- Split View
-
Views
-
Cite
Cite
Rotem Sigall Boneh, Chen Sarbagili Shabat, Henit Yanai, Irit Chermesh, Sivan Ben Avraham, Mona Boaz, Arie Levine, Dietary Therapy With the Crohn’s Disease Exclusion Diet is a Successful Strategy for Induction of Remission in Children and Adults Failing Biological Therapy, Journal of Crohn's and Colitis, Volume 11, Issue 10, October 2017, Pages 1205–1212, https://doi.org/10.1093/ecco-jcc/jjx071
- Share Icon Share
Abstract
Loss of response [LoR] to biologics in Crohn’s disease [CD] is a significant clinical problem. Dietary therapy as a treatment strategy in this setting has not been previously reported. We report the use of dietary strategies using enteral nutrition coupled with the Crohn’s Disease Exclusion Diet [CDED] for LoR to infliximab or adalimumab as a single-centre experience.
Patients with LoR to a biologic despite dose escalation or combination therapy were treated with partial enteral nutrition [PEN] by a polymeric formula and the CDED for 12 weeks. Paediatric patients with severe flares received 14 days of exclusive enteral nutrition followed by PEN + CDED as above. All patients were seen at weeks 6 and 12 for follow up. Current and prior treatment, Harvey Bradshaw Index [HBI], C-reactive protein [CRP] and albumin were recorded. Remission was defined as HBI <5 at week 6.
Twenty-one patients, mean age 22.1 ± 8.9 years [11 adults and ten children] met study criteria. Seventeen patients [81%] had used combination therapy, and 10/21 [47.6%] had failed a second biologic. Seven patients had a prior intestinal resection. Dose escalation had failed in 13/21 [62%] patients. Clinical remission by physician’s global assessment and HBI after 6 weeks was obtained in 13/21 [61.9%]. Mean HBI decreased from 9.4 ± 4.2 to 2.6 ± 3.8 [p < 0.001], mean CRP decreased from 2.8 ± 3.4 to 0.7 ± 0.5 [p = 0.005] and mean albumin increased from 3.5 ± 0.6 to 3.8 ± 0.5 [p = 0.06].
Dietary treatment combining PEN with the CDED may be a useful salvage regimen for patients failing biological therapy despite dose escalation.
1. Introduction
The introduction of biological therapies targeting tumour necrosis factor α (TNFα) was an important advance in the treatment of luminal and extra-intestinal Crohn’s disease [CD].1–7 Treatments with infliximab and adalimumab have been associated with remission, maintenance of remission and mucosal healing in adults and children.2–7 Despite the encouraging results in clinical practice, loss of response [LoR] to biologics is a frequent event.1,8–10 Dose escalation and combination therapy have been shown to improve outcomes for infliximab and adalimumab and to rescue LoR, but LoR can still occur due to difficulty in maintaining trough levels and in other cases often despite high trough levels. An increasing number of patients are developing LoR on their second biologic, leaving physicians and patients with few therapeutic options.4,11,12
Diet appears to play an important role in disease pathogenesis.13,14 Recent and historical studies have shown that exclusive enteral nutrition [EEN] can induce remission and mucosal healing.15,16 This has led to the current guidelines recommending that EEN should be used as a first-line therapy in new onset mild to moderate CD in children.17
Current recommended dietary therapies for induction of remission involve EEN over a prolonged period of time [6–12 weeks] which is difficult to perform. In addition, this therapy is more difficult to perform in adult patients, which has led to decreased compliance and poorer results in adult studies.18 Partial enteral nutrition [PEN] is ineffective for induction of remission,19 and even 8 weeks of 90% PEN is inferior to EEN for inflammatory and clinical remission.20
We have previously hypothesized that the primary effect of dietary therapy is exclusion of dietary components that may cause dysbiosis, alter innate immunity or affect the barrier function such as the mucous layer or intestinal permeability.13,21 We have further demonstrated that a specialized diet that we have developed based on these principles, called the ‘Crohn’s Disease Exclusion Diet’ [CDED], coupled with supportive liquid formulas, can induce remission and mucosal healing, and that these results occurred even in patients who did not take any liquid formula.21 As a result, this diet combined with PEN 50% has replaced EEN as our standard dietary therapy in outpatients.
Critics of EEN have pointed out that EEN works best in children with new onset disease, in a window when the disease may be responsive to all forms of treatment. While dietary therapy has been traditionally used for induction of remission in early disease, the role for dietary treatment of long-duration established disease or in refractory disease has not been established in children or in adults to date. Due to the success of the diet in uncomplicated disease, and an increasing number of patients with LoR to available medications despite dose escalation, combination therapy or in the presence of normal or high trough levels, we expanded its use to anti-TNF biologic refractory patients, a field which has not been explored previously. The aim of the current study is to report our experience with use of dietary therapy other than EEN for induction of remission in patients who have lost response to infliximab or adalimumab.
2. Methods
This is a retrospective analysis of patients using dietary therapy for LoR to biologics between 2013 and 2016. The patient population consisted of patients losing response to biologics from our institution or quaternary referrals for use of dietary therapy. The flow diagram of patients is presented in Figure 1. All paediatric patients and young adults [<20 years] were seen at the Wolfson Medical Center Paediatric Inflammatory Bowel Disease [IBD] centre, or during admission to a paediatric ward for a severe flare, whereas adult cases >20 years of age referred for dietary therapy due to LoR were seen in the IBD centre or in a consultancy setting by a single physician [A.L.] and followed by A.L. and R.S.B. for dietary therapy. All children treated with dietary therapy using the CDED in our centre are seen by a physician and dietician routinely at baseline, weeks 3, 6 and 12 in order to adjust the diet. Adult patients were routinely seen at baseline, week 6 and week 12. Patients had their data recorded into a separate database by one of the authors [R.S.B.] at the time of first follow up 6 weeks after starting the diet. All patients were physically examined, and clinical assessment including number of liquid stools, severity of abdominal pain, bleeding and extra-intestinal manifestations were recorded. Patients were included if they had a flare while using biological therapy at recommended maintenance doses, had evidence for active inflammation by C-reactive protein [CRP], calprotectin or colonoscopy and had received dietary therapy as the only additional therapy for the flare. All the patients who started dietary therapy as the sole change in treatment for this indication were included in this report. Patients of any age could be included. Patients were excluded from this report if they had received any new medication or dose adjustment in concurrent therapy for induction of remission [antibiotics, steroids, immunomodulators or vedolizumab] in addition to dietary and anti-TNF therapy; however, maintenance immunomodulators and biologics could be continued if they had been on therapy for at least 12 weeks while LoR occurred and no change in dose had occurred. Patients were also excluded if they had high fever >38.5 °C, had current bowel obstruction, intra-abdominal abscess, inter-current or opportunistic infection, or refused to use the diet. This study received ethical approval from the hospital ethics committee, which is required for retrospective studies or collection of patient data for clinical reports.
Patients were seen at baseline, week 6 and if responsive to dietary therapy again at week 12.
Patients who deteriorated or refused to continue therapy were considered as failures. Remission was defined by a combination of physician’s global assessment [PGA] and Harvey Bradshaw index [HBI] <5. For this report, complete remission by PGA was defined as complete cessation of symptoms such as diarrhoea, abdominal pain, weight loss and rectal bleeding. The HBI was calculated in all cases from the charts. Response was defined as a drop of at least 3 points or remission.
Specific blood tests such as CRP, and a complete blood count were performed at baseline, week 6 and week 12 for most patients. The normal values for CRP in our lab were <0.5 mg/dL. Patients who had not stopped anti-TNF were asked to perform tests for trough levels and anti-drug antibodies if they were approaching scheduled therapy, otherwise they instituted the diet right away without these tests. Since trough levels of biologics were not available or covered by insurance during the early period of this report, or if patients had desisted from their biologic due to LoR, trough levels were not obtained in all patients.
2.1. Dietary therapy
Patients were treated with one of two dietary strategies dependent upon disease severity at presentation. Those who were treated as outpatients received the CDED with PEN 50% of calculated calories [maximum 1250 kcal/day] for the first 6 weeks. Patients could choose one of two liquid 1 kcal/mL polymeric formulas [Modulen, Nestle; or Pediasure, Abbott Nutrition], based on preference, availability and insurance coverage.
Hospitalized paediatric patients with severe relapses were placed on an abbreviated course of EEN with one of the three polymeric formulas [those mentioned above or Osmolite, Abbott Nutrition] for 2 weeks, based on preference and institutional availability. Once improvement was established, they were transitioned to the CDED with PEN for an additional 6 weeks as above.
Patients who were admitted because of severe flares could either drink the formula orally or receive it by nasogastric tube, but all patients were switched to oral formulas and the CDED by the end of week 2.
The CDED is a whole food diet consisting of fruits, vegetables, meats, and complex and simple carbohydrates currently being evaluated in randomized controlled trials. It comprises two stages over 12 weeks. The principles and many of the food components for the first stage of the diet have been published previously.21 Briefly, foods that contain dietary components that could hypothetically degrade the mucous layer, increase intestinal permeability or have been hypothesized to induce dysbiosis with certain organisms are excluded or limited, depending on the food type. The diet limits or eliminates exposure to animal fats, certain cuts and types of meats, gluten, maltodextrins, xantham gum, emulsifiers, sulfites and certain monosaccharides. On the other hand it involves mandatory consumption of certain fruits, sources of resistant starch and specific sources of animal protein on a daily basis for 12 weeks. The first stage lasts for 6 weeks for induction of remission and is more restrictive than the second stage. All patients received dietary counselling regarding use of diet and a three-page handout with clear written instructions, including a list of mandatory, allowed foods, disallowed foods and quantities allowed, as well as an email and hotline for dietary support and recipes. Several of the foods contain specific minimal quantities that must be ingested, and others that specify the maximal amount that may be ingested. All patients were seen at least three times over 12 weeks [baseline, 6 weeks and 12 weeks] for dietary assessment and instruction. Patients who were started on EEN were seen daily during the admission. After the second week patients were transitioned to the CDED. Patients had access to a support system which included a dietary hotline for questions regarding the diet and foods purchased or served, and a special email address staffed by IBD dedicated dieticians trained in use of the diet. Patients could contact the dieticians via email and receive a collection of recommended CDED-compatible recipes that could be used to improve palatability and variety for meals.
All patients received the identical diet for 12 weeks, or until treatment failure was determined and the patient was transitioned to other medical therapies.
Compliance was assessed directly by questions from physicians and dieticians. Patients who could not consume the full amount of recommended formula or alternative formulas could lower the dose to the maximum amount they were willing to drink as long as they kept to the whole food diet.
3.2. Statistical analysis
Data were stored on an Excel spreadsheet [Microsoft] and analysed on SPSS v23 statistical analysis software [IBM]. Distributions of all continuous variables were assessed for normality using the Kolmogorov–Smirnov test. Distributions deviated significantly from normal, so non-parametric measures were used. Continuous variables are presented as mean ± SD as well as median [interquartile range, IQR]. These variables were compared by remission using the Mann–Whitney U test. Nominal variables such as sex are summarized in frequency tables and are presented as n [%]. These variables were compared by remission using the chi square test. HBI at week 6 was modelled using general linear modelling [GLM] including HBI at baseline, age, sex and duration of illness. All tests are two-sided and considered significant at p < 0.05.
3. Results
3.1. Baseline population and treatment at LoR
Twenty-one patients, mean age 22.2 ± 8.9 years [11 adults and ten children] with mean disease duration of 7.6 years were included. Demographic data are presented in Table 1. Seventeen patients [75%] had received dose escalation [10 mg/kg in the case of infliximab] and 16 [76%] patients used dose interval escalation [q 6 weeks for infliximab, q weekly for adalimumab] prior to commencement of dietary therapy. Five patients [all with previous combination therapy] did not receive dose escalation [Table 2]; among these there were two adults who had their therapy discontinued by their physician prior to receiving dietary therapy, and one adult had requested dietary therapy rather than dose escalation. Seven patients had already undergone one intestinal resection and two patients were steroid dependent and active despite receiving a combination of steroids and biologics.
Baseline [n = 21] . | Remission [n = 13] . | Failure [n = 8] . | Total cohort [n = 21] . |
---|---|---|---|
Gender, male n [%] | 9 [69%] | 3 [37.5%] | 12 [57%] |
Age, years [mean ± SD] | 21.2 ± 9 | 23.8 ± 9.2 | 22.2 ± 8.9 |
Patients >18 years, n [%] | 7 [50%] | 4 [50%] | 11 [50%] |
Duration of disease, years [mean ± SD] | 7.2 ± 6.3 | 8.3 ± 6.7 | 7.6 ± 6.3 |
CRP, mg/dL n = 17 [mean ± SD] | 2.4 ± 2.5 | 4.9 ± 4.6 | 3.2 ± 3.4 |
HBI [mean ± SD] | 7.7 ± 2.5 | 11.6 ± 5 | 9.1 ± 4 |
[median, range] | 7 [5.25–9.75] | 11 [7.25–16.75] | 8 [6.25–12] |
Disease severity | |||
Mild 5–7 n [%] | 7 [58%] | 2 [25%] | 9 [45%] |
Moderate 8–16 n [%] | 5 [42%] | 4 [50%] | 9 [45%] |
Severe >16 n [%] | 0 | 2 [25%] | 2 [10%] |
Disease location, n [%] | |||
L1 [ileum] | 5 [38.5%] | 1 [12.5%] | 6 [27.3%] |
L2 [colonic] | 3 [23%] | 3 [37.5%] | 6 [32%] |
L3 [ileocolonic] | 3 [23%] | 1 [12.5%] | 4 [19%] |
L4 [upper GI] | 1 [8%] | 1 [12.5%] | 2 [9%] |
L3 + L4 | 1 [8%] | 2 [25%] | 3 [14%] |
Treatment failed, n [%] | |||
Infliximab [IFX] | 4 [31%] | 3 [37.5%] | 7 [33%] |
Adalimumab [ADA] | 3 [23%] | 1 [12.5%] | 4 [19%] |
IFX + ADA | 6 [46%] | 4 [50%] | 10 [48%] |
Combination therapy | 9 [69%] | 8 [100%] | 17 [81%] |
Baseline [n = 21] . | Remission [n = 13] . | Failure [n = 8] . | Total cohort [n = 21] . |
---|---|---|---|
Gender, male n [%] | 9 [69%] | 3 [37.5%] | 12 [57%] |
Age, years [mean ± SD] | 21.2 ± 9 | 23.8 ± 9.2 | 22.2 ± 8.9 |
Patients >18 years, n [%] | 7 [50%] | 4 [50%] | 11 [50%] |
Duration of disease, years [mean ± SD] | 7.2 ± 6.3 | 8.3 ± 6.7 | 7.6 ± 6.3 |
CRP, mg/dL n = 17 [mean ± SD] | 2.4 ± 2.5 | 4.9 ± 4.6 | 3.2 ± 3.4 |
HBI [mean ± SD] | 7.7 ± 2.5 | 11.6 ± 5 | 9.1 ± 4 |
[median, range] | 7 [5.25–9.75] | 11 [7.25–16.75] | 8 [6.25–12] |
Disease severity | |||
Mild 5–7 n [%] | 7 [58%] | 2 [25%] | 9 [45%] |
Moderate 8–16 n [%] | 5 [42%] | 4 [50%] | 9 [45%] |
Severe >16 n [%] | 0 | 2 [25%] | 2 [10%] |
Disease location, n [%] | |||
L1 [ileum] | 5 [38.5%] | 1 [12.5%] | 6 [27.3%] |
L2 [colonic] | 3 [23%] | 3 [37.5%] | 6 [32%] |
L3 [ileocolonic] | 3 [23%] | 1 [12.5%] | 4 [19%] |
L4 [upper GI] | 1 [8%] | 1 [12.5%] | 2 [9%] |
L3 + L4 | 1 [8%] | 2 [25%] | 3 [14%] |
Treatment failed, n [%] | |||
Infliximab [IFX] | 4 [31%] | 3 [37.5%] | 7 [33%] |
Adalimumab [ADA] | 3 [23%] | 1 [12.5%] | 4 [19%] |
IFX + ADA | 6 [46%] | 4 [50%] | 10 [48%] |
Combination therapy | 9 [69%] | 8 [100%] | 17 [81%] |
CRP, C-reactive protein; HBI, Harvey Bradshaw Index; GI, gastrointestinal.
Baseline [n = 21] . | Remission [n = 13] . | Failure [n = 8] . | Total cohort [n = 21] . |
---|---|---|---|
Gender, male n [%] | 9 [69%] | 3 [37.5%] | 12 [57%] |
Age, years [mean ± SD] | 21.2 ± 9 | 23.8 ± 9.2 | 22.2 ± 8.9 |
Patients >18 years, n [%] | 7 [50%] | 4 [50%] | 11 [50%] |
Duration of disease, years [mean ± SD] | 7.2 ± 6.3 | 8.3 ± 6.7 | 7.6 ± 6.3 |
CRP, mg/dL n = 17 [mean ± SD] | 2.4 ± 2.5 | 4.9 ± 4.6 | 3.2 ± 3.4 |
HBI [mean ± SD] | 7.7 ± 2.5 | 11.6 ± 5 | 9.1 ± 4 |
[median, range] | 7 [5.25–9.75] | 11 [7.25–16.75] | 8 [6.25–12] |
Disease severity | |||
Mild 5–7 n [%] | 7 [58%] | 2 [25%] | 9 [45%] |
Moderate 8–16 n [%] | 5 [42%] | 4 [50%] | 9 [45%] |
Severe >16 n [%] | 0 | 2 [25%] | 2 [10%] |
Disease location, n [%] | |||
L1 [ileum] | 5 [38.5%] | 1 [12.5%] | 6 [27.3%] |
L2 [colonic] | 3 [23%] | 3 [37.5%] | 6 [32%] |
L3 [ileocolonic] | 3 [23%] | 1 [12.5%] | 4 [19%] |
L4 [upper GI] | 1 [8%] | 1 [12.5%] | 2 [9%] |
L3 + L4 | 1 [8%] | 2 [25%] | 3 [14%] |
Treatment failed, n [%] | |||
Infliximab [IFX] | 4 [31%] | 3 [37.5%] | 7 [33%] |
Adalimumab [ADA] | 3 [23%] | 1 [12.5%] | 4 [19%] |
IFX + ADA | 6 [46%] | 4 [50%] | 10 [48%] |
Combination therapy | 9 [69%] | 8 [100%] | 17 [81%] |
Baseline [n = 21] . | Remission [n = 13] . | Failure [n = 8] . | Total cohort [n = 21] . |
---|---|---|---|
Gender, male n [%] | 9 [69%] | 3 [37.5%] | 12 [57%] |
Age, years [mean ± SD] | 21.2 ± 9 | 23.8 ± 9.2 | 22.2 ± 8.9 |
Patients >18 years, n [%] | 7 [50%] | 4 [50%] | 11 [50%] |
Duration of disease, years [mean ± SD] | 7.2 ± 6.3 | 8.3 ± 6.7 | 7.6 ± 6.3 |
CRP, mg/dL n = 17 [mean ± SD] | 2.4 ± 2.5 | 4.9 ± 4.6 | 3.2 ± 3.4 |
HBI [mean ± SD] | 7.7 ± 2.5 | 11.6 ± 5 | 9.1 ± 4 |
[median, range] | 7 [5.25–9.75] | 11 [7.25–16.75] | 8 [6.25–12] |
Disease severity | |||
Mild 5–7 n [%] | 7 [58%] | 2 [25%] | 9 [45%] |
Moderate 8–16 n [%] | 5 [42%] | 4 [50%] | 9 [45%] |
Severe >16 n [%] | 0 | 2 [25%] | 2 [10%] |
Disease location, n [%] | |||
L1 [ileum] | 5 [38.5%] | 1 [12.5%] | 6 [27.3%] |
L2 [colonic] | 3 [23%] | 3 [37.5%] | 6 [32%] |
L3 [ileocolonic] | 3 [23%] | 1 [12.5%] | 4 [19%] |
L4 [upper GI] | 1 [8%] | 1 [12.5%] | 2 [9%] |
L3 + L4 | 1 [8%] | 2 [25%] | 3 [14%] |
Treatment failed, n [%] | |||
Infliximab [IFX] | 4 [31%] | 3 [37.5%] | 7 [33%] |
Adalimumab [ADA] | 3 [23%] | 1 [12.5%] | 4 [19%] |
IFX + ADA | 6 [46%] | 4 [50%] | 10 [48%] |
Combination therapy | 9 [69%] | 8 [100%] | 17 [81%] |
CRP, C-reactive protein; HBI, Harvey Bradshaw Index; GI, gastrointestinal.
Data regarding combination therapy are presented in Table 2. Seventeen patients [80.9%] had used combination therapy with at least one biologic at the time of LoR.
Patient ID . | Remission . | Disease duration [years] . | Gender, age [years] . | Disease location . | Failed biologics . | Dose escalation . | Combination therapy . | Dietary therapy . | HBI week 0 . | HBI week 6 . | CRP/ESR/CAL week 0 . | CRP/ESR/CAL week 6 . |
---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | NO | 18 | F, 36 | L3 + L4 | ADA + INF | YES | AZA | CDED | 17 | 12 | – ESR 26 | 0.6 ESR 10 |
2 | YES | 7 | F, 28 | L2 | INF | YES | AZA | CDED | 5 | 0 | 1.3 | 1.3 |
3 | YES | 8 | M, 20 | L1 | INF | YES | – | CDED + PEN | 5 | 0 | 8.8 | 0.9 |
4 | NO | 4 | F, 33 | L2 | ADA + INF | YES | AZA | CDED + PEN | 16 | 5 | 0.6 | <0.5 |
5 | NO | 12 | F, 32 | L3 + L4b | INF | NO | AZA | CDED + PEN | 7 | 5 | 1.5 | 0.15 |
6 | YES | 6 | M, 22 | L2 | ADA | YES | – | CDED + PEN | 9 | 0 | CRP<0.5 ESR 20 | CRP<0.5 ESR 12 |
7 | YES | 4 | F, 15.5 | L3 | ADA + INF | YES | MTX | CDED + PEN | 8 | 0 | 3.8 | <0.5 |
8 | YES | 11 | F, 36 | L1 | ADA | NO | AZA | CDED | 12 | 0 | CRP<0.5 Cal 470 | <0.5 – |
9* | YES | 5 | M, 15.2 | L4b | INF | YES | – | CDED + PEN | 7 | 0 | <0.5 | <0.5 |
10 | YES | 2.1 | M, 17.5 | L3 | ADA | YES | MTX | CDED + PEN | 10 | 0 | CRP<0.5 CAL>630 | CRP<0.5 CAL 116 |
11 [T1] | YES | 4.8 | M, 13.8 | L2 | ADA + INF | YES | – | CDED | 6 | 0 | 1.09 | 0.9 |
11 [T2] | YES | 5.7 | M, 14.7 | L2 | ADA + INF | YES | MTX | CDED | 7 | 0 | 1.86 | 0.4 |
12 [T1] | YES | 8 | M, 13.6 | L3 | ADA + INF | YES | MTX | M-EEN + CDED + PEN | 5 | 0 | 1.6 | <0.5 |
12 [T2] | YES | 13 | M, 18 | L3 | ADA + INF | YES | MTX | CDED + PEN | – | – | 1.1 | <0.5 |
13*† | NO | 18 | F, 27 | L2 | ADA + INF | YES | MTX | CDED + PEN | 18 | 9 | – | – |
14† | NO | 5.5 | M, 17.5 | L2 | INF | NO | MTX | M-EEN + CDED + PEN | 13 | 8 | 6.6 | – |
15 | NO | 4.9 | M, 12.7 | L4b | ADA + INF | YES | AZA[p] | CDED + PEN | 9 | 5 | 12 | 0.8 |
16 | YES | 0.4 | F, 13.8 | L1 | INF | NO | AZA[p] | CDED + PEN | 6 | 0 | 4.2 | <0.5 |
17 | NO | 2 | M, 16.9 | L1 | INF | YES | MTX[p] | CDED + PEN | 5 | 5 | 1.1 | 0.8 |
18 | YES | 7 | M, 22 | L3 | ADA + INF | YES | MTX[p] | CDED + PEN | 7 | 0 | 3.3 | 0.9 |
19† | NO | 2 | F, 15 | L3 | ADA | YES | MTX | M-EEN + CDED + PEN | 8 | – | 7.5 | – |
20* | YES | 3 | M, 23 | L1 | ADA + IFX | YES | 6MP[p] | CDED + PEN | 2 | – | CRP<0.5 CAL>300 | – CAL 37.1 |
21 | YES | 25 | M, 43 | L1 | ADA + IFX | NO | AZA[p] | CDED + PEN | 12 | 1 | CRP<0.5 CAL 400 | 0.07 CAL 110 |
Patient ID . | Remission . | Disease duration [years] . | Gender, age [years] . | Disease location . | Failed biologics . | Dose escalation . | Combination therapy . | Dietary therapy . | HBI week 0 . | HBI week 6 . | CRP/ESR/CAL week 0 . | CRP/ESR/CAL week 6 . |
---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | NO | 18 | F, 36 | L3 + L4 | ADA + INF | YES | AZA | CDED | 17 | 12 | – ESR 26 | 0.6 ESR 10 |
2 | YES | 7 | F, 28 | L2 | INF | YES | AZA | CDED | 5 | 0 | 1.3 | 1.3 |
3 | YES | 8 | M, 20 | L1 | INF | YES | – | CDED + PEN | 5 | 0 | 8.8 | 0.9 |
4 | NO | 4 | F, 33 | L2 | ADA + INF | YES | AZA | CDED + PEN | 16 | 5 | 0.6 | <0.5 |
5 | NO | 12 | F, 32 | L3 + L4b | INF | NO | AZA | CDED + PEN | 7 | 5 | 1.5 | 0.15 |
6 | YES | 6 | M, 22 | L2 | ADA | YES | – | CDED + PEN | 9 | 0 | CRP<0.5 ESR 20 | CRP<0.5 ESR 12 |
7 | YES | 4 | F, 15.5 | L3 | ADA + INF | YES | MTX | CDED + PEN | 8 | 0 | 3.8 | <0.5 |
8 | YES | 11 | F, 36 | L1 | ADA | NO | AZA | CDED | 12 | 0 | CRP<0.5 Cal 470 | <0.5 – |
9* | YES | 5 | M, 15.2 | L4b | INF | YES | – | CDED + PEN | 7 | 0 | <0.5 | <0.5 |
10 | YES | 2.1 | M, 17.5 | L3 | ADA | YES | MTX | CDED + PEN | 10 | 0 | CRP<0.5 CAL>630 | CRP<0.5 CAL 116 |
11 [T1] | YES | 4.8 | M, 13.8 | L2 | ADA + INF | YES | – | CDED | 6 | 0 | 1.09 | 0.9 |
11 [T2] | YES | 5.7 | M, 14.7 | L2 | ADA + INF | YES | MTX | CDED | 7 | 0 | 1.86 | 0.4 |
12 [T1] | YES | 8 | M, 13.6 | L3 | ADA + INF | YES | MTX | M-EEN + CDED + PEN | 5 | 0 | 1.6 | <0.5 |
12 [T2] | YES | 13 | M, 18 | L3 | ADA + INF | YES | MTX | CDED + PEN | – | – | 1.1 | <0.5 |
13*† | NO | 18 | F, 27 | L2 | ADA + INF | YES | MTX | CDED + PEN | 18 | 9 | – | – |
14† | NO | 5.5 | M, 17.5 | L2 | INF | NO | MTX | M-EEN + CDED + PEN | 13 | 8 | 6.6 | – |
15 | NO | 4.9 | M, 12.7 | L4b | ADA + INF | YES | AZA[p] | CDED + PEN | 9 | 5 | 12 | 0.8 |
16 | YES | 0.4 | F, 13.8 | L1 | INF | NO | AZA[p] | CDED + PEN | 6 | 0 | 4.2 | <0.5 |
17 | NO | 2 | M, 16.9 | L1 | INF | YES | MTX[p] | CDED + PEN | 5 | 5 | 1.1 | 0.8 |
18 | YES | 7 | M, 22 | L3 | ADA + INF | YES | MTX[p] | CDED + PEN | 7 | 0 | 3.3 | 0.9 |
19† | NO | 2 | F, 15 | L3 | ADA | YES | MTX | M-EEN + CDED + PEN | 8 | – | 7.5 | – |
20* | YES | 3 | M, 23 | L1 | ADA + IFX | YES | 6MP[p] | CDED + PEN | 2 | – | CRP<0.5 CAL>300 | – CAL 37.1 |
21 | YES | 25 | M, 43 | L1 | ADA + IFX | NO | AZA[p] | CDED + PEN | 12 | 1 | CRP<0.5 CAL 400 | 0.07 CAL 110 |
Patients failing diet had last HBI on diet registered.
*Active inflammation by colonoscopy prior to diet.
†Patients failing therapy did not reach week 6 on diet.
[p] = previous combination therapy, not at time of treatment; ADA, adalimumab; INF, infliximab; AZA, azathioprine; 6MP, mercaptopurine; MTX, methotrexate; T1, first treatment; T2, second treatment; M, male; F, female; M-EEN, modified exclusive enteral nutrition; CDED, Crohn’s Disease Exclusion Diet; PEN, partial enteral nutrition; CRP, C-reactive protein; CAL, calprotectin; ESR, erythrocyte sedimentation rate.
Patient ID . | Remission . | Disease duration [years] . | Gender, age [years] . | Disease location . | Failed biologics . | Dose escalation . | Combination therapy . | Dietary therapy . | HBI week 0 . | HBI week 6 . | CRP/ESR/CAL week 0 . | CRP/ESR/CAL week 6 . |
---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | NO | 18 | F, 36 | L3 + L4 | ADA + INF | YES | AZA | CDED | 17 | 12 | – ESR 26 | 0.6 ESR 10 |
2 | YES | 7 | F, 28 | L2 | INF | YES | AZA | CDED | 5 | 0 | 1.3 | 1.3 |
3 | YES | 8 | M, 20 | L1 | INF | YES | – | CDED + PEN | 5 | 0 | 8.8 | 0.9 |
4 | NO | 4 | F, 33 | L2 | ADA + INF | YES | AZA | CDED + PEN | 16 | 5 | 0.6 | <0.5 |
5 | NO | 12 | F, 32 | L3 + L4b | INF | NO | AZA | CDED + PEN | 7 | 5 | 1.5 | 0.15 |
6 | YES | 6 | M, 22 | L2 | ADA | YES | – | CDED + PEN | 9 | 0 | CRP<0.5 ESR 20 | CRP<0.5 ESR 12 |
7 | YES | 4 | F, 15.5 | L3 | ADA + INF | YES | MTX | CDED + PEN | 8 | 0 | 3.8 | <0.5 |
8 | YES | 11 | F, 36 | L1 | ADA | NO | AZA | CDED | 12 | 0 | CRP<0.5 Cal 470 | <0.5 – |
9* | YES | 5 | M, 15.2 | L4b | INF | YES | – | CDED + PEN | 7 | 0 | <0.5 | <0.5 |
10 | YES | 2.1 | M, 17.5 | L3 | ADA | YES | MTX | CDED + PEN | 10 | 0 | CRP<0.5 CAL>630 | CRP<0.5 CAL 116 |
11 [T1] | YES | 4.8 | M, 13.8 | L2 | ADA + INF | YES | – | CDED | 6 | 0 | 1.09 | 0.9 |
11 [T2] | YES | 5.7 | M, 14.7 | L2 | ADA + INF | YES | MTX | CDED | 7 | 0 | 1.86 | 0.4 |
12 [T1] | YES | 8 | M, 13.6 | L3 | ADA + INF | YES | MTX | M-EEN + CDED + PEN | 5 | 0 | 1.6 | <0.5 |
12 [T2] | YES | 13 | M, 18 | L3 | ADA + INF | YES | MTX | CDED + PEN | – | – | 1.1 | <0.5 |
13*† | NO | 18 | F, 27 | L2 | ADA + INF | YES | MTX | CDED + PEN | 18 | 9 | – | – |
14† | NO | 5.5 | M, 17.5 | L2 | INF | NO | MTX | M-EEN + CDED + PEN | 13 | 8 | 6.6 | – |
15 | NO | 4.9 | M, 12.7 | L4b | ADA + INF | YES | AZA[p] | CDED + PEN | 9 | 5 | 12 | 0.8 |
16 | YES | 0.4 | F, 13.8 | L1 | INF | NO | AZA[p] | CDED + PEN | 6 | 0 | 4.2 | <0.5 |
17 | NO | 2 | M, 16.9 | L1 | INF | YES | MTX[p] | CDED + PEN | 5 | 5 | 1.1 | 0.8 |
18 | YES | 7 | M, 22 | L3 | ADA + INF | YES | MTX[p] | CDED + PEN | 7 | 0 | 3.3 | 0.9 |
19† | NO | 2 | F, 15 | L3 | ADA | YES | MTX | M-EEN + CDED + PEN | 8 | – | 7.5 | – |
20* | YES | 3 | M, 23 | L1 | ADA + IFX | YES | 6MP[p] | CDED + PEN | 2 | – | CRP<0.5 CAL>300 | – CAL 37.1 |
21 | YES | 25 | M, 43 | L1 | ADA + IFX | NO | AZA[p] | CDED + PEN | 12 | 1 | CRP<0.5 CAL 400 | 0.07 CAL 110 |
Patient ID . | Remission . | Disease duration [years] . | Gender, age [years] . | Disease location . | Failed biologics . | Dose escalation . | Combination therapy . | Dietary therapy . | HBI week 0 . | HBI week 6 . | CRP/ESR/CAL week 0 . | CRP/ESR/CAL week 6 . |
---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | NO | 18 | F, 36 | L3 + L4 | ADA + INF | YES | AZA | CDED | 17 | 12 | – ESR 26 | 0.6 ESR 10 |
2 | YES | 7 | F, 28 | L2 | INF | YES | AZA | CDED | 5 | 0 | 1.3 | 1.3 |
3 | YES | 8 | M, 20 | L1 | INF | YES | – | CDED + PEN | 5 | 0 | 8.8 | 0.9 |
4 | NO | 4 | F, 33 | L2 | ADA + INF | YES | AZA | CDED + PEN | 16 | 5 | 0.6 | <0.5 |
5 | NO | 12 | F, 32 | L3 + L4b | INF | NO | AZA | CDED + PEN | 7 | 5 | 1.5 | 0.15 |
6 | YES | 6 | M, 22 | L2 | ADA | YES | – | CDED + PEN | 9 | 0 | CRP<0.5 ESR 20 | CRP<0.5 ESR 12 |
7 | YES | 4 | F, 15.5 | L3 | ADA + INF | YES | MTX | CDED + PEN | 8 | 0 | 3.8 | <0.5 |
8 | YES | 11 | F, 36 | L1 | ADA | NO | AZA | CDED | 12 | 0 | CRP<0.5 Cal 470 | <0.5 – |
9* | YES | 5 | M, 15.2 | L4b | INF | YES | – | CDED + PEN | 7 | 0 | <0.5 | <0.5 |
10 | YES | 2.1 | M, 17.5 | L3 | ADA | YES | MTX | CDED + PEN | 10 | 0 | CRP<0.5 CAL>630 | CRP<0.5 CAL 116 |
11 [T1] | YES | 4.8 | M, 13.8 | L2 | ADA + INF | YES | – | CDED | 6 | 0 | 1.09 | 0.9 |
11 [T2] | YES | 5.7 | M, 14.7 | L2 | ADA + INF | YES | MTX | CDED | 7 | 0 | 1.86 | 0.4 |
12 [T1] | YES | 8 | M, 13.6 | L3 | ADA + INF | YES | MTX | M-EEN + CDED + PEN | 5 | 0 | 1.6 | <0.5 |
12 [T2] | YES | 13 | M, 18 | L3 | ADA + INF | YES | MTX | CDED + PEN | – | – | 1.1 | <0.5 |
13*† | NO | 18 | F, 27 | L2 | ADA + INF | YES | MTX | CDED + PEN | 18 | 9 | – | – |
14† | NO | 5.5 | M, 17.5 | L2 | INF | NO | MTX | M-EEN + CDED + PEN | 13 | 8 | 6.6 | – |
15 | NO | 4.9 | M, 12.7 | L4b | ADA + INF | YES | AZA[p] | CDED + PEN | 9 | 5 | 12 | 0.8 |
16 | YES | 0.4 | F, 13.8 | L1 | INF | NO | AZA[p] | CDED + PEN | 6 | 0 | 4.2 | <0.5 |
17 | NO | 2 | M, 16.9 | L1 | INF | YES | MTX[p] | CDED + PEN | 5 | 5 | 1.1 | 0.8 |
18 | YES | 7 | M, 22 | L3 | ADA + INF | YES | MTX[p] | CDED + PEN | 7 | 0 | 3.3 | 0.9 |
19† | NO | 2 | F, 15 | L3 | ADA | YES | MTX | M-EEN + CDED + PEN | 8 | – | 7.5 | – |
20* | YES | 3 | M, 23 | L1 | ADA + IFX | YES | 6MP[p] | CDED + PEN | 2 | – | CRP<0.5 CAL>300 | – CAL 37.1 |
21 | YES | 25 | M, 43 | L1 | ADA + IFX | NO | AZA[p] | CDED + PEN | 12 | 1 | CRP<0.5 CAL 400 | 0.07 CAL 110 |
Patients failing diet had last HBI on diet registered.
*Active inflammation by colonoscopy prior to diet.
†Patients failing therapy did not reach week 6 on diet.
[p] = previous combination therapy, not at time of treatment; ADA, adalimumab; INF, infliximab; AZA, azathioprine; 6MP, mercaptopurine; MTX, methotrexate; T1, first treatment; T2, second treatment; M, male; F, female; M-EEN, modified exclusive enteral nutrition; CDED, Crohn’s Disease Exclusion Diet; PEN, partial enteral nutrition; CRP, C-reactive protein; CAL, calprotectin; ESR, erythrocyte sedimentation rate.
3.2. Dietary therapy
Data regarding choice of dietary therapy are presented in Table 3. Four patients [three adults] refused to drink a formula, and these patients continued with CDED alone. One patient had clinical features with mild diarrhoea, elevated calprotectin and a colonoscopy demonstrating endoscopic recurrence; this patient had an HBI <5 but clinical and endoscopic evidence for activity and was included in the report since faecal calprotectin was available before and after therapy.
Dietary therapy . | Remission [n = 13] . | Failures [n = 8] . |
---|---|---|
CDED + PEN, n [%] | 7 [54%] | 5 [62.5%] |
CDED alone, n [%] | 3 [23%] | 1 [12.5%] |
Modified EEN + CDED, n [%] | 3 [23%] | 2 [25%] |
Total, n [%] | 13 [62%] | 8 [38%] |
Dietary therapy . | Remission [n = 13] . | Failures [n = 8] . |
---|---|---|
CDED + PEN, n [%] | 7 [54%] | 5 [62.5%] |
CDED alone, n [%] | 3 [23%] | 1 [12.5%] |
Modified EEN + CDED, n [%] | 3 [23%] | 2 [25%] |
Total, n [%] | 13 [62%] | 8 [38%] |
CDED, Crohn’s Disease Exclusion Diet; EEN, exclusive enteral nutrition; PEN, partial enteral nutrition.
Dietary therapy . | Remission [n = 13] . | Failures [n = 8] . |
---|---|---|
CDED + PEN, n [%] | 7 [54%] | 5 [62.5%] |
CDED alone, n [%] | 3 [23%] | 1 [12.5%] |
Modified EEN + CDED, n [%] | 3 [23%] | 2 [25%] |
Total, n [%] | 13 [62%] | 8 [38%] |
Dietary therapy . | Remission [n = 13] . | Failures [n = 8] . |
---|---|---|
CDED + PEN, n [%] | 7 [54%] | 5 [62.5%] |
CDED alone, n [%] | 3 [23%] | 1 [12.5%] |
Modified EEN + CDED, n [%] | 3 [23%] | 2 [25%] |
Total, n [%] | 13 [62%] | 8 [38%] |
CDED, Crohn’s Disease Exclusion Diet; EEN, exclusive enteral nutrition; PEN, partial enteral nutrition.
3.3. Response to diet
Clinical response [decrease of at least 3 points or remission] was obtained in 19/21 [90.4%] patients, and remission in 13/21 [62%]. Among patients failing combination therapy, 9/17 [53%] entered remission. Improvement in previously elevated inflammatory markers occurred in 17/21 [81%] patients, while normalization occurred in 9/21 [40.9%]. Changes for individual patients are presented in Table 2 and Figure 2. Among the four patients who used the CDED alone without any liquid formula supplementation, three entered clinical remission.
Changes in parameters of severity and inflammation between weeks 0, 6 and 12 are presented in Figure 3. Baseline severity, age and duration of disease were not different among those entering remission or failing to do so, although the four patients with the highest HBI scores [≥13] did not attain remission [Table 2]. Remission was obtained in 9/13 [69.2%] patients with small intestinal involvement compared to 7/13 [53.8%] patients with colonic involvement. Isolated ileal disease had the highest remission rate [5/6, 83.3%] whereas isolated colitis had the lowest remission rate [3/6, 50%]. Three of the eight failures did not reach 6 weeks on diet and had antibiotics added to their therapy prior to week 6. They were registered as treatment failures. Ten of 13 patients in clinical remission at week 6 continued successfully with their biologic therapy through follow up.
Compliance was assessed by patients self-reporting. Four patients reported lack of compliance. One patient was completely non-compliant and the other three consumed disallowed products occasionally. Among those four patients, one achieved clinical remission.
3.4. Trough levels
Trough levels were obtained for 14/21, but only ten were obtained specifically before use of the diet. Of these, five demonstrated normal to high trough levels [>6 µg/mL, range 6.7–23 µg/mL], three were symptomatic despite trough levels of 3–5 µg/mL after dose escalation and two had trough levels below 2 µg/mL despite dose escalation. Among patients with refractory disease despite good trough levels mentioned above, 3/5 entered remission while two did not.
3.5. Repeat dietary therapy for second LoR
Two patients underwent a second round of dietary treatment, at least 12 months after the first successful induction of remission, because of an additional relapse. In both cases patients obtained remission with dietary therapy the second time as well.
4. Discussion
The advent of biologics targeting TNFα was a huge step forward for patients with CD and their physicians. These medications induce rapid response and remission. However, an increasing number of patients are facing LoR to one or more of these agents,22 despite combination therapy with immune modulators, therapeutic drug monitoring and dose adjustments.23,24 Alternative strategies that can induce remission without additional immune suppression [thus avoiding triple immune suppression in patients failing combination therapy] are an unmet need, especially in the paediatric population for whom newer biologics will not become available in the near future. While dietary therapies have been traditionally relegated to induction of remission at disease onset in children or as an adjuvant maintenance therapy utilizing PEN, they may be able to play a larger role in treatment of active established disease in children and adults.
In the current report, we have demonstrated that a specific feasible dietary strategy may be able to play a pivotal role in regaining remission amongst patients failing medical therapy with biologics. This strategy allowed patients to successfully continue biological therapy, and could possibly serve as a bridge to switch strategies involving slower acting trafficking agents which might need to be employed if anti-TNFs fail. This could work by directly inducing remission, or by reducing exposure to factors triggering inflammation, allowing biological therapies to work better due to a reduction in inflammation. We have previously proposed that diet plays a role in pathogenesis and is one of the factors triggering inflammation in CD.13 From a hypothetical standpoint, if an environmental exposure is not reduced, inflammation may persist, recur or be resistant to treatment.25
This salvage therapy worked best for patients with mild to moderate luminal disease [although this was not significant], which is consistent with our hypothesis that dietary therapy works by reducing triggers for inflammation rather than by a direct anti-inflammatory effect. Support for the first scenario presented above [induction of remission is caused by the diet and not by facilitation of anti-TNF] is supported by the fact that 3/5 patients who had already failed biologics and were off therapy entered remission with the diet and no new medical therapy. Surprisingly, we demonstrated that dietary therapy was effective for induction of remission in patients with refractory disease and an upper range of normal or high trough levels, although the documented number of such patients was small.
This report also differs from induction of remission studies reported by us and other paediatric groups, in that none of the patients were treatment naïve or treated after a short duration of disease, which are positive predictors for response to almost all therapies. In fact the median and mean duration of the disease were 5.6 and 7.6 ± 6.3 years, respectively.
Several aspects of the treatment warrant further discussion. Exclusive enteral nutrition for induction of remission usually entails drinking a liquid formula without access to any food for 6–12 weeks. We had five children with moderate to severe relapses requiring hospitalization who were unwilling to drink formulas or consume their required intake. They received only 2 weeks of EEN by tube followed by the CDED diet and oral PEN at 50% of their energy needs: 5/5 demonstrated improvement and full remission was obtained in 3/5 [60%], demonstrating that even in more difficult cases this may be a successful alternative not requiring a full course of EEN. The other two patients both underwent surgery.
Another interesting outcome was that remission could be obtained just with table foods included in the CDED and without supplemental formula. The effect of the diet seemed independent of the amount of formula consumed, since 3/4 patients who refused to consume any formula entered remission with the diet. This also supports the concept that exclusion of dietary components rather than supplementation is the driving force for the decrease in inflammation. At present, we have data for 18 patients who have been treated with CDED alone [including the four in this report] without supplemental formula, and 14/18 [77%] entered remission just with whole foods in the diet.
We believe that this report adds to the growing body of evidence suggesting that dietary therapy with PEN or exclusion diets may be able to play a larger role for induction of remission and disease control across the disease spectrum, and not just in new onset paediatric disease.24,26 One of the clinical research gaps in this field is the lack of studies demonstrating the ability to use dietary therapy for established long-duration disease or for refractory disease.
Studies previously performed in adults were plagued by methodological problems and high early dropout rates, which may have discouraged further research in to this field.18 A recent meta-analysis found that biological therapy [infliximab] coupled with dietary therapy [PEN with low fat or regular diet] was more likely to have better sustained remission at 1 year than biological therapy alone; no information was available about patients presenting with refractory disease.26 Duration of disease has been a contentious factor since a previous paediatric study demonstrated that new onset treatment-naïve patients are more likely to respond to EEN than those with established disease,27 but even so the remission rates in established disease were still high in this study.
The results of our report and the reports by other authors demonstrating an impact of dietary therapy on different outcomes15,16,21,24,26 suggest that we should keep an open mind about dietary therapy. Perhaps the most interesting speculative aspect about dietary therapy is that it may be the only therapy that addresses the causes of or triggers for inflammation.13
There are several limitations to this initial report including the retrospective nature of the study, the heterogeneous population and the relatively small number of patients. In some cases, patients from outside our institution had their biologics stopped without further optimization or without performing trough levels. A second limitation is that we did not have trough levels and antibodies for most of the patients, as this was not covered by insurance and some of the patients started diet before the trough levels could be obtained. It is interesting that an inability to reach therapeutic trough levels despite dose escalation in active disease, or active disease with high trough levels are both indications for changing therapy, the former within class when possible, and the latter outside of class. In our reported patients with low trough levels, switching within class was not an option with LoR to both available anti-TNFs. Several aspects contribute to the novelty of this report, including the ability to achieve remission in these very challenging patients with treatment refractory disease and using dietary modification involving whole foods. Further investigation is warranted to bridge this research gap.
In conclusion, we have demonstrated that dietary therapy based on an exclusion diet with or without PEN or a very short course of EEN may restore remission in patients failing biological therapies. Use of dietary therapy should be considered in these difficult to treat patients.
Funding
The authors have no funding to declare.
Conflict of Interest
The authors have no relevant conflicts of interest to declare. RSB received grants and support from Nestle for ongoing randomized control trials.
Author Contributions
RSB: concept, design, collection of the data and drafting of the article; CSS: dietary treatment of patients; IC, referral and care of adult patients; HY: referral and care of adult patients; MS: statistical analysis; SBA: statistical analysis; NC: dietary treatment of patients; AL: concept, design and drafting of the article as well as patient care with the diet.
References
Author notes
Conference: Data from this study were presented at the European Crohn’s Colitis meeting, February 2016, Tel Aviv and at IOIBD April 2016