Mark:* Oh, good, thanks for calling back. You know, I've lived in San Francisco for 20 years and I've been negative all this time. I never thought this would happen to me. O.K. [pause] I went out last night to this bar and I met this guy. One thing led to another and we went back to my place and ... I asked him to wear a condom but ... well, he didn't. So afterward of course I asked him if he was positive and he hesitated just a bit. He said he was negative but it was pretty uncomfortable and he found some excuse to leave right after that. I mean, maybe he was telling the truth but why wouldn't he wear a condom if he was negative? Do you think I should take some medicine?
* Mark is a fictionalized, composite character
The preceding fictional call is very typical of those we receive on the HIV postexposure prophylaxis (PEP) line.
My grandfather started practicing medicine in the 1920s before the days of specialists, before the days of HIV/AIDS, and well before the advent of highly active antiretroviral therapy (HAART). However, in some ways I think he would have had an easier time answering callers' questions than I do.
In my grandfather's day, doctors made decisions for their patients because it was widely accepted that doctors "knew better" than patients. Today, despite advances in scientific knowledge, people tend to doubt that health-care providers always have the right answer.
In many ways, this is appropriate. Certainly, skepticism on the part of patients should apply in the case of PEP after sexual or needle-sharing exposure, where scientific evidence is not available to permit physicians -- or anyone -- to know the "right" answer. Nonetheless, as director of PEP for the City of San Francisco, I have been asked many times to answer questions like the last one above. Although the answer has rarely been clear, a few principles, which are outlined in this article, help in guiding the caller to make a reasonable choice.
Since October 1997, more than 600 people in San Francisco have received antiretroviral medications within 72 hours of a nonoccupational exposure to HIV (i.e., outside of a health-care setting) in an attempt to reduce their risk of seroconversion. Most of the people who have called the PEP hotline have had similar stories to that told in the introduction. Callers have described exposure to HIV following broken condoms during sex, assaults, or needle-sharing during illicit drug use.
Yet despite a wealth of clinical experience, experts are still unable to evaluate adequately the efficacy of PEP after sexual exposure. The risk of HIV transmission varies according to the likelihood that the source individual is HIV positive and by the type of sexual act. The average likelihood of HIV transmission after a single receptive anal exposure from a known HIV positive insertive partner, the highest risk sexual exposure, is approximately one seroconversion for every 300 exposures (see table, below). In other words, if all of the people who called the PEP hotline were exposed a single time in this way, and if none of them took any anti-HIV drugs, only two or three people would be expected to seroconvert -- too few to estimate efficacy of treatment with any statistical power.
To evaluate adequately the efficacy of PEP after sexual exposure, a clinical trial would need to be conducted. The trial would need to enroll approximately 5,000 people who had had a single high-risk exposure, initiate anti-HIV drug treatment, and then compare results from that group with results from another group of 5,000 people -- people who had similar exposure but did not initiate medication. While such a trial cannot be set up in a prospective (forward-looking) manner, the Centers for Disease Control and Prevention (CDC) has funded a National Nonoccupational PEP Registry to shed some light on this very situation. For more information, call 877-HIV-1PEP or visit www.hivpepregistry.org.
When exposure to HIV occurs among health-care workers -- for instance, when a nurse administering an injection to a hospital inpatient is accidentally pricked by the needle-tip -- it is usually possible to ascertain the HIV status of the source person. The experience among most people with potential sexual exposure who have been seen through the San Francisco PEP project has been different; while the source person is usually found to be from a high-risk group, his or her HIV status is often unknown. This uncertainty makes the design of an efficacy study after sexual exposure even more challenging.
So where does knowing that the source person is potentially high risk but not knowing his or her HIV status leave the clinician? More importantly, where does it leave someone like Mark (see introduction, above) who is trying to stay negative but who may have been exposed?
The best evidence that taking antiretroviral medication like PEP may reduce the risk of seroconversion comes from a study done by D. Cardo and colleagues that was published in the November 20, 1997 issue of the New England Journal of Medicine. This study showed that health-care workers who took AZT (Retrovir) after exposure to HIV reduced the likelihood of seroconversion by 81%. In this case-control study (as opposed to a placebo-controlled trial, considered to be a stronger design), exposed individuals initiated medication an average of four hours after exposure.
In a nonoccupational PEP study conducted in San Francisco, over 400 subjects initiated medication an average of 33 hours after exposure. Not only is the time between possible HIV exposure and the initiation of drug therapy different in the two settings, but the viral response to anti-HIV medication after mucosal HIV exposure in a sexual setting may be different from the viral response after percutaneous (through-the-skin, as with a needle) exposure.
Another example is also relevant. In one study, antiretroviral medications have been shown to reduce the likelihood of transmission from HIV-infected mothers to their babies during childbirth when the newborn is given anti-HIV treatment only after birth. This is a form of PEP, in that the child was exposed to the virus in the womb and during labor and delivery, and after those exposures, received antiretroviral treatment that clearly had some protective effect. This is compelling scientific evidence that PEP can be effective in some circumstances.
However, is it reasonable to make a so-called leap of faith from evidence of benefit in the health-care and perinatal settings, and to apply the same logic to the sexual exposure setting? Could the use of antiretroviral medications after exposure cause more problems than solutions?
An example from HIV treatment history is worth remembering: the early (circa 1988) recommendation to provide AZT monotherapy (single-drug treatment) may have done as much harm as good. While in that stage of the epidemic many well-intentioned and, for the time, well-informed practitioners strongly recommended AZT monotherapy to people with HIV, everyone subsequently learned that that particular treatment did little to improve quality of life or longevity in the long run.
One of the main reasons AZT monotherapy conferred little benefit is simply that it was monotherapy, which, regardless of the agent in question, is now known to lead to drug resistance.The consequences of single-drug therapy, of course, are clearly in opposition to the benefit seen with triple-drug therapy, or highly active antiretroviral therapy (HAART), in which it is clear that therapy increases both length and quality of life.
What about the legitimate concern that people at risk for HIV infection will lower their guard and reduce their vigilance and determination to practice safer sex behaviors because they believe there is an option to simply take antiretroviral medications "the morning after"? This is also a real though perhaps not overwhelming consideration.
Evaluating the effectiveness of HIV prevention is difficult in the best of circumstances. Today, with HAART and viral mutations affecting transmission rates, it is even more difficult. In Western countries where HAART is readily available, infectiousness of the overall population is believed to be decreasing because of treatment with HAART, with the baseline rates established pre-HAART. There has most likely been a change in baseline infectiousness in the entire community because of a reduction in community viral load due to effective antiretroviral therapy. Estimates of the risk of HIV transmission based on studies in the pre-HAART era therefore may not be valid today.
|Exposure||Likelihood of Transmission|
|Receptive anal intercourse||0.1% - 0.3% (1/1,000 - 3/1,000)|
|Receptive vaginal intercourse||0.08% - 0.2% (0.8/1,000 - 2/1,000)|
|Injection-needle sharing||0.67% (6.7/1,000)|
|Insertive anal sex||0.03% (3/10,000)|
|Insertive vaginal sex||0.03% - 0.09% (3/10,000 - 9/10,000)|
|Oral sex||No per-exposure estimates, but cases have been documented in the medical literature|
Amidst all this uncertainty, how does a clinician make a recommendation for or against PEP medications? The answer goes beyond evaluating the science to acting on the belief that the role of the health-care provider is to support each individual in making the best decisions for himself or herself. When talking to people like Mark (see introduction, above), health-care providers should try to provide as much information as they can so that the person with concerns can make an informed decision about whether to initiate medications. Providers can explain what is known about the risk of HIV transmission after sexual exposure and what is known about the potential benefit of taking medications. They can then work with the individual to try to place this treatment decision within the context of individual risk-taking behavior.
At the San Francisco PEP project, after the initial phone conversation, concerned individuals are encouraged to come in to a clinic where they can have an extended face-to-face discussion with both a counselor and a clinician to further assess the risks and benefits of PEP as well as to work on HIV risk reduction.
Almost all the people who call the PEP hotline make what appear to be reasonable decisions. People who call with very low risk exposures are offered only counseling, not medications. Drawing the line between offering and denying medication is a challenge for health-care providers.
Clinicians at the San Francisco PEP project deem it reasonable to offer medication only to people whose risk is greater than that of a man who has penetrated a known HIV positive partner without using a condom (see table, above). However, clinicians need to determine what treatment criteria they want to use. Some of the longest, most challenging counseling sessions on the PEP hotline take place when the clinician decides not to offer antiretrovirals and the person on the other end of the line insists that the counselor not stand in the way of access to medication. It is not easy to find the line between a healthy establishment of boundaries on the one hand and enabling underlying neurosis on the other.
Most providers outside of San Francisco recommend starting three rather than two antiretroviral medications after exposure. However, no scientific evidence supports the use of three drugs instead of two for the recommended 28 days of preventive treatment. The goal of PEP is to prevent infection, not to treat established infection. Even in studies of infected individuals, two antiretroviral medications are as effective as three during the first few weeks of treatment.
The increased cost and side effects from adding either a non-nucleoside reverse transcriptase inhibitor (NNRTI) or a protease inhibitor (PI) to two nucleoside reverse transcriptase inhibitors (NRTIs) should be avoided unless there is evidence of broad antiretroviral resistance in the virus of the infected partner. (Genotypic drug resistance testing is not yet a commonly used clinical tool, and furthermore, detailed information regarding a partner's particular virus is not likely to be discussed by people making sexual decisions.)
The cost of PEP medication is a major concern. The retail cost for 28 days of AZT and 3TC (Epivir) is approximately $600. Who pays for this treatment? Should insurance companies pay for PEP medications despite the lack of scientific evidence to support their use? If government programs do not cover the cost of PEP medications, access to this treatment will not be equitable across socioeconomic groups but will depend on the financial resources of the individual.
Starting the San Francisco PEP program three and one-half years ago may have been tantamount to opening Pandora's box. Now that the box is open, and PEP has shown some promise, even more challenging situations lie ahead. Indeed, it has created difficult and uncomfortable ethical dilemmas.
For instance, should PEP be offered after sexual assault as is recommended in New York, even though it is estimated that only 2% of convicted sex offenders are HIV positive? What about offering pre- and postexposure prophylaxis to serodiscordant male-female couples who desire to lower their risk of HIV transmission during conception? Because the menstrual cycle and the recommended length of PEP treatment are both usually 28 days, HIV negative people who initiate medications in this setting may end up continuously taking antiretroviral medications until successful conception, HIV seroconversion, or both occur.
While some providers are comfortable with the precepts of harm reduction, others feel that providing anti-HIV medications in this setting will encourage behavior that unnecessarily increases the risk of HIV transmission between partners. In the future, HIV positive women need to come forward (as some already have) and work with medical providers so that their right to have biologic children can be supported while reducing the risk of transmission to the greatest extent possible.
For those HIV health-care providers (and others) who watched many of their patients, friends, and family die a miserable death from AIDS throughout the 1980s and early '90s, it is very uncomfortable to hear from people who may be purposefully putting themselves or their children at risk for HIV infection. It is reasonable for providers to share this discomfort with people who ask them for help. However, providers' discomfort should not stand in the way of helping people access the medications, knowledge, and technology that may help them to reduce their risk of HIV transmission.
It is unlikely that scientific studies that either support or refute the use of PEP after sexual exposure will be conducted anytime soon. So, until there is an effective HIV vaccine, both health-care providers and those seeking care will have to make decisions based on their own beliefs and convictions, available financial resources, and personal comfort levels with risk rather than on science alone. Previous to the latter part of the 20th century, physicians provided firm guidance to their patients despite a paucity of scientific evidence to support their recommendations. Perhaps they recognized more readily than people today that hope and science are not entirely incompatible.
Joshua D. Bamberger, MD, MPH, is the Medical Director for Urban Community Health at the San Francisco Department of Public Health. He is also Assistant Clinical Professor of Family and Community Medicine at the University of California, San Francisco.
People who feel they have had an exposure to HIV or who have questions about postexposure prophylaxis can call the following numbers for more information:
AIDS Institute, New York State Department of Health. HIV prophylaxis following sexual assault: guidelines for adults and adolescents. 1998.
Bamberger, J.D. and others. Postexposure prophylaxis for human immunodeficiency virus (HIV) infection following sexual assault. American Journal of Medicine 106: 323-326. March 1999.
Cardo, D.M. and others. A case-control study of HIV seroconversion in health care workers after percutaneous exposure. New England Journal of Medicine 337(21): 1485-1490. November 20, 1997.
Larkin, H. P.A. Alameda County (CA) Sexual Assault Response Team. Personal communication. April 2000.
Martin, J. and others. Post-exposure prophylaxis after sexual or drug use exposure to HIV: final results from the San Francisco post-exposure prevention (PEP) project. 7th Conference on Retroviruses and Opportunistic Infections. San Francisco. January 30-February 2, 2000. Abstract 196.
Wade, N.A. and others. Abbreviated regimens of zidovudine prophylaxis and perinatal transmission of the human immunodeficiency virus. New England Journal of Medicine 339(20): 1409-1414. November 12, 1998.
Back to the SFAF BETA Summer, 2000 contents page.