Existing therapies for the treatment of Crohn disease (CD) in children focus on achieving remission of active disease and then maintaining symptom quiescence. Often this has involved a stepwise pharmaceutical approach focused on improving quality of life and promoting growth while minimising adverse drug effects.
Several meta-analyses have compared the efficacy of EEN to that of conventional corticosteroid therapy for moderate-to-severe CD. Overall, steroids have been found to be somewhat more effective at inducing symptomatic remission (1–3). Other reviews focusing on paediatric studies have found that the difference in efficacy may be minimal (4,5). Regardless, EEN remains an effective means of controlling symptoms of CD and conveys added benefit through the avoidance of steroids and improved nutritional status and growth (6–8). EEN, together with the antitumour necrosis factor agents, and perhaps azathioprine and methotrexate, are the only therapies for CD in which mucosal healing has been documented as a consequence of therapy (9–15).
Despite evidence demonstrating its efficacy and supplementary benefits, the use of EEN in children with CD has not been universally adopted by North American paediatric gastroenterologists. This is in stark contrast to their European counterparts; the majority of western Europeans report frequent use of EEN (62%), whereas a minority of their Canadian (36%) and American (4%) colleagues report frequent use (16). The reasons for such remarkable differences in attitude and practice have not previously been elucidated.
The goal of the present study was to define attitudes and practice patterns relating to the use of EEN by North American paediatric gastroenterologists. Moreover, we aimed to explore various factors that influence the use of EEN and identify obstacles to adopting EEN as a primary therapy for paediatric CD in North America. The findings may provide opportunities to advance the use of EEN in North America and the consequent benefits for patients. Higher rates of EEN were expected in centres with infrastructure for patient education and follow-up and multidisciplinary support teams including nurses and dietitians, and were served by publicly funded health care (due to the costs associated with EEN).
All of the full and associate members listed in the 2006 membership directory of the North American Society for Paediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN) were contacted by e-mail and invited to complete a survey. International and emeritus NASPGHAN members were not included in the survey.
The survey tool was developed through a collaboration of experts. An iterative process of pilot testing and revising was used to develop the final version. It was designed to capture demographic information, attitudes towards the role of EEN in CD, practice patterns relating to the use of EEN, and specific enteral feeding protocols. Respondents who had not treated at least 1 patient with CD in the last 12 months were asked to provide background information only and were excluded from additional analysis. Respondents were asked to identify themselves as either a “never,” “sparse,” or “regular” user of EEN. No criteria to quantify each choice were provided. Three different versions of the survey were administered based on the self-reported pattern of use. Respondents who reported regular use of EEN were asked about the location and features of disease for which they consider EEN, details of the protocol used, and factors they felt influence the success of EEN treatment. Respondents who reported no use of EEN were asked to identify obstacles to EEN and factors that could increase its use. Respondents who reported sparse use of EEN completed both versions of the survey. Participants were contacted by e-mail and invited to follow a hyperlink to a secure Web site where they could log on using an individualised identifier and complete the survey.
A descriptive analysis of demographic characteristics and patterns of EEN use was performed. Respondents were grouped into 3 groups based on their reported use of EEN: regular users, sparse users, and nonusers. Additional analysis of barriers to enteral feeding and the details of EEN and maintenance protocols was performed. Pearson correlation coefficients were calculated to assess the strength of associations between covariables. Pearson χ2 values were calculated to determine the significance of any differences between patterns of use, study group, and background characteristics. Where appropriate, continuity correction using Yates χ2 was used. Odds ratios (OR) with 95% confidence intervals (CI) were calculated for significant differences. All of the tests were considered statistically significant at the 0.05 level. Mexican respondents were excluded from subgroup analysis because their numbers were too few to make comparison possible.
Surveys were completed by 357 of 1162 eligible NASPGHAN members from Canada, the United States, and Mexico. The participation rate was 30.7%. Thirty-one respondents were excluded from analysis for reporting that they had not treated at least 1 patient with inflammatory bowel disease in the previous 12 months (Table 1).
One hundred thirty of 326 (40%) physicians responded that EEN is an appropriate or extremely appropriate therapy for inducing remission in children with CD. One hundred forty-one of 326 (43%) believed that EEN is sometimes an appropriate therapy, whereas 51 of 326 (16%) believed that it is rarely appropriate. No respondents more believed that EEN is never an appropriate therapy.
Canadians more frequently responded than their American counterparts that EEN is an appropriate therapy for paediatric CD (P < 0.001), and were more than 3 times likely to view EEN as appropriate or very appropriate, OR 3.22 (CI 1.60–6.56). Beyond country of practice, no additional information on location of practice was collected.
One hundred of 326 respondents (31%) reported never using EEN to treat CD, 179 (55%) reported using EEN sparingly, and 40 (12%) reported regular use of EEN. There was a significant difference in the use of EEN between Canada and the United States, P < 0.001. Regular use of EEN was reported by 9% of US and 32% of Canadian respondents, whereas 35% of US and 9% of Canadians respondents reported never using EEN to treat paediatric CD. The odds of US physicians reporting never using EEN was more than 5 times compared with physicians from Canada, OR 5.4 (CI 1.8–18.4).
A Pearson correlation established that both currently working in a centre where EEN is used and previously working in a centre where EEN was used were significantly related to the perceived appropriateness of EEN (both P < 0.01) and to how regularly physicians report using EEN (both P < 0.01). The odds of a paediatric gastroenterologist regularly using EEN are increased 8-fold if that physician had previously worked in a centre where EEN was used either regularly or very regularly, OR 7.9 (CI 3.6–17.2). Remarkably, the same odds are increased nearly 40-fold if the physicians currently work in a centre where EEN is used regularly, OR 38.5 (CI 14.5–105.4). No significant differences in patterns of use were found with respect to physician's age or private practice versus hospital-based practice (data not shown).
Benefits and Barriers to EEN
All of the participants were asked to rank the benefits of EEN (aside from the presumed advantage of inducing remission). Steroid sparing and steroid avoidance ranked highest as the main benefits. Nutritional benefits and the use in children with growth failure ranked second and third, respectively. Compliance issues were seen as both the main disadvantages of EEN by a sizable majority of physicians (72%) and also the major barrier to increased use by nonregular users. The perceived obstacles to nonregular users increasing their use of EEN are presented in Figure 1.
When asked to select information they would like to have before considering a more regular use of EEN most nonregular users reported that they would like practice guidelines (58%) over more evidence of efficacy (44%) or more understanding of the mechanisms of action (17%). The role of various treatment and disease factors in the success of EEN is presented in Figure 2.
Physicians who use enteral feeding most would consider EEN for ileocolonic disease (79%), isolated upper gut disease (76%), or upper and lower gut disease (79%). Likewise, the majority would rarely or never consider EEN for perianal disease (80%) or colonic disease (67%).
A variety of professionals may be involved in the administration of EEN. Seventy-four percent of physicians reported that a dietitian is often or always involved in the administration of EEN and 65% reported that a nurse is always or often involved. The involvement of psychologists and primary care physicians is much less common, with only 8% and 13%, respectively, reported to be often or always involved. There was a significant correlation between the country of practice and the involvement of dietitians (P < 0.01). Dietitians were reported to be always involved by 92% of Canadian respondents and 44% of US physicians. In addition to variation between countries, the involvement of both dietitians and nurses was reported either often or always more frequently by physicians in hospital-based practices (P = 0.003 and P = 0.005, respectively) and less common in private practice (P < 0.001 and P = 0.001, respectively).
There was no clear preference for the type of formula, and many physicians reported using more than 1 type; 47% use polymeric formula, 55% semielemental formula, and 47% elemental formula. Thirty percent of physicians recommend an initial duration of EEN of <6 weeks, and 46% reported using a 6- to-8 week course. Only one-quarter of physicians use an initial course >8 weeks; 9% reported 8 to 10 weeks, 13% reported 10 to 12 weeks, and 3% reported >12 weeks.
Two-thirds of physicians reported using in-hospital education and initiation of feeds rather than outpatient education and initiation of feeds. Seventy-one percent of respondents who use EEN reported nasogastric or gastric tube feeding as their usual route of administration: however, physicians in Canada were 20 times more likely to use nasogastric or gastric feeds over oral feeds than Americans, OR 19.9 (CI 2.8–399.9).
Many physicians report programmes to help cover the costs associated with EEN (Fig. 3). There was a significant difference between Canadian and US respondents in terms of the prevalence of programmes for formula, infusion pumps, and other equipment, (P < 0.001 for all comparisons).
Sixty-three percent of American physicians felt that medical therapy concurrent with EEN is necessary to induce remission. This was a significantly higher proportion than the 24% of Canadian physicians who report the same (P < 0.001). Respondents who treat with concurrent medication most commonly use 5-aminosalicylic acid (69%), followed by 6-mercaptopurine (60%), steroids (51%), azathioprine (40%), infliximab (40%), and methotrexate (12.6%).
Most physicians (61%) agreed that the success of EEN should be measured primarily by clinical improvement. They also agreed that nutritional measures such as weight gain are the next most important outcome measures, and biochemical measures such as erythrocyte sedimentation rate and platelet counts are third. The Paediatric Crohn Disease Activity Index was ranked as the fourth most important measure, and radiological evidence was ranked fifth.
Of the physicians who reported using EEN only 7% reported often or always using maintenance formula to maintain remission. Seventy-four percent of physicians reported sometimes or rarely and 19% never use maintenance formula. Although these data suggest that maintenance therapy is not common, overall Canadian respondents were more likely to report often or always using maintenance therapy than US respondents (OR 3.08; CI 1.3–7.2).
The use of polymeric formula as maintenance therapy was reported by 52% of respondents, whereas the use of semielemental and elemental formulas was reported by 32% and 26% physicians, respectively. The majority (84%) report the use of a maintenance regimen that supplements enteral nutrition with a regular diet, whereas 36% cycle periods of EEN with periods of regular diet and 11% use a different regimen. When transitioning back to a regular diet 57% of respondents reintroduce food progressively, 1 meal at a time, 27% reintroduce regular diet immediately, 55% reintroduce food based on fibre/residue, and 32% reintroduce diet based on allergenicity. Sixty-eight percent of physicians would consider gastrostomy devices for long-term supplementary formula.
Most respondents (86%) felt that medical therapy in conjunction with maintenance enteral nutrition is needed to maintain remission. Among physicians who use concomitant maintenance medication, aminosalicylic acid (74%), 6-mercaptopurine (70%), and azathioprine (56%) were the adjunct drugs of choice, followed by infliximab (36%), methotrexate (22%), and steroids (9%).
Enteral nutrition has been extensively used to treat paediatric CD in centres across western Europe (16). Indeed, consensus-based guidelines published by the European Society for Parenteral and Enteral Nutrition have been among the first to recommend the use of EEN as first-line therapy in children with CD (17). The present study reveals a much different picture of EEN in North America. Only 13% of North American paediatric gastroenterologists report regular use of EEN and, more strikingly, one-third of North American physicians report never using EEN to treat childhood CD.
A previous survey of NASPGHAN members revealed significant disparity between the use of EEN in Canada and the United States. In that study Levine and colleagues found that 36% of Canadian and 4% of American paediatric gastroenterologists used EEN regularly to treat childhood CD (16). The present study's results corroborate these earlier findings with 32% of Canadian and 9% of American gastroenterologists reporting regular use of EEN.
As predicted, the current data also reveal that physicians' opinions towards the appropriateness of EEN are associated with how frequently they use EEN. However, there appears to be a divide between opinion and practice. None of the physicians surveyed reported that EEN is never an appropriate treatment for CD; however, many physicians (31%) reported never using EEN for that purpose.
Many factors contribute to the pattern of EEN use in North America. Given the traditional lack of treatment guidelines, and of randomised controlled trials, paediatric practice has been largely based on adult evidence, expert opinion, and small paediatric studies (18). The current evidence suggests that the use of EEN is heavily influenced by the extent to which physicians are exposed to its use at the centres where they work or train.
The data presented in Figure 1 demonstrates that neither the unclear mode of action by which enteral nutrition controls intestinal inflammation nor issues around its efficacy would seem to contribute substantially to its limited use in North America. Overwhelmingly, issues around patient compliance are perceived by physicians as the most important concerns limiting the use of EEN. The next most significant contributor to its limited use is lack of support from the patient and his or her family. Physicians who do not use EEN believe that their lack of experience is as much a barrier as lack of support from the patient and family. Concerns for the costs associated with EEN and support required of a multidisciplinary team also seem to only play minor roles in physicians' reluctance to use EEN. Although a potential factor to account for the difference in use between US and Canadian colleagues may be ease of access for families to reimbursement for EEN, survey respondents did not endorse this concern.
Some of these patterns and concerns have also been raised by Australian paediatric gastroenterologists in a survey of EEN attitudes in that region (13). Nine of the 21 gastroenterologists included in the present study used EEN regularly for CD. Although the population surveyed was much smaller (21 replies from 36 physicians), use of EEN in this region also related to experience with or exposure to EEN administration. Concerns about compliance, lack of efficacy, and poor understanding of mechanisms were raised by the respondents who do not use EEN for their patients.
Only 12% of North American paediatric gastroenterologists report regular use of EEN and, more strikingly, one-third of North American physicians report never using EEN to treat childhood CD. Although much study and debate focuses on the efficacy of EEN on understanding of the mechanisms of action, these physicians want practice guidelines for EEN.
Opponents of EEN highlight several meta-analyses of paediatric and adult studies that found EEN to be less effective than conventional corticosteroid therapy at inducing clinical remission (1–3). Regardless of the fact that there have been no studies comparing enteral nutrition with placebo or medication less effective than corticosteroids, the benefit of EEN can be demonstrated by comparing the EEN response rates (53% to 82%) to usual placebo response rates in clinical trials (18% to 42%) (1). Additionally, a caution that has been raised with regards to interpreting these meta-analyses is that many of the trials allowed concurrent drug therapy in the corticosteroid treatment groups while excluding concurrent drug therapy in the EEN treatment groups. A post hoc analysis of a Cochrane review addressed this issue by excluding studies that allowed concurrent sulfasalazine in the corticosteroid group. Subsequently, corticosteroid was determined to no longer have a significant advantage (3).
Patient, family, and societal/cultural factors undoubtedly play a role in the acceptance and use of EEN. The present study cannot assess these issues beyond the physician's opinion. It is possible that the physician perception regarding the impact of these factors may differ from that of patients and their families. Additional study is required to explore factors that influence patients' and families' acceptance of EEN and to identify obstacles to patient compliance. Ultimately, identifying these issues may provide a framework for education strategies to address concerns and misconceptions, as well as targeted programmes to support families who choose EEN.
The use of e-mail to distribute the survey created additional obstacles. Outdated e-mail lists and antispam utilities that vary widely in their false-positive rates (legitimate e-mail filtered as spam) confound the author's ability to determine which e-mails were delivered to their addressee. The authors were able to confirm that at least 18 members had undeliverable postal/e-mail addresses, had retired, or were deceased. Thus, the uppermost estimate of the sample population is 1144, which results in an actual participation rate of >31%. Although the relatively low response rate does introduce the possibility of nonresponse bias, the authors feel that the sample likely reflects the opinions and attitudes of the larger paediatric gastroenterologist population. The response rate compares favourably with other Web-based surveys (19), and is more than double the rate reported by Levine and colleagues in their previous survey of NASPGHAN members (16).
As physicians it remains important to remember primum non nocere (first do not harm). In an era of increasingly complex medical therapy that involves immunosuppression and immunomodulation we must not lose sight of the potential for serious adverse consequences of our treatments. It behooves us to explore therapies, such as EEN, that have proven efficacy without toxic adverse effects. Proponents of EEN therapy need to be more vocal and advocate for research into the mechanisms of EEN, but also to further explore patient and family factors for seeking or avoiding this form of therapy.
1. Griffiths AM, Ohlsson A, Sherman PM, et al
. Meta-analysis of enteral nutrition as a primary treatment of active Crohn's disease. Gastroenterology 1995; 108:1056–1067.
2. Fernandez-Banares F, Cabre E, Esteve-Comas M, et al
. How effective is enteral nutrition in inducing clinical remission in active Crohn's disease? A meta-analysis of the randomized clinical trials. JPEN J Parent Enteral Nutr 1995; 19:356–364.
3. Zachos M, Tondeur M, Griffiths AM. Enteral nutritional therapy for induction of remission in Crohn's disease. Cochrane Database Syst Rev
4. Heuschkel RB, Menache CC, Megerian JT, et al
. Enteral nutrition and corticosteroids in the treatment of acute Crohn's disease in children. J Pediatr Gastroenterol Nutr 2000; 31:8–15.
5. Dziechciarz P, Horvath A, Shamir R, et al
. Meta-analysis: enteral nutrition in active Crohn's disease in children. Aliment Pharmacol Ther 2007; 26:795–806.
6. Seidman EG, Griffiths AM, Jones A, et al
. Semielemental diet versus prednisone in the treatment of active Crohn's disease in children and adolescents. Gastroenterology 1993; 104:A778.
7. Newby EA, Sawczenko A, Thomas AG, et al. Interventions for growth failure in childhood Crohn's disease. Cochrane Database Syst Rev
8. Shamir R, Phillip M, Levine A. Growth retardation in pediatric Crohn's disease: pathogenesis and interventions. Inflamm Bowel Dis 2007; 13:620–628.
9. Berni Canani R, Terrin G, Borrelli O, et al
. Short- and long-term therapeutic efficacy of nutritional therapy and corticosteroids in paediatric Crohn's disease. Dig Liver Dis 2006; 38:381–387.
10. Day AS, Whitten KE, Sidler M, et al
. Systematic review: nutritional therapy in paediatric Crohn's disease. Aliment Pharmacol Ther 2008; 27:293–307.
11. Schnitzler F, Fidder H, Ferrante M, et al
. Mucosal healing predicts long-term outcome of maintenance therapy with infliximab in Crohn's disease. Inflamm Bowel Dis 2009; 15:1295–1301.
12. van Assche G, Vermeire S, Rutgeerts P. Mucosal Healing and Anti TNFs in IBD. Curr Drug Targets 2010; 11:227–233.
13. D'Haens G, Geboes K, Rutgeerts P. Endoscopic and histologic healing of Crohn's (ileo-) colitis with azathioprine. Gastrointest Endosc 1999; 50:667–671.
14. D'Haens G, Geboes K, Ponette E, et al
. Healing of severe recurrent ileitis with azathioprine therapy in patients with Crohn's disease. Gastroenterology 1997; 112:1475–1481.
15. Kozarek RA, Patterson DJ, Gelfand MD, et al
. Methotrexate induces clinical and histologic remission in patients with refractory inflammatory bowel disease. Ann Intern Med 1989; 110:353–356.
16. Levine A, Milo T, Buller H, et al
. Consensus and controversy in the management of pediatric Crohn disease: an international survey. J Pediatr Gastroenterol Nutr 2003; 36:464–469.
17. Lochs H, Dejong C, Hammarqvist F, et al
. ESPEN guidelines on enteral nutrition: gastroenterology. Clin Nutr 2006; 25:260–274.
18. Beattie RM, Croft NM, Fell JM, et al
. Inflammatory bowel disease. Arch Dis Child 2006; 91:426–432.
19. Cook C, Heath F, Thompson RL. A meta-analysis of response rates in web- or internet-based surveys. Educ Psychol Meas 2000; 60:821–836.