Does the difficulty grade of laparoscopic liver resection for colorectal liver metastases correlate with long-term outcomes?

Validity of the Institut Mutualiste Montsouris classification system for robotic liver resection

BACKGROUND

The Institut Mutualiste Montsouris (IMM) classification system is one of several widely accepted difficulty scoring systems for laparoscopic liver resections. Nothing is yet known about the applicability of this system for robotic liver resections.

METHODS

We conducted a retrospective review of 359 patients undergoing robotic hepatectomies between 2016 and 2022. Resections were classified into low, intermediate, and high difficulty level. Data were analyzed utilizing ANOVA of repeated measures, 3 x 2 contingency tables, and area under the receiving operating characteristic (AUROC) curves. Data are presented as median (mean ± SD).

RESULTS

Of the 359 patients, 117 were classified as low-difficulty level, 92 as intermediate, and 150 as high. The IMM system correlates well with tumor size (p = 0.002). The IMM system was a strong predictor of intraoperative outcomes including operative duration (p<0.001) and estimated blood loss (EBL) (p<0.001). The IMM system also showed a strong calibration for predicting an open conversion (AUC=0.705) and intraoperative complications (AUC=0.79). In contrast, the IMM system was a poor predictor of postoperative complications, mortality, and readmission.

CONCLUSION

The IMM system provides a strong correlation with intraoperative, but not postoperative outcomes. A dedicated difficulty scoring system should be developed for robotic hepatectomy.

Systematic review and meta-analysis of difficulty scoring systems for laparoscopic and robotic liver resections

INTRODUCTION

The ability to stratify the difficulty of minimally invasive liver resection (MILR) allows surgeons at different phases of the learning curve to tackle cases of appropriate difficulty safely. Several difficulty scoring systems (DSS) have been formulated which attempt to accurately stratify this difficulty. The present study aims to review the literature pertaining to the existing DSS for MILR.

METHODS

We performed a systematic review and metanalysis of the literature reporting on the formulation, supporting data, and comparison of DSS for MILR.

RESULTS

A total of 11 studies were identified which reported on the formulation of unique DSS for MILR. Five of these (Ban, Iwate, Hasegawa, Institut Mutaliste Montsouris [IMM], and Southampton DSS) were externally validated and shown to predict difficulty of MILR via a range of outcome measures. The Ban DSS was supported by pooled data from 10 studies (9 LLR, 1 RLR), Iwate by 10 studies (8 LLR, 2 RLR), Hasegawa by four studies (all LLR), IMM by eight studies (all LLR), and Southampton by five studies (all LLR). There was no clear superior DSS.

CONCLUSION

The existing DSS were all effective in predicting difficulty of MILR. Present studies comparing between DSS have not established a clear superior system, and the five main DSS have been found to be predictive of difficulty in LLR and two of these in RLR.

A novel difficulty scoring system of laparoscopic liver resection for liver tumor

Objectives

To develop a novel difficulty scoring system (NDSS) to predict the surgical difficulty of laparoscopic hepatectomy.

Patients and methods

A total of 138 patients with liver tumors performed liver resection (LLR) between March 2017 to June 2022 were selected from Affiliated Hospital of Jiangnan University and Wujin Hospital Affiliated with Jiangsu University.

Patient demographics, laboratory tests, intraoperative variables, pathological characteristics were assessed. We also assessed the Child Pugh score and the DSS-B score.

Results

Patients were divided into training and testing cohort according to their hospital. Patients in training cohort were divided into high and low difficult groups based on operation time, blood loss and conversion. Higher percentage of patients with malignant liver tumor (87.0% vs. 58.1%; P = 0.003) or history of hepatobiliary surgery (24.1% vs. 7.0%; P = 0.043) in high difficult group than in low difficult group. To improve the difficulty scoring system, we incorporated the history of hepatobiliary surgery and nature of the tumor. A novel difficulty scoring system was established. The results showed that the operation time (P < 0.001), blood loss (P < 0.001), ALT (P < 0.001) and AST (P = 0.001) were associated with the novel difficulty score significantly. Compared with DSS-B, the NDSS has a higher area under the receiver operating characteristic (AUROC) (0.838 vs. 0.814). The nomogram was established according to the NDSS. The AUROCs of the nomogram in training and testing cohort were 0.833 and 0.767. The calibration curves for the probability of adverse event showed optimal agreement between the probability as predicted by the nomogram and the actual probability.

Conclusions

We developed a nomogram with the NDSS that can predict the difficulty of LLR. This system could more accurately reflect the difficulty of surgery and help liver surgeons to make the surgical plan and ensure the safety of the operation.