IAS 2007: Sydney, Australia; July 22-25

Key Links:

By Topic:


The Body PRO Covers: The 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention

The Role of Nucleoside Analogs in the Development of Insulin Resistance

July 24, 2007

Given that HIV-infected patients can now survive for decades on antiretroviral therapy, comorbidities such as lipodystrophy and insulin resistance have moved to the forefront of HIV management. Yet, much remains to be discovered about these syndromes. Which medications are responsible? How much are these comorbidities due to HIV itself and how much to genetic predisposition in this aging population?

While lipodystrophy is associated with insulin resistance via multiple metabolic pathways, the exact sequence by which perturbations in glucose metabolism develop are unknown. Nucleoside reverse transcriptase inhibitors (NRTIs) are thought to contribute to insulin resistance primarily by inducing lipoatrophy, but is that the true and/or only mechanism produced by these agents?

For example, in a study conducted by Fleischman and colleagues, 16 healthy volunteers experienced a reduction in peripheral glucose disposal as measured by hyperinsulinemic euglycemic clamp within four weeks of starting stavudine (d4T, Zerit) therapy without ever experiencing changes in fat distribution.1

To further probe the potential mechanisms linking nucleoside analogs with insulin resistance, van Vonderen from the VU University Medical Center in Amsterdam designed a study to examine glucose disposal in HIV-infected patients taking zidovudine (AZT, Retrovir), a thymidine analog similar to stavudine. In this prospective, randomized, single-blind trial,2 20 treatment-naive male patients from Holland were randomized to receive lopinavir/ritonavir (LPV/r, Kaletra; 400/100 mg twice daily) + zidovudine/lamivudine (AZT/3TC, Combivir) or lopinavir/ritonavir (533/133 mg twice daily) + nevirapine (NVP, Viramune). The lopinavir/ritonavir dose was increased in the second arm to overcome pharmacokinetic interactions that develop between protease inhibitors (PIs) and nevirapine.

Insulin sensitivity was measured using the gold-standard hyperinsulinemic euglycemic clamp. This research tool measures the amount of glucose needed to compensate for an increased serum insulin level, which is achieved with a constant infusion of insulin and glucose over a two-hour time period during which blood sugar levels are frequently monitored. If low levels of glucose infusion are needed during the last 30 minutes of the test, this indicates insulin resistance. Insulin sensitivity and body composition (assessed by whole-body dual energy X-ray absorptiometry and abdominal computed tomography scans) were evaluated at baseline and after three, 12 and 24 months of antiretroviral therapy.

Eleven men were enrolled in the zidovudine/lamivudine arm and nine in the nevirapine arm. Median baseline CD4+ cell counts were the same at 200 cells/mm3, and median HIV-1 RNA levels were also comparable at 5.1 and 4.8 log10 copies/mL, respectively. The baseline age of all participants was 42 (+ 9) years, and their body mass index was normal at 23.3 (+ 3.1) kg/m2.

Throughout the 24-month study, serum lopinavir concentrations were similar when measured during the last hour of the euglycemic clamp test. Limb fat decreased in the zidovudine/lamivudine arm and was 1.26 kg lower at 24 months compared with the nevirapine arm (P < .05). Similarly, the limb:total fat ratio also decreased in the zidovudine/lamivudine arm and was significantly lower at 24 months with zidovudine/lamivudine versus nevirapine (P < .05). Changes in visceral adipose tissue were not statistically significantly different between the two arms at 24 months, but the study will continue for another year of follow-up to monitor for such differences.

Results from the peripheral glucose uptake clamp, which primarily measures the use of glucose in muscle, were significantly lower in the zidovudine/lamivudine arm versus the nevirapine arm at 24 months (19.7 versus 25.3 µmol/kg/min; P < .05) and started to decline immediately after the initiation of antiretroviral therapy, prior to any demonstrable loss of limb fat and/or gain in visceral adipose tissue. Moreover, this loss in the rate of glucose disposal occurred in the absence of reduced adiponectin, which increased in both arms during the study, although not as much in the zidovudine/lamivudine arm (at 24 months: 8.6 versus 10.8 µg/mL; P < .05).

Twelve months after the initiation of therapy, insulin resistance associated with lipolysis appeared in the zidovudine/lamivudine arm, but not the nevirapine arm, as measured by the glycerol clamp test (P < .05). Finally, at 24 months, insulin resistance related to hepatic glucose production appeared in the zidovudine/lamivudine arm, but not the nevirapine arm, as evidenced by a significant rise in endogenous glucose production by clamp testing (P < .05).

Zidovudine/lamivudine-associated insulin resistance showed a sequential onset. Peripheral glucose uptake was first reduced within three months of the start of therapy and persisted over time, with possible contributing factors including loss of limb fat and blunted adiponectin increases. This was followed by insulin resistance associated with lipolysis at 12 months and by insulin resistance related to hepatic glucose production at 24 months.

Although our understanding of insulin resistance with highly-active antiretroviral therapy (HAART) is in constant flux, NRTIs to date have been thought to contribute to insulin resistance mainly via induction of lipoatrophy. The authors of this paper put forth another potential mechanism based on their findings: that is, zidovudine-mediated mitochondrial toxicity in skeletal muscle, adipocytes and hepatocytes. This hypothesis is supported by the observation that zidovudine/lamivudine contributed to an early reduction in peripheral glucose uptake at three months prior to any demonstrable loss of limb fat and/or gain in visceral adipose tissue. Moreover, this reduction occurred in the absence of reduced adiponectin levels.

Further larger studies will be needed to better elucidate these findings, but these data certainly constitute an intriguing addition to our increasing knowledge of NRTI toxicity. Understanding the toxicities associated with specific antiretroviral agents and classes of agents will better enable practitioners to decide which agents to use when designing antiretroviral regimens, especially as new drugs become approved and treatment paradigms shift. The lead author of this study reported that this cohort is part of a larger, ongoing study of 50 patients, so expect to see more on this topic at future meetings.


  1. Fleischman A, Johnsen S, Systrom DM, et al. Effects of a nucleoside reverse transcriptase inhibitor, stavudine, on glucose disposal and mitochondrial function in muscle of healthy adults. Am J Physiol Endocrinol Metab. June 2007;292(6):E1666-E1673.

  2. van Vonderen MGA, Blümer RME, Hassink E, et al, and the MEDICLAS study group. Persistent reduction in peripheral glucose disposal starting prior to fat distribution changes, only during first line ART with AZT/3TC/LPV/r but not NVP/LPV/r, suggests contribution of AZT/3TC to insulin resistance by a body composition-independent mechanism. In: Program and abstracts of the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; July 22-25, 2007; Sydney, Australia. Abstract TUPEB077.

This article was provided by It is a part of the publication 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention.

Please note: Knowledge about HIV changes rapidly. Note the date of this summary's publication, and before treating patients or employing any therapies described in these materials, verify all information independently. If you are a patient, please consult a doctor or other medical professional before acting on any of the information presented in this summary. For a complete listing of our most recent conference coverage, click here.