Pre-operative Sarcopenia Identifies Patients at Risk for Poor Survival After Resection of Biliary Tract Cancers

Introduction: Biliary tract cancers (BTC) are aggressive malignancies that require complex surgical procedures. Patients with BTC can present with skeletal muscle depletion, yet the effects of muscle wasting (sarcopenia) on outcomes have not been well studied. The objective of the current study was to define the impact of sarcopenia on survival among patients undergoing resection of BTC. Methods: Patients who underwent exploration for BTC who had a pre-operative CT scan available for review were identified. Body composition variables including total and psoas muscle area (cm2), muscle density (Hounsfield units), visceral fat area, subcutaneous fat area, and waist-to-hip ratio were analyzed at the level of L3. Outcomes were assessed according to the presence or absence of sarcopenia defined using sex- and BMI-specific threshold values for Psoas Muscle Index (PMI, cm2/m2). Results: Among 117 patients with BTC, 78 (67%) underwent curative-intent resection and 39 (33%) were explored but did not undergo resection due to metastatic/locally advanced disease. Tumor type included distal cholangiocarcinoma (n = 18, 15.4%), hilar cholangiocarcinoma (n = 27, 23.1%), gallbladder carcinoma (n = 52, 44.4%), and intrahepatic cholangiocarcinoma (n = 20, 17.1%). Median patient age was 65.6 years and 43.6% were male. Mean patient BMI was 26.1 kg/m2 among men and 27.5 kg/m2 among women. Overall, 41 (35.0%) patients had sarcopenia. Sarcopenia was associated with an increased risk of death among patients who underwent resection (HR 3.52, 95%CI 1.60–7.78, p = 0.002), which was comparable to patients with unresectable metastatic disease. Other factors such as low serum albumin (HR 3.17, 95% CI 1.30–7.74, p = 0.011) and low psoas density (HR 2.96, 95% CI 1.21–7.21, p = 0.017) were also associated with increased risk of death. Survival was stratified based on sarcopenia, psoas density, and serum albumin. The presence of each variable was associated with an incremental increased risk of death (0 variables ref.; 1 variable HR 3.8, 95% CI 1.0–14, p = 0.043; 2 variables HR 13.1, 95% CI 3.0–57.7, p = 0.001; 3 variables HR 14.6, 95% CI 2.5–87.1, p = 0.003). Patients who had no adverse prognostic factors had a 3-year OS of 67% versus no survival among patients with all 3 factors. Conclusions: Sarcopenia was common among patients undergoing resection of BTC, occurring in 1 of every 3 patients. Sarcopenia was associated with poor survival after resection, particularly among patients who experienced a recurrence. Body composition metrics such as sarcopenia and low psoas muscle density in addition to low albumin level were able to stratify patients into different prognostic categories.