Ulcerative colitis (UC) is an incurable chronic disease that commonly develops at young ages and lasts throughout life, impacting quality of life. Although a recent systematic review showed that more than 80% of studies from high prevalence regions such as North America and Western Europe reported a plateau or decreased incidence of UC, the prevalence of UC remains high. In contrast, a low prevalence region, such as Asia, which has newly industrialized countries, is facing increasing incidence [1]. Therefore, changing the epidemiology of UC in the next decade will challenge health care systems worldwide.

The burden of UC varies in different countries and populations. Although UC is not associated with an increased risk of mortality [2], it is associated with high morbidity [3]. Because of the complex and incurable nature of the disease, understanding the natural history of UC is important to prepare health care resources and policies for this chronic and costly disease. A systematic review of the natural history of UC that included several studies from high incidence countries reported that approximately 50% of patients with UC required corticosteroids at 10 years, the rate of UC-related hospitalization at 10 years was 50%, and the rate of colectomy at 10 years was 10% to 15% [3]. However, long-term studies from low incidence countries are lacking. In Asia-Pacific region, Thailand had the lowest incidence of UC (crude incidence of 0.28 per 100,000 persons) [4].

Hence, we aimed to evaluate the natural history of adult UC in a low incidence area and changes of disease course regarding the rates of endoscopic healing, hospitalization and colectomy between 2000 and 2018.

In the present study, we observed that the patient’s characteristics and the disease extension at diagnosis of UC remained stable over time. There was a less aggressive disease course of UC. Approximately, 60% of patients had E1/E2 and about 14% of those patients developed E3 on follow-up. Although twothirds of patients experienced active disease that required systemic corticosteroid during their disease course, approximately the overall 5-year cumulative risk of requiring hospitalization was 25% and the overall 5-year cumulative risk of colectomy was 5%. These rates have been declined over time that corresponds with an increase in the use of thiopurine (63%), but not biologic use (1%).

There was no sex difference in our study population. Our findings are in line with a pooled analysis of incidence from Western studies [6]. The disease locations of UC in our study were generally similar to those in the Western [3,7], Australian [4], and Asian populations [4]. However, the overall rate of progression to extensive colitis was 14%, which was comparable to the Asia population (12%) [8] but less than that in the Copenhagen study (28%) [9]. In addition, the overall rate of requiring corticosteroid use in our study was approximately 63%, which was relatively less than that in the Sweden study (71%) [7].

The paradigm shifts in UC management such as early initiation of thiopurine and treatment target for endoscopic remission during the two decades were observed. We found a significantly increased proportion of patients with UC diagnosed in the 2010s who received more combinations of oral and topical mesalamine. The shorter time to initiate thiopurine after diagnosis and the proportion of thiopurine users has significantly increased during two decades. In the 2000s, approximately 20% of patients with UC used thiopurine in the first year after diagnosis and increased to 30% of patients with UC in the 2010s. This is in accordance with Dutch study, where there was a 2- to 3-fold increased thiopurine user from 8% in 1991-1997 to 23% in 1998-2006 [10]. For biologic agents, we did not find the changes in the use of biologic agent in our population. The overall biologic agent use was only 1% for both the 2000s and 2010s in our population. In contrast, an increase in the first year use of biologic agent (5%-10%) was found more in the West population [9-11]. A Canadian cohort in 2014 reported that the cumulative incidence of biologic agent use within 5 years of diagnosis was 7.8% [12]. This may explain by the disease course in our population, which was less severity, and then was able to control inflammation with conventional treatment. Regarding endoscopic remission, we demonstrated the significantly increased proportions of mucosal healing during two decades. The cumulative probability for achieving endoscopic healing within 3 years of the disease course was 15% in the 2000s and increased to 46% in the 2010s.

We found that the evolution of UC management, including the use of a combination of oral and topical mesalamine [13], early use of thiopurine [14] and treat-to-target strategy [15], may play role on the changes long-term disease outcomes. The UC-related hospitalization and colectomy rates in our population were lower than those in the West [3]. We also demonstrated a reduction rate of UC-related hospitalization and colectomy over the past two decades. Between the 2000s and 2010s, the 5-year cumulative probabilities for UC-related hospitalization decreased from 30% to 21% and for colectomy decreased from 6% to 2%. A recent systemic review of Western studies showed that the 5-year cumulative probability of requiring UC-related hospitalization was 29% to 54%, and the colectomy rate at 5 years of disease was 10% [3].

There have been no population-based data about the change of UC incidence over time from Thailand, and other countries in Southeast Asia. A hospital-based registry study from Malaysia demonstrated the crude UC incidence in 1980s was 0.3 per 100,000 and increased to 0.9 per 100,000 in 2010-2018 [16]. Disease emergence in developing countries suggests that changing epidemiological trends are related to westernized lifestyle and industrialization. A population-based study from the AsiaPacific region revealed that a high-density population was more likely to develop UC [17]. Thus, in the 21st century, UC is becoming a burden disease across Asia [1]. Our results may help a newly industrialized country with a low incidence of UC with further investigations regarding environmental determinants of UC development, planning for a local health care strategy to modify the disease course of UC, and providing comprehensive information about disease prognosis to a patient for making a treatment decision.

A strength of this study is that all patients were meticulously longitudinally reviewed by 2 experts rather than relying on administrative data. All incident patients were followed from disease onset until the end of the study period. However, this study has several limitations. First limitation is a retrospective nature of the study; any data not recorded in the medical records would have been missed. Second, this study is a hospital-based study. Therefore, our patients would have more severe disease than these from a population-based study. We estimated the natural history of UC would be less aggressive in the national level. Third, changes in disease extent and endoscopic healing may not be accurate. Endoscopic evaluation was not performed routinely for all patients. In our practice, we generally perform endoscopic assessment in patient with flare-up disease. For patients with active disease, we usually perform endoscopic assessment every 6-12 months. Fourth, we did not classify disease activity according to disease activity indexes such as Mayo scores. Because disease course of UC is dynamic from disease in remission to severe disease over time, adjustment for the activity of disease would have been complicated. Fifth, we did not perform the analysis on time to initiation treatment of mesalamine and systemic corticosteroids. Certain therapies, most notably mesalamine was prescribed as an initial treatment for mild to moderate disease activity whereas systemic corticosteroid was prescribed as an initial treatment for severe disease activity. However, our study allows for an assessment of ever used of combination oral and topical of mesalamine and ever used of systemic corticosteroid in UC patients. Last, the ethnic diversity in Bangkok was less than in the Asia region. Therefore, these results would limit the generalization to the country with more diverse ethnicities.

In conclusion, the disease course of UC in Thailand was less aggressive in term of the rates of systemic corticosteroid use, biologic use, colectomy and UC-related hospitalization than in the West. Over the two decades, the rates of UC-related hospitalization and colectomy have been decreasing which were similar to the West.