| Abstract |
Background
The majority of Crohn’s disease (CD) patients with B1 phenotype (i.e. non-stricturing non-penetrating disease) will develop over time a stricturing or a penetrating pattern (i.e. B2 and B3 respectively). Ileal location, active smoking, and number of flares per year were reported to be the major determinants of phenotypic changes. Some authors suggested that intra-abdominal penetrating disease and perianal penetrating disease represent distinct clinic entities which should be studied independently. Although in certain patients with proctitis and left-sided UC, the inflammation spreads proximally, studies on the stability of disease extent over time and risk factors influencing disease extent progression have failed to offer conclusive information. We aimed to study the long-term outcome of a homogeneous and well-defined group of Cretan patients with IBD. Particular emphasis was given on the study of rate and risk factors that might influence: change of disease behaviour over time in B1 CD patients; and proximal disease extension in UC patients with proctitis and left-sided colitis at diagnosis;
Patients and Methods
The patient cohort seen at the Gastroenterology Department of the University Hospital of Heraklion institution has been enriched with cases followed-up at the Gastroenterology Department of Venizelion General Hospital of Heraklion (since its opening at 1996). Of the 736 patients listed in the IBD registry (498 with UC and 238 with CD), 372 (256 with UC and 116 with CD) met the following criteria for inclusion in this analysis: A confirmed diagnosis of UC or CD by standard clinical, radiologic, endoscopic and histologic criteria; A satisfactory assessment of the extent of disease at the initial examination in patients with UC and of disease behaviour at diagnosis in patients with CD; and a follow-up period of at least 60 months; The extent of UC was defined as proctitis, left-sided colitis and extensive colitis. Proximal disease extension was defined as the finding at any follow-up colonoscopy of macroscopic inflammation extending beyond 15 cm from the anus in case of proctitis or beyond the splenic flexure in case of left-sided colitis. The clinical classification of patients with CD was carried out in strict accordance with the Vienna criteria. Survival analyses, using the Kaplan-Meier method with log-rank test, and Cox’s proportional hazards model for multivariate analyses were used to study the predefined end-points. In the multivariate procedure, p ≤ 0.05 was considered as the level of significance. Results of analysis are presented as hazard ratios (HR) with 95% confidence intervals (95% CI).
Results
Crohn’s disease patients
A total of 116 patients with CD were included in the study. The mean follow-up time was 10.5 ± 6.3 years. The actuarial risk of relapse was 97% (95% CI: 95.3%-98.7%) at 10 years after diagnosis. B1 phenotype accounted for 68.9% of CD patients at diagnosis. However, at the end of study, only 36.8% of cases (i.e. 42 patients) had B1 disease. Five years after diagnosis, the proportion of B1 CD patients decreased from 68.9% to 55.1% (i.e. 64 patients). Survival analysis showed that 10 and 20 years after diagnosis, the proportion of CD patients still remaining B1 was 38.9% (95% CI: 34%-43.8%) and 12.1% (95% CI: 5.3%-18.9%), respectively. During follow-up, the behaviour of CD was reclassified in 47.5% of patients with B1 phenotype at diagnosis (i.e. 38 out of 80). The proportion of B1 patients having their disease behaviour reclassified at 5 years after diagnosis was 20% (i.e. 16 out of 80 patients). The actuarial risk of disease behaviour reclassification in patients with B1 phenotype at diagnosis was 43.6% (95% CI: 37.5%-49.7%), and 82.5% (95% CI: 72.7%-92.3%), at 10, and 20 years after diagnosis, respectively. Active smoking (HR: 3.01; 95% CI: 1.406-6.457; P: 0.005) and non-L2 disease (non-colonic disease) (HR: 3.01; 95% CI: 1.165-7.785; P: 0.023) were found to be significantly associated with B2 and B3 reclassification in patients with B1 phenotype at diagnosis. The actuarial risk of B3 reclassification in patients with B2 phenotype at diagnosis (27.8% (95% CI: 12.7%-42.9%) at 10 years after diagnosis) was not significantly different from that of overall reclassification of patients with B1 phenotype at diagnosis (however, P value was borderline for significance, i.e. 0.073). The actuarial risk of appearance of the other penetrating phenotype in an already B3 patient was 22.7% (95% CI: 16%-29.4%), 34% (95% CI: 23.2%-44.8%), and 43% (95% CI: 30.1%-55.9%), at 5, 10, and 20 years after diagnosis of B3 behaviour, respectively. The cumulative probability of receiving immunosuppressives was 52.4% (95% CI: 45.6%-59.2%) at 10 years after diagnosis. A total of 35 patients (30.1%) underwent 44 operations during the observation period. At 5 years of follow-up, 23 out of 116 CD patients (19.8%) had experienced a major surgery. The actuarial risk of surgery in the entire cohort of our CD patients was 29.7% (95% CI: 25.1%-34.3%) and 67.9% (95% CI: 52.5%-83.3%), at 10, and 20 years after diagnosis, respectively. Patients with L2 (colonic) disease had a lower risk of surgery than patients with disease in other locations (P: <0.001, 0.011, and 0.026 for comparisons between L2, and L1, L3 and L4, respectively). Τhe risk of surgery was significantly higher in patients with current B2 and B3 phenotypes when compared to that of patients with current B1 phenotype, whereas the difference between B2 and B3 curves was insignificant. Active smoking (HR: 4.06, 95% CI: 1.778-9.311, P: <0.001) and non-B1 behaviour at diagnosis (HR: 5.87, 95% CI: 2.586-13.338, P: <0.001) were significantly associated with surgery.
Ulcerative colitis patients
A total of 256 patients with UC were included in the study. The mean follow-up time was 12.1 ± 7.5 years. The actuarial risk of relapse was 92.3% (95% CI: 90.5%-94.1%) at 10 years after diagnosis. At 5 years of follow-up, proximal extension was documented in 11 out of 62 patients with proctitis (17.7%) and 7 out of 124 patients with left-sided colitis (5.6%). The actuarial risk of proximal extension in patients with proctitis was 36.8% (95% CI: 29.7%-43.9%) and 51.8% (95% CI: 43.3%-60.3%), at 10, and 20 years after diagnosis, respectively. In contrast, the actuarial risk of proximal extension in patients with left-sided colitis was significantly lower, 17.1% (95% CI: 13.3%-20.9%) and 38.8% (95% CI: 30.5%-47.1%), at 10 and 20 years after diagnosis, respectively. Regression model analysis selected non-smoking as the sole factor significantly associated with disease progression in patients with proctitis (HR: 4.39; 95% CI: 1.025-18.799; p value: 0.046). None of the variables under study was significantly associated with disease progression in patients with left-sided colitis. The cumulative probability of receiving immunosuppressives was 14.1% (95% CI: 10.9%-17.3%), and 32.2% (95% CI: 27.1%-37.3%), at 10, and 15 years after diagnosis, respectively. The actuarial risk of surgery was 2.9% (95% CI: 1.2%-4.6%), και 10.5% (95% CI: 5.8%-15.2%) at 10, and 20 years respectively (years refer to time interval after the initial period of 5 years after diagnosis).
Conclusions
Classification of Crohn’s disease patients in B1 phenotype should be considered as temporary. Smoking and non-colonic disease are risk factors for behavioural change in B1 Crohn’s disease patients. Proximal extension is more common in ulcerative colitis patients with proctitis than in those with left-sided colitis. Among proctitis patients, proximal extension is more common in non-smokers.
|
Κλινική πορεία ελκώδους κολίτιδας και νόσου Crohn στην Κρήτη κατά την διάρκεια της δεκαετίας 1990-1999
