Two-point measurement of amylase in drainage fluid predicts severe postoperative pancreatic fistula after gastric cancer surgery

Early identification of patients at risk of postoperative pancreatic fistula (POPF) allows appropriate management after gastrectomy. Although some reports have suggested a correlation between POPF and the concentration of amylase in drained abdominal fluid (D-AMY), this has not been proven to impact sufficiently on clinical decision-making. A sustained high level of D-AMY is often assumed to be due to unsatisfactory drainage or excessive pancreatic leakage. We assessed the clinical utility of measuring D-AMY on postoperative day (POD) 1 and POD3 for prediction of POPF. Starting in April 2014, 801 patients who underwent radical gastrectomy with prophylactic drain placement were consecutively enrolled. We routinely measured D-AMY on POD1 and POD3, and compared the incidence of problematic POPF and clinical factors including D-AMY. We also attempted to clarify whether such two-point D-AMY measurement was clinically useful for patient management after gastrectomy. Fifty-one of the patients (6.4%) developed Clavien–Dindo grade III or worse POPF. Using D-AMY cutoffs of 2218 IU/L on POD1 and 555 IU/L on POD3, the patients were successfully classified. The highest risk group, in which D-AMY was higher than the cut-off value on both POD1 and POD3, showed a significantly high rate of occurrence (33/105, 31.4%) and high positive likelihood ratio (6.74). Multivariate analysis showed that classification into this highest risk group was an independent risk factor for development of severe POPF (odds ratio 15.2, 95% CI 7.92–29.0). Two-point measurement of D-AMY may be an efficient tool for achieving individualized management of POPF following radical gastrectomy.