The aim of the present study was to compare the new combination of intraarticular + subacromial injection, with intraarticular, subacromial injection and interscalenic brachial plexus block as postoperative analgesia in shoulder arthroscopy. One hundred and twenty patients scheduled for shoulder arthroscopy were enrolled and randomly assigned to one of five groups: intraarticular, subacromial, interscalenic brachial plexus block (IBPB), intraarticular + subacromial (intraarticular + subacromial) injection or a control group. All patients received standardized general anaesthesia and all the injections were given with the same dose and volume of local anaesthetic. The number of boluses (fentanyl 1 microg kg(-1) delivered by a patient-controlled analgesia pump applied at the end of the surgery and the visual analogue pain score (VAPS) at 0, 2, 4, 6, 12, 18 and 24 h after the intervention were recorded. A patient satisfaction score was also assessed at 24 h. Mean bolus consumption, compared with control group, was significantly less in all groups (P < 0.01). Intraarticular + subacromial group utilized fewer boluses compared with subacromial group and significantly lower boluses than intraarticular group (P < 0.01), but IBPB group utilized significantly fewer boluses than intraarticular + subacromial group. Patients in IBPB, intraarticular + subacromial and subacromial groups showed VAPSs that were significantly better than that of the control group at all time points (P < 0.01). The VAPS in intraarticular + subacromial group was statistically comparable with those in IBPB and subacromial groups at each time interval. IBPB and intraarticular + subacromial groups showed comparable patient satisfaction scores. These results confirm the analgesic efficacy of IBPB for shoulder surgery. Nonetheless, the combination of intraarticular and subacromial infiltration, studied for the first time, appears to be a clinically valid alternative with no clinical meaningful adverse effects.