Type of fistula determines response to infliximab in patients with fistulous Crohn's disease.

OBJECTIVES

Infliximab has been shown to be efficacious for treating perianal fistulae in patients with Crohn's disease. There is limited information regarding response to infliximab in patients with other types of fistulae.

METHODS

Sixty consecutive patients with fistulous Crohn's disease and at least three months of follow-up after three infliximab infusions were evaluated. Patients with enteroenteric fistulae were excluded. Complete response was defined as complete closure of the fistulae or complete cessation of fistula drainage.

RESULTS

Thirty-five patients had external fistulae, 16 had internal fistulae, and 9 had mixed (both external and internal) fistulae. Complete response rates were significantly higher in patients with external fistulae (69%) compared to those with internal fistulae (13%); p= 0.001, or those with mixed fistulae (11%); p= 0.01. In the external fistula group, patients with perianal fistulae had a higher rate of complete response (78%) compared to those with abdominal wall fistulae (38%); p= 0.04. The rate of complete response to infliximab was significantly lower among 14 patients with rectovaginal fistulae (14%) compared to those with perianal fistulae (78%); p= 0.0007. In the mixed fistula group only 11% of the patients achieved complete response. This is significantly lower than the rate observed for patients with perianal fistulae (78%); p= 0.004. The Cox proportional hazards model showed that the hazard of relapse for smokers who achieved complete response was nearly twice that of nonsmokers; however, this difference did not reach statistical significance.

CONCLUSION

There is an association between type of fistulae and complete response to infliximab in patients with fistulous Crohn's disease. External fistulae in general and perianal fistulae in particular have a higher rate of closure in response to infliximab compared to other types of fistulae.