BACKGROUND

Analytical morphometric assessment has recently been proposed to improve preoperative risk stratification. However, the relationship between body composition and outcomes following pancreaticoduodenectomy is still unclear. The aim of this study was to assess the impact of body composition on outcomes in patients undergoing pancreaticoduodenectomy for cancer.

METHODS

Body composition parameters including total abdominal muscle area (TAMA) and visceral fat area (VFA) were assessed by preoperative staging CT in patients undergoing pancreaticoduodenectomy for cancer. Perioperative variables and postoperative outcomes (mortality or postoperative pancreatic fistula) were collected prospectively in the institutional pancreatic surgery database. Optimal stratification was used to determine the best cut-off values for anthropometric measures. Multivariable analysis was performed to identify independent predictors of 60-day mortality and pancreatic fistula.

RESULTS

Of 202 included patients, 132 (65·3 per cent) were classified as sarcopenic. There were 12 postoperative deaths (5·9 per cent), major complications developed in 40 patients (19·8 per cent) and pancreatic fistula in 48 (23·8 per cent). In multivariable analysis, a VFA/TAMA ratio exceeding 3·2 and American Society of Anesthesiologists grade III were the strongest predictors of mortality (odds ratio (OR) 6·76 and 6·10 respectively; both P < 0·001). Among patients who developed major complications, survivors had a significantly lower VFA/TAMA ratio than non-survivors (P = 0·017). VFA was an independent predictor of pancreatic fistula (optimal cut-off 167 cm(2) : OR 4·05; P < 0·001).

CONCLUSION

Sarcopenia is common among patients undergoing pancreaticoduodenectomy. The combination of visceral obesity and sarcopenia was the best predictor of postoperative death, whereas VFA was an independent predictor of pancreatic fistula.