Secondary Logo

Journal Logo

Short Communications: Gastroenterology: Inflammatory Bowel Disease

Detection of Cytomegalovirus in Colonic Mucosa of Children With Inflammatory Bowel Disease: Inflammatory Bowel Disease

El-Matary, Wael; Stefanovici, Camelia; Van Caeseele, Paul; Deora, Vini; McCurdy, Jeff§

Author Information
Journal of Pediatric Gastroenterology and Nutrition: August 2018 - Volume 67 - Issue 2 - p 221-224
doi: 10.1097/MPG.0000000000001976


What Is Known

  • Cytomegalovirus is a double-stranded DNA virus and one of the most common infectious organisms associated with inflammatory bowel disease exacerbations, particularly in patients with acute severe corticosteroid-refractory ulcerative colitis.
  • The prevalence of cytomegalovirus disease in adult inflammatory bowel disease patients with corticosteroid-refractory colitis ranges from 5% to 42%, depending mainly on the method of detection.

What Is New

  • Children with moderate-to-severe colitis had a low prevalence of cytomegalovirus in their colonic mucosa.
  • Patients with colitis and positive colonic mucosa to cytomegalovirus responded well to escalated medical therapy without needing antiviral therapy or colectomy.
  • All children who had colectomy during the study period did not have cytomegalovirus detected in colonic mucosa.

Patients with inflammatory bowel disease (IBD) have inherent impaired natural killer (NK) cell activity and defects in mucosal immunity, which may create a favorable environment for opportunistic infections (1). Moreover, such patients are often treated with immunosuppressants that can be associated with reactivation of latent viruses (2). Cytomegalovirus (CMV) is one of the most common infectious organisms associated with IBD exacerbations, particularly in patients with corticosteroid-refractory ulcerative colitis (UC) (3). CMV infection, defined as detection of CMV antigens or antibodies in blood, is common, with an estimated prevalence ranges from 30% to 100% worldwide. The wide variation in prevalence is likely dependent upon age, ethnicity, and geographical location. CMV “disease,” on the other hand, infers symptomatic end-organ detection and most commonly occurs in immunocompromised patients such as patients with IBD leading to symptoms that may mimic those of IBD relapse (2). Whenever detected in colonic tissue, CMV is associated with increased risk of colectomy and mortality (4). The prevalence of CMV disease in IBD patients with corticosteroid-refractory colitis ranged from 5% to 42% depending on method of detection (3,5). The diagnosis of CMV presence in colonic tissue is likely to be more specific in denoting CMV's involvement in colitis. CMV can be detected by using hematoxylin and eosin (H&E) staining, immunohistochemistry (IHC), and tissue polymerase chain reaction (PCR) (2).

The risk of CMV disease in children with IBD is largely unknown. A recent pediatric study concluded a high prevalence of CMV colitis in children with UC who had colectomy within 12 months of presenting with acute severe colitis (6). A number of differences may exist between adult and pediatric IBD patients, such as disease duration, immunosuppression exposure, and the underlying seroprevalence of CMV. Therefore, the aim of this study was to determine the frequency of CMV in colonic mucosa of children with acute moderate-to-severe colitis and to determine the clinical management and outcome in these patients.


A retrospective observational study was performed between December 2011 and May 2017. Consecutive patients younger than 17 years old with moderate-to-severe IBD were assessed for eligibility. We included patients with: new diagnosis of IBD, established IBD with a clinical flare, and patients who underwent colectomy for medically refractory IBD. Given the predilection of CMV for colonic mucosa, we included only IBD patients with colonic involvement (UC, Crohn colitis [CC] [L2 using Paris Classification] (7), or IBD-U). Finally, patients required an endoscopic examination (colonoscopy or sigmoidoscopy) with at least 1 biopsy to evaluate for CMV by PCR and IHC.

The study protocol was approved by the health research ethics board of the University of Manitoba.

Our primary outcome was the frequency of CMV detection in our patient populations, stratified by IBD type, and whether the biopsy samples for CMV were collected at initial presentation or relapse. A diagnosis of CMV was considered positive if there was evidence of viral cytopathic effects histologically from H&E-stained biopsies, and/or 1 or more inclusion bodies by immunohistochemistry (IHC), and/or a positive tissue sample by PCR. A cutoff of 100 copies or more per milligram of colonic tissue was used to denote the presence of CMV by tissue PCR. Severe endoscopic colitis was decided by the physician's global assessment. Moderate-to-severe UC was defined as a pediatric ulcerative colitis activity index (PUCAI) of 35 points or more (8). Corticosteroid refractoriness was defined as no change in clinical disease activity indices after 3–5 days of 1–2 mg/kg (maximum 40–60 mg/day) of intravenous corticosteroids (9).

Medical records, endoscopy reports, and pathology reports were reviewed for all patients through pediatric gastroenterology, pathology, and microbiology databases. Data abstraction included: patient and IBD disease characteristics—age, sex, IBD type, disease extent, disease severity, disease phenotype, and hospitalization status; medication exposure—medication type, and response; endoscopic examination and CMV characteristics—type of endoscopic examination, details of initial colonoscopy, location and number of biopsies, and method of CMV diagnosis; management of CMV and clinical outcomes—use of antiviral medications, and requirement for surgery. See Supplemental Digital Content,


A total of 67 patient encounters from 58 patients met our inclusion criteria and were evaluated for CMV disease. Patients’ demographics, disease characteristics, and previous medications are summarized in Table 1.

Demographic and disease characteristics of the included children with inflammatory bowel disease

The mean duration of follow-up was 2.5 ± 2.2 years. At the time of the endoscopic examination, 40 (69%) participants were above the age of 10 years. Fifty-two (89.6%) patients were diagnosed with UC/IBD-U whereas the rest had CC. Thirty-six (62.1%) patients had preexisting IBD and presented with symptoms of relapse, whereas the rest (22 patients) were newly diagnosed with IBD. Twenty-two (38%) patients were hospitalized at the time of the endoscopic examination and 8 patients (13.8%) underwent colectomy during the study period for medical treatment-refractory UC had colonic biopsy samples from endoscopic examinations (before colectomy), and from colectomy specimens evaluated for CMV. Seven out of those 8 patients had PUCAI ≥65 points.

The most common presenting symptom was diarrhea with rectal bleeding (47 patients [81%]). Thirteen (25%) of the 52 patients with UC/IBD-U had corticosteroid-refractory colitis.

A total of 90 rectosigmoid biopsy samples were analyzed. Biopsies were obtained from the most inflamed part of the rectosigmoid colon as documented by the endoscopist. A median of 1 biopsy sample per patient was analyzed for CMV using PCR (IQR 1–2). One biopsy sample was obtained from 29 (60.4%) patients with UC whereas the rest had more than 1 biopsy. Three biopsy samples were obtained from the most inflamed part of each colectomy specimen. Seventy-nine (80.6%) biopsy samples were obtained from participants with UC with 50 (63.3%) of them from patients with known UC presenting with a relapse and the rest were obtained from newly diagnosed UC patients. Eleven biopsy samples were obtained from children with CC: 6 (55%) from known CC and 5 (45%) from newly diagnosed patients. Four (80%) out of 5 biopsy samples obtained from participants with IBD-U were obtained from known patients with IBD.

Four patients overall had a positive biopsy for CMV, all by PCR (Tables 2 and 3). None of these patients had PUCAI ≥65 points or evidence of viral cytopathic effect by H&E histology or immunohistochemistry. Each of the patients with CMV had UC/IBD-U (6.6% of patient encounters with UC/IBD-U), and no patients with CD had positive biopsy samples for CMV. Three of the 4 patients with CMV had preexisting IBD, and the remaining patient had detectable CMV at the time of diagnosis. Two of the 4 patients with CMV were refractory to corticosteroids, representing 15.4% of the patients with corticosteroid refractory disease (Table 2). As 9 patients had biopsies obtained at diagnosis followed by biopsies obtained during relapses, the total number of UC/IBD patient encounters were 61 whereas the actual number of patients with UC/IBD-U who were included in the study was 52.

Summary of frequency of cytomegalovirus in different subgroups of children with inflammatory bowel disease
Characteristics of patients with positive cytomegalovirus polymerase chain reaction at the time of taking the positive sigmoid biopsy sample

The mean PUCAI score for those with UC and positive CMV PCR was 52 (±5) compared with 42.5 (±13) in those with UC and negative CMV PCR (P < 0.01). All 4 patients responded to escalated medical therapy, without needing antiviral therapy, and none required colectomy over a median duration of follow-up of 1.1 year (IQR 1–1.6). None of those with CC had positive biopsies for CMV by any diagnostic technique. Furthermore, none of the biopsy samples obtained from those with PUCAI >65 or colectomy specimens from those who had colectomy were positive for CMV by H&E, IHC, or PCR.


Studies assessing the incidence of CMV reactivation in IBD have been limited largely to adult populations from tertiary-care centers. To our knowledge, this is the first report in pediatric patients with IBD. Here we report the frequency of CMV in symptomatic, pediatric IBD patients from a single tertiary-care center. Using PCR for diagnosing CMV in colonic tissue, we found a frequency of 6.6% in our overall UC/IBD-U cohort and 0% in our CD cohort. CMV disease was rare in patients with newly diagnosed IBD, and undetectable in patients with CD. Whenever patients were stratified by corticosteroid refractory disease, CMV was detectable in 15.4% of these encounters. None of the patients had detectable CMV by H&E histopathology or IHC.

The lower frequency of CMV detected in our pediatric population is not surprising, given that age is a known risk factor for CMV reactivation (10). This may be related to immunosenescence, disease duration, duration of immunosuppression exposure, and/or the underlying prevalence of latent CMV. The seroprevalence of CMV was recently reported in a cross-sectional cohort of pediatric IBD patients and was found to be between 22% and 38% depending on patients’ age (11). The lower seroprevalence of CMV in pediatric patients compared with adults is not surprising, given that disease acquisition occurs by transmission of bodily fluids, most commonly in teenage years and beyond when individuals become sexually active. Only 1 patient younger than the age of 11 in our study had detectable CMV, supporting the concept that CMV disease is rare in this age group. We also found that CMV reactivation was rare at the time of the initial presentation of IBD, and not detectable in patients with CD. These findings are consistent with the adult literature (3,5).

Noninvasive blood-based testing and/or serology are generally not performed at our center as primary tests to evaluate for CMV. A recent systematic review that examined the diagnostic accuracy of blood-based tests for predicting colonic CMV by IHC or tissue PCR in patients with IBD concluded that PCR of whole blood and the pp65 antigenemia assay should not replace an endoscopic examination for detecting CMV reactivation (12). This conclusion is in an agreement with several published guidelines that recommended tissue examination as the gold standard for diagnosing CMV reactivation (13–15). Serology for CMV, however, can identify individuals with latent virus who are at risk for CMV reactivation (10). Given the lower baseline, seroprevalence of CMV in pediatric patients compared with adults, fewer pediatric patients would be at risk of CMV reactivation. Therefore, CMV serology may serve as a particularly useful tool in this population to identify unexposed patients who do not require an endoscopic evaluation.

The clinical relevance of CMV in IBD exacerbations has been debated extensively (2). In the current study, none of our patients with detectable CMV were felt to have clinically relevant CMV disease, given the absence of virus by histopathology (H&E or IHC). Each of these patients responded to conventional immunosuppression, without anti-viral therapy, and none required colectomy within a median of 14-month follow-up. The only other study to date to assess the impact of CMV on clinical outcomes in pediatric patients with acute severe UC was a retrospective, multicenter study. Five of 15 (33%) patients with CMV required colectomy at 1 year compared with 5 of 41 (13%) age- and sex-matched controls with acute severe UC without detectable CMV. All but 1 patient with CMV received antiviral therapy, which precluded an assessment of the impact of antiviral therapy on colectomy rates (6). In a recent meta-analysis of adult studies, antiviral therapy was associated with an 80% reduction in colectomy risk in patients with corticosteroid refractory disease (16). Similar studies in pediatric patients are needed.

Our study has limitations including the retrospective design and small sample size. CMV assessments were left to the discretion of the individual clinicians and did not follow a standardized protocol with respect to biopsy location and number. However, more than 95% of procedures in our study were performed by a single experienced endoscopist.

In conclusion, we found a low prevalence of CMV in pediatric patients with moderate-to-severe colitis. CMV was most common in patients with corticosteroid-refractory UC, rare in newly diagnosed IBD, and absent in patients with CD. The low frequency of CMV in our study challenges current guidelines that recommend an assessment for CMV in all pediatric patients with acute severe UC refractory to corticosteroids. Future studies are required to confirm these findings, and to develop predictive models to determine high-risk patients that may benefit from an endoscopic evaluation for CMV.


1. Ginsburg CH, Dambrauskas JT, Ault KA, et al. Impaired natural killer cell activity in patients with inflammatory bowel disease: evidence for a qualitative defect. Gastroenterology 1983; 85:846–851.
2. Lawlor G, Moss AC. CMV in IBD: pathogen or bystander. Inflamm Bowel Dis 2010; 16:1620–1627.
3. Cottone M, Pietrosi G, Matorana G, et al. Prevalence of cytomegalovirus infection in severe refractory ulcerative colitis and Crohn's colitis. Am J Gastroenterol 2001; 96:773–775.
4. Kishore J. Infection with cytomegalovirus in patients with inflammatory bowel disease: prevalence, clinical significance and outcome. J Med Microbiol 2004; 53:1155–1160.
5. Takahashi Y, Tange T. Prevalence of cytomegalovirus infection in inflammatory bowel disease patients. Dis Colon Rectum 2004; 47:722–726.
6. Cohen S, Martinez-Vinson C, Aloi M, et al. Pediatric IBD Porto Group of ESPGHAN. CMV infection in pediatric severe ulcerative colitis- a multicenter study from the Pediatric IBD Porto group of ESPGHAN. Pediatr Infect Dis J 2018; 37:197–201.
7. Levine A, Griffiths A, Markowitz J, et al. Pediatric modification of the Montreal Classification for inflammatory bowel disease: the Paris Classification. Inflamm Bowel Dis 2011; 17:1314–1321.
8. Turner D, Otley AR, Mack D, et al. Development, validation, and evaluation of a pediatric ulcerative colitis activity index: a prospective multicenter study. Gastroenterology 2007; 133:423–432.
9. Assche GV, Vermiere S, Rutgeert P. Treatment of severe steroid refractory ulcerative colitis. World J Gastroenterol 2008; 14:5508–5511.
10. McCurdy JD, Jones A, Enders FT, et al. A model for identifying cytomegalovirus in patients with inflammatory bowel disease. Clin Gastroenterol Hepatol 2015; 13:131–137.
11. Hradsky O, Copova I, Zarubova K, et al. Seroprevalence of Epstein–Barr virus, cytomegalovirus, and polyomaviruses in children with inflammatory bowel disease. Dig Dis Sci 2015; 60:3399–3407.
12. Tandon P, James P, Cordiero E, et al. Diagnostic accuracy of blood-based tests and histopathology for cytomegalovirus reactivation in inflammatory bowel disease: a systematic review and meta-analysis. Inflamm Bowel Dis 2017; 23:551–560.
13. Turner D, Travis SPL, Griffiths AM, et al. European Crohn's and Colitis Organization; Porto IBD Working Group, European Society of Pediatric Gastroenterology, Hepatology, and Nutrition. Consensus for managing acute severe colitis in children: a systematic review and joint statement from ECCO, ESPGHAN and the Porto IBD Working Group of ESPGHAN. Am J Gastroenterol 2011; 106:574–588.
14. Bitton A, Buie D, Enns R, et al. Canadian Association of Gastroenterology Severe Ulcerative Colitis Consensus Group. Treatment of hospitalized adult patients with severe ulcerative colitis: Toronto consensus statements. Am J Gastroenterol 2012; 107:179–194.
15. Kornbluth A, Sacher DB. Ulcerative colitis practice guidelines in adults (update): American College of Gastroenterology, Practice Parameters Committee. Am J Gastroenterol 2004; 99:1371–1385.
16. Shukla T, Singh S, Loftus EV, et al. Anti-viral therapy in steroid-refractory ulcerative colitis with cytomegalovirus. Systematic review and meta-analysis. Inflamm Bowel Dis 2015; 21:2718–2725.

colitis; Crohn; cytomegalovirus; inflammatory bowel disease

Supplemental Digital Content

Copyright © 2018 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition