Chapter 55. Preventing HIV Infection: Postexposure Prophylaxis

Until an effective HIV vaccine is in widespread use, taming the HIV epidemic will require the prevention of exposure to the virus. Activities such as on-site rapid HIV testing with early treatment of infected persons, universal precaution training, safer sex education, and needle exchange programs all are important aspects of an effective preventative strategy. However, in cases where exposure has already occurred, it is possible to decrease the risk of transmission of HIV infection through the use of postexposure prophylaxis (PEP).

The best-documented success of HIV PEP has been through the prevention of maternal to child transmission. In the National Institutes of Health, U.S. Department of Health and Human Services funded Pediatric AIDS Clinical Trials Group 076 trial, a three-arm intervention with zidovudine (prenatal, intrapartum, and postpartum) decreased the risk of neonatal HIV infection from 21% to 8%.1 Subsequent studies indicated that a significant part of the decrease in the transmission risk was based on the receipt of postnatal medication. In situations where mothers did not receive any prenatal antiretroviral therapy, the provision of antiretroviral medication to the HIV-exposed newborns within 48 hours of life decreased the risk of transmission by up to 50%.2

HIV-related PEP is now used routinely in hospitals following occupational exposure and in emergency departments following accidental or unprotected exposure to body fluids—termed nonoccupational exposure. It is important for all health care providers to be aware of current PEP recommendations, and to know how to rapidly access such information and/or consultation. This chapter will discuss HIV PEP as it relates to occupational and nonoccupational exposure to HIV.

Through June 2000, the Centers for Disease Control of the U.S. Department of Health and Human Services reported 56 cases of HIV infection in health care workers (HCWs) documented to follow occupational exposure, with an additional 138 possible cases.3 The risk of occupational exposure to potentially infectious body fluid is highest among surgical staff, followed by nonsurgical nurses (especially those in the Emergency Department), phlebotomists, and resident physicians. General surgeons report an average of 4 intraoperative percutaneous injuries per year.4 Among pediatric residents, 13–70% report at least one needle stick exposure per year.5,6

The risk of transmission of HIV from a single percutaneous injury from a contaminated needle is approximately 1 in 300 (0.3%) (Table 55–1).7 An increased risk of transmission has been associated with exposures to patients with late stage HIV infection/AIDS, hollow bore needles, a deep puncture wound, and blood observed in the involved needle.8,9 The risk of transmission of HIV from nonpercutaneous exposures is much less. For exposure to bloody fluid across mucus membranes, the risk is estimated to be less than 1/1000 (0.1%), and even less when the exposure is across intact skin.

The interest in using antiretroviral agents in an attempt to abort HIV infection following exposure arose from animal studies. In these studies, there was some evidence of protection through the administration of zidovudine if given immediately after exposure to retroviral infections.10,11

The Centers for Disease Control of the U.S. Department of Health and Human Services performed a retrospective case-control study of HCWs with occupational exposure to known HIV-infected blood. Many of the HCWs had received zidovudine for PEP. The results showed an 81% decrease in the risk of transmission if the employee received zidovudine.9 With these results, and the proven success of zidovudine and other agents in maternal to child transmission prevention trials, the use of PEP for high-risk occupational exposures has become the standard of care.

When evaluating an employee with an exposure to blood or other potentially infectious body fluid, the immediate task is an assessment of the risk involved with the exposure (Table 55–1). The exposure must be categorized as to its potential for transmission of HIV. As noted, deep injury from visibly contaminated blood carries the highest risk of transmission of HIV. In contrast, mucus membrane exposure to nonbloody body fluids carries a much lower risk, and exposure across intact skin does not carry a risk of HIV-infection.

Of equal importance is whether or not the source patient is known to be HIV-infected. If the patient's serostatus is unknown, efforts should be made to immediately obtain HIV testing, consistent with existing local policies on testing and consent. The use of HIV rapid assays for testing of the source patient can result in cost savings (through decreased medications dispensed to exposed HCWs) and alleviating emotional stress, as it allows for immediate discontinuation of PEP should the source patient test negative. For exposures where the patient's status is unknown, and testing is not possible, an assessment must be made of the potential HIV risk status of the source patient, based on local HIV seroprevalence.

Epidemiology

In urban emergency departments, the most common scenario for consideration of nPEP involves a patient presenting for evaluation following a high-risk sexual encounter, consensual or not. Again, the first issue is to assess the degree of risk, based on exposure characteristics (Table 55–1). In instances where one partner is known to be HIV infected, the risk of anal intercourse is higher than for vaginal intercourse. Receptive anal intercourse with an HIV-infected partner carries an estimated risk of HIV transmission of 0.3–0.5%. Vaginal intercourse carries a risk estimated at one-tenth that of anal sex (0.03–0.09%), with receptive partners more at risk than the insertive partner. Oral sex is associated with a minimal, but not zero, risk of transmission. Intercourse during menses, and traumatic intercourse, as in cases of sexual assault, carry an increased risk of transmission.

For the pediatrician, the consideration of nPEP is generally in relation to exposures from common childhood behaviors, such as biting between children, or puncture wounds from discarded needles found on playgrounds. Although in certain conditions viable virus can be recovered from 8% of syringes and needles up to 21 days later, the risk is quite low. To date there have been no cases of documented transmission of HIV from puncture wounds from discarded needles.

There are several cases of purported transmission of HIV through biting or other in-home exposures, but there are no well-documented cases of HIV transmission between children through biting. HIV is rarely isolated from the saliva, and there are inhibitory molecules in saliva that reduce infectivity, therefore the risk of transmitting HIV through biting is believed to be extremely low. The few cases of in-household nonsexual transmission of HIV have involved direct exposure to infected blood.

In a pediatric hospital setting, inadvertent exposure to breast milk may occur as well, through the inadvertent offering of stored breast milk to the wrong infant. A meta-analysis of prospective cohort studies estimated the risk of transmission of HIV-1 through breast-feeding in infancy as 16%.12 The risk increases with longer duration of breast-feeding, high maternal viral load, the presence of mastitis and mixed formula and breast-feeding.13 However, the risk from a single feeding of infected breast milk is quite low, estimated to be 1–4/100,000. In addition, in the United States most women are HIV-tested during pregnancy and, if HIV-infected, strongly instructed to avoid breast-feeding. The likelihood, therefore, that a breast-feeding woman is HIV-infected, and that a one time exposure to her milk will transmit HIV, is minimal.

In the Emergency Department, or the Pediatric office, the first step in assessing the need for nPEP is in evaluating the significance of the exposure, as the risk of transmission of HIV varies greatly based on the type of exposure (Table 55–1). The risk is highest following transfusion of HIV-infected blood, where up to 90% of recipients will convert to HIV positive status. Other common exposures carry a risk of infection ranging from 13% to 25%, for perinatal exposure in the absence of breast-feeding, to 0.01% (for receptive oral sexual encounters).

PEP is recommended for exposures with a significant risk of HIV transmission and should be started as soon as possible after the exposure (Tables 55–2 and 55–3). It must be stressed, however, that for occupational as well as nPEP this must be done in consultation with experts after a detailed analysis of the characteristics of each individual case. In occupational settings, procedures should be in place to allow for administration of the first dose within hours of exposure. If the mode of exposure is significant, initiation of PEP should not be delayed while the source patient is tested. From animal and case review studies, treatment started 48–72 hours after exposure is much less likely to be effective. If HIV testing of the source patient does occur postexposure, ideally by rapid assay methods, PEP therapy can be discontinued if the results are negative.

Employees should have a baseline HIV antibody drawn, along with hepatitis B surface antibody (to confirm immune status) and hepatitis C antibody (Figure 55–1). Treatment should not be delayed while waiting to obtain or receive the results of baseline blood tests.

Current guidelines recommend a two or three-drug regimen given for 4 weeks. There is no evidence that a three-drug regimen for PEP is more efficacious than a two-drug regimen. However, based on results from HIV-infected adult and pediatric treatment trials, three-drug regimens result in more rapid and prolonged suppression of viral replication than two-drug regimens. Theoretically, three-drug regimens might offer an increased chance of efficacy. The difficulty is that adherence to three-drug regimens is much more difficult than to two-drug regimens, related to the increased toxicity associated with the most common third agents employed. All antiretroviral agents have side effects, and there is a high rate of discontinuation associated with three-drug regimens, particularly those containing a protease inhibitor (PI). This dilemma, the potential for more pronounced side effects, has to be balanced with the potentially increased efficacy, and should be discussed with both the patient and parents.14,15

The choice of antiretroviral agents for PEP must be individualized, and is best done in consultation with an HIV treatment specialist. Emergency departments and hospital employee health offices must be able to dispense a starter supply of medications, based on usual prescribing patterns, but then the treatment regimen should be reviewed as soon as possible with treatment experts.

The selection of the PEP regimen is more complicated when the source patient or partner is known to be HIV-infected, and on treatment. Ideally, when selecting a treatment regimen for PEP, the source patient will be naïve to at least two of the agents used to treat the exposed person.

Most treatment experts recommend a minimum of two nucleoside reverse transcriptase (NRTI) agents, most commonly zidovudine and lamivudine, each given twice daily. For older children and adolescents, the two drugs come in a coformulated tablet (Combivir). For the younger-aged child, they are available in suspension form with a reasonably palatable taste. Alternative NRTI agents include stavudine and didanosine (although not given together, as a result of overlapping toxicities). Abacavir is not recommended for PEP because of its high rate of early hypersensitivity reactions.

Tenofovir is widely used for adults, and in animal studies had a high rate of success in experimental PEP trials.16 However, tenofovir is not yet approved for those younger than 18 years, and is only available in tablet form. For those older than 18 years, the combination of a once daily tablet coformulation of tenofovir and emtriva is one of the preferred regimens.

Prescribing physicians must balance the risk of transmission versus risk of side effects from therapy. At times enhanced potency of three-drug regimens is offset by the increased risk of side effects leading to patients discontinuing therapy prematurely. Some authors feel that the optimal strategy to modify the risk of transmission is to recommend a two-drug regimen for improved compliance and follow-up.

When constructing a three-drug regimen, the U.S. Public Health Service recommends adding a PI. Among the PIs, lopinavir/ritonavir is the preferred agent. However, its’ use is associated with significant gastrointestinal side effects. In addition, the pediatric liquid formulation is significantly unpalatable. Nelfinavir is a PI available as both a powder and tablet. In very young infants and children, however, it requires 3-times-a-day dosing. Atazanavir is a newly released PI, taken once daily, with minimal gastrointestinal side effects (with the exception of asymptomatic elevations of total bilirubin levels). Although available only in capsule formulation at this time, it is expected a powder formulation will be available by 2009.

One alternative to a PI-based three-drug regimen would be using the nonnucleoside reverse transcriptase inhibitor (NNRTI), efavirenz. Efavirenz has an easy dosing schedule, one tablet, once daily. For infants and children, the small capsules can be sprinkled on food. However, in primate drug trails, efavirenz was associated with a high rate of birth defects, so it should not be used in women of childbearing potential. Nevirapine, the other widely used NNRTI, has been associated with severe hepatic toxicity, including several cases of liver failure, when included in PEP regimens.

In recent years, many practitioners have used a three-drug nucleoside and nucleotide regimen, generally zidovudine, lamivudine and tenofovir. In 2001, the New York State Department of Health AIDS Institute issued new guidelines that listed this combination as the preferred regimen for PEP. The rationale for this approach included the efficacy noted in primate studies, the ease of administration and low incidence of side effects, compared to NRTI–PI or NRTI–NNRTI combinations. Despite the expectation that this would be a relatively well-tolerated regimen, one center reported that for the initial 65 patients prescribed the zidovudine/lamuvidine/tenofovir combination for nPEP or oPEP, 66% complained of nausea, and 25% of emesis.17 In contrast, a study from Australia found a better 28-day completion rate for nPEP (85%) using a tenofovir, stavudine and lamuvidine regimen, with a 23% incidence of nausea.18

PEP always follows stressful events, whether related to occupational or nonoccupational exposures. Follow-up measures should include provision or referral for psychological support services. Sexually active adolescents and adults must be cautioned about the potential for transmission of newly acquired HIV through sexual contact. Because of the teratogenic potential of PEP regimens, women of childbearing potential should use reliable birth control contraception during treatment.

Once PEP is prescribed, patients should be carefully monitored for side effects (Figure 55–1). More than 50% of those on PEP report side effects, with gastrointestinal symptoms and general fatigue the most common. Among HCWs prescribed oPEP, more than 20% fail to complete the 28-day regimen, most commonly from medication side effects.

During the 28-day PEP regimen, ideally the patient's blood counts and chemistries should be monitored at baseline, 2 weeks and 4 weeks.

Testing for HIV infection should be by antibody, and not by direct viral methods such as HIV DNA or RNA assays. The DNA and RNA assays have false positive rates of 1–2%, which generally exceeds the risk of infection in the exposed patient. Antibody testing should be performed at 6 weeks, 12 weeks, and 24 weeks postexposure. Although recent data from one study group suggests that virtually all infected patients will seroconvert to antibody positive status by 12 weeks following exposure,19 the present guidelines recommend testing at least through 24 weeks postexposure.20

It is important to realize that there have been failures of PEP both for occupational and nonoccupational exposure.21 Through 2005, there have been at least six cases of transmission of HIV among HCWs despite PEP. When following patients post exposure, providers must be attuned to symptoms of acute retroviral syndrome. Although nonspecific, symptoms noted in 70% or more of patients with the acute retroviral syndrome include fever, adenopathy, pharyngitis and rash.22 Should these occur, urgent consultation with an HIV treatment specialist is necessary for evaluation and management.

Among adolescents and adults receiving nPEP for sexual assault there is a very high rate of failure to complete the prescribed 28-day therapy. Of the 97 patients referred from two emergency departments in Boston for follow-up, only 37 (38%) made at least one follow-up visit, and only 16% finished 28 days of PEP. Almost half of the patients reported an adverse event from the medication.23 In a study form Vancouver, 71 of 258 sexual assault victims were prescribed nPEP regimens, but only eight patients completed a 4-week course of therapy and follow-up visits.24 The reasons for poor compliance with prescribed regimens and scheduled follow-up visits were multifactorial, but illustrate the difficulties in implementing successful nPEP programs.

Emergency departments and hospital employee health divisions must have established policies to provide immediate assessment, support, and treatment to employees and patients with potentially significant exposures to HIV. The treatment of HIV is a rapidly evolving area, even in this third decade of the epidemic. Providers must remain current with new treatment guidelines, and be able to access up-to-date changes. Online references are invaluable, as treatment options are always expanding (Box 55–1).20