Investigation of treatment and analysis of prognostic risk on enterocutaneous fistula in China: a multicenter prospective study / 中华胃肠外科杂志

ABSTRACT

Objective@#To investigate the diagnosis and treatment for enterocutaneous fistula (ECF) in China, and to explore the prognostic factors of ECF.@*Methods@#A multi-center cross-sectional study was conducted based on the Registration System of Chinese Gastrointestinal Fistula and Intra-Abdominal Infections to collect the clinical data of ECF patients from 54 medical centers in 22 provinces/municipalities from January 1, 2018 to December 31, 2018. The clinical data included patient gender, age, length of hospital stay, intensive care unit (ICU) admission, underlying diseases, primary diseases, direct causes of ECF, location and type of ECF, complications, treatment and outcomes. All medical records were carefully filled in by the attending physicians, and then re-examined by more than two specialists. The diagnosis of ECF was based on the clinical manifestations, laboratory/imaging findings and intraoperative exploration.@*Results@#A total of 1521 patients with ECF were enrolled, including 1099 males and 422 females, with a median age of 55 years. The top three primary diseases of ECF were malignant tumors in 626 cases (41.2%, including 540 gastrointestinal tumors, accounting for 86.3% of malignant tumors), gastrointestinal ulcers and perforations in 202 cases (13.3%), and trauma in 157 cases (10.3%). The direct causes of ECF were mainly surgical operation in 1194 cases (78.5%), followed by trauma in 156 (10.3%), spontaneous fistula due to Crohn′s disease in 92 (6.0%), radiation intestinal injury in 41 (2.7%), severe pancreatitis in 20 (1.3%), endoscopic treatment in 13 (0.9%) and 5 cases (0.3%) of unknown reasons. All the patients were divided into three groups: 1350 cases (88.7%) with simple ECF, 150 (9.9%) with multiple ECF, and 21 (1.4%) with combined internal fistula. Among the patients with simple ECF, 438 cases (28.8%) were jejuno-ileal fistula, 313 (20.6%) colon fistula, 170 (11.2%) rectal fistula, 111 (7.3%) duodenal fistula, 76 (5.0%) ileocecal fistula, 65 (4.3%) ileocolic anastomotic fistula, 55 (3.6%) duodenal stump fistula, 36 (2.4%) gastrointestinal anastomotic fistula, 36 (2.4%) esophagogastric/esophagojejunal anastomotic fistula, 29 (1.9%) gastric fistula and 21 (1.4%) cholangiopancreatiointestinal. Among all the simple ECF patients, 991 were tubular fistula and 359 were labial fistula. A total of 1146 patients finished the treatment, of whom 1061 (92.6%) were healed (586 by surgery and 475 self-healing) and 85 (7.4%) died. A total of 1043 patients (91.0%) received nutritional support therapy, and 77 (6.7%) received fistuloclysis. Infectious source control procedures were applied to 1042 patients, including 711 (62.0%) with active lavage and drainage and 331 (28.9%) with passive drainage. Among them, 841 patients (73.4%) underwent minimally invasive procedures of infectious source control (replacement of drainage tube through sinus tract, puncture drainage, etc.), 201 (17.5%) underwent laparotomy drainage, while 104 (9.1%) did not undergo any drainage measures. A total of 610 patients (53.2%) received definitive operation, 24 patients died within postoperative 30-day with mortality of 3.9% (24/610), 69 (11.3%) developed surgical site infection (SSI), and 24 (3.9%) had a relapse of fistula. The highest cure rate was achieved in ileocecal fistula (100%), followed by rectal fistula (96.2%, 128/133) and duodenal stump fistula (95.7%,44/46). The highest mortality was found in combined internal fistula (3/12) and no death in ileocecal fistula. Univariate prognostic analysis showed that primary diseases as Crohn′s disease (χ2=6.570, P=0.010) and appendicitis/appendiceal abscess (P=0.012), intestinal fistula combining with internal fistula (χ2=5.460, P=0.019), multiple ECF (χ2=7.135, P=0.008), esophagogastric / esophagojejunal anastomotic fistula (χ2=9.501, P=0.002), ECF at ileocecal junction (P=0.012), non-drainage/passive drainage before the diagnosis of intestinal fistula (χ2=9.688, P=0.008), non-drainage/passive drainage after the diagnosis of intestinal fistula (χ2=9.711, P=0.008), complicating with multiple organ dysfunction syndrome (MODS) (χ2=179.699, P<0.001), sepsis (χ2=211.851, P<0.001), hemorrhage (χ2=85.300, P<0.001), pulmonary infection (χ2=60.096, P<0.001), catheter-associated infection (χ2=10.617, P=0.001) and malnutrition (χ2=21.199, P<0.001) were associated with mortality. Multivariate prognostic analysis cofirmed that sepsis (OR=7.103, 95%CI:3.694-13.657, P<0.001), complicating with MODS (OR=5.018, 95%CI:2.170-11.604, P<0.001), and hemorrhage (OR=4.703, 95%CI: 2.300-9.618, P<0.001) were independent risk factors of the death for ECF patients. Meanwhile, active lavage and drainage after the definite ECF diagnosis was the protective factor (OR=0.223, 95%CI: 0.067-0.745, P=0.015).@*Conclusions@#The overall mortality of ECF is still high. Surgical operation is the most common cause of ECF. Complications e.g. sepsis, MODS, hemorrhage, and catheter-associated infection, are the main causes of death. Active lavage and drainage is important to improve the prognosis of ECF.