The aim of this study was to determine if substitution of daily methadone with second daily levo-alpha-acetylmethadol (LAAM) would convert non-holders on methadone into holders on LAAM, and to compare plasma concentration-time profiles of (R)-methadone with LAAM and its two metabolites. Sixteen stable methadone maintenance treatment participants (non-holders, n=8) were randomly allocated to continue methadone for 3 months or switch to LAAM for 3 months, and then crossed over to the alternative drug for 3 months. At steady state, there were two testing sessions (24h for methadone and 48h for LAAM), during which opioid withdrawal severity, respiration rate and pupil diameter were measured 10-11 times and venous blood was collected 13-15 times. Ten age- and gender-matched controls underwent one 48-h test session. Areas under the withdrawal severity score versus time curve (AUC 0 - 4 7 hours for LAAM and controls; AUC 0 - 2 4 x 2 for methadone) were similar in holders on methadone and LAAM (P=0.62), but were greater in non-holders when they were taking methadone than LAAM (P<0.001). Respiratory depression and pupillary constriction were similar for LAAM and methadone. In comparison to (R)-methadone, plasma nor- and dinor-LAAM concentrations fluctuated little over the dosing interval. LAAM converted methadone non-holders into LAAM holders. LAAM may therefore be useful in selected MMT non-holders and improve retention in opioid treatment programs.