Secondary Logo

Differences in HIV Disease Progression by Injecting Drug Use in HIV-Infected Persons in Care

Moore, Richard D., MD; Keruly, Jeanne C., MS, CRNP; Chaisson, Richard E., MD

JAIDS Journal of Acquired Immune Deficiency Syndromes: January 1st, 2004 - Volume 35 - Issue 1 - p 46-51
Clinical Science
Free

Background In the United States and many Western countries, injecting drug use continues to be an important cause of HIV infection. This has important clinical and public health implications if injecting drug users (IDUs) have greater barriers to antiretroviral effectiveness than other risk groups. We assessed if there were differences between HIV-infected IDUs and non-IDU patients in the development of AIDS-defining illnesses (ADIs) from the time the patients started their first combination antiretroviral therapy (CART) regimen.

Methods We compared clinical outcomes for IDU patients (n = 827) with those for non-IDU patients (n = 1314) after they started CART. We controlled for financial access, because all patients had access to CART through insurance or a drug assistance program. The incidence (number of ADI cases per 100 person-years) was compared for IDUs and non-IDUs from 1995 through 2002. Incidence ratios were calculated for IDUs compared with non-IDUs. Risk factors for development of ADIs were assessed using negative binomial regression.

Results From 1995–1996 to 2001–2002, there was a decline in ADI incidence among IDUs from 31.9 to 16.2 cases per 100 person-years of follow-up. Over the same time, there was a decline in ADI incidence among non-IDUs from 37.0 to 9.7 cases per 100 person-years. The incidence ratio (incidence among IDUs compared with that among non-IDUs) increased from 0.87 (95% confidence interval [CI], 0.65–1.15) to 1.67 (95% CI, 1.25–2.18) from 1995–1996 to 2001–2002. By negative binomial regression, the incidence ratio for ADIs among IDUs versus non-IDUs increased to 1.45 (95% CI, 1.21–1.75), after 1998, adjusting for differences in demographic, clinical, and treatment factors.

Conclusions The relative incidence of ADIs among IDUs with access to treatment increased ∼50% compared with non-IDUs since 1999. This suggests greater barriers to the effective use of CART for IDUs, resulting in a higher individual and public health burden of clinical HIV disease. It will be important to understand reasons for this growing difference and to implement appropriate interventions to improve the effective use of CART for IDUs.

From Johns Hopkins University School of Medicine, Baltimore, Maryland.

Received for publication June 13, 2003; accepted September 23, 2003.

Supported by the National Institute on Drug Abuse (RO1-DA-11602 and K24-DA-00432).

Reprints: Richard D. Moore, Johns Hopkins University, 1830 East Monument Street, Room 8059, Baltimore, MD 21287 (e-mail: rdmoore@jhmi.edu).

Use of combination antiretroviral therapy (CART) has resulted in a significant decline in AIDS-defining illnesses (ADIs) and death in HIV-infected persons. 1–4 Yet, there is evidence that those persons who acquired HIV infection through injecting drug use may not be achieving the same benefit from CART as those who have not injected drugs. 5–8

We previously reported data from the Johns Hopkins HIV Clinical Cohort that showed that injecting drug users (IDUs) in our cohort had lower rates of adherence to CART than non-IDUs. 9 This lower level of adherence was associated with a lower rate of HIV-1 RNA suppression and smaller increases in CD4 lymphocyte levels than among non-IDU patients. 10 Other researchers have also shown lower antiretroviral adherence rates among HIV-infected IDUs. 11,12

We wanted to assess whether differences in HIV-1 RNA suppression and CD4 cell response are translating into differences in clinical disease progression. If IDUs have not achieved the same benefits of CART as other HIV transmission risk groups, this has important implications in regard to the clinical disease burden in individually HIV-infected IDUs and for the public health in areas such as the northeastern United States, where the rate of HIV transmission from injection drug use is particularly high. To assess the impact of CART on HIV-infected IDUs and non-IDUs, we conducted a further analysis of longitudinal data from the Johns Hopkins HIV Clinical Cohort. We hypothesized that development of an ADI might be increasing in IDUs compared with non-IDUs as the use of CART enters its second 5 years.

Back to Top | Article Outline

METHODS

The Johns Hopkins AIDS Service provides care for a large proportion of HIV-infected patients in the Baltimore metropolitan area. Longitudinal primary care and subspecialty care are integrated on this service. An observational, longitudinal, clinical cohort of patients receiving HIV care in our HIV/AIDS Clinic has been maintained since 1990. 13 Enrollment into the cohort corresponds to enrollment in the HIV/AIDS Clinic. Comprehensive demographic, clinical, therapeutic, and laboratory data are collected at baseline (enrollment) and are subsequently updated at 6-month intervals using structured data collection forms and coding criteria. Professionally trained abstractors update these data using the records from clinic visits and inpatient admissions (at Johns Hopkins and elsewhere), laboratory testing, pharmacy, social services, and all other available clinical sources. A 5% random sample of abstracted data is validated by independent trained individuals, with correction of data and retraining of primary abstractors as necessary. Dropouts from our cohort are relatively uncommon, cumulatively averaging only 9% (defined as >18 months out of care) since 1990. There is no difference in the methods of data collection or in dropout rates for IDUs compared with non-IDUs from our cohort. The Johns Hopkins Institutional Review Board approved this cohort study.

Back to Top | Article Outline

Definitions

We defined HIV transmission risk group based upon the records of the primary care provider, nursing staff, and social worker. If there was >1 possible route of HIV transmission and injection drug use was 1 of these routes, then HIV transmission risk was coded as injection drug use. ADIs were assessed using the 1993 Centers for Disease Control and Prevention's definition of an ADI from outpatient and inpatient records from Johns Hopkins and outside facilities. 14

The antiretroviral drugs that defined first use of CART were based on recent guidelines for the use of antiretroviral therapy for adults. 15 A protease inhibitor, nonnucleoside reverse transcriptase inhibitor, or use of 3 nucleoside reverse transcriptase inhibitors was required. We assessed only those patients who first started CART at enrollment or beyond. Patients who initiated CART before their first visit to our clinic were excluded from analysis.

We restricted this analysis to patients who either had insurance that provided for all available antiretroviral drugs or had received antiretroviral therapy through the Ryan White CARE Act–supported Maryland AIDS Drug Assistance Program.

Back to Top | Article Outline

Analysis

We first calculated the incidence of developing any ADI among patients who had or did not have injection drug use as their mode of HIV transmission. Incidence was calculated as the number of ADI cases per person-years (where person-years was the longitudinal time from the start date of CART to the last known follow-up date for the patient). Incidence was computed biannually from 1995 through 2002. Both unadjusted incidence and adjusted incidence were computed. Adjusted incidence was compared between IDUs and non-IDUs using negative binomial regression analysis. Variables adjusted for included age, sex, race, enrollment CD4 cell count, enrollment HIV-1 RNA level, and history of an ADI before enrollment.

Negative binomial regression was also used to assess the risk of development of an ADI with an interaction term for IDU and calendar year of 1999 or later. This analysis adjusted for the variables listed above plus prior use of antiretroviral drugs before use of the first CART regimen, type of CART regimen (single protease inhibitor, a ritonavir-boosted protease inhibitor, an nonnucleoside reverse transcriptase inhibitor, or a nucleoside reverse transcriptase inhibitor–containing “backbone”), whether the patients achieved a suppressed HIV-1 RNA level (<400 copies/mL), whether the patients had durable suppression of the HIV-1 RNA level (eg, maintained a suppressed HIV-1 RNA level after first HIV-1 RNA level of <400 copies/mL), and the total duration of CART use.

All statistical analyses were done using SAS (SAS Institute, Cary, NC).

Back to Top | Article Outline

RESULTS

A total of 827 IDUs and 1314 non-IDUs were assessed. A comparison of demographic, clinical, and therapeutic variables between IDUs and non-IDUs is shown in Table 1. IDUs were significantly more likely to be African American, and on average, they were younger and had a shorter duration of CART use.

TABLE 1

TABLE 1

The biannual incidence of development of an ADI is shown in Table 2. There was a significant decrease in the incidence of ADI from 1995 through 2002 among both IDUs and non-IDUs. However, a relative increase in ADI rates among IDUs compared with non-IDUs also occurred over this time. Among IDUs, the incidence of an ADI declined from 31.9 to 16.2 cases per 100 person-years from 1995 through 2001–2002. Among non-IDUs, the incidence of ADIs declined from 37.0 to 9.7 cases per 100 person-years over this same period. The unadjusted and adjusted incidence ratio (calculated as the incidence among IDUs divided by the incidence among non-IDUs) significantly increased from 1995–1996 through 2001–2002. The incidence ratio of an ADI adjusted for sex, race, baseline CD4 cell count, and baseline HIV-1 RNA level was 1.03 (95% confidence interval [CI], 0.76–1.43) in 1995–1996 and increased to 1.66 (95% CI, 1.25–2.18) in 2001–2002.

TABLE 2

TABLE 2

Results of the overall multivariate analysis of risk of development of an ADI are shown in Table 3.

TABLE 3

TABLE 3

The interaction term of injection drug use and calendar year of 1999 or later was significant, with a risk ratio of 1.45 (95% CI, 1.21–1.75). This indicates that the relative incidence of an ADI for IDUs versus non-IDUs was 45% higher after 1998 than during or before that year. The main effects show that calendar year of 1999 or later was associated with a reduction in risk to 0.76 (95% CI, 0.66–0.87) and that injection drug use was associated with no further difference in risk after accounting for these factors.

Other covariates that were associated with an increased risk of developing of an ADI were a lower initial CD4 lymphocyte count, higher HIV-1 RNA level, a previous ADI before enrollment, and failure to achieve and maintain an HIV-1 RNA level of <400 copies/mL. IDU patients were less likely to have HIV-1 RNA suppression than were non-IDU patients. Using a threshold of <400 copies/mL, HIV-1 RNA suppression occurred in 354 IDUs (43%) and 464 non-IDUs (56%) (P < 0.001). Durable HIV-1 RNA suppression occurred in 354 IDUs (43%) and 640 non-IDUs (49%) (P < 0.01). In addition, HIV mutations were genotypically measured in our patients. A total of 201 IDUs (24.3% of all IDUs) and 434 non-IDUs (33.0% of all non-IDUs) had genotype assays performed. Among these patients, we found similar rates of selected mutations, including the following: M184V (31% of IDUs and 34% of non-IDUs), K103N, and other nonnucleoside reverse transcriptase inhibitor mutations (31% of IDUs and 30% of non-IDUs); multi–protease inhibitor resistance mutations M46, V82, I84V, and L90M (18% of IDUs and 13% of non-IDUs); and total protease mutations (74% with at least 1 [median, 3], for IDUs; 73% with at least 1 [median, 2], for non-IDUs). Nucleoside analog mutations—M41L, D67N, K70R, L210W, T215Y/F, and K219Q/E—were more common in IDUs (44% with at least 1) than in non-IDUs (34% with at least 1; P < 0.05).

Finally, we compared the rates of individual ADIs to determine if there were differences in specific ADIs that might account for the increasing incidence ratio for IDUs compared with non-IDUs (Table 4). The relative rank of the ADIs was unchanged for IDUs, comparing 1995–1996 with 2001–2002, suggesting that there was no differential increase in any specific ADI in IDUs over this period. IDUs had higher rates of ADIs that tend to occur both relatively earlier (eg, candidal esophagitis and bacterial pneumonia) and later (Mycobacterium avium complex bacteremia and Pneumocystis carinii pneumonia) in HIV disease.

TABLE 4

TABLE 4

Back to Top | Article Outline

DISCUSSION

CART has been highly effective in suppressing HIV-1 RNA levels, increasing CD4 cell counts, and reducing the clinical burden of HIV infection. Several studies demonstrated a decline in the incidence of ADIs after the introduction of CART that affected all HIV transmission risk groups. 1–4 A previous study from the Johns Hopkins HIV Clinical Cohort suggested possible early differences between IDU and non-IDU patients in clinical outcomes after the introduction of CART 6 that had not existed before CART. 16 Other investigators also found early differences between IDU and non-IDU patients in the development of opportunistic illnesses and survival. 5–8 Our current analysis demonstrated that over the long term, IDUs receive relatively less benefit from CART than patients who have other HIV transmission risk factors. In fact, our data indicate that the difference between IDUs and non-IDUs in clinical event rates progressively increases over time and that there is now an ∼1.5-fold difference in clinical effectiveness between these two groups of patients. This difference is occurring even with similar access to treatment, as controlled in our cohort where public insurance and the Ryan White CARE Act–supported Maryland AIDS Drug Assistance Program were similarly available to both IDUs and non-IDUs. In addition, we only assessed patients who had received CART, because prescription of CART may be less likely to occur for IDU patients than for non-IDU patients. 17 The difference between IDUs and non-IDUs also appears to be affecting most ADIs, because no specific ADI was occurring more frequently in IDUs over time.

Illicit drug use is associated with poorer adherence with antiretroviral therapy. This association between drug use and nonadherence was shown previously in our cohort 9,10 and by other researchers. 11,12,18 Unfortunately, we did not have a direct measure of adherence on more than a small subset of our IDU and non-IDU patients; therefore, we were unable to fully adjust for this variable in our analysis. It is active drug use that is most commonly associated with nonadherence. Previous data from our cohort for a subset of patients suggest that continuing drug use and relapse occur in at least 50% of our patients. 9

Although suppression of HIV-1 RNA is a surrogate marker for adherence to CART, 19 other factors related to prior use of antiretroviral therapy and selection of current therapy, pharmacokinetic levels of certain antiretroviral drugs, and infection with resistant virus can affect HIV-1 RNA response. After adjusting for HIV-1 RNA (any suppression and durable suppression of <400 copies/mL) and several other therapeutic variables related to the type and duration of antiretroviral therapy, the interaction between injection drug use and later calendar year remained significant. Because we did not have a direct measure of adherence, it is certainly possible that nonadherence with antiretroviral therapy and prophylaxis for opportunistic infections may play a substantial role in explaining the difference in ADI rates between IDUs and non-IDUs. However, other factors, such as a poorer immune response related to illicit drug use, 20–22 may also serve as barriers to effective antiretroviral therapy for IDUs.

Data from the EuroSIDA Study showed that death preceding the first ADI occurs more frequently in IDUs than in non-IDUs. 23 If death before occurrence of the first ADI substantially decreased in IDUs or increased in non-IDUs, then this could explain the difference in incidence ratios of ADI that were found. In fact, comparing 1995–1998 with 1999–2002, the pre-ADI mortality rate increased from 2.6 to 2.8 cases per 100 person-years among IDUs and from 0.9 to 1.5 cases per 100 person-years among non-IDUs. These small changes would not account for the larger ADI rate differences seen for IDUs and non-IDUs. Finally, we doubt that there was a relative increase in ascertainment of ADIs in IDUs over time, because our methods have not changed and our patient population is geographically stable.

A relative decline in the benefit of CART for IDUs has important implications for the management of HIV infection, because transmission through injection drug use accounts for ∼25% of the cases of HIV infection in the United States 24 and an estimated 2 to 3 million HIV-infected persons worldwide. HIV-infected IDUs who develop opportunistic diseases despite receiving CART will require additional medical treatment, either as an outpatient or in the hospital. This will add to the public health burden because most of these patients have public insurance. 25 The lower rate of HIV suppression that we found among IDUs could result in continued transmission of both wild-type and drug-resistant viruses if there is continued drug use, sharing of drug paraphernalia and unprotected sex. 26,27

Our results are from a single large HIV clinic in Maryland and may not generalize. Most of the drug use in our patients is injection of heroin, and as many as 50% of our patients have continuing active drug use. It is possible that other patient samples may have different patterns of ADI development. However, we would point out that attempts are made to follow antiretroviral therapy and other HIV therapeutic guidelines for all of our patients, and loss to follow-up is no different when comparing our IDU patients with our non-IDU patients. We recommend that other researchers who have access to longitudinal observational clinical data in other settings determine whether there are similar differences between IDU and non-IDU patients.

Our results suggest a higher burden of comorbidities of HIV disease in IDUs. The early effectiveness of CART may be waning in this HIV transmission group. Unfortunately, our results did not reveal why this difference in effectiveness is now occurring, although previous work suggests that poorer adherence may be an important factor. Adjusting for a number of therapeutic, clinical, and laboratory variables did not diminish the association. If this difference in ADI rates between IDU and non-IDU patients generalizes, further studies will be needed to better understand the reasons and to design interventions to improve the effectiveness of antiretroviral treatment for this important group of HIV-infected patients.

Back to Top | Article Outline

REFERENCES

1. Palella Jr, FJ Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med. 1998; 338:853–860.
2. Hogg RS, Heath KV, Yip B, et al. Improved survival among HIV-infected individuals following initiation of antiretroviral therapy. JAMA. 1998; 279:450–454.
3. Moore RD, Chaisson RE. Natural history of HIV infection in the era of combination antiretroviral therapy. AIDS. 1999; 13:1933–1942.
4. Mocroft A, Katlama C, Johnson AM, et al. The changing spectrum of AIDS across Europe 1994–1998. Results from the EuroSIDA Study. Lancet. 2000; 356:291–296.
5. Mocroft A, Madge S, Johnson AM, et al. A comparison of exposure groups in the EuroSIDA Study: starting highly active antiretroviral therapy (HAART), response to HAART, and survival. J Acquire Immune Defic Syndr. 1999; 22:369–378.
6. Poundstone K, Chaisson RE, Moore RD. Differences in HIV progression by sex and injection drug use in the era of highly active antiretroviral therapy. AIDS. 2001; 15:1115–1123.
7. Perez-Hoyos S, del Amo J, Muga R, et al. Effectiveness of highly active antiretroviral therapy in Spanish cohorts of HIV seroconverters: differences by transmission category. AIDS. 2003; 17:353–359.
8. Palepu A, Tyndall M, Yip B, et al. Impaired virologic response to highly active antiretroviral therapy associated with ongoing injection drug use. J Acquir Immune Defic Syndr. 2003; 32:522–526.
9. Lucas GM, Cheever LW, Chaisson RE, et al. Detrimental effects of continued illicit drug use on the treatment of HIV-1 infection. J Acquire Immune Defic Syndr. 2001; 27:251–259.
10. Lucas GM, Gebo KA, Chaisson RE, et al. Longitudinal assessment of the effects of drug and alcohol abuse on HIV treatment outcomes in an urban clinic. AIDS. 2002; 16:767–774.
11. Stein MD, Rich JD, Maksad J, et al. Adherence to antiretroviral therapy among HIV-infected methadone patients: effect of ongoing illicit drug use. Am J Drug Alcohol Abuse. 2000; 26:195–205.
12. Arnsten JH, Demas PA, Grant RW, et al. Impact of active drug use on antiretroviral therapy adherence and viral suppression in HIV-infected drug users. J Gen Intern Med. 2002; 17:377–381.
13. Moore RD. Understanding the clinical and economic outcomes of HIV therapy: the Johns Hopkins HIV Clinical Practice Cohort. J Acquir Immune Defic Syndr. 1998; 17(suppl 1):S38–S41.
14. Centers for Disease Control and Prevention. 1993 revised classification system for HIV-infection and expanded case definition for AIDS among adolescents and adults. MMWR Morb Mortal Wkly Rep. 1992; 41( RR-17):1–19.
15. February 4, 2002, Guidelines for the Use of Antiretroviral Agents in HIV-Infected Adults and Adolescents. Department of Health and Human Services (DHHS) and the Henry J. Kaiser Family Foundation. (http://aidsinfo.nih.gov).
16. Chaisson RE, Keruly JC, Moore RD. Race, sex, drug use, and progression of human immunodeficiency virus disease. N Engl J Med. 1995; 333:751–756.
17. McNaghten AD, Hanson DL, Dworkin MS, et al. Adult/Adolescent Spectrum of HIV Disease Group. Differences in prescription of antiretroviral therapy in a large cohort of HIV-infected patients. J Acquir Immune Defic Syndr. 2003; 32:499–505.
18. Zaccarelli M, Barracchine A, De Longis P, et al. Factors related to virologic failure among HIV-positive injecting drug users treated with combination antiretroviral therapy including two nucleoside reverse transcriptase inhibitors and nevirapine. Aids Patient Care STDS. 2002; 16:67–73
19. Walsh JC, Mandalia S, Gazzard BG. Responses to a 1-month self-report on adherence to antiretroviral therapy are consistent with electronic data and virological treatment outcome. AIDS. 2002; 16:269–277.
20. Pellegino T, Bayer BM. In vivo effects of cocaine on immune cell function. J Neuroimmunol. 1998; 83:139–147.
21. Novick DM, Ochshorn M, Kreek MJ. In vivo and in vitro studies of opiates and cellular immunity in narcotic addicts. Adv Exp Med Biol. 1991; 288:159–170.
22. Madden JJ, Wang Y, Lankford-Turner P, et al. Does reduced DNA repair capacity play a role in HIV infection and progression in the lymphocytes of opiate addicts? J Acquir Immune Defic Syndr. 2002; 31:S78–S83.
23. Mocroft A, Brettle R, Kirk A, et al. Change in the cause of death among HIV positive subjects across Europe: results from the EuroSIDA Study. AIDS. 2002; 16:1663–1671.
24. HIV/AIDS Surveillance Report. U.S. HIV and AIDS cases reported through December 2000, year-end edition. 2000;12 (accessed January 31, 2003, at 222.cdc.gov/hiv/stats/hasr/202.htm).
25. Bozzette SA, Joyce G, McCaffrey DF, et al. Expenditures for the care of HIV-infected patients in the era of highly active antiretroviral therapy. N Engl J Med. 2001; 344:817–823.
26. Vlahov D, Safaien M, et al. Sexual and drug risk-related behaviors after initiating highly active antiretroviral therapy among injection drug users. AIDS. 2001; 15:2311–2316.
27. Miquez-Burano MJ, Peneda-Medina L, Lecusay R, et al. Continued high risk behaviors in HIV infected drug abusers. J Addict Dis. 2002; 21:67–80.
Keywords:

opportunistic illness; antiretrovirals; injecting drug use

© 2004 Lippincott Williams & Wilkins, Inc.