For a study, researchers sought to assess the development and validation of prediction models for AKI following heart surgery. Multivariable prediction models were developed based on a retrospective observational cohort of adult patients undergoing cardiac surgery at a US academic medical center between January 2000 and December 2019 (n= 58,526) and validated on an external cohort from three US community hospitals (n = 4,734). The last follow-up was scheduled for January 15, 2020. The initial metabolic panel following heart surgery revealed perioperative changes in serum creatinine and postoperative blood urea nitrogen, serum sodium, potassium, bicarbonate, and albumin. According to Kidney Disease: Improving Global Outcomes (KDIGO), the area under the receiver-operating characteristic curve (AUC) and calibration measures for moderate to severe AKI, as well as AKI requiring dialysis prediction models within 72 hours and 14 days following surgery. 

The rates of moderate to severe AKI and AKI requiring dialysis were 2,674 (4.6%) and 868 (1.48%) within 72 hours and 3,156 (5.4%) and 1,018 (1.74%) within 14 days after surgery in a derivation cohort of 58,526 patients (median [IQR] age, 66 [56-74] years; 39,173 [67%] men; 51,503 [91%] White participants). The median (IQR) time from the end of the surgical surgery to the first metabolic panel was 10 (7-12) hours. The metabolic panel-based models had excellent predictive discrimination in the derivation cohort for moderate to severe AKI within 72 hours (AUC, 0.876 [95% CI, 0.869-0.883]) and 14 days (AUC, 0.854 [95% CI, 0.850-0.861]) after the surgical procedure and for AKI requiring dialysis within 72 hours (AUC, 0.916 [95% CI, 0.907-0.926]) and 14 days (AUC,  0.900 [95% CI, 0.889-0.909]) after the surgical procedure. From the completion of the surgical operation to the first metabolic panel, the median (IQR) delay was 10 (7-12) hours. The metabolic panel-based models had excellent predictive discrimination in the derivation cohort for moderate to severe AKI within 72 hours (AUC, 0.876 [95% CI, 0.869-0.883]) and 14 days (AUC, 0.854 [95% CI, 0.850-0.861]) after the surgical procedure and for AKI requiring dialysis within 72 hours (AUC, 0.916 [95% CI, 0.907-0.926]) and 14 days (AUC, 0.900 [95% CI, 0.889-0.909]) after the surgical procedure. In the validation cohort of 4,734 patients (median [IQR] age, 67 (60-74) years; 3,361 [71%] men; 3,977 [87%] White participants), the models for moderate to severe AKI after the surgical procedure had AUCs of 0.860 (95% CI, 0.838-0.882) within 72 hours and 0.842 (95% CI, 0.820-0.865) within 14 days, and the models for AKI requiring dialysis and 14 days Calibration was examined using the Spiegelhalter z test, which yielded a P >.05, suggesting appropriate calibration for both validation and derivation models.

A prediction model based on perioperative basic metabolic panel laboratory data showed high predictive accuracy for moderate to severe acute renal damage within 72 hours and 14 days following the surgical procedure in patients having heart surgery. More study was required to evaluate whether the risk prediction tool improves clinical outcomes.

Reference:jamanetwork.com/journals/jama/article-abstract/2789659