Skip Navigation

Association of Hepatitis C Infection and Antiretroviral Use with Diabetes Mellitus in Drug Users

  1. Andrea A. Howard1,2,
  2. Robert S. Klein1,2, and
  3. Ellie E. Schoenbaum1,2
  1. 1AIDS Research Program, Department of Epidemiology and Social Medicine, Bronx, New York
  2. 2Division of Infectious Diseases, Department of Medicine, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, New York
  1. Reprints or correspondence: Dr. Andrea A. Howard, AIDS Research Program, Montefiore Medical Center, 111 E. 210th St., Bronx, NY 10467 (ahoward{at}montefiore.org).
 
Next Section

Abstract

We performed a cross-sectional analysis of factors associated with diabetes mellitus among 557 drug users who were recruited from a methadone treatment program and were participating in a longitudinal study of hepatitis C virus (HCV) infection. We found that HCV infection was strongly associated with diabetes mellitus (adjusted odds ratio [OR], 2.9; 95% confidence interval [CI], 1.3–6.4) after controlling for age, race, unemployment, and body mass index. Among HIV-infected drug users, receipt of antiretroviral therapy (ART) for >1 year was also associated with diabetes (adjusted OR for receipt of ART without a protease inhibitor, 4.1 [95% CI, 1.1–15.5]; adjusted OR for receipt of ART including a protease inhibitor, 5.5 [95% CI, 1.5–20.4]).

Hepatitis C virus (HCV) infection is the most common chronic bloodborne infection in the United States [1]. Chronic HCV infection has been associated with type 2 diabetes mellitus in several observational studies [2–11], particularly among individuals aged ⩾40 years [11]. However, most of these studies did not consider such factors as illicit drug use and HIV infection, which may have confounded the observed relationship. Because of their common routes of transmission, injection drug users are at increased risk for infection with both HCV and HIV [12]. An increased prevalence of diabetes among HIV-infected individuals has also been reported, which has largely been attributed to protease inhibitor (PI) use [13–15]. However, in most studies of diabetes in HIV-infected individuals, the role of HCV infection as a cofactor has not been determined. The objective of this study was to determine the relationship between HCV infection, PI use, and diabetes in a cohort of current and former drug users with and without HIV infection in the Bronx, New York.

Previous SectionNext Section

Patients and Methods

Study participants. We performed a cross-sectional analysis of factors associated with diabetes mellitus among 557 current and former drug users. The subjects were recruited from methadone treatment and were participating in a longitudinal study of HCV infection. Participants were observed with semiannual research visits, at which a standardized interview was conducted that elicited information on sociodemographic characteristics, drug use behaviors, and medical history, including receipt of antiretroviral therapy (ART). At each visit, participants were asked whether they had diabetes. Participants were also screened for alcoholism by means of the CAGE questionnaire [16], with a positive response to ⩾2 items indicating alcoholism [17]. In addition, each participant's height and weight were measured, permitting calculation of the Quetelet body mass index. All subjects provided written informed consent, and the study was approved by the Institutional Review Board of Montefiore Medical Center (Bronx, NY).

Laboratory testing. HIV-1 antibody testing was conducted by EIA with Western blot test confirmation. For HIV-seropositive participants, T lymphocyte subset assays were performed by flow cytometry, and HIV-1 loads were measured by the Quantiplex HIV-1 RNA v3.0 Assay (Chiron), which has a lower limit of detection of 50 copies/mL. HCV antibody testing was performed by EIA (ELISA 3.0; Ortho Diagnostic Systems). HCV RNA quantification was performed by PCR using the COBAS Amplicor HCV Monitor assay, version 2.0 (Roche Diagnostics), which has a lower limit of detection of 600 IU/mL. HCV genotype was assessed by examination of E1 Haming distances between subjects by using prototype sequences acquired from GenBank, as described elsewhere [18, 19].

Statistical analysis. The associations of diabetes mellitus with sociodemographic characteristics, drug and alcohol use behaviors, and clinical variables, including HCV infection and HIV serostatus, were determined for the entire study sample. In addition, the associations of diabetes mellitus with ART use, CD4+ lymphocyte count, and HIV load were determined for the HIV-infected subgroup. Diabetes was defined by self-report of a diabetes diagnosis, and HCV infection was defined by positive HCV antibody status for all analyses. Comparisons were performed using the χ2 test or Fisher's exact test, as appropriate.

Multivariate logistic regression analysis was performed to assess independent predictors of diabetes mellitus and included HCV antibody status as well as established diabetes risk factors, including age, race/ethnicity, unemployment, and body mass index [20]. A separate multivariate logistic regression analysis was performed to assess independent predictors of diabetes among HIV-infected participants and included ART use, age, race/ethnicity, unemployment, and body mass index. Corresponding ORs and 95% CIs were computed. SPSS software, version 10.0 (SPSS), was used for all analyses.

Previous SectionNext Section

Results

Characteristics of the participants are listed in table 1. Of the 557 participants, 418 (75%) were infected with HCV; of these patients, 313 (75%) had a detectable HCV RNA level. Compared with the remainder of the study sample, the HCV-infected participants were significantly less likely to be Hispanic and were more likely to be male, older, unemployed, and HIV seropositive; also, they were more likely to report a history of injection drug use. Two hundred twenty-eight patients (41%) in the study sample were HIV seropositive. Among 216 HIV-infected persons (95%) with available data, the median CD4+ lymphocyte count was 283 cells/mm3 (range, 3–2730 cells/mm3), and the mean HIV load was 3.1 log10 copies/mL. Although 110 (48%) of 228 HIV-infected individuals reported a history of any use of a PI, prolonged exposure was uncommon: only 16 subjects (7%) reported taking a PI for >1 year.

Table 1
Table 1

Characteristics of participants in a study of the association of hepatitis C virus (HCV) infection and antiretroviral use with diabetes mellitus in drug users.

Of the 557 drug users, 72 (13%) reported a diagnosis of diabetes mellitus. The prevalence of diabetes was greatest among participants who were older, were nonwhite, were unemployed, had a high body mass index, and reported a history of injection drug use (table 2). The prevalence of diabetes was also higher among persons who were infected with HCV, compared with those who were not infected with HCV (15% vs. 6%; P = .004). An association between diabetes and either a history of alcoholism or current alcoholism was not detected.

Table 2
Table 2

Factors associated with diabetes among 557 drug users.

Because a single undetectable HCV RNA level does not exclude viremia [21], to assess the association between active HCV infection and diabetes, we repeated analyses including only persons with known active HCV infection (i.e., detectable HCV RNA levels) and those known to be uninfected (i.e., negative for HCV antibody). Compared with uninfected individuals, individuals with active HCV infection were nearly 3 times as likely to have diabetes (OR, 2.7; 95% CI, 1.2–5.8). An association between diabetes and HCV genotype was not detected.

To examine the relationship between HCV infection, injection drug use, and diabetes, a multivariate logistic regression model was constructed that included injection drug use history and HCV infection as covariates, with diabetes as the outcome. In this model, HCV infection was independently associated with diabetes (adjusted OR, 2.6; 95% CI, 1.1–6.3), whereas injection drug use was not (adjusted OR, 1.2; 95% CI, 0.6–2.6).

To further examine the association between HCV infection and diabetes, a multivariate analysis was performed that included established risk factors for diabetes mellitus. As shown in table 3, age, race, unemployment, body mass index, and HCV infection remained independently associated with diabetes. Individuals with HCV infection were nearly 3 times as likely as those without HCV infection to have diabetes (adjusted OR, 2.9; 95% CI, 1.3–6.4). These estimates were essentially unchanged when either a history of alcoholism or current alcoholism was included in the multivariate model.

Table 3
Table 3

Relative odds of diabetes among drug users, calculated by multivariate logistic regression analysis.

The association between HCV infection and diabetes remained significant after stratifying by HIV serostatus. Among HIV-seronegative participants, the prevalence of diabetes was higher in HCV-infected individuals than among uninfected individuals (17% vs. 7%; P = .01). Among HIV-seropositive participants, the prevalence of diabetes was also higher among HIV-HCV-coinfected individuals than among individuals with HIV infection alone (13% vs. 0%; P = .052, by Fisher's exact test).

Among HIV-infected participants, the prevalence of diabetes was also greater among persons who received ART for >1 year and among those who had an undetectable viral load (table 2). Although any history of PI use was not associated with diabetes, PI use for >1 year was.

The association between ART and diabetes was examined further by multivariate analysis that included established risk factors for diabetes (table 4). Because all HIV-infected patients with diabetes were also infected with HCV, we were unable to include HCV infection in the model. Compared with individuals who had received ART for ⩽1 year, individuals who had received ART without a PI for >1 year were 4 times as likely to have diabetes (adjusted OR, 4.1; 95% CI, 1.1–15.5), and those who had received ART including a PI for >1 year were >5 times as likely to have diabetes (adjusted OR, 5.5; 95% CI, 1.5–20.4). These estimates were essentially unchanged when detectable HIV load was also included in the multivariate model.

Table 4
Table 4

Relative odds of diabetes among HIV-infected drug users, calculated by multivariate logistic regression analysis.

Previous SectionNext Section

Discussion

In this cohort of current and former drug users, HCV infection was strongly associated with diabetes mellitus after controlling for other established risk factors for type 2 diabetes and HCV infection. This finding is consistent with the findings of previous observational studies, which have demonstrated an increased prevalence of diabetes among individuals who have chronic HCV infection (21%–50%), compared with individuals who have other forms of liver disease (2%–12%) and other control groups (2%–6%) [2–10]. Previous studies of HCV infection and diabetes either did not include drug users or did not consider HIV serostatus; the present study demonstrates that the association between diabetes and HCV infection is independent of injection drug use and HIV infection. After adjusting for potential confounding variables, we found that drug users with HCV infection remained nearly 3 times as likely as those without HCV infection to have diabetes mellitus. A significant association between HCV infection and diabetes was also present among the subset of drug users with HIV infection; in fact, all of the HIV-seropositive patients who had diabetes were coinfected with HCV.

In addition, we found that, among HIV-infected drug users, exposure to ART for >1 year was also associated with diabetes mellitus. Although the prevalence of diabetes was greatest among individuals treated with a PI, an increased prevalence of diabetes among those who received a PI-sparing regimen was also found. The observed association between PI use and diabetes is consistent with previous studies [13–15]; this has been attributed to a direct inhibitory effect on cellular glucose transport by PI medications [22]. Insulin resistance has been reported among PI-naive persons with HIV infection in association with fat redistribution [23–25], which may occur as a result of nucleoside analogue-induced mitochondrial toxicity [26]. Both insulin resistance and peripheral fat wasting have also been associated with chronic HCV infection in HIV-infected persons receiving ART [27]. Because body composition was not measured in this study, we were unable to assess the extent to which changes in fat distribution may have accounted for our findings. Body mass index measurements demonstrated that obesity was also independently associated with diabetes in this population.

Our study has several limitations. First, we relied on self-report rather than a laboratory screening test to measure diabetes prevalence. If participants who had HCV infection or who used ART were more likely to be screened for and thus self-report diabetes than were those without these characteristics, then differential misclassification may have contributed to the observed increase in diabetes risk associated with HCV infection and ART use. In addition, information about family history of diabetes was not available; thus, we could not control for this risk factor in our analysis. Furthermore, the extent of liver damage among HCV-infected participants could not be determined, because liver biopsies were not performed. Finally, as a result of the cross-sectional nature of this study, causal relationships between HCV infection or ART use and diabetes could not be established. Despite these limitations, our findings support the hypothesis that chronic HCV infection and PI use play significant roles in the development of diabetes mellitus.

In summary, we found an increased prevalence of diabetes mellitus among drug users with HCV infection and among those who had received ART for >1 year. The biological mechanism by which HCV infection and diabetes are associated remains unknown but appears to be independent of HIV infection and injection drug use. Additional studies are needed to determine the temporal relationship between these disorders and to determine whether, among HIV-infected individuals, HCV coinfection and PI use have a synergistic effect on the risk of diabetes. Given that ∼2.7 million persons in the United States are chronically infected with HCV [1], and given that an estimated 240,000 of these persons are coinfected with HIV [12], prospective studies on this subject are a high priority for additional research.

Previous SectionNext Section

Acknowledgments

All hepatitis C assays were performed in the laboratory of David Thomas, under the supervision of Dale Netski, at The Johns Hopkins University School of Medicine, Baltimore, Maryland.

Previous SectionNext Section

Footnotes

  • Financial support: National Institute of Drug Abuse (grants RO1 DA13248 and R01 DA04347).

  • Received September 19, 2002.
  • Accepted January 30, 2003.
Previous Section
 

References

  1. ↵
    1. Alter MJ,
    2. Kruszon-Moran D,
    3. Nainan OV,
    4. et al
    . The prevalence of hepatitis C virus in the United States, 1988 through 1994. N Engl J Med 1999;341:556-62.
    CrossRefMedlineWeb of Science
  2. ↵
    1. Allison MED,
    2. Wreghitt T,
    3. Palmer CR,
    4. Alexander GJM
    . Evidence for a link between hepatitis C virus infection and diabetes mellitus in a cirrhotic population. J Hepatol 1994;21:1135-9.
    CrossRefMedlineWeb of Science
    1. Macedo G
    . Cirrhosis and diabetes: C, the difference. Am J Gastroenterol 1999;94:3088-9.
    MedlineWeb of Science
    1. Zein N,
    2. Abdulkarim AS,
    3. Wiesner RH,
    4. Egan KS,
    5. Persing DH
    . Prevalence of diabetes mellitus in patients with end-stage liver cirrhosis due to hepatitis C, alcohol, or cholestatic disease. J Hepatol 2000;32:209-17.
    MedlineWeb of Science
    1. Mason AL,
    2. Lau JLN,
    3. Hoang N,
    4. et al
    . Association of diabetes mellitus and chronic hepatitis C virus infection. Hepatology 1999;29:328-33.
    CrossRefMedlineWeb of Science
    1. Fraser GM,
    2. Harman I,
    3. Meller N,
    4. Niv Y,
    5. Porath A
    . Diabetes mellitus is associated with chronic hepatitis C but not chronic hepatitis B infection. Isr J Med Sci 1996;32:526-30.
    MedlineWeb of Science
    1. Knobler H,
    2. Schihmanter R,
    3. Zifroni A,
    4. Fenakel G,
    5. Schattner A
    . Increased risk of type 2 diabetes in noncirrhotic patients with chronic hepatitis C virus infection. Mayo Clin Proc 2000;75:355-9.
    Abstract/FREE Full Text
    1. Caronia S,
    2. Taylor K,
    3. Pagliaro L,
    4. et al
    . Further evidence for the association between non-insulin dependent diabetes mellitus and chronic hepatitis C virus infection. Hepatology 1999;30:1059-63.
    CrossRefMedlineWeb of Science
    1. Grimbert S,
    2. Valensi P,
    3. Marchal-Lévy C,
    4. et al
    . High prevalence of diabetes mellitus in patients with chronic hepatitis C. Gastroenterol Clin Biol 1996;20:544-8.
    MedlineWeb of Science
  10. ↵
    1. Ozylikan E,
    2. Arslan M
    . Increased prevalence of diabetes mellitus in patients with chronic hepatitis C virus infection. Am J Gastroenterol 1996;91:1480-1.
    MedlineWeb of Science
  11. ↵
    1. Mehta SH,
    2. Brancati FL,
    3. Sulkowski MS,
    4. Strathdee SA,
    5. Szklo M,
    6. Thomas DL
    . Prevalence of type 2 diabetes mellitus among persons with hepatitis C virus infection in the United States. Ann Intern Med 2000;133:592-9.
    Abstract/FREE Full Text
  12. ↵
    1. Sulkowski MS,
    2. Mast EE,
    3. Seeff LB,
    4. Thomas DL
    . Hepatitis C virus infection as an opportunistic disease in persons infected with human immunodeficiency virus. Clin Infect Dis 2000;30:77-84.
    CrossRef
  13. ↵
    1. Carr A,
    2. Samaras K,
    3. Thorisdottir A,
    4. Kaufmann GR,
    5. Chisholm DJ,
    6. Cooper DA
    . Diagnosis, prediction, and natural course of HIV-1 protease-inhibitor-associated lipodystrophy, hyperlipidaemia, and diabetes mellitus: a cohort study. Lancet 1999;353:2093-9.
    CrossRefMedlineWeb of Science
    1. Behrens G,
    2. Dejam A,
    3. Schmidt H,
    4. et al
    . Impaired glucose tolerance, beta cell function and lipid metabolism in HIV patients under treatment with protease inhibitors. AIDS 1999;13:63-70.
    CrossRef
  15. ↵
    1. Dever LL,
    2. Oruwari PA,
    3. Figueroa WE,
    4. O'Donovan CA,
    5. Eng RH
    . Hyperglycemia associated with protease inhibitors in an urban HIV-infected minority patient population. Ann Pharmacother 2000;34:580-4.
    Abstract
  16. ↵
    1. Ewing JA
    . Detecting alcoholism: the CAGE questionnaire. JAMA 1984;252:1905-7.
    Abstract/FREE Full Text
  17. ↵
    1. Mayfield D,
    2. McLeod G,
    3. Hall P
    . The CAGE questionnaire: validation of a new alcoholism screening instrument. Am J Psychiatry 1974;131:1121-3.
    CrossRefMedlineWeb of Science
  18. ↵
    1. Ray SC,
    2. Arthur RR,
    3. Carella A,
    4. Bukh J,
    5. Thomas DL
    . Genetic epidemiology of hepatitis C virus throughout Egypt. J Infect Dis 2000;182:698-707.
    Abstract/FREE Full Text
  19. ↵
    1. Thomas DL,
    2. Rich JD,
    3. Schuman P,
    4. et al
    . Multicenter evaluation of hepatitis C RNA levels among female injection drug users. J Infect Dis 2001;183:973-6.
    Abstract/FREE Full Text
  20. ↵
    1. National Diabetes Data Group
    1. Cowie CC,
    2. Eberhardt M
    . Sociodemographic characteristics of persons with diabetes. In: National Diabetes Data Group, editor. Diabetes in America. 2nd ed. Bethesda, MD: National Institute of Diabetes and Digestive and Kidney Diseases; 1995. p. 85-116.
  21. ↵
    1. National Institutes of Health Consensus Development Conference statement
    . Management of hepatitis C: 2002. 10-12 June. 2002. Available at: http://consensus.nih.gov/cons/116/116cdc_intro.htm. Accessed 24 April 2003.
  22. ↵
    1. Murata H,
    2. Hruz PW,
    3. Mueckler M
    . The mechanism of insulin resistance caused by HIV protease inhibitor therapy. J Biol Chem 2000;275:20251-4.
    Abstract/FREE Full Text
  23. ↵
    1. Hadigan C,
    2. Miller K,
    3. Corcoran C,
    4. Anderson E,
    5. Basgoz N,
    6. Grinspoon S
    . Fasting hyperinsulinemia and changes in regional body composition in human immunodeficiency virus-infected women. J Clin Endocrinol Metab 1999;84:1932-7.
    Abstract/FREE Full Text
    1. Shikuma CM,
    2. Waslien C,
    3. McKeague J,
    4. et al
    . Fasting hyperinsulinemia and increased waist-to-hip ratios in non-wasting individuals with AIDS. AIDS 1999;13:1359-65.
    CrossRefMedlineWeb of Science
  25. ↵
    1. Hadigan C,
    2. Meigs JB,
    3. Corcoran C,
    4. et al
    . Metabolic abnormalities and cardiovascular disease risk factors in adults with human immunodeficiency virus infection and lipodystrophy. Clin Infect Dis 2001;32:130-9.
    Abstract/FREE Full Text
  26. ↵
    1. Brinkman K,
    2. Smeltink JA,
    3. Romijn JA,
    4. Reiss P
    . Mitochondrial toxicity by nucleoside-analogue reverse-transcriptase inhibitors is a key factor in the pathogenesis of antiretroviral-therapy-related lipodystrophy. Lancet 1999;354:1112-5.
    CrossRefMedlineWeb of Science
  27. ↵
    1. Duong M,
    2. Petit JM,
    3. Piroth L,
    4. et al
    . Association between insulin resistance and hepatitis C virus chronic infection in HIV-hepatitis C virus-coinfected patients undergoing antiretroviral therapy. J Acquir Immune Defic Syndr 2001;27:245-50.
    MedlineWeb of Science
| Table of Contents

This Article

  1. Clin Infect Dis. (2003) 36 (10): 1318-1323. doi: 10.1086/374838
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most