Incidence of treatment-limiting toxicity with stavudine-based antiretroviral therapy in Cambodia: a retrospective cohort study.

BACKGROUND: Although stavudine (D4T) remains frequently used in low-income countries in Asia, associated long-term toxicity data are scarce. The aim of this study was to determine the long-term incidence of severe D4T-toxicity (requiring drug substitution) and associated risk factors in HIV-infected Cambodians up to six years on antiretroviral treatment (ART). METHODOLOGY/PRINCIPAL FINDINGS: This is a retrospective analysis of an observational cohort, using data from an ART program with systematic monitoring for D4T-toxicity. Probabilities of time to D4T substitution due to suspected D4T toxicity (treatment-limiting D4T toxicity) were calculated, a risk factor analysis was performed using multivariate Cox regression modelling. Out of 2581 adults initiating a D4T-containing regimen, D4T was replaced in 276 (10.7%) patients for neuropathy, 14 (0.5%) for lactic acidosis and 957 (37.1%) for lipoatrophy. The main early side effect was peripheral neuropathy (7.0% by 1 year). After the first year, lipoatrophy became predominant, with a cumulative incidence of 56.1% and 72.4% by 3 and 6 years respectively. Older age (aHR 1.8; 95%CI: 1.4-2.3) and lower baseline haemoglobin (aHR 1.7; 95%CI: 1.4-2.2) were associated with the occurrence of neuropathy. Being female (aHR 3.8; 95%CI: 1.1-12.5), a higher baseline BMI (aHR 12.6; 95%CI: 3.7-43.1), and TB treatment at ART initiation (aHR 8.6; 95%CI: 2.7-27.5) increased the likelihood of lactic acidosis. Lipoatrophy was positively associated with female gender (aHR 2.3; 95%CI: 2.0-2.6), an older age (aHR 1.3; 95%CI: 1.1-1.4), and a CD4 count <200 cells/µL (aHR 1.3; 95%CI: 1.1-1.5). CONCLUSIONS: Stavudine-based treatment regimens in low-income countries are associated with significant long-term toxicities, predominantly lipoatrophy. Close clinical monitoring for toxicity with timely D4T substitution is recommended. Phasing-out of stavudine should be implemented, as costs allows.

Abstract licence: CC BY

27 more references, whose wording the paper’s licence keeps private.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.