SummaryCritical incident reporting was introduced into the intensive care unit (ICU) as part of the development of a quality assurance programme within our department. Over a 3-year period 281 critical incidents were reported. Factors relating to causation, detection and prevention of critical incidents were sought. Detection of a critical incident in over 50% of cases resulted from direct observation of the patient while monitoring systems accounted for a further 27%. No physiological changes were observed in 54% of critical incidents. The most common incidents reported concerned airway management and invasive lines, tubes and drains. Human error was a factor in 55% of incidents while violations of standard practice contributed to 28%. Critical incident reporting was effective in revealing latent errors in our 'system' and clarifying the role of human error in the generation of incidents. It has proven to be a useful technique to highlight problems previously undetected in our quality assurance programme. Improvements in quality of care following implementation of preventative strategies await further assessment.

We have compared the recovery profiles of 163 healthy Chinese children after general anaesthesia for minor surgical procedures. Patients were allocated randomly to receive one of four anaesthetic techniques: propofol infusion for induction and maintenance using a pharmacokinetic model-controlled syringe pump set initially at a target concentration of 8 micrograms ml-1 and then adjusted according to clinical requirements; propofol 2.5-3.5 mg kg-1, thiopentone 4-5 mg kg-1 or 2-3% halothane for induction of anaesthesia followed by 1-2% halothane for maintenance of anaesthesia. All patients breathed a mixture of 70% nitrous oxide in oxygen through a laryngeal mask airway and received an appropriate regional anaesthetic block. Recovery was assessed using the time to achieve full Steward score, open eyes on command, orientation and the time required to complete a simple puzzle. Recovery was slowest with the propofol infusion (mean 39.8 (SD 12.9) min when eyes opened on command). The recovery times were significantly shorter with the three other techniques (propofol bolus 21.9 (9.9) min, thiopentone 23.4 (11.3) min, halothane 20.1 (8.9) min), and the choice among these three methods had no significant influence on the recovery profile.

We studied the efficacy and safety of intravenous ondansetron 4 mg for the prevention of postoperative nausea and vomiting after minor gynaecological laparoscopic surgery in Oriental women. This double-blind randomised study compared ondansetron with placebo, given before the induction of anaesthesia. The anaesthetic technique used thiopentone, fentanyl, atracurium, nitrous oxide and isoflurane. Patients were studied for 24 h with nausea assessed using a verbal numeric scale from 0-10 and emetic episodes recorded as they occurred. Results were available for 102 patients in each group. In the first postoperative hour, fewer patients in the ondansetron Group (12%) had emetic episodes compared with the placebo group (33%, P <0.01). Nausea score over the first hour (sum of three readings at 0, 30 and 60 min) was lower in the ondansetron group (median 1.6) compared with the placebo group (3.1, P <0.05). Over the 24 h period, fewer patients in the ondansetron group had emetic episodes (25%) or nausea (43%) compared with patients in the placebo group (56%, P <0.01) and (58%, P <0.05) respectively. No adverse events were seen. Ondansetron 4 mg was more effective than placebo in preventing postoperative nausea and vomiting throughout the 24 h after minor laparoscopic surgery.

No abstract