You are here
Improved outcome of acute severe ulcerative colitis while using early predictors of corticosteroid failure and rescue therapies
Digestive and Liver Disease, In Press, Corrected Proof, Available online 5 March 2016, Available online 5 March 2016
Background and aim
Intravenous corticosteroids remain the first line therapy for severe attacks of ulcerative colitis although up to 30–40% of patients do not respond to treatment. The availability of alternative therapies to colectomy and the knowledge of early predictors of response to corticosteroids should have improved the clinical outcomes of patients with severe refractory ulcerative colitis. The aim of the study is to describe the current need, way of use, and efficacy of rescue therapies, as well as colectomy rates in patients with severe ulcerative colitis flares.
Between January 2005 and December 2011, all patients admitted in three referral centres for a severe ulcerative colitis flare who received intravenous corticosteroids were identified and clinical and biological data were accurately collected. Patients were followed-up until colectomy, death, or date of data collection.
Sixty-two flares were included. Initial efficacy of intravenous corticosteroids (mild activity or inactive disease without rescue treatment, at day 7 after starting intravenous corticosteroids) was achieved in 50% of flares, and rescue therapies were used in 27 episodes (43%). After a median follow-up of 18 months, the colectomy rate was 6.5%. Failed oral corticosteroids for the index flare were the only baseline feature that predicted the need for rescue therapy and colectomy.
There is a marked reduction in the colectomy rate and an increased use of medical rescue therapies as compared to historical series. Patients worsening while on oral corticosteroids for a moderate flare are at high risk of rescue therapy and colectomy and, therefore, should be directly treated with rescue therapies instead of attempting intravenous corticosteroids.
Keywords: Colectomy, Corticosteroids, Cyclosporine, Infliximab, Ulcerative colitis.
Ulcerative colitis (UC) is a chronic inflammatory condition of the colon characterized by a relapsing-remitting course. Up to 25% of UC patients will require hospital admission in their lifetime for intensive intravenous treatment because of severe flares . Intravenous corticosteroids (IV-CS) remain the treatment of choice for severe attacks of UC  though it has been repeatedly reported that up to 30–40% of patients do not respond to treatment , a clinical situation known as steroid-refractory UC. Although a universally accepted definition of steroid-refractoriness is still unavailable, the most accepted criterion is the lack of clinical improvement after 7–10 days of intravenous prednisolone at a dose of 1 mg/kg/day  or persistent active disease after treatment with oral prednisolone 0.75 mg/kg/day for 4 weeks . However, waiting for 1–4 weeks to define treatment refractoriness seems too long for patients with severe inflammatory activity because it may lead to clinical deterioration and the development of disease-related complications such as toxic megacolon, malnutrition, sepsis, or thromboembolic events.
Clinicians’ main concern in the management of severe attacks of UC is to identify those patients who are unlikely to respond to IV-CS therapy. Great efforts have been made to discern early predictors of CS failure in order to allow the identification of candidates for rescue therapies. Travis et al.  found that patients with more than 8 bowel movements per day or with 3–8 stool passages and serum C-reactive protein levels higher than 45 mg/L, on the third day of intensive intravenous treatment, had a 85% risk of colectomy. Similar results were subsequently reported , , and . In addition to clinical variables, the presence of severe endoscopic lesions has also been associated with a worse prognosis .
For many decades, the only treatment option in this clinical setting was colectomy. Twenty years ago, cyclosporine A (CsA) became the first alternative to colectomy for steroid-refractory UC , avoiding colectomy in up to 70% of patients as reported in a systematic review . A decade later, infliximab (IFX) became a new medical alternative for steroid-refractory UC, as shown in the first controlled study by Järnerot et al. . Both drugs were compared recently, reporting similar short-term  and long-term  efficacy.
The availability of both rescue therapies and predictors of poor response to CS should have improved the prognosis of patients with steroid-refractory UC, mainly in terms of the colectomy rate. Nevertheless, large series reporting the long-term outcomes of patients with severe UC flares in the era of biological agents are scarce. Therefore, the aims of the present study were to describe the need, type, the way of use, and efficacy of rescue therapies in this clinical setting, as well as the short and long-term colectomy rate in patients admitted for an acute severe UC flare.
2. Patients and methods
Between January 2005 and December 2011, all patients admitted for UC flares who received IV-CS were identified from the electronic records of three referral University hospitals in the Barcelona area. Patients were only included in the study if they had severe activity at the time treatment with IV-CS was started, according to the Montreal classification of UC severity (6 or more bloody stools per day together with at least one symptom or sign of severity – tachycardia, fever, anaemia, or raised ESR or CRP-) . Patients whose disease extent was limited to the rectum were excluded.
All patients were treated with intravenous methyl-prednisolone at a dose of 1 mg/kg/day. Rescue therapies (CsA or IFX) were prescribed at the discretion of treating physician. Patients receiving CsA were treated intravenously at a dose of 2–4 mg/kg/day (with dose adjustment according to therapeutic range of drug serum trough levels measured every 48–72 h). Patients achieving clinical remission with CsA were thereafter maintained on oral azathioprine 2.5 mg/kg/day. Patients receiving IFX were treated with 5 mg/kg infusions at weeks 0, 2 and 6; in case of clinical response, IFX 5 mg/kg every 8 weeks, azathioprine 2.5 mg/kg/day, or both, were prescribed maintenance therapy at the discretion of their treating physician. In cases of gastrointestinal intolerance to azathioprine, mercaptopurine was prescribed.
Collected data included epidemiological data (gender, age, smoking habit, family history of IBD), clinical (UC duration, previous CS courses, UC extent, failed oral CS for the index flare, extraintestinal manifestations, disease severity at day 3 and 7 of intravenous CS, steroid-dependency criteria and/or colectomy during follow-up), biological data (C-reactive protein at baseline – CRP-, 3, and 7 days), and treatment data (previous maintenance therapy, need for rescue therapy during the index flare, colectomy during admission) parameters. Finally, Truelove's severity criteria, in addition to the number of the daily bloody motions at the time CS treatment was started, was also recorded and converted into dichotomous variables (CRP > 35 mg/L, haemoglobin concentration < 10 g/L, presence of any systemic symptom – heart rate > 90 bpm, arterial hypotension, temperature > 37.5 °C, or extraintestinal manifestations).
For the purposes of the study, and in line with previous studies , we defined initial efficacy as mild activity or inactive disease, according to the Montreal severity score for UC , together with the lack of the need for rescue treatment, at day 7 after starting IV-CS. Steroid-dependency was defined as either the impossibility of reducing the corticosteroid dose to less than 10 mg/day after 3 months of treatment, or disease relapse within 3 months after steroid discontinuation . Patients were followed-up until colectomy, death, or date of data collection. The development of steroid-dependency or the need for colectomy during follow-up was also recorded.
The study was approved by the Institutional Review Board of the coordinating centre (Hospital Universitari Germans Trias i Pujol).
All statistical analyses were carried out in SPSS12.0 for Windows (SPSS Inc., Chicago, IL, USA). Data are expressed as median and interquartile range (IQR) or absolute and relative frequencies. Chi-square analysis and Student's t-test were performed for the between-group comparison of qualitative and quantitative variables, respectively. Variables reaching statistical significance (P-value ≤0.05) upon univariate analysis were included in the binary logistic regression analysis to identify predictors of the need for rescue therapy and colectomy. The cumulative probability of colectomy was calculated by the Kaplan–Meier method.
A total of 62 acute severe UC flares were included and their main clinical features are summarized in Table 1. As expected, most patients had extensive disease, were not current smokers, and had high CRP levels at the time of hospital admission. Interestingly, up to 23% of the index flares were treated with oral CS for an initially moderate flare but required hospital admission due to clinical worsening (failed oral CS). Finally, 33% of patients had been previously exposed to thiopurines at the time of the index flare.
|Female gender||30/62 (48)|
|Age (years)||37 (29–47)|
|Ulcerative colitis duration (months)||46 (0–96)|
|Ulcerative colitis extent|
|Prior corticosteroid courses||28/62 (45)|
|Failed oral corticosteroids for the index flare||14/62 (23)|
|Maintenance treatment with thiopurines||18/62 (33)|
|Baseline C-reactive protein (mg/L)||99 (55–163)|
3.1. Early outcome
As shown in Table 2, 77% of episodes showed clinical improvement after 3 days of intravenous CS therapy, with only mild to moderate activity according to the Montreal's classification. CRP levels at baseline and after 3 days of intravenous CS were available in 79% of flares. There was a decrease in CRP levels in 96% of cases, with 67% of flares showing CRP levels ≤45 mg/L at day 3. Nevertheless, only 50% of episodes met initial efficacy criteria as defined in the study.
|Clinical improvement at day 3 (mild or moderate activity)||48/62 (77)|
|Initial efficacy of corticosteroids (inactive or mild disease at day 7)||31/62 (50)|
|Decreased CRP levels at day 3 from baseline||47/49 (96)|
|CRP levels ≤45 mg/L at day 3||38/57 (67)|
|Rescue therapy||27 (43)|
|During the index admission||1|
In all, rescue therapies were used in 27 episodes (43%), twelve of which were initially treated with CsA and 15 with IFX, with 3 episodes receiving sequential rescue therapies (1 CsA after IFX and 2 IFX after CsA). Median time from intravenous CS to rescue therapy was 7 days (IQR 4–9 days), with no differences between those episodes initially rescued with CsA and those rescued with IFX. There were no significant differences in terms of adverse effects between patients who required rescue therapy and patients who did not (14% vs. 25%, respectively). Infections during hospital admission were also similar among patients who needed rescue therapies (1 urinary infection) and those who did not (1 pharyngitis, 1 gastrointestinal infection by Aeromonas spp., and 1 cystitis by Streptococcus viridans).
Only one patient (1.6%) required colectomy during hospital admission: a 29 year-old man, with a newly diagnosed left-sided UC. CMV infection was ruled out because of lack of response to intravenous CS and intravenous CsA was started as rescue therapy. One week later, colectomy was performed because of clinical worsening.
3.2. Long-term outcome
After a median follow-up of 18 months (IQR 7–36), four additional patients in the whole series underwent elective colectomy due to medical treatment failure during follow-up, all of them within the first 6 months after discharge. All patients undergoing colectomy required rescue treatment during the index flare (3 with IFX and one CsA followed by IFX).
Among those episodes not needing rescue therapy, 61% underwent maintenance therapy with mesalazine and 36% with azathioprine. At the end of follow-up, 37% of them met steroid-dependency criteria, with a non-significant trend towards a higher risk among those following azathioprine treatment (16% vs. 54%, P = 0.084).
3.3. Predictors of treatment outcomes
Many epidemiological, clinical and biological variables were included in the univariate analysis to identify those factors associated with poorer outcomes (Table 3). Failed oral CS for the index flare, the lack of systemic symptoms and presenting less than 2 additional severity criteria were initially associated with a higher need of rescue therapy, but only failed oral CS for the index flare remained as an independent predictor of the need for rescue therapy in the multivariate analysis. Interestingly, 3 out of the 4 patients failing oral CS for the index flare who did not need rescue therapy developed steroid-dependency, suggesting that patients with moderate flares who worsen while on oral CS, have a poorer prognosis (Fig. 1). Finally, both the need for rescue therapy and failed oral CS for index flares were also independent predictors of colectomy.
|Responders to corticosteroids
(n = 35)
|Rescue therapy (n = 27)||Univariate||Multivariate
OR (95% CI)
|Gender (% females)||43||56||0.32|
|Age (years)||36 (30–47)||39 (26–48)||0.2|
|Extensive ulcerative colitis (%)||74||63||0.33|
|Tobacco exposure (%)||12||26||0.08||–|
|Ulcerative colitis duration (months)||50 (0–108)||36 (0–65)||0.5|
|Prior exposure to thiopurines (%)||16||37||0.15|
|Failure to oral corticosteroids for the index flare (%)||11||37||0.017||7.8 (1.5–41)|
|No baseline systemic symptomsa (%)||10||30||0.031||–|
|Baseline haemoglobin (g/L)||119 (104–130)||115 (97–130)|
|Baseline C-reactive protein (mg/L)||94 (58–126)||98 (53–168)||0.43|
|C-reactive protein > 35 mg/L (%)||80||74||0.7|
|Haemoglobin < 10 g/L (%)||26||30||0.7|
|<2 baseline criteria of severityb (%)||97||89||0.01||–|
a Systemic symptoms are: temperature > 37.5 °C, or heart rate > 90 bpm.
b Haemoglobin level < 10.5 g/L, C-reactive protein > 35 mg/L, systemic symptoms.
The introduction of CS in the 50s dramatically reduced mortality from severe attacks of UC from 30–60% to less than 1%. Since then, CS has remained the first-line treatment for severe attacks of CU . However, the response rate to intravenous CS has not changed over the years, with up to 30–40% of patients showing an inadequate response to treatment . Early prognostic studies  and  searched for those parameters associated with the need for colectomy and led to the implementation of simple clinical and biological parameters for decision-making. Thus, the number of bowel movements, the persistence of rectal bleeding and/or persistent high CRP levels after three days of intravenous CS became a simple guide for the introduction of rescue therapies such as CsA or IFX.
Beyond the need for early colectomy, the Oxford group showed the relevance of achieving an early complete clinical response to CS in patients suffering from severe acute UC . In the long-term follow-up of their prospective cohort of patients with acute severe UC, Bojic et al. reported that patients who were not colectomized for the index flare but who showed only a partial clinical response after 7 days of intravenous CS had a significantly higher cumulative probability of colectomy in the follow-up as compared to those who achieved early clinical remission. Therefore, once we have efficient alternatives to colectomy in steroid-refractory UC, prediction must be focused on early clinical response instead of colectomy. Our group reported that the same clinical and biological risk factors described for colectomy also predicted early clinical remission . This is the reason why we decided to define initial efficacy of CS as mild or inactive disease after 7 days of intravenous CS.
Scarce data addressed the issue of whether the availability of both early predictors of CS failure and efficient rescue therapies had an impact on the short- and mid-term prognosis of severe UC flares. Using similar criteria for clinical response, two prospective  and  and two retrospective  and  series reported an initial efficacy of intravenous CS of 40–57%. It is noteworthy that, although the 3-day prediction rule was not used, in any of these series their figures were very close to our 50% rate of initial CS efficacy, suggesting that the initial efficacy of CS remains stable over time and that the implementation of the assessment of predictors of CS failure does not have an impact on it.
Conversely, our colectomy rate in the short (6.5%) and long term (8%) is considerably lower than previously reported. Historical series reported colectomy rates during admission for acute severe UC of 14–36% , , , , and , and up to 64% in the long-term . The early response rate to rescue therapy has been reported to be high in a recent RCT, with 85% of patients showing clinical response after 7 days , and this may impact on the colectomy rate during hospital admission. The short follow-up time in our series might influence our low colectomy rate in the long-term. Conversely, the coexistence of many factors may explain this dramatic fall in colectomy rates. Firstly, since the publication of the prediction study by Travis et al. , early assessment of response to CS has been strongly recommended , allowing for the earlier introduction of rescue therapies. Although it has not been evaluated specifically, it has been suggested that early introduction of rescue therapies might be associated with an increase in the therapeutic response to such drugs  and even with a decrease in the risk of disease-related complications (malnutrition, infections, thromboembolic events). In a Scottish nationwide study that showed a decrease in mortality following hospitalization for UC in recent years, the authors related this reduction solely to the identification of early predictors of response given the scarce use of rescue therapies in Scotland during the study period . To the best of our knowledge, this is the first reported cohort in which response to CS was systematically assessed at day 3, with a median interval from the beginning of CS therapy and introduction of rescue therapy of 7 days, with 25% of cases being rescued four days after starting CS. Secondly, the widespread use of rescue therapies in order to achieve a complete response to CS may have played an important role. Forty-three per cent of our patients received rescue therapy during admission; this is similar to the 47% and 46% recently reported in the United Kingdom audits performed in 2008 and 2010 respectively . It is interesting to note that our study covered the same period (2005–2011) in which a recent retrospective Canadian study demonstrated a marked decrease in both urgent and elective colectomy rates among UC inpatients, which the authors attributed to the increasing use of IFX for steroid-refractory UC . However, in our cohort, CsA was used as much as IFX with similar efficacy, suggesting that this may not be the only reason for such a reduction in colectomy rates. Moreover, the possibility of sequential treatment if no clinical response is achieved with a second-line therapy (switch to CsA from IFX or vice versa) may also impact on patients’ outcome as recently showed in a systematic review .
We also looked for predictors of response to intravenous CS. Former smokers were independently associated with an initial efficacy of intravenous CS. Although smoking cessation is a susceptibility risk factor for developing UC and also a risk factor for a worse outcome among patients with known UC  and , the relationship between smoking status and response to drug therapies in UC has been scarcely assessed. We did not observe a worse response to thiopurine treatment among ex-smokers with steroid-dependent UC .
One of the most interesting findings of our study is that failed oral CS for the index flare is an independent predictor of both the need for rescue therapy and colectomy. Only a few studies have evaluated the role of oral CS failure in the efficacy of intravenous CS. In patients with moderately active UC, Järnerot et al. showed that intravenous CS treatment was more effective in those patients initially treated through the intravenous route as compared to those who previously received oral CS treatment . We recently evaluated the efficacy of intravenous CS treatment for moderate flares of UC. We did not observe differences in the initial response to intravenous CS treatment between patients treated directly with CS and those for whom oral CS treatment had failed (persistent moderate activity), although the latter cases were significantly prone to developing steroid-dependency in the long-term . To the best of our knowledge, no study has assessed the role of oral CS failure in patients with acute severe UC but our results strongly suggest that patients whose clinical status worsens while on oral CS for a moderate flare, should be directly treated with rescue therapies instead of attempting intravenous CS treatment.
Contrarily to what was recently reported , additional Truelove's criteria of severity were not associated with a worse outcome in our patients. In fact, in the univariate analysis the absence of systemic manifestations as well as the presence of less than two additional Truelove's criteria of severity was associated with a poorer outcome in our series. Although we do not have an explanation for this association, similar findings were shown by Jess et al. in a study reporting the long-term outcomes of population-based UC cohorts in Denmark, in which the presence of systemic symptoms (weight loss and fever) at disease diagnosis predicted a subsequent quiescent course .
In summary, the efficacy of intravenous CS treatment for acute severe UC has not changed since its introduction more than 60 years ago, but patients’ prognosis has dramatically improved due to a marked decrease in short and long-term colectomy rates. This is probably a consequence of the early and widespread use of rescue therapies. The lack of endoscopic features is the main limitation of the present study. Severe endoscopic lesions demonstrated their prognostic value in two independent studies  and . However, the retrospective design of our study ruled out the collection of suitable endoscopic features.
Conflict of interest
M.M., E.C., E.G.-P, J.G., and E.D. served as speakers and/or received research grants from MSD and AbbVie.
E.D. received a research grant (Beca d’intensificació 2013) from the Catalonian Society of Gastroenterology (Societat Catalana de Digestologia) that partly supported this study.
-  L.C. Dinesen, A.J. Walsh, M.N. Protic, et al. The pattern and outcome of acute severe colitis. Journal of Crohn's and Colitis. 2010;4:431-437 Crossref
-  A. Dignass, J. Lindsay, A. Sturm, et al. Second European evidence-based consensus on the diagnosis and management of ulcerative colitis part 2: current management. Journal of Crohn's and Colitis. 2012;6:991-1030 Crossref
-  D. Turner, C.M. Walsh, A.H. Steinhart, et al. Response to corticosteroids in severe ulcerative colitis: a systematic review of the literature and meta-regression. Clinical Gastroenterology and Hepatology. 2007;5:103-110 Crossref
-  S. Meyers, P.K. Lerer, E.J. Feuer, et al. Predicting the outcome of corticoid therapy for acute ulcerative colitis. Results of a prospective, randomized, double-bind clinical trial. Journal of Clinical Gastroenterology. 1987;9:50-54 Crossref
-  S.P. Travis, J.M. Farrant, C. Ricketts, et al. Predicting outcome in severe ulcerative colitis. Gut. 1996;38:905-910 Crossref
-  S.C. Lindgren, L.M. Flood, A.F. Kilander, et al. Early predictors of glucocorticosteroid treatment failure in severe and moderately severe attacks of ulcerative colitis. European Journal of Gastroenterology and Hepatology. 1998;10:831-835
-  G.T. Ho, C. Mowat, C.J.R. Goddard, 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. Alimentary Pharmacology and Therapeutics. 2004;19:1079-1087 Crossref
-  M. Mañosa, E. Cabré, E. Garcia-Planella, et al. Decision tree for early introduction of rescue therapy in active ulcerative colitis treated with steroids. Inflammatory Bowel Diseases. 2011;17:2497-2502
-  F. Carbonnel, D. Gargouri, M. Lémann, et al. Predictive factors of outcome of intensive intravenous treatment for attacks of ulcerative colitis. Alimentary Pharmacology and Therapeutics. 2000;14:273-279 Crossref
-  S. Lichtiger, D.H. Present, A. Kornbluth, et al. Cyclosporine in severe ulcerative colitis refractory to steroid therapy. New England Journal of Medicine. 1994;330:1841-1845 Crossref
-  S. García-López, F. Gomollón-García, J. Pérez-Gisbert. Cyclosporine in the treatment of severe attack of ulcerative colitis: a systematic review. Gastroenterologia y Hepatologia. 2005;28:607-614
-  G. Järnerot, E. Hertervig, I. Friis-Liby, et al. Infliximab as rescue therapy in severe to moderately severe ulcerative colitis: a randomized, placebo-controlled study. Gastroenterology. 2005;128:1805-1811
-  D. Laharie, A. Bourreille, J. Branche, et al. Cyclosporin versus infliximab in patients with severe ulcerative colitis refractory to intravenous steroids: a parallel, open-label randomised controlled trial. Lancet. 2012;380:1909-1915 Crossref
-  J. Naves, J. Llaó, A. Ruiz-Cerulla, et al. Long term comparative efficacy of cyclosporine or infliximab-based strategies for the management of steroid refractory ulcerative colitis attacks. Inflammatory Bowel Diseases. 2014;20:1375-1381 Crossref
-  M. Silverberg, J. Satsangi, T. Ahmad, et al. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Canadian Journal of Gastroenterology. 2005;19(Suppl. A):5A-36A
-  A. Dignass, R. Eliakim, F. Magro, et al. Second European evidence-based consensus on the diagnosis and management of ulcerative colitis part 1: definitions and diagnosis. Journal of Crohn's and Colitis. 2012;6:965-990 Crossref
-  D. Bojic, Z. Radojicic, M.D. Nedeljkovic-Protic, et al. Long-term outcome after admission for acute severe ulcerative colitis in Oxford: the 1992–1993 cohort. Inflammatory Bowel Diseases. 2009;15:823-828 Crossref
-  L. Benazzato, R. D’Inca, F. Grigoletto, et al. Prognosis of severe attacks in ulcerative colitis: effect of intensive medical treatment. Digestive and Liver Disease. 2004;36:461-466 Crossref
-  M. Daperno, R. Sostegni, N. Scaglione, et al. Outcome of a conservative approach in severe ulcerative colitis. Digestive and Liver Disease. 2004;36:21-28 Crossref
-  J. Allison, L.J. Herrinton, L. Liu, et al. Natural history of severe ulcerative colitis in a community-based health plan. Clinical Gastroenterology and Hepatology. 2008;6:999-1003 Crossref
-  A. Gustavsson, J. Halfvarson, A. Magnuson, et al. Long-term colectomy rate after intensive intravenous corticosteroid therapy for ulcerative colitis prior to the immunosuppressive treatment era. American Journal of Gastroenterology. 2007;102:2513-2519 Crossref
-  R. Panaccione, P. Rutgeerts, W.J. Sandborn, et al. Review article: treatment algorithms to maximize remission and minimize corticosteroid dependence in patients with inflammatory bowel disease. Alimentary Pharmacology and Therapeutics. 2008;28:674-688 Crossref
-  N.T. Ventham, N.A. Kennedy, A. Duffy, et al. Comparison of mortality following hospitalisation for ulcerative colitis in Scotland between 1998–2000 and 2007–2009. Alimentary Pharmacology and Therapeutics. 2014;39:1387-1397 Crossref
-  R.W. Lynch, D. Lowe, A. Protheroe, et al. Outcomes of rescue therapy in acute severe ulcerative colitis: data from the United Kingdom inflammatory bowel disease audit. Alimentary Pharmacology and Therapeutics. 2013;38:935-945 Crossref
-  K.M. Reich, H.J. Chang, A. Rezaie, et al. The incidence rate of colectomy for medically refractory ulcerative colitis has declined in parallel with increasing anti-TNF use: a time-trend study. Alimentary Pharmacology and Therapeutics. 2014;40:629-638 Crossref
-  N. Narula, M. Fine, J.F. Colombel, et al. Sequential rescue therapy in severe ulcerative colitis: do the benefits outweigh the risks. Inflammatory Bowel Diseases. 2015;21:1683-1694 Crossref
-  S.S. Mahid, K.S. Minor, R.E. Soto, et al. Smoking and inflammatory bowel disease: a meta-analysis. Mayo Clinic Proceedings. 2006;81:1462-1471 Crossref
-  L. Beaugerie, N. Massot, F. Carbonnel, et al. Impact of cessation of smoking on the course of ulcerative colitis. American Journal of Gastroenterology. 2001;96:2113-2116 Crossref
-  E. Domènech, S. Carrión, E. Garcia-Planella, et al. Smoking status and response to thiopurines in steroid-dependent inflammatory bowel disease. Inflammatory Bowel Diseases. 2011;17:971-975
-  G. Järnerot, P. Rolny, H. Sandberg-Gertzen. Intensive intravenous treatment of ulcerative colitis. Gastroenterology. 1985;89:1005-1013
-  J. Llaó, J. Naves, A. Ruix-Cerulla, et al. Intravenous corticosteroids in moderately active ulcerative colitis refractory to oral corticosteroids. Journal of Crohn's and Colitis. 2014;8:1523-1528
-  T. Jess, L. Riis, I. Vind, et al. Changes in clinical characteristics, course, and prognosis of inflammatory bowel disease during the last 5 decades: a population-based study from Copenhagen, Denmark. Inflammatory Bowel Diseases. 2007;13:481-489 Crossref
-  C. Corte, N. Fernandopulle, A.M. Catuneanu, et al. Association between the ulcerative colitis endoscopic index of severity (UCEIS) and outcomes in acute severe ulcerative colitis. Journal of Crohn's and Colitis. 2015;9:376-381 Crossref
a Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona, Barcelona, Spain
b Hospital Universitari Germans Trias i Pujol, Centro de Investigaciones Biomédicas en Red de Enfermedades Hepáticas y Digestivas, Badalona, Spain
c Hospital Universitari de Bellvitge, IDIBELL, L’Hospitalet de Llobregat, Spain
⁎ Corresponding author at: Hospital Universitari Germans Trias i Pujol, Gastroenterology Department, Carretera del Canyet s/n, 08916 Badalona, Catalonia, Spain.
© 2016 Editrice Gastroenterologica Italiana S.r.l., Published by Elsevier B.V.