Validation of a clinical score to determine the risk of acute renal failure after cardiac surgery

Abstract

Objective: Preoperative assessment of risk factors for postoperative acute renal failure (ARF) is a key point in the management of cardiac surgery patients. A reliable evaluation of individual risk permits the efficient application of prophylactic and therapeutic measures to patients with increased risk of ARF. In 2005, Thakar published a scoring system to determine the ARF risk prior to surgery. The score allowed calculation of the incidence of postoperative ARF requiring dialysis (ARF-D) with a high level of precision. Methods: In order to examine the reproducibility of the results on different patient populations, we retrospectively applied the scoring system to 3508 cardiac surgery patients at the Goettingen University Hospital. Results: We found a qualitative correlation between the score value and ARF-D, but prognostic significance, when judged by the area under the receiver operating characteristic (ROC) curve, was considerably reduced. Also, the incidence of ARF-D in the Goettingen population was remarkably higher than the incidence shown by Thakar for patients with the same score. Conclusions: The correlation between score values and incidence of ARF-D shows that the Thakar score is qualitatively transferable to cardiac surgery patients from other centres. Though the score allows the discrimination between patients with higher or lower risks of ARF-D within the Goettingen collective, it was not suitable to estimate the real incidence of postoperative ARF-D with sufficient precision.