Serum creatinine values underestimate surgical risk

Renal insufficiency is associated with increased mortality and morbidity from cardiac surgery. Serum creatinine (SCr) values are routinely used for the assessment of renal dysfunction. However, this parameter can overestimate renal function, especially in low-weight and elderly patients, who can have normal creatinine values despite impaired renal function.

1) To evaluate the prevalence and prognostic impact of different degrees of preoperative renal dysfunction (RD); 2) to assess the prevalence of normal SCr values among patients in different stages of RD.

This was a retrospective study of 1314 consecutive adult patients (836 male, mean age 66 +/- 11 years) undergoing cardiac surgery. Patients were assigned to one of the five stages of RD of the National Kidney Foundation classification according to their glomerular filtration rate (GFR), estimated by the Cockcroft-Gault equation and indexed to body surface area. The impact of each stage of RD on in-hospital mortality was assessed after adjusting for all the other EuroSCORE components in multivariate analysis.

The median logistic EuroSCORE was 3.8 (interquartile range: 1.9-7.0). In-hospital mortality was 3.4% (n = 35). The prevalence of stages 3, 4 and 5 RD was 30.2% (n = 397), 3.4% (n = 45) and 4.3% (n = 56) respectively. Increasing in-hospital mortality was observed across ascending stages of RD. After adjustment for other EuroSCORE risk factors. stage 3 or higher RD was an independent predictor of in hospital mortality--OR 2.0 (95% CI: 1.1-3.9, p = 0.03). Among patients with stage 3 or higher RD, 61% (n = 304) had SCr values < 1.50 mg/dl and 83% (n = 414) had SCr < 2.26 mg/dl (the EuroSCORE cutoff value).

Renal impairment is common among patients undergoing cardiac surgery and the presence of even mild forms of RD is associated with increased mortality. SCr values within the normal range frequently correspond to moderate or even severe RD, indicating that GFR should be calculated systematically to avoid underestimation of surgical risk.