The impact of myosteatosis on postoperative outcomes and survival of patients undergoing pancreatoduodenectomy for suspected/confirmed malignancy.

ABSTRACT

Backgrounds/Aims: While the effects of myosteatosis are emerging, the evidence for its use as a predictor of outcomes in patients undergoing pancreatoduodenectomy (PD) still needs to be established. The study aims to evaluate the effect of myosteatosis on the short- and long-term outcomes of PD. Methods: We analyzed the effect of myosteatosis on the short- and long-term outcomes of patients who underwent PD between July 2006 and May 2013. Myosteatosis was measured retrospectively from preoperative computed tomography (CT) at the L3 vertebra level, and dichotomized as a binary exposure variable as < 38.5 Hounsfield unit (HU) for males, and < 36.1 HU for females. Results: A total of 214 patient (median age 62 years, range 41-80 years) CTs were analyzed for myosteatosis. Overall, 120/214 (56.1%) patients were classed as having myosteatosis. Both groups had similar comorbidity profiles. The presence of myosteatosis was not shown to increase the rate of any short- or long-term complication. However, pancreatic leak (29.8% vs. 13.3%; p = 0.006) and postoperative bleeding (13.8% vs. 5.0%; p = 0.034) were higher in the non-myosteatosis group. The median intensive care (2 days) and hospital stay (12 days) were the same in both groups. The 30-day mortality (myosteatosis 3.3% vs. non-myosteatosis 3.2%; p = 0.95), and 5-year overall survival (myosteatosis 26.7% vs. non-myosteatosis 31.9%; p = 0.5), were similar in both groups. Conclusions: We have found no evidence supporting myosteatosis affecting either the short-term or long-term outcomes of patients undergoing PD for suspected/confirmed malignant tumors.