The clinical course of ulcerative colitis (UC) in children is unpredictable. Compared to patients with adult-onset disease, children with UC have more extensive disease and often a more severe course, manifest by higher rates of corticosteroid use and dependency and shorter time to surgery [1, 2].

Severe exacerbations of UC are common in both children and adults, and cause significant morbidity. These exacerbations can occur both at disease onset and as relapse in patients with established disease.

In 2008, the European Crohn’s and Colitis Organization (ECCO) defined acute severe colitis (ASC) in adults as an exacerbation with more than six bloody stools per day with at least one of the following: tachycardia (>90 b/min), temperature >37.8 °C, anemia (hemoglobin <10.5 g/dL), or an erythrocyte sedimentation rate (ESR) >30 mm/h [3]. In children, ASC is generally defined by a Pediatric Ulcerative Colitis Activity Index (PUCAI) score ≥65 [4], a cutoff that has been validated in independent cohorts and has predictive value with regard to response to intravenous corticosteroid (IVCS) therapy [5, 6] (see Chap. 46, Appendix C-3 for more details regarding PUCAI scoring). In adults, fulminant colitis has been defined by >10 stools per day with continuous bleeding, abdominal tenderness and distension, systemic toxic symptoms such as fever and anorexia, and blood transfusion requirement; this can progress to toxic megacolon with severe colonic distension (>6 cm), hypotension, altered mental status, and high mortality [7]. While colonic dilation is a hallmark of current or impending toxic megacolon (TMC), precise criteria for TMC in children have not been established. One study showed that in children ≥10 years of age, a transverse colon diameter ≥5.6 cm was suggestive of TMC [8], while in children younger than 10 years of age, a diameter >4 cm is concerning for toxic megacolon [9].

The frequency of ASC in children with UC is not fully known, but it is suggested that rates are as or even higher than the rates in adults. For example, over a 3-year period, in the greater Toronto area, it was estimated that 28% of all children with UC developed a severe exacerbation requiring hospitalization for intravenous corticosteroids before the age of 15 [10]. The remainder of this chapter addresses the management and ASC in children.

ASC is a serious and potentially life-threatening exacerbation of pediatric UC. As such, care for patients with ASC should be in the hospital setting so that frequent monitoring of clinical status, disease progression, and potential complications can take place. The goals of management are medical stabilization, treating exacerbating factors such as certain infections, and implementing a stepwise active treatment approach typically beginning with intravenous corticosteroids (IVCS) in order to control gastrointestinal hemorrhage while avoiding/limiting complications from the disease and/or therapy. Response to therapy should be frequently reassessed by a multidisciplinary team of providers, including, in many cases, surgeons with experience in IBD, in order to help guide plans for subsequent treatment.

The initial management of ASC includes a complete history and physical examination, beginning with assessment of vital signs and general appearance which may reveal signs of systemic toxicity such as hypotension, fever, significant tachycardia, or altered mental status. Abdominal tenderness should be assessed, keeping in mind that tenderness and even colonic perforation and peritoneal signs may be masked in patients on high-dose corticosteroids. The absence of bowel sounds is an ominous prognostic indicator. Frequent reassessment is necessary as progression to fulminant disease may be rapid. A PUCAI score should be calculated at the onset of symptoms and then daily during the exacerbation until improvement and disposition. A PUCAI score over 65 correlates with severe disease. This validated scoring system not only gives the provider an idea of the general well-being of the child, but also predicts response to IVCS and helps guide the timing of subsequent “rescue” therapy [6].

Hospitalized patients with ASC should have intravenous access and be fluid-resuscitated to assure adequate hydration. Laboratory studies including a complete blood count, serum electrolytes, albumin, ESR, and CRP should be obtained and repeated frequently. Despite the lack of randomized controlled clinical trials to provide evidence-based guidance for optimal therapy, expert opinion suggests that mucosal healing is best achieved by keeping the hematocrit over 30%, the albumin over 3 g/dL, and the electrolytes in the normal range. Although not evidence-based, in theory, avoiding anemia and hypoalbuminemia may enhance delivery of oxygen to the intestinal tissues and improve mucosal blood flow. Normal electrolytes decrease the likelihood of stasis related to poor motility. Measurement of fecal calprotectin or lactoferrin may be useful to establish a baseline so that repeated assessment can help define response to medical therapy.

Patients with IBD are at higher risk for being diagnosed with Clostridium difficile infection (CDI). In one single-center study, 18.4% of children with UC had a positive polymerase chain reaction (PCR) for the toxin B gene of C. difficile [11]. The percentages may be even higher in hospitalized children with IBD [12]. CDI is implicated in disease exacerbation and increases the risk for complications such as colectomy in adults with UC [13, 14]. Although there is not yet direct evidence that treating CDI in children with ASC improves outcomes, testing for and treatment of CDI is current standard practice and was recommended in the joint ECCO/European Society for Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN) ASC guidelines [9]. Stools should be screened for both toxins A and B. Stool should also be cultured for other potentially treatable bacterial pathogens.

Plain films of the abdomen are recommended as part of the initial evaluation of severe colitis if there are any signs of systemic toxicity that may suggest fulminant disease or TMC [9]. However, since examination findings can be masked by corticosteroid, our practice is to obtain a KUB on every hospitalized patient with ASC at baseline. As previously mentioned, transverse colon dilation ≥ 56 mm and ≥ 40 mm is suggestive of TMC in children ≥10 and <10 years of age, respectively. Colonic dilation has also been shown to predict response to IVCS therapy in this setting [15].

Although children with ASC may not wish to eat or drink due to their physical symptoms, unless surgery is imminent, they should be allowed to do so, since available evidence from the adult literature shows that while bowel rest may decrease stool frequency and volume, it does not improve outcomes and may worsen nutritional status [16]. If a regular diet cannot be tolerated by the third or fourth day, then enteral or parenteral nutrition should be considered, as a further malnourished state may prevent healing and clinical improvement. The risks of parenteral nutrition, including complications from central venous catheters (e.g., infection, thrombus) and electrolyte abnormalities, need to be balanced with potential benefit. There is no evidence to support any particular oral diet or diet restrictions in ASC.

Unlike in Crohn’s disease, antibiotics are generally not indicated in ulcerative colitis, unless there is evidence for toxicity or infection. Since bowel perforations may be silent in patients on high doses of corticosteroids, any clinical sign of infection should be investigated and treated. In well-controlled trials in adults with ASC, intravenous (IV) antibiotics including ciprofloxacin [17] and metronidazole [18] have not been shown to improve ASC outcomes when used as adjunctive therapy to corticosteroids. No large or controlled pediatric studies directly address the efficacy of antibiotics in ASC; however, the recommendations are to treat with IV antibiotics if infection is suspected or while awaiting confirmatory testing [9]. ASC patients with fulminant disease or suspicion or diagnosis of TMC should be treated with IV antibiotics. The antibiotics agent(s) used should target enteric bacteria, including anaerobes.

Adults who are hospitalized with ASC are routinely treated with anticoagulants for venous thromboembolism (VTE) prophylaxis. Hospitalized children with IBD are also at increased risk for VTE [19]. The prothrombotic tendency in IBD is thought to be attributable to many different factors including an increase in procoagulants, decrease in anticoagulants, thrombocytosis, as well as endothelial and immunologic factors [20]. VTE is more common in children with active IBD than in those who have quiescent disease [21]. This risk may be augmented by relative immobility of sick, hospitalized IBD patients. In one study, risk factors for VTE in hospitalized children with IBD included older age, central venous catheters, parenteral nutrition, and the presence of a hypercoagulable condition [19]. Children with colonic IBD appear to be at higher risk for VTE [22]. Despite this, the overall incidence of VTE in hospitalized children remains low (11.8/1000 hospitalizations) [19], and there have been no pediatric studies assessing the benefits and risks of prophylactic anticoagulation in ASC or in IBD in general. As such, the routine use of anticoagulation in children with ASC is not currently recommended [9]. However, noninvasive methods of VTE prophylaxis like frequent mobilization, adequate hydration, and pneumatic/mechanical devices are advised, as they are of low risk, even if not well supported by current evidence. It is reasonable to consider anticoagulation in patients with other risk factors for VTE, including known hereditary causes of thrombophilia, smoking, and the use of oral contraceptives. When used, anticoagulation does not seem to worsen bleeding during IBD flares.

Intravenous corticosteroids (IVCS) are the recommended first-line treatment for ASC in children. IVCS have been used for acute exacerbations of UC for more than 60 years and have been shown to reduce mortality in adults [23]. There are no randomized trials evaluating the comparative efficacy of various CS doses in children. The current recommendations for CS dosing are for 1–1.5 mg/kg/day of methylprednisolone up to 40–60 mg/day [9]. The daily dose is often divided over two daily doses. Doses above 60 mg/day have not been found to be more effective in adults with ASC [24]. More recently, a prospective pediatric cohort study which followed 283 children with ASC for 1 year concluded that an IVCS dose of 2 mg/kg/day was not more effective than doses of 1–1.25 mg/kg/day in preventing the need for salvage therapy during the hospitalization or by 1 year, although day 5 PUCAI scores were improved in the high-dose CS group before sensitivity analysis [25]. In this study, IVCS dosing was at the discretion of the provider (not randomized), but propensity matching was performed to limit bias. Interestingly, glucocorticoid bioactivity in serum did not predict response to IVCS in a study of children with ASC [26].

Not all children with ASC improve with IVCS. A systematic review found a 34% (range 9–47%) IVCS failure rate in a pooled analysis of five studies of children with ASC [27]. In the one prospective study included in the analysis, 37 of 128 children (29%) failed to respond to IVCS and required second-line treatment [6]. Multiple predictors for poor response to IVCS in children have been identified. A multicenter prospective study that followed 128 children with ASC found response to IVCS less likely in older patients and in patients with established disease [6]. The same study showed that after multivariate analysis, additional day 3 and day 5 predictors of IVCS failure included high stool frequency and large amount of blood in the stool. A high CRP at day 5 also predicted CS failure. The PUCAI score outperformed other clinical indices in predicting IVCS failure at both days 3 and 5. A PUCAI score >45 on day 3 predicted CS failure with a sensitivity of 92% and negative predictive value (NPV) of 94%, indicating a high likelihood of response if the PUCAI score is ≤45. On day 5, a PUCAI score of >70 had a specificity and positive predictive value (PPV) of 100% for CS failure, while a score of >65 has specificity and PPV of 96% and 82%, respectively. The addition of fecal calprotectin or CRP to the model did not improve the accuracy of the PUCAI score. The findings from this study and others have formed the basis of recommendations for disease monitoring and for the timing of second-line/rescue therapy in children with ASC. Additional predictors of poor response to IVCS have also been identified. A prior single-center study found that a high number of nocturnal stools and high CRP were predictive of CS failure at days 3 and 5 [10]. The presence of megacolon, defined as a transverse colon diameter >40 mm and >60 mm in children <12 and >12 years of age, respectively, and ulceration on abdominal X-ray may also predict IVCS failure [15]. A separate study showed that day 3 interleukin (IL)-6 levels predicted IVCS failure, although this did not hold true after multivariate analysis [28]. Finally, there is limited evidence that IVCS nonresponders have decreased fecal microbial richness/diversity compared to responders, though this is not yet clinically applicable [29].

In general, monitoring for response to initial therapy begins with careful and frequent reassessment of vital signs, stool frequency, volume, blood loss, and abdominal pain as well as of changes in the abdominal examination. The validation of the PUCAI score in predicting IVCS failure has led to a suggested algorithm and the following recommendations for disease monitoring and for the timing of second-line, also referred to as “rescue” or “salvage”, therapy [9]. A PUCAI score >45 on day 3 of IVCS should initiate preparation for second-line therapy, including discussion of potential risks and benefits with patients and families and inclusion of a surgeon with experience in IBD. A PUCAI score >65 on day 5 should prompt initiation of second-line therapy. Patients with PUCAI scores between 35 and 65 on day 5 can continue IVCS for an additional 2–5 days, at which point further recommendations are based on PUCAI score at that time. Patients who improve on IVCS and have a PUCAI score <35 on day 5 are unlikely to require rescue therapy before discharge [9]. Thiopurines can be considered in IVCS responders, particularly in those who were previously naive, but therapeutic benefit is often delayed for 2–3 months; so, they have little role in the acute setting.

There is no current evidence to support the value of repeat colonoscopic evaluation in ASC patients who are improving on IVCS in the clinical setting. However, repeat sigmoidoscopy is suggested if the day 3 PUCAI score is >45 in order to search for evidence of Crohn’s disease such as granulomas and to exclude cytomegalovirus (CMV) colitis, which can complicate ASC and may alter therapy. While the prevalence of CMV colitis in children with ASC is not known, it is relatively common in adults with UC, particularly in those with steroid refractory disease [30]. Mucosal biopsies should be obtained and evaluated for signs of CMV disease (deep ulcerations and viral inclusions) as well as immunohistochemistry [9]. CMV colitis should prompt an infectious disease consultation, and antiviral treatment should be considered [31].

Patients with ASC with poor response to IVCS require rescue therapy. About one-third of children with ASC require rescue therapy before discharge from the hospital. In adults, the earlier use of rescue therapy appears to decrease mortality [24], and extending IVCS without rescue treatment beyond 14 days is unlikely to provide benefit and may increase the risk for complications, including, but not limited to, opportunistic infections, metabolic and electrolyte abnormalities, osteopenia/porosis, and psychiatric disturbance. The goals of rescue therapy are to improve symptoms and allow for eventual discontinuation of CS. Current rescue therapy options for children with ASC include infliximab, calcineurin inhibitors (cyclosporine and tacrolimus), and colectomy. Although the data supporting these rescue therapies are primarily in CS refractory patients, these treatments are also used without IVCS in patients with contraindications or prior lack of response to CS.

Infliximab (IFX) is a monoclonal antibody against TNF-α that can induce and maintain remission in pediatric UC [32]. Pooled data from six pediatric case series (n = 126) of ASC patients treated with IFX showed a 75% (67–83%, 95% CI) response rate by the time of hospital discharge and a 64% colectomy-free rate during follow-up which ranged from a few months to a few years [27]. In one prospective study, 76% (25/33) of children with ASC refractory to IVCS had short-term response to IFX [6]. The remaining 24% underwent colectomy. At 1 year, 55% had sustained response to IFX and 45% had CS-free sustained response, while an additional 28% required colectomy by 1 year. In a more recent retrospective study from a single center in Italy, 80% of ASC patients had short-term response to IFX, but 50% of these patients went on to colectomy by 24 months [33]. Predictors of IFX failure may include shorter disease duration and more active disease at the time of admission and day 3 of IVCS [6]. IFX is typically dosed at 5 mg/kg at baseline and then repeated at 2 and 6 weeks following the initial dose. The pharmacokinetics of IFX in children with moderate-to-severe UC appears to be similar to that in adults [34]. However, many pediatric centers use higher doses (10 mg/kg) and/or shorter dosing intervals of IFX in ASC. While there is currently a lack of direct evidence to support this practice, some have suggested that IFX clearance may be higher in patients with acute severe disease leading to a requirement for higher dosing [35]. A recent retrospective study of children with IBD (CD and UC) showed that patients with a larger colonic inflammatory burden were more likely to require IFX dose escalation by 12 months than patients with limited or moderate disease and that 43% of patients who started at 5 mg/kg dosing did not improve with dose escalation [36]. Although this study was not limited to ASC patients, it does provide some indirect evidence that children with more extensive disease may benefit from higher IFX doses at the start of treatment. ECCO/ESPGHAN guidelines recommend IFX as the preferred rescue therapy in patients with previous thiopurine failure as IFX can also be effective as a maintenance agent in UC [9]. Prior to starting IFX, tuberculosis and Hepatitis B status should be documented.