Rate of viral rebound according to specific drugs in the regimen in 2120 patients with HIV suppression

Background: It is currently unclear whether the tendency for viral rebound in patients with viral load <50 copies/ml differs according to the specific drug regimen being used. Methods: To follow 2120 patients in EuroSIDA who had attained <50 copies/ml on highly active antiretroviral therapy (HAART), without previously virologically failing HAART. Results: The rate of viral rebound (two consecutive values > 400 copies/ml) was 4.9/100 person-years [95% confidence interval (CI), 4.0 - 5.8] for patients who were naive pre-HAART and 8.0/100 person-years (95% CI, 7.0 - 9.0) for those who were experienced with nucleoside analogue reverse transcriptase inhibitors (NRTI) pre-HAART. The rate of rebound was significantly higher in those taking nelfinavir than in those taking efavirenz, both in patients who were naive pre-HAART and those who were NRTI experienced [adjusted rate ratios, 2.83 (95% CI, 1.51 - 5.31) and 2.86 (95%, CI, 1.65 - 5.00), respectively]. Among patients who were naive pre-HAART, those on abacavir had no evidence of a raised risk of viral rebound (adjusted rate ratio 1.17; 95% CI, 0.51 - 2.69), but in those with pre-HAART NRTI experience the rate was markedly raised (adjusted rate ratio, 4.48; 95% CI, 2.51 - 8.00). A similar picture was seen when comparing those on nevirapine with those on efavirenz, although the elevated rate ratio in pre-HAART experienced patients was of lower magnitude (adjusted rate ratio, 1.93). There was no strong evidence that rebound rates differed significantly for any NRTI pairs compared with zidovudine/lamivudine. Conclusion: Viral rebound rates in patients who have attained <50 copies/ml appear to differ according to the specific drugs being used. (C) 2004 Lippincott Williams Wilkins