Abstract: Whether HIV itself boosts diabetes risk remains controversial, but no one questions that HIV populations bear a weighty load of classic diabetes risk factors, including HCV infection, nonwhite race, older age, and treatment with drugs that affect glucose metabolism. Although currently favored antiretrovirals carry little or no risk of dysregulated glucose, many people with HIV take other drugs that may heighten diabetes risk, including corticosteroids, thiazides, statins, atypical antipsychotics, and niacin. Obesity remains the biggest driver of the U.S. diabetes epidemic. Although a recent CDC study found lower obesity prevalence in people with HIV than in the general population, nearly one quarter of the HIV group was obese. The CDC calculated that 40% of HIV-positive U.S. women are obese. Meta-analysis linked HCV infection to a two thirds higher diabetes prevalence or incidence, and one quarter of HIV-positive people in the United States have HCV infection.

Diabetes has many causes and many consequences (Figure 1). The immediate cause -- upset glucose metabolism -- is solitary. But factors that may contribute to runaway glucose readings are manifold. U.S. Health Resources and Services Administration (HRSA) HIV guidelines list nine HIV risk factors,1 including two HIV-specific factors, use of protease inhibitors (PIs) or nucleoside reverse transcriptase inhibitors (NRTIs), though currently favored antiretrovirals are kinder to glucose metabolism (Figure 1).

Diabetes Risk and Consequences
Diabetes Risks and Consequences Figure 1. Risk factors for type 2 diabetes (left) may be highly prevalent in HIV populations. Consequences of diabetes mellitus (right) often prove more frequent and severe in HIV groups than in the general population. (Pancreas from Servier PowerPoint image bank.)

Whether HIV itself boosts diabetes risk remains hotly contested, but no one doubts that people with HIV bear a weighty risk burden. Besides taking antiretrovirals, HIV populations often have demographic and clinical features that pose a risk of unruly glucose, including nonwhite race, hepatitis C virus (HCV) coinfection, abnormal lipids, high blood pressure, and lipodystrophy. Older age ranks as an inescapable diabetes risk factor, and HIV populations across the world are aging. Although HIV-positive people in the United States tend to be leaner overall than people without HIV, obesity remains a problem in HIV-positive U.S. women.

Myriad studies in people with and without HIV trace links between diabetes and dire clinical consequences:1-6

  • Cardiovascular disease (coronary artery disease, stroke, peripheral vascular disease)
  • Kidney disease
  • Liver disease
  • Cognitive dysfunction and dementia
  • Peripheral neuropathy
  • Retinopathy

National Institutes of Health (NIH) metabolic experts Colleen Hadigan and Sarah Kattakuzhy counsel that "disorders of glucose metabolism" in people with HIV can culminate in "increased prevalence and worsened outcomes in a diverse array of conditions ranging from neurocognitive changes to renal impairment and albuminuria."2

Diabetes also has a telling impact on the ability of people with HIV to stay on the job, according to a study of 376 working-age HIV-positive people in France.7 Among these people diagnosed with HIV from 2004 through 2010, having diabetes boosted the risk of quitting work almost 6 times (adjusted hazard ratio [aHR] 5.7, 95% confidence interval [CI] 1.7 to 18.8, P = 0.005). The finding may sound self-evident, except that HIV disease severity had no impact on work.7 Neither did antiretroviral therapy or HIV-related discrimination. And diabetes proved a stronger job-ender than hypertension or depression.

Weighty Impact of Body Mass Index

Obesity and overweight stand at the top of any diabetes risk list, yet HIV populations generally weigh less than HIV-negative comparison groups. But HIV-positive women in the United States often do weigh too much. In a nationally representative 2009 sample of people with HIV, the Centers for Disease Control and Prevention (CDC) found that 23% are obese.8 In contrast, obesity prevalence stood at 36% in the U.S. general population. Regardless of that difference, the finding that one quarter of U.S. residents with HIV are obese is ominous.

The CDC calculated that 40% of HIV-positive U.S. women are obese. Only 17% of HIV-positive men met standard obesity criteria in this analysis. Statistical analysis adjusted for age, race, poverty, years since HIV diagnosis, and durable viral suppression determined that HIV-positive women had a twice higher obesity rate than HIV-positive men (adjusted prevalence ratio 2.12, 95% CI 1.13 to 1.89). Almost everyone in this HIV sample, 95%, was taking antiretroviral therapy, 41% were non-Hispanic black, 35% non-Hispanic white, and 19% Hispanic.

Obesity emerges as a routine diabetes risk factor in HIV populations. In a 16,632-person DAD Study analysis, every additional mg/kg2 of body mass index (BMI) independently raised the risk of newly diagnosed diabetes 10% (incidence rate ratio 1.10, 95% CI 1.08 to 1.13, P < 0.001).9 In an Italian comparison of 4249 people with HIV and 9148 healthy controls, diabetes prevalence rose from 3.2% in normal-weight HIV-positive people to 3.9% in overweight people with HIV and to 12.7% in the obese.10 Diabetes prevalence in the control group measured 1.1% in those of normal weight, 3.1% in the overweight, and 7.8% in the obese.

A person with HIV doesn't have to be obese, or even greatly overweight, to run a high risk of insulin resistance and diabetes because of weight abnormalities. Unhealthy visceral and subcutaneous fat build-ups affected insulin resistance in an analysis of 926 HIV-positive people representative of the U.S. HIV population in the Fat Redistribution and Metabolic Change in HIV Infection (FRAM) study.11 Body mass index in this HIV group (73% men, mean age 42.7) averaged only 25.1 kg/m2, on the lower end of overweight spectrum of 25 to 30 kg/m2. Splitting HIV-positive people into three groups with the highest, medium, and lowest visceral and subcutaneous adipose tissue levels (VAT and SAT), FRAM investigators used multivariable analysis to determine that those in the highest VAT tertile had 3-fold higher odds of insulin resistance (HOMA above 4) than those in the lowest tertile (odds ratio [OR] 3.12, 95% CI 2.0 to 4.8, P < 0.0001). HIV-positive people in the highest trunk SAT tertile had twice higher odds of insulin resistance than those in the lowest tertile (OR 2.09, 95% CI 1.36 to 3.19, P = 0.001). A study of 345 antiretroviral-treated adults in Portugal confirmed a link between lipodystrophy and abnormal insulin and glucose.12

Some (but not all) studies buttress these intuitive associations between weight, fat, insulin resistance, and diabetes. A Women's Interagency HIV Study (WIHS) of 178 women with and at risk for HIV traced significant ties between higher BMI (P < 0.001) or more kilocalories from sweets (P = 0.025) and greater HOMA-measured insulin resistance.13 "Heavy-intensity physical activity" was associated with lower HOMA values (P = 0.006). The link between higher BMI and insulin resistance remained significant in an analysis limited to women with HIV (P < 0.001), while heavy-intensity activity proved marginally associated with lower HOMA (P = 0.06).

A cross-sectional Boston study of 207 antiretroviral-treated adults (more than 80% men) found an inverse association between Mediterranean Diet Score and HOMA-calculated insulin resistance -- the higher (better) the score, the lower the insulin resistance (beta = -0.15, P = 0.03).14 The inverse association between Mediterranean Diet Score and insulin resistance held true in study participants with fat redistribution (beta = -0.13, P = 0.02).

But in an Australian study of 84 antiretroviral-treated men with lipodystrophy, dietary fat intake was not tied to total cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides, fasting insulin, glucose, or HOMA-measured insulin resistance.15 And fat subtypes could not be linked to fasting insulin, insulin resistance, glucose, total cholesterol, HDL cholesterol, or triglycerides.

But no one questions the central role of diet and exercise in maintaining a healthy body weight to avoid or control abnormal glucose. In the section on insulin resistance and diabetes, 2014 HRSA guidelines for HIV care stress the benefits of both a good diet (with advice from a dietician) and regular cardiovascular exercise.1 HRSA lists "habitual physical inactivity" as a diabetes risk factor. For pointers on integrating diet and exercise into a diabetes care program, see the box in this article.

Is HIV Itself a Diabetes Risk Factor?

Worthy statisticians who tried to disentwine HIV from other variables as a diabetes risk factor have reached two conclusions on whether HIV alone can lead to diabetes: Yes, it can. And no, it can't. One problem in any such analysis, no matter how large or refined, is residual confounding -- the unassailable possibility that factors not melded into the covariate mix are swaying results one way or the other. And HIV populations typically shoulder a fugal array of diabetes risk factors. So it may be impossible to say whether HIV infection -- independently of other risk factors common to people with HIV -- propels them toward a diabetes diagnosis.

Certainly one can imagine a rationale for why HIV, by itself, makes diabetes more likely. NIH metabolic mavens Colleen Hadigan and Sarah Kattakuzhy note that HIV-induced inflammation riles chemokines that regulate insulin and so could explain why HIV abets diabetes development.2 As an example, they cite an AIDS Clinical Trials Group (ACTG) study that linked systemic inflammation after antiretroviral therapy (ART) began to new-onset diabetes.16 This case-control study compared 55 people diagnosed with diabetes within 48 weeks of starting their first antiretroviral regimen and 55 controls matched for baseline BMI and race/ethnicity who remained free of diabetes. After adjustment for 48-week BMI and CD4 count, baseline marker levels, and indinavir use, people with higher initial levels of inflammation signals hsCRP, sTNFR1, and sTNFR2 had higher odds of a diabetes diagnosis. After further adjustment for week-48 glucose, higher sTNFR1 remained an independent predictor of diabetes (highest versus lowest quartile aOR 23.2, 95% CI 1.28 to 423, P = 0.03).

A more recent analysis of 3695 antiretroviral-treated people in the SMART and ESPRIT studies buttressed these findings, discerning links between higher inflammatory marker levels and newly diagnosed diabetes.17 During an average 4.6 years of follow-up, diabetes developed in 137 people. A regression model adjusted for baseline diabetes- and HIV-related factors determined that every doubling of IL-6 or hsCRP, two inflammation markers, nudged up the risk of incident diabetes (aHR 1.29, 95% CI 1.08 to 1.55, P = 0.005; aHR 1.22, 95% CI 1.10 to 1.36, P < 0.001).

But a study of 214 antiretroviral-naive people in France found no significant links between these same inflammation markers and insulin resistance in an analysis adjusted for age, sex, geographic origin, BMI, and waist circumference.18 This French analysis differs from the ACTG and SMART/ESPRIT studies in focusing on antiretroviral-naive people and in choosing a softer endpoint -- insulin resistance rather than new-onset diabetes. But one would prefer concordant results in all three studies to nail down the HIV→inflammation →diabetes hypothesis.

And if you look at studies weighing the impact of HIV as an independent diabetes risk factor, results are all over the map: Some find HIV does boost diabetes risk,10,19 others find it does not,20-22 and others find a lower diabetes risk in antiretroviral-treated people with HIV than in comparison groups.19,21,23 Two studies reached two of these three conclusions, depending on the study period and antiretroviral use.19,21

Two big cohort studies -- one in Italy and one in Denmark -- tagged HIV infection as an independent diabetes risk factor in comparisons with general-population groups. A 2008 Italian study compared 4249 HIV-positive people attending a Milan infectious diseases clinic and 9148 healthy controls from 15 Italian regions (Table 1).10 About three quarters of both groups were men, median age was significantly younger in the HIV group (45.7 versus 46.6, P < 0.0001), and BMI was significantly lower (23.2 versus 25.3 kg/m2, P < 0.0001). Median triglycerides were higher in the HIV group (126 versus 100 mg/dL, P < 0.0001). Diabetes prevalence proved significantly greater in the HIV group (4.1% versus 2.5%, P < 0.0001). After statistical adjustment for age, gender, BMI, triglycerides, and total, HDL, and low-density lipoprotein (LDL) cholesterol, chances of prevalent diabetes were 55% higher in people with HIV (aOR 1.55, 95% CI 1.02 to 2.31, P = 0.035).

Table 1. Independent Risk Factors for New-Onset or Prevalent Diabetes in People With HIV
Authorn, Year(s), AgeM/F, Race, BMI, ARTRisk Factors* (95% CI)
Yoon,25 New York City (case-control study)49, 1991-2000, A 45.163/37, 39% black, 35% Hispanic, 27% white, M 28.5, 82% NRTI use, 10% NNRTI use, 71% PI useBMI: OR 1.13/kg/m2 (1.03-1.23)
Family history: OR 5.55 (1.41-21.85)
ALT: OR 1.16/10-unit higher (1.03-1.30) PI use, HCV not associated
Rasmussen,19 Denmark (HIV+/HIV- matched-cohort comparison)3540, 1996-2010, M 38.784/26, 97.5% white, 18% overweight or obese, 86% on ARTHIV+ vs HIV- in 1996-1998: IRR 2.83 (1.57-5.09)
HIV+ vs HIV- in 1999-2010 before ART begins: IRR 0.45 (0.21-0.96)
Obese vs normal weight: IRR 9.25 (5.37-15.94) Age 60+ vs <30: IRR 8.16 (1.91-34.74)
Lipoatrophy: IRR 2.30 (1.39-3.80)
SQV use: IRR 1.53 (1.01-2.34)
d4T use: IRR 1.81 (1.19-2.75)
De Wit,24 Europe, U.S., Australia, Argentina (DAD Study)33,389, 1999-2005, M 3874/26, 13% nonwhite, M 23.0, 73% on ART, 58% on PI

A 33,389-person DAD Study analysis linked nevirapine use to lower risk of new diabetes.24
d4T use: RR 1.13 (1.08-1.15)
ZVD use: RR 1.05 (1.01-1.10)
ddI use: RR 1.06 (1.01-1.11)
RTV use: RR 0.90 (0.85-0.95)
NVP use: RR 0.92 (0.86-0.99)
HDL: RR 0.75/mmol/L higher (0.58-0.96)
Trig: RR 1.64/doubling (1.50-1.80)
Fat gain: RR 1.36 (1.09-1.68)
Tien,20 U.S. (WIHS)1524, 2000-2006, M 39.20/100, 56% black, 28% Hispanic, 16% white, M 26.8, 84% on ARTNRTIs >3 y: aRH 2.64 (1.11-6.32)
3TC >1 y: aRH 2.81 (1.33-5.95) HIV, PIs not associated
Galli,10 Milan4299, 2008, M 45.776/24, race not reported, 29.6% overweight or obese, 91.5% on ARTHIV: aOR 1.55 (1.02-2.31)†
Age >50 vs ≤50: aOR 3.77 (2.82-5.10)
BMI 25-29.9 vs <25: aOR 1.59 (1.12-2.29)
BMI ≥ 30 vs <25: aOR 4.03 (2.72-5.99)
HTN: aOR 1.34 (1.01-1.79)
Petoumenos,9 Europe, U.S., Australia, Argentina (DAD Study)16,632, 2000-2010, M 46.273/27, 7% nonwhite, M 23.0, 95% on ARTAge/5 y: IRR 1.16 (1.10-1.21)
Glucose ≥141/100 nonfasting/fasting: IRR 12.89 (10.43-15.92)
Trig 150-200: IRR 1.87 (1.34-2.60)
Trig 200-500: IRR 2.91 (2.23-3.79)
Trig ≥ 500: IRR 5.91 (4.23-8.27)
BMI: IRR 1.10/kg/m2 (1.08-1.18)
Lipodystrophy: IRR 1.27 (1.02-1.56)
CD4 200-349 vs <200: IRR 0.52 (0.36-0.77)
CD4 ≥ 350 vs < 200: IRR 1.27 (1.02-1.56)
BP ≥ 130/85: IRR 1.37 (1.09-1.72)
Ledergerber,26 Switzerland (Swiss HIV Cohort Study)6513, 2000-2006, M3869/31, 15.5% nonwhite, M 22.5, 73% on ARTMale: IRR 2.54 (1.53-4.21)
40-49 vs < 40: IRR 1.93 (1.22-3.05)
50-59 vs < 40: IRR 2.29 (1.30-4.09)
≥ 60 vs < 40: IRR 4.32 (2.28-8.16)
Black vs white: IRR 2.10 (1.11-4.00)
Asian vs white: IRR 4.88 (2.17-10.9)
CDC stage C vs A/B: IRR 1.56 (1.04-2.35)
Central obesity: IRR 4.69 (3.14-7.00)
HCV, HTN not associated
Tripathi,21 South Carolina (HIV+/HIV-matched-cohort comparison)6816, 1994-2011, M 3957/43, 79% nonwhite, 10% documented obese, 80% on ARTART-treated vs HIV-neg: aHR 0.55 (0.46-0.65)
ART-naive vs HIV-neg: aHR 0.82 (0.63-1.07, NS)
Cumulative PI: aHR 1.35 (1.03-1.78)
Female: aHR 1.32 (1.06-1.65)
Age: aHR 1.09/year (1.04-1.15)
HTN: aHR 2.01 (1.59-2.55)
Dyslipidemia: aHR 1.71 (1.29-2.26)
Obesity: aHR 1.57 (1.15-2.13)
HCV not associated (aHR 1.41, 0.98-2.03)
Butt,23 U.S. (Veterans Aging Cohort Study)3227, 2002-, A 49.697.5/2.5, 67% black, 10% Hispanic, 20% white, A 25.2, proportion on ART not reportedHIV: aOR 0.84 (0.72-0.97)†

In HIV+:
ART use: aOR 1.11 (1.05-1.17)
NRTI/y: aOR 1.06 (1.02-1.10)
NNRTI/y: aOR 1.09 (1.02-1.10)
Older age (increasing with stratum)
Black vs white: 1.65 (1.22-2.22)
Hispanic vs white: 1.55 (1.01-2.37)
Higher BMI (increasing with stratum)
HCV: aOR 1.36 (1.06-1.73)

* Only independently associated variables listed.

† Adjusted odds ratios for prevalent diabetes, not new-onset diabetes.

A, average (mean); aHR, adjusted hazard ratio; ALT, alanine aminotransferase; aOR, adjusted odds ratio; aRH, adjusted relative hazard; aRR, adjusted relative risk; BMI, body mass index; BP, blood pressure; d4T, stavudine; ddI, didanosine; HCV, hepatitis C virus; HDL, high-density lipoprotein cholesterol; HTN, hypertension; IRR, incidence rate ratio; M, median; M/F, proportion male/female; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; NS, not significant; NVP, nevirapine; PI protease inhibitor; RR, relative rate; RTV, ritonavir; SQV, saquinavir; 3TC, lamivudine; Trig, triglycerides; WIHS, Women's Interagency HIV Study; ZVD, zidovudine.

The Milan investigators took a snapshot of diabetes prevalence in their study populations in 2008.10 A Danish study of similar size had the advantage of tracking diabetes incidence through more than a decade, from 1996 through 2010, in people with HIV and the general population.19 Denmark provides a unique platform for HIV research because everyone diagnosed with HIV in this country of 5.6 million people receives care at one of eight government HIV centers and gets tracked in a national database. The country also keeps detailed health records of the entire population that facilitate comparisons of HIV-positive and negative people.

For the diabetes study, researchers identified 4984 Danish-born people with HIV and matched them by age and gender to 19,936 people in the general population.19 Median age in the two groups stood at 38.7, 84% were men, and 97.5% were white. Through 8 years of follow-up in the HIV group and 11.2 years in the comparison group, incidence of type 2 diabetes measured 2.7% in people with HIV and 3.6% in the control group.

From 1996 through 1998, the early years of combination ART, diabetes incidence proved significantly higher in people with HIV both before they started ART (adjusted incidence rate ratio [aIRR] 2.40, 95% CI 1.03 to 5.62) and after they started therapy (aIRR 3.24, 95% CI 1.42 to 7.39). From 1999 through 2010, diabetes incidence was 55% lower in the HIV group before ART began (aIRR 0.45, 95% CI 0.21 to 0.96) and equivalent in the two groups after HIV-positive people started ART (aIRR 1.00, 95% CI 0.79 to 1.28).

Thus the Danish team discerned a marked shift in diabetes risk among HIV-positive people relative to the general population starting in 1999.19 What happened? One factor must be dwindling prescriptions of antiretrovirals that bollix glucose metabolism. This study also linked heightened diabetes risk to stavudine, didanosine, indinavir, and saquinavir. But that can't be the whole story because diabetes risk was higher in HIV-positive people before 1999 even if they had not started antiretrovirals. The Danish investigators noted that a large DAD analysis of HIV-positive people charted waning diabetes risk as combination ART got safer (1999-2006).24 (The DAD study also implicated stavudine and didanosine -- as well as zidovudine -- in diabetes risk.)

The Danish researchers suggested that in these earlier years people with HIV were generally sicker than in later years and so perhaps more prone to metabolic upheavals.19 At the same time they got treated with toxic nucleoside monotherapy and with diabetogenic drugs like pentamidine. They could have also mentioned corticosteroids, thiazides, statins, certain antipsychotics, niacin, and megestrol acetate -- and among illicit drugs, opiates. The lower diabetes risk with than without HIV starting in 1999, the authors noted, could partly reflect lower obesity prevalence documented in the HIV-positive Danish population than in the general population, a finding mirrored in the United States.8

HIV did not emerge as a new-onset diabetes risk factor in a comparison of 222 HIV-positive and 155 HIV-negative but at-risk people in the Bronx, New York, with diabetes rates of 5% in the HIV group and 8% in the comparison group.22 New-onset diabetes or prediabetes combined proved significantly less likely with than without HIV (15% versus 26%, P = 0.038).

A 2006 WIHS analysis of 1524 women with HIV and 564 at risk for HIV infection found a new-onset diabetes rate of 1.96 per 100 person-years in HIV-negative women, slightly but nonsignificantly more than the 1.53 per 100 person-years in HIV-positive women without recent antiretroviral therapy and slightly but nonsignificantly less than in women taking a PI regimen (2.50 per 100) or women taking a non-PI regimen (2.89 per 100).20

A comparison of 6816 HIV-positive adults in South Carolina matched by age, race, gender, and total months of enrollment to 6816 HIV-negative Medicaid recipients also discovered differing diabetes risks with and without HIV depending on whether the HIV group was taking antiretroviral therapy.21 This analysis of people in care at some point from January 1994 through December 2011 found that antiretroviral-naive people with HIV did not differ from the HIV-negative group in diabetes incidence (aHR 0.82, 95% CI 0.63 to 1.07). But people taking combination antiretroviral therapy had a 45% lower risk of new-onset diabetes than did HIV-negative controls (aHR 0.55, 95% CI 0.46 to 0.65).

Finally, analysis of diabetes prevalence in 3227 HIV-positive and 3240 HIV-negative members of the Veterans Aging Cohort Study (VACS) in care after 2002 found a significantly lower diabetes rate in the HIV group (14.9% versus 21.4%, P < 0.0001).23 Logistic regression analysis adjusted for pertinent risk factors determined that veterans with HIV had 16% lower odds of prevalent diabetes (aOR 0.84, 95% CI 0.72 to 0.97). Most of this difference reflected lower BMI in the HIV group (43.5% with HIV overweight or obese versus 48% without HIV). Within the HIV group, however, overweight doubled the odds of prevalent diabetes (aOR 2.02, 95% CI 1.30 to 3.13) and obesity quadrupled the odds (aOR 4.10, 95% CI 2.57 to 6.53).

Together these findings suggest that in most populations studied, people with HIV run an equivalent or lower risk of diabetes than comparable cohorts without HIV -- perhaps because HIV groups (especially those dominated by men) tend to be leaner than their general-population counterparts, and perhaps because more routine care and monitoring of people with HIV catches rising glucose early and lets clinicians reverse or at least stall the surge. As awareness of glucose abnormalities in people with HIV grew, keener monitoring probably grew apace. Among people with HIV, though, greater weight certainly boosts diabetes risk, as discussed above9,10 and confirmed in other studies.19,21,23,25 At the same time, Todd Brown argues in this interview, "HIV patients have several diabetes risk factors that HIV-uninfected patients don't have," and people with multiple classical risk factors "may be at higher risk for diabetes." Brown also notes that HIV-positive people with an undetectable viral load have residual inflammation, which can contribute to diabetes risk.

Ten Nobel Prizes for Research on Diabetes
Nobel Prize

In an article on diabetes history, the University of Chicago's Kenneth Polonsky notes that 10 Nobel Prizes have gone to scientists working on some aspect of diabetes:

  • 1923: Frederick G. Banting, John J.R. Macleod: Discovery of insulin
  • 1947: Carl F. Cori, Gerty T. Cori: Course of catalytic conversion of glycogen
  • 1947: Bernardo A. Houssay: Role of hormones released by anterior pituitary lobe in sugar metabolism
  • 1958: Frederick Sanger: Structure of proteins, especially insulin
  • 1971: Earl W. Sutherland: Mechanisms of action of hormones
  • 1977: Rosalyn Yalow: Radioimmunoassays for peptide hormones
  • 1992: Edmond H. Fischer, Edwin G. Krebs: Reversible protein phosphorylation as biologic regulatory mechanism

Source: Polonsky KS. The past 200 years in diabetes. N Engl J Med. 2012;367:1332-1340.

Risk With Antiretrovirals and Other Drugs

The 2014 edition of HRSA guidelines for HIV care do not count HIV infection as a diabetes risk factor.1 They do stress that rarely used antiretrovirals -- indinavir and stavudine -- induced insulin resistance in short-term studies of healthy volunteers and that other antiretrovirals "also perturb glucose metabolism."1 The 1994-2011 population-based comparison of people with and without HIV in South Carolina found a similar new-onset diabetes risk in antiretroviral-naive people and the HIV-negative group, a lower diabetes risk in antiretroviral-treated people with HIV than in the HIV-negative group, but a one third higher risk with cumulative PI use than in people without HIV (aHR 1.35, 95% CI 1.03 to 1.78).21

The latest (2013) Infectious Diseases Society of America HIV guidelines say "previously reported adverse effects" of antiretroviral therapy -- including diabetes -- "are much less frequent with the use of the newer agents,"27 an opinion endorsed by Todd Brown in the interview. Indeed, a 4-week placebo-controlled trial of atazanavir or lopinavir/ritonavir in healthy HIV-negative adults found that neither affected insulin sensitivity.28 And a 48-week comparison of first-line atazanavir/ritonavir and darunavir/ritonavir found no clinically relevant differences between these PIs in fasting glucose or insulin sensitivity.29 But lopinavir/ritonavir prescribing information warns that people taking these PIs may experience new-onset or worsening diabetes mellitus. And in a recent review article, Brown and two colleagues suggest the only antiretrovirals clinicians may want to switch to manage glucose upsets are lopinavir/ritonavir, stavudine, and zidovudine.30 In light of findings on currently preferred PIs, it may be time for HSRA HIV guidelines to reconsider the blanket caveat that PIs raise diabetes risk.1

Two studies in people with HIV have linked efavirenz to abnormal glucose,31,32 although the impact was small (4 mg/dL).31 In the Veterans Aging Cohort Study, every year of nonnucleoside therapy hoisted diabetes risk 9%.23 But a 33,389-person DAD Study analysis linked nevirapine use to lower risk of new diabetes.24 Prescribing information for rilpivirine, the newest nonnucleoside, does not mention insulin resistance, glucose, or diabetes.

As for current nucleosides/nucleotides, tenofovir did not affect insulin sensitivity in HIV-negative volunteers.33 Prescribing information for coformulated tenofovir/emtricitabine (Truvada) includes no warnings on insulin resistance, glucose abnormalities, or diabetes mellitus. Elevated blood glucose is listed as a lab abnormality seen in people taking abacavir in clinical trials. But neither insulin resistance nor diabetes appears in prescribing information as an abacavir or abacavir/lamivudine (Epzicom) side effect. Alone among diabetes risk studies, a 2000-2006 WIHS analysis linked more than 1 year of lamivudine therapy to an almost tripled risk of diabetes, while finding no such association for stavudine, zidovudine, or abacavir.20 Since this study also tied cumulative nucleoside use to higher diabetes risk, longer lamivudine use could be a surrogate for longer nucleoside use.

Prescribing information for three licensed integrase inhibitors -- dolutegravir, elvitegravir, and raltegravir -- does not list insulin resistance, glucose abnormalities, or diabetes mellitus as side effects.

Research links several non-HIV drugs to glucose abnormalities, including corticosteroids, thiazides, statins, atypical antipsychotics, niacin, and illicit opiates. Because of their wide use in people with HIV, statins deserve special attention, but research so far has yielded mixed results on how these anti-lipid agents affect diabetes risk. HIV/diabetes mavens Anne Monroe, Marshall Glesby, and Todd Brown note that statins can increase insulin resistance.30 But "given the cardiovascular event reduction benefit from statins," they suggest, "increases in insulin resistance/diabetes mellitus likely do not outweigh the benefit of statin therapy in the general population or in HIV-infected patients."30

Opiate use boosted diabetes incidence in a study of 1713 HIV-positive women and 652 HIV-negative women in the Women's Interagency HIV Study.34 In an analysis adjusted for HCV infection, HIV status, antiretroviral status, and classic diabetes risk factors, current opiate use raised the risk of new-onset diabetes about 60% (aRH 1.61, 95% CI 1.02 to 2.52).

HCV: Special Caution Warranted

Among classic diabetes risk factors tagged in HSRA HIV guidelines (Figure 1),1 HCV infection merits special attention in HIV populations because of its high prevalence among HIV-positive people. The CDC figures that one quarter of HIV-positive U.S. residents has HCV infection, and 80% of HIV-positive people who inject drugs have HCV as well.35

HCV infection could add to diabetes risk by the same mechanism ascribed to HIV infection -- ongoing inflammation. Unbridled inflammation spurred by untreated HCV infection could heighten diabetes risk even more. At the same time, HCV-induced liver steatosis and fibrosis -- and consequent insulin resistance -- could drive diabetes risk in HIV-positive people, NIH experts Hadigan and Kattakuzhy suggest.2 A study of 432 people with HIV but without HCV, Hadigan and Kattakuzhy observe, linked fibrosis (estimated by aspartate aminotransferase to platelet ratio index, or APRI) to prevalent diabetes.36

A 2008 meta-analysis of 34 studies determined that HCV infection boosts chances of diabetes compared with uninfected people in both retrospective studies (aOR 1.68, 95% CI 1.15 to 2.20) and prospective studies (aHR 1.67, 95% CI 1.28 to 2.06) (Figure 2).37 People with HCV also had independently higher odds of diabetes than people with HBV (aOR 1.80, 95% CI 1.20 to 1.40). And in an unadjusted analysis, HCV/HIV coinfection nearly doubled the odds of diabetes compared with HIV monoinfection (OR 1.82, 95% CI 1.27 to 2.38).

HCV and Diabetes Risk in Meta-Analysis
HCV and Diabetes Risk in Meta-Analysis Figure 2. A 34-study meta-analysis determined that HCV infection boosted odds of diabetes by two thirds in both retrospective and prospective studies (first two cones). HCV raised odds of diabetes 80% more than hepatitis B virus (HBV) infection (third cone). In an unadjusted analysis, coinfection with HCV and HCV raised odds of diabetes 82% compared with HIV alone. (Source: White DL et al.37)

Not all cohort studies tie HCV to diabetes in people with HIV. For example, the Swiss HIV Cohort Study of 6513 HIV-positive people seen in 2000-2006 linked age, race, CDC stage, and central obesity -- but not HCV -- to new-onset diabetes.26 A New York City case-control comparison of 49 HIV-positive people with diabetes and 49 without diabetes tied higher alanine aminotransferase to new-onset diabetes (aOR 1.16 per 10 units higher, 95% CI 1.03 to 1.30), but not HCV infection.25 A study of 6816 South Carolina residents with HIV found a strong trend toward higher diabetes risk with HCV (aHR 1.41, 95% CI 0.98 to 2.03).21

An array of other studies have linked HCV to diabetes in people with HIV, including 1389 people naive to antiretroviral therapy,38 1230 people taking their first antiretroviral regimen,39 3227 HIV-positive U.S. veterans,23 and young, lean HIV-positive people without a family history of diabetes.40 A 2001-2008 study of 78 HIV-positive pregnant women taking a PI or nevirapine regimen tied HCV coinfection to a quadrupled risk of glucose abnormalities defined by National Diabetes Data Group criteria (aOR 4.16, 95% CI 1.22 to 14.1, P = 0.022).41 (HSRA HIV care guidelines list current pregnancy, gestational diabetes, delivery of an infant weighting more than 4.1 kg (9 lb), and polycystic ovary syndrome as diabetes risk factors.1)

Results of studies like these, and the plausible mechanisms outlined above, should heighten vigilance for insulin and glucose abnormalities in HIV-positive people coinfected with HCV.

An Age-Old Addendum

HIV populations around the world are aging, with a growing sample of studies showing survival of antiretroviral responders who do not inject drugs approaching or reaching rates in the general population. Aging makes diabetes more likely in the general population, and it does in people with HIV. Six of the nine studies in Table 1 identified older age as an independent diabetes risk factor. A DAD Study analysis of 16,632 people in Europe, the United States, Australia, and Argentina determined that every 5 years of age boosted diabetes incidence 16%.9 In a South Carolina study every 1 year of age hoisted diabetes risk 9%.21 Current HRSA HIV care guidelines fail to list age as a diabetes risk factor,1 perhaps because it is so obvious. But as HIV clinicians succeed in stewarding more of their patients into older age, they would do well to keep the obvious in mind.


  1. U.S. Department of Health and Human Services Health Resources and Services Administration. Guide for HIV/AIDS clinical care. April 2014.
  2. Hadigan C, Kattakuzhy S. Diabetes mellitus type 2 and abnormal glucose metabolism in the setting of human immunodeficiency virus. Endocrinol Metab Clin N Am. 2014;43:685-696.
  3. Worm SW, De Wit S, Weber R, et al. Diabetes mellitus, preexisting coronary heart disease, and the risk of subsequent coronary heart disease events in patients infected with human immunodeficiency virus: the Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D Study). Circulation. 2009;119:805-811.
  4. Valcour VG, Shikuma CM, Shiramizu BT, et al. Diabetes, insulin resistance, and dementia among HIV-1-infected patients. J Acquir Immune Defic Syndr. 2005;38:31-36.
  5. Valcour VG, Sacktor NC, Paul RH, et al. Insulin resistance is associated with cognition among HIV-1-infected patients: the Hawaii Aging With HIV cohort. J Acquir Immune Defic Syndr. 2006;43:405-410.
  6. McCutchan JA, Marquie-Beck JA, Fitzsimons CA, et al. Role of obesity, metabolic variables, and diabetes in HIV-associated neurocognitive disorder. Neurology. 2012;78:485-492.
  7. Dray-Spira R, Legeai C, Le Den M, et al. Burden of HIV disease and comorbidities on the chances of maintaining employment in the era of sustained combined antiretroviral therapies use. AIDS. 2012;26:207-215.
  8. Thompson-Paul A, Wei S, Mattson C, Skarbinski J. Prevalence of obesity in a nationally representative sample of HIV+ adults receiving medical care in the U.S.: Medical Monitoring Project, 2009. 20th Conference on Retroviruses and Opportunistic Infections. March 3-6, 2013. Atlanta. Abstract 777.
  9. Petoumenos K, Worm SW, Fontas E, et al. Predicting the short-term risk of diabetes in HIV-positive patients: the Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) study. J Int AIDS Soc. 2012;15:17426.
  10. Galli L, Salpietro S, Pellicciotta G, et al. Risk of type 2 diabetes among HIV-infected and healthy subjects in Italy. Eur J Epidemiol. 2012;27:657-665.
  11. Grunfeld C, Rimland D, Gibert CL, et al. Association of upper trunk and visceral adipose tissue volume with insulin resistance in control and HIV-infected subjects in the FRAM study. J Acquir Immune Defic Syndr. 2007;46:283-290.
  12. Freitas P, Carvalho D, Santos AC, et al. Lipodystrophy defined by fat mass ratio in HIV-infected patients is associated with a high prevalence of glucose disturbances and insulin resistance. BMC Infect Dis. 2012;12:180.
  13. Hessol NA, Ameli N, Cohen MH, et al. The association between diet and physical activity on insulin resistance in the Women's Interagency HIV Study. J Acquir Immune Defic Syndr. 2013;62:74-80.
  14. Tsiodras S, Poulia KA, Yannakoulia M, et al. Adherence to Mediterranean diet is favorably associated with metabolic parameters in HIV-positive patients with the highly active antiretroviral therapy-induced metabolic syndrome and lipodystrophy. Metabolism. 2009;58:854-859.
  15. Samaras K, Wand H, Law M, Emery S, Cooper DA, Carr A. Dietary intake in HIV-infected men with lipodystrophy: relationships with body composition, visceral fat, lipid, glucose and adipokine metabolism. Curr HIV Res. 2009;7:454-461.
  16. Brown TT, Tassiopoulos K, Bosch RJ, Shikuma C, McComsey GA. Association between systemic inflammation and incident diabetes in HIV-infected patients after initiation of antiretroviral therapy. Diabetes Care. 2010;33:2244-2249.
  17. Béténé A, Dooko C, De Wit S, et al. Interleukin-6, high sensitivity C-reactive protein, and the development of type 2 diabetes among HIV-positive patients taking antiretroviral therapy. J Acquir Immune Defic Syndr. 2014;67:538-546.
  18. Boufassa F, Goujard C, Viard JP, et al. Immune deficiency could be an early risk factor for altered insulin sensitivity in antiretroviral-naive HIV-1-infected patients: the ANRS COPANA cohort. Antivir Ther. 2012;17:91-100.
  19. Rasmussen LD, Mathiesen ER, Kronborg G, Pedersen C, Gerstoft J, Obel N. Risk of diabetes mellitus in persons with and without HIV: a Danish nationwide population-based cohort study. PLoS One. 2012;7:e44575.
  20. Tien PC, Schneider MF, Cole SR, et al. Antiretroviral therapy exposure and incidence of diabetes mellitus in the Women's Interagency HIV Study. AIDS. 2007;21:1739-1745.
  21. Tripathi A, Liese AD, Jerrell JM, et al. Incidence of diabetes mellitus in a population-based cohort of HIV-infected and non-HIV-infected persons: the impact of clinical and therapeutic factors over time. Diabet Med. 2014;31:1185-1193.
  22. Polsky S, Floris-Moore M, Schoenbaum EE, Klein RS, Arnsten JH, Howard AA. Incident hyperglycaemia among older adults with or at-risk for HIV infection. Antivir Ther. 2011;16:181-188.
  23. Butt AA, McGinnis K, Rodriguez-Barradas MC, et al. HIV infection and the risk of diabetes mellitus. AIDS. 2009;23:1227-1234.
  24. De Wit S, Sabin CA, Weber R, et al. Incidence and risk factors for new-onset diabetes in HIV-infected patients: the Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) study. Diabetes Care. 2008;31:1224-1229.
  25. Yoon C, Gulick RM, Hoover DR, Vaamonde CM, Glesby MJ. Case-control study of diabetes mellitus in HIV-infected patients. J Acquir Immune Defic Syndr. 2004;37:1464-1469.
  26. Ledergerber B, Furrer H, Rickenbach M, et al. Factors associated with the incidence of type 2 diabetes mellitus in HIV-infected participants in the Swiss HIV Cohort Study. Clin Infect Dis. 2007;45:111-119.
  27. Aberg JA, Gallant JE, Ghanem KG, et al. Primary care guidelines for the management of persons infected with HIV: 2013 update by the HIV Medicine Association of the Infectious Diseases Society of America. Clin Infect Dis. 2014;58:1-10.
  28. Dubé MP, Shen C, Greenwald M, Mather KJ. No impairment of endothelial function or insulin sensitivity with 4 weeks of the HIV protease inhibitors atazanavir or lopinavir-ritonavir in healthy subjects without HIV infection: a placebo-controlled trial. Clin Infect Dis. 2008;47:567-574.
  29. Aberg JA, Tebas P, Overton ET, et al. Metabolic effects of darunavir/ritonavir versus atazanavir/ritonavir in treatment-naive, HIV type 1-infected subjects over 48 weeks. AIDS Res Hum Retroviruses. 2012;28:1184-1195.
  30. Monroe AK, Glesby MJ, Brown TT. Diagnosing and managing diabetes in HIV-infected patients: current concepts. Clin Infect Dis. 2015;60:453-462.
  31. Erlandson KM, Kitch D, Tierney C, et al. Impact of randomized antiretroviral therapy initiation on glucose metabolism. AIDS. 2014;28:1451-1461.
  32. Dave JA, Lambert EV, Badri M, et al. Effect of nonnucleoside reverse transcriptase inhibitor-based antiretroviral therapy on dysglycemia and insulin sensitivity in South African HIV-infected patients. J Acquir Immune Defic Syndr. 2011;57:284-289.
  33. Randell PA, Jackson AG, Zhong L, Yale K, Moyle GJ. The effect of tenofovir disoproxil fumarate on whole-body insulin sensitivity, lipids and adipokines in healthy volunteers. Antivir Ther. 2010;15:227-233.
  34. Howard AA, Hoover DR, Anastos K, et al. The effects of opiate use and hepatitis C virus infection on risk of diabetes mellitus in the Women's Interagency HIV Study. J Acquir Immune Defic Syndr. 2010;54:152-159.
  35. Centers for Disease Control and Prevention. HIV and viral hepatitis.
  36. DallaPiazza M, Amorosa VK, Localio R, et al. Prevalence and risk factors for significant liver fibrosis among HIV-monoinfected patients. BMC Infect Dis. 2010;10:116.
  37. White DL, Ratziu V, El-Serag HB. Hepatitis C infection and risk of diabetes: a systematic review and meta-analysis. J Hepatol. 2008;49:831-844.
  38. Visnegarwala F, Chen L, Raghavan S, et al. Prevalence of diabetes mellitus and dyslipidemia among antiretroviral naive patients co-infected with hepatitis C virus (HCV) and HIV-1 compared to patients without coinfection. J Infect. 2005;50:331-337.
  39. Mehta SH, Moore RD, Thomas DL, et al. The effect of HAART and HCV infection on the development of hyperglycemia among HIV-infected persons. J Acquir Immune Defic Syndr. 2003;33:577-584.
  40. Jain MK, Aragaki C, Fischbach L, et al. Hepatitis C is associated with type 2 diabetes mellitus in HIV-infected persons without traditional risk factors. HIV Med. 2007;8:491-497.
  41. Pinnetti C, Floridia M, Cingolani A, et al. Effect of HCV infection on glucose metabolism in pregnant women with HIV receiving HAART. HIV Clin Trials. 2009;10:403-412.