ABSTRACT
BACKGROUND: Chronic kidney disease is an important preoperative risk factor. However, the association between renal dysfunction and risk of death has not been well explored in non-cardiac surgery. METHODS: Two prospective observational studies in non-cardiac surgery were analysed: the European Surgical Outcomes Study (EuSOS) and the UK National Confidential Enquiry into Patient Outcome and Death (NCEPOD). The relationship between preoperative estimated glomerular filtration rate (eGFR) and postoperative mortality was examined using multivariable Cox proportional hazards models. RESULTS: In EuSOS, 1580 (4·3 per cent) of 36 779 patients died in hospital; in NCEPOD, 298 (2·8 per cent) of 10 466 patients had died by 60 days after surgery. Chronic kidney disease (eGFR below 60·0 ml per min per 1·73 m(2) ) was present in 6415 patients (17·4 per cent) in EuSOS and 2262 (21·6 per cent) in NCEPOD. Preoperative chronic kidney disease was associated with older age, men, diagnosis of diabetes, cardiovascular or respiratory disease, and non-elective surgery. Preoperative eGFR categories below 60·0 ml per min per 1·73 m(2) were associated with increasing adjusted hazard ratios (HRs) for death compared with a value of 90·0 ml per min per 1·73 m(2) and above. In EuSOS, the risk of death increased with lower eGFR category, to a maximum with eGFR 15·0-29·9 ml per min per 1·73 m(2) (HR 3·37, 95 per cent c.i. 2·70 to 4·22). In NCEPOD, the risk of death also increased with declining eGFR and was maximal for eGFR below 15·0 ml per min per 1·73 m(2) (HR 3·40, 1·78 to 6·50). CONCLUSION: Renal dysfunction is an important risk factor for death after non-cardiac surgery and the risk increases steeply for patients with moderate to severe kidney dysfunction.
