Postoperative Mortality in Patients With Cirrhosis: Reconsidering Expectations

Background A diagnosis of cirrhosis increases a patient's risk of postoperative mortality. Surgeons are reticent to operate when cirrhosis is known unless no option is available. This study aimed to identify the modern perioperative risk in cirrhotic patients undergoing intervention under general anesthesia for non-transplant operations. Methods A retrospective chart review was conducted utilizing the Rush Medical Center electronic medical record. All patients over 18 years of age with a diagnosis of cirrhosis undergoing intervention between 2009 and 2019 were reviewed. 90-day mortality rates in patients grouped by Child's score, Model for End-Stage Liver Disease (MELD), and Model for End-Stage Liver Disease with sodium incorporated (MELDNa) were compared to previously accepted rates. Results 93 patients (46% women) aged 22-72 years of all Child-Turcot-Pugh (CTP) (40% A, 36% B, and 25% C) classifications and MELD/MELDNa ranging 6-40 were analyzed. 90-day mortality of the entire population was 16%, significantly lower than expected based on CTP score (16% vs. 32%; P = .0005), MELD (16% vs. 41%; P < .0001), and MELDNa (16% vs. 46.8%; P < .0001). This was also true for CTP-B patients (12% vs. 30%; P = .025), CTP-C patients (35% vs. 70%; P = .0002), patients with MELD >14 (27% vs. 70%; P < .001), and patients with MELDNa >14 (23% vs. 70%; P < .0001). Conclusion Data indicate that perioperative mortality is lower than widely accepted. This suggests the need for a national database study using a representative population to determine the risk of mortality for patients with cirrhosis having surgery in recent times. Accurate estimation of this risk allows for meaningful discussion between physicians and patients when deciding to proceed with elective, necessary operations.