We conducted a study to compare the effectiveness of patient-controlled analgesia (PCA) technique to conventional analgesic therapy (CAT) after coronary artery bypass graft (CABG). The PCA group received hydromorphone 0.1 mg.hr-1 basal infusion and bolus doses of 0.2 mg Q 5 min (maximum 1.2 mg.hr-1) while the CAT group received morphine 2.5 mg iv Q 30 min prn until extubation followed by prn meperidine 1 mg.kg-1 im Q 4 hr or acetaminophen 325 mg with codeine 30 mg po (1 or 2 tablets) when oral intake was possible. The degree of pain was assessed using a Visual Analogue Scale (VAS) starting after extubation and every 6-8 hr for the next 60 hr. Holter monitoring was initiated one hour after patient arrival in the Intensive Care Unit (ICU) and continued for 72 hr. Other measured variables were pulmonary function, sedation, side effects and total opioid requirements. Results show that the day-to-day VAS pain score decreased in the PCA group (P < 0.001) while it remained unchanged in CAT patients. The PCA patients had lower VAS pain scores at extubation (P < 0.05). During the third postoperative day, the PCA group had a lower VAS pain score, a lower incidence of severe pain defined as a score > 5 on the VAS scale, and a reduced incidence of myocardial ischaemia (P < 0.01). However, there was no difference in the duration, severity, area under the curve (AUC), or heart rate during ischaemic events. Postoperative pulmonary function was abnormal in both groups (NS) with minimal recovery by the fourth day.(ABSTRACT TRUNCATED AT 250 WORDS)