Factors that predict the occurrence of and recovery from non-alcoholic fatty liver disease after pancreatoduodenectomy.

BACKGROUND

The manifestations of non-alcoholic fatty liver disease in patients who have undergone pancreatoduodenectomy differ from those associated with obesity-related non-alcoholic fatty liver disease. This study aimed to identify factors that predicted the occurrence of and recovery from non-alcoholic fatty liver disease after pancreatoduodenectomy.

METHODS

This retrospective study included 120 patients who underwent pancreatoduodenectomy between April 2004 and December 2013. Non-alcoholic fatty liver disease was diagnosed using unenhanced computed tomography as a value of <40 Hounsfield units. Recovery from non-alcoholic fatty liver disease was based on increases in liver computed tomographic attenuation values. Pre-, intra-, and postoperative factors were analyzed using univariate analysis and multivariable logistic regression models.

RESULTS

Non-alcoholic fatty liver disease occurred after pancreatoduodenectomy in 45 patients (38%), and in 11 of 41 patients (27%) who received prophylactic pancreatic enzyme supplementation therapy and in 34 of 79 patients (43%) who did not (P = .082). Six patients received therapeutic supplementation after diagnosis. The non-alcoholic fatty liver disease recovery rates in patients who did and did not receive pancreatic enzyme supplementation therapy were 100% and 58%, respectively (P = .069). Multivariable analysis identified a high body mass index, small pancreatic volume, long operative time, and a high aspartate aminotransferase/alanine aminotransferase ratio 1 month after pancreatoduodenectomy as independent risk factors. A small diameter main pancreatic duct, a low serum amylase level at postoperative day 28, and a high minimum liver computed tomographic value predicted recovery from non-alcoholic fatty liver disease.

CONCLUSION

The non-alcoholic fatty liver disease occurrence rate in patients undergoing pancreatoduodenectomy is high, but in about half of these patients, non-alcoholic fatty liver disease will resolve without any enzyme supplementation. Prophylactic supplementation in the postoperative management of pancreatoduodenectomy patients should be based on risk factors, and therapeutic supplementation should be based on recovery factors.