Early identification of HIV infection is of paramount importance for appropriate counseling, risk reduction, partner notification, and consideration of early aggressive therapy with combination antiretroviral medications. Appropriate counseling of a person with an indeterminate HIV test result is affected by the probability that the test reflects seroconversion and is not a false-positive enzyme-linked immunosorbent assay (ELISA). The rate of HIV seroconversion in persons with an indeterminate test result depends on prior HIV exposure (1-3). Rates of seroconversion for individuals who receive indeterminate Western blot (WB) test results are <1% for blood donors and other low-risk individuals (1-3). Confusion often arises when interpreting indeterminate HIV test results and how to best evaluate and counsel patients, particularly in high-risk populations.
In 1989, the Centers for Disease Control and Prevention (CDC) announced their clinical recommendations and guidelines for interpretation of indeterminate HIV test results (4). The CDC noted that serologic testing is only one component of the diagnosis and that physicians should consider medical history (including high-risk behavior or exposure to HIV) and physical examination (4). The CDC recommended serial HIV serologic testing of ELISA and WB (i.e., at 6 weeks, 3 months, 6 months, and 12 months) after an indeterminate HIV test result (4).
HIV is the leading medical problem in prisons in several states (5-8). Vlahov et al. (8) examined HIV-1 seroprevalence in 10 correctional institutions in the United States. Prevalence ranged from 2.1% to 7.6% for men and 2.5% to 14.7% for women, with seroprevalence higher in women than in men in 9 of the 10 institutions (8).
Many of the behaviors that put people at risk of HIV infection also put them at risk of incarceration. Therefore, many people may become infected near the time of incarceration when drug use, HIV risk-taking behavior, and illegal activities are escalating. Because of the high risk of HIV in this population, we hypothesized that indeterminate HIV tests taken on entrance to correctional institutions are more likely to represent true HIV infection in prison inmates than in other populations.
Mandatory HIV testing on all sentenced inmates at the Rhode Island state prison was implemented in 1990. Inmates are generally tested on entrance into the prison. Before testing, a nurse interviews inmates regarding risk and arranges for medical follow-up if needed. Serum samples were analyzed by ELISA (Abbott Laboratories, Abbott Park, IL, U.S.A.), and only if repeatedly reactive, with confirmatory WB (Bio-rad Diagnostics, Hercules, CA, U.S.A.). The Association of State and Territorial Public Health Laboratory Directors criteria for WB analysis were used (4). All HIV-infected persons were referred for medical intervention.
We conducted a medical chart review on all HIV indeterminate tests through October 1996 and examined follow-up HIV test results and information on demographics and HIV risk behavior reported to the nurse on incarceration.
The Rhode Island prison system completes HIV testing on ∼11,000 individuals each year: 45% of men and 39% of women are from minority groups. The average length of stay for those awaiting trial is 24 days for men and 10 days for women. Many of these individuals, particularly women, are released into the community before repeat serologic testing. Forty-five percent of the women and 22% of the men are sentenced on prostitution or drug-related charges. A majority of the inmates are Rhode Island residents.
From the inception of mandatory testing in January 1990 through October 1996, 35 individuals had an indeterminate HIV test. Thirty-one of 35 (89%) of the individuals with indeterminate HIV tests had follow-up HIV testing. Twenty-three of 31 (74%) individuals with a follow-up HIV test seroconverted. The characteristics of individuals with an indeterminate HIV test result and those who seroconverted are shown in Table 1.
Risk factors for HIV were examined to identify variables that were strongly associated with seroconversion. Venereal disease, abnormal discharge, tuberculosis, hepatitis, sexual risk behaviors, and race were not associated with seroconversion. Alcohol and drug use, including injection drug use, cocaine, and crack use, was significantly associated with seroconversion (p < 0.01, odds ratio [OR] = 11.8, relative risk [RR] = 2.04). Injection drug use was also significantly associated with seroconversion (p = 0.03, OR = 9.3, RR = 1.56).
Of those who seroconverted, 13 of 23 (57%) had a fully reactive WB at three or more bands. Twenty-one of 23 (91%) of the seroconverters were fully reactive at p24, 0 of 23 (0%) at gp41, 0 of 23 (0%) at gp120, and 9 of 23 (39%) at gp160. Twenty-one of 23 (91%) had partial reactivity at other bands. Among those who did not seroconvert on follow-up, 1 of 8 (13%) had a fully reactive WB at three or more bands, 2 of 8 (25%) at p24, 0 of 8 (0%) at gp41, 0 of 8 (0%) at gp 120, and 3 of 8 (38%) at gp160. Seven of 8 (88%) had partial reactivity at other bands.
Twenty-one of 23 (91%) of those who seroconverted were either never incarcerated previously (8 persons) or were not incarcerated in the 3 months prior to their first indeterminate HIV test result (13 persons). Two individuals had been incarcerated 2 months before their first indeterminate HIV test, one for 9 days and the other for 6 weeks. Both had been HIV-negative at the beginning of their prior incarceration, and the latter denied any risk behavior while in prison.
Twenty-three of 31 (74%) of all incarcerated persons with an indeterminate WB who had follow-up HIV testing seroconverted. This is the highest rate of seroconversion ever reported for persons with indeterminate WB test results. Many incarcerated individuals are at high risk of HIV infection on the basis of their prostitution and drug-related activities. Therefore, correctional institutions offer an opportunity to identify acute HIV infection for early medical intervention in a population that underuses health care. Based on our findings and the percentage of high-risk individuals in prisons, we propose that indeterminate HIV serology results in the prison setting be interpreted differently than in the community.
Correctional institutions nationwide have revealed that 67% of the inmates who have been tested and counseled have reported a history of injection drug use, making it the most common risk factor for HIV in prisoners (8-10). We found injection drug use to be a significant predictor of HIV seroconversion in this population of inmates with indeterminate HIV tests. Because seroconversion typically occurs within 3 months of infection (4), most, if not all, of these prisoners were infected while outside of the prison.
As a consequence of strict drug laws that are intended to curtail drug use, this highly elusive population of substance abusers is more heavily concentrated in prisons than in all drug treatment programs, hospitals, and social service areas combined (6-8). Incarceration may be the only opportunity to link many of these individuals with supportive social services that may allow them to break the cycle of substance abuse and enter into long-term primary medical care (9,11).
Incarcerated individuals with an indeterminate test result need to be appropriately counseled regarding the implications of their test. If these individuals are in the process of seroconverting, they may have very high viral load and may be highly infectious, either through sexual contact or through needle-sharing behaviors. Counseling of partners can begin during the time of incarceration. Because the majority of individuals seroconvert, persons should be counseled that in all probability they have HIV infection. Recent information regarding the optimistic outlook of aggressive treatment of early HIV infection with combination medications can be provided to these individuals during this critical period (12). If these services are provided with appropriate support, it is our experience that persons who by and large are disenfranchised from the medical and social service agencies can be linked with intensive support that can continue in the community (9-10,13). This can provide hope and continuity of care for individuals who otherwise would be released into the community and may never be contacted until they develop serious HIV-related complications due to immunosuppression.
We recommend that individuals in the correctional setting who are found to have indeterminate HIV results undergo immediate testing with HIV viral load quantification. For those persons who are truly infected, the viral load will most likely be very high (>5000 copies/ml). This test can be done with a short turnaround time. If the HIV viral load test is negative, repeat HIV serologic testing is still necessary (14).
We report a very high risk of subsequent HIV seropositivity in persons who are incarcerated and who have an indeterminate HIV test result. These data should result in a modification of counseling recommendations within prisons and jails. All incarcerated persons with a history of high-risk behaviors and an indeterminate test should be counseled that it is likely they are HIV-positive, and confirmatory HIV viral load testing should be conducted as soon as possible. One-on-one counseling with a trained HIV counselor along with supportive case management and comprehensive HIV-related care must be provided to these individuals both within the incarcerated setting, and then on follow-up in the community.
Acknowledgments: We thank Phil Barber, Rhode Island Department of Health Office of AIDS/STD; Ann Marie Roberti, Glennis Channon, and Paul Berrigan, Eleanor Slater Laboratory; Patricia Amata, Ann Marie Bandieri, and Finian Murphy, Rhode Island Department of Corrections; Charles C.J. Carpenter, The Miriam Hospital Immunology Center; and Lucie Coughlin, The Miriam Hospital, for their invaluable assistance in this project. This work is supported by the National Institutes of Health, National Institute on Drug Abuse K20 grant DA00268 to J.D.R.
1. Jackson JB, Hanson MR, Johnson GM, Spahlinger TG, Polesky HF, Bowman RJ. Long-term follow-up of blood donors with indeterminate human immunodeficiency virus type 1 results on Western blot. Transfusion
2. Klienman S, Fitzpatrick AL, Secord K, Wilke D. Follow-up testing and notification of anti-HIV Western blot atypical (indeterminate) donors. Transfusion
3. Lee J. Follow-up investigation of indeterminate Western blot results for antibody to human immunodeficiency virus type 1. J Formos Med Assoc
4. Centers for Disease Control. Interpretation and use of the Western blot assay for serodiagnosis of human immunodeficiency virus type 1 infections. MMWR Morb Mortal Wkly Rep
5. HIV prevention in the U.S. correctional system, 1991. MMWR Morb Mortal Wkly Rep
6. Smith P, Miki J, Truman B, et al. HIV infection among women entering the New York State correctional system. Am J Public Health
7. Hammet T, Widom R, Epstein J, Gross M, Sifre S, Enos T. 1994 update: HIV/AIDS and STDs in correctional facilities. U.S. Department of Justice, Office of Justice Programs, National Institute of Justice, December 1995.
8. Vlahov D, Brewer F, Castro K, Narkunas J, Salive M, Ullrich J, Munoz A. Prevalence of antibody to HIV-1 among entrants to US correctional facilities. JAMA
9. Flanigan TP, Rich JD, Dickinson BP, Vigilanle K, Spaulding A. Incarceration as a unique opportunity for HIV diagnosis, initiation of comprehensive care, and linkage to the community. Presented at the 4th Conference on Retroviruses and Opportunistic Infections, Washington, DC, January 22-26, 1997.
10. Dixon P, Flanigan T, DeBuono B, et al. Infection with the human immunodeficiency virus in prisoners: meeting the health care challenge. Am J Med
11. Brewer T, Derrickson J. AIDS in prison: a review of epidemiology and preventative policy. AIDS
12. Markowitz M, Cao Y, Vesanen M, et al. Tennerracz K, Ho DD. Recent infection treated with AZT, 3TC, and a potent protease inhibitor. Presented at the 4th Conference on Retroviruses and Opportunistic Infections, Washington D.C, January 22-26, 1997.
13. Flanigan TP, Kim JY, Zierler S, Rich J, Vigilante K, Bury-Maynard D. A prison release program for HIV-positive women: linking them to health services and community follow-up. Am J Public Health
14. Schwartz DH, Layendezner OB, Arango-Jaramillo S, Castillo RC, Reynolds M. Extensive evaluation of a seronegative participant in an HIV-1 vaccine trial as a result of false positive PCR. Lancet