Vacuum assisted closure system in the management of postoperative complex enterocutaneous fistulae: Postoperative Cohort Study‎

INTRODUCTION

An enterocutaneous fistula (ECF) is defined as an abnormal connection between skin and intestine. A fistula is simple when it had a short track that communicates directly with the skin, or more complex, when it is associated with abscess formation or involvement of multiple loops of bowel with multiple openings on the skin.¹

The etiology of ECF can be iatrogenic in (75–85%) or spontaneous in (15-25%). Iatrogenic causes trauma, operations for malignancy associated with extensive adhesiolysis…etc. Spontaneous fistulae occur from IBD, malignancy, radiation …etc.²

There is controversy in the use of vacuum assisted closure (VAC) system in management of ECF. It had been contraindicated as it increases the risk of new fistula formation, but as time went by, VAC therapy revealed itself to be a “Swiss knife multi-tool”; a single device with multiple functions in ECF.³

Several studies of VAC reported successful closure of ECF but the main advantages of VAC were decreasing fistula effluent and enhancing healing of excoriated skin by keeping skin dry.⁴

The aim of the study is the evaluation of  the efficacy of the VAC therapy in treatment of the complex postoperative enterocutaneous fistula.

 PATIENT AND MATERIALS

This prospective present study was carried out at Al-Hussein University hospital and Al-Mokattam Health insurance hospital in the period from January 2018 to June 2019.

The patients enrolled in this study were those who had postoperative complex enterocutaneous fistulae, non-surgical candidates due to bad general conditions with no chronic debilitating diseases, age ranging from 20-60 years in both sexes and no evidence for the presence of distal obstruction, malignancy, IBD, diverticulitis, foreign body or visible intestinal mucosa while patients with postoperative simple fistula, pancreatic- or biliary atmospheric fistula, spontaneous fistulae or fistulae with exposed organs, intestinal loops or blood vessels, patients that had initial surgery for fistula repair, impaired renal function, hepatic function or hypoalbumenimia or patients on steroids, chemotherapy or radiotherapy were excluded from the study

After counseling, consenting and approval of ethical committee, all patients were subjected to complete history taking, full general and local examination and full laboratory and imaging investigations to identify origin, length and course of fistula, the presence of multiple openings or intervening abscess cavity:

After proper resuscitation, all patients had VAC therapy by the use of the VAC device manufactured by (KCI, San Antonio, TX) for their enterocutaneous fistulae with regular dressing changes  Figure (1).

Figure 1: Components of VAC system⁵

A drainage tube was located on top of the foam then a second piece of foam was placed on top of it or a Sensa TRAC pad was applied over a hole in the drape opposite to the fistula.

The foam, the first few inches of the drainage tube and the surrounding area of healthy skin were covered with an adhesive membrane to form a good seal. The drain was then connected to the VAC.

 Once it is switched on, the air is sucked out causing drawing the edges of the wound in. Fluid was taken out and transported to the disposable container.

VAC pressure was adjusted to 125 mmHg negative pressure with a cycle of 5 min suction and 2 min off suction. Higher negative pressure up to 200 mmHg was used in high output fistula to prevent leakage below the drape and detachment of the dressing and thus decompression of the VAC system. The dressing was changed at 48 hour intervals initially then the period was adjusted Figure (2).

Figure 2: Male patient with intestinal fistula at the level of the umbilicus originated from the jejunum after leakage from an eneterotomy that was made to extract a solid object that caused intestinal obstruction after its ingestion. VAC was applied after 3 days from the onset and the fistula took 18 days to close. The patient then sought plastic surgery for wound closure.

After one month, response to VAC therapy in terms of time from fistula VAC, general condition, the reduction in fistula output, the occurrence of fistula closure, the wound size, the skin condition, time to start oral feeding and the occurrence of complications or mortality was recorded. On follow-up, time to complete closure and type of closure either spontaneously or surgically were recorded. The cost of treatment was then recorded.

Data were collected and entered to the computer using SPSS (Statistical Package for Social Science) program version 26.0, SPSS Inc., Chicago, Illinois, USA for statistical analysis. Descriptive statistics was expressed as mean, SD and range in quantitative data and as frequency and percentage in qualitative analysis. Analytical statistics for qualitative data was done by using Chi-square test while Student t-test was used for quantitative data, Kaplan-Meier survival analysis and multiple logistic regression analysis for multivariate analysis. P (probability) value less than 0.05 was considered statistically significant.

RESULTS

Sixty patients had VAC for ECF. After the use of VAC, mortality occurred in 12 patients. 8 patients died of ongoing sepsis despite of VAC and 4 patients died from pulmonary embolism (Figure 3). According to mortality, patients were classified into two groups; survival group that consisted of 48 patients and mortality group that consisted of 12 patients.

Figure 3: Mortality after VAC therapy

In this study, 44 patients were male and 16 patients were female with age range 27-65 years. Smoking was found in 20 patients. Surgical history was found in 42 patients while family history of IBD was found in 4 patients. The presence of past surgical history and BMI were statistically different (p <0.001, 0.038 respectively). It was found that the amount of hemoglobin,platelets and the level of serum sodium and calcium was significantly higher in the survival group (p 0.001, <0.001, 0.033 and 2 cm) (p <0.001). The time that elapsed from the onset of fistula to VAC were different (p <0.001) Table (1).

Regarding the clinical outcome, no difference in terms of the time to start oral feeding but the improvements in general and perifistular skin condition, the reduction in fistula output and wound size, the need to segregate and the occurrence of complications were all significantly better in the survival group (p <0.001, <0.001, <0.001, <0.001, 0.023, 0.006 respectively) Figure (4).

Table 1: Demographic and postoperative data

Figure 4: Response to VAC in both groups

Figure 5: Complications after VAC therapy

40 patients (66.67%) had complications. They varied from pain in 16 patients, new fistula formation in 12 patients, development of pulmonary embolism in 4 patients and ongoing sepsis in 8 patients Figure (5).

The survival probability in this study was 83.6% after 30 days and 59.8% after 42 days Figure (6).

By multivariate analysis, the most important variable that affect the mortality was the amount of calcium (OR 61.46) followed by past surgical history (OR 36.22) and time from fistula to VAC (OR 10.99) (Table 2)

Figure 6: Kaplan-Meier Survival analysis

Table 2: Multivariate analysis

Regarding survived patients, spontaneous closure occurred in 32 patients while surgical intervention to close the fistula by resection/anastomosis of the bowel or plastic intervention to close the defect were required in 16 patients to accomplish complete closure. The mean time to complete closure was 32.25 ± 12.17 days while the mean cost of treatment in all included patients was 7540 ± 1001.46 Egyptian pounds Table (3).

Table 3: Outcome of VAC in survived patients