Protease inhibitor (PI)-containing antiretroviral regimens are potent suppressors of HIV replication. Increasingly, metabolic abnormalities such as hypercholesterolemia and triglyceridemia are associated with PI use, reasons cited for switching to PI-sparing regimens. Yet optimal switch regimens that take into account both improvements in side effects and continued virologic suppression have not been defined. This study will compare the effect on chemical metabolic abnormalities of switching to an all nucleoside regimen versus dual nucleoside plus nonnucleoside reverse transcriptase inhibitor (NNRTI) therapy. Determining the effects of each regimen on chemical metabolic abnormalities and maintenance of virologic suppression will define which of the switch strategies being studied improves metabolic abnormalities without compromising viral suppression.

Patients are stratified on the basis of fasting non-HDL cholesterol and triglyceride levels and on ritonavir- or nonritonavir-containing pre-entry PI regimens. Patients are assigned randomly to add to their pre-entry regimen 1 of the following 3 treatments: Arm A - ABC; Arm B - NVP; or Arm C - EFV.

Patients discontinue pre-entry PIs after Day 14. Patients are followed to determine the effect of the maintenance regimens on fasting non-high-density lipoprotein (HDL), cholesterol, and triglycerides at Week 24. Fasting total cholesterol, HDL cholesterol, direct low-density lipoprotein, and triglycerides are measured at Weeks 12, 24, and 48. Fasting glucose, insulin, C-peptide, apolipoproteins A-1 and B, lipoprotein a, and homocysteine are measured at Weeks 24 and 48. Anthropometrics, body mass index, and body image are measured at Weeks 12, 24, and 48. HIV viral load is measured at Weeks 2, 4, 8, 12, 16, 20, 24, 32, 40, and 48. If HIV RNA stays below 200 copies/ml, therapy continues unchanged. If confirmed HIV RNA of 200 copies/ml or higher is found, an HIV genotype is obtained providing the viral load is sufficient to yield results, the best medical therapy is instituted (not supplied by the study), and off treatment/on study follow-up is continued. If patients are intolerant to a study drug, an alternate study drug (ABC, EFV, or NVP supplied by the study) is permitted and switched treatment/on study follow-up continued, or the best medical therapy is instituted (not supplied by the study), and off treatment/on study follow-up is continued. Patients are followed until the last patient enrolled has completed 48 weeks on study.