Background
Ulcerative colitis (UC) is a dynamic disease that can progress to involve increasing segments of the colon over time [
1]. In population-based studies, around one-third of UC patients with limited disease at diagnosis will have proximal disease extension within 10 years [
2]. As one of the major determinants of long-term disease course, proximal disease extension is associated with a more aggressive disease course evident by higher rate of therapeutic requirements and colectomy [
3], when compared with those presented extensive colitis at diagnosis [
4]. Till now, the indentified risk factors included younger age, extra-intestinal manifestations, refractory disease at diagnosis and non-smoking, [
3,
5,
6] but these factors varied between studies which merits further study.
Studies on the changes in the disease course over time of Asian patients with UC are still lacking giving the rising incidence of UC in Asia [
7]. According to previous studies, phenotypic differences exist between Asian patients with UC and Caucasians [
7‐
9]. However, the specific way of disease extent progression in Asian UC populations has been poorly described with limited sample size and short-term follow-up [
10‐
12]. It remains unclear whether Asian patients with UC present as a different way of disease extent progression from Caucasians.
The aim of this study was to evaluate the dynamic disease evolution of a large cohort of UC patients from an Asian tertiary center. We also assessed risk factors and long-term outcomes of patients with disease extension.
Discussion
To our knowledge, this study has the largest sample size to present disease extent progression in Asian patients with UC. We have shown that in a Chinese cohort of UC cases, one in six UC patients experienced E1 to E2/E3 or E2 to E3 disease extension, and one in four patients with limited UC experienced disease extension to extensive colitis, after 7.5 years of follow-up. Only disease extent at diagnosis was identified as a clinical predictor for disease extension. Progression to extensive colitis resulted in a increased therapeutic requirements.
According to a recent retrospective study aiming to evaluate the natural disease course between pancolitis and non-pancolitis E3, pancolitis was associated with higher probabilities of cumulative relapse or hospitalization [
18]. In the present study, we further classified E3 lesion based on hepatic flexure as well and found that, of patients diagnosed with E3, patients with pancolitis were associated with increased treatment requirements. Specifically, 38(33.3%) initiated steroids and 14(12.3%) initiated immunosuppressive agents in the 114 patients with pancolitis during the follow-up, whereas 16 (26.7%) initiated steroids and 6(10%) initiated immunosuppressive agents in the 60 patients with non-pancolitis E3.
Previous estimates of disease extension have varied between one-fifth to one-third of patients with E1 or E2 showed disease extension to E3 [
6,
19,
20]. According to a recent meta-analysis, approximately one quarter of patients with limited UC demonstrated disease extension over time with most extension occurring during the first 10 years [
21]. Disease extension may occur anytime after initial diagnosis with 31.1% within the first decade [
22]. In our study, 91 (17.6%) had disease extension during a median follow-up of 7.5 years. The median time for any progression of disease extent was 16.1 months (IQR: 8.3–42.2 months). The cumulative rate of disease extension was 9.9, 14.9, 19.6, 24.6 and 30.5% at 1–5 year post-diagnosis, which was comparable to that of previous studies [
3,
12].
In the present study, we found that only initial disease location had an impact on the risk of extension. Patients with initial E1/2 were at greater risk of extending to pancolitis. Several studies have demonstrated that patients with proctitis are at greater risk to extend to pancolitis [
6]. In a recent study [
20], the only predictor for UC extension was having E2 extent at baseline compared to E1. According to a recent meta-analysis, extension was 17.8% (95% CI 11.2–27.3) from E1 to E3, 27.5% (95% CI 7.6–45.6) from E2 to E3 and 20.8% (95% CI 11.4–26.8) from E1 to E2 [
21]. In our study, though patients with E1 at baseline had the highest rates of disease extension of which the majority belongs to E1 to E2 extension, patients with E2 had a higher risk of progression to E3 compared to E1. This finding suggests that clinicians need to follow E2 patients more closely who are on higher risk for progression to pancolitis. Other previously reported risk factors for disease extension have also been examined. In the current study, there was a trend, albeit not statistically significant, toward increased risk of disease progression in patients diagnosed at a younger age. In contrast, smoking may have a protective effect against proximal disease extension [
6]. Albeit there was a trend, we failed to find a statistically significant difference between smokers and non-smokers (
p = 0.171). Altogether, risk factors have been largely inconsistent across studies and better predictors of disease progression are needed.
We also found that UC disease extension was associated with increased therapeutic requirements. Prior studies have examined the relationship between disease extension and subsequent outcomes [
3,
5,
6,
10,
22]. Patients with disease extension tent to have a poor prognosis [
6,
10]. For example, proximal progression was preceded by a flare-up in the majority of patients with UC according to a retrospective study [
22]. In the population-based IBSEN cohort, there cumulative rates of colectomy were higher in extenders as compared to non-extenders [
3]. Disease flare associated with progression also follows a refractory course with higher therapeutic needs [
5]. In a Danish population-based inception cohort study, patients with proximal disease progression were more likely to be steroid-refractory, with greater need of immunotherapy, hospitalization and colectomy, as compared to non-extenders [
20]. In the present study, UC extension resulted in poorer outcomes evidenced by increased risk of flare-ups, increased need for steroids and immunosuppressive agents, and higher rate of steroid-dependence.
In the current study, we failed to demonstrate a direct causal relationship between disease extension and increasing colectomy rates probably due to the relative low rate of colectomy (1.4%). Although our study was conducted in a referral center, the colectomy rate appears to be comparable with that of previous Eastern population-based studies [
23,
24]. Previous studies have suggested that Chinese UC patients tent to have a milder disease course as compared with Caucasians [
12,
25]. Although the cause of this variation remains unclear, possible factors included less extensive disease, better response to medical therapy, and less acceptance of colectomy by patients and/or physicians [
26]. Another potential explanation might be that disease extent per se maybe not really a dictating factor for colectomy and it is more about other predictors ie, underlying severity of inflammation and/or refractoriness to medical treatments. Nonetheless, further studies are warranted to elucidate the correlation between disease extension and colectomy.
There were several limitations to our current study. First, our findings should be interpreted carefully due to the retrospective and referral center–based design of our investigation. Further long-term population-based study are required in order to reduce the selection bias. Second, due to the relative low rate of colectomy, we failed to find a relationship between disease extension and increased chance of colectomy. Last but not least, it is unknown whether any treatments or interventions decrease the risk of disease extension. According to our study, none of the medical treatment during follow-up associated with reduced disease extension, future research should investigate treatments or interventions that could prevent disease progression from E1/E2 to E3.