You are on page 1of 7

Dig Dis Sci (2010) 55:1059–1065

DOI 10.1007/s10620-010-1126-4


Cytomegalovirus Infection in Patients with Active Inflammatory
Bowel Disease
John J. Kim • Nicole Simpson • Nancy Klipfel
Renee DeBose • Nancy Barr • Loren Laine

Received: 26 August 2009 / Accepted: 11 January 2010 / Published online: 29 January 2010
Ó Springer Science+Business Media, LLC 2010

Background The reported prevalence of cytomegalovirus
(CMV) infection with active inflammatory bowel disease
(IBD) is highly variable, and whether CMV negatively
impacts the clinical course is controversial.
Aims The aim of this study was to determine the prevalence of CMV in patients with active ulcerative colitis
(UC) or Crohn’s disease (CD) and compare the course of
disease in patients with and without CMV.
Methods Consecutive patients with acute exacerbations
of active IBD colitis had immunohistochemistry staining
for CMV antigen performed on archived specimens. Clinical features were retrospectively reviewed.
Results Twelve (10%) of 122 UC patients had CMV, and
0/20 patients with CD had CMV. Of 12 UC patients with
CMV infection, seven were not taking steroids or immunosuppressives at their index flare. UC patients with CMV
were more likely to have leukocytosis (OR = 5.3, 95% CI
1.5–18.2), require hospitalization (OR = 4.9, 95% CI 1.2–
19.0), and be hospitalized C7 days (OR = 5.0, 95% CI
1.6–21.3) compared to patients without CMV. Of 12 UC
patients with CMV, ten (83%) were treated for their index
flare with steroids or 6-mercaptopurine. Only one patient

J. J. Kim  N. Simpson  R. DeBose  L. Laine
Department of Medicine, Keck School of Medicine University
of Southern California, Los Angeles, CA, USA
N. Klipfel  N. Barr
Department of Pathology, Keck School of Medicine University
of Southern California, Los Angeles, CA, USA
L. Laine (&)
GI Division, Keck School of Medicine, 2025 Zonal Ave,
Los Angeles, CA 90033, USA

(8%) was treated for CMV infection which occurred
14 months after index endoscopy. Over the 6 months after
the index flare, UC patients with CMV had a higher
frequency of IBD-related hospitalizations compared to
patients without CMV (50 vs. 24%, P = 0.021), but none
required surgery or died.
Conclusions The prevalence of CMV with active UC was
10%. Although CMV infection may be a marker of disease
severity, our results suggest it does not cause severe morbidity or mortality in a general population of patients with
a UC flare.
Keywords Cytomegalovirus 
Inflammatory bowel disease  Ulcerative colitis 
Crohn’s disease
CMV Cytomegalovirus
Inflammatory bowel disease
Ulcerative colitis
Crohn’s disease
H&E Hematoxylin and eosin

Although cytomegalovirus (CMV) infection is well recognized in immunocompromised hosts (e.g., HIV infection,
bone marrow transplant recipients) [1, 2], the reported
prevalence and pathogenicity of CMV infection in patients
with inflammatory bowel disease (IBD) are variable. CMV
infection is frequently identified in the colonic mucosa of


We hypothesize that a subset of patients previously thought to have IBD flares also had concurrent CMV infection. All colonic specimens for each case were reviewed for every patient. Medical records including inpatient and outpatient charts. the patients had histological reexamination with H&E. No other evaluation for CMV. The most severely inflamed specimens for each case were selected for IHC examination [10]. and after their IBD flare. CMV infection may be diagnosed if typical intranuclear inclusion bodies are seen on standard hematoxylin and eosin (H&E) stain. Patients who underwent endoscopy for screening or surveillance of dysplasia were excluded. the first documented procedure at our institution or the procedure that yielded the most histologically-active specimen on hematoxylin & eosin (H&E) stain was selected. Patient Population Statistics An Institutional Review Board waiver of consent was obtained prior to initiating this retrospective study. The clinical significance of finding CMV infection on colonic biopsy is controversial.4% when routine investigation for CMV infection was not done [3. and pathology reports were reviewed to confirm patients who met enrollment criteria and to characterize the patients’ clinical course before.0 g/dL. Anemia was defined as hemoglobin \10. 4]. In addition to IHC. Univariate analysis to determine patient and . For the most part. patients who 123 Descriptive statistics are reported as mean ± standard deviation. some patients with steroid refractory disease have concurrent CMV infection associated with high morbidity and mortality [12]. and endoscopic evidence of idiopathic IBD and active colitis were included. Since the clinical history and endoscopic findings of CMV colitis can resemble those of lBD. By utilizing IHC staining for CMV antigen. CMV infection was defined by the presence of CMV immediate early antigen by immunoperoxidase staining on colonic specimens.5 g/dL and elevated ESR was defined as [30 mm/h according to Truelove and Witts’ classification for severe ulcerative colitis [13].1060 patients with active IBD and often presents a diagnostic and therapeutic challenge. However. Leukocytosis was defined as white blood cell count [10. in these patients. such as viral culture. and whether CMV negatively impacts the clinical course is controversial. clinical improvement may be achieved in some patients only with antiviral therapy and steroid reduction [3]. CMV likely caused clinically significant disease.5–3. Potentially eligible patients were initially identified by searching the endoscopy records for patients who underwent colonoscopy or sigmoidoscopy for suspected IBD colitis based on symptoms of new onset or worsening hematochezia and/or diarrhea. we evaluated the pathology specimens from all IBD patients who underwent diagnostic endoscopy for IBD flare. CMV infection may be an innocent bystander. the sensitivity of H&E exam for CMV infection is low [5]. Methods Dig Dis Sci (2010) 55:1059–1065 underwent colonoscopy or sigmoidoscopy for suspected IBD flare at Los Angeles County ? University of Southern California Medical Center who were confirmed to have clinical. IBD-related hospitalization was defined by an IBD-related discharge diagnosis per ICD-9CM code and verified by the medical chart. Thus. resulting from secondary infection in an inflamed mucosa [11]. However. Currently IHC is considered the gold standard for detecting CMV [7]. Immunohistochemistry (IHC) exam performed on colon biopsy specimens is more sensitive than routine histologic exam and has excellent specificity [6]. Between January 2002 and January 2006. was performed. However. The reported prevalence of cytomegalovirus (CMV) infection with active IBD is highly variable and not well described in patients with mild to moderate disease. a level reported to be associated with medical treatment failure in ulcerative colitis [14]. during. examination of pathology specimens is crucial for making the correct diagnosis. The aim of this study was to determine the prevalence of colonic CMV infection in patients with active IBD and also to compare characteristics and clinical outcomes of active IBD patients with and without CMV infection. Definition and Study Endpoints The paraffin blocks from archived specimens of subjects in this study were reprocessed and stained with immunoperoxidase to determine the prevalence of CMV infection in IBD patients. The primary endpoints of the study were IBD-related hospitalizations and requirement of surgery within 6 months of the index IBD flare. laboratory test results. Surgery was defined by an operation of the intestine per ICD-9-CM code and verified by the operative record. When multiple procedures were performed on the same patient. endoscopic findings. Hypoalbuminemia was defined as albumin \3. histologic. this is not routinely done in general practice. Furthermore. the prevalence of colonic CMV has been reported to be as high as 14–57% in patients with IBD who had IHC examination [8–10].300/mm [3]. previous studies have only evaluated patients with steroid-refractory or steroiddependent IBD. Some investigators have estimated the prevalence of CMV infection in patients hospitalized for IBD at 0. In fact.

none of the patients with CMV infection had viral cytopathic changes. 22 (26%) were being treated with steroids. 142 patients underwent endoscopy for indication of suspected IBD flare and had endoscopic and histological evidence of IBD with active colitis.2–19. All tests were two-sided. and hospital length of stay [7 days (OR 5. P value \0. the remaining analysis was conducted only on UC patients.3–15.. including steroids. and comparisons were made using the generalized log-rank test.9 ± 7. UC Patients with CMV Infection Twelve of the 122 patients (10%) with UC had CMV infection (Fig. Chicago.6. Results Patient Population During the study period. a median of 16 specimens (range 1–65) was obtained for histologic evaluation during endoscopy. Treatment with immunosuppressive medications prior to endoscopy. Statistical analysis was performed using SPSS 16. TNF tumor necrosis factor. and 98 (69%) were Latino. 1). leukocytosis (odds ratio [OR] 4. None of the 20 patients with CD had CMV infection.0 (SPSS Inc. Due to the absence of CMV infection and the small number of patients with CD. On retrospective review of H&E examination. On univariate analysis (Table 2). and one (1%) was being treated with biologics at the time of endoscopy. MTX methotrexate. and biologics. hospitalization at time of index endoscopy (OR 4. 95% CI 1. Nine patients (75%) with CMV infection who were admitted for UC flare were discharged after clinical improvement with a mean length of stay of 9. he was admitted for severe UC and an endoscopy revealed active colitis with positive IHC staining for CMV.Dig Dis Sci (2010) 55:1059–1065 1061 disease characteristics associated with CMV infection in patients with active ulcerative colitis (UC) was conducted by utilizing a Pearson chi-square test. four (20%) were Table 1 Characteristics of patients at time of endoscopy for index flare Characteristic Overall n = 142 Ulcerative colitis n = 122 Crohn’s disease n = 20 Mean age (years) 40 ± 13 40 ± 13 39 ± 14 Male (%) 90 (63%) 78 (64%) 12 (60%) Latino 98 (69%) 84 (69%) 14 (70%) African American (%) 19 (13%) 14 (11%) 5 (25%) Race Caucasian 18 (13%) 17 (14%) 1 (5%) Asian (%) 6 (4%) 6 (5%) 0 (0%) Native American Indian (%) New diagnosis (%) 1 (1%) 68 (48%) 1 (1%) 61 (50%) 0 (0%) 7 (35%) Hospitalization (%) 61 (43%) 51 (42%) 10 (50%) 5-ASA (%) 52 (37%) 46 (38%) 6 (30%) Steroids (%) 30 (21%) 22 (18%) 8 (40%) 6-MP/AZA/MTX (%) 12 (8%) 9 (7%) 3 (15%) Anti-TNF (%) 3 (2%) 1 (1%) 2 (10%) being treated with immune-modulators. The mean age was 40 ± 13 years (Table 1). ten of 12 patients (83%) with CMV infection were treated with steroids and/or immune-modulators (Table 3) for presumptive UC flare. and in patients not on steroids and not hospitalized for their flare it was one out of 59 (2%). Kaplan–Meier estimates of the days without hospitalization and days to surgery were produced. AZA azathioprine. Sixty-one patients (43%) were hospitalized at the time of endoscopy.0).6) were associated with CMV infection in patients with active UC. IL). 1 Flow chart of 142 consecutive patients undergoing colonoscopy or sigmoidoscopy for acute inflammatory bowel disease (IBD) colitis flare 123 . 95% confidence interval [CI] 1. At the time of the index flare. was not associated with CMV infection. Only one out of 12 patients (patient 5) was later diagnosed with CMV infection. IHC exam was not performed on any of the biopsy specimens and none of the patients were diagnosed with CMV colitis based on H&E examination.2 days. He had clinical improvement with intravenous ganciclovir followed by oral valganciclovir for Active IBD N=142 1/02-1/06 Medications 6-MP 6-mercaptopurine. 90 (63%) were male. and two (10%) were being treated with biologics at the time of endoscopy. Of the 20 patients with CD.9. One hundred twenty-two patients (86%) had UC and 20 patients (14%) had CD. Consequently. whereas in patients not hospitalized for their index flare it was three of 71 (4%).8). For each index case.0. 5-ASA 5-aminosalicylic acid UC CD N=122 N=20 CMV+ CMV- CMV+ CMV- N=12 N=110 N=0 N=20 Fig. The CMV prevalence in UC patients not on steroids was eight of 100 patients (8%). nine (7%) were being treated with immunemodulators. eight (40%) were being treated with steroids. Multivariate analysis was not possible due to insufficient number of patients with CMV infection. and 68 (48%) were newly diagnosed with IBD.05 was considered significant. Of the 122 patients with UC. Fourteen months after the index endoscopy. 95% CI 1.3–19. immune-modulators.

9) ESR [30 mm/h 6 (50%) 32 (27%) 2. S Left-sided 5-ASA. either by causing CMV colitis directly or by exacerbating underlying IBD.4) 6MP/AZA 1 (8%) 8 (7%) 1. S steroid 123 . 5-ASA 0 104 49.9) Distal colitis 1 (8%) 42 (38%) 0. 5-ASA 18 Not hospitalized 46. 5-ASA 8 8 32.5 9 43/F 0 None Pancolitis S. Discussion The role of CMV in active IBD has been controversial. none of the patients died.0 10 57/M 0 None Pancolitis S. 6-MP 6-Mercaptopurine. Based on the association of CMV with severe steroid-refractory IBD.4 (0. other studies have challenged the pathogenicity of CMV in active IBD by demonstrating a lack of correlation of CMV and clinical severity of IBD [21]. 2).2) Pancolitits 4 (33%) 33 (30%) 1. Early studies have highlighted the association of CMV infection with severe active IBD and high doses of immunosuppressive medications. In addition.4 6 30/F 0.6 (0.2 5. 5-ASA 5-aminosalicylic acid. 6MP Left-sided 5-ASA.8–8.0) Hgb \10. he underwent total proctocolectomy due to steroid-refractory UC despite induction with infliximab.4 19/M 0 None Left-sided 5-ASA 2 15 41. 20 of 82 patients [24%].7 (0.9 UC ulcerative colitis.1 8 34/M 0.2) 1. 5-ASA 7 Not hospitalized 0.2) Left-sided colitis 7 (58%) 35 (32%) 3. M male.0–1.1062 Dig Dis Sci (2010) 55:1059–1065 Table 2 Baseline clinical.9) Anti-TNF 0 (0%) 1 (1%) 1. none of the UC patients with CMV infection required surgery while three of 80 patients (4%) without CMV infection underwent surgery.8–8. 15. F female. cyclosporine [18]. TNF tumor necrosis factor. it is also possible that CMV may be an ‘‘innocent bystander’’ detected in the Table 3 Summary of the 12 ulcerative colitis patients with cytomegalovirus (CMV) infection Patient Age/ gender Duration of UC (years) Medication Endoscopic involvement Treatment Length of stay (days) Days to subsequent hospitalization Follow-up (months) 1 65/M 8.2–2.0 (1. biochemical and endoscopic characteristics of patients with ulcerative colitis flare related to cytomegalovirus (CMV) status UC Patients Undergoing IBD-Related Hospitalization and Surgery Characteristic CMV? n = 12 CMVn = 110 OR (95% CI) Age [40 years 5 (42%) 54 (49%) 0.4 (0.4 5-ASA.5–7. Hgb hemoglobin.8–8.6 5-ASA.7–9.9 7 33/M 0 None Pancolitis S.0 (1.2 2. However. 5-ASA 6 18 9.300/mm3 7 (58%) 23 (21%) 5. S Distal S.8 (0.8 5-ASA Pancolitis S.3) OR odds ration.3) Steroids 4 (33%) 18 (16%) 2. Oral/topical 5-ASA 7 (58%) 40 (36%) 2. However.2 (0.6) WBC [10. 5-ASA 24 Not hospitalized 3 32/M 14. During a mean followup of 23 ± 19 months in patients with CMV infection and 29 ± 23 months in patients without CMV infection. 6-MP 0 Not hospitalized 13.3–4. one can hypothesize that CMV plays a role in flares of IBD.9 (1. CI confidence interval. 5-ASA 5-ASA 5 5 Not hospitalized 25 0.2 5-ASA. S 0 Not hospitalized 14.0 5-ASA.2–19.1) Hospitalization required 9 (75%) 42 (38%) 4.0 5 53/M 0.7 (0.9–3. AZA azathioprine. 5-ASA 5-aminosalicylic acid 10 months. S Left-sided S. Initial studies reported CMV inclusion bodies in colectomy specimens with fulminant or steroid-refractory UC [8. at 42 months from the index flare. 6-MP 6-Mercaptopurine.0) Hospitalization [7 days 5 (42%) 12 (11%) 5. ESR erythrocyte sedimentation rate. The colectomy specimen did not demonstrate evidence of reactivation of CMV on H&E stain or IHC exam. S Left-sided S.0 0 5-ASA None Left-sided Left-sided S.0 (0. 16].0 2 29/M 0.0) 0. treatment with steroids [17]. 5-ASA 14 Not hospitalized 4 52. WBC white blood cell count. anti-TNF agents [19].5–18.5) Male New diagnosis 9 (75%) 5 (42%) 69 (63%) 56 (51%) 1.4–5.0–1. In addition. UC patients with CMV infection required more frequent hospitalizations compared to patients without CMV infection (5 of 10 patients [50%] vs. However.6–21.5 5-ASA.021).2–2.2–11.3 (1.3 11 12 33/M 40/F 2.0) In patients with at least 6 months follow-up from the time of index endoscopy (Fig. P = 0. at 6 months from the time of index endoscopy.7 (0.5 g/dL 4 (33%) 29 (26%) Albumin \3.5 (0. and leukapharesis [20] has been associated with an increased risk of CMV infection.0 g/dL 6 (50%) 30 (27%) 2.2 (0.7 (0.

83% of the patients with CMV infection were treated with immunosuppressive medications and none received antiviral medications for the flare. Other authors have challenged the pathogenic role of CMV in patients with active IBD and supported the theory of CMV as an innocent bystander. a higher incidence of CMV in UC compared to CD is consistent with prior reports [8. CMV disappeared without treatment in most of the CMV reactivation-positive patients. none of the UC patients with CMV infection required surgery during the 6-month follow-up and none developed fulminant colitis or died despite the lack of immediate antiviral medication and the use of immunosuppressive medication. all the patients recovered from their IBD exacerbation without associated morbidity or mortality. some previous studies suggested that 1063 CMV infection may be associated with the development of IBD in a subset of patients [27–29]. Interestingly. 22]. is not generally of great clinical importance in our population. Although our study had a limited number of patients with CD. steroids were used twice as often in patients with CMV infection in our study. previously compared 25 moderate to severe UC patients with reactivation of CMV detected by serum antigenemia or real-time polymerase chain reaction assay and 45 moderate to severe UC patients without CMV reactivation or naive to CMV infection [21]. 95% CI 0. Our results suggest that CMV infection is associated with more severe UC. The increased severity of UC with CMV is consistent with either CMV being an innocent bystander in UC (with CMV more common in patients with more severe disease) or with CMV inducing more severe disease in patients with UC. CMV was not diagnosed at the time of index flare in any of our patients. univariate analysis showed significant associations of CMV infection with elevated white blood cell count. In contrast. serum IgM antibody. This is in stark contrast to some case series of CMV infection in IBD patients that have reported a colectomy rate of 64% and mortality rate of 44% [11. We found the prevalence of CMV infection in patients with active UC to be lower than previously reported in patients with acute severe colitis (21–38%) [20–24] and steroid-refractory colitis (32–36%) [17. Furthermore. or histological inclusion body. Nearly half of our patients were newly diagnosed with IBD. we found a higher rate of hospitalization in CMV-positive patients during the 6 months following their index flare. However. 31]. demonstrated that rapidly proliferating 123 . 2 Kaplan–Meier estimates of ulcerative colitis-related hospitalization rates for patients with CMV infection (n = 12) and without CMV infection (n = 110) background of colonic mucosa during active IBD. In our study. or death in the face of immunosuppressive therapy without antiviral therapy suggests that CMV.7. Our results are close to the 16% prevalence of CMV infection in a study from India that also included both outpatients and inpatients [26]. 30]. even if pathogenic. The absence of CMV-directed therapy allowed examination of the natural history of CMV infection in UC patients.2–2. a clear understanding of the role of CMV infection in active IBD is needed. Therefore. in that study diagnosis of CMV was made with serum CMV DNA. and hospitalization C7 days. 23. unlike previous studies that selected patients with more severe disease. However. and/or do not provide information on long-term clinical outcomes. the clinical outcomes including rates of remission and colectomy were not significantly different whether or not there was reactivation of CMV infection. Many previous studies were limited in size. thus comprising a large inception cohort. In contrast to some prior reports. This finding may be due to the inclusion of patients with mild to moderate disease activity in our study. With emerging therapies that increase immunosuppression. However. 12 of the 120 patients (10%) with UC and none of the 20 patients with CD had CMV infection. For example. In addition to increased disease severity in UC patients with CMV infection. were done in highly selected populations. 25].Dig Dis Sci (2010) 55:1059–1065 Fig. our study did not show a significant association of steroid or other immunosuppressive therapy with CMV infection [17. Matsuoka et al. the discrepancy in the prevalence of CMV infection may be even greater. Considering the higher sensitivity of IHC exam for CMV infection compared to the routine H&E exam utilized by some of the previous studies. A new diagnosis of UC was not associated with an increased incidence of CMV infection (OR 0. fulminant colitis. In this series. However. hospitalization at the time of index endoscopy. Pfau et al.3) in our study. raising the possibility of a type II error. the lack of surgery.

Pallares J.14:1373–1379.114:201–304. Iwao Y. Our results do not provide support for recommending routine IHC to detect CMV in IBD patients. required surgery. et al. Chacha SG. J Clin Virol.96:895–899. Keruly J. The Zidovudine Epidemiology Study Group. Cytomegalovirus infection in inflammatory bowel disease patients undergoing anti-TNFa therapy. et al. 9. [23] also suggested that a possible difference in clinical manifestations of CMV infection may be explained by the existence of pathogenic and non-pathogenic genotype.72:1253– 1256. Raffensperger Ec. Vega R. et al. Cytomegalovirus colitis complicating ulcerative colitis in the steroid-naive patient. despite the frequent use of immunosuppressive therapy and the absence of antiviral treatment. 2001. Dome`nech E. Another limitation is that not all patients received full colonoscopy. et al. TNF-a and IGN-c [32]. Cytomegalovirus is frequently reactivated and disappears without antiviral agents in . et al. British Medical Journal. Cytomegalovirus in colorectal cancer and idiopathic ulcerative colitis. J Gastro. 13. Cytomegalovirus infection in ulcerative colitis: a prospective. 21. Dig Dis Sci (2010) 55:1059–1065 2. Management of human cytomegalovirus infection and disease after allogeneic bone marrow transplantation. 6. 7. Intern Med. none of the patients developed fulminant colitis. In summary. Ward KN. Osaki R. Tung JK. 3.e. our data do not allow us to determine whether CMV detection and treatment would improve outcomes. Cytomegalovirus infection in steroid-refractory ulcerative colitis: a casecontrol study. et al. Furthermore. 2001. 1992. Thus. J Acquir Immune Defic Syndr Hum Retrovirol. Casellas F. colonic CMV may be more reflective of IBD disease severity rather than viral pathogenicity. Cytomegalovirus inclusions in patients with ulcerative colitis and toxic dilation requiring colonic resection. 2008. our results suggest it does not cause severe morbidity or mortality in a general population of patients with a UC flare. given the more frequent and longer hospitalizations at their index IBD flare.2:1041–1048. which precluded us from conducting a multivariate analysis to identify independent risk factors for CMV infection. Kochman ML. Infection by cytomegalovirus in patients with ulcerative colitis requiring colonic resection.94:1053–1056. et al. Okamoto H. Acute cytomegalovirus infection superimposed on corticosteroid-naı¨ve ulcerative colitis. Wu GD. Am J Gastroenterol.34:68–71. Furth EE. Beaugerie L. et al. 2008. Tsujikawa T. Hatakeyma K. Incidence and natural history of cytomegalovirus disease in patients with 123 20. surgery. advanced human immunodeficiency virus disease treated with zidovudine. Rev Inst Med Trop Sao Paulo. Nakamura T. Kambham N. 1977. if so. Refractory ulcerative colitis complicated by a cytomegaloviral infection requiring surgery: report of a case. 2006.. Andoh A. Stocchi R. if CMV therapy leads to significant benefit it would indicate that CMV does play a meaningful role in the clinical course of IBD. 15. or death in a general population of patients with a UC flare. Chakravarty BJ. Cooper HS. Dome`nech E. Caprilli R. et al. Conflict of interest statement None of the authors have conflict of interest or funding source to declare in connection with this study. Lashner B. Cytomegalovirus infection in patients with inflammatory bowel disease.96:2137–2142. et al.1064 cells in granulation tissue are susceptible to CMV infection [11]. We cannot rule out that detection and treatment of CMV might have decreased disease severity and the rate of repeat hospitalization. A limitation of our study is the relatively small number of patients with CMV infection. 2008. Truelove SC. 10. 1955. Papadakis KA. Ohta M. 1999. Criscuoli et al. References 1. Our study does suggest that routine IHC and treatment of CMV-positive patients is unlikely to decrease fulminant colitis. i. Kuwabara A. Inflamm Bowel Dis. et al. the prevalence of CMV infection in patients with active UC was 10%. 1989. Gastroenterology. Alcala MJ. Cortisone in ulcerative colitis: final report on a therapeutic trial. or died with their IBD flare. Gallant JE.88: 852–855. J Infect Dis. et al. However. UC patients with CMV infection may have more severe disease. 16. A randomized placebo-controlled trial of CMV therapy should provide the best assessment of the clinical importance of CMV in IBD. Outcome of cytomegalovirus infections in patients with inflammatory bowel disease. Surg Today. our results would underestimate CMV prevalence. Because patients did not receive anti-CMV therapy at the time of their flare. Cunha Ade A. Cytomegalovirus infection in severe ulcerative colitis patients undergoing continuous intravenous cyclosporine treatment in Japan. 1993.43:180–183. Haematologica. Maruyama K. Chien K. et al. Mariguela VC. Pfau P. Minami M. et al. Shintaku IP. although CMV infection in the setting of active UC may be a marker of disease severity. Fanin R.84:1517–1520. Cywiner-Golenzer C. and in situ hybridization for the detection of cytomegalovirus in gastrointestinal biopsies. Binder SW. Clinicopathologic characteristics of clinically relevant cytomegalovirus infection in inflammatory bowel disease. et al. 2007. 2008. 2004.28:365–373. D’Ovidi V. 17. Am J Gastroenterol. Ohkura T. Bertra´n X.50:83–87.84(1):71–79. Vernia P. Monfort L. It is possible that some patients may have isolated right sided involvement [33] and. 18. Predictors and the rate of medical treatment failure in ulcerative colitis. Nakamura R. Definition and diagnosis of cytomegalovirus colitis in patients infected by human immunodeficiency virus. 1999. Am J Gastroenterol. 14. 8.166:1223–1227. comparative study on prevalence and diagnostic strategy. CMV tropism for site of inflammation [3] and increased levels of procytokines. Ojanguren I. Witts LJ. Am J Gastroenterol. immunohistochemistry. 2000. 19. Moore RD. 12. Cytomegalovirus colitis complicating inflammatory bowel disease. Am J Surg Pathol. World J Gastroenterol. Kandiel A. Am J Gastroenterol.47:1341–1344. Mori T. Vega R. Matsuoka K. In conclusion. 11. Med Clin (Barc). Richman DD.14:423–429. Menacho M.42(10):823–829. Jonas L. 1997. Vij R.101: 2857–2865. Cartwright CA. Longacre T. 2004. 5. 2007. et al. A comparison of routine light microscopy. 4. Am J Gastroenterol. et al.13:754–760. may activate viral latency in the setting of IBD exacerbation.

et al. Prevalence of cytomegalovirus infection in severe refractory ulcerative and Crohn’s colitis.100:3154–3163. The pathogenicity of cytomegalovirus in inflammatory bowel disease: a systematic review and evidence-based recommendations for future research. Maher M. et al. Severe acute colitis associated with CMV: a prevalence study. 2009. Dig Dis Sci.102: 331–337. Ghoshal UC. J Clin Invest.Dig Dis Sci (2010) 55:1059–1065 22. Md Med J. and treatment. Nelson JA. Casa` A. et al. 25. Inflamm Bowel Dis. Cottone M. 1997. Sterringa G. Cytomegalovirus infection associated with onset of inflammatory bowel disease. Inflamm Bowel Dis. 33. Am J Gastroenterol.12: 879–884. Hofkin GA. Murray J.10:245–250. Orlando A. 32. Martorana G.54:2456–2462. 29. Ghoshal U. Orlando A. Kishore J. Inflamm Bowel Dis. 26. 2003. 2007. 2004. Frequent detection of cytomegalovirus in the intestine of patients with inflammatory bowel disease. Dimitroulia E. et al. Ting CD. Dis Colon Rectum.36:818–820. et al. 123 . 2006. 1065 28. Hommes DW. 23.53:1155–1160. Spanakis N. clinical significance and outcome. 2004. et al. Nassar M. ulcerative colitis patients. Iwamoto GK. et al. Orvar K. 24. 1993. Case report: recurrence of chronic ulcerative colitis induced by intercurrent cytomegalic virus infection. Infection with cytomegalovirus in patients with inflammatory bowel disease: prevalence.38:2307–2310. Matsui T. 2008. Pietrosi G. Carmen G.15:810–811. 30. 27.46:59–65.44:1047–1048. Soderberg-Naucler C. Wada Y. Cytomegalovirus disappearance after treatment for refractory ulcerative colitis in 2 patients treated with infliximab and 1 patient with leukapheresis. Mocciaro F. diagnosis. Dig Dis Sci. et al. Monick MM. Interferon-gamma and tumor necrosis factor-alpha specifically induced formation of cytomegalovirus-permissive monocyte-derived macrophages that are refractory to the antiviral activity of these cytokines. J Med Microbiol.85:1853–1857. Intractable ulcerative colitis caused by cyomegalovirus infection: a prospective study on prevalence. 2001. 2004. Modulation of interleukin 1 beta gene expression by immediate early genes of human cytomegalovirus. 1995. Konstantinidou AE. Clark BD. 1990. Fish KN. Dig Liver Dis. 31. Am J Gastroenterol. et al. J Clin Invest. Matake H.96:773–775. Criscuoli V. van Deventer SJ. Acute cytomegalovirus infection is a risk factor in refractory and complicated inflammatory bowel disease. Criscuoli V.