HIV-positive women with classic risk factors for diabetes, including obesity and a family history of diabetes, should be screened by their HIV care providers for the presence of the condition, according to a US study to be published in the May 15th edition of Clinical Infectious Diseases. The recommendation was made after investigators in New York found that HAART regimens which did not include a protease inhibitor were associated with glucose intolerance and diabetes in HIV-positive women, seemingly contradicting earlier studies which found an association between disorders in glucose metabolism and protease inhibitors.
Earlier studies have found evidence of an increased prevalence of glucose intolerance and diabetes amongst HIV-positive individuals in the HAART era and this has been attributed to the use antiretroviral drugs, particular protease inhibitors. However, classic risk factors for diabetes may also have been a significant confounding factor, and most of the earlier studies examining glucose metabolism in HIV-positive individuals have focused on men, or have lacked HIV-negative controls.
Investigators therefore wished to determine the associations of HIV, HAART, and other factors with both diagnosed and undiagnosed diabetes in a cohort of 620 middle-aged women who were either HIV-positive or at risk of HIV infection.
The women were recruited between autumn 2001 and early 2003 to the Menopause Study in the Bronx, New York City. A total of 332 women (54%) were HIV-positive. At an interview all the women provided a personal and family medical history, about their drug use patterns, and if they had ever been diagnosed with diabetes. HIV-positive women were asked about their use of HAART.
In addition, the investigators recruited women to a metabolic substudy. According to their HIV status and use of antiretroviral use the women were allocated into one of four arms: an HIV-negative group; HIV-positive but antiretroviral naïve; HIV-positive and taking a HAART regimen which did not include a protease inhibitor; HIV-positive and taking a HAART regimen including a protease inhibitor.
Previously diagnosed diabetes
Diabetes risk factors were common across the cohort. In total, 458 women (74%) had a body mass index above 25, 82 women (13%) had delivered a baby weighing more than 4.08kg, and 44% (242) had a family history of diabetes.
Diabetes had been previously diagnosed in 79 women (13%). The median age at diabetes diagnosis was 42 years, and 67 women (85%) were taking medication for their diabetes. There was no significant difference in the prevalence of previously diagnosed diabetes between HIV-negative women (13%) and HIV-positive women (12%).
After controlling for age, HIV status, and taking HAART, the factors independently associated with diagnosed diabetes were current methadone treatment (adjusted odds ratio, 1.8), BMI above 25 (adjusted odds ratio, 2.6), family history of diabetes (adjusted odds ratio, 2.7).
Impaired glucose tolerance
Of the women recruited to the metabolic substudy, 12% (26 individuals) had oral glucose tolerance test results indicative of impaired glucose tolerance, and 13 (6%) had results indicating diabetes. Compared to HIV-negative women, HIV-positive women receiving a non-protease inhibitor containing HAART regimen were significantly more likely to have impaired glucose tolerance (p = 0.02).
After controlling for race, HIV infection, use of HAART, use of protease inhibitors, and waist-to-hip ratio, the factors significantly associated with an abnormal oral glucose tolerance test result were age above 50 years, family history of diabetes, physical inactivity, and smoking.
Insulin resistance
The prevalence of insulin resistance was highest amongst HIV-positive women receiving a non-protease inhibitor HAART regimen (26% versus 12% protease inhibitor arm, 9% HIV-negative, 3% HIV-positive, no HAART, p < 0.05). After HIV infection and receipt of antiretrovirals was controlled for, the factors which remained significantly associated with insulin resistance included waist circumference (p < 0.001), Hispanic ethnicity (p = 0.01) and physical inactivity (p = 0.01).
When the investigators restricted their analysis to HIV-positive women they found that waist circumference (p < 0.001) and taking a HAART regimen that did not include a protease inhibitor (p = 0.03) were significantly associated with insulin resistance.
Insulin secretion
Insulin secretion was lower amongst HIV-negative women than HIV-positive women (p = 0.04). There was no difference in the level of insulin secretion, however, between HIV-positive women taking a protease inhibitor and those taking a HAART regimen not including this class of drug. Controlling for enrollment group, the investigators established that the factors associated with lower insulin secretion included older age (p = 0.03) and current use of opiates (p = 0.02).
Why no relationship found between protease inhibitors and glucose intolerance?
“Using an oral glucose tolerance test, we found that non-protease inhibitor HAART, and not protease inhibitor-HAART was associated with impaired glucose tolerance and insulin resistance”, write the investigators.
The investigators suggest that the association between opiate use and glucose intolerance could be a marker for hepatitis C virus infection, a risk factor for diabetes that was not assessed in the study.
They conclude by recommending that all HIV-positive women, with risk factors for diabetes, should be screened by their HIV care providers for the presence of the condition.
Howard AA et al. Disorders of glucose metabolism among HIV-infected women. Clin Infect Dis 40 (on-line edition), 2005.