Key result
Adding intraoperative variables to preoperative data improved the AUC for predicting postoperative AKI from 0.804 to 0.817 (P<.001), though clinical improvements in risk classification were modest.
Why the study?
Postoperative acute kidney injury is a common complication of noncardiac surgery, but current risk stratification models have substantial limitations, including limited use of perioperative data.
Does adding intraoperative data to preoperative data improve the prediction of postoperative AKI in adults undergoing major noncardiac surgery?
Observational (n=42,615)
Yes
Does adding intraoperative data to preoperative data improve the prediction of postoperative AKI in adults undergoing major noncardiac surgery?
Effect estimate: AUC 0.817 (95% CI 0.802-0.832)
p-value: p=<.001
Adding intraoperative data to preoperative models provides statistically significant but clinically modest improvements in predicting postoperative AKI after major noncardiac surgery.
No takes yet. Share an insight, caveat, or question.
Modest AKI prediction gains from intraoperative data should not yet change practice; leaves open need for prospective outcome validation.
Lei et al. (2019) conducted an observational in Postoperative acute kidney injury in major noncardiac surgery (n=42,615). Addition of intraoperative variables to preoperative data vs. Preoperative variables alone was evaluated on Postoperative AKI within 7 days after surgery (AUC 0.817, 95% CI 0.802-0.832, p=<.001). Adding intraoperative variables to preoperative data improved the AUC for predicting postoperative AKI from 0.804 to 0.817 (P<.001), though clinical improvements in risk classification were modest.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: