329 E Didanosine (ddI) is a nucleoside analogue that competitively inhibits the HIV reverse transcriptase enzyme. The lability of the molecule in acid milieu reduces its bioavailability to 10% when administered orally.1 A first approach to overcome this limitation was the formulation of ddI as buffered tablets, so that the absorption of the drug was improved by gastric-acid neutralization.2 Gastrointestinal symptoms such as dyspepsia, nausea/vomiting, flatulence or diarrhea, are among the most frequent complaints in subjects receiving ddI. In addition to the inner toxicity of ddI, the citrate/phosphate/sucrose vehicle in buffered tables seems to play a significant part in these gastrointestinal toxicities.3 Since the year 2001, ddI has been available in the form of capsules containing gastro-resistant pellets that liberate the drug’s active ingredient within the intestinal lumen. The main pharmacokinetic parameters are approximately equivalent for enteric-coated and buffered preparations, but the former was expected to be more well tolerated.4 Herein, we present the results of a small retrospective study, in which we assessed the incidence of side effects in HIV-positive subjects who initiated a triple regimen containing buffered versus enteric-coated ddI. All HIV-positive patients who initiated a ddI-containing triple regimen, either with buffered tablets or enteric capsules (400 mg once daily in both instances), at one HIV/AIDS reference center in 2001 and 2002 were retrospectively analyzed. All recruited patients had not had prior ddI experience and a minimum follow-up of 12 months. Subjects with prior history of chronic gastrointestinal disease, pancreatitis, neuropathy, or alcohol abuse were not considered eligible. The main laboratory parameters were recorded through chart review. Adverse reactions were identified from clinical records. Treatment adherence was ascertained examining the pharmacy records, because all subjects had to obtain the medication monthly from the pharmacy. A total of 133 HIV-positive patients starting a ddI-containing triple regimen were retrospectively analyzed. Overall, 83 received buffered tablets (b-ddI) and 50 received entericcoated capsules (e-ddI). Overall, baseline demographics were comparable between both groups. Mean age was 38 8 years old, 80% were male, 55% had acquired HIV through sexual contacts, and 45% were former intravenous drug users. The virologic and immunologic status when ddI was started were comparable between both treatment groups: patients on b-ddI had an average CD4 count of 429 272 cells per microliter and a median viral load of 2249 HIV-RNA copies per milliliter, while these parameters