Abstract: Objective:To verify the accuracy of the mathematical model of gastrointestinal decompression after esophagectomy of esophageal cancer and explore predictive value of the mathematical model in the postoperative complications.
Methods:The retrospective casecontrol study was conducted. The clinicopathological data of 192 patients with esophageal cancer who underwent esophagectomy in the Beijing Chaoyang Hospital of Capital Medical University between October 2013 and October 2016 were collected. Among 192 patients, 160 didn′t have postoperative complications and 32 had postoperative complications (7 with postoperative anastomotic leakage, 9 with pulmonary infection and 16 with dysfunction of gastralintestinal tract). Patients selected the appropriate surgical procedures according to individual conditions, and then volume of gastrointestinal decompression was recorded daily. According to the regression equation of influencing factors of volume of postoperative gastrointestinal decompression: average daily drainage volume within 5 days (mL)=262.287+ 132.873×tubular stomach-72.160×smoking history-27.904×pathological type of tumor-36.368×age，predictive value of postoperative gastrointestinal decompression was calculated and compared with real volume of gastrointestinal decompression. Observation indicators: (1) comparison between predictive value and real volume of postoperative gastrointestinal decompression in patients without complications; (2) comparison between predictive value and real volume of postoperative gastrointestinal decompression in patients with complications. Measurement data with normal distribution were represented as ±s and comparison was analyzed using the paired samples t test. Measurement data with skewed distribution were described as M (range), and comparison was analyzed using the Wilcoxon signed rank tests.
Results:(1) Comparison between predictive value and real volume of postoperative gastrointestinal decompression in patients without complications: predictive value and real volume of postoperative gastrointestinal decompression in 160 patients without complications were respectively 187 mL (range, 58-392 mL) and 207 mL (range, 20-570 mL), with no statistically significant difference (Z=-1.106, P>0.05). (2) Comparison between predictive value and real volume of postoperative gastrointestinal decompression in patients with complications: 7 patients had postoperative anastomotic leakage, including 1 with cervical anastomotic leakage and 6 with chest anastomotic leakage. The predictive value and real volume of postoperative gastrointestinal decompression in 7 patients with anastomotic leakage were respectively (215±58)mL and (338±106)mL, with a statistically significant difference (t=-3.139, P<0.05). The predictive value and real volume of postoperative gastrointestinal decompression in 9 patients with postoperative pulmonary infection were respectively (176±61)mL and (239±111)mL, with no statistically significant difference (t=-1.805, P>0.05). The predictive value and real volume of postoperative gastrointestinal decompression in 16 patients with dysfunction of gastralintestinal tract were respectively (236±60)mL and (357±107)mL，with a statistically significant difference (t=-4.716, P<0.05).
Conclusions:The mathematical model of gastrointestinal decompression after esophagectomy of esophageal cancer is correct and feasible. There is a predictive value for patients with postoperative anastomotic leakage and dysfunction of gastralintestinal tract.