Perioperative Dynamics of Renal Functional and Structural Damage Markers in Children, Undergoing Open Heart Surgery. Doctoral Thesis

Abstract

Introduction. Acute kidney injury (AKI) is a serious complication in the perioperative period and is consistently associated with increased morbidity and case fatality rate. This has been best researched in the cardiac surgery setting where it has been shown that up to 11.5–86% of patients exposed to cardiopulmonary bypass (CPB) will develop AKI, with 2–18.9% requiring renal replacement therapy (RRT). Depending on the criteria used to define AKI and the postoperative period studied, mortality ranges from 1% to 30% although this is consistently higher, approaching 80%, if RRT is required. AKI may contribute to chronic kidney disease and negative long–term health outcomes. There is not a clear understanding of the pathogenesis of CPB associated AKI and no effective treatment or prevention has yet been established for this syndrome. Aim. The aim of the study was to investigate perioperative dynamics of kidney injury markers and to identify early and sensitive marker of kidney injury, suitable for application in children, undergoing open heart surgery for correction of congenital heart lesions. Material and methods. Prospective uncontrolled cohort study was conducted between 2011 and 2015, 93 children with various congenital heart lesions undergoing CPB were enrolled. Serum creatinine (SCr) level was determined by Jaffé’s method (Cobas 6000 analyzer, Roche), serum Cystatin C (Cys C) was analized using particle–enhanced turbidimetric immunoassay (PETIA) for the quantitative determination, using COBAS C 501 analyzer, urine NGAL was determined by ARCHITECT system (Abbott Diagnostics, Illinois, USA). All hemodinamical data were extracted from anesthesia charts and transfered to MS Excell spredsheets for further processing. Then an average of all validated MAP recordings was calculated for each patient. Hypotension was defined as a MAP < 15% less than average MAP value. All MAP recordings were compared to average (except MAP recordings during CPB, when pulsatile blood flow was absent). Ratio of hypotensive MAP recordings (MAP < 15% less than average MAP value) to total MAP recordings was calculated and expressed as %. Results. AKI developed in 42 patients (45.6%) by meeting at least KDIGO stage I criteria (with SCr rise by more than 50% from the baseline). 38 patients comply with the 1st stage of AKI, 3 with 2nd and 3rd stage by 2 patients according the KDIGO classification and staging system. One patient having severity stage II and two patients having severity stage III of AKI required initiation of RRT, using peritoneal dialysis. Two patients from the RRT group survived, one died. Median intraoperative urine output was 2.32 mL/kg/h, (Range from 0.42–5.87 mL/kg/h). Median CPB time was 163 min., median aortic cross–clamping time was 97.9 min., cooling during CPB to 29.5°C. The diagnosis of AKI using SCr was delayed by 48 hours after CPB. In contrast, maximum increase in urine NGAL levels was 400–fold within 12 hours after CPB in patients having AKI. Median level of urine NGAL in the sample collected 12 hours after CPB was 132.85 ng/mL (IQR 60.78–257.23). In the group of patients with intact renal function the corresponding level of NGAL was 22.90 ng/mL (IQR12.3–75.65), p < 0.05. At 12 hours after CPB, AUC was 0.911, sensitivity of 88%, specificity of 92%, CI 95% 0.852– 0.971 and cut-off value of 70 ng/mL, p < 0.001. SCys C has maximum expression at 24 hours after CPB with median level of 1.31 mg/L (IQR 1.06–1.48) in AKI group vs. 0.77 mg/L (IQR 0.63–1.06) in non-AKI group, p < 0.05. ROC curve analysis showed AUC of S Cys C at 12 hours after CPB was 0.837 (95% CI 0.724–0.950), p < 0.001). Median fluid balance (FB) on the first postoperative day in non-AKI patients was 13.58 mL/kg (IQR 0.00–37.02) vs. 49.38 mL/kg (IQR 13.20–69.32) in AKI patients, p < 0.001. Ratio of hypotensive episodes (MAP < 15% from average MAP recordings) in AKI group was 19.66, (IQR 12.91–25.80) versus 11.03, (IQR 8.28–14.05) in non-AKI patients, p = 0.075. Conclusions. 1. AKI is a frequent complication after open heart surgery in children with congenital heart lesions. From 93 patients included in the study, 42 (45.2%) met at least KDIGO Stage I criteria for AKI. 2. Renal tubular injury marker- urinary NGAL has best predictive performance at 12 hours after CPB with AUC of 0.91, sensitivity of 88%, specificity of 92%, CI 95% 0.85–0.97 and cut-off value of 70 ng/ml, p < 0.001. Renal glomerular filtration marker – SCys C showed best performance at 24 hours: AUC of 0.84, sensitivity of 81%, specificity of 72%, CI 95% 0.84–0.93, p < 0.001. 3. FB is a sensitive marker of kidney dysfunction. Median FB in the 1st postoperative day has statistically significant difference between AKI patients: 49.38 mL/kg (13.20–69.32) versus 13.58 mL/kg in patients with intact kidney function (AUC = 0.84; p = 0.0011) and can be used as a marker of AKI. 4. The hypothesis of the association of intraoperative hypotension and postoperative AKI was not confirmed, using methodology, applied in this study. 5. AKI had severe impact on clinical outcome: Patients in AKI group had longer MV times, lengths of ICU stay and LOS in the hospital.