Abstract
BACKGROUND:
Severe ulcerative colitis is associated with significant morbidity. Multidetector computed tomography (MDCT) scans are frequently obtained upon hospital admission but the ability of radiographic findings to predict steroid failure is unknown.
AIM:
To identify MDCT features predictive of inpatient rescue in hospitalized UC
METHODS:
Patients hospitalized with UC who underwent a CT scan within 48 hours of hospitalization were retrospectively identified. Radiologists blinded to the outcome prospectively evaluated CT scans for presence of bowel wall thickening, stranding, and hyperenhancement as well as mural stratification, mesenteric hyperemia, and proximal dilation. Logistic regression adjusting for potential confounders was used to test the independent association between radiographic findings and need for rescue therapy.
RESULTS:
The study cohort included 74 patients. The mean age of the group was 45 years and twothirds (66%) were male. Twenty-eight (38%) patients required either inpatient medical rescue or colectomy. The mean number of positive radiographic findings was 4.4 (range 2–6) with a higher median number of findings in those who required rescue therapy (5 vs 4, p=0.03). Mural stratification was significantly more common among those who required rescue therapy (92% vs. 49%, p = 0.001). No other radiographic findings were independently associated with inpatient rescue. On multivariable analysis, mural stratification (OR 14.9, 95% CI 2.76–80.2) and number of positive findings (OR 2.10, 95% CI 1.06–4.16) remained independently predictive of need for rescue therapy.
CONCLUSIONS:
Mural stratification was highly predictive of steroid-refractoriness and need for medical or surgical rescue therapy in hospitalized UC.
Keywords: Mural Stratification, Ulcerative Colitis, Rescue Therapy
INTRODUCTION
Ulcerative colitis is a chronic inflammatory disease of the colorectum affecting nearly 1 million individuals in the United States and many more globally.1,2 About one in four of patients will develop acute severe ulcerative colitis (ASUC), which is characterized by bloody diarrhea and systemic signs or symptoms.3 While the mortality rate is low, morbidity including need for hospitalization, escalation of immunosuppressive therapy, and/or need for colectomy remains.4–8 Traditionally, assessment of disease severity has relied on clinical parameters and endoscopic findings while therapy response is assessed through dynamic predictors such as change in C-reactive protein (CRP) and improvement in symptoms.9 However, such predictors perform inadequately in prognosticating short-term outcomes in ASUC. One such important short-term outcome is failure to respond to intravenous steroids. Rather than relying on change in symptoms or laboratory parameters after 3–7 days of steroid treatment, identification of predictors of poor response to steroids at the outset at hospitalization can result in early initiation of effective medical or surgical rescue therapy and prevent protracted morbidity and disease-related complications.
Abdominal multidetector computed tomography (MDCT) scans are frequently performed in hospitalized UC patients to evaluate for complications of disease activity such as infection, perforation and/or toxic megacolon. However, the role of CT findings in stratifying severity of UC and/or predicting failure of intravenous steroid therapy is not well established. Several studies have attempted to quantify the utility of performing CT scans in UC with rates of clinically significant findings ranging from 8% to 69%.10–13 One study specifically evaluated the impact of CT scans on the clinical outcome of colectomy during same admission for ASUC.13 Here, authors found that CT directly influenced a surgical decision in 8% of cases. In the single case that prompted surgical intervention, a disease complication (pneumoperitoneum) influenced the decision. While the authors found that many CT findings were not different between those who required colectomy and those who did not, additional radiographic parameters such as bowel wall stranding were not evaluated.
The aim of this study was to identify radiographic feature(s) on an abdominal CT scan which predict need for inpatient rescue therapy as defined by the need for inpatient medical therapy (anti-tumor necrosis factor (TNF) or cyclosporine) or colectomy at hospital admission.
METHODS
Cohort
The study cohort included patients with ulcerative colitis who were hospitalized, initiated on intravenous corticosteroid therapy, and underwent an abdominal CT scan within 48 hours of admission. Compared to hospitalized ulcerative colitis patients who did not undergo CT scan during admission, hospitalized ulcerative colitis patients who did undergo CT scan had higher CRP levels (72 vs 38, p=0.001), increased need for inpatient colectomy (13% vs 4%, p=0.04), and longer lengths of stay (10 vs 6.8 days, p=0.0008). 14 However, there were no differences in albumin level, presence of pancolitis, need for inpatient medical rescue therapy, or need for inpatient colectomy or medical rescue. Exclusion criteria were a diagnosis of Crohn’s disease or indeterminate colitis, a prior history of surgical resection, and a CT scan performed more than 48 hours after the admission date for suspected disease- or therapy-related complications. This retrospective study was approved by the institutional review board of Partners Healthcare. Informed consent was waived by the Institutional Review Board.
Clinical Variables
Clinical variables were retrospectively collected from the electronic medical record. Variables of interest included age, gender, weight, height, body mass index (BMI), history of smoking, disease extent (Montreal classification15), endoscopic severity, medication use, length of hospitalization as well as erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), hemoglobin, and albumin values at hospitalization. Variables including number of bowel movements, bloody stools, temperature, and pulse were not collected.
Outcome
The primary outcome was need for inpatient medical rescue therapy (anti-tumor necrosis factor (TNF)α or cyclosporine) and/or colectomy during hospitalization.
Multidetector computed tomography (MDCT) Interpretation
All the patients in the study underwent abdomino-pelvic CT exam on 16-slice, 64-slice and 128-slice MDCT scanners after the administration of oral and intravenous contrast media. The oral contrast media (OCM) protocol included administration of positive or neutral oral contrast media. For the positive OCM, 900ml of either 2% barium (Readi-cat 2%, Bracco Diagnostics, Princeton, NJ) or 2% diatrizoate meglumine (Gastrograffin, Bracco Diagnostics, Princeton, NJ) was used. In patients receiving neutral OCM protocol, a total of 1350ml of low density barium (0.1% barium suspension, VoLumen, E-Z-Em; Lake Success, New York) was administered over a 60-minute period. Following OCM intake, contrast enhanced CT of the abdomen and pelvis was performed in the portal venous phase after injection of 80–120 ml of iodinated contrast media (Isovue 370, Bracco Diagnostics, Princeton, NJ) at a rate of 2.5 – 3ml/sec. The technical parameters for the scan included 100–120kVp, automA range 100–600 mA, 5mm collimation, pitch of 0.9–1.375 and 0.5–0.8 sec rotation time. Coronal and sagittal reformatted images of 2.5–3mm thickness were generated at the scanner console by the technologist following completion of axial CT data acquisition.
Two radiologists (AK, MG) independently reviewed all CT scans on a PACS workstation (AGFA, Richmond, VA). The readers were blinded to the technical parameters of the scans, clinical data, laboratory results and outcome of interest. The radiologists reviewed both the axial and coronal/sagittal reformatted datasets for assessment of MDCT findings of UC. Image analysis was performed using a predesigned template created to document findings such as bowel segments involved (small vs large bowel segments), extent of involvement, mural findings, extramural features and presence of complications. The number of involved colonic segments was calculated as the sum of all segments of the colon demonstrating colitis with a maximum total score of 6 (cecum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum). The mural findings included mural thickening (wall thickness >3mm), mural hyperenhancement (increased attenuation of the bowel wall when compared to that of normal appearing bowel) and mural stratification (identification of two or three layers within the bowel wall). (Figure 1) The extramural findings included peri-enteric inflammatory changes (surrounding peri-colonic/peri-enteric fat stranding and prominent vasa recta), stricture, abscess, fistula, bowel obstruction, free fluid/loculated fluid collections and lymphadenopathy. (Figure 2) The diagnosis of active disease was based on recognition of mural hyperenhancement, mural stratification, mural thickening, peri-enteric inflammatory changes and presence of complications such as abscess and toxic megacolon. Four CT scans could not be evaluated because of absence of intravenous contrast. Positive oral contrast in the lumen restricted evaluation of certain MDCT features in up to 13% of the remaining patients.
Figure 1.
Mural stratification and hyperenhancement in a patient hospitalized for ulcerative colitis.
Axial (a) and coronal reformatted (b) contrast enhanced CT images demonstrate mural thickening, stratification and hyperenhancement (arrow) of the descending and recto-sigmoid colon.
(c) Axial contrast enhanced CT image in a different patient with normal appearing colon without mural abnormalities.
Figure 2.
Peri-colonic abnormalities in a patient hospitalized with ulcerative colitis.
a. Contrast enhanced coronal and sagittal reformatted images demonstrating mild to moderate peri-enteric hyperemia and small adjacent lymph nodes.
Statistical Analysis
Baseline demographics and clinical characteristics were compared between those who required inpatient rescue therapy and those who did not. Continuous variables were reported as means and standard deviations or medians and interquartile ranges. Comparisons were made using the Student’s t test or Wilcoxon rank sum test, as appropriate. Categorical variables were reported as absolute numbers and percentages. Comparisons were made using the Chi-square or Fisher’s exact test, as appropriate. Variables with a high pre-test probability of affecting the outcome of interest or selected novel variables found to be significant in the univariate analysis were entered in a multivariable logistic regression model. Patients with missing CRP and albumin levels were assigned the median values of the cohort. A p-value < 0.05 indicated statistical significance. All analyses were performed using R Studio version 1.0.153.16
RESULTS
Study Cohort
The study cohort included 74 patients hospitalized for severe ulcerative colitis. Participants had a mean age of 45 years (range 14–86 years) and males comprised 66% of the cohort. Pancolitis was present in 43% of patients and the mean CRP was 67.1 mg/dL. Endoscopic data was available on 56 patients and severe disease was demonstrated in 61%. The mean hemoglobin was 11.4 g/dL and the mean albumin was 3.4 g/dL. Participants had a mean length of stay of 9 days (range 2–27 days). The median time to CT scan was 0 days (IQR 0–1) and CT scans were performed within the first day of admission in 89% of patients. MDCT features consistent with disease activity were common in the examined CT scans: mesenteric hyperemia (100%), bowel wall thickening (98.6%), bowel wall hyperenhancement (90.2%), bowel wall stranding (70%), mural stratification (65.6%) and proximal dilation (10%). The mean number of involved colon segments was 4.5 (range 2–6). The mean number of total MDCT features present among the study cohort was 4.4 (range 2–6). Any rescue therapy was required in 28 (37.8%) patients. Of these patients, 3 (10.7%) required both medical therapy and colectomy, 22 (78.6%) required medical therapy, and 3 (10.7%) required colectomy. Medical rescue therapy included cyclosporine in 4 patients (16%), infliximab in 14 patients (56%), adalimumab in 6 patients (24%), and golimumab in 1 patient (4%).
CT Findings and Rescue therapy
There was no significant difference in age, gender, BMI, smoking status, disease extent, and/or disease duration between those who required inpatient rescue and those that did not (Table 1). There was, however, a significantly higher use of prior biologic therapy in those that required inpatient rescue (50% vs 24%, p=0.04) with a trend toward increased endoscopic disease severity (64% vs 35%, p =0.05). Length of stay was also significantly longer in those that required inpatient rescue (12 vs 6 days, p < 0.001).
Table 1.
Demographic, clinical and serologic variables by need for any medical or surgical rescue therapy in hospitalized patients with acute severe ulcerative colitis
| Characteristic | Any Rescue (n=28) [N(%)] | No Rescue (n=46) [N(%)] | p-value |
|---|---|---|---|
| Age | 42 +/− 26 | 44 +/− 28 | 0.52†† |
| Gender | 0.63 | ||
| Male | 20 (71%) | 29 (63%) | |
| Female | 8 (29%) | 17 (37%) | |
| Body Mass Index (BMI) | 25 +/− 5 | 24 +/− 8 | 0.81†† |
| Smoking Status (n=21) | 0.33† | ||
| Never Smoked | 5 (50%) | 5 (45%) | |
| Smoker | 2 (20%) | 0 | |
| Past Smoker | 3 (30%) | 6 (55%) | |
| Disease duration (years) | 3 +/− 8 | 3 +/− 6 | 0.25†† |
| Pancolitis | 11 (39%) | 21 (46%) | 0.77 |
| Severe Endoscopic Disease (n=56) | 18 (64%) | 16 (35%) | 0.05 |
| Prior Mesalamines | 26 (93%) | 38 (83%) | 0.30† |
| Prior Immunomodulators | 15 (54%) | 16 (35%) | 0.18 |
| Prior Biologics | 14 (50%) | 11 (24%) | 0.04 |
| C-reactive protein (in mg/L) | 50 +/− 52 | 46 +/− 74 | 0.91†† |
| Erythrocyte Sedimentation Rate (in mm/hr) | 44 +/− 41 | 31 +/− 33 | 0.08†† |
| Albumin (in g/dL) | 3.4 +/− 0.5 | 3.5 +/− 0.7 | 0.38 |
| CRP/albumin ratio | 1.4 +/− 2 | 1.1 +/− 2.9 | 0.96†† |
| Hemoglobin (in g/dL) | 11 +/− 2 | 12 +/− 3 | 0.49 |
| Length of Stay (in days) | 12 +/− 9 | 6 +/− 4 | < 0.001†† |
Fishers exact test
Wilcoxon rank sum test
Individually, mural stratification was the only MDCT feature significantly associated with need for rescue therapy (Table 2). In those who required rescue, mural stratification was seen in 92% of patients compared to 49% of those who did not require rescue therapy, p=0.001 (Figure 3). This remained significant on multivariable analysis (OR 14.9, 95% CI 2.76–80.2) when controlling for age, gender, albumin, CRP, and pancolitis (Table 3). Furthermore, addition of mural stratification to endoscopic severity alone led to an increase in area under the curve, 0.65 to 0.79 (Figure 4). The parameters of mural stratification as a diagnostic test included a sensitivity of 0.92 (0.74, 0.99), a specificity of 0.51 (0.35, 0.68), a positive predictive value of 0.55 (0.39, 0.70), and a negative predictive value of 0.91 (0.71, 0.99). Presence of bowel wall hyperenhancement was not different by need for rescue therapy (96% vs. 86%, p=0.23). Bowel wall thickening was frequent in both those requiring inpatient rescue as well as those who did not require rescue with no significant difference between the groups (100% vs. 98%, p=1.00). Bowel wall stranding also demonstrated no significant difference by need for rescue therapy (78% vs. 65%, p=0.39). Mesenteric hyperemia was present in all examined CT scans (n=69) and therefore no comparative conclusions could be drawn. Finally, proximal dilation was not significantly different between those requiring and those not requiring rescue therapy (7% vs. 12%), p=0.70).
Table 2.
Presence of MDCT features by need for any medical or surgical rescue therapy in hospitalized patients with acute severe ulcerative colitis
| Characteristic | Any Rescue (n=28) [N(%)] | No Rescue (n=46) [N(%)] | p-value |
|---|---|---|---|
| Bowel Wall Thickening (n=70) | 1.00† | ||
| Yes | 27 (100%) | 42 (98%) | |
| No | 0 | 1 (2%) | |
| Bowel Wall Stranding (n=70) | 0.39 | ||
| Yes | 21 (78%) | 28 (65%) | |
| No | 6 (22%) | 15 (35%) | |
| Bowel Wall Hyperenhancement (n=61) | 0.23† | ||
| Yes | 25 (96%) | 30 (86%) | |
| No | 1 (4%) | 5 (14%) | |
| Mural Stratification (n=64) | 0.001 | ||
| Yes | 23 (92%) | 19 (49%) | |
| No | 2 (8%) | 20 (51%) | |
| Mesenteric Hyperemia (n=69) | N/A | ||
| Yes | 26 | 43 | |
| No | 0 | 0 | |
| Proximal Dilation (n=70) | 0.70† | ||
| Yes | 2 (7%) | 5 (12%) | |
| No | 25 (93%) | 38 (88%) | |
| Median Number of Positive Findings | 5 +/− 0 | 4 +/− 2 | 0.03†† |
| Median Number of Involved Colonic Segments | 4 +/− 1 | 4 +/− 1 | 0.78†† |
Fishers exact test
Figure 3.
Presence of mural stratification by need for rescue therapy in hospitalized ulcerative colitis patients
Table 3.
Multivariable analysis, predictors of inpatient medical or surgical rescue therapy in patients hospitalized with acute severe ulcerative colitis
| Variable | Odds Ratio | 95% Confidence Interval | p-value | |
|---|---|---|---|---|
| Model #1 | Mural Stratification | 14.9 | 2.76–80.2 | 0.002 |
| Age | 1.00 | 0.96–1.04 | 0.94 | |
| Male Gender | 1.73 | 0.47–6.40 | 0.41 | |
| Albumin | 0.52 | 0.17–1.62 | 0.26 | |
| CRP | 1.00 | 0.99–1.01 | 0.85 | |
| Pancolitis | 0.89 | 0.27–3.01 | 0.85 | |
| Model #2 | Number of Positive Radiographic Findings | 2.10 | 1.06–4.16 | 0.03 |
| Age | 0.99 | 0.96–1.03 | 0.65 | |
| Male Gender | 1.71 | 0.50–5.90 | 0.39 | |
| Albumin | 0.65 | 0.22–1.94 | 0.44 | |
| CRP | 1.00 | 0.99–1.01 | 0.90 | |
| Pancolitis | 0.78 | 0.25–2.50 | 0.68 |
CRP: C-reactive Protein
Figure 4.
Comparison in the receiver operating curves between endoscopic severity, mural stratification, and mural stratification with endoscopic severity in predicting need for inpatient medical or surgical rescue
The median number of positive MDCT findings was higher for those who required any rescue therapy (5 versus 4, p=0.03). The median number of involved colonic segments did not differ between the two groups, however (4 vs. 4, p=0.79). The predictive value of the number of positive MDCT predictors remained in multivariable analysis when controlling for age, gender, albumin, CRP, and pancolitis (OR 2.10, 95% CI 1.06–4.16, p=0.03) (Table 3). The clinical variables remained non-significant as in the prior multivariable analysis.
Inclusion of prior biologic therapy as a covariate in the model did not significantly change findings [mural stratification OR 12.04, 95% CI 2.14–67.66; number of positive findings OR 2.20, 95% CI 1.18–4.10]. In addition, excluding missing CRP or albumin (rather than assigning median values of the cohort) did not significantly change the findings [mural stratification OR 40.04, 95% CI 3.34–480.18; number of positive findings OR 2.77, 95% CI 1.30–5.91].
DISCUSSION
There are several predictive indices which have been developed to stratify the risk of steroid failure in ASUC. However, many rely on dynamic assessment of clinical status after 3–5 days of therapy and, consequently, can result in delay in initiation of rescue therapy thereby prolonging morbidity.9,17,18 Here, we show that MDCT predictors at admission can be important asset in stratifying patients at risk for worse outcomes. Specifically, we demonstrate that mural stratification was independently predictive of the need for inpatient rescue therapy, as defined by medical rescue or colectomy, in hospitalized UC patients. This finding highlights the unique role of MDCT in predicting steroid-refractoriness at hospital admission.
Mural stratification appears as a target like sign on imaging with the inner ring being composed of mucosa, lamina propria and muscularis mucosa, the middle ring composed of submucosa, and the outer ring composed of muscularis propria.19 Generally, IV contrast is necessary to identify mural stratification but severe submucosal edema may be seen on noncontrast CT.20 This finding is not specific to the inflammatory bowel diseases and can also been seen with infectious, ischemic, vascular and radiation etiologies.19,20 Mural stratification was seen in 65.6% of examined CT scans in this study cohort, which is consistent with previous reports of 61 to 70% in UC.19,21 Patel, et al evaluated the correlation between intestinal radiographic features and colonoscopic or clinical severity.21 In this study, mural stratification was found to demonstrate the strongest correlation with clinical severity of all radiographic predictors evaluated as well as the second strongest correlation with colonoscopic severity behind mucosal hyperenhancement. Our work not only confirms that mural stratification is significantly correlated with clinical severity in UC but also extends these findings to clinical outcomes in UC.
Radiologic assessment has long held an important role in the management of UC. However, imaging has traditionally been reserved for assessment of disease complications in ASUC rather than offering prognostic value. Standard CT can determine the anatomical extent of disease as well as the presence of bowel wall thickening, colonic dilation and/or colonic perforation. These features make CT particularly useful in patients with severe inflammatory disease such as in ASUC. In a study by Mege et al, investigators sought to determine the association of CT findings and the outcome of colectomy (versus medical therapy).22 They found no difference in disease location, wall enhancement, and wall thickness among other CT features between the two groups. Inclusion of all patients who received medical therapy regardless of corticosteroids only or need for inpatient biologic therapy may have limited the role in evaluating the impact of radiographic features on predicting a more clinically severe course, however. In our study, we group need for biologic therapy with need for colectomy to identify all those who failed first line medical therapy, which is a clinically impactful and relevant outcome.
There are several strengths of this study. First, it expands upon previous work evaluating the relationship between radiographic features and disease severity in UC. 21 In addition to replicating the correlation between mural stratification and disease severity, we demonstrated the independently predictive value of mural stratification on outcomes in hospitalized severe colitis patients, a group that is at high risk for poor outcomes.21 Mural stratification was found to be significantly predictive of the outcome of interest when controlling for important potential confounders such as age, albumin, and CRP. Therefore, mural stratification adds to existing independent clinical and biomarker predictors of ASUC outcomes. Furthermore, mural stratification was the only independent predictor identified in our cohort therefore it outperformed existing clinical variables including CRP and albumin. Next, radiologists were blinded to the outcome of interest thereby limiting potential bias in the study design. Finally, we isolated our study cohort to only those patients who had a CT scan within 48 hours of admission to limit the confounding effect of corticosteroid therapy on findings. Here, 89% of patients had CT scans performed within the first day of admission. Therefore, these findings are likely reflective of presenting disease severity and not influenced by subsequent therapy.
There are also several limitations of our work. First, the cohort was a relatively small sample size which may have limited the ability to detect more modest radiographic predictors of inpatient outcomes. However, this sample size is larger than previously used to detect correlation between CT features and disease severity.21 Next, we did not use dedicated colon CT imaging but performance of this has been poor, bowel preparation is not appropriate for hospitalized severe colitis patients, and it is not performed in this setting.23 Similarly, while a few studies have evaluated the role of MR in detecting acute inflammation, bowel preparation remains a concern and it is also not performed in this setting. Finally, there may have been a selection bias towards patients with more severe disease preferentially receiving MDCT scans. However, nearly 50% of our hospitalized UC cohort received CT scans suggesting that this is rather frequent diagnostic intervention among patients hospitalized with UC. 14
The results of this study do not suggest that routine CT scans should be performed in all hospitalized UC patients for prognosis.28 However, with increasing low dose radiation in CT scans, they can be valuable to use as additional tools in risk prognostication when available. 24–26 Furthermore, mural stratification was demonstrated to be the radiographic feature most predictive of UC inpatient outcomes and this feature can also be assessed by ultrasound.27 Thus, ultrasound could offer an alternative imaging option in patients with hospitalized UC. However, further studies would need to replicate these findings as well as determine differences in sensitivity and specificity between the two imaging modalities prior to implementation.
In conclusion, mural stratification identified on index CT scan during an admission for severe UC was highly predictive of the need for inpatient medical therapy or colectomy. There is a need for further trials evaluating early medical rescue therapy in those with clear predictors of steroid failure to prevent morbidity and obtain early control of disease activity. Incorporation of radiologic predictors into existing decision making algorithms may allow for improved risk stratification.
Acknowledgments
Sources of Funding: Ananthakrishnan is funded by the Crohn’s and Colitis Foundation and National Institutes of Health (R03 DK112909). This work is supported by the National Institutes of Health (P30 DK043351) to the Center for Study of Inflammatory Bowel Diseases.
Footnotes
Conflicts of Interest:
Cushing, Kordbacheh, Gee, and Kambadakone have no conflicts of interests to declare.
Ananthakrishnan has served on scientific advisory boards for Abbvie, Gilead, Takeda, and Merck.
REFERENCES
- 1.Shivashankar R, Tremaine WJ, Harmsen WS, Loftus EV Jr. Incidence and Prevalence of Crohn’s Disease and Ulcerative Colitis in Olmsted County, Minnesota From 1970 Through 2010. Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association. 2017;15(6):857–863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Molodecky NA, Soon IS, Rabi DM, et al. Increasing incidence and prevalence of the inflammatory bowel diseases with time, based on systematic review. Gastroenterology. 2012;142(1):46–54.e42; [DOI] [PubMed] [Google Scholar]
- 3.Dinesen LC, Walsh AJ, Protic MN, et al. The pattern and outcome of acute severe colitis. J Crohns Colitis. 2010;4(4):431–437. [DOI] [PubMed] [Google Scholar]
- 4.Lichtiger S, Present DH, Kornbluth A, et al. Cyclosporine in Severe Ulcerative Colitis Refractory to Steroid Therapy. New England Journal of Medicine. 1994;330(26):1841–1845. [DOI] [PubMed] [Google Scholar]
- 5.Jarnerot G, Hertervig E, Friis-Liby I, et al. Infliximab as rescue therapy in severe to moderately severe ulcerative colitis: a randomized, placebo-controlled study. Gastroenterology. 2005;128(7):1805–1811. [DOI] [PubMed] [Google Scholar]
- 6.Laharie D, Bourreille A, Branche J, et al. Ciclosporin versus infliximab in patients with severe ulcerative colitis refractory to intravenous steroids: a parallel, open-label randomised controlled trial. Lancet (London, England). 2012;380(9857):1909–1915. [DOI] [PubMed] [Google Scholar]
- 7.Turner D, Walsh CM, Steinhart AH, Griffiths AM. Response to corticosteroids in severe ulcerative colitis: a systematic review of the literature and a meta-regression. Clin Gastroenterol Hepatol. 2007;5(1):103–110. [DOI] [PubMed] [Google Scholar]
- 8.Kedia S, Ahuja V, Tandon R. Management of acute severe ulcerative colitis. World J Gastrointest Pathophysiol. 2014;5(4):579–588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Travis SP, Farrant JM, Ricketts C, et al. Predicting outcome in severe ulcerative colitis. Gut. 1996;38(6):905–910. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Gashin L, Villafuerte-Galvez J, Leffler DA, Obuch J, Cheifetz AS. Utility of CT in the emergency department in patients with ulcerative colitis. Inflamm Bowel Dis. 2015;21(4):793–800. [DOI] [PubMed] [Google Scholar]
- 11.Yarur AJ, Mandalia AB, Dauer RM, et al. Predictive factors for clinically actionable computed tomography findings in inflammatory bowel disease patients seen in the emergency department with acute gastrointestinal symptoms. J Crohns Colitis. 2014;8(6):504–512. [DOI] [PubMed] [Google Scholar]
- 12.Israeli E, Ying S, Henderson B, Mottola J, Strome T, Bernstein CN. The impact of abdominal computed tomography in a tertiary referral centre emergency department on the management of patients with inflammatory bowel disease. Aliment Pharmacol Ther. 2013;38(5):513–521. [DOI] [PubMed] [Google Scholar]
- 13.da Luz Moreira A, Vogel JD, Baker M, Mor I, Zhang R, Fazio V. Does CT influence the decision to perform colectomy in patients with severe ulcerative colitis? Journal of gastrointestinal surgery : official journal of the Society for Surgery of the Alimentary Tract. 2009;13(3):504–507. [DOI] [PubMed] [Google Scholar]
- 14.Gupta V, Rodrigues R, Nguyen D, et al. Adjuvant use of antibiotics with corticosteroids in inflammatory bowel disease exacerbations requiring hospitalisation: a retrospective cohort study and meta-analysis. Aliment Pharmacol Ther. 2016;43(1):52–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Satsangi J, Silverberg MS, Vermeire S, Colombel JF. The Montreal classification of inflammatory bowel disease: controversies, consensus, and implications. Gut. 2006;55(6):749–753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.RStudio: Integrated Development for R [computer program]. Version 1.0.153. Boston, MA: RStudio, Inc.; 2016. [Google Scholar]
- 17.Ho GT, Mowat C, Goddard CJ, et al. Predicting the outcome of severe ulcerative colitis: development of a novel risk score to aid early selection of patients for second-line medical therapy or surgery. Aliment Pharmacol Ther. 2004;19(10):1079–1087. [DOI] [PubMed] [Google Scholar]
- 18.Gibson DJ, Hartery K, Doherty J, et al. CRP/Albumin Ratio: An Early Predictor of Steroid Responsiveness in Acute Severe Ulcerative Colitis. J Clin Gastroenterol. 2018; 52(6)48–52. [DOI] [PubMed] [Google Scholar]
- 19.Gore RM, Balthazar EJ, Ghahremani GG, Miller FH. CT features of ulcerative colitis and Crohn’s disease. AJR American journal of roentgenology. 1996;167(1):3–15. [DOI] [PubMed] [Google Scholar]
- 20.Ahualli J The target sign: bowel wall. Radiology. 2005;234(2):549–550. [DOI] [PubMed] [Google Scholar]
- 21.Patel B, Mottola J, Sahni VA, et al. MDCT assessment of ulcerative colitis: radiologic analysis with clinical, endoscopic, and pathologic correlation. Abdominal imaging. 2012;37(1):61–69. [DOI] [PubMed] [Google Scholar]
- 22.Mege D, Monsinjon M, Zappa M, et al. Is abdominal CT useful for the management of patients with severe acute colitis complicating inflammatory bowel disease? A study in 54 consecutive patients. Colorectal Dis. 2017;19(4):O97–o102. [DOI] [PubMed] [Google Scholar]
- 23.Andersen K, Vogt C, Blondin D, et al. Multi-detector CT-colonography in inflammatory bowel disease: prospective analysis of CT-findings to high-resolution video colonoscopy. Eur J Radiol. 2006;58(1):140–146. [DOI] [PubMed] [Google Scholar]
- 24.Ordas I, Rimola J, Garcia-Bosch O, et al. Diagnostic accuracy of magnetic resonance colonography for the evaluation of disease activity and severity in ulcerative colitis: a prospective study. Gut. 2013;62(11):1566–1572. [DOI] [PubMed] [Google Scholar]
- 25.Hafeez R, Punwani S, Pendse D, et al. Derivation of a T2-weighted MRI total colonic inflammation score (TCIS) for assessment of patients with severe acute inflammatory colitis-a preliminary study. Eur Radiol. 2011;21(2):366–377. [DOI] [PubMed] [Google Scholar]
- 26.Yu LL, Yang HS, Zhang BT, et al. Diffusion-weighted magnetic resonance imaging without bowel preparation for detection of ulcerative colitis. World J Gastroenterol. 2015;21(33):9785–9792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kucharzik T, Kannengiesser K, Petersen F. The use of ultrasound in inflammatory bowel disease. Annals of gastroenterology. 2017;30(2):135–144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Swanson G, Behara R, Braun R, Keshavarzian A. Diagnostic medical radiation in inflammatory bowel disease: how to limit risk and maximize benefit. Inflamm Bowel Dis. 2013;19(11):2501–2508. [DOI] [PubMed] [Google Scholar]