Safety and Economics of Parenchymal Transection in Minimally Invasive Liver Surgery

LIVER-MIND 2 is a multicentre, observational, retrospective registry coordinated by the SICE HPB and Minimally Invasive Surgery Working Group, enrolling consecutive adults undergoing elective minimally invasive (laparoscopic, robotic or hybrid) liver resection between January 2021 and December 2025 across ≥15 high-volume or dedicated centres (target >30 centres). The first 65 robotic cases per centre are excluded to account for the robotic HPB learning curve. The primary exposure is the transection strategy/primary device, with the principal contrast between cavitron ultrasonic surgical aspirator (CUSA) and energy devices. The primary outcome is a 90-day composite transection-related complication: significant bleeding (estimated blood loss >500 mL and/or transfusion and/or reintervention for haemostasis) and/or clinically relevant bile leak (ISGLS grade B-C). Secondary outcomes include conversion, Clavien-Dindo ≥IIIa morbidity, post-hepatectomy liver failure (PHLF), operative time, length of stay, 90-day readmission and mortality, oncological radicality, and direct costs. The target sample size is ≈900 patients (80% power, two-sided α = 0.05, expected composite event rate 12%, minimal clinically relevant difference 5%, 5% drop-out inflation). The primary analysis uses multivariable logistic regression with propensity-score matching and inverse-probability-of-treatment weighting; pre-specified subgroup analyses and a statistical analysis plan are deposited before database lock. An exploratory cost analysis adopts a hospital perspective combining DRG-based standard costing and device-level micro-costing.

Inclusion criteria

• Age ≥ 18 years
• Elective minimally invasive liver resection
• Laparoscopic, robotic or hybrid approach
• Benign or malignant disease
• Signed informed consent for use of anonymised data
• Data completeness ≥ 70% in the CRF

Exclusion criteria

• Urgent or emergency surgery
• Liver transplantation (donor / recipient)
• Associated major non-hepatic biliary or vascular resection
• Lack of consent or follow-up
• Participation in conflicting interventional trials
• Surgery before January 2021
• Robotic procedures performed during the learning-curve phase (first 65 robotic cases per centre, per Rahimli et al.)