The formation of the reduced metabolite of haloperidol — reduced haloperidol (RH) — was evaluated after intramuscular haloperidol injection in 13 chronic schizophrenic patients. After an overnight fast, each patient received a 5mg dose. Plasma samples were obtained prior to injection and 0.25, 0.5, 1.0, 1.5, 2, 2.5, 3, 6, 9, 12, 24, 36, 48, 72 and 96 hours postinjection. Plasma haloperidol, RH and homovanillic acid (HVA) levels were assayed by high performance liquid chromatography with electrochemical detection. Plasma haloperidol concentrations were detectable in each patient at 96 hours postinjection. The mean time to maximum (tmax) plasma concentrations was 0.86 hours. A wide interpatient variability was observed in the pharmacokinetic parameters for the population. RH was detected in only 6 patients. The mean tmax of RH occurred at 1.37 hours. Although statistically not significant, the mean volume of distribution of haloperidol patients with RH was slightly larger compared with patients with only haloperidol (32.89 vs 28.33 L/kg). The mean elimination half-life of haloperidol was longer but also not statistically significant in the RH/haloperidol patients compared with haloperidol only patients (43.03 vs 34.37 hours). Intramuscular haloperidol injection produced a characteristic decrease in plasma HVA levels with a tmax at 0.75 hours. Plasma HVA levels remained decreased and returned to baseline values 12 hours later. Plasma HVA levels were slightly lower in RH/haloperidol patients compared with haloperidol only patients.