Objectives: To compare the efficacy, tolerability and safety of a ritonavir 400 mg/saquinavir hard gel fomulation 400 mg twice daily versus an indinavir 800 mg once every 8 In containing first-line protease inhibitor (PI) treatment regimen. Methods: Open, randomized, multicentre clinical trial. PI-naive patients received either ritonavir/saquinavir and one nucleoside reverse transcriptase inhibitor (NRTI) or indinavir and two NRTIs. Intention-to-treat (ITT) and on-treatment (OT) analyses were performed. Results: The baseline characteristics of the study participants were similar in both arms, 67 patients (37%) were naive to antiretroviral treatment. The proportion of patients who achieved a plasma viral load below the level of detection of 400 copies/ml at week 48 was 43% (39/90) in the ritonavir/saquinavir arm and 63% (57/90) in the indinavir arm (P=0.005, ITT analysis). Using an OT analysis, these percentages were 84% and 88%, respectively (P=0.6). There were more drop-outs in the ritonavir/saquinavir arm than in the indinavir arm (35.6% (32/90) versus 15.6% (14/90), P=0.002), mainly due to gastro-intestinal side-effects. Abnormal liver tests and increased lipids levels were more frequently reported in the ritonavir/saquinavir arm than in the indinavir arm. Conclusion: In PI-naive patients, indinavir in combination with two NRTIs was more effective and better tolerated than ritonavir/saquinavir plus one NRTI. Both treatments were very effective for patients who were able to tolerate them.