The pharmacokinetics of methadone in healthy subjects and opiate users

There is some evidence that monitoring methadone plasma concentration may be of benefit in dosage adjustment during methadone maintenance therapy for heroin (opiate) dependence. However, the kinetics of oral methadone are incompletely characterized. We attempted to describe the latter using a population approach combining intensive 57 h sampling data from healthy subjects with less intensive sparse 24 h data from opiate users.

Single oral doses of rac-methadone were given to 13 drug-naive healthy subjects (7 men and 6 women) and 17 opiate users beginning methadone maintenance therapy (13 men and 4 women). Plasma methadone concentrations were measured by h.p.l.c. Kinetic analysis was performed using the P-Pharm software.

Comparison of kinetic models incorporating mono- or biexponential disposition functions indicated that the latter best represented the data. The improvement was statistically significant for the data from healthy subjects whether the full 57 h or truncated 24 h profiles were used (P<0.031 and P<0.024, respectively), while it was of borderline significance for the more variable data from opiate users (P=0.057) or for pooled (healthy subjects and opiate users) data (P=0.066). The population mean oral clearance of methadone was 6.9+/-1.5 s.d. l h(-1) (5.3+/-1.2 s.d. l h(-1) using 0-24 h data) in the healthy subjects. The results of separate analyses of the data from opiate users and healthy subjects were in contrast with those obtained from pooled data analysis. The former indicated a significantly lower clearance for opiate users (3.2+/-0.3 s.d. l h(-1), P<0.001); 95% CI for the difference = -3 to -6 l h(-1) and no difference in the population mean values of V/F (212+/-27 s.d. l and 239+/-121 s.d. l, P=0.15), while according to the latter analysis addiction was a covariate for V/F but not for oral clearance. A slower absorption of methadone in opiate users was indicated from the analysis of both pooled and separate data. The median elimination half-life of methadone in healthy subjects was 33-46 h depending on the method used to calculate this parameter.

Estimates of the long terminal elimination half-life of methadone (33-46 h in healthy subjects and, possibly, longer in opiate users) indicated that accurate measurement of this parameter requires a duration of sampling longer than that used in this study. Our analysis also suggested that parameters describing plasma concentrations of methadone after a single oral dose in healthy subjects may not be used for predicting and adjusting dosage in opiate users receiving methadone maintenance therapy unless coupled with feedback concentration monitoring techniques (for example Bayesian forecasting).