ObjectiveTo evaluate the pharmacokinetic interactions between efavirenz and rifampicin (rifampin) in patients with HIV infection and tuberculosis.DesignNonblind, randomised, pharmacokinetic study.Patients24 patients (21 male, 3 female; mean age 37 years) with HIV infection and tuberculosis.InterventionsPatients were randomised to one of the following treatments: group A (n = 16) received antituberculosis drugs without rifampicin, plus highly active antiretroviral therapy (HAART) including efavirenz 600mg once daily, on days 1 to 7. Patients were then switched to rifampicin in bodyweight-adjusted fixed-dose combination plus HAART including efavirenz 600mg once daily (group A-1; n = 8) or efavirenz 800mg once daily (group A-2; n = 8). Group B (n = 8) received rifampicin in bodyweight-adjusted fixed-dose combination on days 1 to 7; on day 8, HAART including efavirenz 800mg once daily was added. Blood samples were obtained on days 7 and 14.MethodsPlasma concentrations of efavirenz and rifampicin were quantified by using validated high performance liquid chromatography assays, and pharmacokinetic parameter values were determined by noncompartmental methods. The differences between pharmacokinetic parameters on days 7 and 14 were used to assess interactions.ResultsThere was a correlation between the pharmacokinetic parameters of efavirenz and the dose/kg administered. For efavirenz, mean (median) peak concentration, trough concentration and area under the concentration-time curve over the administration interval decreased 24% (24%), 25% (18%) and 22% (10%), respectively, in the presence of rifampicin. Large interpatient variability was observed, suggesting that plasma concentration monitoring of efavirenz may be advisable. Overall, the pharmacokinetics of efavirenz 800mg plus rifampicin were similar to those of efavirenz 600mg without rifampicin. The pharmacokinetics of rifampicin did not change substantially in the presence of efavirenz. Differences in patients' bodyweight appeared to cause further differences in exposure to efavirenz. Plasma concentrations of efavirenz in patients weighing <50kg were similar to those previously described in HIV-infected patients without concomitant tuberculosis. However, plasma concentrations in patients weighing ≥50kg were almost halved compared with those in patients weighing <50kg.ConclusionsAlthough the minimal effective efavirenz plasma concentration that assures virological success is not currently known, it may be advisable to increase the dosage of efavirenz to 800mg once daily when it is coadministered with rifampicin. Rifampicin can be used with efavirenz without dosage modification.