Sethi, Ajay K. PhD, MHS*; Celentano, David D. ScD, MHS*; Gange, Stephen J. PhD*; Gallant, Joel E. MD, MPH*†; Vlahov, David PhD*‡; Farzadegan, Homayoon PhD*
In January 2001, the Centers for Disease Control and Prevention (CDC) issued its strategic plan for HIV prevention, with the overarching objective of reducing new infections in the United States by at least 50% by 2005. 1 The CDC plan recommends the “delivery of targeted, sustained and evidence-based HIV prevention interventions to all persons at high risk for acquiring or transmitting HIV by unprotected sex and/or needle sharing.” 1 In the era of highly active antiretroviral therapy (HAART), AIDS death rates in the United States have declined dramatically and the number of persons living with HIV has increased. 2,3 There is a growing need to counsel infected individuals, who now live longer healthier lives and are thus more capable of engaging in high-risk behavior, such as unprotected sex or needle sharing, that might transmit HIV to others. Moreover, individuals who develop resistance to antiretroviral therapy (ART) can transmit drug-resistant virus to others, which can limit treatment options in those who become infected with these strains. 4
Several studies have measured the prevalence of genotypic and/or phenotypic resistance among recent HIV seroconverters. 4–14 Although injection drug users (IDUs) have been underrepresented in these studies, transmission of resistant virus among IDUs is a growing public health concern and warrants an examination of the problem from a prevention perspective. Therefore, we investigated the predictors of unprotected sex and needle sharing among HIV-infected IDUs participating in an ongoing cohort study in Baltimore, Maryland. Based on reported high-risk behavior, ART use, and viral load, we classified individuals as being at higher or lower risk of HIV and drug-resistant HIV transmission and determined the prevalence of clinically significant drug resistance among those in the latter group.
This investigation was conducted among participants in the AIDS Links to the Intravenous Experience (ALIVE) study, a longitudinal study of the natural history of HIV infection among IDUs in Baltimore, Maryland. Details of the study methods and characteristics of the cohort have been published. 15 Briefly, a total of 3360 persons who were at least 18 years old were recruited in 1988 through 1989 and 1994. Participants had to have a history of nonmedical injection of illicit substances within the previous 10 years but could not have AIDS at the time of recruitment. Subjects provided informed consent (approved by the Committee on Human Research) and were followed at 6-month intervals with repeat interviews, physical examinations, and blood specimen collection stored in a repository. The interviews obtained information regarding demographics, health care utilization, current drug and sexual behavior, and ART use.
In this analysis, we studied HIV-infected IDUs who contributed visits between December 1996 and February 2000. To determine predictors of high-risk behavior among HIV-infected IDUs, we restricted the sample to visits with complete information on drug and sexual behaviors. To examine the potential for HIV and drug-resistant HIV transmission, we further restricted the sample to visits with complete information on viral load and ART use.
Predictors of Drug and Sexual Behavior
Our first analysis examined predictors of 4 outcomes: (1) having sexual intercourse in the past 6 months; (2) unprotected sex (vaginal or anal intercourse without using a condom) among those reporting recently having sex; (3) injection drug use in the past 6 months; and (4) needle sharing, defined as lending or borrowing nondisinfected needles during recent injection of illicit drugs. The univariate and multivariate association of these outcomes with sociodemographic, alcohol and drug use, sexual risk behavior, clinical, and health care access and utilization variables measured at the concomitant visit were examined. To account for the correlation between repeated measures from each subject, generalized estimating equation (GEE) regression analyses were conducted using a logit link function, an unstructured correlation structure, and a robust estimator of variance. 16,17 Covariates whose probability values were less than 0.10 in univariate analysis were entered into multivariate models. For the final multivariate model, only covariates whose probability values were less than 0.05 were retained.
Potential Transmission of HIV and Drug-Resistant HIV
We identified visits at which HIV-infected IDUs reported engaging in unprotected sex or needle sharing in the previous 6 months and examined plasma viral load at that visit and the prior visit. We classified IDUs as being at higher risk of HIV transmission if they had an “elevated” viral load during the period of high-risk behavior, defined as having a plasma viral load ≥1000 copies/mL at both visits or rising to greater than 1000 copies/mL between visits. Participants who reported ART use when they were at higher risk of HIV transmission were classified as being at higher risk of drug-resistant HIV transmission.
Although a study by Quinn and colleagues 18 observed no seroconversions among HIV-discordant couples in which the infected sexual partner’s serum viral load was <1500 copies/mL, we chose a plasma viral load cutoff of 1000 copies/mL; this was also the minimum concentration needed for drug resistance testing. No studies have estimated the minimum viral load associated with increased risk of parenteral transmission, which is considered to be more efficient than heterosexual HIV transmission. 19–22
ART regimens reported were classified as either non-HAART or HAART. HAART was defined as one of the following regimens 23: 2 nucleoside reverse transcriptase inhibitors (NRTIs) plus 1 protease inhibitor (PI), 2 PIs, 2 NRTIs plus 1 nonnucleoside reverse transcriptase inhibitor (NNRTI), 2 NRTIs plus abacavir, or 1 PI plus 1 NRTI plus 1 NNRTI. Non-HAART regimens were defined as 2 NRTIs, any monotherapy, 1 PI plus 1 NRTI, or 1 NNRTI plus 1 NRTI. For the purposes of this analysis, tenofovir disoproxil fumarate was included in the NRTI class.
Clinically significant resistance was determined by expert interpretation of genotypes obtained by VirtualPhenotype (Vircolab, Rockville, MD) and compared with VirtualPhenotype interpretation. 24 The expert was an experienced HIV clinician and researcher, was blinded to all data except the genotype report, and sometimes supplemented or confirmed his interpretation using the Stanford algorithm (available at: http://hivdb.stanford.edu). 25 The expert indicated the mutations associated with no, low-level, intermediate-level, or high-level resistance to 16 antiretroviral agents approved by the US Food and Drug Administration. Intermediate- and high-level resistance to a drug was deemed clinically significant such that the expert would be less likely to use that drug in a subsequent regimen.
Between December 1996 and February 2000, 638 IDUs contributed 2745 visits in the ALIVE study (mean = 3 visits, interquartile range [IQR]: 2, 4). For the analysis of predictors of high-risk behaviors, the sample consisted of 638 (100%) IDUs with 2731 (99.5%) visits. Participants reported unprotected sex, needle sharing, and both high-risk behaviors at 13%, 7%, and 3% of visits, respectively. For the cross-sectional analysis to characterize the potential for HIV and drug-resistant HIV transmission, the sample consisted of 616 (96.6%) IDUs contributing 2448 (89.2%) visits.
Factors Associated With Drug and Sexual Behaviors
Factors associated with reporting sex and unprotected sex are shown in Table 1. In multivariate analysis, older age (P < 0.001), unemployment (P < 0.001), and homelessness (P = 0.029) were independently associated with not having sex, whereas alcohol use (P < 0.001) and higher CD4+ cell count (P < 0.001) were independently associated with being sexually active. Among sexually active IDUs, women were almost twice as likely as men to report unprotected sex after adjusting for alcohol use, trading sex for drugs, frequency of sex, health insurance, and HIV status of sex partners (P < 0.001). HIV-infected IDUs whose sex partners were also infected were almost 80% more likely to engage in unprotected sex compared with IDUs who had at least 1 sex partner not known to be infected in multivariate analysis (P < 0.001).
Several factors were associated with being sexually active or having unprotected sex by univariate analysis but not in the final multivariate models. Although having an AIDS diagnosis was associated with not having sex (P = 0.007), sharing needles (P = 0.042) and recently seeing a doctor (P = 0.002) were associated with being sexually active. Lower income, which was associated with not having sex (P = 0.033), was associated with having unprotected sex among sexually active IDUs (P = 0.042). Injection drug use (P = 0.015), needle sharing (P < 0.001), and having a sexual partner who was an IDU (P < 0.001) were associated with engaging in unprotected sex.
Factors associated with reporting injection drug use and needle sharing are shown in Table 2. In a multivariate analysis, unemployment (P < 0.001), homelessness (P < 0.001), alcohol use (P < 0.001), trading sex for drugs (P < 0.001), and having no health insurance (P = 0.027) were independently associated with injection drug use, whereas having an undetectable viral load (P = 0.003), a marker of effective ART use, was associated with not injecting drugs. Among recent injectors, homelessness (P = 0.002), daily injection (P = 0.001), and trading sex for drugs (P < 0.001) were independently associated with sharing needles.
Lower income and having a sexual partner who was an IDU were associated with injection drug use (P = 0.004 and P < 0.001, respectively) and needle sharing (P = 0.013 and P < 0.001, respectively) in univariate analyses but were not retained in the multivariate models. A recent visit to an emergency department (P < 0.001) and drug detoxification treatment (P < 0.001) were associated with recent injection drug use, whereas methadone maintenance (P = 0.015) and a recent doctor visit (P = 0.001) were more common among those who did not recently inject drugs. Similarly, higher viral load (P = 0.034) and having an opportunistic infection (P < 0.001) were associated with injection drug use, whereas receiving ART (P < 0.001), having a higher CD4+ cell count (P = 0.076), and having an undetectable viral load (P < 0.001) were associated with not injecting drugs. Alcohol use (P = 0.013), being sexually active (P = 0.014), and having a higher CD4+ cell count (P = 0.007) were significantly associated with sharing needles, whereas receiving ART (P = 0.043) was associated with not sharing needles in univariate analysis.
Potential HIV and Drug-Resistant HIV Transmission
Of 2448 visits contributed, participants reported sex or injection drug use at the majority (80%) of visits (Fig. 1). At 19% of visits, IDUs were at higher risk of transmitting HIV because they reported recent unprotected sex or needle sharing while having an elevated viral load. The median viral load at the visit before the period of high-risk behavior was 4.22 log10 copies/mL (IQR: 3.37, 4.81), and at the visit when recent high-risk behavior was assessed, it was 4.26 log10 copies/mL (IQR: 3.47, 4.86). At 61% of all visits, IDUs were considered to be at lower risk for transmitting HIV because they did not engage in high-risk behavior (45%), did not have an elevated viral load at the time (5%), or both (11%). At one third of visits when IDUs were at higher risk of HIV transmission, they were also at higher risk of drug-resistant HIV transmission, because ART use was reported. Non-HAART regimens were reported at one half of those visits (30% monotherapy, 59% 2 NRTIs, and 11% other 2-drug combinations).
We obtained complete VirtualPhenotype results on 121 (80%) specimens. According to the expert interpretation of the genotypes obtained (Figs. 2 and 3), 57 (47%) were considered to have any clinically significant resistance, 50 (41%) were considered to have low to no resistance, and 14 (12%) showed wild-type virus. Thirty-nine (32%) showed clinically significant resistance to at least 1 antiretroviral agent in 1 drug class, 15 (12%) to agents in 2 drug classes, and 4 (3%) to agents in all 3 drug classes. Most IDUs with clinically significant resistance to specific agents were also exposed to those agents during the study period. For example, among those with lamivudine resistance as indicated by the presence of the M184V mutation, 98% had prior exposure to that drug. Cross-resistance was also evident; whereas 12% of genotypes indicated clinically significant abacavir resistance, no one reported abacavir use. Similarly, whereas 12% to 13% of genotypes indicated resistance to all NNRTIs, past exposure to delavirdine, efavirenz, and nevirapine was 0%, 33%, and 63%, respectively. This was due to the fact that the K103N mutation, which causes class-wide resistance, was present in 13 of the 16 genotypes indicating NNRTI resistance. The L90M and D30N major mutations were present in 10 of 12 genotypes indicating resistance to a PI.
Excluding IDUs who had wild-type virus and duplicate mutation patterns, agreement between expert and Virtualphenotype interpretation for reverse transcriptase inhibitors (RTIs) ranged from 75% for stavudine to 100% for lamivudine and all NNRTIs. For PIs, agreement ranged from 74% for indinavir to 83% for nelfinavir and amprenavir. For all RTIs that were disagreed on, the expert was more likely than VirtualPhenotype to classify resistance as clinically significant. There was no obvious pattern of disagreement between expert and VirtualPhenotype interpretation for PIs.
Based on our findings, among IDUs who were at higher risk of HIV transmission, clinically significant drug-resistant HIV was found at 14% of visits, after adjusting for the number of specimens for which resistance test results were not obtained and assuming that those not receiving ART were not at higher risk of drug-resistant HIV transmission. Of these, IDUs reported engaging in unprotected sex at 68% of visits, sharing needles at 25% of visits, and engaging in both behaviors at 7% of visits. Among visits when recent high-risk behavior was reported and clinically significant drug resistance was found, IDUs reported engaging in high-risk behavior at 44% and 43% of visits 6 and 12 months later, respectively.
The strongest evidence for transmission of drug-resistant HIV comes from studies showing that up to 27% of HIV seroconverters have drug resistance mutations. 5–13 Although comparison between studies is difficult because of the use of different assays, definitions of resistance used, populations, and time periods studied, our study corroborates earlier findings. For someone to become infected with drug-resistant HIV, that individual must engage in high-risk behavior with a partner who has drug resistance and a viral load that is high enough to allow for HIV transmission. In our study, among HIV-infected IDUs who engaged in high-risk behavior and had an elevated viral load, 14% had clinically significant resistance, placing their partners at risk for acquiring a drug-resistant HIV strain.
Only 1 of the studies that examined resistance among seroconverters was conducted in an IDU population; however, no primary resistance was found. 5 In a previous study, the prevalence of primary resistance mutations among recent seroconverters between 1988 and 1999 in the ALIVE study was 5%. 26 It is unclear why resistance among IDU seroconverters has been lower than in other populations of HIV seroconverters. Given the low uptake of ART in our Baltimore IDU cohort during that time period, 27,28 lower levels of resistance may have been circulating in this community. Nevertheless, the findings in the present study are consistent with our expectation that the prevalence of resistance among HIV-infected IDUs who engaged in high-risk behavior should be higher than the prevalence of resistance among those who recently seroconverted.
The prevalence of resistance among individuals who engage in high-risk behavior may not accurately estimate the likelihood of transmission of drug-resistant HIV. For transmission of drug resistance to occur from an HIV-infected person to a sexual or drug-using partner, several factors must occur simultaneously. First, the individual must engage in a high-risk behavior such as unprotected sex or needle sharing. Second, drug-resistant HIV must be present in genital secretions or blood. Third, the resistant strain must be present in adequate quantities and be sufficiently fit to be infectious and to not be out-competed by wild-type strains, if present. Finally, infection or reinfection with the drug-resistant HIV strain in the recipient must be established. Other factors known to affect the risk of HIV transmission in general may also affect the transmission of drug-resistant HIV strains. These include the viral load in the HIV-infected person 18 and the presence of sexually transmitted infections. 29
The low social desirability of reporting unprotected sex and needle sharing may have led to an underestimation of high-risk behavior, thus making our estimates of the potential transmission of drug-resistant HIV conservative. Furthermore, drug resistance was measured in stored plasma, which may better estimate the risk of transmission through needle sharing than through unprotected sex. Previous studies have shown that the presence of drug-resistant HIV in plasma may not reflect the susceptibility of genital tract virus because of varying selection pressures due to poor bioavailability of some antiretroviral agents in some tissues. 30,31 Also, expression of resistance mutations may vary over time due to changing selection pressures. 32 Thus, an individual with detectable resistance at the time of a study visit may have had susceptible virus during high-risk behavior and vice versa.
Because we tested previously collected plasma, we could not apply the recommendation that resistance testing be conducted while an individual is taking ART or within a short time after stopping therapy to maximize the opportunity to detect resistance mutations that might be present. 32 In the cohort study setting, where study visits are conducted and plasma specimens are stored semiannually, we still found that 47% of specimens tested showed clinically significant resistance mutations. We believe that this study demonstrates that drug resistance testing, which has already been shown to be useful in the clinical management of patients, also has utility in the epidemiologic study setting. Any estimates of prevalence of clinically significant drug resistance among previously collected plasma specimens must be considered conservative, however.
This study revealed that HIV-infected IDUs engaged in unprotected sex or needle sharing at 23% of visits during the study period. Without a comparison group, one should not single out HIV-infected IDUs as the only population that engages in high-risk behavior. Among HIV-infected men who have sex with men (MSM), approximately one quarter reported recent unprotected anal intercourse. 33 In a study of HIV-infected substance users, of whom 74% were heterosexual, 61% and 20% reported recent unprotected vaginal and anal intercourse, respectively. 34 In a study of 236 HIV-infected patients in clinical care, 56 (24%) reported 1 or more high-risk sexual acts over the previous 3 months, and of these, 12 (22%) had HIV drug resistance. 35 In another study conducted by the CDC at 12 state or city health departments, among HIV-infected IDUs who had injected in past year, 50% had shared needles during that year. 36 In a cohort of IDUs in France, 24% of HIV-infected users who injected in the previous 6 months had also shared needles at least once. 37
We found that women were nearly twice as likely as men to report unprotected sex, even after adjusting for multiple factors that included having only HIV-infected sexual partners, trading sex for drugs, and frequency of sex. It is possible that women have difficulty in convincing their male sexual partners to use condoms, indicating a need for interventions that aim to reduce high-risk behavior designed specifically for female IDUs. Alternatively, a reporting bias may also explain our finding: men may have been less likely than women to admit to engaging in recent unprotected sex.
HIV-infected IDUs whose sexual partners were also infected were twice as likely to have unprotected sex, which has also been reported among MSM. 38 This behavior increases the potential risk of HIV superinfection, 39–41 including superinfection with drug-resistant virus, which could have an adverse impact on future response to ART. Thus, it is important that HIV-infected individuals understand the potential risks and consequences of continuing to engage in unprotected sex regardless of the serostatus of their sexual partners.
Interventions are needed to reduce the transmission of drug-resistant HIV in communities. One approach would be to improve adherence among individuals prescribed HAART to avoid the development of resistance. The use of non-HAART regimens may also have contributed to the development of resistance in this cohort. Although the study sample comprised only those visits in the HAART era, 47% of the IDUs found to have clinically significant drug-resistant HIV were receiving non-HAART regimens. This was not surprising, because we previously showed that IDUs in Baltimore have been slow to receive HAART despite having indications for ART, and many continue to receive non-HAART regimens. 27,28 Those participants in the ALIVE cohort who are in care receive their care in a variety of settings and from a variety of clinicians, many of whom are not HIV experts. Given that a goal of HAART is to reduce HIV transmission, 42 this study reinforces the negative public health consequences of treating HIV infection with non-HAART regimens. Not surprisingly, we found that treatment with a non-HAART regimen was associated with having a viral load >1000 copies/mL (odds ratio [OR] = 2.6, 95% confidence interval [CI]: 1.9, 3.4).
Perhaps the most important and effective way to reduce transmission of HIV and drug-resistant HIV is through safer sexual and drug-using practices. For illicit drug-using populations, availability of methadone maintenance and needle exchange programs can reduce drug addiction and needle sharing, respectively. HIV care providers also need to incorporate risk reduction counseling during routine clinic visits with their patients, particularly those who are receiving ART or are infected with drug-resistant HIV. 43 Given that one of the strongest predictors of engaging in unprotected sex or needle sharing is reporting that behavior previously, 44 emphasizing and maintaining risk reduction is a challenging task. In our study, IDUs who engaged in unprotected sex were also more likely to share needles and vice versa. Thus, counseling IDUs about sexual and drug-using risk behaviors is critical. A recent study by Marks and colleagues 45 showed that discussion of safer sex varies across clinics. In their survey, 52% to 94% of HIV-infected patients sampled from 5 public HIV clinics in California reported that their providers had talked with them at least once about safer sex. 45
The transmission of drug-resistant HIV has been well documented and has recently been shown to be increasing over time 4,14; further studies like ours are needed to examine this problem from a prevention perspective in other populations. Improving ART adherence, appropriate prescribing of HAART regimens to prevent or overcome resistance, and risk reduction counseling will be the most important ways to reduce transmission of drug-resistant HIV in communities.
1. Centers for Disease Control and Prevention. HIV Prevention Strategic Plan Through 2005
. Atlanta: Centers for Disease Control and Prevention; 2001.
2. Karon JM, Fleming PL, Steketee RW, et al. HIV in the United States at the turn of the century: an epidemic in transition. Am J Public Health
3. Centers for Disease Control and Prevention. Update: AIDS—United States, 2000. MMWR Morb Mortal Wkly Rep
4. Little SJ, Holte S, Routy JP, et al. Antiretroviral-drug resistance among patients recently infected with HIV. N Engl J Med
5. Alexander CS, Dong W, Schechter MT, et al. Prevalence of primary HIV drug resistance among seroconverters during an explosive outbreak of HIV infection among injecting drug users. AIDS
6. Boden D, Hurley A, Zhang L, et al. HIV-1 drug resistance in newly infected individuals. JAMA
7. Little SJ, Daar ES, D’Aquila RT, et al. Reduced antiretroviral drug susceptibility among patients with primary HIV infection. JAMA
8. Salomon H, Wainberg MA, Brenner B, et al. Prevalence of HIV-1 resistant to antiretroviral drugs in 81 individuals newly infected by sexual contact or injecting drug use. Investigators of the Quebec Primary Infection Study. AIDS
9. Weinstock H, Respess R, Heneine W, et al. Prevalence of mutations associated with reduced antiretroviral drug susceptibility among human immunodeficiency virus type 1 seroconverters in the United States, 1993–1998. J Infect Dis
10. Briones C, Perez-Olmeda M, Rodriguez C, et al. Primary genotypic and phenotypic HIV-1 drug resistance in recent seroconverters in Madrid. J Acquir Immune Defic Syndr.
11. Duwe S, Brunn M, Altmann D, et al. Frequency of genotypic and phenotypic drug-resistant HIV-1 among therapy-naive patients of the German Seroconverter Study. J Acquir Immune Defic Syndr
12. UK Collaborative Group on Monitoring the Transmission of HIV Drug Resistance. Analysis of prevalence of HIV-1 drug resistance in primary infections in the United Kingdom. BMJ
13. Harzic M, Pellegrin I, Deveau C, et al. Genotypic drug resistance during HIV-1 primary infection in France (1996–1999): frequency and response to treatment. AIDS
14. Grant RM, Hecht FM, Warmerdam M, et al. Time trends in primary HIV-1 drug resistance among recently infected persons. JAMA
15. Vlahov D, Anthony JC, Munoz A, et al. The ALIVE study, a longitudinal study of HIV-1 infection in intravenous drug users: description of methods and characteristics of participants. NIDA Res Monogr
16. Diggle PJ, Liang KY, Zeger SL. Analysis of Longitudinal Data
. New York: Oxford University Press; 1994.
17. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics
18. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med
19. Kaplan EH, Heimer R. A model-based estimate of HIV infectivity via needle sharing. J Acquir Immune Defic Syndr Hum Retrovirol
20. Wiley JA, Herschkorn SJ, Padian NS. Heterogeneity in the probability of HIV transmission per sexual contact: the case of male-to-female transmission in penile-vaginal intercourse. Stat Med
21. Padian N, Marquis L, Francis DP, et al. Male-to-female transmission of human immunodeficiency virus. JAMA
22. Fischl MA, Dickinson GM, Scott GB, et al. Evaluation of heterosexual partners, children, and household contacts of adults with AIDS. JAMA
23. Department of Health and Human Services. Guidelines for the Use of Antiretroviral Agents in HIV-Infected Adults and Adolescents
. Rockville, MD. Department of Health and Human Services; 2002.
24. Harrigan PR, Montaner JS, Wegner SA, et al. World-wide variation in HIV-1 phenotypic susceptibility in untreated individuals: biologically relevant values for resistance testing. AIDS
25. Shafer RW, Jung DR, Betts BJ. Human immunodeficiency virus type 1 reverse transcriptase and protease mutation search engine for queries. Nat Med
26. Wendel K, Quinn TC, Celis L, et al. HIV drug resistance in intravenous drug users undergoing seroconversion. Presented at the Infectious Disease Society of America 2001 Meeting, San Francisco, October 2001.
27. Celentano DD, Vlahov D, Cohn S, et al. Self-reported antiretroviral therapy in injection drug users [see comments]. JAMA
28. Celentano DD, Galai N, Sethi AK, et al. Time to initiating highly active antiretroviral therapy among HIV-infected injection drug users. AIDS
29. Grosskurth H, Mosha F, Todd J, et al. Impact of improved treatment of sexually transmitted diseases on HIV infection in rural Tanzania: randomised controlled trial. Lancet
30. Venturi G, Catucci M, Romano L, et al. Antiretroviral resistance mutations in human immunodeficiency virus type 1 reverse transcriptase and protease from paired cerebrospinal fluid and plasma samples. J Infect Dis
31. Si-Mohamed A, Kazatchkine MD, Heard I, et al. Selection of drug-resistant variants in the female genital tract of human immunodeficiency virus type 1-infected women receiving antiretroviral therapy. J Infect Dis
32. Devereux HL, Youle M, Johnson MA, et al. Rapid decline in detectability of HIV-1 drug resistance mutations after stopping therapy. AIDS
33. Kalichman SC, Kelly JA, Rompa D. Continued high-risk sex among HIV seropositive gay and bisexual men seeking HIV prevention services. Health Psychol.
34. Kalichman SC, Greenberg J, Abel GG. HIV-seropositive men who engage in high-risk sexual behaviour: psychological characteristics and implications for prevention. AIDS Care
35. Kozal M, Amico R, Chiarella J, et al. Prevalence of antiretroviral resistance mutations in patients in clinical care reporting high risk HIV transmission behavior. Presented at the XIV International AIDS Conference, Barcelona, July 2002.
36. Diaz T, Chu SY, Weinstein B, et al. Injection and syringe sharing among HIV-infected injection drug users: implications for prevention of HIV transmission. Supplement to HIV/AIDS Surveillance Group. J Acquir Immune Defic Syndr Hum Retrovirol
. 1998;18(Suppl 1):S76–S81.
37. Reynaud-Maurupt C, Carrieri MP, Gastaud JA, et al. Impact of drug maintenance treatment on injection practices among French HIV-infected IDUs. The MANIF 2000 Study Group. AIDS Care
38. Marks G, Ruiz MS, Richardson JL, et al. Anal intercourse and disclosure of HIV infection among seropositive gay and bisexual men. J Acquir Immune Defic Syndr Hum Retrovirol
39. Ramos A, Hu DJ, Nguyen L, et al. Intersubtype human immunodeficiency virus type 1 superinfection following seroconversion to primary infection in two injection drug users. J Virol
40. Jost S, Bernard MC, Kaiser L, et al. HIV-1 super-infection: AE subtype supplanted by B subtype. Presented at the XIV International AIDS Conference, Barcelona, July 2002.
41. Walker BD. Harnessing the immune system to fight HIV infection. Presented at the XIV International AIDS Conference, Barcelona, July 2002.
42. Bartlet J, Gallant J. 2003 Medical Management of HIV Infection
. Baltimore: Johns Hopkins University, Department of Infectious Diseases; 2003.
43. Hecht FM. Approaches to HIV prevention among seropositive patients in the clinical care setting. Top HIV Med
44. Celentano DD, Munoz A, Cohn S, et al. Dynamics of behavioral risk factors for HIV/AIDS: a 6-year prospective study of injection drug users. Drug Alcohol Depend.
45. Marks G, Richardson JL, Crepaz N, et al. Are HIV care providers talking with patients about safer sex and disclosure?: a multi-clinic assessment. AIDS
© 2004 Lippincott Williams & Wilkins, Inc.