Incidence of microscopic colitis in the Netherlands. A nationwide population-based study from 2000 to 2012

Abstract

Background

Incidence rates of microscopic colitis are mainly based on regional data from a limited number of countries. To evaluate geographical differences and changes over time, more nationwide incidence rates are needed.

Aims

The aim of this retrospective study was to assess the incidence rate of microscopic colitis in the Netherlands in a nationwide cohort.

Methods

A search was performed in the Dutch pathology registry, covering records of all approximately 16.5 million inhabitants. Incident cases were defined as a first diagnosis of microscopic colitis (collagenous or lymphocytic colitis) between 2000 and 2012.

Results

In total, 7228 incident cases were identified with a mean annual incidence rate of 3.4 per 100,000 person years. Collagenous colitis was present in 3741 cases and lymphocytic colitis in 2718 cases, with a mean annual incidence rate of 1.8 and 1.3 per 100,000 person years, respectively. Remaining 769 cases were described as undefined microscopic colitis. Collagenous and lymphocytic colitis incidence rates increased significantly over time (p < 0.001) with a male:female ratio of 1:3 and 1:2, respectively.

Conclusion

The Dutch mean annual incidence rates of collagenous and lymphocytic colitis were considerably lower than previously reported by other countries. However, incidence rates increased gradually over time, with a clear female predominance.

Introduction

Microscopic colitis (MC) is an umbrella term for chronic watery diarrhoea with a normal endoscopic appearance of the colonic mucosa, and characteristic histological abnormalities. MC includes two subtypes, i.e. collagenous colitis (CC) and lymphocytic colitis (LC) [1], [2]. Histologically, the hallmark of CC is the presence of a thickened subepithelial collagen layer (10 μm or more), whereas LC is characterized by an increased number of intraepithelial leukocytes [3], [4], [5]. Both MC entities are predominantly found in females over 65 years of age [6]. Whether CC and LC can be considered histopathological variants of the same disease or two different clinical entities is still matter of discussion [7], [8], [9]. Clinically, both CC and LC are characterized by a chronic relapsing course [10], [11]. Although MC is successfully treated with oral budesonide in most cases, the relapse rate after cessation of treatment is 50–80% [12]. The chronic, relapsing, watery diarrhoea is the main contributor to the significantly decreased quality of life in MC patients [13], [14], [15].

Epidemiological studies performed in Sweden reported mean annual incidence rates up to 5.4 per 100,000 person years for CC in 2005–2010 [16] and 4.5 per 100,000 person years for LC in 2004–2008 [17]. Data collected in the US (Olmsted County, Minnesota) between 2002 and 2010, showed incidence rates of 7.1 and 9.5 per 100,000 person years for CC and LC, respectively [18], while a recent Spanish study [19] reported incidence rates of 2.6 and 2.2 per 100,000 person years between 2004 and 2008 for CC and LC, respectively. Most of the epidemiological studies described an increase in incidence rates over time [12], [19], [20], [21], but recent follow-up data of two long-term, regional cohorts showed a stabilization of the incidence rates of both CC and LC over the last 10–15 years [17], [18]. Although differences between countries seem to be present, the number of studies is too limited to draw firm conclusions on the geographical distribution of MC and changes over time.

The majority of epidemiological studies reporting MC incidence rates were based on regional data with populations up to 650,000 inhabitants [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27]. Only one nationwide study was published so far, covering a rather small population of 277,184 inhabitants [28]. To evaluate potential geographical difference and changes over time, and to estimate the disease burden of MC in future, more data on long-term, nationwide incidence rates are warranted. Therefore, the aim of the present study was to assess the mean annual incidence rate of MC in the Netherlands in a nationwide, population-based cohort, spanning a thirteen-year period.