Investigation of Epidemiology and Clinical Characteristics of Patients with Enterocutaneous Fistula in China

Introduction

A fistula is an abnormal communication between two epithelial surfaces. It often starts from an offending side and makes its way to an adjacent surface. A digestive tract fistula can originate anywhere in the digestive tract, such as the mouth, esophagus, stomach, duodenum, small intestine, colon, rectum, or anal canal [1]. Digestive tract fistulas include external and internal fistulas. External fistulas penetrate the abdominal wall and communicate with the outside world, and internal fistulas connect with other cavity organs. However, internal and external fistulas can also co- exist [2].

An enterocutaneous fistula (ECF) is one of the most common and serious fistulas. It refers to an abnormal passage between the gastrointestinal tract and other hollow organs, body cavities, or outside the body, through which gastrointestinal contents enter other organs, body cavities, or outside the body, causing a series of pathophysiological changes, such as infection, fluid loss, malnutrition, and organ dysfunction. Enterocutaneous, enterenteric, enterovesical, enterocolic, enteroatmospheric, and rectovaginal fistulas are the most frequent fistulas [3, 4, 5]. ECFs are usually caused by underlying diseases and surgical procedures [6]. Therefore, they are considered a complication rather than a distinct disease [7, 8, 9]. Surgical complications are the most common cause of intestinal fistula formation [10, 11]. Except for surgical events, the common causes of ECFs are diverticular disease, Crohn’s disease, malignancy, radiation, non-surgical injuries, and foreign bodies [10, 12, 13, 14, 15, 16, 17]. A fistula’s etiology should be confirmed before planning treatment. It is important to define the cause of an ECF to develop a treatment plan for it. Treatment of an ECF involves treatment of the fistula itself and its underlying disease. Although the outcomes of therapeutic treatments have improved significantly, in general, the mortality of patients with ECFs remains high. The purpose of this study was to understand the clinical characteristics and treatment of ECFs in our hospital and to identify high-risk factors influencing the prognosis of patients with ECFs, to improve their treatment outcomes.

Statistical Analysis

The statistical software package SPSS (version 20.0; IBM, Armonk, NY, USA) was used for all analyses. The association between the ECF cure rate and the first visited hospital, definitive operation, malnutrition, and other clinical characteristics was analyzed using the Chi-square test. Student’s t-test was used to evaluate significant differences between two groups of data in all of the relevant analyses. The level of statistical significance was set to p < 0.05.

Discussion

This study analyzed the general demographic characteristics of patients with ECFs. Studies have found that the proportion of male patients among all patients with ECFs is 48.1-72.3% [2, 18, 19, 20, 21]. In this study, approximately 69.7% of the patients with ECFs were male. No significant differences were found in comparisons with the female patients. The median age in this study was 52 years (range: 7-86 years), which was similar to results of other studies [19, 20, 21]. However, we included a 7-year-old ECF patient in our study, which shows that ECF can also occur in children. Approximately 40% of the patients in this study were from Changsha, and the rest were from other cities or regions in China. At the same time, 84.4% of patients were referred from other hospitals throughout the country. More importantly, since the first case of ECF was admitted to our department in 2013, the number of admitted patients has increased annually, reaching 65 cases in 2019.

Our study showed that 79.8% of the patients had a single ECF, which is between the percentages reported in other studies [2, 20]. Moreover, our results showed the main direct cause of ECF was surgical treatment (92.2%), which was higher than other causes [2, 21]. Previous research reported the most common causes of ECFs were malignant tumors (41.2%), gastroenteric ulcers and perforations (13.3%), and trauma (10.3%) [2]. However, our research found that the most common causes were gastroenteric tumors (25.7%), internal obstruction and abdominal infection (14.2%), and abdominal trauma (14.2%). So, it is clear that abdominal tumors and trauma are the main causes of ECFs. Furthermore, a study by Zheng found that the most common types of ECFs were the jejuno-ileal fistula (28.8%), colon fistula (20.6%), and rectal fistula (11.2%) [2]. However, the most common types found in our study were small intestine fistula (44.0%), duodenum fistula (20.2%), and colon fistula (18.8%). There are two reasons for this difference: First, there are different standards for classification of ECFs. In this study, we recorded fistulas between the stomach and the small intestine as gastric fistulas, fistulas between the small intestine and colon as intestinal fistulas, and fistulas between the colon and rectum as colon fistulas. Second, the study’s sample size was small.

As reported in this study, 84.4% of patients were referred from other hospitals throughout the country, 43.1% had visited a secondary care hospital first, and 51.8% had visited a tertiary care hospital first. There were no significant differences between the two groups selected. However, we found that the level of care of the hospital that patients visited first was associated with the cure rate (p = 0.039). This finding can be attributed to two reasons: (1) the patients who had visited a tertiary care hospital first might have done so because their conditions were more serious than those of the other patients; and (2), it is possible that the lymphadenectomies performed in the tertiary care hospitals were more extensive and prone to serious and fatal complications, than those performed in the secondary care hospitals.

In our study, 22.5% of patients received a definitive operation, which was lower than other studies (23.4- 53.2%) [2, 21]. Moreover, 81.6% of patients had received the operation one week after the diagnosis of the fistula, which is consistent with the results of Visschers and Zheng [18, 19]. More importantly, the healing rate was 57.3% for the patients in this study, which is consistent with the results of Quinn (53.8%) [21]. We found that the cure rate was related to having the definitive surgery after the diagnosis of the fistula (p = 0.024) (Table 1). However, surgery is no longer the first choice for patients with intestinal fistulas, given the widespread use of somatostatin, growth hormone, enteral and parenteral nutrition, and negative pressure suction. In addition, proximal enterostomy has improved the healing rate of patients with ECF. Though studies have reported an in-hospital mortality rate of 19.1% and an operative mortality from definitive surgery of 9.8-15.1%, [19, 21] we were unable to assess the mortality rate of patients with ECFs in the present study. First, some patients were discharged or transferred to another hospital before the fistula healed. Second, because of the long treatment period and high costs, it was difficult to conduct outpatient follow-up care. However, one year after treatment for an appendiceal stump fistula in December 2019, a 16-year-old girl suffered from severe abdominal infection, which resulted in rupture of the splenic artery and bleeding, leading to her death. Wu L found that some patients had intra-abdominal bleeding events during hospitalization, and the incidence of intra-abdominal bleeding was 10.6% [22].

Figure 4A shows the time from the operation to the diagnosis of ECF. The time for 40.0% of the patients was within 1 week after the initial operation, and that the time for 28.9% of patients was 8 to 31 days after the initial operation. Martinez and Visschers reported that the median time from the initial surgery to fistula recurrence was 6-21 days (range: 1-494 days) [19, 20]. In our study, approximately 68.9% of patients with ECF received a clear-cut diagnosis within 31 days after the first operation. However, due to inconsistencies in stratification standards, we could not compare our results with the results of Martinez and Visschers [19, 20]. Our study found that the length of stay for the first hospitalization of 90.8% of the patients was one week to half a year. Visschers found that the median period of treatment was 70 days (range: 6-497 days) [19]. However, it was not possible for us to measure the precise length of total treatment for those patients with ECF because some of them received treatment at home.

This study identified patients with complications of diabetes (10.1%), hypertension (11.5%), hypoproteinemia (11.0%), and other common diseases. More importantly, we found a strong relationship between preoperative albumin levels and the cure rate, which is consistent with the results of Visschers [19]. However, no relationship was found between the cure rate and diabetes, hypertension, hypoproteinemia, or anemia. Furthermore, in our study, only 3.2% of the patients had Crohn’s disease, whereas 6-8% of patients in other studies had Crohn’s disease [2, 21, 22].

Conclusion

In our study, we had investigated the epidemiology and clinical characteristics of 218 patients with enterocutaneous fistulas. We found that the cure rates of enterocutaneous fistulas were associated with the first visited hospital, definitive surgical treatment, and malnutrition.