Prediction of pancreatic fistula after pancreatoduodenectomy by preoperative dynamic computed tomographic imaging and fecal elastase-1 levels

Abstract

Objective: The aim of this study was to evaluate the stand-alone and combined abilities of preoperative dynamic computed tomography (CT) and fecal elastase-1 levels to predict the developement of pancreatic fistulae after pancreatoduodenectomy.

Materials and Methods: For 146 consecutive patients from January 2006 to March 2015, the preoperative dynamic CT images and medical records of the preoperative fecal elastase-1 levels and postoperative pancreatic fistulae (POPF) were reviewed. The CT attenuation values were measured on the unenhanced images (P) and images obtained in the arterial (A) and equilibrium phases (E) after contrast administration. The three CT enhancement ratios were calculated as (A - P)/P, (E - P)/P and (E - P)/(A - P). The correlation of the CT enhancement ratios and preoperative fecal elastase-1 levels with the development of POPF was performed using the independent two-sample t-test, logistic regression models, receiver operating characteristic (ROC) curve analysis and Youden method. The combined ability of the CT and fecal elastase-1 level findings for the prediction of POPF developement was assessed by ROC comparison and tree analysis.

Results: The mean values of (E - P)/P and (E - P)/(A - P) were significantly higher among the patients without POPF (n = 107) than among the patients with POPF (n = 39) (2.256 ± 3.633 vs. 1.038 ± 0.508; P = 0.001 and 1.116 ± 1.395 vs. 0.713 ± 0.294; P = 0.006, respectively). In the logistic regression analyses, (E - P)/P and (E - P)/(A - P) were significant predictors for the development of pancreatic fistulae (odds ratio [OR] = 0.243, P = 0.002 and OR = 0.176, P = 0.014, respectively). The mean preoperative fecal elastase-1 levels were higher (OR = 1.003, P = 0.034) in the POPF group than in the non-POPF group. In the ROC comparison, there were no significant differences in the areas under the curve (AUC) between the prediction values of CT enhancement ratios and fecal elastase-1 combined and those of CT enhancement ratios alone (P = 0.897, P = 0.917). Tree analysis revealed that the CT enhancement ratios of the equilibrium phase were more powerful and effective predictors of POPF than the fecal elastase-1 levels.

Conclusion: The CT enhancement ratios of the equilibrium phase and the preoperative fecal elastase-1 levels might be useful predictors of the risk of developing a pancreatic fistula following pancreatoduodenectomy. Furthermore, CT enhancement ratios of the equilibrium phase alone could be powerful enough to predict the pancreatic fistula.