Truelove and Witts [5] were the first investigators to suggest the addition of corticosteroid use in the medical treatment of ulcerative colitis in 1955. Their preliminary findings in 1954 [14] included 210 patients from five different hospitals with chronic ulcerative colitis that would normally have received 6 weeks of treatment in the hospital. Of those patients, 109 were treated with oral cortisone (up to 100 mg per day) compared to placebo (n = 101) for 6 weeks. At the end of the study period, patients were classified into three categories: clinical remission, improved, and no change or worse. Clinical remission was defined as one or two nonbloody stools daily, weight gain, and without fever or tachycardia. Hemoglobin and ESR values were also either normal or returning to normal. The improved group included all intermediate cases, and the no change or worse group is self-explanatory. Cortisone-treated patients appeared to do better than placebo-treated patients and were more likely to be in clinical remission (41.3 % vs. 15.8 %) at the end of the 6-week study (p-value <0.001). These findings held particularly true in patients treated with cortisone on their first attack of disease but also true of disease relapse as well as all grades of disease severity. In a subset of patients, sigmoidoscopic or barium enema examinations were also assessed at the end of treatment and seemed to suggest that the cortisone-treated group did better than the control group; however, the numbers of patients in these groups were a small sample of the total patients and therefore only a preliminary conclusion [5].

In 1974, it was discovered that a 5-day intensive intravenous cortisone regimen for treatment of severe attacks similarly resulted in higher remission rates [6]. A total of 49 patients over a 5-year period received intravenous prednisolone 60 mg in divided doses for 5 days. Nearly 75 % (n = 36) of the patients experienced rapid improvement and were symptom-free after 5 days, and 92 % (n = 33) of these patients remained without symptoms during the next 6 weeks. The patients were followed on an average of 3 years, and approximately 50 % remained in remission (p-value <0.02). Factors that influenced the success of the intravenous regimen included patients who were treated during their first attack and those with radiological evidence of extensive or distal disease. The authors also concluded that complete failure to respond to intravenous therapy or deterioration during the 5-day course of treatment was an absolute indication for emergency surgery in patients with severe disease.

A number of subsequent studies have also demonstrated similar remission rates in patients treated with corticosteroid therapy. Furthermore, a recent systematic review [9] identified five randomized controlled trials (RCTs) [5, 15–18], which involved 445 patients and determined the efficacy and safety of corticosteroid therapy in ulcerative colitis. Of the 226 patients that received corticosteroids, 46 % achieved remission compared to 21 % of the 219 patients that were treated with placebo after 2–8 weeks. Corticosteroids induced remission in active UC with a number needed to treat (NNT) of 3 (95 % CI 2–9). When oral corticosteroids that were thought to act mainly topically and therefore poorly absorbed, such as oral fluticasone or beclomethasone, were excluded, there were three remaining trials [5, 16, 17] with a NNT of 2 (95 % CI 1.4–6) [19]. Similarly, intravenous corticosteroids were also successful at inducing remission [20], but given the route of administration and short duration therapy, it was not included in the meta-analysis.

Severity and extent of disease is based on both clinical and endoscopic findings and is characterized as mild, moderate, severe, as well as fulminant colitis [21]. Although the natural history of ulcerative colitis is variable, it is estimated that approximately 15 % of UC patients will develop severe symptoms that require hospitalization and intensive medical therapy [22]. Acute severe colitis is potentially a life-threatening condition that requires early appropriate medical therapy with a primary goal to induce remission and decrease the risk for colectomy. Indeed, corticosteroids have helped decrease the mortality risk associated with severe or fulminant colitis from 75 % in the 1930s [1] to less than 1 % in the twenty-first century [2]; however, since the introduction and widespread use of corticosteroids, there has been no significant change in colectomy rates over the last 30 years [2].

Once the decision to use corticosteroids has been made, there is unfortunately little evidence-based data available to indicate the optimal dose and route for corticosteroid administration in management of active colitis.

In one study of outpatients with moderate ulcerative colitis treated with corticosteroids (oral prednisone 20, 40, and 60 mg), remission was achieved in two-thirds of the patients who received 40 or 60 mg but only in one-third of those given 20 mg daily. In addition, the side effects were more frequent in patients that received 60 mg daily than 40 mg and with no significantly greater efficacy [29].

Another randomized trial performed on patients with active proctocolitis found no difference in response rate or side effects produced in those that received oral prednisolone at 10 mg four times a day or 40 mg as once per day dosing [30].

With the lack of evidence that doses higher than 60 mg per day result in significantly higher efficacy, we recommend that patients with moderate-to-severe active colitis that require corticosteroid use for active disease be treated with a course of prednisone 40–60 mg orally once per day. In addition to lack of evidence that higher doses result in higher efficacy, higher doses of corticosteroids result in significantly more frequent and severe adverse effects. Finally, another reason to use a single daily dosing regimen is that studies have shown that lowering the daily frequency of any given medication increases compliance rates. Thus, given the lack of evidence regarding improved efficacy over single daily dosing regimen, in all likelihood this dose regimen will increase compliance and hence increase the likelihood of efficacy.

As noted above, there are a significant percentage of UC patients whose disease is nonresponsive to corticosteroid therapy. Prior to diagnosing a UC patient as steroid-resistant or nonresponsive, a trial of corticosteroids is administered in intravenous form when there is a lack of response to oral treatment at 40–60 mg daily [31] or when there are systemic signs of toxicity, and the patient requires hospitalized care. This is based on the pharmacokinetics of oral versus parenteral corticosteroids. It has been demonstrated that peak plasma levels of oral corticosteroid absorption is delayed in patients with active colitis [32], and parenteral corticosteroid administration as a continuous infusion is associated with a higher plasma concentrations and less variability in plasma levels [33]. Hence, changing patients from oral to intravenous steroids will insure that the lack of response is not due to any issues with absorption of oral corticosteroids.

If there is no clinical improvement with the initial oral corticosteroid regimen after 5–7 days, or if there are systemic signs of toxicity or evidence of fulminant colitis, then the patient requires hospitalization for parenteral corticosteroid treatment, commonly methylprednisolone at a dose equivalent to 1 mg/kg per body weight. If significant clinical improvement is achieved following administration of parenteral treatment, then the intravenous corticosteroid therapy can be transitioned to oral form, and a similar taper schedule as outlined above can be followed [24].

While a continuous infusion results in less variability in plasma concentrations, there is no improved efficacy of continuous corticosteroid infusion compared to a single intravenous daily bolus. In a double-blind randomized trial of patients with severe ulcerative colitis that either received 1 mg/kg/day of methylprednisolone administered as continuous infusion or given as a bolus dose, no significant difference in the efficacy or safety profile is found [20].

In a systematic review [2] of 32 cohort and controlled clinical trials assessing the efficacy of corticosteroids in severe ulcerative colitis, 24 studies reported the dose administered of intravenous corticosteroids, standardized as methylprednisolone equivalent (using mean adult weight 70 kg). In this meta-regression analysis that controlled for disease severity, there was no correlation found between corticosteroid dose used and colectomy rate (R ² < 0.01, p-value 0.98). Further, the risk for colectomy did not decrease any further when corticosteroid doses beyond 60 mg daily were used.

In patients that have received systemic steroid therapy for active colitis and achieved clinical improvement, physicians should at this point implement a plan to discontinue corticosteroids since there is no benefit for its use in maintenance therapy and long-term use results in a higher risk of adverse effects [24]. As there is a very high rate of disease relapse after corticosteroids are stopped, discontinuation of corticosteroids should be done with a plan to implement a steroid-sparing maintenance medication.

There are no randomized trials that have studied an optimal corticosteroid taper schedule after clinical symptoms improve in a patient with active disease. We recommend that if oral prednisone 40–60 mg daily for moderate-to-severe active colitis is given and if there is significant clinical improvement, a taper dose schedule can then be initiated with 5–10 mg taper weekly until 20 mg daily dose is reached [24]. Then, tapering should proceed by 2.5–5.0 mg per week. In addition to concerns regarding relapse of disease, clinicians should be cautious of the development of symptoms secondary to adrenal insufficiency, especially in those patients that have been on corticosteroids for a prolonged time. Adrenal insufficiency may manifest up to 9 months after steroid cessation, particularly in times of stress [34].

It is estimated that one-third of patients hospitalized with severe ulcerative colitis will fail to respond to corticosteroid therapy and may require urgent colectomy [2]. The development of predictive measures for identifying patients that are nonrespondent to corticosteroid treatment has helped determine who may be suitable for second-line medical therapy (calcineurin inhibitors such as cyclosporine or antitumor necrosis factor antibodies such as infliximab) or colectomy. Since much of the morbidity and mortality associated with severe UC is related to delayed surgery, it is prudent to identify early which patients are likely to fail corticosteroid treatment and determine when to initiate other rescue medical therapy, so that if surgery is necessary, it is not inappropriately delayed [35, 36].

There are two predictive models that have helped determine which patients may need early colectomy. In one prospective study by Travis et al. [37], patients with severe colitis defined as 8 stools per day or 3–8 stools per day and an elevated C-reactive protein (CRP) of >45 mg/L on the third day of intravenous therapy had a positive predictive value (PPV) of 85 % for colectomy. Similarly, Lindgren et al. [36] developed a regression formula to predict the likelihood of medical failure (fulminant colitis index = number stool frequency/day + 0.14 × CRP mg/L) and found that a cutoff score greater than 8 at day 3 had a PPV of 72 % for colectomy.

More recently, Ho et al. [38] developed predictive factors of nonresponse to corticosteroid therapy and a risk score to help identify patients within the first 3 days of medical therapy for either early second-line medical therapy or early surgery. This was developed from a retrospective chart review of 167 patients with severe ulcerative colitis. The scoring system was based on a 0–9 point scale and included an assessment of mean stool frequency (<4, 4–6, 7–9, >9), presence of colonic dilation (>4 cm), and hypoalbuminemia (<30 g/L). Patients with a score of 0–1 had a low likelihood of medical therapy failure (11 % risk), and those with a score of 2–3 had an intermediate likelihood of failure to medical therapy (43 % risk). A score of ≥4 predicted a high likelihood of nonresponse to medical therapy failure with a sensitivity of 85 % and specificity of 75 % (AUC, area under the curve 0.88).

Clinical evaluation after 3 days of systemic corticosteroid therapy appears to be the best tool in assessing the short-term prognosis of active colitis in a patient [39]. These are some clinical parameters described that may be used when assessing for treatment response and consideration for second-line medical therapy or surgery. We recommend that if the patient has not responded after 3–5 days of intravenous corticosteroids, considerations be given toward the aforementioned options and importantly steroids be tapered as rapidly as feasible, as the patient will still develop the inherent side effects from the ongoing corticosteroid use without benefit.

The lifetime colectomy risk in severe colitis is estimated to be 30–35 % [2, 31]. Based on well-established clinical parameters, the general response to parenteral corticosteroid treatment in short-term studies (5–14 days) is reported between 45 and 80 % [24]. Patients are typically deemed steroid refractory if there is a lack of response to intensive intravenous corticosteroid treatment at adequate doses (1 mg/kg/day) for 7–10 days [40]. The prolongation of corticosteroid therapy beyond 10 days did not improve remission rates. In fact, longer corticosteroid treatment may result in more deleterious outcomes as surgery is delayed in the patient with severe colitis. Unfortunately and despite our current medical therapy, the short-term colectomy rate in severe colitis has not changed over the past three decades in a systematic review from 1974 to 2006 [2].

In a large population-based Olmsted County cohort [41], patients with ulcerative colitis that did not respond to corticosteroids in the short term (less than 30 days) had a 90 % colectomy rate. However, even among the patients that initially responded to corticosteroid treatment in the short term, only 49 % of the patients maintained remission without surgery or prolonged corticosteroid use over the following 1-year study period. The data also approximated that a quarter of patients with UC are steroid dependent at the end of 1 year.

These findings raise concern for the less than optimal long-term outcomes in some patients with active colitis that require a course of systemic corticosteroids and also highlight the high risk for colectomy in patients nonresponsive to corticosteroids. Other medical treatment modalities and surgery should be considered once it is determined that a patient is steroid refractory. These data also emphasize that even those patients that respond to corticosteroids are still at significant risk for colectomy. Thus, once a patient has had improvement in symptoms, while the patient is beginning to taper corticosteroids, he or she should also be placed on a long-term maintenance non-corticosteroid therapy.

The potential for adverse effects related to corticosteroids is both dose and duration of therapy dependent. Corticosteroids resulted in side effects in over 50 % of patients that were receiving high-dose steroid treatment and in 30 % of patients on prophylactic doses [42].