Predicting postoperative adverse events. Clinical efficiency of four general classification systems - The project perioperative risk

Background: The Project Perioperative Risk in Gothenburg is a prospective clinical-epidemiological study designed to elucidate factors affecting the perioperative risk in unselected adult patients undergoing elective general or orthopaedic surgery. In this report on postoperative adverse events of varying severity, the predictive ability of four simple classification systems, ASA physical status, patient age, surgical stress and a visual analogue scale for intuitively appreciated global risk (RISK-VAS) is described. Methods: 1361 patients undergoing 1471 surgical procedures were subjected to a detailed and standardised preoperative assessment and classified according to ASA, age, procedure magnitude and RISK-VAS. Postoperative adverse events were prospectively registered during the entire hospital stay. Results: The four classifications all correlated to postoperative adverse events. ASA physical status, RISK-VAS and patient age all appear to be more efficient in predicting severe than mild events, while surgical stress predicted severe events and mild ones equally well. Conclusion: The most efficient predictor of severe events appeared to be RISK-VAS, where a relative risk of 28.1 of acquiring a severe postoperative adverse event could be demonstrated for those who had a score of 4 or more compared with those who had scores less than 4.