What Is Known
- Inflammatory bowel disease in the pediatric population, particularly the youngest patients with very early-onset inflammatory bowel disease is frequently limited to the colon, making classification between Crohn disease and ulcerative colitis challenging.
- Severe colitis in pediatric patients is often refractory to medical therapy.
- Colectomy in patients with known or later declared Crohn disease can result in postoperative complications.
What Is New
- Temporizing diverting ileostomy in medically refractory colonic pediatric inflammatory bowel disease can improve clinical status.
- Diversion ileostomy can be a steroid-sparing treatment option.
- A subset of pediatric patients with inflammatory bowel disease with severe colonic disease can be successfully reanastomosed after temporizing ileal diversion.
In a subset of pediatric patients with severe colonic inflammatory bowel disease (IBD) refractory to medical therapy, surgical intervention may be necessary. Colectomy is most often followed by ileal pouch-anal anastomosis (IPAA). Although this procedure is often curative in patients with severe ulcerative colitis (UC), there is a higher risk of postoperative complications in patients with Crohn disease (CD) including disease recurrence, chronic refractory pouchitis, and pouch failure (1,2). Therefore, the decision to proceed with colectomy in pediatric IBD is most frequently limited to when the diagnosis is determined to be UC to avoid these sequelae.
Despite efforts to determine the diagnosis of UC before surgery, reclassification from UC to CD can occur, with rates in the literature varying between 2% and 16% (2–5). In a recent study performed on 202 pediatric patients with UC who underwent IPAA, 33 (16%) were later found to have CD. Over a 20-year follow-up period, patients were significantly more likely to retain their pouch if they remained diagnosed with UC (92%) compared to those who underwent a change in diagnosis to CD (61%) (3). In an older study of 151 pediatric patients with UC who underwent IPAA, 15% were later found to have CD and suffered complications of chronic pouchitis and pouch failure (2). Furthermore, a proportion of pediatric patients with colonic IBD (5%–30%) cannot be characterized definitively as CD or UC and are thus classified as IBD-unclassified (IBD-U) (1,6,7). Surgical management decisions in these patients are particularly challenging.
Colonic fecal diversion by creation of a temporary ileostomy has been well described as a therapeutic measure in perianal CD, leading to improved symptoms. The procedure diverts the fecal stream from the colon, and may decrease downstream inflammation while medical therapy is optimized (8). At our institution, temporizing diverting ileostomy is also used for the indication of colonic IBD refractory to medical therapy in pediatric patients. This procedure may be particularly beneficial in patients with IBD-U, as it can provide time for the appropriate diagnosis to be determined, and thus potentially preventing the complications associated with IPAA in patients who in fact have CD as described above. Thus far, there is no available published data on the success of temporizing diverting ileostomy performed in pediatric patients with severe colonic disease.
The primary aim of the present study was to characterize clinical and surgical outcomes of ileal diversion for refractory colitis in pediatric patients with IBD at our institution. Secondary aims were to determine whether diversion allowed for a change or clarification in IBD diagnosis and to describe adverse events associated with ileal diversion.
Study Design and Patients
This was a retrospective study of pediatric patients followed at The Children's Hospital of Philadelphia (CHOP), who underwent diverting ileostomy between 2000 and 2014 for the indication of severe and refractory colonic IBD. The study was approved by the CHOP institutional review board and consent was not required for this retrospective chart review. Potential subjects were primarily identified through a query of the operating room (OR) manager database for the surgery terms “ileostomy,” “ileostomy, revision,” “laparoscopic-assisted ileostomy,” and “ileostomy, continent, creation.” This list of patients was crossreferenced with a clinical database of patients with IBD.
Inclusion criteria included a documented diagnosis of IBD based on endoscopic, radiologic, laboratory, and clinical findings. The diagnosis of CD, UC, or IBD-U was determined by the treating physician and was based on accepted clinical, endoscopic, and histologic criteria. Specifically, CD was diagnosed if the following criteria were met: presence of granuloma, fistula, and/or perianal disease. Patients were characterized as IBD-U if these criteria were absent but there were other histological findings, such as duodenal inflammation, or other clinical phenotype potentially consistent with CD. The remainder of the cohort with isolated colonic disease was classified as UC. Patients who underwent diverting ileostomy were included only if the indication for surgery included severe, refractory colonic disease. Patients were excluded if they had a diagnosis of infectious colitis, diverting ileostomy for indication other than severe colonic disease, such as severe perianal disease, history of prior or concurrent colonic resection, or if medical documentation was incomplete. Patients were included if they had severe colitis and perianal disease. Medical records were reviewed to obtain baseline demographic, clinical, and outcome data including age at diagnosis, disease phenotype, anthropometric measurements, laboratory studies, imaging studies, endoscopy and pathology results, medication history, surgical history, and adverse events or complications related to disease, medication, or surgery. Clinical variables were reviewed in the 1-year period leading up to diversion, and compared with the 1-year period after surgery. Follow-up surgical procedures and any changes to diagnosis were evaluated through 2015. Study data were collected and managed using REDCap (9), a secure, Web-based application of electronic data capture tools hosted at the CHOP.
Disease Activity and Clinical Variables
To assess disease severity before diversion, Pediatric Crohn Disease Activity Index (PCDAI) (10), and Pediatric Ulcerative Colitis Activity Index (PUCAI) (11), scores were calculated retrospectively via chart review based on available data. PCDAI scores were calculated for patients with CD and IBD-U, and PUCAI was calculated for subjects with UC. Given that the disease distribution in this patient population was primarily colonic, PUCAI scores were also calculated for patients with IBD-U; there is no known validated disease activity score in IBD for patients with IBD-U. Prediversion endoscopy reports were also reviewed, if available.
In the postoperative period, the PUCAI, PCDAI, and other standardized indexes were not calculable, because stool variables are no longer evaluable in diverted patients. In place of such an index, other measurable outcomes related to disease activity were analyzed pre- and postdiversion. These measures included height and weight velocities, blood transfusion requirement, hospitalizations, chronic steroid use, hemoglobin, and albumin. Patients were excluded from analysis if complete data were not available for any given variable, and these instances are noted in the text.
Height and weight velocities were calculated for subjects with complete records over the 1 year prior and 1 year after diversion. Two growth points separated by 6 months to 1 year were required to calculate growth velocity. Height velocity z scores were also calculated for patients with available data.
Number of days hospitalized and number of hospitalizations in the 1 year before diversion compared with 1 year after diversion were compared.
Chronic Steroid Use
We defined chronic steroid use as either 3 courses of steroids in 1 year or a prolonged single course of steroids for 3 months or longer with failure of taper. Chronic steroid use was compared in the 1 year before diversion compared with 1 year after diversion.
Requirement for Blood Transfusion
Average number of blood transfusions required in the year before diversion was compared to average number of transfusions required in the year after diversion.
Average hemoglobin 6 months and 1 year before diversion were compared with average hemoglobin 6 months and 1 year after diversion.
Average albumin 6 months and 1 year before diversion were compared to average albumin 6 months and 1 year after diversion.
Follow-up surgeries including colectomy and ostomy takedown with bowel reanastomosis were recorded through 2015.
Change in Diagnosis
Reclassification of IBD subtype (CD, UC, IBD-U) was recorded through 2015.
Complications of ileal diversion were noted for the period of time of diversion for each subject. These included ostomy prolapse, ostomy revision, bowel resection, toxic megacolon, malignancy, and death.
Clinical variables pre- and postdiversion were compared using the Wilcoxon signed-rank test. Statistical significance was determined at the 2-sided α = 0.05 level (P ≤ 0.05). All statistical analyses were carried out in the R software package (version 3.1.1) (12).
One hundred three potential subjects were identified for the study, and 24 were included in the final analysis. Thirty-two patients were excluded because of a non-IBD diagnosis and 44 patients were excluded because of concurrent or previous resection. Two patients with very early-onset (VEO)-IBD who would otherwise have met inclusion criteria were removed from analysis because of underlying immunodeficiency causing their IBD phenotype, one with chronic granulomatous disease, and 1 with immunodysregulation polyendocrinopathy enteropathy X-linked syndrome. One patient who otherwise met inclusion criteria was excluded because of incomplete medical records. One patient who had diverting ileostomy performed for the indication of severe colitis, also had severe perianal disease, and was included in the analysis.
Demographic, baseline clinical characteristics, and disease phenotype of the 24 included subjects are summarized in Table 1. Average age at diagnosis was 6.72 years (range 6 months to 14 years). Twelve patients (50%) were diagnosed with IBD at age 5 years or younger. Ten (42%) had an initial diagnosis of CD, 1 had an initial diagnosis of UC (4%), and 13 (54%) had IBD-U. All patients had primarily colonic disease. All but 1 patient, an infant, had failed therapy with 1 or more biologic agents before diversion. Mean length of time from diagnosis to diversion was 2.2 years (range 3 months to 7 years).
One surgeon performed the majority of the procedures (17/24). The initial surgical procedure performed in all patients was a loop ileostomy. In 22 of the 24 patients a Turnbull modified loop ileostomy was used. In this procedure, the proximal end of the bowel is brought out like an end and the distal end is sutured flush to the skin (13). Mean length of follow-up from time of diversion was 3.4 years (range 8 months to 10.3 years).
Of the 19 subjects with complete data available, PCDAI was calculated for 18 patients with CD or IBD-U, and PUCAI was calculated for 1 subject with UC. Mean PCDAI before diversion was 49.31 (range 32.5–67.5); all patients with CD and ileocecal valve had moderate to severe disease based on these scores. PUCAI before diversion was 65 in the subject with UC, which is severe. PUCAI was also calculated for all subjects with IBD-U for comparison, given that all subjects had primarily colonic disease. When PUCAI scores were calculated for 11 patients with IBD-U with available data, the mean was 66.4 (range 25–70). One patient (9%) had mild disease (PUCAI 10–34) and was steroid dependent, and 10 patients had moderate to severe disease (PUCAI 35–70) (11).
Endoscopy was frequently performed shortly before diversion to assess extent and severity of disease, and to exclude a diagnosis of cytomegalovirus infection as an etiology of refractory colitis. These procedures were performed at the discretion of the treating physician. Results from colonoscopy or flexible sigmoidoscopy performed before ileal diversion were reviewed by the authors in 19 of 24 patients with available data. Fourteen of the 19 patients were deemed to have severe colitis and 5 were moderate, based on review of described findings from the procedure and available images. There was insufficient information to use a validated scoring system for endoscopic severity.
Average height velocity 1 year before diversion was 2.91 cm/year, which improved to 5.58 at 1 year after diversion (P = 0.0495) (Fig. 1A). Average weight velocity 1 year before diversion was −1.56 kg/year, which improved to 5.94 in the 1 year after diversion (P = 0.0222) (Fig. 1C). Twenty subjects had available data for these calculations. Average height velocity z score before diversion was –3.19 and average height velocity z score after diversion was 1.21 (P = 0.009181) (Fig. 1B). Fourteen subjects had available data for this calculation, based on established reference values for height velocity z scores for children.
The average number of days hospitalized in the 1 year before diversion was 31.94, which decreased to 6.82 in the 1 year after diversion (P = 0.0006) (Fig. 1D), for 17 patients with available data. Average number of hospitalizations 1 year before diversion was 2.89, which decreased to 1.56 in the 1 year after diversion (P = 0.0132) for 18 patients with available data (Fig. 1E). If patients were hospitalized at another institution, this may have been noted in the chart, but duration of hospitalization may not have been specified. Hospitalization for the purpose of the diverting ileostomy was not included. Hospitalizations at other institutions before or after diversion, if not documented in the chart, would have been omitted from our data collection.
Chronic Steroid Use
The majority of patients in this cohort required prolonged treatment with systemic corticosteroids. Using this definition, 17 of 24 patients (70.83%) had chronic steroid exposure in the 1 year before diversion. After diversion, 4 of 18 (22.22%) patients who completed 1 year data collection required chronic steroids.
During the period of diversion, the majority of patients (18/24) were treated with biologic therapy. Additional details regarding medical therapy before and during the time of diversion can be found in Table 2.
Requirement for Blood Transfusion
Nine patients in the cohort (37.5%) required blood transfusion in the year before diversion, with an average of 4.33 transfusions per year. This is compared to 0.67 in the 1 year after diversion (P = 0.0076) in that subset of patients (Supplemental Digital Content 1, Fig. 1, http://links.lww.com/MPG/A869).
Average hemoglobin 6 months before diversion was 9.92 g/dL, compared to 10.43 after diversion (P = 0.1088) in 16 patients with available data. Average hemoglobin 1 year before diversion was 9.94 compared with 10.62 after diversion (P = 0.0207) in 12 patients with complete data (Supplemental Digital Content 2, Fig. 2, http://links.lww.com/MPG/A870).
Average albumin 6 months before diversion was 3.37 g/dL, compared with 3.65 post diversion (P = 0.11) in 19 patients with available data. Average albumin 1 year before diversion was 3.45 compared with 3.73 after diversion (P = 0.19) in 15 patients with complete data. These trends in improvement did not reach statistical significance.
Ten patients underwent subsequent colectomy after diversion, with an average length of time from diversion to colectomy of 1.04 years (range 2 months to 4.4 years). Seven patients underwent successful bowel reanastomoses and takedown of their ileostomy after an average period of 1.94 years diverted (range 5 months to 2.9 years). Seven patients in this cohort remain diverted at this time, with an average length of time diverted of 3.4 years (range 1.3–10.3 years). One patient has a prolonged diversion secondary to parental decision despite medical recommendation to proceed with colectomy. When excluding this patient from analysis, average length of time diverted of those patients still with ileostomy decreases to 1.17 years (range 1.3–4.3 years).
Change in Diagnosis
Seven patients underwent change in diagnosis during the period of diversion. Of the 10 patients who had a colectomy, 6 underwent a change or clarification of diagnosis. Five patients who were determined to have CD ultimately required colectomy due to persistent severe colitic symptoms. These included 3 patients (subjects 1, 4, and 5) classified as very early-onset IBD (VEO-IBD), who are suspected to have an underlying immunodeficiency with ongoing genetic and immunologic evaluation. Subject 2, had successful IPAA and had an intact pouch at the time of transition to adult gastroenterology (GI), 6 years after pouch creation. Subject 24 also underwent colectomy for severe persistent colitis <6 months ago. The 7 patients who underwent successful bowel reanastomoses and takedown of their ileostomy, maintained their initial diagnoses of CD or IBD-U. All patients with confirmed diagnosis of UC underwent colectomy. Details of diagnosis, length of time diverted, and surgical outcome can be found in Table 2.
Seven patients experienced postoperative complications, which are summarized in Table 3. These included reduction of prolapse in the OR, ostomy revision, and resection of intussuscepted bowel. There were no other serious complications such as toxic megacolon, malignancy, or deaths associated with ileal diversion.
The present study describes our experience using temporizing diverting ileostomy for refractory colitis in pediatric patients with IBD. The data support the notion that diversion can be beneficial in stabilizing disease to allow for response to medical therapy, and in some cases, more definitive classification of disease type.
In patients with severe colonic IBD, medical options are frequently exhausted and become extremely limited. Diverting ileostomy is considered at our institution to stabilize these patients who may be poor candidates for colectomy by diverting the fecal stream from a severely inflamed colon. This management option was first described in a study by Harper et al (14) who demonstrated clinical improvement in adult patients with severe Crohn colitis or ileocolitis who underwent loop ileostomy before definitive colectomy. In a subsequent study, adult patients with Crohn colitis experienced worsening of clinical status after instillation of fecal contents to the distal limb of a loop ileostomy (15).
In the present study, ileostomy was performed in patients with moderate to severe colitis as defined by the PCDAI/PUCAI and endoscopic findings at the time of diverting ileostomy. We found improvement in several measures of disease activity postdiversion. These included decreased steroid exposure (88% steroid-free after diversion), decreased requirement for blood transfusion, significant improvement in height and weight velocities, height velocity z score, and decrease in number of hospital admissions and hospitalization days. These clinical improvements allowed for successful reanastomosis in 7 patients. In those who ultimately went on to colectomy, the majority of patients achieved disease stabilization allowing for improved growth and nutrition, normalization of laboratory studies, and successful steroid taper possibly allowing for better surgical outcomes.
It is also worth noting that certain variables would be affected more significantly in 1 disease process compared with another. For example, it may be expected that growth failure would be a more significant issue in patients with CD compared with UC. Our small sample size did not, however, allow for statistically meaningful subgroup analysis by disease phenotype or eventual surgical outcome. We also acknowledge that improvements in clinical variables seen after diversion may not be fully attributable to the surgical intervention. In many cases medications were also modified during the period of time of diversion, although in several cases, patients were maintained on their previous medical therapy and did demonstrate clinically meaningful improvement while diverted.
Because pediatric IBD, particularly in young children, can present with isolated colonic distribution of disease, it can be difficult to correctly classify the disease as UC or CD at diagnosis. Patients with IBD-U and severe refractory disease are unique candidates for diversion due to the potential complications of colectomy and IPAA in patients who may have CD. If despite a thorough evaluation, the disease subtype remains unclear in the setting of severe colitis, at our institution a temporizing diverting ileostomy is considered. Diversion achieved the goal of more definitive diagnosis for a subset of patients in the present study. Fifty-four percent of the cohort carried the diagnosis of IBD-U at the time of diversion (13/24). By the time of reconnection, colectomy, or the end of the study period if still diverted, only 7 patients remained classified as indeterminate, and 6 of those with IBD-U had their disease type determined during the period of diversion.
There are potential caveats to conclusions regarding disease phenotype, which has implications for those patients who underwent a change in diagnosis during the study period. In 3 cases, there were CD-specific findings noted during the period of time diverted (granulomas in 2 subjects and development of perianal disease in another), which allowed for a definitive change in diagnosis. For 3 additional patients whose disease was indeterminate and later called UC, the period of diversion provided additional time to assess for CD-specific features and none were seen; however, as previously stated, features of CD may appear at any time during a patient's disease course.
Another striking feature noted in our study was that 50% (12/24) of the patients were diagnosed with IBD at the age of 5 years or younger. Patients with VEO-IBD often present with more severe refractory disease compared with older-onset IBD (1). Furthermore, it is often difficult to clearly establish the diagnosis of CD versus UC in these patients due to the predominant colonic involvement at presentation (1). A subset of patients with VEO-IBD have an underlying primary immunodeficiency or single genetic defect driving the development of disease, affecting therapeutic options, and surgery is often not curative in this cohort. Indeed, 2 patients who were initially identified for the present study were excluded based on diagnoses of chronic granulomatous disease and immunodysregulation polyendocrinopathy enteropathy X-linked syndrome. Only 3 of the 12 patients with VEO-IBD had a clear diagnosis of CD at the time of diversion. Although single gene defects or underlying immunodeficiency have not been identified thus far in the study population, several patients with VEO-IBD in the cohort are presently undergoing genetic and immunodeficiency evaluation and are part of a separate study examining the role of genetics in IBD. Thus, ileal diversion for medically refractory colitis in VEO-IBD should be strongly considered as a temporizing measure to allow for improved growth, avoidance of steroids, and time for further diagnostic workup, whereas medical management is optimized.
An additional potential benefit of diverting ileostomy is the opportunity for directed therapy to the colon via the distal limb of the ileostomy. Half (12/24) of the patients in this cohort were recommended to infuse antibiotics, 5-aminosalicylic acid preparations, or short-chain fatty acid enemas through their loop ileostomies to treat distal disease. There were, however, limitations in obtaining these therapies and challenges with administering them such that we could not reliably determine their efficacy in this retrospective analysis.
There is limited data on the long-term success of diverting ileostomy for colonic disease in both pediatric and adult literature. In a recent large meta-analysis of adult patients with perianal disease, 2 of 3 patients demonstrated improvement, and successful reconnection of the bowel was achieved in 17% (16). There is less published experience with this procedure performed for the indication of refractory colonic disease. A retrospective study, however, described its use in adult patients with colonic CD, in which 6 of the 29 patients underwent fecal diversion for the indication of colitis only (17). Successful reanastomosis was not achieved in any of these patients (17). In contrast, our study demonstrated successful reanastomosis in 7 of 24 (29%) patients with follow-up ranging from 2 months to 7.3 years. All patients in the cohort who were determined to have UC underwent colectomy. Those who were found to have CD were attempted to be stabilized and reconnected when possible. Ultimately, as we saw in our population, a subset of patients with severe colonic disease, including CD and IBD-U, will require colectomy.
We investigated whether there were objective criteria to determine the success of reanastomosis in our cohort; however, due to the small sample size and limited data available, we did not identify specific biomarkers or disease characteristics that were predictive of success or failure. In our cohort, factors involved in the decision to takedown the ileostomy and reconnect the bowel were determined by clarity of the diagnosis and restaging of disease by colonoscopy. Based on our analysis of experience at our institution, we believe that reconnection or decision to proceed to colectomy should be considered by 12 months after diversion, with most successful reconnections occurring by about 2 years after diversion, but acknowledge that decisions regarding subsequent surgeries will be made based on each patient's evolving clinical status.
The present study identified several potential surgical complications of diverting ileostomy, including the need for surgical revision of the ostomy and prolapse reduction in the OR. There were 3 instances of bowel intussusception leading to partial resections of ischemic bowel. The Turnbull ileostomy was the procedure of choice for several reasons. This type of ileostomy is completely diverting with minimal spill over of intestinal contents to the distal limb, it provides decompression of the distal colon, and allows infusion of enemas or medicated solutions through the distal bowel, as mentioned above. In addition, it is technically easier to close than other types of loop ileostomies, because closure can usually be done without a laparotomy and often without a circumferential anastomosis. Disadvantages to any loop ileostomy are prolapse and parastomal hernias. Adult data on complication rates of intestinal ostomies ranges widely from 21% to 70% (18). Anecdotally, at our institution, the frequency of complications and revisions in the present study population is at least comparable to pediatric patients who have ileostomies for other indications, and may even be more favorable than other patient populations, such as those with Hirschsprung's disease, based on experience at our center. There were no life-threatening adverse events after diversion surgery such as toxic megacolon or malignancy, and there were no deaths.
Limitations of the present study include the small sample size and retrospective design with reliance on accurate and complete documentation and chart review to obtain data. Data on several clinical variables were missing for various patients, as was noted in the text. Another limitation was the difficulty to objectively define disease severity during the time patients were diverted using any of the previously validated disease activity scores. PCDAI (19), abbreviated PCDAI (20), PUCAI (11), and Harvey-Bradshaw scores (21), all include frequency and/or consistency of bowel movements to calculate score. Diverted patients do not have true bowel movements to quantify or qualify. We relied on other surrogate measures of disease severity and found improvement in most of these measures.
Based on our experience of ileal diversion in the patients described in the present study, we conclude that there is benefit to temporizing ileostomy in a subset of pediatric patients with IBD and medically refractory colitis. Further prospective studies to investigate the benefit of short-chain fatty acid enemas and other therapies through an ileostomy will also add to our understanding of this temporizing option in pediatric patients with severe colonic disease.
The authors would like to acknowledge Kristen Plantz and Cassandra Spengler for their help with data collection and organization.
1. Heyman MB, Kirschner BS, Gold BD, et al. Children with early-onset inflammatory bowel disease (IBD): analysis of a pediatric IBD consortium registry. J Pediatr
2. Alexander F, Sarigol S, DiFiore J, et al. Fate of the pouch in 151 pediatric patients after ileal pouch anal anastomosis. J Pediatr Surg
3. Polites SF, Potter DD, Moir CR, et al. Long-term outcomes of ileal pouch-anal anastomosis for pediatric chronic ulcerative colitis. J Pediatr Surg
4. Pakarinen MP, Natunen J, Ashorn M, et al. Long-term outcomes of restorative proctocolectomy in children with ulcerative colitis. Pediatrics
5. Durno C, Sherman P, Harris K, et al. Outcome after ileoanal anastomosis in pediatric patients with ulcerative colitis. J Pediatr Gastroenterol Nutr
6. Carvalho RS, Abadom V, Dilworth HP, et al. Indeterminate colitis: a significant subgroup of pediatric IBD. Inflamm Bowel Dis
7. Kugathasan S, Judd RH, Hoffmann RG, et al. Epidemiologic and clinical characteristics of children with newly diagnosed inflammatory bowel disease in Wisconsin: a statewide population-based study. J Pediatr
8. D’Haens GR, Geboes K, Peeters M, et al. Early lesions of recurrent Crohn's disease caused by infusion of intestinal contents in excluded ileum. Gastroenterology
9. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform
10. Hyams J, Markowitz J, Otley A, et al. Evaluation of the Pediatric Crohn Disease Activity Index: a prospective multicenter experience. J Pediatr Gastroenterol Nutr
11. Turner D, Otley AR, Mack D, et al. Development, validation, and evaluation of a Pediatric Ulcerative Colitis Activity Index: a prospective multicenter study. Gastroenterology
12. R Development Core Team. A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2015.
13. Anderson DN, Driver CP, Park KG, et al. Loop ileostomy fixation: a simple technique to minimise the risk of stomal volvulus. Int J Colorectal Dis
14. Harper PH, Lee EC, Kettlewell MG, et al. Role of the faecal stream in the maintenance of Crohn's colitis. Gut
15. Zelas P, Jagelman DG. Loop illeostomy in the management of Crohn's colitis in the debilitated patient. Ann Surg
16. Singh S, Ding NS, Mathis KL, et al. Systematic review with meta-analysis: faecal diversion for management of perianal Crohn's disease. Aliment Pharmacol Ther
17. Mennigen R, Heptner B, Senninger N, et al. Temporary fecal diversion in the management of colorectal and perianal Crohn's disease. Gastroenterol Res Pract
18. Shabbir J, Britton DC. Stoma complications: a literature overview. Colorectal Dis
19. Hyams JS, Ferry GD, Mandel FS, et al. Development and validation of a pediatric Crohn's disease activity index. J Pediatr Gastroenterol Nutr
20. Loonen HJ, Griffiths AM, Merkus MP, et al. A critical assessment of items on the Pediatric Crohn's Disease Activity Index. J Pediatr Gastroenterol Nutr
21. Harvey RF, Bradshaw JM. A simple index of Crohn's-disease activity. Lancet