Occupational Exposure to HIV: Revised Hopkins Policy

Methods to deal with occupational exposure to HIV are undergoing radical revision as a result of the CDC analysis of risk factors for infection and utility of AZT prophylaxis. The case-control study analyzed data from health care workers from 191 facilities who experienced needle stick injuries from an HIV-infected source during the period from January, 1988 through September, 1994. There were 31 health care workers who seroconverted and 679 who did not. The results of this analysis showed the following odds ratios for HIV transmission [MMWR 44:929,1995]:

  • Deep Injury - not further defined
  • 16.1
  • Visible blood on the device
  • 5.2
  • Procedure involving a needle in a vein or artery
  • 5.1
  • Terminal illness in the source case
  • 8.4
  • AZT prophylaxis
  • 0.2

    This study showed the anticipated enhanced risk associated with a high inoculum size (volume of blood and high viral load) and the depth of the injury. However, the five-fold reduction in transmission associated with AZT prophylaxis was not anticipated. The CDC is expected to issue new guidelines for post-exposure prophylaxis within the next 2 to 3 months. In the interim we have felt it important to implement strategies that reflect these new findings. Pending the CDC recommendations, the current Johns Hopkins policy is as follows:

    Risk Assessment

    Source: Patients who have recent negative serologic tests for HIV are presumed to be uninfected. Patients in high risk categories are presumed to be infected if they have not been tested recently. We arbitrarily define high risk patients as those in categories with HIV seroprevalence rates exceeding 10%: injection drug users, hemophiliacs, and gay men. Most would extend this definition to regular sexual partners of those individuals and those known to have HIV infection. An alternative method to deal with this issue is to use one of the rapid tests such as the Single Use Diagnostic System (SUDS; Mirax Corp., Norcross, GA.) which will provide a tentative answer within minutes. Many states permit testing the source without informed consent in the event of potential HIV exposure in the workplace, but this is not allowed in Maryland.

    Exposure: Exposure is defined as exposure of skin or mucous membranes to blood or bloody body fluids. The extent of the risk is defined on the basis of the nature of exposure characteristics as defined above. The most important factors are penetrating injury and the inoculum size as determined by the volume and viral burden (predicted by late-stage disease).

    Prophylactic Regimen:

    1. AZT (200 mg tid) plus 3TC (150 mg bid) x 4 weeks.
    2. For high risk exposures or probable HIV resistance: indinavir (800 mg tid) or ritonavir (300 mg bid x 2 days, 400 mg bid x 2 days, 500 mg bid x 3-10 days, then 600 mg bid) to complete 4 weeks of treatment.


    • AZT needs to be included because it is the only drug with established merit in this setting. The previously recommended dose was often 1.2 gm/day, but 75% experienced subjective side effects, and 31% found these to be sufficiently severe to prompt premature discontinuation of the drug. The dose now recommended is 600 mg/day because this appears to be the maximum effective dose and is substantially less toxic.
    • 3TC is added to the regimen based on the assumption that two drugs are probably better than one in this setting. Nevertheless, this opinion has not been verified and probably will not be for a long time. 3TC is well-tolerated and augments the effect of AZT on viral burden.
    • A protease inhibitor is added in two settings: when the exposure is felt to be high risk, as defined by the variables already discussed, and when the source is likely to carry drug resistant virus. Resistance to the drug administered is most prominent with prior exposure to 3TC; resistance is less likely and less harsh with exposure to AZT. Resistance is much more unusual with source exposure to ddI, ddC or d4T. Less is known about the extent or clinical relevance of resistance to protease inhibitors.
    • The choice of protease inhibitors is somewhat arbitrary. It is clear that indinavir is more easily tolerated. The major side effect is nephrolithiasis, which some would conclude is more threatening than the GI intolerance to ritonavir. Most people have difficulty tolerating full dose AZT plus ritonavir when these drugs are initiated in full dose simultaneously. An alternative strategy is the graduated dose of ritonavir, which does achieve therapeutic levels very rapidly. With intolerance to AZT, the common recommendations are to reduce the dose to 300 mg/day or to change therapy to d4T (40 mg twice daily). Intolerance to 3TC is unusual, but could be managed by combining AZT with ddI. With intolerance to the protease inhibitors, the alternative is to switch to another agent in this class or discontinue this component of treatment.
    • There is limited experience with all of these drugs in healthy persons without HIV infection. The most extensive is with AZT given in this setting, and the prior record shows no serious adverse reactions. 3TC is generally devoid of side effects. Indinavir is occasionally complicated by nephrolithiasis requiring the warning to drink large volumes (2-4 L/day). The major problem with ritonavir is GI intolerance and an enormous number of drug interactions that require patient warning. The only drug that seems to have an extensive track record during pregnancy is AZT through ACTG 076; this seemed to verify safety during the second and third trimesters.


    Counseling: We provide counseling for the health care worker through the Infectious Diseases Service. There are nine persons who manage this emergency beeper, including five ID fellows and four attendings. The fellows field the initial calls and then refer to the attending when additional information or advice is required.

    Routine care and follow-up: This is done by the Employee Health Service.

    Laboratory monitoring: Serologic testing is performed at baseline and three months and six months following exposure. A CBC and a chemistry panel is performed at baseline, two weeks and four weeks. To our knowledge, there has been a single report of a health care worker who seroconverted nine months after exposure, and this remains the singular exception to the six month rule.

    Informed consent: The health care worker gives written informed consent for use of antiretroviral agents in this fashion, but this may not be necessary in many or most institutions. A document is provided that summarizes much of the information given above, as well as the potential side effects of the drugs. The health care worker is also warned about the possibility of the acute retroviral syndrome at 2-4 weeks after exposure, and the reduced probability of transmission in the event that this illness is not detected. Females of child bearing potential have a pregnancy test prior to prophylactic treatment. All patients are advised to eliminate or reduce risk of HIV exposure to sexual contacts for the six month period.

    Other bloodborne pathogens: These are important issues that are necessary to address but are beyond the scope of this review. The major concerns are preventable disease as with hepatitis B virus infection.

    Starter pack: We offer a 72-hour supply of AZT and 3TC for health care workers who are undecided about the decision for prophylaxis. The timeliness of the initiation of treatment is an important variable in disease prevention, and an attempt is made to initiate treatment within one hour of the exposure. The starter pack permits us to achieve this goal while permitting the health care worker to seek additional information from authoritative sources and/or to assess the tolerability of the drugs. The CDC Guidelines for HIV exposures in the workplace are based on a Consensus Conference held 2/29-3/1/96. The final version is expected in July. In the meantime, the policy noted above will need to suffice.