Machine learning, deep learning and hernia surgery. Are we pushing the limits of abdominal core health? A qualitative systematic review

INTRODUCTION

This systematic review aims to evaluate the use of machine learning and artificial intelligence in hernia surgery.

METHODS

The PRISMA guidelines were followed throughout this systematic review. The ROBINS-I and Rob 2 tools were used to perform qualitative assessment of all studies included in this review. Recommendations were then summarized for the following pre-defined key items: protocol, research question, search strategy, study eligibility, data extraction, study design, risk of bias, publication bias, and statistical analysis.

RESULTS

A total of 13 articles were ultimately included for this review, describing the use of machine learning and deep learning for hernia surgery. All studies were published from 2020 to 2023. Articles varied regarding the population studied, type of machine learning or Deep Learning Model (DLM) used, and hernia type. Of the thirteen included studies, all included either inguinal, ventral, or incisional hernias. Four studies evaluated recognition of surgical steps during inguinal hernia repair videos. Two studies predicted outcomes using image-based DMLs. Seven studies developed and validated deep learning algorithms to predict outcomes and identify factors associated with postoperative complications.

CONCLUSION

The use of ML for abdominal wall reconstruction has been shown to be a promising tool for predicting outcomes and identifying factors that could lead to postoperative complications.

A comparison of outcomes between class-II and class-III obese patients undergoing robotic ventral hernia repair: a multicenter study

BACKGROUND

Morbid obesity has been considered a contraindication to ventral hernia repair (VHR) in the past. However, the relationship between a greater body mass index (BMI) and adverse outcomes has yet to be established in the minimally invasive sphere, particularly with robotics, which may offer an effective surgical option in these high-risk patients. We sought to investigate this relationship by comparing the outcomes of class-II (BMI: 35-39.9 kg/m²) and class-III (BMI: ≥ 40 kg/m²) obese patients after robotic VHR (RVHR).

METHODS

Data were analyzed from two centers and six surgeons who performed RVHR between 2013 and 2020. Patients with a BMI > 35 kg/m² were included in the study. A 1:1 propensity score match (PSM) analysis was conducted to obtain balanced groups and univariate analyses were conducted to compare the two groups across preoperative, intraoperative, and post-operative timeframes. Postoperative complications and morbidity were reported according to the Clavien-Dindo Classification and comprehensive complication index (CCI^®) systems.

RESULTS

From an initial cohort of 815 patients, 228 patients with a mean BMI of 39.7 kg/m² were included in the study. PSM analysis stratified these into 69 patients for each of the class-II and class-III groups. When comparing matched groups, there were no differences in any of the variables across all timeframes, except for a higher rate of Polytetrafluoroethylene (PTFE)-based mesh use in the class-III group (39.1% vs 17.4%, p = 0.008). The estimated recurrence-free time was 76.4 months (95% CI = 72.5-80.4) for the class-II group and 80.4 months (95% CI = 78-82.8) for the class-III group.

CONCLUSION

This multicenter study showed no difference in outcomes after RVHR between matched class-II and class-III obese patients.

Orthotopic liver transplantation without venovenous bypass: 125 cases from a single center

AIM

This study analyzed a 10-year single-center experience in orthotopic liver transplantation (OLT) without venovenous bypass (VVB).

METHODS

We retrospectively analysed a nonrandomized series (1999-2008) of 125 adult OLT patients without VVB.

RESULTS

The main causes of liver failure were viral hepatitis (n = 39), alcoholic liver disease (n = 22), and liver cancer (n = 17). One-year survival was 76.4%. The most common postoperative complications were bile duct stenosis (n = 12), postoperative bleeding (n = 8), hepatic artery thrombosis (n = 7), and primary liver failure (n = 6). Twelve patients required hemodialysis and four underwent retransplantations of the liver. Fourteen patients died before postoperative day 30(th). Univariate analysis showed significant differences between patients who did and did not survive 30 days among donor death diagnoses (P = .05), red blood cell units transfused (P = .03), aspartate aminotranferase on the first postoperative day (P = .002), ABO type (P = .04), time of orotracheal intubation (P = .001), hemodialysis (P = .001), and period of postoperative vasoactive drug use (P = .006). The total length of orotracheal tube intubation showed a significant independent association with mortality before 30 days (P < .001).

CONCLUSION

OLT without VVB can be safely performed even in severe cases of chronic liver failure.