Primary and secondary liver failure after major liver resection for perihilar cholangiocarcinoma

P.B Olthof,J Erdmann,S Buettner,O Busch,J De Jonge,A.-M Van Keulen,R.-J Swijnenburg,T Van Gulik,W Polak,M Besselink,J Ijzermans,B Groot Koerkamp

doi:10.1016/j.hpb.2021.08.312

Abstract

Purpose: The aim of this study was to investigate the incidence and risk factors of primary and secondary LF after major liver resection for pCCA. Method: All patients who underwent a major liver resection for suspected pCCA between 2000 and 2020 at two tertiary-referral hospitals were included. LF was defined according to the International Study Group for Liver Surgery (ISGLS) criteria and only grade B/C was considered clinically relevant. Primary LF was defined as failure without any underlying postoperative cause, and secondary as LF with an onset after an underlying postoperative complication as a cause. Results: The incidence of LF and 90-day mortality were 20.9% and 17.0% in the 253 included patients. The incidences of primary LF was 9.1% and secondary 11.9%. Abdominal sepsis, portal vein thrombosis, and arterial thrombosis were the most frequent causes. The absence of preoperative remnant liver assessment and blood loss were independent risk factors for primary LF. Independent risk factors for secondary LF were ECOG performance status, percutaneous biliary drainage, and preoperative cholangitis. Conclusion: LF after major liver resection for pCCA occurred in one of every five patients. Preoperative assessment of the remnant liver seems essential to avoid primary LF with concomitant postoperative mortality.

Highlights

Residual positive margins or undiagnosed lesions, among others impact on disease-free survival after liver resection for liver malignant tumors
We aim to ascertain the accuracy of indocyanine green with intraoperative ultrasonography for intraoperative detection of nodules, mainly in the first two cm of liver parenchyma
The accuracy for malignant tumor detection for intraoperative ultrasonography alone, for indocyanine green alone and the coincidence of lesions detected with ultrasonography and indocyanine green together was, 65 (S 80%, E 50%), 70 (S 100%, E 40%) and 85 (S 80%, E 90%), respectively (Figure 1)

Summary

Introduction

Residual positive margins or undiagnosed lesions, among others impact on disease-free survival after liver resection for liver malignant tumors. We aim to ascertain the accuracy of indocyanine green with intraoperative ultrasonography for intraoperative detection of nodules, mainly in the first two cm of liver parenchyma. Methods: Between July 2019 until December 2020, indocyanine green was administered intravenously 1 day before surgery in 20 patients. A total of 42 liver tumors were identified in preoperative imaging. Iwate Score was 6.6 (DE 2)(Table 1). Comprehensive complication index was 13.9 (20.7) (Table 1). Thirty-four out of 42 tumors detected in preoperative imaging were histologically confirmed as cancer. Ten out of 20 tumors detected intraoperatively were confirmed histologically as cancer. Conclusions: We strongly recommend the use of preoperative indocyanine green because it improves the accuracy of detecting malignant tumors when it is used in combination with ultrasonography

Objectives

Findings

Methods