2026/05/18 by Per M. Hellström · 1 voice
Medicine · #Microscopic Colitis #Liver Diseases and Immunity #Spondyloarthritis Studies and Treatments
paper · doi:10.1111/joim.70099
openalex publication_date 2026/05/18 · openalex created_date 2026/05/19 · openalex updated_date 2026/06/03
Primary sclerosing cholangitis (PSC) and microscopic colitis (MC) are two separate conditions with distinctly different symptoms, where epidemiological data now show growing evidence that these two entities can co-occur more than expected by chance. In the current issue of Journal of Internal Medicine, Bergman and collaborators [1] describe a bidirectional epidemiological analysis employing first a cohort study showing an incidence rate of PSC and MC of 11.9 (confidence interval, 7.8–18.3, compared with controls 1.7; CI: 1.0–2.8). Then, an added case-control study showed PSC to be more common in MC patients with a frequency of 0.08% as compared to controls with 0.01%, yielding an odds ratio of 7.26; CI: 3.5–15.1. Taken together and based on background epidemiological data of more than 22,000 cases of MC versus more than 108,000 healthy people, the results show a possible association between the two diseases. This is in line with a previous systematic review of twelve cases [2]. Even though the absolute risks of co-occurrence are low, MC may have a diagnostic relevance for symptoms and signs of PSC. The Bergman paper [1] in this issue explores the co-occurrence of PSC and MC and forwards the question of what—if anything—connects the two conditions. It reviews proposed mechanisms, be it due to genetic susceptibility through linkage of HLA alleles between PSC and MC or changes of the microbiome with increased bacterial translocation across the intestinal mucosa, immune dysregulation or enterohepatic bile acid circulation as possible common disease mechanisms. Hence, this framework opens a possibility for clinicians and researchers to move from association to explanation. So, why is this pairing intriguing? Both conditions are uncommon, possibly immune-mediated and often chronic, making repeated co-occurrence clinically noticeable. PSC is strongly linked to inflammatory bowel disease (IBD) (classically ulcerative colitis), but MC sits outside the ‘classic IBD’ category, raising questions about shared pathways. Understanding possible shared drivers for MC and PSC could clarify risk stratification, surveillance strategies and treatment choices. MC is characterized by chronic, watery, non-bloody diarrhoea with macroscopically normal colonoscopy and characteristic histology; it includes lymphocytic colitis and collagenous colitis. It occurs mainly in postmenopausal women, but there are exceptions to this rule. The symptoms are usually managed with budesonide at a low dosage, along with removal of non-steroid anti-inflammatory drugs, proton pump inhibitors and sertraline, but may require additional treatment with cholestyramine, 5-aminosalicylic acid or immunomodulators. The clinically recognized association between active coeliac disease and MC, as well as activation of the immune responses by rectal instillation of Fraser's gluten fraction, speaks in favour of a general immune response directed towards gluten-sensitive targets throughout the gastrointestinal tract. PSC is a chronic cholestatic liver disease that appears mainly in middle-aged men. The disease is characterized by inflammation and fibrosis of intra- and extrahepatic bile ducts, with a variable course involving bouts of severe abdominal pain, repeated cholangitis and development of cholangiocarcinoma. From a common perspective, the two entities have little in common. A scientific goal could be distinguished by (a) shared causal mechanisms, (b) one condition contributing to the other and (c) non-causal explanations, including ascertainment bias and medications. To the present day, we have little knowledge of an association between MC and PSC. Two studies have focused on the issue: one case series of 12 patients [2] and a single case report, collectively 13 cases. The case series revealed that 75% of MC diagnoses occurred after PSC was diagnosed, as many cases were asymptomatic at the time of MC diagnosis, suggesting potential underdiagnosis of PSC. In line with this, in the single case report, the PSC was disclosed 9 months after MC. Thus, there is a possible time lag between the appearances of the two diagnoses. PSC alone, or in combination with IBD, is well characterized [3]. The MC–PSC combo is not systematically studied, so the apparent ‘rarity’ may reflect under-biopsying in chronic diarrhoea with normal macroscopic endoscopy of those with slightly elevated liver enzymes. Detection of an association between PSC and MC, the key methodological issue is to verify as follows: ascertainment bias (more colonoscopies), confounding by medications and competing diagnoses (bile acid diarrhoea). The clinical takeaway would be to treat co-occurrence as plausible, but not yet mechanistically defined. In order to do so, clear evidence is needed. The candidate connecting mechanisms between MC and PSC are tentative. Of many possibilities, shared autoimmunity, or immune skewing, may cause an overlap of autoimmune comorbidities in both PSC and MC, suggesting systemic immune predisposition. Here, the question remains whether there is a shared T-cell phenotype or cytokine signature between the two? Commensurate with this, a genetic susceptibility may prevail where HLA associations or other immune-related loci may create a shared risk background [4]; current evidence for PSC is stronger than for MC. The gut–liver axis and microbial products may be involved, as altered intestinal permeability or microbial composition could increase portal delivery of pathogen-associated molecules that shape biliary inflammation. Similarly, cholestatic disease and altered bile acid pools can change colonic bile acid exposure. Bile acids may also act as immunomodulatory signalling molecules (FXR/TGR5 pathways) that can influence mucosal inflammation. Medication effects and reverse causality: drugs associated with MC (e.g., certain PPIs/NSAIDs/SSRIs in observational studies) and treatments used in PSC/IBD could confound the association. Diagnostically, misattribution of diarrhoea can occur, and bile acid malabsorption, small intestinal bacterial overgrowth or pancreatic insufficiency may mimic or coexist with MC and must be evaluated separately. Even if shared mechanisms with the same alleles and bacterial translocation are at hand, we are yet to define the disease mechanisms in detail. Even with the same genotypes, the phenotypes seem to greatly differ, which complicates the explanation of the disease mechanism. Even though the present epidemiological article by Bergman et al. [1] reveals an association, a mechanistic principle would be compulsory to establish true evidence. Even though the pathophysiology of MC and IBD is largely unknown, there are similarities in potentially shared pathogenic mechanisms that can blur the distinction between the two diagnostic entities. This paves the way for mistakes in the clinically featured diagnostic process and diagnoses being reported to the health authorities. Earlier, three previous studies have examined cases of IBD misdiagnosed in cohorts of patients with MC. Kao and collaborators [5] used the Kaiser Permanente database, identifying 547 cases of MC between 1996 and 2005. Over an average follow-up of less than 3 years, none were reported with IBD. Similarly, Thörn et al. [6] reported 272 cases of MC with up to 4 years of follow-up, where only two cases of IBD occurred. Other studies have estimated 1.2%–2.1% of patients of MC also may have a diagnosis of IBD [7-9]. Finally, an epidemiological study by Khalili et al. investigated the possible misclassification of IBD as MC, arriving at a maximal misclassification rate of 4.1% [10]. Employing Poisson distributed 95% prediction intervals, a fair estimate would be 7 (2–10; 95% confidence intervals) cases of MC misdiagnosed as IBD. Hence, there is a substantial risk that the reported significance of patients sharing MC and PSC diagnoses should be at stake. Define a clear phenotype of the patient. Establish prospective cohorts of PSC patients with systematic diarrhoea assessment and protocol biopsies (even with normal mucosa) to estimate the true MC prevalence. Consider immune mapping with mucosal and peripheral immune profiling to identify shared inflammatory circuits that might resemble the PSC–UC patterns. Consider also the temporal relationships between symptoms and a clear diagnosis. As it seems, MC usually precedes PSC, which might give a clue to an underlying mechanism. To find evidence for disease connections, the search for mechanistic explanations through samples for paired stool microbiome/metabolome plus serum and bile acid profiling is necessary. Findings should be correlated with histology and cholestatic biomarkers. Standardized diagnosis of bile acid malabsorption and small intestinal bacterial overgrowth should be carried out to reduce misclassification. Time-stamped drug exposure and onset timing of diagnosis to test whether MC follows particular treatments and precedes the PSC diagnosis are necessary. Finally, overall reports of the number of biopsies, colon segments sampled and diagnostic criteria used for MC and PSC should be provided. MC and PSC can co-occur, but the literature is still thin and vulnerable to bias. The most plausible bridges run through the gut–liver axis, layered on immune susceptibility. Clinically, persistent watery diarrhoea in PSC warrants biopsies and a deliberate differential diagnosis of MC. The epidemiological results we read today are eye-openers to possible connections between different disease entities. Research progress depends on systematic case-finding, careful phenotyping and mechanistic sampling that can distinguish a ‘shared cause’ from a ‘shared context’. The author declares no conflicts of interest.