++
The initial management of an unstable patient with suspected meningitis must focus on assuring airway, breathing, and cardiovascular stability. Supplemental oxygen is always administered. (See Chapter 19 for management of shock). If signs of increased intracranial pressure develop, clinicians should elevate the head slightly and initiate controlled hyper-ventilation (target Paco2 between 30 and 35 mm Hg). Children unresponsive to initial therapy may benefit from the use of mannitol (0.25–1 g/kg). Associated seizures are controlled with rapid-acting benzodiazepines followed by an appropriate second-line agent (see Chapter 52). If hypoglycemic (defined as a blood glucose of <40 mg/dL), children require an intravenous glucose infusion and monitoring.
++
If the child with suspected meningitis is unstable, lumbar puncture should be delayed.3 Although the early administration of antibiotics may prevent recovery of the organism from CSF culture,4 appropriate antibiotics should be administered early. Of note, bacterial pathogens are isolated from blood culture in approximately half of children with bacterial meningitis and may help guide decisions about duration and type of parenteral antibiotics.
++
Stable patients with suspected meningitis should have prompt blood testing and diagnostic lumbar puncture, especially in children at higher risk of bacterial infections (e.g., neonates, immunocompromised children, and close contacts of a confirmed bacterial meningitis case). The initial laboratory evaluation should include a complete blood count (CBC) with differential,5 peripheral glucose, as well as CSF white blood cell (WBC) with differential, red blood cell count (RBC), glucose, protein, and Gram stain. In children without bacterial meningitis, CSF glucose should be approximately 60% of peripheral glucose.6 Cultures of blood and CSF should also be obtained.
++
Normal CSF parameters are age-related (Table 58-1).6–9 Traumatic lumbar punctures occur commonly, especially in the youngest children, which may complicate the interpretation of CSF cell counts. Correction formulas are designed to correct CSF WBC for the presence of CSF RBCs. Either the standard 500:1 or the actual peripheral RBC: WBC ratio has a can be applied. The introduction of peripheral blood also elevates the CSF protein by approximately 1.1 mg/dL per 1000 RBCs per mm.10
++
++
For children with a CSF pleocytosis, additional testing should be dictated by the clinical scenario. Enteroviral polymerase chain reaction (PCR) testing can confirm a viral infection and shorten duration of hospitalization and of parenteral antibiotics11 as bacterial meningitis would be very unlikely.12 Newer enteroviral tests can return results within a few hours and may guide ED decision-making for children with CSF pleocytosis.13 For the youngest infants, HSV PCR testing should be considered. In Lyme disease endemic areas such as the Northeast, Mid-Atlantic and upper Midwest, Lyme serology should be ordered. Due to low sensitivity, clinicians should not order CSF Lyme PCR.14
+++
Distinguishing Bacterial from Aseptic Meningitis
++
Bacterial cultures take several days to either identify a bacterial pathogen or to reliably exclude bacterial growth. Meningitis prediction models, such as the Bacterial Meningitis Score (Table 58-2), combine presenting signs and symptoms to calculate the risk of bacterial meningitis.15,16 Children with none of the six included high-risk predictors are at very low risk of bacterial meningitis. The Bacterial Meningitis Score has now been validated in several published studies with a high degree of diagnostic accuracy: sensitivity of 99.3% (95% confidence interval [CI] 98.7%–99.7%), specificity 62.1% (95% CI 60.5%–63.7%), and negative predictive value (NPV) 99.5% (95% CI 99.3%–99.9%).3
++
+++
Antibiotic Pretreatment
++
Antibiotic pretreatment prior to diagnostic lumbar puncture may render bacterial cultures falsely negative. The time to culture sterilization depends on bacterial pathogen with almost immediate sterilization for N. meningitidis, but several hours of culture positivity for S. pneumoniae or Group B streptococcus.5 In children with proven bacterial meningitis, antibiotic pretreatment lowered CSF protein and raised CSF glucose.3 Therefore, meningitis prediction rules should not be applied to children pretreated with antibiotics.
+++
Distinguishing Lyme from Aseptic Meningitis
++
In Lyme endemic areas such as Northeast, Mid-Atlantic and upper Midwest, children with CSF pleocytosis who are at low risk of bacterial meningitis may still have Lyme meningitis. The treatment for Lyme meningitis is 3 to 4 weeks of parenteral antibiotics.5,17 The “Rule of 7's” is a previously validated clinical decision rule to distinguish Lyme from septic arthritis (Table 58-3).18 Children with none of the three high-risk clinical predictors are at low risk of Lyme arthritis: sensitivity 96% (95% CI 90%–99%), specificity 41% (95% CI 36%–47%), and NPV 91% (95% 86%–94%).19 Application of a validated prediction rule for Lyme meningitis has the potential to decrease unnecessary hospital admissions and parenteral antibiotic treatment for low-risk patients while awaiting Lyme serology results.
++