Back to Annual Meeting Program

Background and Objectives:
Recent reports suggested that early removal of surgical drains following pancreaticoduodenectomy (PD) reduce the postoperative complications including postoperative pancreatic fistulas (POPFs) with decreased hospital stay and costs. However, the indicator for proper drain management still remains unclear. The aim of this study was to identify the indicators for proper drain management after PD.
Methods:
Prospectively collected data from 200 consecutive patients who underwent PD were evaluated. (86 females and 114 males, median age 66 years; range 19-88). The pancreatic anastomosis was reconstructed with a two-layered duct-to-mucosa pancreaticogastrostomy with internal stent. POPF was assessed using the criteria of International Study Group Pancreatic Fistula (ISGPF). The surgical complications were classified according to the Clavien-Dindo (C-D) classification. Predictive clinical factors for clinically relevant POPFs (Grade B and C) were analyzed by logistic regression analysis. Management of surgical drains was also analyzed.
Results:
Of 200 patients, 44 developed pancreatic fistulas; grade A in 29 patients, grade B in 12, and grade C in 3. Thus, clinically relevant POPFs were occurred in 15 (8%). Severe surgical complications (over C-D classification Grade3) were occurred in 17 (9%).
By univariate analysis, drain amylase on postoperative day (POD) 2, 3, 4, 5, the color of surgical drain fluid (dark red) on POD1,3,4, WBC on POD3,4, serum C-reactive protein (CRP) on POD 3,4, and body temperature on POD3,4,5 were found to be significantly associated with clinically relevant POPFs (p<0.05). By multivariate analysis on POD4, the color of surgical drain fluid (dark red) [p=0.01, Odds ratio 9.8, 95%CI 1.7-58.3] and serum CRP [p=0.03, Odds ratio 1.2, 95%CI 1.1-1.4] were found to be independent predictive factors for clinically relevant POPFs.
Based on the receiver operating characteristic curve analysis, serum CRP>15.6 mg/dl on POD 4 displayed the optimal sensitivity (80%) and specificity (87%).
In the patients with serous fluid in surgical drain, and serum CRP levels<15.6mg/dl on POD4 (n=163, drains to be removed on POD5), clinically relevant POPFs were occurred in 2 (1%). 5 patients (3%) required additional percutaneous drainage. Severe surgical complications were occurred in 12 (7%). On the other hand, in the patients with the dark red fluid in surgical drain, or serum CRP levels>15.6mg/dl on POD4 (n=37, drains to be removed on POD6 or longer), clinically relevant POPFs were occurred in 13 (35%). 5 patients (14%) required additional percutaneous drainage. Severe surgical complications were occurred in 6 (16%) including one surgical mortality.
Conclusions:
A combination of serum CRP levels and the color of surgical drain fluid on POD4 may be indicators for proper management of surgically placed drains following PD.