The Association between Surgeon and Hospital Variation in Use of Laparoscopic Liver Resection and Short-Term Outcomes

Abstract

Introduction: Laparoscopic liver resection (LLR) offers equivalent oncologic outcomes to open resection while reducing complications and hospital stays in selected patients. However, LLR uptake may vary across providers and centres with differing expertise given the constraints of laparoscopy for liver resection. We examined how surgeon and hospital LLR uptake associates with short-term outcomes. Methods: We performed a population-based study including patients undergoing elective hepatectomy for gastrointestinal cancer over 2007-2019 at regionalized hepato-pancreato-biliary centres. For each surgeon and hospital, hierarchical regression estimated case-mix-adjusted rates of LLR, 90-day major morbidity/mortality, and prolonged length of stay. Linear regression estimated the association between rates of LLR and outcomes. Results: 5,015 patients (median age 63; 38.4% female; 61.7% major resections), 62 surgeons, and 11 hospitals were included. 17.7% of patients had LLR. Adjusted LLR rates ranged from 2.7-100% for surgeons and 6.9-56.6% for hospitals. We observed no association between LLR and 90-day morbidity/mortality rates across surgeons (β=0.03, 95%CI-0.03-0.09) and hospitals (β=0.06, 95%CI-0.24-0.37). Similarly, there was no association between rates of LLR and prolonged stays (surgeon: β=-0.02, 95%CI-0.05-0.01; hospital: β=-0.11, 95%CI-0.29-0.08) (Figure). Conclusions: Increased uptake of LLR for surgeons or hospitals was not associated with better short-term outcomes. While LLR associates with patient-level benefits in trial settings, this was not observed when considered from the provider and institution perspective using real-world data. While laparoscopy may contribute to better patient outcomes, further systematic processes of care should be identified to support ongoing incremental improvements.