DOP50 Recommencement of Inflammatory Bowel Disease medications following colectomy for Ulcerative Colitis
Ledder, O.(1);Lujan, R.(1);Orlanski-Meyer, E.(1);Friss, C.(1);Loewenberg Weisband, Y.(2);Greenfeld, S.(3);Kariv, R.(3);Lederman, N.(4);Matz, E.(5);Focht, G.(1);Yanai, H.(6);Ollech, Y.(6);Dotan, I.(6);Turner, D.(1)
(1)Shaare Zedek Medical Center, Department of Pediatric Gastroenterology and Nutrition, Jerusalem, Israel;(2)Clalit Health Services, Clalit Research Institute, Tel Aviv, Israel;(3)Maccabi Healthcare Services, Maccabi Healthcare Services, Tel Aviv, Israel;(4)Meuhedet Health Services, Meuhedet Health ServicesTel, Tel Aviv, Israel;(5)Leumit Health Services, Leumit Health Services, Tel Aviv, Israel;(6)Rabin Medical Center, Division of Gastroenterology, Petah Tikva, Israel
Background
Total colectomy (TC) is often considered a curative procedure for patients with chronic refractory ulcerative colitis (UC) or acute severe colitis. Chronic pouchitis and de novo Crohn’s disease (CD) are well recognized sequelae following TC, yet the true incidence of these are poorly characterized in large population models. We assessed the rate of subsequent utilization of IBD medications as a proxy marker of clinically significant pouchitis or de novo CD following TC for UC.
Methods
This study utilized data from the Epi-IIRN project, a meta-database incorporating patient data from all four health maintenance organisations (HMO) in Israel, representing 98% of the population. We included all patients identified in the prevalence cohort with an initial diagnosis of UC who underwent TC from January 2000, with ≥ 6 month follow-up. Primary outcome was utilization of IBD medications (including thiopurines, methotrexate, infliximab, adalimumab, vedolizumab, ustekinumab and tofacitinib) following TC. Secondary outcomes were time to commencement of medications, hospitalizations and repeated surgery. Potential predictors of IBD medication use were identified using multivariable models.
Results
Overall 23,506 patients with UC were identified in the prevalence cohort of whom 456 patients underwent TC for UC and were included in our analysis. 51% of our sample were female, with a median follow up of 8.5 years (IQR 3.8-13.2) and 3956 patient-years. Median age at UC diagnosis was 44.1 (26.5-56.2) years and at TC was 50.2 (34.2-61.7). IBD medications were commenced in 88 (19%) of patients, including 54 (12%) biologics, 56 (12%) immunomodulators and 3 (1%) tofacitinib. A diagnosis of CD was formally assigned to 65 (75%) of these patients. The need for IBD medications was gradual (figure 1).
Patients recommenced on IBD medications were younger at diagnosis (30.3 years (18.3-49.7) vs 41.4 (24.2-55.6), p<0.001), at colectomy (34.3 (21.5-54.1) vs 47.5 (30.4-60.6), p<0.001) and with shorter interval from UC diagnosis to TC (2.2 years (1.0-3.3) vs 2.8 (1.4-5.1), p=0.03). TC during childhood was performed in 33 (7%) patients and these had higher utilization of IBD medications post TC (16/33 (49%) vs 72/423 (17%), p=0.001).
Conclusion
In this nationwide analysis we show that almost 20% of patients require ongoing IBD medications following TC for UC. Younger age is associated with higher rate of subsequent medication use. Patient expectations need be adjusted to account for the potential ongoing requirement of long-term medication following TC.