A Comparison of AKIN, KDIGO, and RIFLE Definitions to Diagnose Acute Kidney Injury and Predict the Outcomes after Cardiac Surgery in a South Asian Cohort

Background: Acute kidney injury (AKI) after cardiac surgery is associated with an increased mortality and morbidity. Different definitions for AKI have been used such as Acute Kidney Injury Network (AKIN), Kidney Disease: Improving Global Outcomes (KDIGO), or risk, injury, failure, loss, end-stage kidney disease (RIFLE). Each of these definitions has their own benefits and limitations for predicting the degree of AKI and adverse outcomes following cardiac surgery. This study was aimed to compare the three AKI definitions to diagnose AKI and their predictive ability for mortality and morbidity after isolated coronary artery bypass surgery (CABG) in a South Asian cohort. Material and Methods: A single-center retrospective review was conducted on 1,508 patients having undergone isolated CABG surgery from January 2015 to January 2019. AKI was assessed on three definitions, and comparative receiver operating characteristics curves were built against the outcomes to assess discriminative power of each. Results: Mean age of participants was 59.43 (+/- 1.12) years, predominantly males (82.6%). Patients with AKI were elder, more likely to be diabetic and hypertensive. AKI by any definition occurred in 58.7% (885/1,508) patients. Frequency of AKI was 508 (33.7%), 517 (34.4%), and 871 (57.8%) on AKIN, KDIGO, and RIFLE criteria, respectively. The proportion of patients with RIFLE-risk 619 (41%) was greater compared with AKIN stage 1 (342 [22.7%]) and KDIGO stage 1 (330 [21.9%]), while for stages 2 and 3 the proportions were comparable across the three systems. Area under the curve (AUC) for 30-day mortality for AKIN was (0.786, [95% CI: 0.764-0.806]), KDIGO: (0.796, [95% CI: 0.775-0.816]) and for RIFLE (0.844, [95% CI: 0.825-0.862]). AUC for overall morbidity was in undesirable ranges (i.e., >0.5-<= 0.7) for all three definitions. Conclusion: In a South Asian cohort, performance of AKIN and KDIGO criteria was comparable to diagnose AKI, while RIFLE definition, though overestimated the incidence of AKI particularly stage 1, and had an excellent discriminatory power to predict mortality compared to other two definitions.