The Association of Coeliac Disease and Microscopic Colitis

A Large Population-based Study

M. Stewart; C. N. Andrews; S. Urbanski; P. L. Beck; M. Storr


Aliment Pharmacol Ther. 2011;33(12):1340-1349. 

In This Article


Coeliac Disease

Over the 5-year study period, 42 950 upper endoscopies were preformed. Coeliac disease was the stated indication for 1342 endoscopies with an additional 4058 done for symptoms that could be consistent with CD; diarrhoea, malabsorption, weight loss, failure to thrive and iron-deficiency anaemia. Our search of the pathology data identified 2244 patients who underwent duodenal biopsy to confirm CD. Of these, 763 patients had biopsies demonstrating histopathological features consistent with CD. The age-standardised and gender-standardised incidence of CD ranged from 10.4 per 100 000 population in 2004 to 15.7 per 100 000 population in 2007 (Figure S1). Between 2004 and 2008, there was a trend towards more upper endoscopies being performed and as a result more biopsies being taken for CD (Table 1). The increase in biopsies paralleled the increase in upper endoscopies, which indicates that the number of biopsies per upper endoscopy did not increase. Duodenal biopsy results for males and females are shown in Figure 1. The age-adjusted and gender-adjusted incidence of CD increased by an average of 8.6% annually (Table 2). The number of cases of CD per 1000 upper endoscopies performed did not increase suggesting that the increased incidence of CD was related to the increased number of upper endoscopies and biopsies rather than increased frequency of CD in the same population (Table 1).

Figure 1.

Duodenal biopsy results per 1000 Upper Endoscopies (UE) reported normal or where coeliac disease was diagnosed. (a) Duodenal biopsies in males. (b) Duodenal biopsies in females.

Among patients who had a duodenal biopsy for CD, the main indications for the procedure were the presence of upper gastrointestinal symptoms and suspected CD (Table S1). Over half (54.4%) of the patients who were diagnosed with CD underwent endoscopy with CD as the indication, compared with only 6.3% of those with normal duodenal biopsies (P < 0.001). The number of men and women who underwent upper endoscopy during the study period for any reason was approximately equal (50.2% women vs. 49.8% men). Women were more likely to undergo upper endoscopy with CD as the stated indication (67.6% female vs. 32.4% male; P < 0.001) and consequently were proportionally more likely to have a duodenal biopsy for CD (65.2% women vs. 34.8% men; P < 0.001).

Patients diagnosed with CD had a mean age of 34.0 years, with almost half (45%) being under the age of 30. Patients who had normal duodenal biopsies had a mean age of 50.3 years. Those diagnosed with CD were significantly younger than those with normal duodenal biopsies (P < 0.0001).

Microscopic Colitis

Over the study period, 1106 patients were diagnosed with MC, 347 of whom were diagnosed with CC and 759 with LC. The age-standardised and gender-standardised incidence of MC ranged from 16.9 per 100 000 population in 2004 to 26.2 per 100 000 population in 2008, which indicates that the incidence of MC is still rising. This study demonstrates that the increasing incidence of MC was driven largely by increased diagnosis of LC with relatively stable incidence of CC (Figure 2). Patients diagnosed with MC had a mean age of 57.8 years with almost half (44%) being over 60 years of age. Women were more likely to be diagnosed with MC than men (P < 0.0001).

Figure 2.

Incidence rates of lymphocytic colitis and collagenous colitis, per 100 000 population, age-standardised and gender-standardised to the Canadian population. (a) Incidence in males by year. (b) Incidence in females by year.

Coeliac Disease and Microscopic Colitis Concomitance

Over the 5-year study, 120 patients had both a colonic and duodenal biopsy performed. These biopsies established the concomitant diagnosis of CD and MC in 40 patients (32 LC and 8 CC). Ninety percent of these patients were women (36 of 40), which is a significantly higher rate of disease coincidence compared with men (P = 0.004). With a mean age of 50.1 years, the MC and CD patients were also significantly older than the CD patient cohort, and significantly younger than the MC cohort, (P < 0.000 and P = 0.002, respectively). Accordingly, the majority of patients diagnosed with both CD and MC were women in their forties and fifties (52.5%) (Figure 3). Three-quarters of patients (77.5%) were younger than 60 years with the youngest being diagnosed at 8 years of age. Most patients with both MC and CD underwent upper endoscopy for either confirmation of CD or unexplained diarrhoea (98%), and lower endoscopy for unexplained diarrhoea (68%). As expected in MC, 95% of these patients had no endoscopic findings on lower endoscopy (Table S2).

Figure 3.

Histogram of age at the time of diagnosis for males and females with both microscopic colitis and coeliac disease.

Same-day upper and lower endoscopy led to the concurrent diagnosis of both CD and MC in 19 of the 40 patients with both diagnoses. The remaining patients had CD and MC diagnosed on separate occasions; 11 of 40 were diagnosed with CD first and 10 of 40 were diagnosed with MC first. Interestingly, 47 patients had multiple negative duodenal biopsies taken on multiple occasions, but did not undergo colon biopsy on these occasions.

Within the CD population, MC occurred at an age-standardised and gender-standardised annual rate of 11.4 per 1000 cases of CD (LC = 9.0 per 1000; CC = 2.4 per 1000). The gender-specific standardised annual rates of MC were 15.1 per 1000 cases of CD for women (LC = 11.6 per 1000; CC = 3.6 per 1000) and 3.5 per 1000 cases of CD for men (LC = 3.5 per 1000; CC = 0.0 per 1000). The overall SIR for MC within the CD cohort was 52.7 (95% CI = 37.7, 71.8). Conversely, within the MC population, CD occurred at an overall standardised annual rate of 7.9 per 1000 cases of MC with an overall SIR of 58.9 (95% CI = 42.1, 80.2; Table 3). The gender-specific standardised annual rates of CD were 9.4 per 1000 cases of MC for women and 3.2 per 1000 cases of MC for men. Within the LC cohort, the standardised annual rate of CD was 9.1 per 1000 cases of LC with gender specific rates of 10.9 per 1000 cases of LC in women and 4.0 per 1000 cases of LC in men. The overall standardised annual rate of CD in CC was 5.3 per 1000 cases of CC with a female specific rate of 6.4 per 1000 cases of CC. No men were diagnosed with CD and CC. These incidence rates reveal an association between CD and both LC and CC that was strongly statistically significant (P < 0.0001).

Endoscopy Factors

Eighty of the 120 patients who had upper and lower gastrointestinal biopsies underwent both procedures on the same day. The 40 patients who had the procedures on different days had a mean of 237.5 days between procedures (range 1–1050). There was no association between the order of biopsies and diagnosis (data not shown). Thirty-nine of the 40 patients who had normal colon and duodenal biopsies had both procedures on the same day. In contrast, 38 of 80 patients diagnosed with either, or both, CD and MC had biopsies taken on different days (P < 0.0001).

Physicians with an academic practice obtained both duodenal and colonic biopsies in 5.9% of patients in whom they performed endoscopy, compared with 2.0% of patients seen by their private practice colleagues (P < 0.0001). Sixty-six percent of the patients who had both colon and duodenal biopsies were seen by an academic endoscopist. Surgeon endoscopists obtained both duodenal and colonic biopsies in 0.6% of patients, whereas gastroenterologists obtained both biopsies in 4.0% of patients (P = 0.020). Surgeon endoscopists also performed fewer same-day upper and lower endoscopic biopsies (0.6% of patients) than gastroenterologists (2.8% of patients; P = 0.045). The age group of endoscopists (< or >50 years old) did not have a significant association on endoscopy patterns (data not shown).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.