Abstract
3′-Azido-3′-deoxythymidine (AZT; zidovudine) either alone or in combination with didanosine or another nucleoside analogue is the first-line treatment for patients with human immunodeficiency virus (HIV) infection, many of whom have concurrent gastrointestinal (GI) disease. This study assessed whether the absorption of AZT was affected by GI disease. The absorption and pharmacokinetics of AZT in 23 HIV-infected individuals was measured after a single dose of AZT and was related in 12 patients to small intestinal function. Levels of AZT and its metabolite 5′-glucopyranuronosylthymidine (G-AZT) were measured by radioimmunoassay. Intestinal permeability was assessed by differential urinary excretion of orally administered lactulose/1-rhamnose; absorptive capacity was measured simultaneously by the urinary excretion of 3-o-methyl-D-glucose, d-xylose and 1-rhamnose. Small intestinal inflammation was assessed by faecal excretion of indium-labelled neutrophils. Peak levels of AZT in serum varied between 170 and 1820 ng mL−1. The metabolite G-AZT was present in serum at peak concentrations varying from 1020 to 9930 ng mL−1. There was up to a sevenfold variation in the area under the curve (AUC). The time to maximum serum concentration for AZT was between 30 and 120 min, with an absorption half-life of between 2 and 38 min. The median elimination half-life was 57 min (range 46–72 min), close to the predicted half-life of 60 min. Intestinal permeability was increased in six of the 12 subjects studied and eight had evidence of reduced absorptive capacity. Ten of the 12 patients had evidence of small intestinal inflammation. We concluded that neither changes in permeability nor absorptive capacity influenced the absorption of AZT. There was no relationship between the presence of intestinal inflammation and AZT absorption. This study showed a significant intra-individual variation of serum AZT levels which cannot be explained on the basis of altered intestinal absorptive capacity or intestinal inflammatory changes.