Is it Full Speed Ahead with Postexposure Prophylaxis After Nonoccupational HIV Exposure?

abstract & commentary

Source: Lurie P, et al. Postexposure prophylaxis after nonoccupational HIV exposure. Clinical, ethical and policy considerations. JAMA 1998; 280:1769-1773.

This special communication to JAMA about postexposure prophylaxis (PEP) after nonoccupational exposure is bound to stir controversy. Although the title of the article implies that what will be presented are only considerations, this paper is a set of recommendations that, in fact, readers may find useful.

Lurie and colleagues begin with a review of PEP for occupational exposures. The risk for such exposures is fairly well known. After exposure to a known HIV-positive source involving body fluid, the risk of transmission is about 0.25%, but increases with exposure to large-bore needles or delivery of larger volumes of body fluid. Mucous membrane exposures are less risky (about 0.09%).

The CDC has produced its most recent guidelines for PEP in the occupational setting, which now include a more helpful algorithm to assist decision making.1 In the hospital, the HIV status of the source patient is often known or can be obtained with rapid HIV testing that determines the HIV serostatus within several hours and at least helps modify the decision about duration of therapy.

The risk of seroconversion in nonoccupational exposures is generally not known. Nonetheless, several principles are key to understanding the approach to persons seeking PEP for their nonoccupational exposures. There are three important factors to consider: 1) frequency of exposure; 2) the probability that the source patient is HIV positive; and 3) the probability of transmitting HIV if the source patient is infected.

Exposure is considered to be either sporadic or continuing. Those exposures that are continuing (e.g., repeated sexual exposures), particularly if they occur more than once a month, require ongoing PEP. Regarding the serostatus of the source, only occasionally is that information available. Normally, the physician will have to estimate the likelihood that the source is infected from estimates of seroprevalence based on geographic location or databases gathered by the CDC in their National HIV Seroprevalence Surveys. The probability of transmission from an HIV-positive source for men who have single-encounter sex with men may approach 0.5%, which drops to 0.1% for exposures involving penile-vaginal intercourse. Other factors that may increase risk include a higher titre of virus, mucosal trauma, and concomitant genital tract infection.

Three case studies are presented for illustrative purposes. One involves an injection drug user who is HIV-negative and uses his own needles but has just shared a needle in New York City. It turns out the geographic location of the exposure would prompt PEP. The second case involves continuing exposure through sex with source status being unknown. Since there are few data from clinical trials, PEP cannot be recommended, but the patient should be referred for risk-reduction services. The last case involves sporadic exposure through sex for a college man who attends school in Nebraska. He and his regular partner were HIV-negative but last night’s exposure involved another female student whose serostatus was unknown. Since the seroprevalence among college students is about 0.2%, the risk of transmission should be several fold lower than occupational mucous membrane exposure. The recommendation here is not to give PEP; there are, however, some contingencies. If the patient insists on PEP, he and his partner should be tested for HIV infection again and PEP stopped if both he and his partner are negative.

Other examples of sporadic heterosexual vaginal exposure produce more difficult decisions since they fall under medium probability. If the source is positive, PEP is appropriate. If the source is negative, the physician may have to weigh other factors such as the incidence of bleeding, trauma, concomitant genital infection, and geographic location of the source.

In instances of rape, the clinician may calculate that the risk is lower than in other incidences of sexual exposure but, because of psychological trauma and anxiety over pregnancy, PEP may be appropriate.

Lurie et al review the downside of PEP. We certainly do not want to send out a "double message" (on the one hand saying abstinence is best and safe sex is mandatory but if you slip up we will give you PEP). Lurie et al feel that this concept of "disinhibition" is not strong enough at this time to negate the value of PEP in the appropriate settings. Lurie et al emphasize that although some clinicians may be reluctant to give PEP for exposures that they deem irresponsible, clinicians still have an ethical duty to administer PEP when indicated. To quote Lurie et al, "Clinically, it is irrelevant whether the patient is an "innocent victim" or a "knowing participant." Their example of giving surgical care to an intoxicated traffic victim serves their argument.

Lurie et al also discuss the issues of timely access to PEP, difficulties with adherence and development of antiviral drug resistance, and cost considerations. Special sites for distribution of PEP may be necessary in some geographic locations. Adverse effects in my experience lead to a significant number of patients on PEP discontinuing their medication and adherence should be monitored in some way.

Regarding cost, Lurie et al calculate that the cost of $13,650 per year-of-life saved with PEP should make insurance companies willing to pay for PEP involving sporadic high-risk exposures. Consider some of their other cost estimates. For zidovudine (AZT) prophylaxis, the cost of preventing one HIV infection is $136,500. If a protease inhibitor is added to the regimen, the cost increases to $199,000. If the source patient’s HIV status is unknown, the cost per year-of-life saved increases.

Comment by Joseph F. John, MD

JAMA has provided us with a true service in publishing this perspective. Lurie et al at UCSF have been involved with this issue for several years, their clinical setting being at one of the epicenters of the American HIV epidemic. Lurie et al have thrown out a challenge to clinicians who will be confronted with patients seeking PEP after nonoccupational exposures.

Recommendations for Nonoccupational Postexposure Prophylaxis (PEP) ________________
Exposure Frequency*
Probability of Transmission†
Low Medium  High 
Sporadic Inform, but do not recommend  Consider Recommend‡
Continuing Inform but do not recommend; consider referral to risk reduction services Inform but do not recommend; refer to risk reduction services Inform but do not recommend; refer to risk reduction services 
*Exposure frequency: continuing, more than once a month, and sporadic, less than once a month. 
†Probability of transmission (the product of the probability the source is human immunodeficiency virus [HIV] positive and the probability of transmission for that act, if the source is HIV positive): high, unprotected anal intercourse or sharing of syringes with someone known to have HIV infection or of unknown HIV status but from a population with high HIV prevalence (about 30% or greater); medium, unprotected anal intercourse or sharing of syringes with someone of unknown HIV status from an intermediate-risk population (about 10-30% prevalence); unprotected penile-vaginal sex with known HIV-positive partner; and low, other risks such as oral sex, penile-vaginal sex, unprotected anal intercourse, or sharing of syringes with a partner of unknown serostatus in low-prevalence population. 
‡Recommendations: inform, explain the concept of PEP, review benefits and risks, and explain that the likelihood of infection is low and that PEP is not advised; consider, inform patient and tailor decision to individual patient’s desires and specific clinical situation; and recommend, inform patient and make clear that expected benefits outweigh the risks in his or her case and that PEP is, therefore, advisable. 
Reprinted with permission from: JAMA 1998;280:1769-1773. 

Exposures to blood and body fluids in the hospital have been a challenge to employee health services and infectious diseases sections across the country. Better guidelines and experience have allowed us to feel more comfortable about managing these types of exposures. Still, they require time, a great deal of organization, and full support from the hospital administration. After many years of overseeing services for PEP at our combined institutions, we succeed more times than not in successfully caring for healthcare workers with blood and body exposures.

PEP for nonoccupational exposures seems a horse of a different color, requiring a new mindset and new epidemiologic tools. Surely, those in charge of making decisions about PEP for occupational exposure will face some tough decisions. It will be interesting to see how sexually active populations gauge the value of PEP in the face of the degree of their risk taking.

The biggest public health risk in applying widespread PEP for nonoccupational exposures is probably the spread of resistant HIV. Although there is now documentation of transmission of resistant virus, it is not known if a short duration of exposure such as that with PEP can select a resistant quasispecies of HIV. Lurie et al make no definitive statements that PEP should in fact be HAART, but it would certainly be hard to offer patients less. The pressure on the clinician in such situations probably precludes any regimen other than HAART. One advantage of HAART is the likelihood that the combination would eradicate what should be a small inoculum of virus, thus, further limiting the emergence of resistant HIV.

It will be interesting to watch the evolution of these recommendations. At an earlier time when we had only reverse transcriptase inhibitors to offer healthcare workers with occupational exposure, few were eager to take PEP. Now that we have better scientific estimates of risk after exposure and much better antiretroviral regimens, clinicians will face patients that desire or even demand PEP. Physicians will need to weigh many ethical, social, and virologic variables in what may become a common outpatient dilemma.

Implementation of such a program has already been reported by the group from the Department of Public Health in San Francisco at the 1998 12th World AIDS Conference in Geneva. That health department became the first to offer medication, in-depth counseling, and HIV antibody testing to individuals who experienced non-occupational exposure via sexual exposure or needle sharing.2

To quote the abstract of the article, "Initially, there was a vociferous negative response by community groups about societal disinhibition and the erosion of established prevention messages, contrasted with resistance by HIV activists to any exclusion criteria established by research protocols."2 Yet, through vigorous radio and other media awareness programs, the project gradually met with more acceptance. The experience of this ongoing San Francisco experience should become increasingly valuable for other cities planning nonoccupational PEP.

We clinicians all should have a zero tolerance for HIV risk taking. Nevertheless, PEP for nonoccupational exposures is becoming a reality that will put new and unique pressures on HIV care givers.


1. Public health service guidelines for the management of healthcare worker exposures to HIV and recommendations for postexposure prophylaxis. MMWR Morb Mortal Wkly Rep 1998;47[No.RR-7]:1-34.

2. Kegebein VR, et al. 12th World AIDS Conference, Geneva, June 28-July 3, 1998. Abstract 251.