Read et al (1) introduced the term lymphocytic colitis (LC) in 1980 to describe patients with chronic or relapsing non-bloody watery diarrhea and an endoscopically and/or radiologically normal colon. However, a histopathological examination of colonic biopsy samples from these patients shows increased intraepithelial infiltration of lymphocytes and increased inflammatory cells within the lamina propria (1). The reported prevalence of LC in population-based investigations has varied from 14.2 to 45 per 100,000 individuals (2–8), while its incidence has been reported to be between 0.6 and 16 per 100,000 individuals (7–13). Although LC can occur at all ages, its incidence peak occurs at approximately 60 years of age, and females constitute 68% of patients with LC (14). Analysis of the prevalence and incidence of LC in different populations should, therefore, consider the age structure of these population pyramids (15).

Long-term epidemiological studies have found that the incidence of LC is increasing (14,16); this could be a real increase or the result of an increased awareness of the necessity of examining colonic biopsy samples in patients with diarrhea without obvious etiology (14). In Örebro county (Sweden) and Olmsted county (MN, USA) where these studies were conducted, the awareness of LC has been high and constant during all of the investigation period, making it unlikely that this factor played a role in the observed increase in LC incidence. However, change in the age distribution during the long observation periods (>15 years) may have an effect on this increase, since official Swedish statistics indicate a considerable increase in the proportion of individuals >60 years of age in Örebro county.

There have been several recent comprehensive reviews on LC (5,14,17–21). The present review takes a different approach from the others by aiming not merely to present an updated account of LC, but also to highlight the debated issues of clinical presentation, diagnosis, pathogenesis and treatment options.

The cardinal symptom for LC is chronic or recurrent non-bloody diarrhea (22,23). The diarrhea is chronic in 33% and intermittent in 67% of patients with LC (22,23). Patients with LC have 4.9 bowel movements per day on average; however, 22% of patients have up to 10 bowel movements daily (5,14,21,23). Other symptoms occurring in patients with LC include abdominal pain and weight loss (24–27) (Table I). However, in some patients, constipation is the main symptom of LC (24–27). It is noteworthy that in clinical practice, colonic biopsies are routinely taken during colonoscopy of patients with diarrhea and a macroscopically normal colon, but not in those with constipation. It is therefore, not clear whether LC is exclusively a diarrheal disease.

The reported rate of spontaneous symptomatic remission in LC in follow-up studies has varied between 59 and 93% (23,27). Histopathological normalization reportedly occurs in 82% of patients with LC after approximately 38 months of follow-up (23). A follow-up study over a 6.4-year period also found a high rate of symptomatic remission in patients with LC (28). Moreover, the risk of colon cancer or inflammatory bowel disease is not increased in patients with LC (29).

The results of a physical examination, laboratory tests, small and large bowel radiology and colonoscopy are normal in LC, with the diagnosis instead being based on histological changes in colonic biopsy samples. The typical histological changes include an increase in intraepithelial lymphocytes (≥20/100 enterocytes) and infiltration of inflammatory cells (mainly lymphocytes and plasma cells) in the lamina propria (21,30). Although surface epithelial damage can be observed, the architecture of the mucosa is normal (Fig. 1) (21,30). In clinical practice, the increase in intraepithelial lymphocytes is usually clearly evident with manual counting only required in cases with a borderline number of lymphocytes (21). The surface mucosa overlying lymphoid aggregates always contains higher numbers of intraepithelial lymphocytes; thus, these areas should be avoided when an LC diagnosis is being made (21).

While the histopathological diagnosis of LC is straightforward, it is unclear whether the histopathological changes observed in LC are diffusely distributed throughout the colon or limited to a certain segment(s). Certain studies have found the histopathological changes to be more prominent and easier to detect in the right colon (cecum and the ascending and right transverse colon) (11,31,32), while other authors have maintained that biopsies obtained from the left colon (sigmoid and descending colon) are reliable for the diagnosis of LC (10,33,34). These controversial results question whether colonoscopy with a segmental biopsy is required for the diagnosis of LC, or whether sigmoidoscopy is sufficient. Solving this issue is important for clinical practice, since the risk of complications, the clinicians labor and the economic burden are higher for colonoscopy than for sigmoidoscopy. Since this matter has not yet been resolved, several gastroenterologists perform colonoscopy with a segmental colonic biopsy to ensure an accurate diagnosis.

Chromogranin A is a 68-kDa protein comprising 439 amino acid residues, which was isolated from secretory granules of the adrenal medulla (35). Chromogranin A is co-stored and co-released with monoamines and peptide hormones present in the adrenal medulla, pituitary gland, parathyroid, thyroid C-cells, pancreatic islets, endocrine cells of the gastrointestinal tract and sympathetic nerves (35–37). Chromogranin A is therefore considered to be a general marker for all endocrine cells. Some of the patients in a cohort of irritable bowel syndrome (IBS), showed an extremely high density of colonic chromogranin A-positive cells (Fig. 2), and were subsequently identified as having LC (38). This unexpected observation was confirmed in a larger cohort of patients with LC, and led to the colonic chromogranin A-positive cell density being proposed as a diagnostic marker for LC (Figs. 3 and 4) (39). The use of the chromogranin A-positive cell density as a marker for the diagnosis of LC has a high sensitivity (97 and 100%, in the right and left colon, respectively) and specificity (98 and 94%, in the right and left colon, respectively) (39). The time and cost associated with the use of this marker are the same as those for the detection of lymphocytes. Using the chromogranin A-positive cell density as a marker has several advantages: i) it has a high sensitivity and specificity in the left colon, thus sigmoidoscopy with biopsy can be used in the diagnosis; ii) it is not affected by the artifact caused by lymphoid aggregates and is thus useful in borderline cases where lymphocyte infiltration does not completely fulfill the definition of LC; and iii) it can be used to differentiate patients with LC from those with IBS, refractory celiac disease, lymphocytic enterocolitis, autoimmune enteropathy and lymphoma.

LC and IBS have similar symptoms and normal radiological and endoscopic appearances of the colon (40), which means that patients with LC are at risk of being misdiagnosed as IBS (24,40–43). The task of differentiating between LC and IBS is made even more difficult by the fact that some patients with IBS exhibit an increase in the number of intraepithelial lymphocytes (40,44). The use of colonic chromogranin A-positive cell density as a marker for LC may prove to be beneficial in such cases, since the density of this marker in the colon is low in patients with IBS (45).

Celiac disease has been reported to be associated with LC; between 4.3 and 31% of patients with celiac disease exhibit histopathological changes typical of LC (46–48). Approximately 7% of patients with LC have celiac disease (6,10,23,25,49–51). It has been speculated that colonic lymphocyte infiltration observed in the colon of patients with celiac disease is part of the same autoimmune pathogenesis observed in the small intestine. The coexistence of celiac disease and LC makes it necessary to further examine patients with celiac disease who do not respond to a gluten-free diet. Thus, one of the first tasks in clinical practice is to exclude the possibility of celiac disease in patients exhibiting LC symptomology.

It is generally accepted that LC is secondary to an abnormal immune reaction in predisposed individuals. Gastrointestinal infections, drugs and/or autoimmunity possibely trigger this reaction. An abnormality in the colon neuroendocrine system has also been reported and seems to play a role in the pathophysiology of LC.

The drugs commonly used in the treatment of LC are loperamide, 5-aminosalicylates, budesonide, prednisolone, bismuth subsalicylate, immunosuppression and anti-TNF-α therapies (17,93–95) (Fig. 9). Cholestyramine has also been recommended for the treatment of LC, but its clinical response was possibly due to the fact that the patients treated suffered from bile acid malabsorption (93). Loperamide, which is a symptomatic non-specific anti-diarrheal agent, is often the first-line therapy prescribed. It is administered as a tablet of 2 mg (up to 8 tablets taken daily if necessary). A complete or partial response to loperamide has been observed in 70–73% of patients (11,22). 5-Aminosalicylates, such as sulfasalazine and mesalamine are a common therapy for LC in clinical practice due to positive responses and tolerance in the majority of patients (22,96). However, there have been several reports of 5-aminosalicylates inducing remission in <50% of patients (25,97). Budesonide is the most documented drug used in the treatment of LC. It is a glucocorticoid with no significant adverse events that are known to occur in systemic treatments with corticosteroids. This is due to its limited systemic availability caused by the extensive first-pass hepatic metabolism by cytochrome P-450 enzymes. Clinical remission has been reported in 82–86% of patients with LC following 6-weeks of treatment with 9 mg of budesonide daily (11,98). Relapse after cessation of budesonide treatment has been shown to occur in 14–44% of patients after 1–7 months (11,98). The relapsed patients responded to new treatment and remained in clinical remission with a maintenance therapy of a median daily budesonide dose of 3 mg (11,98). Four weeks of treatment with prednisolone at 40 mg daily tapered by 5 mg/week resulted in complete or partial clinical remission in 87–94% of patients with LC (11,99). However, the majority of these patients relapsed after termination of the therapy (11,22). Bismuth subsalicylate at a dose of 2 or 3 tablets (262 mg each) 3 or 4 times daily for 8 weeks induced remission in the majority of patients which was sustained up to 7–23 months (22,100). No toxicity was absorbed despite the high dose of bismuth, which may have been due to the short treatment duration or low bismuth absorption. However, treatment with bismuth is not without any toxic effects and its availability is limited in certain countries. Treatment with azathioprine at 2–2.5 mg/kg/day, with 6-mercaptopurine at 1 mg/kg/day both indefinitely, and anti-TNF-α have produced good results in patients with refractory LC (94,99,101). The indications for surgical therapy are limited (102).

Concomitant drug use should be assessed and drugs that potentially cause drug-induced LC should be discontinued, if possible. Considering the possibility that LC and PI-IBS can be the same disorder, diet, regular exercise and probiotic intake recommendations provided to patients with IBS (103–105) should also be recommended for patients with LC. The benign course and high rate of spontaneous remission of LC denotes that drugs with severe side-effects should be avoided if possible. Furthermore, the drug cost and drug coverage may be limiting factors for some patients. These aspects should be taken into account when selecting treatment. An algorithmic approach to the treatment of LC is recommended (Fig. 9).

LC is not a rare disorder as previously thought. Although LC is considered a diarrheal disorder, it can present with other symptoms, such as constipation or abdominal pain. Colonic biopsy samples in patients presenting with symptoms other than diarrhea and with a macroscopically normal colon should be taken more frequently than routinely done at present. The density of colonic chromogranin A-positive cells may be used as a diagnostic tool for LC. The cause of LC may be a low-grade inflammation with an interaction with the colonic neuroendocrine system. There are good reasons to suspect that LC and PI-IBS are one and the same disorder. The benign course and high rate of spontaneous remission, as well as the drug cost and drug coverage should be taken into accout when making decisions regarding treatment. An algorithmic approach to the treatment of LC is preferable.