P340 Real-world treatment patterns and recurrence rates of rectovaginal fistulas in patients with Crohn’s disease: A retrospective cohort database analysis

Karki, C.(1);Latremouille-Viau, D.(2);Gilaberte, I.(3);Hantsbarger, G.(4);Romdhani, H.(2);Lightner, A.(5);

(1)Takeda Pharmaceuticals USA- Inc., Global Evidence and Outcomes, Cambridge, United States;(2)Analysis Group- Inc., Analysis Group, Montreal, Canada;(3)Takeda Pharmaceuticals USA- Inc., Clinical Sciences, Cambridge, United States;(4)Takeda Pharmaceuticals USA- Inc., Statistical and Quantitative Sciences, Cambridge, United States;(5)Cleveland Clinic, Department of Colon and Rectal Surgery, Cleveland, United States


Rectovaginal fistulas (RVF) are a difficult to treat perianal complication of Crohn’s disease (CD) with a significant impact on quality of life. Management of RVF includes both medical and surgical interventions. Few studies have assessed real-world treatment patterns for CD-related RVF. This retrospective US cohort database analysis aimed to assess the rate of, and time to RVF episodes of care in CD, and describe treatment patterns.


IBM Truven Health MarketScan® databases (Commercial and Medicare supplemental administrative claims [01/01/01 to 31/12/19]) were used. Adult females with a diagnosis code for CD, ≥1 medical claim with a RVF-specific diagnosis/procedure ICD-9/10 code on or after the date of first observed CD diagnosis, and ≥180 and ≥720 days of continuous health plan enrolment before and after index (date of first RVF-related code after first CD diagnosis date), respectively, were included. Treatment patterns were assessed during the 6 months pre-index (baseline period) and any time post-index. The first RVF episode of care (episode) started on the date of the first RVF-related code after CD diagnosis. RVF-related codes ≤90 days apart were considered as the same episode. Time to subsequent episodes was assessed by Kaplan–Meier (KM) analysis. Only descriptive statistics were reported for the proportion of patients with episodes and subsequent treatment patterns.


Of 274 096 adult females diagnosed with CD during a continuous eligibility period, 2540 (0.9%) had a RVF-specific code. Overall, 963 patients met the inclusion criteria (median age: 46.0 years). The median follow-up (time between the index date and end of continuous health plan enrolment or data availability) was 48.5 months. During the follow-up period, 963 (100%), 430 (44.7%) and 217 (22.5%) patients had at least one, two or three episodes, respectively. KM analysis showed the probability of having a subsequent episode within 1, 2 and 5 years of their first or second episode (Figure). During the baseline period, 775 (80.5%) and 287 (29.8%) patients received non-biologic or biologic therapies, respectively, and 929 (96.5%) and 494 (51.3%) at any time post-index. At any time post-index, 587 (61.0%) patients had ≥1 RVF-related surgery.

Figure. Time to subsequent RVF episode of care


This study showed patients with CD-related RVF required recurrent episodes of medical and surgical care in a US real-world setting. In addition, post-index, an important proportion of patients were observed with biologics use (>50%) and nearly two-thirds of patients had RVF-related surgery. Patients with CD-related RVF had varied treatment patterns and more studies are needed to inform the standard of care for patients with RVF.

Sponsor: Takeda Pharmaceuticals, Inc.