Mechanisms and Predictors of Acute Kidney Injury with Perioperative Rosuvastatin in Patients Undergoing Cardiac Surgery

Background In patients undergoing cardiac surgery perioperative statin therapy has been associated with an unexpected increase in postoperative plasma creatinine. Here we investigated mechanisms and predictors of acute kidney injury (AKI) in 1922 patients enrolled in the Statin Therapy in Cardiac Surgery (STICS) randomized placebo-controlled trial of perioperative rosuvastatin (20 mg once daily). Methods AKI was defined according to international guidelines (KDIGO) using plasma creatinine, and also by cystatin C. Potentially mechanistically relevant plasma/serum biomarkers of muscle injury, inflammation, and kidney injury were investigated, including total creatine kinase (CK), growth differentiation factor 15 (GDF-15), interleukin-6 (IL-6), procalcitonin (PCT), placental growth factor (PLGF), kidney injury molecule-1 (KIM-1), and neutrophil gelatinase-associated lipocalin (NGAL). Results At 48 hours post-surgery, the incidence of AKI was greater in the rosuvastatin group than in the placebo group when defined by a rise in creatinine (24.7% vs 19.3%, p=0.005) or cystatin C (9.2% vs 5.1%, p<0.001); the majority of AKI was stage 1 in severity (87% when defined by creatinine, and 80% when defined by cystatin C). Compared with placebo, rosuvastatin led to higher postoperative serum levels of KIM-1 (278±5 pg/ml versus 259±5 pg/ml, P=0.01), and to more frequent elevations in CK to >10x and >40x the baseline level (30.9% versus 26.5%, p=0.032, and 2.1% versus 0.7%, p=0.016, respectively), whereas postoperative concentrations of GDF-15, IL-6, PCT, PLGF, and NGAL were similar between groups. In multivariable analyses, insulin treatment, baseline KIM-1, combined coronary artery bypass grafting (CABG) and aortic valve replacement (AVR) surgery, and allocation to rosuvastatin were all independently associated with AKI as defined by creatinine or cystatin C. Odds ratios for rosuvastatin compared to placebo for both creatinine- and cystatin C-defined AKI were not materially altered by further adjustment for post-randomization increases in CK. Conclusions Perioperative rosuvastatin initiation increased the absolute risk of AKI after cardiac surgery by 4-5%, whether defined by creatinine or cystatin C, and led to higher post-operative KIM-1, suggesting a deleterious effect on renal function, possibly mediated by proximal tubular injury. Insulin treatment, baseline KIM-1, combined CABG/AVR surgery, and allocation to rosuvastatin were all independently associated with AKI by any definition.