently positive results to prevent or treat AKI.

Renal Replacement Therapy

The common indications for acute RRT in AKI include (1) fluid overload that is unresponsive to diuretics or prevents adequate nutrition; (2) hyperkalemia that is unresponsive to nondialytic therapy; (3) refractory metabolic acidosis; (4) hypertension refractory to medical management, especially in the setting of oliguria; and (5) symptomatic uremia, including pericarditis, pleuritis, and neurologic symptoms. The choices available include hemodialysis, peritoneal dialysis, and continuous renal replacement therapy (CRRT). The choice of dialysis modality depends on the patient’s clinical status, the physician’s expertise, and the availability of appropriate resources. Extracorporeal therapies (hemodialysis and CRRT) require central vascular access, specialized equipment and technical personnel, anticoagulation, and the ability to tolerate a large extracorporeal volume. Critically ill patients frequently require blood pressure support with vasoactive amines for effective hemodialysis. Hemodialysis provides the most rapid correction of electrolyte derangements and acid-base status. However, due to the rapid fluid shifts that characterize a limited hemodialysis duration of 3 or 4 hours, it is poorly tolerated in patients with hemodynamic instability and diminished cardiac reserve. Therefore, CRRT has become the standard of care in resource-rich settings, allowing for gentle, continuous management of fluid overload. Accurate ultrafiltration and blood flow rates are crucial for pediatric CRRT because the extracorporeal circuit volume can comprise more than 15% of a child’s total blood volume. Small inaccuracies in ultrafiltration may represent a large percentage of the patient’s total body water. The advent of hemofiltration machines with volumetric control allowing for accurate ultrafiltration flows has led to an increased use of CRRT compared to peritoneal dialysis as the preferred modality for treating pediatric AKI, especially in the intensive care setting. Miniaturization of the extracorporeal circuits and incorporation of smart software will enable safe and effective CRRT in neonates and young infants.

Peritoneal dialysis is the other continuous modality and is frequently the preferred therapy for neonates and small infants. The advantages of peritoneal dialysis include ease of performance and no requirement for specialized equipment, personnel, or anticoagulation, rendering it an attractive and life-saving treatment to use especially in resource-limited settings. In critically ill children with respiratory failure on mechanical ventilation, dwell volume in the peritoneal cavity might be poorly tolerated, creating restrictive physiology. Certain conditions such as recent abdominal surgery with compromise of the peritoneum, intra-abdominal infection, and ostomy placement render the peritoneal route unsuitable for dialysis.

Timing of initiation of RRT has been the topic of constant debate during the past few years. Currently, there is no convincing evidence to support early initiation of RRT in the absence of the conventional indications in pediatric patients.

PREVENTION

Ensuring adequate resuscitation aimed at early reversal of shock, avoiding hypotension, and sparing use of nephrotoxins with careful therapeutic drug monitoring appear to be the most effective strategies in preventing AKI in critically ill patients. Vigorous fluid administration has been successfully employed to prevent AKI in patients at high risk, including those who have undergone cardiac surgery or renal transplantation; those with hemoglobinuria, myoglobinuria, and early tumor lysis syndrome; and those who received nephrotoxic agents such as radiocontrast, cisplatin, and amphotericin. Adjunctive pharmacologic agents for the prevention of AKI, including diuretics, have not been systematically tested. Administering diuretics early in the course of AKI to “test” the tubular response of the kidney and potentially convert the syndrome from an oliguric to a nonoliguric form has been proposed by some investigators; although it does not alter the natural history of the disease, it can simplify fluid, electrolyte, and nutritional therapy. One common and reasonably safe clinical approach is to use high-dose furosemide (2–5 mg/kg per dose bolus followed by a continuous drip) for oliguria of less than 48 hours in duration that has not responded to adequate hydration. Rasburicase, a recombinant urate oxidase, has become the standard of care to prevent AKI in pediatric tumor lysis syndrome. Aminophylline, a xanthine oxidase inhibitor, is a renal vasodilator that is proposed to offset adenosine-mediated afferent arteriole vasoconstriction through competitive antagonism. It has been studied for prevention of cardiac surgery–associated AKI. Although aminophylline use postoperatively did not decrease the incidence of AKI, intraoperative aminophylline treatment of oliguria has been associated with improved urine output and decreased need for RRT when compared with furosemide. In fact, a closely related agent, theophylline, has been recommended to treat perinatal asphyxia-induced AKI. Further studies in more heterogenous groups are required before use can be generalized. N-Acetylcysteine, dopamine, and sodium bicarbonate have not been found to be effective in preventing or treating AKI. Probenecid, by competitive binding to proximal tubule organic anion transporter, decreases the tubular uptake of certain nephrotoxins such as the antiviral cidofovir. Although beneficial in the experimental setting, spironolactone, atorvastatin, and other vasodilators have not impacted the incidence of AKI in clinical trials. Levosimendan, a calcium-sensitizing agent, might be protective against AKI in adult patients undergoing cardiac surgery. Remote ischemic preconditioning, defined as brief periods of induced ischemia (usually in a limb) prior to a known insult, has been used in cardiac surgery; although a meta-analysis has demonstrated protective effects in adults undergoing interventional cardiac procedures, use in neonates with single-ventricle physiology was not associated with decreased AKI.

PROGNOSIS: RECOVERY OR PROGRESSION

The clinical course of AKI has classically been divided into 3 phases: initiation, maintenance, and recovery. To this paradigm, the addition of an “extension” phase following the initiation phase has been proposed to reflect previously underestimated amplification processes. Recent advances in the pathogenesis of AKI allow us to postulate temporal relationships between the clinical phases and the cellular alterations. The initiation phase is the period during which initial exposure to the insult occurs, kidney function begins to fail, and parenchymal injury is developing. Intracellular ATP depletion occurs, sublethal injury to the tubule epithelial and endothelial cells predominates, generation of reactive oxygen molecules is initiated, and activation of inflammatory mechanisms commences. If the injury is alleviated at this stage, complete restitution and recovery are the rule. Prolongation of injury ushers in the extension phase. Blood flow returns to the cortex, and tubules undergo reperfusion-dependent cell death but also commence the regeneration process. In contrast, medullary blood flow remains severely reduced, resulting in more widespread tubule cell death, desquamation, and luminal obstruction. Injured endothelial and epithelial cells amplify the raging inflammatory cascades, and the endothelial denudation potentiates the intense vasoconstriction. The GFR continues to decline. This phase probably represents the most optimal window of opportunity for early diagnosis and active therapeutic intervention. This is followed by the maintenance phase, during which the parenchymal injury is established and the GFR is maintained at its nadir, even though renal blood flow begins to normalize. Both cell injury and regeneration occur simultaneously, and the balance between cell survival and death may determine the duration and severity of this phase. Repair of both epithelial and endothelial cells appears to be critical to overall recovery. Measures to accelerate the endogenous regeneration processes may be effective during this phase. The recovery phase is characterized functionally by improvement in GFR and structurally by reestablishment of tubule integrity, with fully differentiated and polarized epithelial cells. However, the repair process frequently is incomplete, and both microvascular and tubular dropout have been demonstrated in animal studies.

Children who appear to recover from AKI secondary to hemolytic-uremic syndrome exhibit a 10% to 25% rate of progression to CKD or end-stage renal disease. Long-term follow-up of premature infants with neonatal AKI has shown a 45% rate of renal insufficiency. In children who had an episode of AKI, 34% had either reduced kidney function or were dialysis-dependent upon hospital discharge. Long-term (3–5 years) follow-up of children who survived an episode of AKI found patient survival to be 57%, with the majority of mortality occurring within 2 years of the AKI episode. In addition, approximately 60% of the patients studied in a follow-up clinic visit demonstrated evidence of CKD. Pediatric patients who had AKI that completely resolved still had higher mortality rates than did patients who never had AKI. Among 1-year survivors of bone marrow transplant (BMT), CKD incidence was almost 30%, higher in those who had AKI in the first 3 months following BMT. In a report from Taiwan, 7% of patients discharged with AKI from a renal specialty hospital progressed to CKD or end-stage renal disease (ESRD). In adults, multicause AKI is associated with more CKD, likely due to the presence of comorbidities. In survivors of RRT, long-term quality of life was poor, with increased mortality rates and development of CKD and ESRD. Children who receive solid organ transplants are at increased risk of developing CKD and ESRD, particularly if they develop AKI in the perioperative period. Collectively, these data strongly suggest that long-term follow-up is warranted for children who survive an AKI episode. As the decline in renal function might not happen until after adolescents and young adults transition from pediatric to adult health care settings, strong collaborations between pediatric and adult primary care providers are essential to ensure the long-term adequate follow-up of at-risk individuals.

SUGGESTED READINGS

INTRODUCTION

Glomerular diseases present clinically in several different ways depending on the nature and severity of the primary disease and the extent to which the normal physiologic functions of the glomerulus are perturbed. Some children with glomerulopathies are found incidentally to have microscopic hematuria or proteinuria but are otherwise asymptomatic. At the other extreme, children may become critically ill with oligoanuric rapidly progressive glomerulopathy in need of urgent dialysis. Whereas numerous glomerular diseases are inherited (see Chapter 469), most forms are acquired and are generally considered to be immunologically mediated. There are 3 classical clinical syndromes that develop from glomerular injury: acute and chronic glomerulonephritis (GN), defined by the triad of hematuria, hypertension, and acute kidney injury (AKI); nephrotic syndrome (NS), defined by proteinuria and hypoalbuminemia; and hemolytic-uremic syndrome (HUS), defined by microangiopathic hemolytic anemia, thrombocytopenia, and AKI.

GLOMERULONEPHRITIS

The pathophysiologic sequence of events that lead to the development of the nephritic triad (hematuria, hypertension, and AKI) are shown in Figure 468-1. In each of the clinical entities with glomerular proliferation, inflammation leads to decreased glomerular perfusion, resulting in compromised kidney function and retention of salt and water with potential development of hypertension and edema.

APPROACH TO A CHILD WITH GLOMERULONEPHRITIS

The patient with glomerular disease presents clinically with a constellation of features that may include hematuria, proteinuria, edema, hypertension, and AKI. The urinary sediment is characterized as active when dysmorphic erythrocytes and cellular casts are present. A series of questions guides the initial diagnostic and management plan.

1. Is the process acute or chronic? Many patients with chronic GN appear relatively asymptomatic until the disease is advanced. Clues of chronicity include evidence of chronic kidney disease (CKD): significant anemia, renal osteodystrophy (abnormal bone radiographs or an elevated intact parathyroid hormone level), or small echogenic kidneys on ultrasound examination. Acute onset of severe hypertension often causes neurologic symptoms such as headaches and seizures, whereas long-standing hypertension of insidious onset may be clinically silent, but left ventricular hypertrophy may be present.

2. Is the kidney disease isolated, or are additional organ systems involved? A careful review of the systems and physical examination will help determine whether the investigation should move in the direction of primary (acquired) GN or toward secondary GN due to multisystem disease (Table 468-1). Relevant extrarenal involvement may be clinically silent. For example, postinfectious serologies (such as a streptozyme or anti–hepatitis B or anti–hepatitis C antibodies) may be indicated if infection-associated GN is a possibility. When patients present with a small-vessel vasculitis, involvement of the lung parenchyma or sinuses usually requires radiologic confirmation.

3. Is there hypocomplementemia? A low level of complement component C3 in the serum generally indicates 1 of 5 diseases in children: acute postinfectious GN, lupus nephritis (LN), C3 glomerulopathy (C3G), atypical HUS, or GN associated with chronic infections (subacute bacterial endocarditis or shunt nephritis), as indicated in Figure 468-2.

4. Is there recurrent painless macroscopic hematuria? If yes, immunoglobulin (Ig) A nephropathy is a likely diagnosis, although it also is a fairly common presentation of Alport syndrome during the first decade of life.

5. What is the age of the patient? Although most glomerular diseases can occur in almost any age group, some have a characteristic age range for onset of disease (Table 468-2). GN in the newborn period is extremely rare and often is the result of a congenital infection.

6. Is the onset rapidly progressive? Rapidly progressive glomerulonephritis (RPGN) is suggestive of a crescentic GN. More rarely, AKI may result from acute GN with superimposed acute tubular necrosis. RPGN is an emergency that needs urgent histologic diagnosis. Most of these patients have aggressive diseases that are successfully treated only with immunosuppressive therapy initiated early and in high doses. In patients with anti–glomerular basement membrane (GBM) nephritis or severe vasculitis, early plasmapheresis may be lifesaving.

7. Is NS present? Although proteinuria is a hallmark of many glomerular diseases, NS is suggestive of a more high-risk glomerular histopathology. In a patient presenting clinically with acute GN and hypocomplementemia, the presence of NS increases the concern for membranoproliferative GN, C3 glomerulopathy, or lupus nephritis. NS occurs less commonly in acute poststreptococcal glomerulonephritis (APSGN) and rarely in GN associated with chronic infections.

8. Is a kidney biopsy needed now? This procedure is generally not necessary when the diagnosis is obvious and the disease is self-limited without specific therapy (ie, typical APSGN and mild nephritis in patients with Henoch-Schönlein purpura), or for isolated steroid-sensitive NS. However, for most glomerular diseases, a definitive diagnosis can be made only by kidney biopsy. Moreover, for many patients, the most rational and evidence-based therapies are based on a histopathologic rather than a clinical diagnosis. Based on the severity and reversibility of the damage to the kidney, a kidney biopsy often predicts response to therapy and prognosis.

ACUTE POSTSTREPTOCOCCAL GLOMERULONEPHRITIS (APSGN)

APSGN is the most frequent and best-characterized acute postinfectious glomerulonephritis (PIGN). However, many other bacterial, viral, and parasitic pathogens may also induce acute PIGN. Identifying a specific causative pathogen often is difficult because the infection usually precedes the nephritis by a few weeks. A history of pharyngitis or pyoderma suggests a previous infection by group A β-hemolytic streptococcal species that may be confirmed by serologic testing. APSGN accounts for 80% to 90% of PIGN cases and is used as the prototype for this group of disorders. In approximately 10% of cases, PIGN follows other antecedent diseases caused by a variety of infectious agents as summarized in Table 468-3.

EPIDEMIOLOGY

APSGN is primarily a disease of school-age children and is more common in boys. Although reported in an 8-month-old child, the disease rarely occurs in children younger than 3 years of age and may occur as a sporadic or epidemic disease. APSGN follows infection with specific “nephritogenic” strains of group A β-hemolytic Streptococcus. APSGN has been shown to be nephritogenic following pharyngitis (strains 1, 3, 4, 12, 18, 25, and 49) or impetigo (strains 2, 49, 55, 57, and 60). It is extremely rare for APSGN and acute rheumatic fever (associated with M strains 1, 3, and 12) to occur simultaneously in the same patient.

Antibiotic treatment of the prodromal disease does not prevent acute GN, but treatment is important as a public health measure to prevent the spread of the nephritogenic bacteria. A study from Macedonia reported that 23% of sibling contacts had abnormal urinalyses and 9% developed clinical features of nephritis. Although difficult to determine with certainty, the overall risk of developing APSGN after infection with a nephritogenic strain is in the range of 10% to 15%. Because 50% to 85% of patients with APSGN are asymptomatic, the true incidence is difficult to determine. Rates of APSGN have decreased in the United States and Europe during the past 2 decades, likely due to earlier recognition and treatment of streptococcal infections, preventing spread of infection. The prevalence of certain nephritogenic stains is decreasing as well. A seasonal pattern exists for APSGN in North America that mirrors that of the pathogenic organisms: pharyngitis in the winter and spring and impetigo in the summer and fall.

PATHOPHYSIOLOGY

The precise nature of the antigen–antibody complex that causes nephritis remains unclear. The latency period between the acute infection and the onset of nephritis represents the time required to generate sufficient IgG antistreptococcal antibodies to trigger immune complex formation. Although several streptococcal antigens have been identified in the glomerular immune deposits, 2 proteins are of particular interest, based on currently available evidence. First is the cationic cysteine proteinase exotoxin B (SpeB) that co-localizes with complement and IgG within glomerular subepithelial immune deposits. The presence of circulating anti-SpeB antibodies is strong evidence of a recent nephritogenic streptococcal infection. Second is the nephritis-associated plasmin receptor, a protein that does not co-localize with glomerular immune deposits but is associated with increased intraglomerular plasmin activity and is thought to facilitate immune complex formation and inflammation.

Additional antigens are of interest and suggest that a group of streptococcal proteins may be involved in the initiation and progression of glomerular injury. It is currently thought that the target streptococcal antigen is initially trapped within glomeruli, with subsequent immune complex formation occurring in situ in the kidney. Once glomerular immune deposits are formed, the alternative and lectin complement pathways are activated, followed by neutrophil infiltration and glomerular damage. APSGN is an “exudative” GN characterized by the presence of many intraglomerular neutrophils. Pyuria and even white cell casts may be observed in the urinary sediment.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

Clinical symptoms begin abruptly. Patients usually are afebrile with a latency period of 1 to 2 weeks after having pharyngitis and 3 to 6 weeks after having a skin infection. The most common presenting features in symptomatic patients are edema (85%) and gross hematuria (30–50%). Almost all the patients have microscopic hematuria. The urine often has a unique cola or tea color. Hypertension is common (60–80%) but usually is mild to moderate; rarely, hypertensive encephalopathy and posterior reversible encephalopathy syndrome (PRES) can develop, even without a significant elevation in the serum creatinine concentration. Hypertension is not entirely explained by the degree of intracellular volume expansion. Echocardiographic evidence of left ventricular dysfunction is a common finding in children hospitalized with APSGN. The degree of AKI usually is mild. Rapidly progressive GN can occur, but only rarely. Although many patients have significant proteinuria and a slightly depressed serum albumin level (at least in part due to intravascular volume expansion), fewer than 5% of symptomatic patients develop frank NS. Severe complications, including pulmonary hemorrhage and cerebrovascular accidents, have been reported and can require kidney biopsy to distinguish APSGN from systemic vasculitis or Goodpasture syndrome. Occasionally, patients may have significant extrarenal manifestations suggestive of Henoch-Schönlein purpura. Spontaneous improvement typically begins within 1 week, with resolution of the edema occurring in 5 to 10 days and resolution of hypertension occurring in 2 to 3 weeks; however, the urinalysis may be abnormal for 1 to 2 years or, rarely, longer.

DIAGNOSTIC EVALUATION

The serum C3 level usually is below 50% of normal levels in patients with APSGN. However, as many as 10% of APSGN patients have a normal C3 level at the time of assessment. In patients with other forms of acute PIGN, this percentage is even higher. Decreased serum C4 levels are unusual, as consumption of C3 is due primarily to activation of the alternative pathway. A significant and persistent decrease in both C3 and C4 levels should suggest an alternative diagnosis, such as systemic lupus erythematosus, bacterial endocarditis, shunt nephritis, or membranoproliferative GN (see Fig. 468-2). The serum C3 level typically returns to normal within 8 to 12 weeks after APSGN. Urinalysis generally shows hematuria, proteinuria, and cellular casts (both red and white cells); rarely, patients may have a normal urinalysis.

Confirmation of a recent streptococcal infection renders the diagnosis of APSGN most likely, with a few caveats. A positive throat culture is insufficient to confirm the diagnosis, as 20% of normal schoolchildren are carriers who will have positive throat cultures without disease. An antibody to the streptolysin O enzyme (ASO) is detected in 80% of children with antecedent pharyngitis, but the test is also positive in 16% to 18% of healthy children. After skin infections, a positive ASO is less frequent (50%), possibly because of the ASO enzyme binding to lipids in the skin. The streptozyme assay combines serologic testing to 5 different streptococcal antigens (streptolysin O, streptokinase, hyaluronidase, DNase B, and NADase) and increases the likelihood of a positive test to greater than 95% in patients with pharyngitis and to 80% in patients with skin infections. The DNase B test also is quite reliable to establish a prior streptococcal infection. In the majority of patients, the clinical presentation and laboratory tests render the diagnosis of APSGN quite evident. There is usually no indication to perform a kidney biopsy to confirm the diagnosis in such children. However, for children with persistently low C3 levels or those who fail to recover along the typical timeline (Fig. 468-3), kidney biopsy may reveal a predominance of C3 deposits suggestive of C3 glomerulopathy.

TREATMENT AND COMPLICATIONS

APSGN is an acute disease that resolves without specific medical therapy. Close monitoring and supportive care are essential to manage the acute nephritis syndrome until the glomerular injury spontaneously resolves. If not previously administered, antibiotics should be given to prevent the spread of the nephritogenic strain of Streptococcus to other individuals. Prophylactic antibiotics do not prevent future recurrences (an extremely rare event) and are thus not recommended. During the acute phase, patients may need to be hospitalized for observation and treatment of hypertension, edema, oliguria, elevated serum creatinine, or electrolyte abnormalities. Fluid-overloaded patients often respond to loop diuretics. Urine output increases spontaneously within 5 to 10 days; severe oliguria lasting beyond 2 weeks is extremely rare and is cause to question the diagnosis. Hypertension usually is not severe and can be managed with short-acting calcium channel blockers.

In patients with more severe AKI, hyperkalemia, hyperphosphatemia, and acidosis are likely to occur and will require medical management. These patients rarely need dialysis. Corticosteroids and cytotoxic agents have no substantiated therapeutic role, although there is limited anecdotal evidence of benefit for patients with rapidly progressive GN associated with crescents on biopsy. This is such a rare presentation that performing prospective clinical trials to evaluate the benefit of such therapies is impossible.

PROGNOSIS AND OUTCOMES

Spontaneous improvement should begin within 1 week, with the gross hematuria, oliguria, azotemia, and hypertension generally resolving within 4 weeks; low C3 within 2 months; proteinuria by 6 months; and microscopic hematuria by 2 years following diagnosis (see Fig. 468-3). APSGN in children has been considered a completely reversible disease, which is still generally true. In the exceptional patient with crescentic disease, prolonged oliguria, and massive proteinuria, recovery may not be complete. Currently, no data are available on the evaluation of outcomes 2 to 3 decades after uncomplicated childhood APSGN. In the meantime, it seems reasonable to anticipate a good long-term outcome in children with this disease, although those hospitalized with severe acute disease deserve long-term follow-up care, as they may not have a completely benign long-term outcome. Recurrent APSGN is extremely rare but has been reported.

GLOMERULONEPHRITIS WITH CHRONIC INFECTIONS

Immune responses may also induce GN in the face of certain chronic persistent infections. In pediatric patients, the most important examples are infective endocarditis, shunt nephritis, hepatitis B and C, and human immunodeficiency virus (HIV), as well as congenital infections with cytomegalovirus or toxoplasmosis that may present clinically as infantile NS. In endemic tropical areas of the world, chronic infection with Plasmodium malariae, Schistosoma mansoni, and filariasis may cause more chronic forms of GN.

INFECTIVE ENDOCARDITIS

Subacute bacterial endocarditis (usually caused by Streptococcus viridans) is less common with the decline in rheumatic fever, but acute bacterial endocarditis is now more common, especially among intravenous drug abusers. Staphylococcus aureus is the most common pathogen, but several other organisms are occasionally implicated. Both serum C3 and C4 levels are depressed in 90% of patients. A subset of patients may develop circulating anti-neutrophil cytoplasmic antibodies (ANCA). Histologically, the disease resembles APSGN, even in the small ANCA-positive subset. With appropriate antibiotic therapy, the glomerulopathy improves, but chronic kidney injury may persist. Normalization of the serum complement levels is a good prognostic indicator. The possibility of other kidney lesions being present, especially drug-induced interstitial nephritis and septic emboli, needs to be considered in this population.

SHUNT NEPHRITIS

Shunt nephritis was first recognized in children with ventriculoatrial shunts for hydrocephalus. The risk of development of nephritis is reportedly 4% among patients with infected shunts. This complication is extremely rare in patients with infected ventriculoperitoneal shunts. A similar form of nephritis occurs with infected vascular access grafts. Coagulase-negative Staphylococcus is the most common organism. Blood cultures may be negative, but the organism often is cultured from the graft itself. The kidney disease frequently has an indolent presentation: gross hematuria occurs frequently, and 25% to 30% of patients may develop NS. Serum C3 and C4 levels usually are depressed (90%). Histologically, the lesion is a diffuse proliferative GN that resembles membranoproliferative GN (MPGN). Treatment includes antibiotic therapy and removal of the infected shunt. The glomerular disease slowly resolves, but residual CKD is a frequent occurrence.

CHRONIC HEPATITIS

An association exists between chronic hepatitis B (HBV) and membranous nephropathy. Less common kidney manifestations associated with chronic HBV infection are MPGN, crescentic GN, and kidney disease related to polyarteritis nodosa (PAN). Introducing the hepatitis B vaccination during infancy has decreased the incidence of HBV-associated kidney disease. Children present clinically with NS or with nonnephrotic proteinuria during the chronic carrier stage. Depressed serum C3 and C4 levels are not characteristic but may occur. In children, most kidney complications remit spontaneously within 5 to 7 years, concomitant with the disappearance of HBe antigenemia and the appearance of HBe antibodies. The KDIGO (Kidney Disease: Improving Global Outcomes) guidelines recommend antiviral therapy for HBV infection and membranous nephropathy with interferon alpha or nucleoside analogs for persistent kidney disease. Because these agents are nephrotoxic, kidney function must be followed closely during therapy. Immunosuppressive therapy is contraindicated due to an increased risk of developing chronic active hepatitis, although corticosteroids have been used in a select subset of patients with severe vasculitis or crescentic GN.

In adults, chronic active infection with hepatitis C virus also can cause MPGN, especially in association with cryoglobulinemia, membranous nephropathy, and IgA nephropathy. Unlike in adults, MPGN has been associated with viral infections in only a very limited number of children. KDIGO guidelines suggest combined antiviral treatment for hepatitis C virus–associated MPGN using PEGylated interferon and ribavirin, titrated according to the patient’s tolerance and level of kidney function.

HIV-ASSOCIATED NEPHROPATHY

Prior to the development of highly active anti-retroviral therapy (HAART), approximately 2% to 10% of patients infected with HIV developed nephropathy. Most patients present with proteinuria, usually with NS. It may be the first manifestation of HIV infection, or it may occur in patients with acquired immunodeficiency syndrome (AIDS) or AIDS-related complex. A unique variant of focal segmental glomerulosclerosis (FSGS) is found in 60% of patients, the vast majority of whom are of African descent with advanced HIV disease. In addition to the typical FSGS lesion, “collapsed” glomeruli (also seen in a small subset of patients with idiopathic FSGS), degenerated and microcystic tubules, interstitial edema and inflammation, and glomerular endothelial tubuloreticular inclusions (reminiscent of those associated with lupus nephritis) are common features. Carrying the risk allele for the apolipoprotein L1 (APOL1) gene increases the chance for developing HIV-associated nephropathy among African Americans. HAART slows the rate of progression and may even prevent the development of nephropathy. Anecdotal case reports suggest a beneficial outcome in patients treated with steroids or cyclosporine. Because opportunistic infections are common occurrences, the relative risks and merits of these more aggressive treatments need to be carefully considered. Nephrotoxicity due to urinary crystal formation or tubular damage may occur with the use of certain antiretroviral drugs. In addition to the “collapsing” variant of FSGS, HIV-infected patients have an increased risk of developing other glomerulopathies, including IgA nephropathy, thrombotic microangiopathy, PIGN, MPGN, and immune complex GN. HAART has been less effective for treating these forms of kidney diseases in HIV-infected individuals. Hepatitis C co-infection and drug-induced nephrotoxicity are also common occurrences. Kidney transplantation, once contraindicated in patients with HIV, is now the most effective renal replacement therapy, provided rigorous criteria are met.

MEMBRANOPROLIFERATIVE GLOMERULONEPHRITIS (MPGN)

MPGN, less commonly called mesangiocapillary GN, is not a distinct disease. It is a biopsy diagnosis characterized morphologically by (1) thickening of the GBM caused by immune complex deposition or interposition of mesangial cell cytoplasm in the GBM and (2) hypercellularity caused by proliferation of mesangial cells and the influx of leukocytes, which produces a typical lobular appearance of the glomerular tuft. MPGN is frequently idiopathic in children but may be secondary to a variety of other entities, as shown in Figure 468-4.

The idiopathic varieties are classified as 2 distinct and unrelated diseases based on kidney biopsy findings. Immunoglobulin-mediated MPGN (Ig-MPGN, formerly MPGN type I) is an immune complex disease characterized by a predominance of IgG deposition along the subendothelial space of the GBM. In contrast, C3 glomerulopathy (C3G) is a disease of dysregulated complement activation characterized by predominantly C3 deposits in the glomerulus. C3G can be further subdivided into dense deposit disease (DDD, formerly type II MPGN) and GN with dominant C3 (C3GN) based on abnormalities on electron microscopy.

EPIDEMIOLOGY

MPGN usually presents between the ages of 8 and 30 years; it rarely occurs in patients younger than the age of 5 years (although cases as young as 15 months have been reported) and mainly affects Caucasians. Although uncommon, familial clusters are increasingly recognized and associated with inherited complement gene variants. Overall, male and female patients are equally affected, but the condition is slightly more common in female adolescents. The incidence of idiopathic Ig-MPGN appears to have decreased during the past 2 decades. In adults, this may be explained by the recognition that many patients previously classified as having Ig-MPGN actually are chronically infected with hepatitis C. The reason for the declining pediatric incidence of idiopathic MPGN is unclear.

PATHOPHYSIOLOGY

Ig-MPGN is an immune complex–mediated glomerular disease. The close association with hepatitis C infection in some adult populations and the declining overall incidence of this disease suggest that exogenous antigens trigger the disease. C3G is caused by abnormalities in the regulation of the alternate pathway of complement activation, resulting in predominant deposition of complement C3 fragments within the glomerulus and subsequent injury. Complement factor H (CFH) and CFH-related gene variants associate with disease, as CFH is the major regulator of the alternative pathway of complement. In C3G, the abnormal CFH-related proteins interfere with CFH regulation of complement activation predominantly along the GBM, which results in abnormal glomerular C3 deposition in glomeruli with little to no effect on plasma C3 regulation.

DDD is characterized by the presence of dense ribbon-like deposits within the GBM, as seen by electron microscopy. The nature of this dense material remains unknown, but it appears to be biochemically similar to GBM itself. This material appears to activate the alternative complement cascade, with C3 characteristically lining the outer aspect of these deposits. Immunoglobulins usually are absent. C3 nephritic factor (C3NeF) is an autoantibody that causes unregulated activation of the alternative complement pathway. C3NeF may be present in all types of MPGN but is most common in DDD.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

Twenty to 30% of children with MPGN/C3G present with an acute nephritic syndrome that initially mimics APSGN. In fact, many patients have had a preceding upper respiratory illness, and as many as 40% have evidence of having had a recent streptococcal infection. The presence of NS and a depressed serum C4 level are early clues that differentiate MPGN from APSGN. Another 20% to 40% of children are diagnosed following an incidental finding of proteinuria and hematuria. NS is a presenting feature in 30% to 50% of children, and even more children will develop the syndrome during the course of the disease. Much rarer presentations include recurrent episodes of gross hematuria that mimic IgA nephropathy and rapidly progressive GN. Subtle, but not substantial, differences exist between the clinical features at presentation of Ig-MPGN, C3GN, and DDD. Some patients with DDD have partial lipodystrophy (Barraquer-Simons disease).

DIAGNOSTIC EVALUATION

Although a kidney biopsy is required to establish the diagnosis of MPGN/C3G, persistent hypocomplementemia is highly suggestive of the diagnosis. Patients with Ig-MPGN have evidence of activation of the classical complement cascade with low serum C3 levels and borderline or low C4 levels. The complement profile in patients with C3G suggests activation of the alternative pathway with low C3 and normal C4 levels. The presence of C3NeF is most characteristic of DDD. However, the serum C3 level is normal in 25% of all patients with MPGN/C3G at the time of initial presentation. The levels are variable over time and appear not to have prognostic significance.

TREATMENT AND COMPLICATIONS

Optimal therapy remains uncertain. Clinical trials have shown no clear benefit of any treatment, but the duration of many trials has been relatively short, which is significant because MPGN/C3G usually is a chronic, slowly progressive disease process, so the benefit of treatment may not be evident unless given for years. For many children without NS, kidney function may remain stable for many years without specific therapy. The consensus in North America is that children with Ig-MPGN and significant proteinuria will do better if treated with alternate-day steroids, based on data from a cohort of children in Cincinnati and the International Study of Kidney Disease in Children (ISKDC). A variety of different protocols have been tried, but most involve a dose of 2 mg/kg every other day (maximum 60 mg) for at least 1 year, with various tapering schedules thereafter. Hypertension is a frequent complication that requires treatment. The role of antiplatelet therapy with aspirin or dipyridamole has been evaluated and is generally not recommended. Limited and conflicting data on the use of cytotoxic drugs and complement inhibition preclude recommendation.

There is limited but emerging evidence to support the use of immunosuppression or complement inhibition for C3GN or DDD. Angiotensin-converting enzyme inhibitors (ACEi) and angiotensin II receptor blockers (ARB) are recommended for proteinuria and hypertension control. Plasma exchange therapy may have a role in the patient subgroups with inherited CFH deficiency and possibly those with significant glomerulopathy associated with C3NeF. In an uncontrolled trial of pediatric patients, treatment with the anti-C5 antibody eculizumab was associated with reduction in proteinuria and increase in kidney function, but many children at other centers were nonresponders. The increasing number of small case series reporting beneficial outcomes in patients with severe C3G treated with rituximab highlights the need for a well-designed clinical trial.

PROGNOSIS AND OUTCOMES

Limited data are available on the long-term outcome of children with MPGN/C3G. For many patients, the disease carries a poor prognosis, particularly if associated with NS. Overall, 50% to 60% of untreated patients develop end-stage renal disease within 10 to 15 years; fewer than 10% undergo spontaneous remission. The prognosis is worse in patients with persistent NS and significant crescent formation and interstitial fibrosis on kidney biopsy. The outcome may have improved in recent years, but because of the indolent course of this disease, long-term follow-up will be necessary to confirm this impression. Following kidney transplantation, recurrent disease is common and can lead to graft failure.

Treatment response and prognosis of patients with C3GN appear to be worse than those of patients with Ig-MPGN.

LUPUS NEPHRITIS

Systemic lupus erythematosus (SLE) is considered the prototype of multisystem diseases that cause immune complex–mediated tissue injury to the kidney. At diagnosis, a finding of a completely normal kidney without IgG and C3 deposits is extremely rare, although kidney biopsies are performed generally only in patients with clinical or laboratory evidence of nephropathy, which includes 40% to 60% of children at the time of diagnosis and another 10% to 20% within 5 years of diagnosis.

PATHOPHYSIOLOGY

The mechanism of tissue injury is the glomerular deposition or in situ formation of immune complexes. Most tissue deposits contain IgG and C3, but the presence of IgM or IgA is also common. The presence of circulating autoantibodies has been shown to predate (by an average of 3 years) the onset of clinical symptoms. Autoantibodies to double-stranded DNA, nucleosomes, and α-actinin are most closely associated with GN in patients with SLE. Nucleosomes (chromatin fragments with a histone core) that are released from apoptotic cells have been of particular interest as antigenic triggers. Through the activation of complement, the glomerular immune deposits are thought to trigger an inflammatory response that causes kidney damage.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

Symptoms of lupus nephritis span the spectrum of glomerular injury, ranging from asymptomatic microscopic hematuria or proteinuria to rapidly progressive GN. AKI is a rare but recognized presentation. Usually other systemic manifestations of SLE are apparent, and a kidney biopsy is almost never required to establish the diagnosis of this disorder. The presence of 4 of 11 American College of Rheumatology classification criteria for SLE confers a 96% sensitivity and specificity for the diagnosis of SLE and has been validated in children. Although the diagnosis of SLE usually is established by clinical criteria, a kidney biopsy is needed to determine the histologic pattern of renal injury, which guides initial therapy.

Several distinct patterns of histopathology occur at comparable rates in both childhood and adult-onset SLE. The revised classification of the International Society of Nephrology/Renal Pathology Society (ISN/RPS) defines 6 classes of lupus nephritis that are further subdivided based on the degree of disease activity (indicating the potential for therapeutic responsiveness) and chronicity (nonreversible damage). Class I is characterized by mesangial immune deposits with normal-appearing glomeruli. Class II lesions include mild mesangial matrix expansion and/or mesangial hypercellularity and predominantly mesangial immune complex deposits. The proliferative forms of lupus nephritis are the most common and most severe. Class III is focal proliferative nephritis (15–24% of the cases), with involvement in fewer than 50% of the glomeruli. Class IV, also called diffuse proliferative lupus nephritis (DPLN), has involvement of greater than 50% of the glomeruli and unfortunately accounts for 40% to 50% of the patients with lupus nephritis. In class III or IV nephritis, immune deposits occur in the mesangium and along the capillary wall, mainly in a subendothelial location. Class V is also called membranous lupus nephritis (10–20% of cases) and involves immune complex deposition in a subepithelial location. Combinations of proliferative and membranous lesions are possible (mixed class). Class VI is an irreversible advanced disease in which more than 90% of the glomeruli are destroyed by sclerosis.

Clinically, patients with class I or II usually have mild hematuria with or without low-grade proteinuria. Patients with class V usually present with proteinuria, often in the nephrotic range. As many as half the patients have microscopic hematuria, but cellular casts are not seen in the urine sediment. Although typically patients with the most severe clinical disease will have DPLN, the same histologic pattern may be observed in patients with relatively mild clinical features. The presence of isolated NS is most likely to be associated with class V disease, whereas nephritis and NS together is most commonly caused by DPLN. However, either can be seen in patients with class III or even class II nephritis.

DIAGNOSTIC EVALUATION

Almost all patients have positive antinuclear antibodies (ANA), although these autoantibodies can be detected in individuals who do not have SLE (especially in a speckled pattern). True ANA-negative SLE is a rare occurrence in children. Hypocomplementemia is another important finding in patients with SLE and active disease. Anti–double-stranded DNA titers and serum C3 levels are useful parameters for monitoring the activity of the kidney disease and response to therapy. No combination of these serologic studies will effectively predict the histologic type or severity of kidney involvement.

TREATMENT AND COMPLICATIONS

SLE nephritis may be controlled with medical therapy, but currently there is no known cure. The treatment of patients with classes I and II should be determined by the extrarenal manifestations. This is a mild glomerular lesion that has an excellent prognosis. However, for many patients, the class and severity of the nephritis guide the approach to immunosuppression.

DPLN is an aggressive disease that will ultimately destroy the kidney if left untreated. Consensus treatment recommendations developed for children, supported by several studies, indicate that the combined use of corticosteroids with immunosuppressive drugs results in a better long-term outcome than does treatment with corticosteroids alone. DPLN treatment is now divided into 2 phases: induction therapy to achieve remission and maintenance therapy to prevent relapses. If focal necrotizing lesions or crescents are present, patients with class III should be treated similarly. Induction includes corticosteroids initiated at high doses and tapered to low doses (often in the range of 0.1–0.5 mg/kg/d) over several months. Intravenous pulse therapy (30 mg/kg; maximum 1000 mg/dose) may be included, especially if the initial kidney disease is severe, as there is a delayed onset of action of the cytotoxic drugs. Six doses of intravenous cyclophosphamide (0.5–1.0 g/m²) administered monthly or twice-daily oral mycophenolate mofetil (MMF) is used in addition to corticosteroids.

The maintenance phase of therapy includes low-dose corticosteroids, discontinuation of cyclophosphamide (if used for induction therapy), and addition/continuation of MMF or azathioprine. Hydroxychloroquine is first-line therapy for any patient with SLE unless there is a contraindication, as its use is proven to prevent disease flares and to prolong survival. The optimal duration of maintenance therapy is uncertain, but 2 years after achieving remission without evidence of a relapse appears to be the minimum. During this period, serologic studies are monitored. Rising anti-dsDNA titers and falling serum complement levels are warning signs of an impending flare. Even with current treatment plans, only 80% of patients achieve remission, and 35% to 50% will experience at least 1 relapse.

Evidence is lacking regarding the best treatment approach for class V nephritis. Patients with a pure membranous lesion have a fairly good renal prognosis and often are managed like idiopathic membranous nephropathy with low-dose prednisone and MMF or cyclosporine A. In patients with superimposed proliferative lesions (mixed class), treatment depends on the severity of the proliferative lesions because they infer a higher risk for developing progressive CKD.

Other therapy has been used in patients with severe lupus nephritis, especially in the group that does not achieve an initial remission and in those who relapse early. There is insufficient evidence to recommend their use at this time. Although large clinical trials have failed to show efficacy of rituximab as an add-on therapy, there is evidence from large case series in adults and children that B-cell–depleting agents should be a first consideration for refractory lupus nephritis. Monoclonal antibody therapy targeting B-cell activating factor (belimumab) is approved for only adults with nonrenal SLE, but is now being studied in adults with lupus nephritis and children with nonrenal SLE. Intravenous immunoglobulin (IVIg) is beneficial for severe disease associated with hypogammaglobinemia. Plasmapheresis has not been shown to alter kidney outcomes, and its use in SLE should be limited to the rare subset of patients with thrombotic microangiopathy associated with antiphospholipid antibodies.

PROGNOSIS AND OUTCOMES

End-stage renal disease (ESRD) remains a problem in patients with DPLN, especially in those with advanced kidney scarring at the time of initial diagnosis and in noncompliant individuals. A small minority of patients have aggressive disease that is difficult to control with current treatment strategies. The incidence of ESRD in patients with SLE and DPLN has decreased from 50% in the 1950s to 5% to 10% today. Following kidney transplantation, deposits occasionally recur in the allograft, but they do not lead to graft failure. Overall patient survival is now in the range of 80% at 15-year follow-up. Several factors predict a worse prognosis and include ethnicity (African American, Hispanic), gender (male), failure to achieve disease remission, kidney disease severity (elevated creatinine, hypertension, or NS at diagnosis), and kidney histology (class VI disease, DPLN with crescents, significant interstitial fibrosis). Death within the first 5 years is usually due to active disease, infection, or thrombosis, whereas late mortality usually is caused by accelerated cardiovascular disease and malignancy.

IgA NEPHROPATHY

IgA nephropathy is the most common primary cause of GN throughout the world. Even though the course of the disease is indolent in most patients, a significant number of patients are at risk of ultimately developing ESRD.

EPIDEMIOLOGY

The true incidence of IgA nephropathy is uncertain, as there are no reliable serologic markers; a kidney biopsy is the only way to establish the diagnosis. The disease appears to be more prevalent in Asians and Caucasians and is relatively rare in blacks, with the male-to-female ratio ranging from 2:1 to 6:1. Although it may occur at any age, it most commonly presents in the second and third decades of life and rarely in the first decade. However, it has been reported in children as young as 3 years of age. Although IgA nephropathy is generally considered to be a sporadic disease, familial clustering has been observed in a pattern suggesting an autosomal dominant inheritance with incomplete penetrance. Specific disease-causing gene mutations have yet to be identified.

Most cases of IgA nephropathy are idiopathic, but numerous secondary causes are described and include severe liver disease (with regression following liver transplant); portosystemic shunts; enteropathies such as celiac disease and Crohn disease; chronic lung diseases, including obstructive bronchitis, cystic fibrosis, and idiopathic interstitial pneumonia; neoplasias such as small cell carcinomas and lymphoma; infections such as HIV, Mycoplasma, toxoplasmosis, and leprosy; dermatitis herpetiformis; and seronegative arthropathies. Lupus nephritis may also be characterized by significant IgA deposits. Many children with Henoch-Schönlein purpura have evidence of GN that histopathologically resembles IgA nephropathy. Some consider these 2 entities to be different manifestations of the same disease. An overlapping pathogenesis is suggested by the fact that both diseases share similar geographic and racial distribution; the 2 disorders have been reported to coexist in different individuals in the same family (including monozygotic twins), and rare patients with the idiopathic form of IgA nephropathy have been reported to develop Henoch-Schönlein purpura several years later.

PATHOPHYSIOLOGY

IgA nephropathy is a systemic autoimmune disease that manifests exclusively in the kidney. Glomerular IgA deposits are polymeric IgA1 derived from systemic rather than mucosal immune responses. It is proposed that 4 processes must come together to induce IgA nephropathy. First, compared to IgA from normal individuals, the IgA molecule itself is different, with a unique glycosylation pattern in the hinge region of the antibody characterized by reduced galactose residues. The serum level of galactose-deficient IgA1 is a heritable trait in diverse racial or ethnic groups. Second, loss of tolerance to galactose-deficient IgA1 results in production of antibodies specific for the aberrant IgA1. A dysregulated mucosal immune system with defective responses to commonly encountered alimentary pathogens or dietary substances is hypothesized to be the key trigger. The third and fourth processes occur when galactose-deficient IgA1 forms immune complexes and deposit in the mesangium, where they may interact with mesangial cells, activate complement via the mannose-binding lectin pathway, and trigger glomerular injury. These complexes may react with naturally occurring IgG antibodies. Indeed, IgG glomerular deposits often are present together with IgA. Glomerular IgA deposits are present in 3% to 16% of healthy individuals, suggesting that an abnormality of immune complex responsiveness or clearance by glomerular cells is required for disease.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

There are 3 classic clinical presentations of IgA nephropathy. Most frequently, 40% to 50% of patients present with recurrent episodes of painless macroscopic hematuria. The onset of macrohematuria often is preceded 24 to 48 hours earlier by an upper respiratory tract infection; associations with gastroenteritis, sinusitis, and strenuous exercise also have been reported but are less common. The latency period between the onset of prodromal symptoms and macrohematuria is considerably shorter than that observed in patients with APSGN. The gross hematuria episodes typically resolve within a few days, but microscopic hematuria and variable degrees of proteinuria typically persist. Some patients complain of flank or loin pain, presumably because of swelling of the kidney capsule. The first episode of macrohematuria may mimic APSGN. Recurrent episodes of macroscopic hematuria may occur in patients with Alport syndrome during the first decade of life, mimicking the clinical presentation of IgA nephropathy. Painless gross hematuria also may be the initial presentation of children with MPGN/C3G. The interval between episodes in IgA nephropathy is highly variable, ranging from a few months to many years. The frequency of the episodes decreases with age.

Almost as frequently, between 20% and 40% of patients present with an incidental finding of microscopic hematuria and mild proteinuria. A subset of this group will develop future episodes of gross hematuria, which occurs more frequently in children. Less commonly, fewer than 10% of patient will present with NS, occasionally mimicking minimal change NS clinically, but with IgA deposits on kidney biopsy. Finally, rare patients will present with an acute nephritic episode or RPGN. Most of the acute nephritic episodes resolve without specific therapy, and the long-term prognosis remains relatively good. If a kidney biopsy is performed during the acute phase, the degree of glomerular disease usually fails to explain the degree of kidney dysfunction. Superimposed acute tubular necrosis often is evident, with tubular toxicity from erythrocyte products such as hemoglobin and iron being possible contributory factors. The very small number of patients who develop RPGN caused by a severe crescentic form of IgA nephropathy have a poor outcome.

DIAGNOSTIC EVALUATION

There are currently no diagnostic serologic tests for IgA nephropathy. With the first episode of gross hematuria, serologic studies usually are performed to rule out other causes of acute GN. Serum IgA levels may be elevated in 8% to 16% of children with IgA nephropathy; the percentage is higher in adults, but this test is not sufficiently specific to establish a diagnosis. Although IgA and C3 have been found in the dermal capillaries of forearm skin biopsies in 30% to 50% of patients with IgA nephropathy, this test is not currently recommended. Testing for galactose-deficient IgA1 or urinary IgA1-IgG immune complexes remains investigational. Many pediatric nephrologists would recommend a biopsy for suspected patients with IgA nephropathy who have impaired kidney function, hypertension, or significant proteinuria, as these individuals are at increased risk of ultimately developing ESRD. The Oxford classification is an evidence-based schema devised by the International IgA Nephropathy Network and the Renal Pathology Society to predict potential progression of the disease that scores 4 histologic features of kidney injury: mesangial hypercellularity, endocapillary hypercellularity, segmental glomerulosclerosis, and tubular atrophy/interstitial fibrosis.

TREATMENT AND COMPLICATIONS

Optimal therapy for IgA nephropathy is controversial, but there is emerging evidence to guide treatment decisions (Fig. 468-5). The disease usually follows a benign course so that no therapeutic intervention is necessary. However, a subset of patients can develop progressive CKD. Risk factors include male sex, older age, elevated serum creatinine, hypertension, persistent proteinuria greater than 1.0 g/24 h, and the severity of the proliferative and sclerotic lesions on kidney biopsy. Current management recommendations include the following:

1. Patients with isolated microscopic hematuria and urinary protein-to-creatinine ratios less than 0.6 to 0.8 mg/mg (males and females, respectively) require no treatment except monitoring.

2. Tonsillectomy has been proposed as a therapeutic option, but there is a lack of evidence of long-term benefit when performed in patients without evidence of tonsil-associated clinical symptoms.

3. Hypertensive patients should be treated with an ACEi or ARB, as these drugs control the blood pressure and may slow the rate of decline in glomerular filtration rate. Normotensive patients with proteinuria greater than 0.5 to 1.0 g/24 h may also benefit from such treatment with a decrease in proteinuria.

4. The benefit of treatment with omega-3 fatty acids in the form of fish oil has not been firmly established due to conflicting clinical trial outcomes. It is thought to act as an anti-inflammatory agent. Compliance is poor due to fishy aftertaste; otherwise, it is well tolerated and may add some benefit, especially when combined with an ACEi or ARB in proteinuric patients with well-preserved kidney function.

5. Alternate-day steroids appear to benefit patients with progressive CKD. The optimal treatment duration is unclear; 6 months has been used in several clinical trials. Addition of cytotoxic agents, azathioprine, or MMF has yet to prove beneficial in preventing progression of CKD, but several clinical trials are under way to evaluate the efficacy of alternative protocols and new agents.

6. Patients with aggressive crescentic GN usually are treated with corticosteroids and an immunosuppressive agent, as the prognosis is otherwise poor. One approach is cyclophosphamide for 3 months followed by azathioprine or MMF for 1 to 2 years; the latter is based on results in small case series.

PROGNOSIS AND OUTCOMES

IgA nephropathy is a disease with a highly variable outcome. Complete sustained remission occurs in fewer than 10% of patients. For patients with self-limited episodes of recurrent gross hematuria who do not develop significant proteinuria, the long-term outcome is quite good. Risk factors for progressive disease include persistent and significant degrees of proteinuria, hypertension, and an elevated serum creatinine. In the high-risk patient group, kidney function slowly deteriorates. After 20 to 25 years of follow-up, 20% to 30% develop ESRD and another 20% have progressive CKD. Following kidney transplantation, IgA deposits commonly recur in the allograft (20–75%), but recurrent disease is not a common cause of graft failure.

HENOCH-SCHÖNLEIN PURPURA (HSP) NEPHRITIS

Henoch-Schönlein purpura (HSP) or anaphylactoid purpura is the most common vasculitis in children. It is a self-limited systemic vasculitis syndrome characterized clinically by a tetrad of clinical features that occur in any order and at any time over the course of several days to weeks: a purpuric rash (may initially look urticarial), arthralgias, abdominal pain, and GN. Unlike arthritis and gastrointestinal involvement, nephritis rarely, if ever, precedes the onset of purpura. Histologically, the GN is indistinguishable from idiopathic IgA nephropathy. Depending on the diagnostic criteria, the incidence of GN ranges from 20% to 40%. The kidney complications of HSP are discussed here, whereas the diagnosis and management of HSP are discussed elsewhere (see Chapter 200).

PATHOPHYSIOLOGY

Although HSP nephritis and IgA nephropathy are related diseases, they display pathophysiologic differences with important therapeutic implications. Whereas IgA nephropathy manifests exclusively in the kidney, HSP leads to many extrarenal manifestations. The pathophysiology is similar in its involvement of glomerular deposition of polymeric IgA1 immune complexes. The serum IgA1 in patients with HSP is also galactose-deficient. Circulating immune complexes containing galactose-deficient IgA1 also are detectable in HSP. The role of galactose-deficient IgA1 molecules in kidney lesions is suggested by the observation that they are found only in the serum of patients with HSP when they are having an episode of nephritis. The mucosal immune system is dysregulated, at least in part due to increased intestinal permeability to gut contents. Familial clustering of individuals with HSP nephritis and IgA nephropathy has been described in those who have similar circulating IgA abnormalities.

The variability of disease course between patients with HSP nephritis and IgA nephropathy might be explained by differences in the duration of production, amount, composition, and localization of the galactose-deficient IgA immune complexes and/or different predispositions to parenchymal cell injury within and beyond the kidney.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

Whereas most patients with HSP nephritis have asymptomatic hematuria and mild proteinuria, some (∼20%) develop gross hematuria. Approximately 10% of the children with kidney disease present with an acute nephritic syndrome. NS and RPGN with AKI are much less common. Many patients develop nephritis after the appearance of the rash; glomerulopathy usually is evident within 1 month of disease onset in the remainder. Rare clinical presentations include hypertension with a normal urinalysis and microscopic hematuria due to urethritis. Occasionally, patients with ANCA-positive vasculitis, SLE, and even APSGN may present clinically with a syndrome that mimics HSP, which should be ruled out by serologic studies in atypical patients. Seventy percent of children with HSP are clinically well within 4 weeks, but recurrent episodes occur fairly commonly for the first 4 months (∼33%) and have been reported after an interval as long as 5 years. GN is often more severe in those with recurrent disease.

DIAGNOSTIC EVALUATION

The diagnosis depends on the presence of palpable purpura with a normal platelet count and coagulation studies, plus 1 or more of the following: diffuse abdominal pain, arthritis/arthralgia, or a biopsy with prominent IgA deposition. There are no specific diagnostic laboratory tests for HSP. Testing for galactose-deficient IgA1 or urinary IgA1-IgG immune complexes remains investigational. Elevated serum IgA levels and IgA rheumatoid factor are detected in approximately 50% of patients. Serum complement levels usually are normal or elevated. Although not routinely performed, a skin biopsy shows small vessel leukocytoclastic vasculitis with IgA deposition. Kidney biopsy is not necessary for diagnosis but can be helpful in making treatment decisions when patients develop NS or RPGN. The histologic lesions of HSP nephritis can be categorized as defined by the ISKDC. Grade I has minimal histologic alterations; grade II, pure mesangial proliferation without crescents; grade III, mesangial proliferation with < 50% crescentic glomeruli; grade IV, 50% to 75% crescentic glomeruli, grade V,> 75% crescentic glomeruli; and grade VI, membranoproliferative-like GN.

TREATMENT AND COMPLICATIONS

There is no curative treatment for HSP nephritis. Most patients recover spontaneously with supportive care. However, a small subset present with severe GN are at risk of developing CKD and even ESRD. Current management recommendations are based on KDIGO guidelines (KDIGO Glomerulonephritis Work Group 2012). If proteinuria is > 0.5g/1.73 m²/d (including those with NS), ACEis or ARBs should be prescribed. Patients with ISKDC grades IV to VI have a less favorable prognosis and usually are treated with steroids with or without an immunosuppressive agent (cyclophosphamide and/or plasma exchange), depending on the severity of AKI. Long-term follow-up of a pediatric cohort failed to show a benefit for steroids in grade III HSP nephritis in which there were fewer than 50% crescents. If abnormal proteinuria persists for longer than 3 to 6 months in patients with grade I, II, or III nephritis, corticosteroids also should be considered.

Steroid treatment appears to decrease the severity of abdominal pain and may decrease recurrence rates. Nonsteroidal anti-inflammatory drugs (NSAIDs) may help relieve joint pain, but careful monitoring to detect nephrotoxicity is important. Studies in children have shown that early treatment with a short course (2–4 weeks) of prednisone does not reduce the risk of developing nephropathy within the first 12 months.

PROGNOSIS AND OUTCOMES

HSP is a self-limited disease that usually lasts 1 to 2 weeks. Typically, the prognosis is excellent. HSP morbidity is determined in most cases by gastrointestinal complications during the acute phase and by nephritis in the long term. Although the urinalysis remains abnormal in 50% of the children at 3 months and in 10% to 20% after 2 years, more than 90% of patients eventually achieve full clinical remission. A few children (3–4%) have a catastrophic acute illness and develop a rapidly progressive clinical course with ESRD within months. Another group (∼5%) has evidence of CKD, usually persistent proteinuria, with progression over a period of several years. Poor prognostic features include an acute nephritic presentation, heavy proteinuria, and the presence of glomerular crescents (> 50%) on biopsy. All patients with severe nephritis need monitoring for many years. In an outcomes study of 78 patients with HSP (after mean follow-up of a remarkable 23 years), 44% of individuals with NS or acute nephritis at presentation had developed hypertension or a progressive CKD, whereas 82% with isolated hematuria at presentation had normal renal function.

ANCA-ASSOCIATED VASCULITIS (AAV)–ASSOCIATED NEPHROPATHY

Systemic vasculitis with ANCA is associated with pauci-immune (minimal evidence of immune complex deposition) focal necrotizing GN. Two target ANCA antigens, both present in neutrophil granules, have been identified: proteinase-3 (PR3) produces a diffuse cytoplasmic staining pattern (cANCA), and myeloperoxidase provides a perinuclear staining pattern (pANCA). There are 2 small vessel vasculitic syndromes associated with ANCA and focal necrotizing GN: microscopic polyangiitis (MPA) and granulomatosis with polyangiitis (GPA; formerly Wegener granulomatosis). Eosinophilic granulomatosis with polyangiitis (EGPA; also known as Churg-Strauss) is a third type of small vessel vasculitis defined by a triad of asthma, eosinophilia, and eosinophilic inflammation; it has not been reported to cause glomerulopathy.

EPIDEMIOLOGY

Whereas ANCA-associated vasculitis (AAV) occurs mainly in middle-aged adults, it can present in adolescents and has been reported in children as young as 4 years of age. Nephritis is seen in 70% to 80% of children with both GPA and MPA, and in some cases, disease is renal-limited.

PATHOPHYSIOLOGY

These disorders are all associated with the formation of autoantibodies that react with antigens normally “hidden” from immune surveillance within neutrophil and monocyte cytosolic primary granules and lysosomes. The most common targets for ANCAs are thought to be myeloperoxidase (MPO) and PR3. The PR3 antigen also has been detected on renal tubular epithelia. The critical cryptic epitopes become expressed and are visible on the cell membrane in response to neutrophil “priming.” Infections (eg, S aureus), drugs (eg, antithyroid drugs, hydralazine, minocycline, and several others), environmental factors (eg, silica dust and heavy metals), and genetic risk factors all have been implicated as priming events. The ensuing immune response results in the formation of ANCA and sensitized mononuclear cells, both implicated in the pathogenesis of vascular inflammation and subsequent kidney damage. Neutrophil extracellular traps (NETs), Fc-receptors, microparticles, and localized complement activation also play roles in the ensuing nephritis.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

The Chapel Hill Consensus Conference Nomenclature of Systemic Vasculitis provides a framework for diagnosis and classification but has not been validated in children with vasculitis. The Paediatric Rheumatology European Society (PRES) has developed classification criteria for GPA but not MPA, and it has been endorsed by both the European Society of Paediatric Nephrology and the European League Against Rheumatism. The features of GPA and MPA overlap, except for necrotizing granulomatous lesions of the upper and lower airway in GPA but absent in patients with MPA. Both diseases may represent a clinical spectrum of the same primary pathologic process. They both typically present with a flu-like prodrome and systemic symptoms such as fever, anorexia, malaise, weight loss, and myalgias. Skin involvement (purpura or urticaria) and arthritis/arthralgia are common occurrences. Lung disease occurs in both syndromes, ranging from clinically silent involvement to life-threatening pulmonary hemorrhage. In MPA, the lung disease results from alveolar capillaritis, whereas in GPA, necrotizing granulomas form pulmonary nodules or even cavitating lesions. Granulomatous involvement of the upper airway, sinuses, nasal passages, and ears also is present in many patients with GPA. Frequently associated symptoms include nasal discharge, pain due to local ulcers, sinus pain, hoarseness, stridor, earache, and sensorineural hearing loss. Kidney involvement is very common in both, with rates of approximately 90% in MPA and 80% in GPA. There is a small subset of patients who present with renal-limited vasculitis. Histologically, the pauci-immune focal necrotizing GN is similar and often severe. All patients with nephritis have hematuria, most have proteinuria, and more than 70% have a reduced glomerular filtration rate at diagnosis. Some patients present with RPGN due to severe glomerular damage caused by fibrinoid necrosis and crescent formation. Even though GPA is classified as a granulomatous disease, granulomas are only rarely detected in perhaps 5% of kidney biopsies. Immune deposits are sparse or completely absent (justifying the name pauci-immune GN).

DIAGNOSTIC EVALUATION

ANCAs are detected in 85% to 95% of patients with active GPA (80–90% PR3-ANCA) and 70% with MPA (60% MPO-ANCA). Most patients (75–80%) with renal-limited vasculitis are MPO-ANCA positive. A small subset of patients with GN caused by anti-GBM antibodies and with autoimmune connective tissue disorders may also be ANCA positive. Although there is some overlap, the type of ANCA may predict the nature of the disease. ANCA may be present in other disorders, including inflammatory bowel disease, cystic fibrosis, and autoimmune hepatitis, for which the target antigen is typically neither PR3 nor MPO.

TREATMENT AND COMPLICATIONS

ANCA-associated systemic vasculitis is an aggressive disease. Without treatment, the 2-year mortality rate is 90%. The introduction of cyclophosphamide therapy has resulted in a dramatic improvement in outcomes, but significant morbidity and mortality are still encountered. Treatment plans are completely based on evidence from adult studies and small pediatric cohort studies. There is little role for corticosteroid therapy alone except to control the extrarenal localized disease. The first goal of therapy is to induce remission of the disease; it generally involves 3 to 6 months of corticosteroids and either cyclophosphamide or rituximab. Intravenous pulse steroids (30 mg/kg; maximum 1000 mg/dose) may be included, especially if the initial kidney disease is severe. Plasmapheresis is beneficial in patients with severe acute nephritis or severe pulmonary hemorrhage. Once the disease is in remission, maintenance immunosuppression with azathioprine, methotrexate, or rituximab is administered for at least 18 months to prevent relapse. Prophylactic therapy to prevent Pneumocystis pneumonia, cystitis, candidiasis, gastritis, osteopenia, and amenorrhea usually is prescribed. Trimethoprim-sulfamethoxazole appears to be beneficial for patients with GPA to prevent airway disease relapses. Relapse rates of 50% within 5 years (the risk is higher in the PR3-ANCA group) are typical.

Monitoring the ANCA titer can be useful in evaluating disease activity. Relapses are unusual when the titer is persistently negative, whereas disease relapses often are accompanied by the reappearance of ANCA or a significant titer increase. The Birmingham Vasculitis Activity Score (BVAS) is a validated tool for clinical treatment trials.

PROGNOSIS AND OUTCOMES

Patient and kidney survival have improved dramatically with intensive immunosuppressive therapy. There are few long-term follow-up studies of children. Five-year survival rates in adults are now in the range of 75% to 85%. The published pediatric series are small, but they often report 5-year survival rates greater than 90%. CKD persists in as many as half of the patients, with 20% to 25% eventually developing ESRD. Following kidney transplantation, recurrence in the allograft has been reported, but too few cases have been reported to know the effect on graft survival. Early deaths are commonly due to active disease, infectious complications, and thromboembolism; late deaths are often due to treatment sequelae. Long-term complications include an increased risk of infertility and malignancy.

ANTI–GLOMERULAR BASEMENT MEMBRANE DISEASE

A rare and most aggressive form of crescentic GN is mediated by anti-GBM antibodies. Pulmonary involvement, typically causing pulmonary hemorrhage, occurs in 40% to 70% of patients. This pulmonary-renal syndrome is called Goodpasture syndrome.

EPIDEMIOLOGY

Anti-GBM disease typically occurs in Caucasian adults with 2 peak incidences, 1 in the third and 1 in the seventh decade of life. It is rare in the first decade of life but has been reported in children as young as 2 years. Although most cases are sporadic, mini-epidemics have occurred. A genetic predisposition (HLA-DR15 and HLA-DR4 associated) has been noted in some patients. Anti-GBM nephritis may occur as a rare complication of other kidney diseases (after extracorporeal shock wave lithotripsy or in association with staghorn calculi, membranous nephropathy, or minimal change disease). Many patients had an antecedent flulike illness, and an association with influenza A2 has been reported. Another unique situation in which anti-GBM nephritis may occur is in the small subset (5–10%) of patients with Alport syndrome who receive a kidney transplant.

PATHOPHYSIOLOGY

The Goodpasture target antigen is a normal structural component of glomerular and pulmonary basement membranes. It consists of the last 36 amino acid residues of the carboxy-terminal region (NC1 domain) of the α3 chain of type IV collagen. Basement membrane type IV collagen consists of 6 distinct polypeptide chains (α1 to α6), 3 of which combine to form triple helices. The relative abundance of the α3 chain in glomeruli and alveoli likely explains the restriction of antibody-mediated disease to these organs. The antibodies produced after transplantation in patients with Alport syndrome recognize an antigen on the α5 NC1 domain of collagen IV, distinct from the Goodpasture target antigen. Most anti-GBM antibodies are of the IgG isotype, although a few patients with IgA antibodies have been reported. Antigen-sensitized T cells also contribute to the disease’s pathogenesis.

Why some, but not all, patients develop pulmonary disease is unclear, because the target antigen (α3 chain of type IV collagen) is present in all alveolar basement membranes. Experimental models suggest that a second factor in addition to the anti-GBM antibodies may be required before the lung tissue becomes affected. Exposure to tobacco, hydrocarbons, cocaine, and viral pneumonitis, all of which may cause lung injury, has been associated with an enhanced likelihood of pulmonary hemorrhage.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

Patients present with acute nephritis or with symptoms caused by CKD. Progression to irreversible kidney injury can occur within weeks. Pulmonary hemorrhage presenting as hemoptysis often precedes or accompanies the onset of nephritis. In 30% of patients, the pulmonary involvement may be clinically silent or may follow the onset of GN. The episodes of hemoptysis may be life-threatening. Pulmonary-renal syndrome may also occur in patients with SLE, systemic vasculitis, GPA, HSP, cardiovascular disease, and various infectious diseases.

DIAGNOSTIC EVALUATION

More than 90% of patients have an anti-GBM antibody detected in the plasma at the time of clinical presentation; however, the titer of the antibody does not correlate with the severity of the disease. A subset of patients (10–40%) may also have a positive ANCA due to anti-myeloperoxidase antibodies. Some of these overlap patients have evidence of extrarenal disease caused by vasculitis, and this group seems to have a slightly better prognosis. Iron deficiency anemia may occur as a complication of pulmonary hemorrhages. Short-lived acute pulmonary hemorrhage can be detected by the presence of hemosiderin-laden macrophages in bronchoalveolar lavage fluid. The diagnosis is confirmed by a kidney biopsy showing crescentic nephritis with linear deposits of IgG and C3 along the GBM.

TREATMENT AND COMPLICATIONS

If the patient has pulmonary disease or evidence of reversible (typically acute dialysis-independent) kidney disease, the mainstay of therapy is plasmapheresis to remove the circulating antibody. Immunosuppressive therapy (prednisone plus cyclophosphamide) is added to reduce new production of antibodies. For reversible disease, the duration of therapy may be guided by anti-GBM antibody titers.

PROGNOSIS AND OUTCOMES

The kidney disease is unlikely to be reversible unless treatment is started early. In most patients, production of anti-GBM antibody is of limited duration, with the antibody disappearing from the circulation within 8 to 14 weeks; it is rarely detected after 6 months. Nonetheless, the ability of this antibody to cause irreversible kidney injury is remarkable. ESRD in adults is almost inevitable if the initial creatinine is greater than 5 mg/dL or if more than 75% of the glomeruli have crescents on kidney biopsy. Patients who recover from the acute disease generally do well. Disease relapses are extremely rare, in the range of 2%.

RAPIDLY PROGRESSIVE (CRESCENTIC) GLOMERULONEPHRITIS (RPGN)

RPGN is a clinical syndrome characterized by sudden onset and rapid decline in kidney function. Kidney histology shows extensive crescent formation, typically in more than 50% of the glomeruli. Without specific therapy, most patients progress to ESRD within a period of weeks to months. Early treatment offers the best chance for renal recovery, so in suspected cases, early diagnosis is a nephrologic emergency.

EPIDEMIOLOGY

RPGN is a rare disorder in childhood that is classified into 3 groups based on the immunopathologic features (Fig. 468-6): (1) anti-GBM nephritis (5–10%), (2) immune complex nephritis (45–75%), and (3) pauci-immune disease associated with ANCA (10–40%). Less commonly, other forms of proliferative GN may present as severe crescentic disease.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

Children with RPGN may present with gross hematuria (50–85%), edema (13–80%), anemia (70%), and hypertension (63–85%). NS is uncommon, likely because of the rapid reduction in glomerular filtration leading to AKI. The urinalysis usually reveals an “active” urine sediment with hematuria, proteinuria, and cellular casts. Depending on the etiology, extrarenal manifestations also may be present. Pulmonary hemorrhage suggests a diagnosis of anti-GBM nephritis or ANCA vasculitis and may be fatal if treatment is delayed.

DIAGNOSTIC EVALUATION

Serologic studies can be very helpful in establishing a tentative diagnosis, but biopsy confirmation is essential. Initial tests should include C3, C4, ANA, ANCA, and anti-GBM antibodies. Kidney biopsy establishes the diagnosis in most patients and provides valuable information about the relative degree of acute (potentially treatable) and chronic histologic changes. If the kidney is damaged beyond the point of repair, immunosuppressive therapy is futile, at least for the glomerulopathy.

TREATMENT AND COMPLICATIONS

Therapy is directed to the specific disease entity, but a few general principles apply in all cases. Spontaneous resolution of RPGN is extremely unlikely to occur. The prognosis is poor unless aggressive therapy is started early in the course of the disease. The one exception is crescentic APSGN, which often spontaneously resolves. Because the diseases that cause crescentic nephritis are immunologically mediated, the mainstay of therapy is immunosuppression. Because RPGN is so rare, especially in the pediatric population, evidence-based treatment guidelines are not available. If immunosuppressive therapy is warranted, most nephrologists begin with high-dose intravenous corticosteroids, followed by daily oral prednisone and cytotoxic therapy (usually cyclophosphamide). Plasmapheresis is indicated for patients with anti-GBM disease (before AKI progresses to dialysis dependence) and severe ANCA-associated GN or associated pulmonary hemorrhage; its efficacy in other crescentic diseases is unclear. In general, the degree of kidney failure at diagnosis predicts long-term outcome.

NEPHROTIC SYNDROME

Nephrotic syndrome (NS) is not a disease but a constellation of clinical findings common to several glomerular disorders. By definition, it comprises proteinuria greater than 50 mg/kg per 24 hours (> 40 mg/m² per hour or a urinary protein-to-creatinine ratio > 2.0 mg/mg), hypoalbuminemia (serum albumin < 3.0 g/dL), and hyperlipidemia (elevated very-low-density lipoprotein, intermediate-density lipoprotein, low-density lipoprotein, and triglycerides). The reason for the increased hepatic production of lipoproteins during the nephrotic state is not entirely understood, but it appears to be associated with low plasma oncotic pressure or defects in lipoprotein catabolism. NS may be a manifestation of any of the proliferative glomerular diseases, but in children, it more commonly occurs as an isolated entity without clinical evidence of nephritis or significant glomerular hypercellularity on kidney biopsy. The pathophysiologic sequence of events that lead to the classical clinical features in NS, proteinuria, and hypoalbuminemia are depicted in Figure 468-7.

APPROACH TO A CHILD WITH ISOLATED NEPHROTIC SYNDROME

The NS of childhood has been divided into 3 broad groups: congenital/infantile, primary (inherited or idiopathic), and secondary. Only 10% to 15% of children have an identifiable secondary cause for their NS. The histologic lesions that are associated with secondary causes of NS are frequently indistinguishable from the idiopathic lesions, but the treatment is targeted to the primary underlying cause. During the past 2 decades, mutations in at least 28 genes have been identified in either inherited or sporadic primary NS (Table 468-4).

Primary NS is the occurrence of the constellation of clinical findings that define NS in the absence of an identifiable causative agent or disease. Primary NS is classified into 4 categories based on biopsy findings: minimal change nephrotic syndrome (MCNS), MCNS with proliferative changes, focal segmental glomerulosclerosis (FSGS), and membranous nephropathy (MN). Although it is still debated, one theory is that MCNS and FSGS represent different ends in the spectrum of the same disease rather than distinct disorders. From a prognostic perspective, the histologic pattern is less important than is the responsiveness to corticosteroids. Most children with steroid-sensitive disease have MCNS, so patients with new-onset NS do not undergo routine kidney biopsy at the time of diagnosis.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS OF STEROID-SENSITIVE NEPHROTIC SYNDROME (SSNS)

Most patients (95%) initially present with dependent edema that is most obvious in the eyelids, scrotum, and labia. Early morning swelling of the eyelids (periorbital edema) is a common occurrence, and frequently new patients are misdiagnosed as having allergic conjunctivitis. Increasing abdominal girth from ascites also frequently occurs. Most episodes of NS are triggered by an antecedent upper respiratory tract infection, as are most relapses. Less commonly, patients with NS present with symptoms secondary to complications of the NS, such as infections (peritonitis, cellulitis) or thromboembolic events. Hypertension is not typical, but a mild elevation in blood pressure occurs in 10% of patients. Microscopic hematuria is seen in approximately 20% of patients. Macroscopic hematuria is a distinctly uncommon manifestation and suggests another diagnosis.

DIAGNOSTIC EVALUATION

In addition to proteinuria, the urinalysis may show oval fat bodies (lipid-containing tubular cells) and waxy or hyaline casts. Granular casts are seen frequently when the patients are volume-contracted. Cellular casts are absent in idiopathic NS. When patients first present, the serum creatinine level is sometimes elevated, especially in the presence of intravascular volume contraction (frequently associated with an elevated hematocrit). Hyponatremia suggests dehydration with an appropriate increase in antidiuretic hormone level. The serum C3 level usually is normal or elevated. Urinary protein losses may result in low serum levels of IgG, thyroxine-binding globulin (nephrotic infants in particular are at risk of clinical hypothyroidism), vitamin D–binding globulin, alternative complement pathway factor B, and antithrombin III. Serum fibrinogen levels often are elevated and can be used as a surrogate marker for the risk of thrombosis.

Corticosteroids will induce a remission in the vast majority of children with NS. Therefore, for those with a typical clinical presentation, a kidney biopsy is not indicated prior to initiating steroid therapy. Atypical features that may require a kidney biopsy prior to treatment include age younger than 1 year, macroscopic hematuria, hypertension, hypocomplementemia, extrarenal symptoms such as a rash of arthritis, or AKI not caused by volume concentration.

SUPPORTIVE TREATMENT

Edema is mild in most patients and can be managed with dietary salt restriction. Fluid intake is not usually restricted except in the face of severe hyponatremia. Diuretics are effective but rarely are needed and must be used cautiously to avoid intravascular volume contraction. Intravenous albumin infusions may be hazardous but can be lifesaving in patients with clinical signs and symptoms caused by severe hypovolemia or generalized anasarca. A low serum albumin level alone is never sufficient reason to administer intravenous albumin. Hypertension seldom is present, even during corticosteroid therapy, but when present, it requires treatment with antihypertensive medications. In these situations, the hypertension usually is transient, and medications often can be discontinued once the disease is in remission.

A major aspect of the initial management is education for the parents. The majority of patients will follow a chronic relapsing course, and the families need to be prepared in advance. All families should be taught how to use a dipstick to check the urine for protein. The family should keep a diary of the urinary-dipstick results, along with drug doses and significant clinical events. The importance of testing the urine during concurrent illness must be emphasized. Whereas relapses are frequently triggered by an upper respiratory tract infection, transient low-grade proteinuria may also occur and resolve spontaneously. Using nonsteroidal anti-inflammation drugs as antipyretics is best avoided in these patients due to the potential nephrotoxicity of these drugs.

PHARMACOLOGIC THERAPY

Although spontaneous remissions may occur, the current standard of care is to treat these children with corticosteroids. Screening for tuberculosis and ascertainment of varicella immune status are recommended before high-dose corticosteroids are given. The optimal dose, schedule of administration, and total duration of steroid therapy are unknown. Most children receive initial treatment with prednisone (60 mg/m²/d or 2 mg/kg/d; maximal dose 80 mg/d) based on the ISKDC. Prednisone can be given in a single daily dose or can be divided into 2 or 3 daily doses. Ninety percent of children who will respond do so within 4 weeks and 98% by 8 weeks. The current definition of steroid-responsiveness is response within 8 weeks. Once the urine becomes negative for protein, the prednisone is switched to alternate days and tapered over the course of 6 weeks or longer. The total duration of daily and alternate-day steroid therapy used to treat the initial episode influences the subsequent relapse rate. However, 3 recent well-designed clinical trials showed no significant reduction in the risk of having a relapse when extending the therapy past 3 months. Alternate-day dosing of prednisone during the taper has been shown in pediatrics to cause less growth suppression.

Almost 50% of children with steroid-sensitive nephrotic syndrome (SSNS) experience several relapses. There is no agreement on a standard protocol for treating relapses. A commonly used protocol is prednisone 60 mg/m²/d until the urine is free of protein for 5 to 7 days; this is followed by alternate-day therapy that is tapered over several weeks.

FREQUENTLY RELAPSING NEPHROTIC SYNDROME (FRNS)

If a child with NS experiences frequent relapses (2 or more relapses within 6 months of initial response or 4 or more relapses within any 12-month period), then the potential adverse effects of cumulative steroid doses required to achieve remission begin to exceed the presumed benefits of maintaining remission of the NS.

The possibility of medication noncompliance, including use of lower than recommended prednisone doses, and the presence of occult infections (dental infections or sinusitis) should always be considered in patients with frequent relapses. Although some nephrologists perform kidney biopsy on children after diagnosis of frequently relapsing nephrotic syndrome (FRNS), nearly all have proven to be minimal change disease.

In cases of FRNS, the alternate-day dose can be tapered to a “threshold dose” (a dose below which relapses occur) to reduce the number of relapses and the total cumulative dose of steroid therapy. This dose is often in the range of 15 to 20 mg/m² and is continued for 12 to 18 months. There is evidence that taking this dose every day during upper respiratory tract infections may reduce relapse rates. Regular assessment of growth (both height and weight) and monitoring for cataracts are imperative in children receiving chronic or recurrent corticosteroid therapy.

Steroid-sparing strategies have also been developed to treat FRNS. One strategy involves alkylating agents such as oral cyclophosphamide for 8 to 12 weeks, which is generally well tolerated with minimal risk of gonadal toxicity (total cumulative dose < 200 mg/kg). Chlorambucil also is effective but is used less widely because of the risk of seizures. The alkylating agent is ideally started after induction of remission (to minimize the risks of developing infections and hemorrhagic cystitis) and is used in combination with low-dose prednisone. After undergoing treatment, the majority of patients experience a prolonged remission (35–65% still in remission at 5 years), but relapses do recur in a sizeable subset.

Other steroid-sparing strategies used for treatment include a 6-month course of oral MMF, B-cell depletion with rituximab, and calcineurin inhibitors.

STEROID-DEPENDENT NEPHROTIC SYNDROME (SDNS)

When a child with NS develops 2 consecutive relapses during steroid therapy or when the ability to achieve and/or remain in remission from NS requires administration of corticosteroids (relapse within 14 days of its cessation), then the child is considered to be steroid-dependent. Patients with steroid-dependent nephrotic syndrome (SDNS) may also fit the definition for having FRNS, and similarly to FRNS, the potential adverse effects of the cumulative steroid doses required to maintain remission exceed presumed benefits of achieving remission of NS. MCNS is the most common cause of SDNS. Before initiating steroid-sparing therapies, a biopsy may be performed.

As in FRNS, the alternate-day oral corticosteroid can be tapered to the threshold dose. Other steroid-sparing strategies used for treatment include a 6-month course of oral MMF, B-cell depletion with rituximab, calcineurin inhibitors, and vincristine. Children with SDNS do not respond as well to alkylating agents as does the FRNS subgroup. Patients with SDNS tend to also be medication-dependent to maintain a sustained remission.

STEROID-RESISTANT NEPHROTIC SYNDROME (SRNS)

Steroid-resistant nephrotic syndrome (SRNS) is defined as a failure to achieve remission after 6 to 8 weeks of full-dose daily prednisone. Approximately 15% to 20% of children with idiopathic NS will develop disease that is resistant to steroids. In addition, a small subset of patients who initially are responsive to steroids do evolve into having SRNS. Because MCNS is less common and FSGS is more common in this population, a biopsy is recommended for steroid-resistant patients prior to initiating alternative therapies. Genetic testing also is performed by many nephrologists for patients with SRNS, especially those who present at younger than 2 years of age or when there is a positive family history of NS.

The French Pediatric Nephrology Society recommends 3 doses of intravenous pulse steroids (1000 mg/m²) every other day for children with NS who are unresponsive to 4 weeks of steroids. Lack of response within 7 days is used as the indication for biopsy. If tolerated, a further 4 weeks of therapy will then increase the rate of SSNS by 10% to 20%.

The role of alkylating agents in treating SRNS is unclear, but a subset of patients will respond, especially those who had at least a partial response to prednisone therapy. Combined treatment with high-dose corticosteroids (oral prednisone with or without intravenous pulse steroids) and alkylating agents has reportedly induced a partial or complete remission in 30% to 60% of patients in some case series, but these protocols are associated with significant morbidity and should be limited to patients with well-preserved kidney function. Even a partial response is associated with a better outcome.

Cyclosporine A (CsA) is effective in inducing and sustaining remission in 75% to 80% of patients with SRNS and has had a major impact in a small group of patients debilitated by the disease and by steroid toxicity and in some patients who have had a poor response to cyclophosphamide. A common starting dosage is 5 to 6 mg/kg divided into 2 doses, with target predose blood levels of 50 to 100 ng/mL. Higher blood levels (100–200 ng/mL) are sometimes necessary in the more challenging patients, but the potential benefit of higher levels needs to be weighed against the risk of developing nephrotoxicity, which increases with the dose and duration of use. CsA levels must be carefully monitored, and some recommend repeat kidney biopsy to evaluate the degree of interstitial fibrosis if therapy is continued for longer than 18 months. Unfortunately, relapses commonly occur once CsA is discontinued. Mild side effects frequently occur and include hypertension, gingival hypertrophy, and hirsutism. Tacrolimus appears to be equally effective, avoids the development of the hirsutism and gingival hypertrophy associated with CsA treatment, and is now more commonly used in many programs. Starting doses of 0.05 to 0.1 mg/kg/dose twice daily are usually adjusted to achieve goal predose blood levels of 4 to 10 ng/mL. Steroid- and calcineurin inhibitor–resistant patients rarely respond to the other known immunosuppressant medications, including cytotoxic agents, rituximab, and MMF.

All hypertensive patients should be treated with ACEis or ARBs; these agents may also be recommended in normotensive patients for their antiproteinuric effects as long as serum potassium levels remain normal and the patient is not pregnant or at risk of becoming pregnant.

CONGENITAL NEPHROTIC SYNDROME

Congenital NS is defined as heavy proteinuria, hypoproteinemia, and edema starting within 3 months after birth. It is rare. Cases are caused either by genetic defects, especially nephrin and podocin (see Table 468-4), or by a perinatal infection. The Finnish type of congenital NS is caused by a nephrin gene mutation and is not accompanied by extrarenal malformations. Most of these children are born prematurely, with a birth weight ranging between 1.5 and 3.5 kg. The placental weight is more than 25% of the newborn weight in practically all cases. Proteinuria begins in utero and is detectable in the first urine sample after birth. Microscopic hematuria and normal creatinine values during the first months are typical. Heavy protein loss (up to 100 g/L) results in oliguria and severe edema if not treated. Hyperlipidemia, hypothyroidism, and hypogammaglobulinemia are present, due to urinary losses of lipoproteins, thyroid-binding globulins, and immunoglobulin. By ultrasound, the NPHS1 kidneys are large with increased cortical echogenicity and indistinct corticomedullary borders. The glomerular histopathology can vary and include minimal change, mesangial expansion, focal segmental glomerulosclerosis, or diffuse mesangial sclerosis, and the findings overlap in different entities. If congenital infections are ruled out, genetic analysis is the preferred method for establishing a definitive diagnosis.

For congenital NS, immunosuppression is not recommended. Conservative management of edema with sodium and fluid restriction and intermittent IV albumin and loop diuretics can usually keep these patients out of the hospital. The management of these patients also includes a hypercaloric diet, thyroid hormone replacement, monitoring for thrombotic episodes, and prompt management of infectious complications. The outcome of these patients without major extrarenal manifestations is comparable with those of other patient groups after kidney transplantation. Judicious use of ACEis and indomethacin to limit glomerular filtration and subsequent protein losses can allow for adequate growth until the children are large enough to get a kidney transplant. However, other patients require unilateral or bilateral nephrectomy and chronic dialysis as a bridge to kidney transplant.

MINIMAL CHANGE NEPHROTIC SYNDROME (MCNS)

MCNS, also known historically as nil disease or lipoid nephrosis, is the most common cause of NS in children, accounting for 90% of patients presenting at younger than 10 years of age and 70% to 80% of all pediatric patients with NS. MCNS is characterized by response to corticosteroids (> 90%), a chronic relapsing course (60–80%), and an excellent long-term prognosis. As long as future relapses remain steroid responsive, the long-term prognosis is good.

EPIDEMIOLOGY

The incidence of MCNS is 2 to 7 per 100,000 children under the age of 16, per year. The peak incidence is in preschool children, with a median age of 2.5 years; 80% of patients present before reaching their sixth birthday. Boys are more commonly affected. Familial cases have been reported (3% in some series), but the responsible genes have not yet been identified. There appears to be a higher incidence in the Asian population; however, a recent study suggests that they have less complicated clinical outcomes. A history of atopy is reported in 30% to 60% of these children, and occasional associations with food allergies have been made.

PATHOPHYSIOLOGY

In children, MCNS usually is idiopathic, although several secondary causes are known (Table 468-5). Isolated NS is considered to be a primary disease of the glomerular epithelial cell (podocyte) that is either injured by an unknown systemic factor or defective due to a cell-specific genetic defect (sometimes referred to as podocytopathy). See Table 468-4 for a list of podocyte genes known to cause podocytopathy and steroid-resistant NS. Thus far, the nature of the soluble “nephrotic factor” remains elusive. Evidence for the existence of a circulating “nephrotic factor” in idiopathic NS is supported by the observation that on the rare occasion when an MCNS kidney in relapse was transplanted, proteinuria rapidly resolved in the recipient. Conversely, MCNS has been reported to recur in kidney transplant recipients with a history of MCNS.

TREATMENT

Corticosteroids will induce a remission in approximately 90% of children with MCNS. Of those, approximately 60% relapse. The therapy for children with MCNS and FRNS, SDNS, or SRNS does not differ from that for patients with FSGS.

PROGNOSIS AND OUTCOMES

Before the introduction of antibiotics and corticosteroids, 40% of nephrotic children died within 5 years of being diagnosed. Deaths were most commonly due to infectious complications. Although today more than 95% of these children are alive at 25 years, deaths secondary to infections or thrombotic complications still occur (∼3% overall mortality rate). Late-onset kidney injury is very unusual, but a small number of patients develop acute tubular necrosis if a relapse is accompanied by sepsis or hypovolemia. NSAIDs increase the risk of developing AKI during a relapse.

Eight to 10 years after being diagnosed, 85% of patients achieve a long-lasting remission. Younger age at diagnosis and frequent relapses within the first 6 months predict longer duration of the disease. A number of patients (14–42%) continue to experience relapses into adulthood; the frequency may be more common than previously known and may even occur after a 20- to 25-year remission. Some patients who are initially steroid-responsive later become steroid-resistant. Many of these patients have FSGS, which may have been missed on an early biopsy or may have evolved from MCNS.

FOCAL SEGMENTAL GLOMERULOSCLEROSIS (FSGS)

The majority of children with SRNS (∼75%) have primary FSGS. FSGS is a more serious form of idiopathic NS, because only 20% to 25% of patients respond to steroids. It is the most frequent glomerular disease to cause ESRD in children and is second only to congenital anomalies of the kidney and urinary tract (CAKUT) as a cause of pediatric ESRD. FSGS is diagnosed histologically by the complete collapse of a segment of the glomerulus associated with mesangial sclerosis. Only some of the glomeruli are affected initially (ie, a focal rather than diffuse lesion), and the deep juxtamedullary glomeruli are involved first. FSGS may be primary, or it may develop as a secondary consequence of prior glomerular injury or hypertension (Table 468-6). The majority of children with hereditary or idiopathic FSGS present with SRNS (75–80%). The secondary forms of FSGS more typically present as asymptomatic proteinuria without the NS.

EPIDEMIOLOGY

In children, primary FSGS typically begins when they are between 2 and 7 years of age and occurs with an incidence of 0.3 cases per 100,000 patient years. There is a slight predominance in boys, especially when it develops at a young age. There is also a higher incidence among African Americans, Hispanics, and South Asians. The incidence of FSGS has been increasing in many pediatric and adult populations.

PATHOPHYSIOLOGY

Primary idiopathic FSGS recurs in kidney allografts, often within hours of the surgery, suggesting that a systemic factor(s) leads to the disorder. The nature and source of this factor and the reason for its production remain elusive. Pregnant women with FSGS gave birth to infants with NS that disappeared spontaneously a few weeks after birth. A bioassay based on the ability of plasma from patients with FSGS to increase albumin permeability in isolated rat glomeruli has been developed, but this factor has also been detected in genetic forms of FSGS, suggesting that it may not be a primary etiologic agent.

In a growing percentage of patients, steroid-resistant FSGS has been identified as an inherited disease (one that typically presents before 6 years of age). Mutations in the slit diaphragm protein nephrin (NPHS1) are the most common cause of congenital NS, followed by mutations in the Wilms tumor suppressor gene 1 (WT1) and podocin (NPHS2). Less common genetic causes of congenital NS have also been reported, although many (including WT1 and NPHS2) can present clinically later than 3 months of age. Across all pediatric age groups, mutations in podocin are the most common cause of FSGS. Inheritance is autosomal recessive; both homozygous mutations and compound heterozygous mutations have been reported. Importantly, 10% to 30% of children with no family history of SRNS harbor sporadic podocin mutations. An ever-growing number of genes have been linked to inherited forms of SRNS, and additional genes remain to be identified (see Table 468-4).

Genetic factors also enhance FSGS susceptibility. Variants of APOL1 have been linked to FSGS in African Americans with HIV or hypertension. The risk alleles are common in sub-Saharan Africa due to evolutionary pressure to maximize resistance to trypanosomiasis. Homozygosity for the risk alleles confers an estimated 4% lifetime risk for developing FSGS. Moreover, APOL1-associated FSGS occurs earlier and progresses to ESRD more rapidly.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

Most children (90%) with idiopathic FSGS present with NS that initially may be indistinguishable from SSNS. At the time of disease presentation, approximately 30% of children have mild hypertension, 55% have microscopic hematuria, and 20% have an elevated serum creatinine level. Less commonly, patients may present with asymptomatic proteinuria with or without microscopic hematuria. Macroscopic hematuria is rare.

DIAGNOSTIC EVALUATION

Definitive diagnosis requires a kidney biopsy, as is recommended for all patients with SRNS. Patients with FSGS also have tubulointerstitial damage that may impair proximal tubular function, causing features of Fanconi syndrome such as glycosuria, phosphaturia, renal tubular acidosis, and low-molecular-weight proteinuria. Genetic testing for podocin mutations is recommended for autosomal recessive, familial FSGS and for patients who fail to respond to calcineurin therapy.

TREATMENT

FSGS is a challenging disease to treat. With a poor overall response rate to immunosuppressive therapy and a high rate of recurrence in kidney allografts, this disease can be devastating. The current KDIGO guidelines recommend initial treatment of primary FSGS with high-dose prednisone given for between 4 and 16 weeks or until complete remission. At best, 25% of patients achieve a remission after treatment with a prolonged course of corticosteroids alone. Often the steroid therapy must be continued for several months.

Calcineurin inhibitors are recommended for patients who are resistant or intolerant to corticosteroids. Considerable evidence-based literature supports the use of CsA as the second-line agent together with low-dose prednisone. Initial plasma trough levels may need to be higher than those for SSNS, often in the range of 125 to 225 ng/mL. Tacrolimus is sometimes effective in CsA nonresponsive patients and is increasingly replacing CsA due to a better side effect profile. Rituximab can be effective in steroid-responsive patients but has been disappointing in patients with SRNS. MMF may have a role in the treatment of steroid- and calcineurin-resistant FSGS, but data in children are limited. Triple therapy (steroids, CsA, and MMF), plasmapheresis, protein adsorption columns, and low-density lipoprotein apheresis are all reportedly effective in selected individuals, but data are insufficient to recommend their use at this time.

Although controversial, patients with monogenic secondary forms of FSGS should still receive at least a trial of immunosuppressive drugs, although responsiveness is exceptional. Until genetic testing becomes standard of care, most patients with steroid-resistant FSGS have already tried and failed a variety of immunosuppressive agents by the time the genetic testing is considered.

For recurrent FSGS after kidney transplantation, plasmapheresis and escalation of the dose of calcineurin inhibitors have occasionally been effective. However, many patients will lose their graft and return to dialysis.

PROGNOSIS AND OUTCOMES

The single best prognostic indicator in patients with idiopathic NS is responsiveness to steroid therapy. Approximately 50% of the patients who do not achieve remission of NS develop ESRD within 10 years. The risks of developing progressive CKD are higher among patients of Hispanic and African descent, children with disease onset when younger than 1 year of age, and patients with the collapsing variant of FSGS. The severity of the proteinuria and the degree of interstitial fibrosis on kidney biopsy also correlate with the rate of progression to ESRD. Recurrence of proteinuria and NS after transplantation occurs in 20% to 30% of patients. Patients with podocin mutations have rarely achieved remission with immunosuppressive therapy, so progression to ESRD is predicted.

MEMBRANOUS NEPHROPATHY (MN)

Membranous nephropathy (MN) is a noninflammatory proteinuric glomerular disease characterized by the presence of hallmark subepithelial immune deposits (usually containing IgG and C3). NS secondary to MN does not usually respond to an 8-week course of prednisone; therefore, diagnosis is made on biopsy.

EPIDEMIOLOGY

MN is a rare cause of NS in children, accounting for approximately 1% of all NS cases in North America. Although slightly more common in adolescents, it can present at any age, including in infants. Cases affecting identical twins and siblings have been reported, suggesting a genetic factor. Detailed epidemiologic studies in children have not been conducted.

PATHOPHYSIOLOGY

Unlike the disease in adults, the majority of childhood cases are secondary to an underlying disorder (Table 468-7). Idiopathic MN appears to be an antibody-mediated disease. Immune complexes are thought to form locally within the subepithelial space, where the target antigen is located. In a rat model of MN (Heymann nephritis), the target antigen is megalin, a normal constituent of rat but not human podocytes. Compared to 70% to 80% of adult patients with MN, autoantibodies to M-type phospholipase A2 receptor (aPLA₂R) are seen in 45% to 75% of children and adolescents with primary MN. Noncoding genetic sequence variants in the gene locus encoding PLA₂R also are associated with the development of MN. In a small number of pediatric patients, immune complexes involving cationic bovine serum albumin (BSA) and anti-BSA antibodies have been implicated, raising the possibility of a role for certain food antigens in MN pathogenesis. Other targets for autoantibodies in adult idiopathic MN, such as aldose reductase, manganese superoxide dismutase, alpha-enolase, and thrombospondin type 1 domain–containing 7A, have been described. An unusual antenatal form of MN has been reported and is associated with transplacental passage of antibodies to the podocyte antigen neutral endopeptidase. MN has also been reported rarely in patients with insulin-dependent diabetes mellitus, Crohn disease, and primary biliary cirrhosis, suggesting that other candidate autoantigens will eventually be unveiled. Glomerular inflammation does not explain the proteinuria; rather, the terminal membrane attack complex of the complement cascade C5b-9 appears to be the critical mediator.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

Proteinuria is the hallmark of MN. At presentation, 70% have NS, 70% have microscopic hematuria, and 20% are hypertensive. The disease usually is insidious in its onset and progression. When a kidney biopsy establishes a diagnosis of MN (usually after the patient’s failure to respond to corticosteroid therapy for NS), an exhaustive investigation is warranted to look for an underlying cause. Only 3% of children have kidney failure at presentation. A rapid deterioration in kidney function mandates a search for an additional cause. For example, a small subset of these patients develops a superimposed crescentic GN, sometimes due to the formation of anti-GBM antibodies. Although renal vein thrombosis is a common complication among adult patients with MN, this complication is rare in children.

DIAGNOSTIC EVALUATION

A kidney biopsy is required to establish a definitive diagnosis. Serologic tests, including serum complement levels, usually are normal. Laboratory investigations are particularly helpful during the search for a secondary cause (see Table 468-7). Hepatitis serology, serum complement levels, and an ANA test should be obtained in all patients. The kidney biopsy suggests a secondary cause if immune deposits are also found in the mesangium (SLE), if significant deposits of IgA are present (IgA nephropathy), or if inflammatory cells are present (APSGN); all these features are atypical of the idiopathic MN.

TREATMENT

Established treatment guidelines for children with idiopathic MN do not currently exist. Children who present with nonnephrotic proteinuria, normal blood pressure, and normal kidney function appear to have a good prognosis without specific therapy, but close follow-up is warranted. Spontaneous remissions have been reported in 3% to 30% of adult patients. Use of ACEi or ARB therapy and a low-sodium diet are recommended for hypertension and as antiproteinuric therapy.

Treatment guidelines for patients at risk of developing CKD (NS, scarring on kidney biopsy, rising serum creatinine level during follow-up) have been developed for adults and often are applied to children for whom there are no evidence-based data, although the long-term prognosis appears to be better in children. In general, corticosteroids alone decrease proteinuria, but effects on long-term kidney function are less clear. For this reason, they are usually combined with an alkylating agent, either at the beginning of therapy or within 6 months, depending on the initial response. Therapy is required for several months. Calcineurin inhibitors are an alternative, especially for patients in whom alkylating agents are contraindicated. Data are emerging to suggest that MMF and rituximab may be useful in MN and that PLA2R-Ab levels are predictive of rituximab effect.

PROGNOSIS AND OUTCOMES

MN usually is an indolent, slowly progressive disease with fewer than 5% of children developing ESRD 5 years after being diagnosed. Good, long-term follow-up studies are not yet available for patients who develop MN in childhood.

MEDICAL COMPLICATIONS OF NEPHROTIC SYNDROME

Persistence of NS is associated with significant morbidity and even mortality. Short-term mortality rates of 3% are most frequently due to infectious complications and less commonly to thromboembolic events. Because NS frequently follows a chronic relapsing course, children are at risk for developing several recognized long-term complications (Table 468-8). Not to be overlooked is the tremendous psychological stress that the patients and their families endure. The side effects of corticosteroids and cytotoxic drugs are also significant for many children.

INFECTIONS

Even in the current medical era, acute bacterial infections are relatively common and potentially fatal complications of NS. The acute onset of fever in a child during a relapse of NS needs urgent evaluation. Primary peritonitis still occurs in 2% to 6% of these children. Patients with serum albumin levels less than 1.5 g/dL have a greater risk of developing peritonitis. Fever associated with abdominal pain must be considered peritonitis until proven otherwise. Prior to the advent of routine immunization against pneumococcal organisms, 50% of peritonitis cases were caused by Streptococcus pneumoniae, whereas other encapsulated or gram-negative organisms were occasionally isolated (especially Escherichia coli). Patients with presumed peritonitis should be treated with antibiotics to cover both gram-positive and gram-negative organisms while culture results are pending. Unfortunately, the peritoneal culture is negative in 15% to 50% of patients with a presumptive diagnosis of peritonitis. Prophylactic penicillin has not proven to be effective for preventing pneumococcal peritonitis. Once a nephrotic patient is in remission and off steroids for a few months, inoculation with the 23-valent pneumococcal polysaccharide vaccine (Pneumovax) and the pneumococcal conjugate vaccine (Prevnar 13) should be done, if not given in infancy as is currently recommended.

Several abnormalities associated with the nephrotic state predispose patients to development of bacterial infections: abnormalities of cellular immunity; low total serum IgG and specific antibody levels (only partly explained by urinary losses); and decreased serum levels of complement proteins B and D, which impair plasma bacterial opsonic activity and may explain the predisposition to infections caused by encapsulated bacteria. Local factors related to tissue edema, breakdown of the skin barrier, and the presence of peritoneal fluid, which is a rich culture medium, may also facilitate infections. The use of immunosuppressive drugs is another contributing factor.

Varicella-zoster infections occurring during a relapse may be severe. The live-attenuated vaccine should be given only after immunosuppressive drugs have been discontinued for a few months. In the interim, significant exposures should be treated with zoster immune globulin if available or antiviral drugs such as valacyclovir. Varicella antibody levels may decline in patients with numerous relapses or with NS that is refractory to treatment. Annual immunization with killed influenza vaccine is recommended.

THROMBOTIC DISEASE

Thromboembolism is one of the most serious complications of NS. Virtually all nephrotic patients are in a hypercoagulable state, and as many as 20% experience thrombotic events that are often clinically silent (only 2–4% have symptoms). The cause of the NS may affect the absolute risk. Thrombosis is a relatively common occurrence in adults with MN but infrequent in children with MCNS, but it is still recommended that children with documented thrombotic complications undergo an evaluation for known genetic causes of thromboembolic disorders. Which abnormalities of the clotting process best correlate with the risk of thrombosis remains unclear. Plasma fibrinogen levels appear to be the best surrogate measure of hypercoagulability. Virtually every component of the hemostatic pathway is perturbed in some way, including increased serum levels of clotting factors (especially factors VIII and V and fibrinogen), urinary loss of anticoagulants (antithrombin III and protein S), and increased platelet number and adhesiveness. Other contributing factors include hyperlipidemia and hyperviscosity. The risk of developing thrombotic disease is significantly enhanced in the presence of hypovolemia and during prolonged periods of immobilization. Hospitalized young adult nephrotic patients have a 7-fold increased risk of experiencing thromboemboli. Indwelling venous catheters impose a major risk and should be avoided if at all possible.

The clinical presentation is highly variable and dependent on the site of the clot. In children, both arterial and venous clots occur. The renal vein and sagittal sinus may be targets, as are the pulmonary and femoral arteries. Aspirin therapy may be recommended for patients with significant thrombocytosis. Prophylactic anticoagulation is not recommended in the pediatric age group due to the risks of bleeding. In those with a documented clot or very high risk factors (such as an indwelling catheter), warfarin or low-molecular-weight heparin is used; the latter requires adequate factor VIII activity levels.

ACUTE KIDNEY INJURY (AKI)

AKI is a common occurrence in children hospitalized with complications of NS (50% in a recent study). Predisposing factors are numerous and may include infection, nephrotoxic medication, and hemodynamic consequences of the nephrotic state.

CARDIOVASCULAR DISEASE

An unresolved issue of significant concern is whether children with chronic relapsing NS are at risk of developing premature atherosclerotic disease. Autopsies done on children as young as 5 years who died from NS have found atheromatous lesions. In adults with NS, the relative risk of myocardial infarction is 5.5 compared with a control population after adjustment for other known risk factors.

Several factors may contribute to the increased risk of having cardiovascular disease and stroke; they include hypertension, the use of steroids, systemic inflammation, and the atherogenic plasma lipid profile. In particular, plasma cholesterol and lipoprotein levels are high. Whether hypercholesterolemia should be treated with 3-hydroxyl-3-methylglutaryl-CoA reductase inhibitor (statin) therapy in children aged 8 to 18 years with hypercholesterolemia is controversial. Use of statins may reduce cardiovascular risks, but there is no strong evidence that it slows the progression of kidney disease. The potential benefits of using statins must be weighed against the concerns about its effects on myelination and the risk of developing statin-induced rhabdomyolysis, which appears to be higher with concurrent use of a calcineurin inhibitor. Although no consensus statement exists to guide decisions in NS, there are expert panel integrated guidelines published on pediatric cardiovascular risk reduction by the National Heart, Lung, and Blood Institute (http://www.nhlbi.nih.gov/health-pro/guidelines/).

BONE DISEASE

Although reduced bone formation rates, osteoporosis, and avascular necrosis are known complications of long-term use of corticosteroids, these complications appear to be rare in patients with SSNS based on an evaluation of bone density by dual energy x-ray absorptiometry. One proposed explanation is that the steroid-associated increase in the body mass index is protective as a consequence of increased biomechanical loading. Children with numerous steroid-treated relapses may not reach their final adult-height potential. A recent study reported that more than 5% of children with NS had radiographic evidence of vertebral fractures at 1 year, indicating that bone health needs ongoing surveillance in patients with frequent relapses.

HEMOLYTIC-UREMIC SYNDROME (HUS)

Historically, HUS is the most common glomerular disease to cause severe AKI in a previously healthy young child. Most children with HUS (90%) have an antecedent diarrheal illness caused by a strain of E coli that produces a Shiga-like toxin. This group of patients had been formerly classified as having “diarrhea-associated” HUS (D+HUS), and this disorder is now known as Shiga toxin–producing E coli–associated (STEC+) HUS.

Approximately 10% of patients develop HUS for other reasons, which are summarized in Table 468-9; these cases are referred to as atypical HUS.

The possibility of a genetic form of the disease should be considered in any child with an atypical presentation. Although both HUS and thrombotic thrombocytopenic purpura (TTP) are characterized by thrombotic microangiopathy, the triggers, primary target organs, response to specific treatments, and clinical outcomes are distinct. These disorders are currently considered as separate entities. A syndrome resembling HUS can also be seen in patients with malignant hypertension.

EPIDEMIOLOGY

HUS occurs worldwide with sporadic and epidemic patterns. More than 70% of children in the United States and Western Europe with STEC+ HUS, which is mainly a disease of infants and young children (9 months to 5 years of age), have been infected with E coli 0157:H7. The incidence is estimated to be 2 to 3 new cases per year per 100,000 children younger than the age of 5 years. In children younger than the age of 15 years, the risk of developing HUS following an E coli 0157:H7 colitis is in the range of 8% to 10%. Other strains of Shiga toxin–producing E coli (0103:H2, associated with HUS after causing a urinary tract infection) and Shigella dysenteriae serotype 1 may also trigger HUS. There is often a seasonal disease pattern, with the greatest incidence in the summer and fall in North America. HUS is an uncommon occurrence in African Americans and occurs more frequently in rural populations. The primary reservoir for E coli 0157:H7 is farm animals, especially cattle; in the United States, approximately 1% of cattle harbor this bacterium. The bacteria can be killed by exposure to adequate heat; undercooked beef and nonpasteurized milk or milk products are frequent sources of exposure. Consumption of fruits, juices, or vegetables that have been exposed to contaminated manure and contaminated lakes and swimming pools all have caused disease outbreaks. Although it is unusual, person-to-person spread has been documented.

PATHOPHYSIOLOGY

STEC+ HUS

The sequence of events that result in the clinical features and differential diagnosis of HUS are depicted in Figure 468-8. HUS is initiated by damage to the endothelium with formation of microthrombi, leading to the diagnostic triad of microangiopathic hemolytic anemia, thrombocytopenia, and AKI. The enteropathogenic E coli are characterized by the production a Shiga-like toxin (first reported in stool of patients with S dysenteriae). These toxins are also referred to as verotoxins because of their cytopathic effects on Vero cells, a cell line that is derived from African green monkey kidney cells. The toxins can bind to a glycosphingolipid receptor (Gb3) that is expressed on human glomerular endothelial and mesangial cells and on tubular epithelial cells. During the enteric prodrome, plasma samples show evidence of a prothrombotic state. Once the toxin damages glomerular endothelial cells, a thrombotic microangiopathic reaction ensues that is characterized by depositions of fibrin and damage to erythrocytes and platelets. A similar cascade of events may cause injury in other vascular tissues as well.

ATYPICAL HUS

The most common cause of atypical HUS is a dysregulation of the alternative complement activation pathway. Loss-of-function mutations in CFH, factor I, and membrane cofactor protein (MCP), complement regulatory proteins, as well as gain-of-function mutations in C3 and complement factor B have all been identified in patients with atypical HUS. In a small subset of patients, autoantibodies against CFH (and less commonly factor I) prevent normal regulation of the alternative pathway. More than 90% of patients with anti-CFH antibodies have a complete deficiency of CFH-related proteins 1 (CFHR1) and 3 (CFHR3) due to a homozygous deletion of CFHR1-R3. It is not yet entirely clear how complement dysregulation is manifest differently in atypical HUS compared to C3 glomerulopathy. It is likely that the initial kidney injury triggers are distinct. Common antecedents of atypical HUS include upper respiratory infections, other febrile episodes, non–E coli diarrhea, and drug exposure.

Another important cause (15–30% cases) of atypical HUS occurs as a complication of S pneumoniae infections in young children. Many of the offending strains are not included in the multivalent pneumococcal conjugate vaccine that is now routinely given in infants. This organism produces neuraminidase, an enzyme that cleaves cell membrane sialic acid residues on erythrocytes, platelets, and glomeruli to expose the cryptic Thomsen-Friedenreich (T) antigen. It is hypothesized that naturally occurring IgM antibodies react with the T-antigen to induce hemolysis, thrombocytopenia, and glomerular capillary damage. These patients have a positive Coombs test, and difficulties are encountered with ABO cross-matching. Use of fresh-frozen plasma, an additional source of the pathogenic antibody, is contraindicated. The outcome has improved in this patient group, but mortality rates remain high, in the range of 12%.

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

Following ingestion of E coli–contaminated food or liquid, most affected children develop abdominal pain and diarrhea that is usually bloody (90%). Half of the patients have nausea and vomiting. Fever is typically low-grade or absent. The colitis usually is self-limited, but complications may occur and include rectal prolapse, toxic megacolon, bowel wall necrosis, and perforation. Strictures may develop later. HUS occurs in approximately 15% of infected children. The onset of HUS usually is abrupt, occurring 5 to 10 days (median 1 week) after the onset of the diarrhea, as the colitis is resolving. The presenting symptoms often are due to AKI and relate to the patient’s intravascular volume status. The severity of the AKI is highly variable, but 50% to 60% develop oligoanuria and require dialysis. Hypertension is variable but may be severe. It is caused by fluid overload and by activation of the renin–angiotensin system within the ischemic kidneys. The mean duration of AKI is 2 weeks. In atypical HUS, the antecedent illness is more nonspecific, often associated with fever, upper respiratory symptoms, and vomiting. Sometimes a prodromal illness is not apparent at all. The clinical onset of HUS can be insidious.

The microangiopathic hemolytic anemia causes pallor, mild icterus, and symptoms secondary to acute anemia. The degree of thrombocytopenia is variable; patients may present with petechiae. Hemolysis may continue for several weeks. There is no correlation between the severity of the hemolysis and thrombocytopenia and the degree of AKI.

Neurologic involvement is a common occurrence and can mimic TTP. Most patients are very irritable and somnolent. However, more serious involvement leading to seizures and coma occurs in 10% of children. The neurologic manifestations may be the presenting feature and may impact acute mortality rates and long-term morbidity.

Although the bowel and the kidneys are always involved, virtually any organ can become damaged by microvascular thrombosis. Approximately 40% of patients have hepatomegaly with elevated serum transaminases, and 20% have pancreatic involvement with elevations of serum amylase and lipase levels. Less commonly, patients develop hyperglycemia and may require insulin therapy. Most of these patients are able to discontinue insulin when the acute illness resolves, but some develop chronic insulin-dependent diabetes. Although primary cardiac involvement not related to volume overload rarely occurs, myocarditis and myocardial ischemia are serious complications. Pericardial effusions may also occur.

DIAGNOSTIC EVALUATION

Biochemical evidence of AKI associated with microangiopathic hemolytic anemia and thrombocytopenia are the hallmark features of HUS. Urinalysis typically shows hematuria and proteinuria. Anemia is characterized by the presence of schistocytes and helmet cells on the smear, negative Coombs test (except in pneumococcus-associated HUS), elevated reticulocytes and lactose dehydrogenase (LDH) level, and low serum haptoglobin concentrations. The prothrombin (PT) and partial thromboplastin time (PTT) are usually normal unless antibiotic therapy has caused vitamin K deficiency. Serum levels of many prothrombotic molecules are increased (tissue factor, platelet-activating factor, plasminogen activator inhibitor, von Willebrand factor, thromboxane A₂), whereas decreased levels have been reported for some antithrombotic systems (prostacyclin, thrombomodulin, plasminogen activator). Hypoalbuminemia a frequent occurrence at presentation, primarily due to gastrointestinal losses. Leukocytosis is a common occurrence, and peripheral white blood cell counts greater than 20,000/μL correlate with a poorer prognosis. Serum C3 level and CH50 (total hemolytic complement, the dilution of serum that lyses 50% of red blood cells in a reaction mixture) may be depressed, but only in those 30% to 40% of patients with atypical HUS associated with genetic mutations in complement regulatory genes. However, a normal serum C3 level does not rule out the possibility of a genetic complement deficiency.

When a patient presents with clinical features suggestive of STEC+ HUS, documentation of a Shiga toxin–producing E coli is in order to identify the source of exposure and minimize the spread of the disease. Stool cultures should be obtained as soon as possible with special sorbitol-MacConkey agar plates used to identify colorless colonies of E coli 0157:H7. The organism is difficult to culture after the first few days of diarrhea. Many laboratories can also test for the presence of the Shiga toxin in the stool.

When atypical HUS is a possibility, it is important to obtain a critical serum sample before initiating transfusion or plasma exchange. ADAMT13 activity and inhibitor testing is important to rule out a diagnosis of TTP (while not common in children but for which urgent plasmapheresis therapy is indicated) as well as to evaluate the complement system: serum C3, C4, CH50, CFH, and complement factor I levels and antibodies to CFH and complement factor I (or activity levels if antibody testing will be delayed), and MCP staining on leukocytes if available. Low serum C3 levels are an early clue to a genetic complement regulatory abnormality, but normal levels do not preclude such a diagnosis. Other clues might include diarrhea-negative HUS, patients with an HUS relapse, a positive family history of nonsynchronous HUS, postpartum HUS, and HUS after bone marrow transplant.

A definitive genetic etiology will require specific gene testing and is important as it has long-term therapeutic and prognostic implications.

TREATMENT AND COMPLICATIONS

For STEC+ HUS, supportive medical care is by far the most important aspect of treatment and is responsible for the drastic decline in mortality rates, down from greater than 50% in the predialysis era. Meticulous attention must be paid to management of fluids and electrolytes. Volume loss from diarrhea and vomiting must be replaced, but care must be taken to avoid intravascular volume overload. Severe oligoanuric AKI requiring dialysis still develops in 40% to 50% of the patients. Packed red blood cell transfusions should be reserved for patients with clinical indications of severe anemia (hematocrit < 18%) but are eventually required in approximately 80% of patients. Packed red blood cells must be infused slowly or during dialysis, with careful monitoring of the patient’s blood pressure. Platelet transfusions should be avoided unless the patients are actively bleeding or the transfusions are needed in preparation for an invasive procedure. Many patients require antihypertensive drugs; angiotensin pathway inhibitors are best avoided during the acute phase due to their effects on kidney perfusion that may aggravate AKI. Nutritional support is important, as many patients with HUS are already catabolic due to several days of poor caloric intake before presentation.

No specific therapeutic interventions have proven beneficial for treating patients with STEC+ HUS. Plasma infusion or plasma exchange therapies do not decrease mortality rates or improve long-term outcome. However, these therapies are still considered in patients with severe central nervous system involvement based on their proven efficacy in adults with TTP. Antiplatelet drugs and fibrinolytic therapy are of no proven benefit but increase the risk of having hemorrhagic complications and should be avoided. During the prodromal phase of colitis, antimotility drugs are contraindicated, and antibiotics should be avoided as they have been reported to increase the probability of developing HUS.

There is a consensus treatment approach for optimal treatment of atypical HUS. Eculizumab, a monoclonal anticomplement component C5 antibody, prevents many of the proinflammatory and prothrombotic effector functions of the complement cascade. Prospective clinical trials in children with atypical HUS show that eculizumab is highly effective in patients with identifiable complement mutations. Prior to the availability of eculizumab, the prognosis of atypical HUS was poor and the recurrence rate was high. Plasma therapy could achieve complete or partial remission in 75% of HUS episodes, but 50% of affected children either died or reached ESRD at 3-year follow-up. If atypical HUS is due to an anti-CFH autoantibody, immunosuppressive agents with or without plasmapheresis must be used in addition to eculizumab in order to eliminate autoantibody-producing cells.

During the recovery phase, the hemolytic process resolves more slowly than do the other manifestations, and patients often are discharged with a persistent anemia from an inappropriately low reticulocyte count that cannot be explained by deficiency of folate, vitamin B₁₂, or iron. A low serum erythropoietin level may be a factor. The hemoglobin level eventually normalizes in most patients without specific intervention. Cholelithiasis has been reported as a late sequel to the hemolytic process.

PROGNOSIS AND OUTCOMES

Acute mortality rates for STEC+ HUS are in the range of 3% to 5%, often related to severe complications such as neurologic disease and cardiac failure or multiorgan involvement. Kidney function spontaneously improves, and almost all patients are able to discontinue dialysis. Approximately 5% of survivors suffer long-term damage to the kidney or brain. However, several recent long-term follow-up studies suggest 5% to 25% of patients will follow a course of slowly progressive CKD with or without hypertension. Patients with proteinuria, hypertension, or kidney dysfunction need long-term follow-up. Poor prognostic indicators include oliguria or anuria for longer than 2 weeks, an initial neutrophil count greater than 20,000, coma on admission, and atypical forms of the disease. Although kidney biopsies are not generally performed, when these data are available, CKD has been predicted by the presence of cortical necrosis or thrombi in more than 60% of the glomeruli and by involvement of extraglomerular renal vessels. Atypical HUS generally follows a more severe clinical course, is more likely to relapse, and may be associated with genetic deficiency of 1 of the key complement cascade regulatory proteins (see Table 468-8).

Recurrent disease in either native or transplant kidney is extremely rare in STEC+ HUS (< 10%) but occurs quite commonly in atypical HUS. Prophylactic eculizumab treatment in patients with atypical HUS and ESRD is recommended at the time of kidney transplantation, and ongoing treatment may be necessary for long-term allograft survival.

PREVENTION

Children actively infected with Shiga toxin–producing E coli should not be treated with antibiotics or antimotility agents, as they may increase the risk of developing HUS. Maintaining a normal intravascular volume during the diarrheal phase (often necessitating hospitalization for intravenous fluids) may decrease the severity of AKI when HUS ensues. Treatment during the diarrheal stage with Synsorb-Pk in an attempt to trap Shiga toxin in the gut failed to improve outcomes. The best way to prevent infection with Shiga toxin–producing E coli is by eating well-cooked meat (especially ground beef) and washed fruits and vegetables. Good hand-washing practices will decrease the (low) risk of person-to-person spread. Routine irradiation of meat and vaccine development offer the best hope for disease control but are currently not available options.

SUGGESTED READINGS

INTRODUCTION

This chapter discusses hereditary disorders of the glomerulus. These conditions can be categorized on the basis of the demonstrated or predicted function(s) of the affected proteins. It should be recognized that this is a somewhat arbitrary categorization that likely simplifies the cell-cell and cell-matrix interactions that produce disease phenotypes.

GLOMERULAR BASEMENT MEMBRANE DISORDERS

FAMILIAL HEMATURIAS: ALPORT SYNDROME COMPLEX

A substantial fraction of children with persistent hematuria (30–50% in several published series) have an inherited disorder of glomerular basement membranes (GBMs). Accurate diagnosis of familial hematuria is crucial for predicting prognosis and providing accurate reproductive counseling.

GENETICS

The Alport syndrome complex (ASC) encompasses genetic disorders of the collagen IV α345 network, the predominant collagen IV species in mature GBM. ASC accounts for most children and adolescents with familial glomerular hematuria. Genetic loci associated with ASC include the COL4A3 and COL4A4 genes on chromosome 2 and the COL4A5 gene on the X chromosome. ASC is genetically heterogeneous, with X-linked and autosomal forms. The majority of patients have X-linked ASC due to mutations in COL4A5, the gene encoding the α5 chain of type IV collagen [α5(IV)]. Autosomal ASC is caused by mutations in 1 or both alleles of COL4A3 or COL4A4. Heterozygous mutations in COL4A3 or COL4A4 are inherited in a dominant fashion, while patients with mutations in both alleles present in a recessive manner. Patients with heterozygous mutations in COL4A3 or COL4A4 and isolated hematuria are commonly classified as having “thin basement membrane nephropathy,” although this term describes a renal pathologic finding rather than a discrete disease entity.

CLINICAL FEATURES

Persistent microscopic hematuria is the hallmark of ASC, occurring in 100% of males and 95% of females with X-linked disease and all patients with autosomal recessive ASC. About 50% of individuals with heterozygous mutations in COL4A3 or COL4A4 are asymptomatic and are considered carriers of autosomal ASC. Episodic gross hematuria is common. Since only 10% to 15% of children with X-linked ASC have de novo mutations, most will have a parent with hematuria. However, normal parental urinalyses does not exclude a diagnosis of X-linked ASC. Each parent of a child with autosomal recessive ASC is heterozygous for a mutation in COL4A3 or COL4A4. Since about 50% of these parents are asymptomatic carriers, hematuria may be found in both, 1, or neither parent of a child with autosomal ASC.

Children with ASC frequently present with gross or microscopic hematuria similar to children with acquired forms of hematuria. A family history of kidney failure or deafness in a child with hematuria should suggest a possible diagnosis of ASC. Sensorineural deafness develops in 80% of males with X-linked ASC and most patients with autosomal recessive ASC, but deafness is unusual in those with heterozygous mutations in COL4A3 or COL4A4. The onset of measurable hearing deficits typically occurs in late childhood. The hearing defect is bilateral and initially involves high-frequency wavelengths, gradually extending into conversational speech over time. About 40% of males with X-linked ASC exhibit ocular anomalies, including maculopathy, anterior lenticonus, posterior polymorphous dystrophy, and recurrent corneal erosions, but the appearance of these anomalies is typically delayed until adolescence or young adulthood. The maculopathy and anterior lenticonus also occur in patients with autosomal recessive ASC.

Rarely, X-linked ASC is associated with smooth muscle tumors (leiomyomata) of the esophagus, tracheobronchial tree, and, in females, the external genitalia. The Alport-diffuse leiomyomatosis complex is a contiguous gene syndrome resulting from deletions involving the adjacent COL4A5 and COL4A6 genes on the X chromosome.

Kidney function is usually well preserved during the first decade of life in males with X-linked ASC. Overt proteinuria frequently appears during late childhood or early adolescence, followed by development of hypertension and renal insufficiency. About 50% of males with X-linked ASC reach end-stage renal disease (ESRD) by age 25, and 90% reach ESRD by age 40. COL4A5 genotype has a strong influence on timing of ESRD, with deletions and truncating mutations associated with more rapid progression to ESRD. The probability of ESRD in females with X-linked ASC is about 15% by age 45 and 30% by age 60. Subjects with autosomal recessive ASC follow a course similar to males with X-linked ASC. Renal outcomes in patients with heterozygous mutations in COL4A3 or COL4A4 range from normal kidney function throughout life to chronic proteinuric kidney disease to ESRD in adulthood.

RENAL PATHOLOGY

The earliest histologic abnormality in patients with ASC is thinning of the GBM as seen by electron microscopy. GBM thinning may be a static abnormality in females with X-linked ASC and in patients with heterozygous mutations in COL4A3 or COL4A4. In males with X-linked ASC and patients with autosomal recessive ASC, the GBM undergoes progressive thickening associated with fragmentation of the lamina densa into multiple lamellae. These changes typically first appear during childhood, and the extent of GBM displaying these alterations increases with age. In females with X-linked ASC and in patients with heterozygous mutations in COL4A3 or COL4A4, the extent of GBM thickening ranges from focal to diffuse, and the impact of aging on GBM thickening is variable.

Light microscopy typically shows no abnormalities or mild proliferation of glomerular mesangial cells in children with ASC. With time, males with X-linked ASC and patients with autosomal recessive ASC develop focal and global glomerulosclerosis, tubular atrophy, and interstitial fibrosis. Light microscopic changes in females with X-linked ASC and patients with heterozygous mutations in COL4A3 or COL4A4 are variable and also impacted by age at biopsy.

Immunostaining of kidney biopsy specimens using monospecific antibodies shows diagnostic abnormalities in expression of α3(IV), α4(IV), and α5(IV) chains in many ASC patients with X-linked or autosomal recessive disease. Abnormal expression of the α5(IV) chain in epidermal basement membranes occurs frequently in patients with X-linked ASC, allowing diagnosis by skin biopsy. Collagen IV expression in the kidney and skin is normal in patients with heterozygous mutations in COL4A3 or COL4A4. Molecular analysis of the COL4A3, COL4A4, and COL4A5 genes by next-generation sequencing is widely available in the United States and Europe.

TREATMENT

Angiotensin-converting enzyme inhibition slows the progression of renal disease in transgenic mice with ASC and appears to exert a similar effect in patients. Treatment is indicated in patients with proteinuria and ideally is initiated while renal function is still normal. Patients with ASC, including symptomatic patients with heterozygous mutations in COL4A3 or COL4A4, need regular monitoring of urine protein excretion to facilitate early institution of treatment. Transplant outcomes in ASC are generally excellent, although a small percentage of patients develop severe crescentic glomerulonephritis of the allograft mediated by antibodies to collagen IV α5 or α3 chains.

PIERSON SYNDROME

Pierson syndrome is a recessive disorder consisting of congenital nephrotic syndrome associated with ocular abnormalities (buphthalmos, microcoria, cataracts) and hypotonia. Pierson syndrome arises from mutations in the LAMB2 gene, which encodes the laminin β2 chain.

PODOCYTE DISORDERS

DISORDERS OF THE SLIT DIAPHRAGM AND ACTIN CYTOSKELETON

CONGENITAL NEPHROTIC SYNDROME, FINNISH TYPE (CNF)

CNF is the predominant cause of massive proteinuria in the first month of life. Proteinuria begins in utero, manifested by elevated amniotic fluid α-fetoprotein levels, and postnatal edema. Untreated, the disease frequently results in early death due to malnutrition, overwhelming infection, and thrombosis. With appropriate therapy, affected children can survive infancy and undergo renal transplantation, with excellent patient and graft survival rates.

GENETICS

CNF is a recessive disorder resulting from homozygous or compound heterozygous mutations in the NPHS1 gene on chromosome 19, which encodes the podocyte slit diaphragm protein nephrin. Nephrin is a cell-surface protein with a single transmembrane domain and 8 extracellular Ig-like domains that is specifically expressed at the slit diaphragms between podocyte foot processes. Nephrin molecules projecting from adjacent foot processes appear to interact to define pores in the slit diaphragm. Nephrin knock-out mice develop massive proteinuria associated with absence of podocyte foot processes and slit diaphragms.

CLINICAL FEATURES

Infants with CNF typically present within the first month of life with edema, hypoalbuminemia and proteinuria. Placental enlargement is common. Glomerular filtration rate (GFR) is typically normal for at least the first 6 months of life.

RENAL PATHOLOGY

Early histologic findings include mesangial hypercellularity and matrix expansion, dilatation of Bowman’s spaces and renal tubules, and podocyte foot process effacement. Later biopsy may show glomerulosclerosis, tubular atrophy, and interstitial fibrosis. Molecular analysis of NPHS1 is commercially available, making renal biopsy unnecessary for diagnosis in most infants with congenital nephrotic syndrome.

TREATMENT

With appropriate therapy, infants with CNF should grow sufficiently to undergo renal transplantation. Management is complex, involving aggressive nutritional support, thyroid supplementation, prompt recognition and treatment of bacterial infection, and early identification and therapy of thrombotic events. Once the child reaches 7 to 8 kg in weight, bilateral nephrectomies are performed, followed by 6 to 8 weeks of hemodialysis to allow optimal nutrition and normalization of plasma proteins in preparation for renal transplantation.

AUTOSOMAL RECESSIVE STEROID-RESISTANT NEPHROTIC SYNDROME

Genetic studies of families exhibiting recessive transmission of steroid-resistant nephrotic syndrome (SRNS) have implicated mutations in a variety of proteins critical to the normal functioning of podocytes and associated slit diaphragms. Among these proteins are podocin (NPHS2), phospholipase C epsilon-1 (PLCE1 or NPHS3), myosin 1E (MYOE1), aarf domain-containing kinase 4 (ADCK4), intraflagellar transport protein IFT139 (TTC21B), Crumbs homolog 2 protein (CRB2), CD2-associated protein (CD2AP), and receptor-type tyrosine protein phosphatase O (PTPRO). Age at onset of proteinuria ranges from early infancy to adolescence, and renal histologic findings include minimal changes, diffuse mesangial sclerosis, and focal segmental glomerulosclerosis. Autosomal recessive SRNS frequently progresses to ESRD, is typically refractory to calcineurin inhibition, and has a low risk of recurrence after kidney transplantation in comparison to nongenetic SRNS.

AUTOSOMAL DOMINANT STEROID-RESISTANT NEPHROTIC SYNDROME

SRNS can also be transmitted as an autosomal dominant disorder with onset of proteinuria in childhood. Proteins implicated in autosomal dominant SRNS include α-actinin-4 (ACTN4), transient receptor potential channel 6 (TRPC6), inverted formin 2 (INF2), RHO GTPase-activating protein 24 (ARHGAP24), anilin (ANLN), and paired box protein 2 (PAX2).

MYH9 DISORDERS: EPSTEIN AND FECHTNER SYNDROMES

Epstein and Fechtner syndromes are autosomal dominant disorders characterized by hematuria, progression to chronic kidney disease/ESRD, and sensorineural deafness, but patients also exhibit thrombocytopenia and giant platelets. Fechtner patients exhibit granulocyte cytoplasmic inclusions (Döhle-like bodies). These conditions, along with 2 other genetic conditions featuring giant platelets, Sebastian syndrome and May-Hegglin anomaly, and nonsyndromic hereditary deafness DFNA17, result from heterozygous mutations in the gene MYH9, which encodes nonmuscle myosin heavy chain IIA (NMMHC-IIA). NMMHC-IIA is a component of the podocyte cytoskeleton.

DISORDERS OF GENE REGULATION

DENYS-DRASH SYNDROME (DDS) AND DIFFUSE MESANGIAL SCLEROSIS

DDS comprises the triad of XY gonadal dysgenesis with ambiguous genitalia, Wilms tumor, and diffuse mesangial sclerosis (DMS). Children with DDS may exhibit DMS and genital abnormalities without Wilms tumor or DMS and Wilms tumor without genital abnormalities. DMS can also occur as an isolated condition.

GENETICS

DDS and some cases of isolated DMS arise from heterozygous mutations in WT1, a 10-exon gene that encodes the transcription factor WT1, or Wilms tumor suppressor. The WT1 protein has a predicted molecular weight of 47 to 49 kDa, according to the presence or absence of 2 alternatively spliced exons. WT1-expressing cells can generate 4 WT1 transcripts, carrying both, 1, or none of the alternative exons. The predominant WT1 transcript contains both alternative exons. The WT1 protein contains an NH3-terminal domain that is essential for transcriptional regulation and also mediates self-association of WT1. The COO-terminal domain consists of 4 zinc finger motifs that mediate DNA binding.

WT1 mRNA is expressed during nephrogenesis in the condensed mesenchyme, renal vesicle, and podocytes. Podocytes of adult kidneys express WT1 mRNA at low levels. WT1 mRNA expression patterns suggest that it may mediate differentiation of nephrogenic mesenchyme to glomerular epithelium and help maintain the epithelial phenotype of the podocyte. Loss of WT1 function is predicted to allow uncontrolled growth of metanephric blastema cells, resulting in Wilms tumor, and aberrant podocyte differentiation. WT1 mRNA is also expressed in primordial tissue of the gonadal ridge, ovarian follicular granulosa cells, and Sertoli cells; normal maturation of the gonads may also depend on the functional integrity of WT1.

Most DDS patients have a mutation in exon 9 of the WT1 gene affecting the third zinc finger. Persistence of nephrogenic rests (undifferentiated mesenchyme) in DDS kidneys likely results from the germline WT1 mutation. A Wilms tumor will then result if a somatic mutation (“second hit”) inactivates the normal WT1 allele within a nephrogenic rest. In contrast to Wilms tumor, DMS and gonadal dysgenesis in DDS do not occur by a recessive mechanism, since DDS patients are heterozygous for their WT1 mutations. DDS mutations behave in a dominant-negative fashion (ie, the mutant protein interferes with the function of the product of the wild-type allele).

CLINICAL FEATURES

In 46,XY DDS patients, genital abnormalities range from hypospadias with cryptorchidism to clitoral enlargement with labial fusion. External genitalia usually appear normal in 46,XX patients. Anomalies of the internal reproductive organs, including atrophy of the vagina and uterus, streak ovaries, or dysgenetic testes, may occur with either karyotype. The abnormal gonads of 46,XY patients may evolve into gonadoblastoma. DDS patients are more likely to have bilateral Wilms tumor and are younger when their tumors are identified than children with sporadic Wilms tumor.

Proteinuria appears during infancy and is frequently of nephrotic range. Hematuria and hypertension are common. Most patients progress to ESRD by 3 years of age.

RENAL PATHOLOGY

The glomerular changes of DDS-associated DMS depend on the cortical depth of the glomerulus (ie, its degree of maturation) and the timing of sampling. The pathognomonic lesion is observed in middle cortical glomeruli, which show mesangial expansion with obliteration of capillaries, producing a solidified mass covered by a corona of hypertrophied podocytes.

TREATMENT

DDS should be suspected in children with proteinuria associated with Wilms tumor and/or ambiguous genitalia and confirmed by renal biopsy and karyotyping. The presence of DMS on kidney biopsy should prompt suspicion of DDS. Evaluation of DDS patients includes screening for WT1 mutations where available, imaging of the internal reproductive organs, and, at some point, examination of gonadal histology.

Because of the high risk of Wilms tumor, any remaining renal tissue should be removed once ESRD occurs (some will have had previous unilateral nephrectomy for Wilms tumor). Some clinicians favor prophylactic bilateral nephrectomies, or removal of the remaining kidney after excision of a unilateral Wilms, once the diagnosis of DDS has been made, and then proceeding to transplant. Others prefer to monitor such patients for development of Wilms tumor in any remaining renal tissue by frequent ultrasound examination, moving to nephrectomy and transplant when ESRD supervenes. Gonadal tissue should be excised in 46,XY DDS patients because of the risk of gonadoblastoma. Gonadectomy is not necessary in those with an XX karyotype.

FRASIER SYNDROME

Frasier syndrome is another WT1-associated disorder characterized by male pseudohermaphroditism and renal failure. Patients have normal female external genitalia but a 46,XY karyotype and often go unrecognized until evaluated for primary amenorrhea. Proteinuria appears during childhood, progressing to nephrotic syndrome and chronic renal failure during adolescence. Renal pathology consists of focal segmental glomerulosclerosis. Gonadoblastoma is a frequent complication, but the risk of Wilms tumor appears to be low. Mutations in the donor splice site in intron 9 have been reported in patients with Frasier syndrome.

NAIL-PATELLA SYNDROME (NPS)

NPS is an autosomal dominant disorder consisting of hypoplasia or absence of the patellae, dystrophic nails, dysplasia of the elbows, and iliac horns. About 50% of NPS patients exhibit renal disease, including hematuria, proteinuria, and hypertension. The risk of progression to ESRD is about 10%. While light and immunofluorescence microscopy show nonspecific changes, electron microscopy reveals multiple irregular lucencies in GBM, resulting in a moth-eaten appearance. Staining with phosphotungstic acid may reveal cross-banded collagen fibrils within these lucent areas; by immunostaining, these fibrils appear to represent type III collagen. Heterozygous mutations in the LIM homeodomain transcription factor gene LMX1B have been found in NPS patients. These mutations appear to result in haploinsufficiency of LMX1B protein.

STORAGE DISORDERS

FABRY DISEASE

Fabry disease is an X-linked disorder caused by deficiency of the lysosomal hydrolase enzyme α-galactosidase A (α-Gal A), resulting in accumulation of glycosphingolipid in various tissues, including the kidney.

GENETICS

Almost all unrelated families with Fabry disease have unique mutations in the α-Gal A gene. The majority of the mutations are single base substitutions. Deficiency of α-Gal A leads to progressive intracellular accumulation of neutral glycosphingolipids, particularly those with a-galactosyl moieties, the most abundant of which is globotriaosylceramide (Gb3). All tissues except red blood cells accumulate Gb3, with the highest concentrations found in the kidney.

CLINICAL FEATURES

Since Fabry disease is X-linked, severe disease occurs in hemizygous males, whereas heterozygous females exhibit a variable but typically less severe course. Affected males become symptomatic in childhood and early adolescence with paresthesias and pain in the hands and feet due to peripheral neuropathy. The classic skin lesion, angiokeratoma corporis diffusum universale, appears in adolescence and consists of 1- to 3-mm, cherry-red to black macules initially involving the scrotal area and trunk. Other skin manifestations include palmar erythema, splinter hemorrhages, and conjunctival or buccal mucosal telangiectasias. Later complications of the disease, such as cerebral infarction, dementia, and coronary artery disease, are not seen in children and adolescents.

The earliest renal manifestation is a concentrating defect. Proteinuria is first detected in young adulthood, and some patients exhibit hematuria. ESRD typically occurs in the third to sixth decade.

PATHOLOGY

Gb3 accumulation affects most renal cells. Podocytes are most prominently affected and appear to contain multiple cytoplasmic vacuoles by light microscopy. The characteristic lesion of Fabry disease is seen on electron microscopy and consists of stacks or whorls of dense lysosomal inclusions.

DIAGNOSIS

Diagnosis of the hemizygote patient can usually be made on clinical grounds. The diagnosis should be confirmed by demonstrating decreased to absent α-Gal A activity in serum, leukocytes, cultured skin fibroblasts, or biopsy samples. Molecular testing is also available.

TREATMENT

Enzyme replacement therapy using recombinant human α-Gal A (agalsidase) offers the potential for delaying or preventing ESRD and other complications of Fabry disease. Symptomatic children with Fabry disease should receive enzyme replacement, while asymptomatic children should be closely monitored for development of symptoms or laboratory changes indicating renal, cardiac, neurologic, or gastrointestinal involvement.

OTHER STORAGE DISORDERS ASSOCIATED WITH GLOMERULAR DYSFUNCTION

Glomerular dysfunction may be detected in patients with various inherited storage diseases, due to accumulation of the storage material in glomerular cells, mesangium, or GBM. Deficiency of the enzyme lecithin-cholesterol acyltransferase (LCAT) prevents esterification of plasma cholesterol and thus interferes with reverse cholesterol transport from tissues to the liver. The accumulation of free cholesterol in renal cells results in proteinuria that frequently progresses to ESRD. Ultrastructural findings are distinctive, consisting of numerous, irregular lucencies in the mesangium and GBM that contain dense granular or membrane-like structures.

Sialidosis results from deficiency of sialidase, which removes terminal sialic acid residues from sialyl-oligosaccharides released during glycoprotein degradation, leading to accumulation of sialic acid–rich material in various tissues, including kidney. Renal involvement, manifested by proteinuria, occurs in type II, or dysmorphic-type, sialidosis. Light and electron microscopy of renal biopsy specimens shows massively swollen podocytes due to the presence of innumerable thin-walled cytoplasmic vesicles. Accumulation of storage material in podocytes has also been described in I-cell disease and in Hurler syndrome.

Although renal disease is not unusual in older patients with glycogen storage disease type 1 (GSD1) due to glucose-6-phosphatase deficiency, it is not clear that the pathologic changes in the kidney arise from storage of glycogen. It has been proposed that proteinuria and glomerulosclerosis in patients with GSD1 result from hyperfiltration injury.

SUGGESTED READINGS