Predictive value of abdominal drain amylase for postoperative pancreatic fistula following gastric resection: A systematic review and meta-analysis

Abstract Background Radical dissection of lymph nodes, accompanying gastric cancer resection, can lead to collateral damage to the pancreas and development of postoperative pancreatic fistula (POPF). Methods We searched the Cochrane Library, MEDLINE, Embase, and Web of Science up to April 21, 2020, to identify studies documenting the value of abdominal drain amylase level (d-AMY) on postoperative day 1 (POD1) as a predictor of POPF after gastric surgery. The quality of selected studies was assessed using the QUADAS-2 questionnaire. The diagnostic value of d-AMY on POD1 for prediction of POPF was first assessed by calculation of pooled estimates of sensitivity, specificity, likelihood ratios (LR), and the diagnostic odds ratio (DOR). Secondly, the accuracy was further demonstrated graphically with the hierarchical summary receiver operating curve (hSROC). PROSPERO registration number: CRD42020181145. Results DOR of nine studies (cases n = 1856) observing the occurrence of POPF after measurement of d-AMY on POD1 was 18.7 (95%CI: 10.0, 34.8), and the area under hSROC was 0.88 ± 0.02. The pooled sensitivity was 0.74 (95%CI: 0.66, 0.81) and specificity 0.84 (95%CI: 0.82, 0.86). The negative LR was at the lowest point of 0.16 (95%CI: 0.07, 0.37) at the cutoff value for d-AMY of 941 IU/L, while the positive LR ranged from 4.4 (cutoff 2119 IU/L) to 6.2 (cutoff 5000 IU/L). Conclusion d-AMY on POD1 can be used as an accurate and non-invasive predictor of POPF in the earliest stage of postoperative course following gastric cancer resection; value ≤ 941 IU/L warrants early drain removal and low probability of POPF (any grade).