Comparison of IWATE, Institut Mutualiste Montsouris, and Southampton Laparoscopic Liver Resection Difficulty Scoring Systems for Predicting Intra and Postoperative Outcomes in Robotic Hepatectomy.

BACKGROUND

IWATE, Institut Mutualiste Montsouris (IMM), and Southampton are established difficulty scoring systems (DSS) for laparoscopic hepatectomy, yet none specifically address robotic hepatectomy. Our study evaluates these 3 DSS for predicting perioperative outcomes in robotic hepatectomy.

METHODS

With IRB approval, we prospectively followed 359 consecutive patients undergoing robotic hepatectomies, assessing categorical metrics like conversions to open, intra/postoperative issues, Clavien-Dindo Score (≥III), 30 and 90-day mortality, and 30-day readmissions using Receiver Operating Characteristic (ROC) curves and Area Under the Curve (AUC) to determine efficacy in predicting their occurrence for each DSS. Continuous metrics such as operative duration, estimated blood loss (EBL), length of stay, and total cost were analyzed using Spearman's correlation and regression. Predictive strength was significant with an AUC or correlation ≥.700 and -value ≤.05.

RESULTS

IMM had highest predictive accuracy for conversions to open (AUC = .705) and postoperative complications (AUC = .481). Southampton was most accurate in predicting Clavien Dindo ≥ III complications (AUC = .506). IWATE excelled in predicting 30-day mortality (AUC = .552), intraoperative issues (AUC = .798), In-hospital mortality (AUC = .450), 90-day mortality (AUC = .596), and readmissions (AUC = .572). Regression showed significant relationships between operative duration, EBL, and hospital cost with increasing scores for all DSS ( ≤ .05).

DISCUSSION

Statistical analysis of the 3 DSS indicates that each has specific strengths that can best predict intra- and/or postoperative outcomes. However, all showed inaccuracies and conflicting relationships with the variables, indicating lack of substantial hierarchy between DSS. Given these inconsistencies, a dedicated comprehensive DSS should be created for robotic hepatectomy.