Background: Oliguria is often viewed as a sign of renal hypoperfusion and an indicator for volume expansion during surgery. However, the prognostic association and the predictive utility of intraoperative oliguria for postoperative acute kidney injury (AKI) are unclear.
Methods: We conducted a retrospective cohort study on patients undergoing major thoracic surgery in an academic hospital to assess the association of intraoperative oliguria with postoperative AKI and its predictive value. To contextualise our findings, we included our results in a meta-analysis of observational studies on the importance of oliguria during noncardiac surgery.
Results: In our cohort study, 3862 patients were included; 205 (5.3%) developed AKI after surgery. Intraoperative urine output of 0.3 ml kg-1 h-1 was the optimal threshold for oliguria in multivariable analysis. Patients with oliguria had an increased risk of AKI (adjusted odds ratio: 2.60; 95% confidence interval: 1.24-5.05). However, intraoperative oliguria had a sensitivity of 5.9%, specificity of 98%, positive likelihood ratio of 2.74, and negative likelihood ratio of 0.96, suggesting poor predictive ability. Moreover, it did not improve upon the predictive performance of a multivariable model, based on discrimination and reclassification indices. Our findings were generally consistent with the results of a systematic review and meta-analysis, including six additional studies.
Conclusions: Intraoperative oliguria has moderate association with, but poor predictive ability for, postoperative AKI. It remains of clinical interest as a risk factor potentially modifiable to interventions.
Keywords: acute kidney injury; anaesthesia; oliguria; postoperative outcome; risk prediction; systematic review and meta-analysis; thoracic surgery.
Copyright © 2020 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.