A total of 10 patients received diverting loop ileostomies (Table 2). A subanalysis excluding these patients resulted in nearly identical findings with regard to LOS and incidence of postoperative complications. Stoma creation was significantly more likely in patients who underwent dual resections (both ileocolic and sigmoid) as compared with those who underwent ileocolic resections with sigmoid primary repairs (p = 0.017). In addition, 7 of the 10 patients who underwent a stoma creation were diagnosed with an ISF preoperatively; however, knowing the diagnosis preoperatively did not significantly correlate with the decision to divert.
Descriptions of surgical management of ISFs dates back to 1974, when surgical approaches to ISFs were first described.6 The first patient series of 63 patients was drawn from a single-institution experience at Mount Sinai Hospital. In that article, the authors posited that the colon may be the origin of the fistula, accounting for why some patients get enterocolonic fistulas, versus others who experienced isolated enteritis or entero-enteric fistulas. With improvements in radiologic imaging and closer examination of pathologic specimens, we now know the colon to most often be an innocent bystander.19
In 1982, a series of 17 patients with CD with ISFs who underwent operations revealed that 13 of these patients were accurately diagnosed on radiologic imaging preoperatively.8 Using imaging (rather than intraoperative assessment), the authors concluded that all of the patients in this series did not have active Crohn’s colitis of the sigmoid. This series was the first to propose primary repair of the colonic end of the fistulous tract as an alternative to colonic resection.
In 1997, a retrospective cohort of 90 surgical cases of ISF repair found that patients were significantly more likely to undergo a sigmoid resection (as opposed to primary repair) if the ISF was found on preoperative radiology studies. Our study population (all of whom had a CT, a colonoscopy, or both) had a preoperative finding of an ISF diagnosed in 70% of patients (59 of 84 patients), which is consistent with other published reports in the recent past.11,12,19 Our study, however, did not identify a significant correlation between knowing the ISF diagnosis preoperatively and the procedure performed for repair of the sigmoid. A total of 25 (42%) of the 59 patients with an ISF diagnosed preoperatively underwent a sigmoid resection versus 8 (32%) of the 25 patients who did not receive an ISF diagnosis until they were in the operating room (p = 0.38). It seems unlikely that knowing the diagnosis preoperatively influenced the surgeon’s decision-making regarding the sigmoid repair technique.
Although only borderline statistically significant, this same study from 1997 also found a small difference in LOS by sigmoid repair type (8.3 d in the primary repair group versus 9.9 d in the resection group; p = 0.15). In addition, 7 patients in this group (8%) were found to have coincident colovesicular fistulas. Our study omitted patients with colovesicular fistulas and found a statistically significant increase in the LOS for patients who underwent double resections.
Two previous studies have attempted to examine CD-specific risk factors for conversion from laparoscopy to an open surgical approach,20,21 and although both of these studies showed increased conversion rates of 9% in cases of complex entero-enteric fistulas, as compared to 3% seen overall, neither study concluded that ISFs are a contraindication to attempted laparoscopy.22
A 2009 study at our institution examined the safety of laparoscopy in 335 laparoscopic bowel resections for CD dating back to 1993. Of these, 117 were found to have entero-enteric fistulas, with 51 found to have ISFs. It was determined that internal fistulizing disease could be managed by experienced laparoscopic surgeons safely, with acceptable morbidity.23 (Although our study looks at cases performed by surgeons included in this study, the cases are unique and were drawn from a later time period.) Subsequent studies further validated these findings.15,16,22
In our study, we found that a preoperative diagnosis of an ISF on imaging was significantly correlated with converting to an open operation and with the creation of a diverting stoma. Interestingly, only 1 patient without a preoperative diagnosis of an ISF underwent laparoscopy requiring a conversion to an open operation. (Of note, there was no significant difference in the presence of active sigmoid disease observed in the operating room between those patients with a preoperative diagnosis of an ISF versus those without.) This may indicate that ISFs detectable on imaging or colonoscopy are reflective of more severe disease of the terminal ileum or that knowing about an ISF preoperatively may have changed surgical decision-making in a way that is not fully understood.
Moreover, our finding that LOS was not significantly different for the laparoscopic versus the open operation group should be thoughtfully considered. The value of laparoscopy is not limited to LOS. Additional benefits, including decreased adhesions and the lower likelihood of subsequent incisional hernia, should still be considered when deciding on a surgical approach. In addition, given the small number of procedures in our series performed via a laparotomy, the lack of a significant difference in LOS or postoperative complications in our study may be attributed to limited statistical power.
In one other series of 104 ISFs from 2000 to 2007, >37% of these patients had coincident internal fistulizing disease requiring additional surgical resections. The authors found that postoperative complications did not differ by surgical approach (laparoscopic versus open) or by sigmoid repair type (primary repair versus resection).12 This series, however, likely experienced significant confounding, given the inclusion of patients with additional internal fistulizing disease. In this series, only 3 of 10 surgeons included performed laparoscopy. In our series, all 9 of the surgeons performed or attempted laparoscopic ISF repair. Only 19% of cases were completed laparoscopically versus 60% in our series, and 51% of the patients in this previous series received a diverting/protective stoma versus 12% in our series. These dramatic differences in surgical approach may reflect the multiple internal fistulas seen in many of these patients (37%) in the former series, indicating that inclusion of synchronous fistulas in a case series intended to analyze ISFs may have introduced significant confounding.
Our study of 84 patients with isolated ISFs contributes to the literature on the surgical management of CD by examining differences in postoperative complications by surgical approach and sigmoid repair type for patients with isolated ISFs and by adjusting for the effect of each surgical variable on the other in a multivariate analysis. These patients experienced no significant differences in the incidence of postoperative complications by operative approach (laparoscopic or open), nor by sigmoid repair type (primary repair versus en bloc resection). Nor did they have a significantly different LOS depending on whether they had an open or a laparoscopic operation. Patients in the series did, however, experience a significantly longer LOS when their sigmoid colon was resected rather than primarily repaired.
Given these findings, we advocate for performing a sigmoid primary repair in ISF cases, when feasible. This decision should be based on the surgeon’s ability to differentiate between diseased versus secondarily involved loops of bowel. If the sigmoid appears diseased, it should generally be resected along with the ileocolic segment. If, however, the sigmoid colon appears to be an innocent bystander, we would advocate for a primary repair, using either a handsewn or stapled primary closure. It should be noted, however, that in cases where the sigmoid does not appear diseased but there is either a large defect or a large abscess (particularly when on the mesenteric side of the bowel), the surgeon may appropriately decide to perform a segmental sigmoid resection. With regard to the decision to perform a diversion, a recent study showed a strong correlation between previous resections for CD and the risk for clinical anastomotic leak,24 suggesting that previous resection should be one variable taken under consideration. The decision, however, should be left up to the surgeon and made in the context of other factors, including patient nutritional status and intraoperative assessments of tissue integrity (Fig. 3).
Our analysis has several limitations that are important to consider. Given the cross-sectional and observational nature of the data, it is not possible to draw conclusions about causation. We are able to deduce associations, but hypotheses remain for why these differences exist. Specifically, questions about surgical decision-making remain unanswered–does preoperative knowledge of an ISF change the surgeon’s threshold for converting from laparoscopy to laparotomy? How does a surgeon decide when a diverting stoma ought to be formed for these patients?
In addition, given the nature of physician and surgeon practice patterns and endoscopic mucosal resection use at Mount Sinai Hospital, the data on a patient’s preoperative disease severity and the degree of immunocompromise are limited. Although we were able to calculate HBI, this measure has been shown to have variable validity when compared with endoscopic findings of disease severity.18,25 As a proxy measure, we determined whether patients were treated with steroids postoperatively; however, this gives an incomplete picture of the severity of a patient’s disease and his or her perioperative substrate.
Our conclusion from these findings is not that they should direct operative management but that they may provide valuable information for counseling patients both before surgery and during their hospital stay. Operative management of the sigmoid colon should be left up to the discretion of the surgeon. In the event that a surgeon must perform en bloc resection of the ISF (ie, both ileocolic and sigmoid resection), that patient can anticipate a longer hospital stay (and recovery) but no increased risk of postoperative complications. In addition, if an en bloc resection is planned preoperatively, the surgeon may wish to also perform a diverting stoma, something that the patient should be counseled on preoperatively, when possible.
Laparoscopic operations for ISF in CD are safe and do not result in an increased incidence of postoperative complications, as compared with open surgeries. Moreover, primary repair of the sigmoid colon when feasible appears safe, with no increase in postoperative complications relative to sigmoid resections and is likely to be cost-effective, given the reduced LOS.
The authors thank Jill Gregory for her illustrations.
1. Fichera A. Crohn’s disease
: how modern is the management of fistulizing disease? Nat Rev Gastroenterol Hepatol. 2009;6:511–512.
2. Levy C, Tremaine WJ. Management of internal fistulas in Crohn’s disease
. Inflamm Bowel Dis. 2002;8:106–111.
3. Strong S, Steele SR, Boutrous M, et al; Clinical Practice Guidelines Committee of the American Society of Colon and Rectal Surgeons. Clinical practice guideline for the surgical management of Crohn’s disease
. Dis Colon Rectum. 2015;58:1021–1036.
4. Fazio VW, Wilk P, Turnbull RB Jr, Jagelman DG. The dilemma of Crohn’s disease
: ileosigmoidal fistula complicating Crohn’s disease
. Dis Colon Rectum. 1977;20:381–386.
5. Korelitz BI, Kesar V, Taunk R, Schneider J. Sequential Crohn’s ileitis, ileosigmoidal fistula, segmental sigmoid polyposis, and sigmoid stricture: the natural history. J Clin Gastroenterol. 2017;51:607–610.
6. Greenstein AJ, Kark AE, Dreiling DA. Crohn’s disease
of the colon: I–fistula in Crohn’s disease
of the colon, classification presenting features and management in 63 patients. Am J Gastroenterol. 1974;62:419–429.
7. Heimann T, Greenstein AJ, Aufses AH Jr.. Surgical management of ileosigmoid fistula in Crohn’s disease
. Am J Gastroenterol. 1979;72:21–24.
8. Broe PJ, Cameron JL. Surgical management of ileosigmoid fistulas
in Crohn’s disease
. Am J Surg. 1982;143:611–613.
9. van Dongen LM, Lubbers EJ. Surgical management of ileosigmoid fistulas
in Crohn’s disease
. Surg Gynecol Obstet. 1984;159:325–327.
10. Bezzi M, Pastore P, Bolignari S, et al. Surgical management of ileosigmoid fistulas
in Crohn’s disease
. Ital J Surg Sci. 1988;18:131–135.
11. Young-Fadok TM, Wolff BG, Meagher A, Benn PL, Dozois RR. Surgical management of ileosigmoid fistulas
in Crohn’s disease
. Dis Colon Rectum. 1997;40:558–561.
12. Melton GB, Stocchi L, Wick EC, Appau KA, Fazio VW. Contemporary surgical management for ileosigmoid fistulas
in Crohn’s disease
. J Gastrointest Surg. 2009;13:839–845.
13. Block GE, Schraut WH. The operative treatment of Crohn’s enteritis complicated by ileosigmoid fistula. Ann Surg. 1982;196:356–360.
14. Dasari BV, McKay D, Gardiner K. Laparoscopic versus open surgery for small bowel Crohn’s disease
. Cochrane Database Syst Rev. 2011;(1):CD006956.
15. Neumann PA, Rijcken EJ, Bruewer M. Current status of laparoscopic surgery for patients with Crohn’s disease
. Int J Colorectal Dis. 2013;28:599–610.
16. Rosman AS, Melis M, Fichera A. Metaanalysis of trials comparing laparoscopic and open surgery for Crohn’s disease
. Surg Endosc. 2005;19:1549–1555.
17. Harvey RF, Bradshaw JM. A simple index of Crohn’s-disease activity. Lancet. 1980;1:514.
18. Zittan E, Kabakchiev B, Kelly OB, et al. Development of the Harvey-Bradshaw Index-pro (HBI-PRO) score to assess endoscopic disease activity in Crohn’s disease
. J Crohns Colitis. 2017;11:543–548.
19. Maconi G, Sampietro GM, Parente F, et al. Contrast radiology, computed tomography and ultrasonography in detecting internal fistulas and intra-abdominal abscesses in Crohn’s disease
: a prospective comparative study. Am J Gastroenterol. 2003;98:1545–1555.
20. Schmidt CM, Talamini MA, Kaufman HS, Lilliemoe KD, Learn P, Bayless T. Laparoscopic surgery for Crohn’s disease
: reasons for conversion. Ann Surg. 2001;233:733–739.
21. Pokala N, Delaney CP, Brady KM, Senagore AJ. Elective laparoscopic surgery for benign internal enteric fistulas: a review of 43 cases. Surg Endosc. 2005;19:222–225.
22. Beyer-Berjot L, Mancini J, Bege T, et al. Laparoscopic approach is feasible in Crohn’s complex enterovisceral fistulas: a case-match review. Dis Colon Rectum. 2013;56:191–197.
23. Nguyen SQ, Teitelbaum E, Sabnis AA, Bonaccorso A, Tabrizian P, Salky B. Laparoscopic resection for Crohn’s disease
: an experience with 335 cases. Surg Endosc. 2009;23:2380–2384.
24. Johnston WF, Stafford C, Francone TD, et al. What is the risk of anastomotic leak after repeat intestinal resection in patients with Crohn’s disease
? Dis Colon Rectum. 2017;60:1299–1306.
25. Vermeire S, Schreiber S, Sandborn WJ, Dubois C, Rutgeerts P. Correlation between the Crohn’s disease
activity and Harvey-Bradshaw indices in assessing Crohn’s disease
severity. Clin Gastroenterol Hepatol. 2010;8:357–363.
Crohn’s disease; Ileosigmoid fistulas; Surgical technique
Supplemental Digital Content
© 2019 The American Society of Colon and Rectal Surgeons