Laparoscopic versus open liver resection in patients aged at least 80 years: retrospective propensity score-matched cohort study

Abstract

BACKGROUND: Laparoscopic liver resection has been associated with less morbidity than, and similar global outcomes to, open liver resection. There is no robust evidence that these outcomes lead to similar clinical outcomes in patients aged over 80 years. The aim of this study was to analyse the short-term outcomes between open and laparoscopic liver resection in patients over 80 years old. METHODS: A retrospective analysis was undertaken. The study population comprised patients aged ? 80 years who underwent laparoscopic or open liver resection between January 2014 and December 2019, and who presented with resectable malignant tumours. The primary outcome was postoperative morbidity, according to Dindo-Clavien grading. Cox regression models were used to compute hazard ratios and 95% confidence intervals. Propensity score matching (1 : 1) was performed to balance the two groups according to independent prognostic factors for morbidity. RESULTS: A total of 988 patients were analysed from 34 centres (16 from Asia, 14 from Europe and 4 from America): 487 in the open group and 501 in the laparoscopic group. Independent risk factors associated with severe morbidity were the open approach (hazard ratio 1.59, 95% confidence interval 1.19 to 2.11; P < 0.001), Charlson Co-morbidity Index score > 7 (HR 1.69, 1.26 to 2.27; P < 0.001), more than one resected tumour (hazard ratio 1.55, 1.13 to 2.11; P = 0.006), major hepatectomy (hazard ratio 1.86, 1.22 to 2.83; P = 0.003), and Iwate score ? 7 (hazard ratio 1.43, 1.02 to 2.01; P = 0.03). Before propensity score matching, severe morbidity, length of intensive care unit stay, 90-day mortality, length of hospital stay, and readmission were better in the laparoscopic group (P < 0.050). These observations were confirmed after propensity score matching. CONCLUSION: The laparoscopic approach is a safe procedure for elderly patients, with better morbidity and mortality outcomes than the open approach, and should be considered as a default option.