Reproducibility and predictive value of three grading systems for intraoperative adverse events in a cohort of abdominal surgery

Intraoperative adverse events (iAEs) are increasingly recognized for their impact on patient outcomes. The Kaafarani classification and Surgical Apgar Score (SAS) were developed to assess the intraoperative course; however, both have their drawbacks. ClassIntra was validated for iAEs of any origin. This study compares the Kaafarani and SAS to ClassIntra considering predictive value and interrater reliability in a cohort of abdominal surgery to support implementation of a classification in clinical practice. The authors made use of the LAParotomy or LAParoscopy and ADhesiolysis (LAPAD) study database of elective abdominal surgery. Detailed descriptions on iAEs were collected in real-time by a researcher. For the current research aim, all iAEs were graded according ClassIntra, Kaafarani, and SAS (score ≤4). The predictive value was assessed using univariable and multivariable linear regression and the area under the receiver operating curve (AUROC). Two teams graded ClassIntra and Kaafarani to assess the interrater reliability using Cohen's Kappa. A total of 755 surgeries were included, in which 335 (44%) iAEs were graded according to ClassIntra, 228 (30%) to Kaafarani, and 130 (20%) to SAS. All classifications were significantly correlated to postoperative complications, with an AUROC of 0.67 (95% CI: 0.62-0.72), 0.64 (0.59-0.70), and 0.71 (0.56-0.76), respectively. For the secondary endpoint, the interrater reliability of ClassIntra with κ 0.87 (95% CI: 0.84-0.90) and Kaafarani 0.90 (95% CI: 0.87-0.93) was both strong. ClassIntra, Kaafarani, and SAS can be used for reporting of iAEs in abdominal surgery with good predictive value for postoperative complications, with strong reliability. ClassIntra, compared with Kaafarani and SAS, included the most iAEs and has the most comprehensive definition suitable for uniform reporting of iAEs in clinical practice and research.