Collagenous and Lymphocytic Colitis

Background

The term microscopic colitis is used to describe both collagenous colitis (CC) and lymphocytic colitis (LC).^[1] Both are relatively rare conditions that are diagnosed when a patient with chronic watery nonbloody diarrhea has an endoscopically or radiographically normal colon, but colonic mucosal biopsies show unique inflammatory changes. Because the mucosa is not ulcerated or otherwise disrupted, the diarrhea generally does not contain blood or pus. CC and LC primarily affect the colon, but the terminal ileum can also be involved.

CC was described concurrently in 1976 by Lindstrom and by Freeman.^[2] In 1980, Read described microscopic colitis, which is clinically indistinguishable from CC but is differentiated from it by specific biopsy features. Later, the term LC was proposed by Lazenby to replace the term microscopic colitis and to distinguish it from infectious colitis and inflammatory bowel disease (ulcerative colitis and Crohn disease).

CC and LC should be considered in any patient with unexplained nonbloody diarrhea. Patients undergoing either sigmoidoscopy or colonoscopy for unexplained diarrhea who have normal endoscopic findings should have biopsy samples taken to diagnose or rule out either form of microscopic colitis.^{[3, 4, 1, 5]}

For patient education resources, see Digestive Disorders Center as well as Colitis, Celiac Sprue, and Crohn's Disease.

• References

Pathophysiology

The characteristic feature of lymphocytic colitis (LC) is an infiltration by lymphocytes into the colonic epithelium. Collagenous colitis (CC) shares this feature but additionally shows a distinctive thickening of the subepithelial collagen table. LC and CC have been suggested to represent different phases of a single pathophysiologic process, with LC possibly being a precursor or earlier phase of CC; however, this has not been proven.

A mixed form, characterized by both thickening of the collagen plate and an increased number of intraepithelial lymphocytes has also been described.

The diarrhea in CC is more likely due to the inflammatory process than to the subepithelial collagen layer, although this layer may serve as a cofactor in the role of a diffusion barrier. Increased levels of immunoreactive prostaglandin E2 and nitric oxide (NO) in the colonic epithelium may contribute to a secretory diarrhea in patients with CC.

Some patients with CC and concurrent collagenous infiltration of the duodenum and/or the ileum have demonstrated altered small bowel dysfunction, diagnosed by reduced D-xylose absorption.

Some individuals have bile acid malabsorption demonstrated by the Se-75-homotaurocholate (SeHCAT) test, in which a positive test result is shown by retention of less than 11% of the administered dose of radioactivity after 7 days.

Some patients with CC may have increased mucosal secretion of transforming growth factor-1 (TGF-1) and vascular endothelial growth factor, a fibrosis-enhancing peptide.

• References

Etiology

No definite etiology has been determined for microscopic colitis, but evidence now indicates that drug consumption may trigger underlying inflammatory factors in the colon of affected individuals, whereas others may exacerbate the diarrhea in patients with idiopathic microscopic colitis. Although many drugs, either alone or in combination, may cause diarrhea as an adverse effect, nonsteroidal anti-inflammatory drugs (NSAIDs) show a strong trend (P = 0.057) toward increasing the risk of collagenous colitis (CC), and rechallenge has been shown to cause recurrence of CC.

Antidepressant agents such as selective serotonin reuptake inhibitors (SSRIs) as a group increase the risk of CC, but, in this class of medications, sertraline alone significantly increases the risk of lymphocytic colitis (LC). Anecdotal reports of a large number of additional drugs have been associated with the onset of microscopic colitides. Among these, ranitidine (confirmed by rechallenge), aspirin, acarbose, ticlopidine, and proton pump inhibitors have high-level evidence of causality, whereas flutamide, simvastatin, carbamazepine and lisinopril have intermediate-level evidence.

Note the following:

• Some patients have had histologic findings consistent with both CC and LC at different points during the course of the disease. This raises the possibility that these may be different manifestations of the same medical condition or different points along a continuum.
• Many case reports describe patients with preexisting presumed autoimmune conditions, such as celiac sprue, psoriasis and rheumatoid arthritis, who subsequently are diagnosed with microscopic colitis. Patients diagnosed with LC also have had uveitis, idiopathic pulmonary fibrosis, juvenile diabetes mellitus, pernicious anemia, autoimmune thyroid disease, and idiopathic thrombocytopenic purpura. Approximately one third of patients with celiac disease have histologic findings consistent with microscopic colitis. For this reason microscopic colitis should be considered in patients diagnosed with celiac disease who have diarrhea that fails to respond to a gluten-free diet. Removal of gluten from the diet, however, is ineffective in treating microscopic colitis in the absence of celiac disease.
• Microscopic colitis and inflammatory bowel disease (Crohn disease or ulcerative colitis) have been diagnosed concurrently or sequentially in a small number of cases, with either diagnosis preceding the other. The two conditions may have similar histologic features, making the distinction between them difficult.
• LC was diagnosed in a patient who failed to respond to antibiotic therapy for diarrhea associated with Campylobacter jejuni infection in the stool. The infection theoretically might have stimulated an autoimmune response that resulted in LC.
• Increased mast cells in some LC patients might indicate that degranulation products could serve as chemoattractants for inflammatory cells.
• Although only 14% of patients with LC were smoking actively when diagnosed, 23% were former smokers, suggesting that smoking may actually be protective against the development of LC. Excessive production of nitric oxide by inducible nitric oxide synthase (iNOS) may result in cytotoxicity through the generation of peroxynitrite that is formed when nitric oxide reacts with superoxide. Corticosteroids inhibit iNOS biosynthesis in some cells, and newer inhibitors of this enzyme may have a future role in the management of LC and other forms of colitis. Nitric oxide diffuses into cells and can induce inflammatory cell damage by causing nitration reactions. Increased nitric oxide generation has been reported in CC. The absence of nitrotyrosine staining, which is an indicator of active inflammation typically seen in ulcerative colitis, may correspond to the lesser degree of inflammation in CC compared with ulcerative colitis.^[6]
• Diarrhea associated with CC has been proven to be due to reduced net Na⁺ and Cl^- absorption, with the thickened collagen layer serving as a diffusion barrier. The epithelium nearly completely loses its ability to absorb these ions. In fact, the shift in net chloride flux is to secretion rather than absorption.
• Secretion of vascular endothelial growth factor, a fibrosis-enhancing peptide, is noted to be increased by the mucosa in some patients with CC. This may play a part in the pathogenesis.
• In some patients with diabetes mellitus and diarrhea, evidence of an increase in the thickness of subepithelial collagen bands has been found. However, there was no correlation reported between collagen thickness, age, and the duration of diabetes.^[7]

• References

Epidemiology

United States statistics

The true incidence of microscopic colitis is not known. The disease has been increasingly diagnosed over the past 20 years, but it is still uncommon. A population-based study found the incidence of microscopic colitis to have increased significantly from 1.1 per 100,000 persons in the late 1980s to 19.6 per 100,000 persons by the end of 2001. The rates increased with age. Prevalence at the end of the study period was 103.0 per 100,000 persons (39.3 for collagenous colitis [CC] and 63.7 for lymphocytic colitis [LC]).

International statistics

Estimates of international incidence rates of CC range from 1.1 to 5.2 per 100,000 per year; those for LC range from 3.1 to 5.5 per 100,000 per year.

Sex- and age-related demographics

LC affects similar numbers of men and women, whereas CC is up to 20 times more frequent in women than in men.

Both conditions are observed more commonly in people older than 40 years, with peak incidence in the sixth and seventh decades of life, and the incidence of both conditions increases with age. Isolated cases have been reported in younger populations, including children.

• References

Prognosis

Approximately 20% of patients with microscopic colitis may have a spontaneous remission without specific therapy.

More than half of patients treated for lymphocytic colitis (LC) experienced resolution of symptoms after 6 months of treatment, while only 15% of patients had significant persistent symptoms.

In some patients, the diarrhea may wax and wane over many years; however, more than 80% of patients can expect diarrhea and histologic abnormalities to resolve within 3 years.

Rare cases with severe and protracted diarrhea have been fatal; these cases are likely thought to be due to epithelial membrane sloughing and resultant altered mucosal permeability.

Although some small studies have suggested otherwise, microscopic colitis (either collagenous colitis [CC] or LC) does not appear to increase the risk of colon cancer.

Mortality/morbidity

Morbidity is limited to the consequences of diarrhea and malabsorption, including metabolic abnormalities such as hypokalemia and dehydration, weight loss, fatigue, and vitamin deficiencies. This is not considered a life-threatening condition; however, profuse watery diarrhea may lead to severe dehydration and electrolyte abnormalities requiring intensive resuscitation.

• References

History & Physical Examination

History

A protracted course of nonbloody diarrhea occurs. Diarrhea may be of high volume; some patients may eliminate more than 1500 mL of fluid per day.

Symptoms may have been present from several months to 2-3 years before medical attention is sought and a diagnosis is made. In fact, perhaps half of patients proven to have microscopic colitis meet the Rome, Rome II, and Manning symptom-based criteria for diarrhea-predominant irritable bowel syndrome and may be so misdiagnosed until endoscopic biopsies of the colon are taken.

Less frequent complaints include abdominal cramping, fecal incontinence, and weight loss, although weight loss may be seen in 40% or more of patients with CC.

Incontinence is probably more a reflection of the advanced age of those individuals who are affected, and patients with this problem may do well if treated with antidiarrheal agents.

Physical Examination

Physical examination usually does not reveal any specific abnormalities. Severely affected individuals may show signs of dehydration, malnutrition, and weight loss.

• References

Laboratory Studies

Blood studies

Results usually are within the reference range, but anemia, hypokalemia, hypoalbuminemia, elevation of the erythrocyte sedimentation rate, or a combination of these findings may be present.

Human leukocyte antigen (HLA)-A1 is more frequent in patients with lymphocytic colitis (LC) (67%) than in controls (28%) or in patients with collagenous colitis (CC) (26%). HLA-B3 has not been reported in patients with LC. A frequency of 26% is expected in controls and also is observed in patients with CC.

The HLA patterns observed in other gastrointestinal conditions, such as Crohn disease, have not been found in LC.

Approximately 50% of patients with CC and those with LC may have circulating autoantibodies, especially rheumatoid factor (RF); antinuclear, antimitochondrial, antineutrophilic cytoplasmic antibodies (ANCA); and antiparietal cell, antithyroglobulin, and antimicrosomal antibodies.

Stool studies

Increased levels of the stool inflammatory markers (eg, eosinophil protein X [EPX], myeloperoxidase [MPO], tryptase] have been detected in stool from patients with CC, whereas the levels of these markers in patients with irritable bowel syndrome did not differ from healthy controls.^[12]

Some affected patients have a diminished ability to absorb water resulting from reduced colonic absorption of sodium and chloride. Chloride/bicarbonate exchange across the colonic mucosa also may be reduced.

Stool evaluation on occasion may show the presence of leukocytes. In these circumstances, the stool should be tested for enteric bacterial pathogens, ova and parasites, and C difficile.

A prolonged (24- to 72-h) stool collection occasionally may demonstrate steatorrhea in affected individuals. A finding of greater than 7 g of fecal fat excretion per 24 hours in an individual ingesting 100 g of fat per day usually is indicative of fat malabsorption, and, even if microscopic colitis is present, a diagnosis of concurrent sprue should be considered.

Other tests

Immunohistochemical studies of biopsies in LC and CC cases demonstrate abnormalities consistent with a mixed histocompatibility-restricted mechanism.^[13]

The excessive intraepithelial lymphocytes observed in LC are predominantly CD4⁺ T cells rather than CD8⁺.

• References

Imaging Studies and Procedures

Imaging Studies

Findings on plain abdominal radiograph, barium enemas, and computed tomography (CT) scans typically are normal or nonspecific and show no evidence of colonic mucosal damage or wall abnormality.

Procedures

Biopsies obtained by sigmoidoscopy or colonoscopy are necessary to diagnose LC or CC.^{[1, 5]}

If a colonoscopy is performed, biopsies should be taken from the rectosigmoid and possibly also from the right side of the colon. Approximately 95% of patients with microscopic colitis will have diagnostic left colon biopsies; but, if these biopsies are nondiagnostic at sigmoidoscopy in a patient in whom clinical suspicion remains high, total colonoscopy for right-sided biopsies may confirm the diagnosis.

In individuals with LC or CC, the mucosa appears normal endoscopically or occasionally mild mucosal edema may be seen. It does not show the more readily apparent changes of inflammatory bowel disease such as friability, ulceration, and pseudopolyps. Most patients have similar degrees of histologic abnormality in the right and left sides of the colon.

CC demonstrates a thick colonic subepithelial collagenous deposit, whereas LC reveals a pronounced intraepithelial lymphocytic inflammation in the absence of a thickened collagen band.^[1]

• References

Histologic Findings

Histologic features of microscopic colitis include the following:

• Surface epithelium shows a chronic inflammatory infiltrate of plasma cells, lymphocytes, and eosinophils in the lamina propria.
• Intraepithelial lymphocytosis, with greater than 20 lymphocytes per 100 epithelial cells, is pathognomonic of the diagnosis of lymphocytic colitis (LC) (see following image), although lesser numbers of intraepithelial lymphocytes may be present in some patients with LC.

  View Image

  Lymphocytic colitis (LC) showing marked chronic inflammatory cell infiltrate of the surface epithelium (on right) with preservation of crypt architect....

   

  

   
• Epithelial cell damage is demonstrated by cell flattening, subepithelial blebs, and denuded epithelium. Fixed specimens may show epithelial loss and detachment.
• Crypts may have minimal architectural distortion as in Crohn disease or ulcerative colitis. However, typically, evenly spaced parallel crypts of equal diameter are present.
• CC shares the histologic features of LC but additionally demonstrates a thickened subepithelial collagen layer (usually >10 µm) in the lamina propria, compared to a normal thickness of 5-7 µm.
• For comparison, a representative biopsy of CC is shown in the following image.

  View Image

  Collagenous colitis (CC) showing similar inflammatory cell infiltration as in lymphocytic colitis (LC), with the characteristically thickened subepith....

   

  

   
• In some biopsy specimens, the surface epithelium may be denuded or partially detached from the collagen layer, even simulating pseudomembranes in rare cases. This either may be artifactual or may be due to a defect in adherence to the subepithelial membrane.

• References

Approach Considerations

If colitis is refractory to continued medical therapy or if effective medication cannot be tolerated, colectomy or ileostomy might be the only effective therapy; however, this is seldom necessary.

• References

Medical Care

Evaluating treatment options is impaired by the fairly frequent occurrence of spontaneous remission of symptoms; however, positive placebo-controlled double-blind randomized trials of budesonide have been performed and reported in both lymphocytic colitis (LC) and collagenous colitis (CC).

Farrukh and Mayberry synthesized the work on the definition of microscopic colitis and the relationship between LC and CC, evaluated the international epidemiology and research on the etiology, and performed a systematic assessment of the available treatment options.^[14] Among their findings: LC and CC have common clinical symptoms but are well defined histopathologically; these conditions are predominantly idiopathic and typically have a benign clinical course, although serious complications and death may occur. They noted a peak incidence at age 60 to 70 years and a female preponderance. There may be a relationship with autoimmune disorders and exposure to nonsteroidal anti-inflammatory drugs (NSAIDs) and proton-pump inhibitors (PPIs). Unfortunately, treatment is primarily based upon anecdotal evidence and findings from limited budesonide clinical trials; there are no well-established long-term follow-up of these patients.^[14] Other reviewers have noted similar features of microscopic colitis.^{[15, 16]}

A trial of dietary restriction and avoidance of potentially aggravating drugs (particularly NSAIDs) may alleviate symptoms in some patients, but many will require medical therapy.

Treatment should be initiated with the least toxic regimen or medication, with stronger medication used only if the milder treatment fails. Generally, 4-6 weeks should be allowed before deeming a particular medication is ineffective in the treatment of CC or LC.

One possible treatment algorithm is as follows:

• First line: Loperamide (Imodium AD) or diphoxylate/atropine (Lomotil) for mild diarrhea.
• Second line: Bismuth subsalicylate, two or three 262 mg tab tid or qid for 1-2 months (effective in up to 90% of patients); mesalamine, 3 g/d for 8 wk; or cholestyramine (especially if bile acid malabsorption is documented), at a mean dose of 8 g/d in moderate disease.
• Third line: If patient's condition is still not responding or if a patient has a more clinically severe colitis, a 6-week course of budesonide at the lowest effective dosage (usually 9 mg each morning) or a 2-week course of high-dose prednisone (60-80 mg/d) before tapering can be prescribed. Longer courses of budesonide may be beneficial, and, while systemic adverse effects may occur, little or no adrenal suppression should be anticipated. Recurrences after discontinuation of budesonide usually respond to reinstitution of the same medication. Longer courses of prednisone (up to 2 mo before tapering) may be needed in some patients, but recurrence is common after its discontinuation.
• Fourth line: Some refractory cases may benefit from azathioprine (approximately 2 g/kg/d) or 6-mercaptopurine, but responses often take months to occur. Methotrexate can alternatively be used in this setting.

Miehlke et al evaluated the treatment of lymphocytic colitis with oral budesonide and found that at week 6, remission was documented by colonoscopy and histology in 86% of the budesonide group compared with 48% of those administered placebo (P = 0.01). Histologic remission was confirmed in 73% of those receiving budesonide and 31% of patients administered placebo (P = 0.03).^[17] Relapse during follow-up was evident in about 44% (15 patients), but 8 of those who had a relapse again had a response to budesonide. Clinical remission and improved histology is achieved in a majority of patients with lymphocytic colitis when treated with budesonide.^[17]

Miehlke et al also performed a phase 3, placebo-controlled, multicenter study to evaluate budesonide and mesalamine as short-term treatments for CC and noted that at week 8, a greater percentage of patients in the budesonide group were in clinical remission than in the placebo group: using the Hjortswang-Criteria, 80% of patients on budesonide and 44% of those on mesalamine achieved clinical remission compared with 37.8% of those given placebo. Budesonide was superior to mesalamine (P = .0035) and significantly improved stool consistency and mucosal histology, as well as relieved abdominal pain. The investigators concluded that short-term treatment of CC with oral budesonide (9 mg once daily) is safe and effective, whereas this was not true of short-term treatment with oral mesalamine (3 g once daily).^[18]

Patients who respond to treatment, but experience a recurrence, will often respond again to the same previously effective medication.

There is no evidence of a benefit from probiotics.

Diet and activity

Patients should avoid or eliminate possible secretagogues, such as caffeine, and, when appropriate, lactose-containing products. A low-fat diet is advisable if steatorrhea is documented.

Consultations

Consultation with a gastroenterologist often is needed to make the diagnosis and to work through the treatment algorithm.

• References

Medication Summary

Medication is indicated only if discontinuing dietary components or medications considered possibly responsible for the illness fail to alleviate the symptoms. As above, treatment is initiated with the least toxic effective agents. If a patient fails to respond to simple antidiarrheal drugs, anti-inflammatory or immunosuppressive medications may be required. Studies of budesonide in LC have shown an 86% response rate in symptoms and in histologic findings, with an 81% response rate in CC. A treatment algorithm is discussed above.

• References

Loperamide (Imodium AD)

• Dosing, Interactions, etc.

Clinical Context:  Poorly absorbable opiate that decreases colonic smooth muscle contraction and propulsive activity. Slows intestinal motility and delays colonic transit. Reduction of gastrointestinal secretion may contribute to the antidiarrheal effect. Well-tolerated and safe drug when taken in recommended dosages.

• References

Diphenoxylate hydrochloride/atropine sulfate (Lomotil)

• Dosing, Interactions, etc.

Clinical Context:  Antidiarrheal agent chemically related to narcotic analgesic meperidine. A subtherapeutic dose of anticholinergic atropine sulfate is added to discourage overdosage, in which case diphenoxylate may clinically mimic the effects of codeine.

Each tab of Lomotil (or 5 cc of elixir) contains 2.5 mg diphenoxylate hydrochloride and 0.025 mg atropine sulfate.

• References

Bismuth subsalicylate (Pepto-Bismol)

• Dosing, Interactions, etc.

Clinical Context:  Controls diarrhea by reducing fluid secretion into intestinal lumen, by binding bacterial toxins, or by acting as an antimicrobial agent.

• References

Class Summary

Antidiarrheal agents may be appropriate in mild cases of microscopic colitis.

• References

Cholestyramine (Questran, LoCholest)

• Dosing, Interactions, etc.

Clinical Context:  Absorbs bile salts in the intestine, resulting in an insoluble complex that is excreted in feces.

• References

Class Summary

Diarrhea in patients with LC and possible bile salt malabsorption may respond to exchange resins.

• References

Hyoscyamine (Levsin SL, Levbid, Symax, Cystospaz)

• Dosing, Interactions, etc.

Clinical Context:  Anticholinergic agent with limited and generally symptomatic utility in patients with colitis. Levsin or Levsin SL (sublingual) is dispensed as 0.125-mg tab, Cystospaz as 0.15-mg tab, and Levbid or Symax as 0.375-mg tabs.

• References

Class Summary

Some patients with cramping pain associated with diarrhea will respond to antispasmodic medication.

• References

Sulfasalazine (Azulfidine EN-tabs)

• Dosing, Interactions, etc.

Clinical Context:  Prodrug complexing the active component of 5-aminosalicylic acid (5-ASA) with an inactive sulfapyridine moiety to prevent systemic absorption in the upper gastrointestinal tract. In the colon, the diazo bond is cleaved by bacterial flora and active 5-ASA is released to function as a topical anti-inflammatory drug. Adverse effects typically are due to sulfapyridine rather than to 5-ASA.

• References

Mesalamine (Asacol, Pentasa)

• Dosing, Interactions, etc.

Clinical Context:  Controlled-released formulations of 5-ASA that cause fewer side effects than sulfasalazine due to the absence of the sulfa moiety. Pentasa is ethylcellulose-coated 5-ASA with an acrylic-based resin that dissolves in neutral or alkaline pH found in the terminal ileum or the colon. Dissolution of the coating of these tablets releases active 5-ASA where it can be topically active. These medications are more costly than sulfasalazine but typically better tolerated. Asacol is dispensed in 400-mg tabs. Pentasa is available in 250-mg tab.

• References

Class Summary

Drugs that reduce inflammatory changes at the level of the colonic wall may be needed in a subset of patients with colitis who fail to respond to antidiarrheal medication.

• References

Prednisone (Deltasone, Orasone, Sterapred)

• Dosing, Interactions, etc.

Clinical Context:  Inexpensive corticosteroid available in many strengths, which simplifies tapering schedules. Methylprednisolone (Medrol), dexamethasone (Decadron), or hydrocortisone can be used instead of prednisone. Dosage should be adjusted based on relative potencies.

• References

Budesonide (Entocort EC)

• Dosing, Interactions, etc.

Clinical Context:  Topical glucocorticoid delivered to the small bowel and ascending colon in a time-dependent manner. Active drug is coated with methylcellulose which dissolves at pH of 5.5 or greater, starting in the duodenum. Does not suppress the hypothalamus-pituitary-adrenal axis to a significant degree.

• References

Class Summary

Systemic anti-inflammatory agents that are readily absorbed from the gastrointestinal tract. Have a multitude of significant side effects when used over a prolonged period of time. Patients who fail to respond adequately to topical anti-inflammatory drugs may benefit from a course of corticosteroid therapy.

• References

Azathioprine (Imuran), 6-mercaptopurine (Purinethol)

• Dosing, Interactions, etc.

Clinical Context:  Azathioprine is an antimetabolite available in tablet form for oral administration or in 100-mg vials for IV injection and is an imidazolyl derivative of 6-mercaptopurine. It is cleaved in vivo to mercaptopurine. Both compounds are eliminated rapidly from blood and are oxidized or methylated in erythrocytes and liver. No azathioprine or mercaptopurine is detectable in urine 8 h after administration.

• References

Methotrexate (Folex, Rheumatrex)

• Dosing, Interactions, etc.

Clinical Context:  Previously known as amethopterin. Antimetabolite that inhibits dihydrofolic acid reductase. Also used in certain neoplastic diseases, severe psoriasis, and adult rheumatoid arthritis.

• References

Class Summary

Have been administered to a small number of patients with LC who have not responded to other medical therapy. Specific indications and recommended dosages have not been established yet.

• References