Article Text

Sudden-onset watery diarrhoea in a middle-aged woman

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

A 55-year-old woman presented with a 2-month history of profuse watery diarrhoea. The onset of diarrhoea was very sudden − she was woken up at 04.00 h one morning with a severe urge to move her bowels. The motions were watery, frequency averaging five times per day. There was no rectal bleeding, abdominal pain or vomiting. She had lost 2 kg in weight but there was no history of iritis, arthritis, rashes or aphthous ulcers. Clinical examination was unremarkable. Laboratory investigations revealed normal full blood count, urea and electrolytes. C-Reactive protein was not raised and there was no growth on stool culture. Colonoscopy showed mild erythema with loss of vascular pattern in the transverse colon, but no ulceration. Multiple biopsies were taken from the region. Histology of the colonic biopsy is shown in the figure.

Figure Histology of the colonic biopsy


What is the diagnosis?
What is the pathogenesis of this condition?



The diagnosis is collagenous colitis. Histologically, there is a characteristic, thick, subepithelial collagen band. There is a chronic inflammatory cell infiltration of the lamina propria, but the crypt architecture is well maintained.


It has been postulated that an initial insult by a luminal agent, which may be an infective agent, gluten, or a drug, eg, a non-steroidal anti-inflammatory, may damage the bowel mucosa leading to an autoimmune response, with ongoing chronic inflammation in susceptible individuals.1


Collagenous colitis was first described by Lindstrom in 1976.2 He reported a patient with chronic watery diarrhoea in whom a subepithelial collagen band was seen on rectal biopsy. More recent reports of collagenous colitis have suggested that it is not as rare as initially supposed. 3-6 There is another group of patients with colorectal biopsies revealing changes very similar to collagenous colitis in terms of lymphocytic inflammation, but without a collagen band. To reduce confusion the term ‘lymphocytic colitis’ has been suggested to describe this entity. ‘Microscopic colitis’, in turn, is a general term denoting conditions characterised by microscopic colonic inflammation, which includes the subsets of collagenous and lymphocytic colitis.1

Collagenous colitis has a striking female predominance (10:1) and the mean age is in the sixth decade.1 There is chronic, watery diarrhoea with no bleeding. The diarrhoea is secretory in nature, with an active colonic secretion of chloride ions associated with a passive secretion of sodium and water. The diarrhoea may be of sudden onset in over 40% of patients.3 Typically, colonoscopy is normal or shows only mild, nonspecific changes. All the other investigations are negative and diagnosis depends on histology. There is an increased prevalence of collagen deposition in the proximal colon. In an initial series, up to 40% of cases were reported to have normal histology of the rectosigmoid colon.4 This has been challenged in a recent study which found that left-sided biopsies alone would have been sufficient for diagnosis in 89% of cases.5 The collagen band consists of type III collagen rather than the usual type IV. Since type III is usually involved in repair processes, the suggestion is that the collagen band is a result of inflammation. This view is supported by the results of sequential biopsy studies showing inflammation preceding the appearance of collagen.4 Most evidence suggests that the band is not important in the pathophysiology of the disease. The presence or thickness of the collagen band correlates poorly with the symptoms, whereas the degree of inflammation correlates with diarrhoea and response to treatment.4 Certain disease associations have been noted in patients with collagenous colitis. These include ulcerative colitis, Crohn's disease, coeliac disease, rheumatoid arthritis, systemic lupus erythematosus, scleroderma, pernicious anaemia, myasthenia gravis and Hashimoto's disease.3

The natural history of collagenous colitis appears to be benign, following a chronic relapsing course. Often spontaneous or treatment induced, sustained resolution of symptoms may occur. The collagen band can disappear during therapy or occasionally in association with spontaneous remission. Sulphasalazine and other 5-ASA derivatives have the best response rate.1 Steroids are also very effective, but their response is not sustained after withdrawal and the doses required are often high.3 Symptomatic treatment with antidiarrhoeals is the only therapy required in most cases. Surgery in the form of ileostomy for faecal diversion may be required in patients with incapacitating, unresponsive disease.6 Our patient responded very well to a short course of steroids. Her diarrhoea settled within a few days of starting prednisolone and uncharacteristically she remained in sustained remission months after stopping the steroids.

Features of collagenous colitis

  • watery diarrhoea with no bleeding

  • female preponderance (F:M, 10:1)

  • mean age: sixth decade (rare in children)

  • colonoscopy normal or only minor non-specific changes

  • histology diagnostic: subepithelial collagen band in colonic biopsy

  • treatment largely symptomatic

  • chronic relapsing course (spontaneous remission in about 1/3 rd)

Final diagnosis

Collagenous colitis.