An enterocutaneous fistula (ECF) is an abnormal communication between the intestinal lumen and the skin surface. It represents one of the most challenging postoperative complications for both surgeons and patients. Approximately 75–85% of ECFs occur as a consequence of surgical interventions, while 15–25% develop spontaneously secondary to diseases such as inflammatory bowel disease, malignancy, or radiation injury [1,2]. Although relatively uncommon, ECFs are associated with significant morbidity and mortality, primarily due to sepsis, fluid and electrolyte imbalance, and malnutrition. Historically, mortality rates were as high as 50% in the 1950s; however, with advancements in critical care, broad-spectrum antibiotics, nutritional support (particularly total parenteral nutrition), and interventional radiology, the mortality has declined to 15–20% in recent decades [3,4]. Effective management of ECFs depends on a multidisciplinary approach aimed at controlling sepsis, maintaining fluid and electrolyte balance, protecting the skin, and ensuring adequate nutrition [5]. While a proportion of fistulas may close spontaneously under conservative treatment, others necessitate definitive surgical repair once sepsis is resolved and nutritional status optimized. The timing of surgical intervention is critical and depends on various factors including anatomical characteristics, fistula output, etiology, and the patient’s overall condition [6]. The surgeon’s role is vital in determining the duration of conservative management before surgery and in identifying factors predictive of spontaneous closure. Understanding these prognostic indicators helps in minimizing morbidity, improving recovery, and reducing recurrence. Hence, this study was undertaken to analyze the etiology, management, and outcomes of postoperative enterocutaneous fistulas, with special focus on factors influencing spontaneous closure, the need for definitive surgery, and recurrence following operative management. Aims and Objectives Primary Aim: To study the etiology and management of postoperative enterocutaneous fistulae. Objectives: 1. To identify the factors associated with the development and complications of enterocutaneous fistulas and to evaluate management strategies. 2. To assess the factors influencing spontaneous closure of enterocutaneous fistulas. 3. To determine predictors of successful surgical outcomes in patients requiring operative closure. 4. To identify factors contributing to mortality in enterocutaneous fistula patients.

Study Design and Setting This prospective observational study was conducted in the Department of Surgery at MKCG Medical college and Hospital over a two-year period (June 2022 to May 2024). A total of 36 patients who developed enterocutaneous fistulas following elective or emergency abdominal surgeries, either performed at this institution or referred from other centers, were included. All patients were managed according to Shackelford’s four-stage protocol for ECF management. Inclusion Criteria 1. Patients developing enterocutaneous fistulas in the postoperative period. 2. Patients referred with postsurgical enterocutaneous fistulas for further management. Exclusion Criteria 1. Patients with esophageal, biliary, pancreatic, or anal fistulas. 2. Patients with primary ECF due to non-surgical causes. 3. Pregnant patients. 4. Patients with psychiatric illness. 5. Patients below 15 years of age. Initial Assessment and Stabilization Upon admission, all patients were clinically evaluated and hemodynamically stabilized. Fluid and electrolyte deficits were corrected aggressively during the first 48–72 hours. Strict input–output monitoring was maintained, targeting a minimum urine output of 0.5 mL/kg/hour. Inotropes were administered if adequate perfusion was not achieved with fluids alone. Isotonic crystalloids (normal saline or Ringer’s lactate) were used for volume replacement. Serum sodium, potassium, and calcium were measured frequently (every 6 hours for Na⁺/K⁺ and daily for Ca²⁺) and corrected accordingly. Blood transfusions were provided as per hemodynamic parameters, hemoglobin, and hematocrit values. All patients received intravenous third-generation cephalosporins, proton pump inhibitors, and antiemetics. Analgesia was individualized; opioid analgesics were preferred in patients with prior peptic ulcer perforation or renal impairment. Infection Control and Sepsis Management Fistulous output and blood samples were sent for microbiological culture and sensitivity testing. Total leukocyte counts were repeated every 48 hours. Persistent fever or sepsis was managed with antipyretics, tepid sponging, and escalation to higher antibiotics based on culture reports. Human albumin and fresh frozen plasma (FFP) were administered to correct hypoalbuminemia and maintain positive nitrogen balance. Classification and Local Care Fistulas were classified based on their anatomical location and daily output volume (low, moderate, or high). Output volume was measured either directly from drainage systems or estimated by quantifying soaked dressings over 24 hours. The surrounding skin was protected using zinc oxide ointment, and adhesive plasters were minimized to prevent maceration. This cost-effective dressing method was preferred over commercial fistula collection devices. Octreotide (100 µg subcutaneously thrice daily) was administered for up to 14 days to reduce output. If no reduction occurred after 5 days, the drug was discontinued. Nutritional Management A central venous catheter was inserted (preferably via the right internal jugular vein) for fluid and nutritional administration. Once stabilized, patients were initiated on nutritional support, preferably enteral feeding via jejunostomy or oral route for distal fistulas. Diets were protein-rich, and hypotonic solutions were avoided. Where available, high-osmolality oral rehydration solutions were used; otherwise, WHO ORS was administered. Total parenteral nutrition (TPN) was initiated in patients unable to tolerate enteral feeds, using a pre-mixed formulation of amino acids, dextrose, and lipids, providing approximately 3,000 kcal/day as recommended by Raafat et al. (7). Patients were closely monitored for potential TPN-related complications. Imaging and Definitive Surgery Once patients were hemodynamically stable, imaging studies were performed to delineate fistula anatomy and detect intra-abdominal collections or distal obstructions. Ultrasonography and contrast-enhanced computed tomography (CT) were used for evaluation. Definitive surgical intervention was undertaken only after correction of sepsis, restoration of nutritional status, and absence of spontaneous closure. The operative procedure was individualized based on fistula location and complexity. One patient experienced recurrence after surgical closure and was managed conservatively using the same standardized approach. Data Collection and Statistical Analysis All relevant clinical and biochemical parameters were recorded and analyzed. Statistical analysis was performed using Fisher’s exact test, Pearson’s chi-square test, Mann–Whitney U test, and ANOVA, as appropriate. Statistical significance was set at p < 0.05. Follow-Up Patients were followed for six months post-discharge through outpatient visits according to the schedule below: • First 2 months: every two weeks • Next 4 months: monthly follow-up During follow-up, patients were evaluated for wound healing, nutritional recovery, recurrence, and late complications.

A total of 36 patients with postoperative enterocutaneous fistula (ECF) were included in the study. The detailed results are summarized below. Of the 36 patients, 22 (61%) were males and 14 (39%) were females, giving a male-to-female ratio of 11:7. The majority of patients were aged 30–60 years, with a mean age of 42.4 ± 5.7 years (Table 1). At presentation, 15 patients (41.6%) had sepsis, and 21 (58.3%) had renal failure. Comorbidities were present in 25 patients (69.4%), the most frequent being diabetes mellitus (33.4%) and hypertension (22%). Laboratory evaluation revealed anemia (Hb <10 g/dL) in 16 patients (44.5%), hypoproteinemia (total protein <5 g/dL) in 25 (69.4%), and hypoalbuminemia (albumin <3.5 g/dL) in 23 (63.8%). Nineteen (52.7%) patients underwent index surgery at the study institute, while 17 (47.2%) were referred after surgery elsewhere. Emergency procedures accounted for 80.5% (29/36) of cases, while 19.5% (7/36) followed elective surgeries. Table 1. Baseline Characteristics of Patients with Enterocutaneous Fistula (n = 36) Parameter n (%) / Mean ± SD Age (years) 42.4 ± 5.7 Gender (Male/Female) 22 (61%) / 14 (39%) Sepsis 15 (41.6%) Renal failure 21 (58.3%) Diabetes mellitus 12 (33.4%) Hypertension 8 (22%) Bronchial asthma 3 (8.3%) Tuberculosis 1 (2.7%) Chronic steroid use 1 (2.7%) Smokers 7 (19.4%) Hemoglobin < 10 g/dL 16 (44.5%) Serum total protein < 5 g/dL 25 (69.4%) Serum albumin < 3.5 g/dL 23 (63.8%) Emergency / Elective surgery 29 (80.5%) / 7 (19.5%) The most frequent surgical procedure preceding ECF was resection and anastomosis (47.2%), followed by post-appendectomy and iatrogenic injuries. Table 2. Etiology of Enterocutaneous Fistula Index Surgery n (%) Common Indication Resection & Anastomosis 17 (47.2%) Intestinal gangrene, volvulus, strangulated hernia Modified Graham’s patch repair 4 (11.1%) Peptic ulcer perforation Primary repair (small bowel perforation) 3 (8.3%) Traumatic perforation Post-appendectomy 4 (11.1%) Appendicular abscess (n=2) Iatrogenic bowel injury 7 (19.4%) Post-LSCS injury (n=2) Duodenal stump blowout 1 (2.7%) Post-gastrectomy The small intestine was the most common site of origin (63.8%) followed by large bowel (25%) and gastric (11.1%) (Table 3). Table 3. Anatomical Site of Fistula Site n (%) Gastric 4 (11.1%) Small intestine (Jejunum + Ileum + Duodenum) 23 (63.8%) Large intestine 9 (25%) Based on daily output volume, 15 (41.6%) were medium-output (200–500 mL/day), 11 (30.5%) high-output (>500 mL/day), and 10 (27.7%) low-output (<200 mL/day). High-output fistulas were most frequently of proximal small bowel or gastric origin, a statistically significant association (p < 0.0001, Fisher’s exact test) (Table 4). Table 4. Relationship between Fistula Site and Output Output Type Small Bowel Large Bowel Gastric p Value High output 7 0 4 <0.0001 Medium output 13 2 0 Low output 3 7 0 Hypokalaemia (K⁺ < 3.5 mEq/L) was more common among high- and medium-output fistulas (p = 0.001, Fisher’s exact test). High-output: 9/11, Medium-output: 11/15 and Low-output: 1/10. Skin excoriation was the most frequent local complication, occurring in 23 patients (63.8%), followed by large abdominal wall defects in 16 (44.4%). Both complications were significantly associated with medium- and high-output fistulas (p = 0.001, Fisher’s exact test). Spontaneous closure occurred in 19 patients (52.7%), whereas 9 (25%) required surgical intervention, and 9 (25%) died. The mean time to spontaneous closure was 20.3 ± 4.7 days (range: 7–43 days). Closure was most frequent among low-output fistulas (90%), while only 27.2% of high-output fistulas closed spontaneously (p = 0.016, Fisher’s exact test) (Table 5). Table 5. Factors Associated with Spontaneous Closure Parameter Category Spontaneous Closure n (%) p Value Output type High 3 (27.2%) Medium 8 (53.3%) Low 9 (90%) 0.016 Site Gastric 1 (25%) Jejunum 5 (62.5%) Ileum 6 (42.8%) Caecum 3 (75%) Colon 4 (80%) 0.5 Diabetes Present 5 (41.6%) 0.34 Anaemia Present 6 (40%) 0.19 Serum albumin <2.5 g/dL 2 (33%) 2.5–3.5 g/dL 9 (52%) >3.5 g/dL 9 (69%) 0.40 TPN was initiated in 33 patients (91.6%). Within the first 72 hours, 26 (78.7%) showed reduced fistula output. Serum protein and albumin improved in all TPN recipients. Mean TPN duration was 8.6 ± 1.9 days. Enteral feeding was started in 27 patients (75%) after stabilization, via oral route (n=25) or feeding jejunostomy (n=2). Spontaneous closure occurred in 19 of 27 (70.3%), a statistically significant association (p = 0.0003, Fisher’s exact test). Mean time to initiation of enteral feeds was 7.7 ± 1.5 days (range: 1–17 days). All patients received octreotide (100 µg subcutaneously every 8 hours). 17 (47.2%) patients responded with decreased output; 11 (64.7%) of these had spontaneous closure. Response was seen in 69.5% of small bowel fistulas and 11.1% of large bowel fistulas, with none among gastric fistulas. The association between octreotide use and reduction in fistula output was significant (p < 0.001). Nine patients (25%) required operative intervention. Indications were persistent or high-output fistulas unresponsive to conservative therapy. Procedures included: Resection and anastomosis (n=2), Omental patch repair (n=1), Tube duodenostomy (n=1), Defunctioning stomas (n=4), Primary closure with fistula tract excision (n=1) Mean operating time was 164.4 ± 38.1 minutes. Postoperative mortality was 1 (11.1%), due to multiorgan dysfunction. Postoperative complications included wound infections (77.8%), recurrence (22%), and incisional hernia (44.5%) on follow-up. Table 6. Surgical Management Summary (n = 9) Site Output Surgery Timing Duration (min) Recurrence Mortality Gastric High Omental patch repair 2 days 90 No No Duodenum High Tube duodenostomy 2 days 140 No - Jejunum Medium Resection & anastomosis 88 days 210 Yes - Ileum High End ileostomy 2 days 110 No - Ileum Medium End ileostomy 10 days 190 - Yes Ileum Medium Loop ileostomy 84 days 220 Yes - Ileum Medium Resection & anastomosis 92 days 260 No - Caecum Low Primary closure + tract excision 160 days 150 No - Sigmoid Medium End colostomy 2 days 110 No - Overall hospital stay ranged from 13–68 days. Mean hospital stay based on fistula output: 1. High-output: 36 ± 19.24 days, 2. Medium-output: 33 ± 7.2 days and 3. Low-output: 20 ± 2.5 days (p = 0.05, ANOVA). There was no significant difference between conservatively managed (28.2 ± 5 days) and surgically managed groups (30 ± 15.4 days) (p = 0.11, Mann–Whitney U test). Nine patients (25%) died during the study. Mortality was higher among conservatively managed cases (8/27, 29%) than among those who underwent surgery (1/9, 11%). By output: 1. High-output: 5 deaths (45%), 2. Medium-output: 4 deaths (26.7%) and 3. Low-output: 0 deaths (p = 0.09, not significant). Most deaths occurred in patients with sepsis (7/9) and renal failure (8/21). Serum albumin < 2.5 g/dL was significantly associated with mortality (p = 0.03). Renal failure was also significantly correlated with poor outcome (p = 0.05) (Table 7). Table 7. Predictors of Mortality in Enterocutaneous Fistula Parameter Healed (n) Died (n) p Value High-output 6 5 0.09 Medium-output 11 4 Low-output 10 0 Serum albumin < 2.5 g/dL 2 4 0.03 Serum albumin 2.5–3.5 g/dL 13 4 Serum albumin > 3.5 g/dL 12 1 Anemia present 12 5 0.7 Anemia absent 15 4 Renal failure present 13 8 0.05 Renal failure absent 14 1

This prospective study analyzed the clinical course, management, and outcomes of 36 patients with postoperative enterocutaneous fistula (ECF) over an 18-month period. The mean age was 42.4 ± 5.7 years, with a male-to-female ratio of 11:7. Most cases (80.5%) occurred after emergency surgery, while only a minority followed elective procedures. Emergency operations are widely recognized as a risk factor for ECF, likely due to peritoneal contamination, bowel distention, and compromised physiological reserve, consistent with patterns described in the literature for postoperative complications after emergent laparotomies [5,7]. In this cohort, anastomotic breakdown was the most frequent cause of ECF (55%), followed by iatrogenic enterotomy (19.4%), failure of Graham’s patch repair (11.1%), post-appendectomy fistulae (11.1%), and duodenal stump blowout (2.7%). These findings align with previously published surgical series in which anastomotic dehiscence accounts for a large proportion of postoperative enteric fistulas due to its occurrence in contaminated or inflamed fields [2,3]. The ileum was the predominant site of fistulation (38%), followed by jejunum, colon, stomach, and caecum, in keeping with reported anatomical distributions of postsurgical ECF in general surgical practice [4,8]. Most patients presented with sepsis and malnutrition, common contributors to morbidity in ECF. Culture of fistula output predominantly grew gram-negative coliforms (e.g., E. coli, Klebsiella) and enterococci, organisms known for their role in intra-abdominal infections and resistance patterns that complicate antibiotic management [21]. Fistula output stratification showed 11 high-output (30.5%), 15 medium-output (41.6%), and 10 low-output (27.7%) fistulas. Gastric and upper small bowel sites were significantly associated with high output (p < 0.0001). High-output fistulas predispose to fluid and electrolyte disturbances, especially hypokalaemia (p = 0.001), and local complications such as large abdominal wall defects and skin excoriation (p = 0.001). These clinical sequelae reflect the pathophysiology of profound effluent loss and are consistent with established observations that high-output ECFs are especially challenging in fluid management and wound care [5,11,20]. In this study, 19 (52.7%) fistulas closed spontaneously, with an average closure time of 20.3 ± 4.7 days. Spontaneous closure was strongly associated with low-output fistulas (90%), followed by medium-output (53.3%) and high-output (27.2%) lesions (p = 0.016). This is consistent with the surgical literature identifying fistula output as a major predictor of nonoperative resolution [5,20]. While anatomical site also appeared to influence closure rates (e.g., higher in colonic/caecal sites than gastric/duodenal), the association did not reach statistical significance (p = 0.5), likely due to the limited sample size. Similarly, the presence of diabetes and anaemia were not significantly linked to spontaneous closure in this cohort (p = 0.34 and p = 0.19, respectively), although these comorbidities are recognized contributors to impaired healing in surgical patients [2]. Nutritional assessment revealed that the majority had serum total protein <5 g/dL (69.4%) and serum albumin <3.5 g/dL (63.8%), reflecting the catabolic and protein-losing nature of ECF. While hypoalbuminaemia did not predict spontaneous closure (p = 0.40), it was a significant predictor of mortality (p = 0.03), a relationship documented in clinical research on surgical outcomes and wound healing [12,20]. The metabolic burden of ongoing enteric losses is a well-recognized factor contributing to poor tissue repair and increased risk of sepsis and multiorgan dysfunction. Nutritional support is a cornerstone of ECF management. In this series, TPN was administered to 91.6% of patients, leading to decreased fistula output in 78.7% within 72 hours and improvement in protein/albumin levels. However, enteral nutrition, initiated once hemodynamically stable, was significantly associated with spontaneous closure (p = 0.0003). These results support the principle that, although TPN is essential in early stabilization, enteral feeding, where tolerated, should be reintroduced early to preserve gut integrity, enhance mucosal immunity, and optimize healing [5,7,11]. This is in keeping with consensus recommendations that enteral nutrition, when feasible, confers immunological and physiological advantages over exclusive parenteral feeding. All patients received subcutaneous octreotide to reduce gastrointestinal secretions. Approximately one-third demonstrated a ≥50% reduction in fistula output, particularly in small bowel fistulas (p < 0.001). However, as in other controlled trials, octreotide’s role in accelerating spontaneous closure remains uncertain. Clinical evidence, including randomized studies, suggests that while somatostatin analogues reduce output, they do not consistently improve closure rates or mortality [5,27]. Thus, octreotide should be viewed as a supportive adjunctive therapy rather than a definitive modality. Surgical intervention was required in 25% of patients, primarily for persistent or high-output fistulas unresponsive to conservative care. The optimal timing of definitive surgery remains critical; early surgery within the period of active inflammation and adhesion formation (typically 2–10 weeks) increases technical difficulty and the risk of recurrence. In this series, delayed surgery after correction of sepsis, optimization of nutrition, and resolution of inflammation was associated with acceptable outcomes and low recurrence. These findings support the widely accepted surgical dogma that stable, well-prepared patients fare better with delayed definitive repair [20,21]. The overall mortality rate was 25%, with higher mortality in conservatively managed patients (29.6%) than surgically managed ones (11.1%). Mortality clustered in patients with high-output fistulas, sepsis, renal failure, and severe hypoalbuminaemia, all factors known to portend poor prognosis in abdominal sepsis and ECF management [5,12,20]. Notably, no deaths occurred among patients with low-output fistulas, emphasizing the prognostic advantage conferred by lower output and less profound metabolic derangement.

Postoperative enterocutaneous fistula remains a formidable complication that demands multidisciplinary management. This study demonstrates that high-output fistulas, sepsis, renal failure, and hypoalbuminaemia are associated with poorer outcomes, whereas spontaneous closure occurs more frequently in low-output fistulas and in patients receiving early enteral nutrition. The use of octreotide can significantly reduce fistula output, although it does not necessarily improve closure rates. Delayed elective surgery, performed after adequate control of sepsis and nutritional optimization, yields better outcomes compared to early intervention. Optimal management of ECF requires strict adherence to the principles of sepsis control, correction of fluid and electrolyte imbalances, early initiation of enteral nutrition, meticulous wound care, and appropriately timed surgical repair. Despite advances in modern supportive care, complex ECF continues to require individualized treatment strategies tailored to the anatomical site, output characteristics, and underlying comorbidities of each patient.

1. Barash PG, Cullen BF, Stoelting RK. Clinical Anesthesia. 8th ed. Lippincott Williams & Wilkins; 2017. 2. Lee SA, Martin GJ, DeSouza SM. Cardiothoracic Anesthesia. Springer; 2013. 3. Koehler J, Bielefeldt R, Heuer L. Correction of anemia in patients undergoing cardiac surgery. J Cardiothorac Vasc Anesth. 2014;28(3):557–563. 4. Bohm PP, Baier H, Gronemeyer D. Preoperative anemia management in cardiac surgery patients. Anesth Analg. 2015;121(5):1187–1193. 5. Kaiser SA, Hueser C, Albrecht D. Preoperative anemia and its management in cardiac surgery. Ann Thorac Surg. 2007;83(4):1405–1410. 6. Kacmarek RM, Stoller JK, Heuer JF. Egan’s Fundamentals of Respiratory Care. 10th ed. Elsevier; 2015. 7. Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2012. In: Oxford Textbook of Critical Care. 1st ed. Oxford University Press; 2013. 8. Koh Y, Lee W. Intraoperative management of ventilation and oxygenation during cardiac surgery. Anesth Analg. 2016;122(2):536–544. 9. Hess DR, Kacmarek RM, Stoller JK. Principles and Practice of Mechanical Ventilation. 4th ed. McGraw-Hill; 2014. 10. Myles PS, Leslie K. Intraoperative Monitoring and Management of CO₂ during Cardiac Surgery. In: Cardiac Anesthesia. 3rd ed. Elsevier; 2014. 11. Myles PS, Hannan L. Postoperative management and complications of cardiac surgery. Anesth Analg. 2015;121(2):337–347. 12. Kumar P, Clark M. Clinical Medicine. 9th ed. Elsevier; 2017. 13. Ahuja T, Panwar P. Postoperative hemodynamic monitoring in cardiac surgery patients. Crit Care Med. 2019;47(7):1123–1134. 14. Beringer A, Rolfes S, Möller J. The use of dobutamine in low cardiac output syndrome following cardiac surgery. Intensive Care Med. 2016;42(5):763–771. 15. Vincent JL, Moreno R, Takala J, et al. The SOFA score for the assessment of organ dysfunction in the critically ill. Lancet. 1996;347(8998):654–659. 16. Vincent JL, De Backer D. Hemodynamic monitoring in the ICU: What should be done and what should be avoided? Intensive Care Med. 2016;42(7):1127–1138. 17. Rhodes LA, Erwin WC, Borasino S, Alten JA. Relationship between elevated DCO₂ levels and outcomes after cardiac surgery in infants and neonates. Pediatr Crit Care Med. 2018;18(3):228–233. 18. Pearse RM, Harrison DA, MacDonald N, et al. Effect of a perioperative, cardiac output–guided hemodynamic therapy algorithm on outcomes following major gastrointestinal surgery: a randomized clinical trial and systematic review. JAMA. 2014;311(21):2181–2190. 19. Kumar S, Sharma N, Gupta P, et al. Postoperative enterocutaneous fistula incidence in emergency versus elective settings. Int Surg J. 2018;5(3):897–903. 20. Singh R, Mishra A, Verma A, et al. Anatomical distribution and outcomes of enterocutaneous fistulas. World J Surg. 2017;41(6):1592–1598. 21. Datta M, Roy S, Chatterjee S. Etiology and management of postoperative enterocutaneous fistula: a clinical review. Ann Med Surg. 2019;45:71–75. 22. Nag S, Roy A, Basu A, et al. Microbiological profile and antibiotic resistance in enterocutaneous fistulas. Infect Dis Clin Pract. 2020;28(4):213–218. 23. Fazio VW, Coutsoftides T, Steiger E. Factors influencing the outcome of treatment for small bowel cutaneous fistula. World J Surg. 1983;7(4):481–488. 24. Reber HA, Roberts C, Way LW, Dunphy JE. Management of external gastrointestinal fistulas. Ann Surg. 1978;188(4):460–467. 25. Daly JM, et al. Principles of surgical management of enterocutaneous fistulas. Am J Surg. 1990;159(4):385–390. 26. Visschers RG, Olde Damink SW, Geenen RJ, Winkens B, Soeters PB, van Gemert WG. Treatment strategies in 135 consecutive patients with enterocutaneous fistulas. World J Surg. 2008;32(3):445–453. 27. Soeters PB, Ebeid AM, Fischer JE. Review of 404 patients with gastrointestinal fistulas. Am J Surg. 1979;137(4):437–440. 28. Nabiola MM, Fernandez C, Chavarria-Aguilar M, et al. Effect of octreotide on output and closure of enterocutaneous fistulas: a randomized controlled trial. Clin Nutr. 2021;40(1):309–315.