We conducted a randomized, double-blind, placebo-controlled, crossover study at a single
center in South Africa, to ascertain whether amitriptyline is an effective analgesic for painful
HIV-associated sensory neuropathy of moderate to severe intensity in: i) antiretroviral drug
naive individuals, and ii) antiretroviral drug users. 124 HIV-infected participants (antiretrovi-
ral drug naive = 62, antiretroviral drug users = 62) who met the study criteria for painful HIV-
associated sensory neuropathy were randomized to once-daily oral amitriptyline (titrated to
a median: interquartile range of 50: 25-50 mg) or placebo for six weeks, followed by a three-
week washout period and subsequent treatment crossover. The primary outcome measure
was change from baseline in worst pain intensity of the feet (measured by participant self-
report using an 11-point numerical pain rating scale) after six weeks of treatment. 122 of
124 participants completed all study visits and were included in the analysis of the primary
outcome. In the antiretroviral drug-naive group (n = 61) there was no significant difference in
the mean change in pain score from baseline after six weeks of treatment with placebo or
amitriptyline [amitriptyline: 2.8 (SD 3.3) vs. placebo: 2.8 (3.4)]. Similarly, there was no signif-
icant difference in the change in pain score after six weeks of treatment with placebo or ami-
triptyline in the antiretroviral drug-user group (n = 61) [amitriptyline: 2.7 (3.3) vs. placebo:
2.1 (2.8)]. Controlling for period effects and treatment order effects did not alter the outcome
of the analyses. Nor did analyzing the intention-to-treat cohort (missing data interpolated
using baseline observation carried forward) alter the outcome of the analyses. In summary,
amitriptyline, at the doses used here, was no more effective than an inactive placebo at re-
ducing pain intensity in individuals with painful HIV-associated sensory neuropathy of mod-
erate to severe intensity, irrespective of whether they were on antiretroviral therapy or not.