胰十二指肠切除术后早期胰瘘的影响及预测因素分析
| 作者: |
1黄涛,
1杨美文,
1张雷达
1 陆军军医大学附属西南医院 肝胆外科，重庆 400038 |
| 通讯: |
张雷达
Email: 2518569931@qq.com |
| DOI: | 10.3978/.2018.03.005 |
摘要
目的：探讨胰十二指肠切除术（PD）后早期胰瘘的影响及预测因素，以指导术后更早拔除腹腔引流管，促进患者康复。
方法：回顾西南医院肝胆外科2013年1月—2015年10月收治的430例PD手术患者的临床资料，用单因素与多因素统计法分析以及受试者操作特性曲线（ROC）分析术后早期胰瘘的影响因素及因素的预测价值。
结果：430例患者中发生术后早期胰瘘116例（26.9%），此外，在未发生术后早期胰瘘的患者中，术后5 d内拔除腹腔引流管患者术后胰瘘、腹腔感染以及总并发症发生率均明显低于5 d后拔管患者（3.1% vs. 12.1%、9.2% vs. 20.3%、24.5% vs. 42.0%，均P<0.05）。单因素及多因素Logistic回归分析显示，术后第1天腹腔引流液淀粉酶水平（DFA1）与术后第1天血清淀粉酶水平（SA1）为术后早期胰瘘的独立影响因素（OR=1.000、1.004，95% CI=1.000~1.001、1.001~1.006，均P<0.05）；ROC分析结果显示，DFA1预测PD术后早期胰瘘发生的曲线下面积（AUC）为0.916，明显大于SA1（0.745），DFA1≥494.75 IU/L敏感度、特异度、阳性预测值、阴性预测值分别为91.7%、80.8%、62.7%、96.5%。
结论：DFA1是PD后早期胰瘘重要的影响与预测因素，对于DFA1<494.75 IU/L的患者，可于术后第3天安全拔除腹腔引流管，并对患者实施快速康复策略。
关键词：
胰十二指肠切除术；胰腺瘘；淀粉酶类；因素分析，统计学
方法：回顾西南医院肝胆外科2013年1月—2015年10月收治的430例PD手术患者的临床资料，用单因素与多因素统计法分析以及受试者操作特性曲线（ROC）分析术后早期胰瘘的影响因素及因素的预测价值。
结果：430例患者中发生术后早期胰瘘116例（26.9%），此外，在未发生术后早期胰瘘的患者中，术后5 d内拔除腹腔引流管患者术后胰瘘、腹腔感染以及总并发症发生率均明显低于5 d后拔管患者（3.1% vs. 12.1%、9.2% vs. 20.3%、24.5% vs. 42.0%，均P<0.05）。单因素及多因素Logistic回归分析显示，术后第1天腹腔引流液淀粉酶水平（DFA1）与术后第1天血清淀粉酶水平（SA1）为术后早期胰瘘的独立影响因素（OR=1.000、1.004，95% CI=1.000~1.001、1.001~1.006，均P<0.05）；ROC分析结果显示，DFA1预测PD术后早期胰瘘发生的曲线下面积（AUC）为0.916，明显大于SA1（0.745），DFA1≥494.75 IU/L敏感度、特异度、阳性预测值、阴性预测值分别为91.7%、80.8%、62.7%、96.5%。
结论：DFA1是PD后早期胰瘘重要的影响与预测因素，对于DFA1<494.75 IU/L的患者，可于术后第3天安全拔除腹腔引流管，并对患者实施快速康复策略。
Analysis of influential and predictive factors for early postoperative pancreatic fistula after pancreaticoduodenectomy
CorrespondingAuthor:ZHANG Leida Email: 2518569931@qq.com
Abstract
Objective: To investigate the influential factors and prediction for early postoperative pancreatic fistula (POPF) after pancreaticoduodenectomy (PD), so as to guide the earlier removal of abdominal drainage tube to accelerate postoperative recovery of the patients.
Methods: The clinical data of 430 patients undergoing PD in Southwest Hospital from January 2013 to October 2015 were reviewed. The risk factors and their predictive values for early POPF were analyzed by univariate and multivariate statistical analysis, and receiver operating characteristic (ROC) curve, respectively.
Results: Early POPF occurred in 116 patients (26.9%) of the total 430 patients. In addition, among the patients without early POPF, the incidences of POPF, intra-abdominal infections and overall complications were all significantly reduced in those with abdominal drainage tube removed within postoperative day (POD) 5 compared with others with abdominal drainage tube retained more than POD 5 (3.1% vs. 12.1%; 9.2% vs. 20.3%; 24.5% vs. 42.0%, all P<0.05). The results of univariate and multivariate Logistic regression analysis showed that drain fluid amylase on POD 1 (DFA1) and serum amylase on POD 1(SA1) were independent risk factors for early POPF following PD (OR=1.000 and 1.004, 95% CI=1.000–1.001 and 1.001–1.006, both P<0.05). The results of ROC curve analysis showed that the area under the curve (AUC) of DFA1 for predicting early POPF was 0.916 which was obviously larger than that of SA1 (0.745), and the sensitivity, specificity, and the positive and negative predictive value were 91.7%, 80.8%, 62.7%, and 96.5% for DFA1≥494.75 IU/L, respectively.
Conclusion: DFA1 is an important risk and predictive factor for early POPF after PD. In patients with DFA1<494.75 IU/L, the abdominal drainage tube can be safely removed and then fast track recovery protocol can be adapted on POD 3.
Keywords：
Pancreaticoduodenectomy; Pancreatic Fistula; Amylases; Factor Analysis
Statistical
Methods: The clinical data of 430 patients undergoing PD in Southwest Hospital from January 2013 to October 2015 were reviewed. The risk factors and their predictive values for early POPF were analyzed by univariate and multivariate statistical analysis, and receiver operating characteristic (ROC) curve, respectively.
Results: Early POPF occurred in 116 patients (26.9%) of the total 430 patients. In addition, among the patients without early POPF, the incidences of POPF, intra-abdominal infections and overall complications were all significantly reduced in those with abdominal drainage tube removed within postoperative day (POD) 5 compared with others with abdominal drainage tube retained more than POD 5 (3.1% vs. 12.1%; 9.2% vs. 20.3%; 24.5% vs. 42.0%, all P<0.05). The results of univariate and multivariate Logistic regression analysis showed that drain fluid amylase on POD 1 (DFA1) and serum amylase on POD 1(SA1) were independent risk factors for early POPF following PD (OR=1.000 and 1.004, 95% CI=1.000–1.001 and 1.001–1.006, both P<0.05). The results of ROC curve analysis showed that the area under the curve (AUC) of DFA1 for predicting early POPF was 0.916 which was obviously larger than that of SA1 (0.745), and the sensitivity, specificity, and the positive and negative predictive value were 91.7%, 80.8%, 62.7%, and 96.5% for DFA1≥494.75 IU/L, respectively.
Conclusion: DFA1 is an important risk and predictive factor for early POPF after PD. In patients with DFA1<494.75 IU/L, the abdominal drainage tube can be safely removed and then fast track recovery protocol can be adapted on POD 3.