Loading...
A 29-year-old emergency medicine resident was suturing a laceration on an HIV-infected patient when she scratched her hand through a latex glove with the solid-bore suture needle. The needle surface had a small amount of visible blood; on removing her glove, the resident saw a small scratch but did not experience any bleeding. After washing the wound with soap and water, she sought evaluation at the hospital’s occupational health office. During her work-up, she was found to be about 5 weeks pregnant.
The source patient was on stable treatment with his first regimen — tenofovir/FTC and ritonavir-boosted atazanavir. At the time of the exposure, his CD4 count was 550 cells/mm3, his viral load was undetectable, and he had no history of treatment failure or documented resistance.
What would you tell the resident about the risk for HIV infection from this exposure? Would you recommend postexposure prophylaxis (PEP)? If so, which regimen? If not, would you change your recommendation if the resident specifically requested PEP?
Decisions regarding PEP initiation must be made on a case-by-case basis after taking into account the risk associated with the exposure, the degree of immunosuppression in the source patient, and the risks and benefits of antiretroviral therapy (ART). According to PEP guidelines from the U.S. Public Health Service (MMWR Recomm Rep 2005; 54:1), the average risk for HIV acquisition after a percutaneous needlestick injury such as the one described here is about 0.3% (and after mucus membrane exposure, 0.09%). In a retrospective case-control study (in which AZT use decreased the risk for occupational HIV acquisition by 80%), the following factors were associated with increased risk: deep injury, a procedure involving a needle placed directly into an artery or vein of the source patient, visible blood on the tip of the needle before the injury, use of a hollow-bore needle, and a source patient with a terminal illness, likely reflecting the higher viral loads typically seen in people with advanced AIDS (N Engl J Med 1997; 337:1485). The resident described here had only one of these risk factors (visible blood on the needle tip). Her risk for HIV acquisition is certainly less than 0.3% but likely not zero, given the visible break in her skin and the remote possibility that the source patient transmitted HIV despite an undetectable viral load.
According to the Public Health Service guidelines, a relatively low-risk exposure such as this one should prompt immediate initiation of a two-drug PEP regimen. (A three-drug combination would be advised if the exposure were more severe or if the source patient had a viral load >1500 copies/mL or symptomatic HIV infection.) However, the decision to initiate PEP in this particular woman is complicated by her pregnancy. Data on the safety of antiretroviral use during pregnancy are both limited and conflicting with respect to specific outcomes, such as preterm delivery and mitochondrial dysfunction. However, overall data from the Antiretroviral Pregnancy Registry are reassuring: Among 3951 infants born to mothers with first-trimester exposure to antiretrovirals, the incidence of birth defects was 3.0%, similar to that seen in the general U.S. population.
Given the relatively low risk associated with this exposure and the concerns about ART during pregnancy, this resident could reasonably defer ART. However, if she were willing, I would recommend that she initiate a two-drug PEP regimen. Guidelines support the use of once-daily tenofovir/FTC (Truvada) or twice-daily AZT/3TC (Combivir). Although concerns exist about tenofovir and fetal bone development, data from the Antiretroviral Pregnancy Registry indicate a relatively low incidence of birth defects (2.6%) following first-trimester exposure to this drug. Regardless, I would favor AZT/3TC because we have more experience with it than with any other NRTI combination. Notably, the major side effects of AZT — headache, fatigue, and nausea — are not inconsequential during the first trimester of pregnancy. If the resident insisted on a third drug in the regimen, I would recommend a boosted PI rather than an NNRTI, because efavirenz is contraindicated during pregnancy, and nevirapine is associated with hepatotoxicity and rash in women with normal CD4-cell counts.
Finally, I would recommend HIV testing, a complete blood count, and a comprehensive metabolic panel at baseline, followed by HIV antibody testing at 6 weeks, 3 months, and 6 months. HIV viral-load testing is not necessary unless the resident begins to exhibit signs or symptoms of acute HIV infection.
Sustaining a percutaneous exposure to a pathogen such as HIV is difficult for any healthcare provider to endure, but, in the setting of a pregnancy, the challenges are even greater. One must consider not only the nature of the exposure and the characteristics of the source patient but also the safety of antiretrovirals during pregnancy. Ultimately, however, decisions regarding PEP must be made based largely on the needs and desires of the affected healthcare provider.
A percutaneous injury involving a suture needle, at a shallow depth and through a gloved hand, confers a lower risk for acquisition of bloodborne pathogens than does an injury involving a hollow-bore needle, a procedure in which the needle is placed directly into a vein or artery, a sharp that is visibly contaminated with the source patient’s blood, or deeper lacerations. Furthermore, risk is considered to be lower if the source patient has a viral load <1500 copies/mL than if he or she has a higher viral load, acute seroconversion, or symptomatic HIV disease. Although the source patient’s undetectable viral load in this case is reassuring, it does not translate into an absolute risk of zero. Plasma viral load is a measurement of only cell-free virus; even in the absence of detectable viremia, latently infected cells could still lead to HIV transmission.
For relatively low-risk exposures such as this one, the U.S. Public Health Service guidelines recommend a basic two-drug PEP regimen (MMWR Recomm Rep 2005; 54:1). Furthermore, the guidelines state that a known or suspected pregnancy in a healthcare provider does not preclude use of PEP and that PEP should not be denied solely on the basis of pregnancy. Teratogenic effects of antiretrovirals are of concern in this setting but have not been seen with AZT and 3TC, two agents that are recommended both for PEP and for treatment during pregnancy. The Antiretroviral Pregnancy Registry provides no evidence of an increased risk for birth defects among infants exposed to AZT or 3TC during the first trimester, compared with infants in the general U.S. population. Data from PACTG 076 also support the safety of AZT use during pregnancy (AIDS 1998; 12:1805). Nevertheless, avoiding these drugs during the first trimester is best when possible.
In this particular case, I would offer the resident PEP with AZT/3TC. This recommendation is obviously not without risk, and the tolerability could be very poor, which could hinder adherence. Studies have demonstrated a higher rate of adverse events in healthy individuals who take antiretrovirals as part of PEP than in those who do so as part of HIV treatment. In particular, AZT/3TC use might exacerbate any pregnancy-related nausea, vomiting, headaches, and fatigue that this woman is already experiencing. These issues must be discussed up front in great detail with the resident, and adverse effects must be managed aggressively, using additional medications that are safe to take during pregnancy.
Given the low risk of this exposure and concerns for the safety of the fetus, this resident might very well refuse PEP. I do not believe that this would be an unreasonable decision and would do my best to reassure her that the likelihood of HIV infection in this scenario is extremely low. I would, however, inform her fully about the data regarding AZT/3TC use during pregnancy, so that her decision is as fact-based as possible. Regardless of her decision, close follow-up, for both medical monitoring and emotional support, is essential.
How would you manage this patient? Join the discussion at our new online forum.