An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
James D. Lewis, MD, MSCE, Robert Sandler, MD, MPH, Carol Brotherton, PhD, Colleen Brensinger, MS, Hongzhe Li, Michael D. Kappelman, MD, MPH, Scott G. Daniel, PhD, Kyle Bittinger, PhD, Lindsey Albenberg, DO, John F. Valentine, MD, John Hanson, MD, David Suskind, MD, Andrea Meyer, Charlene W. Compher, PhD, RD, Meenakshi Bewtra, MD, PhD, Akriti Saxena, MD, Angela Dobes, MPH, Benjamin Cohen, MD, Ann D. Flynn, MD, Monika Fischer, MD, Sumona Saha, MD, Arun Swaminath, MD, Bruce Yacyshyn, MD, Ellen Scherl, MD, Sara Horst, MD, Jeffrey R. Curtis, MD, MS, MPH, Kimberly Braly, RD, Lisa Nessel, MSS, MLSP, Maureen McCauley, BS, Liam McKeever, PhD, Hans Herfarth, MD, PhD; DINE-CD Study Group.
Author Information and AffiliationsCrohn disease (CD) is a chronic, relapsing form of inflammatory bowel disease for which there is no known cure. Chronic immunosuppressive therapy is the main therapeutic approach. Many patients with mild to moderate symptoms are interested in alternatives to chronic immunosuppressive therapy, such as therapeutic diets. Existing evidence supports the therapeutic potential of the Mediterranean diet (MD) and the specific carbohydrate diet (SCD) as treatment for mild to moderately active CD; however, the most effective therapeutic diet for CD has not been established.
The Diet to INducE Remission in Crohn's Disease (DINE-CD) study was designed to compare the effectiveness of the MD vs the SCD for treatment of patients with CD who have mild to moderate symptoms.
The DINE-CD study was a multicenter, randomized, comparative effectiveness trial of the MD vs the SCD. The study included adult patients with CD who had mild to moderate symptoms—that is, a short Crohn's Disease Activity Index (sCDAI) score of 176 to 399. Patients were randomly assigned 1:1, stratified by use of biologic medications, to follow either the MD or the SCD for 12 weeks. For the first 6 weeks, participants received prepared meals and snacks according to their assigned diet. After 6 weeks, they were instructed to follow the diet independently and were given an option to purchase prepared meals. The primary outcome was symptomatic remission (sCDAI score <150) at week 6. Secondary outcomes included symptomatic remission at week 12 and the following factors, assessed at weeks 6 and 12: clinical remission based on a composite of symptoms, examination findings, and laboratory studies using the full CDAI (CDAI score <150); fecal calprotectin (FC) response (FC <250 μg/g and reduction by >50% from baseline among those with baseline FC >250 μg/g); C-reactive protein (CRP) response (high-sensitivity CRP [hsCRP] <5 mg/L and >50% reduction from baseline among those with hsCRP >5 mg/L); and scores on the short Inflammatory Bowel Disease Questionnaire (SIBDQ) and Patient-Reported Outcomes Measurement Information System (PROMIS) measures of fatigue, pain, sleep, and social isolation. Comparisons between groups were made using the Cochran-Mantel-Haenszel χ2 test or t test and, within groups, the paired t test.
A total of 194 patients were randomly assigned, but because of an error, 3 participants had a baseline sCDAI score <150 and were excluded from the effectiveness analyses (n = 92 assigned to MD and n = 99 assigned to SCD). Elevated FC, CRP, or inflammation at endoscopy was confirmed at screening in 46% (38 MD; 50 SCD) of participants. Before week 6, 2 participants in the MD group and 3 participants in the SCD group were lost to follow-up. Twelve and 15 participants in the MD and SCD groups, respectively, withdrew from the study before week 6. Within the MD and SCD groups, from week 0 to week 6, there was improvement in the mean [SD] sCDAI score in the MD group (−59.33 [64.53]; P < .0001) and the SCD group (−71.78 [75.94]; P < .0001). However, the magnitude of change did not differ between the treatment groups (P = .23). The percentage of participants who achieved symptomatic remission at week 6 did not differ by diet (43.5% on the MD and 46.5% on the SCD; P = .77). Clinical remission was achieved in 47.8% of patients on the MD and 48.5% of patients on the SCD (P = .93). FC response was achieved in 4 participants (30.8%) on the MD and 8 participants (34.8%) on the SCD (P = .83). CRP response was achieved in only 1 participant (3.6%) on the MD and 2 participants (5.4%) on the SCD (P = .68). At week 12, the percentage of participants who achieved symptomatic remission was 40.2% with the MD and 42.4% with the SCD (P = .87). There was no evidence of heterogeneity of response based on the presence or absence of confirmed inflammation at screening for the primary outcome (test for interaction, P = .68). Adherence to the diet all the time at week 6 of the study was reported by 59 (64%) of those on the MD and 67 (68%) of those on the SCD. By week 12, 39 participants (42%) and 40 participants (40%) reported adherence all the time to the MD and SCD, respectively.
Symptomatic remission was achieved in ≥40% of participants on both diets regardless of confirmed inflammation. CRP response was uncommon. None of the effectiveness measures were significantly different by treatment group.
Despite being provided with meals and snacks per their assigned diet for the first 6 weeks of the study, <70% of participants reported attempting to adhere to either the MD or the SCD all the time.
Note: The material presented in this section is adapted from the following peer-reviewed publication: Lewis JD, Sandler R, Brotherton C, et al; DINE-CD Study Group. A randomized trial comparing the specific carbohydrate diet to a Mediterranean diet in adults with Crohn's disease. Gastroenterology. 161(3):837-852.e9. doi:10.1053/j.gastro.2021.05.047 [PMC free article: PMC8396394] [PubMed: 34052278] [CrossRef]
Crohn disease (CD) is a chronic, relapsing form of inflammatory bowel disease (IBD) for which there is no known cure. Therapies approved by the FDA are designed for the treatment of patients with moderate to severely active CD. In contrast, optimal management of patients with mild to moderate symptoms is not well defined. Many patients with mild to moderate CD are interested in alternatives to chronic immunosuppressive therapy, such as a therapeutic diet.
The etiology of CD is thought to be linked to a genetically predisposed dysregulated response of the immune system to environmental factors, such as the gut microbiome.1 Diet has been linked to the composition of the gut microbiome and the risk of developing CD.1 Several studies have identified a lower risk of CD among populations that adhere to a diet consistent with the traditional Mediterranean diet (MD) (ie, high in fresh fruits, vegetables, nuts, fish, and whole grains and use of olive oil as the predominant fat source)2,3 or, similarly, a diet with low inflammatory potential.4 Following the MD has also been associated with reduced symptoms and improved quality of life (QOL) after CD diagnosis.5 Numerous cohort studies, randomized controlled trials (RCTs), and systematic reviews have also documented the efficacy of this diet to reduce inflammation,6 cardiovascular disease (CVD),7,8 cancer,9 and mortality.10 This is particularly important for patients with CD, as evidenced by a meta-analysis that linked CD with an increased risk of CVD.11 Given the general health benefits of an MD, there are many reasons to recommend a diet consistent with the principles of an MD for patients with CD.1,12
Other therapeutic diets have been recommended for patients with CD. Among the most commonly used in clinical practice is the specific carbohydrate diet (SCD). The SCD was popularized by Elaine Gottschall in the book Breaking the Vicious Cycle13 and has been used by numerous patients with inflammatory diseases. The SCD employs detailed lists of foods that are permitted and not permitted. Fresh fruits and vegetables are universally acceptable, with the exception of certain starchy vegetables, such as potatoes and yams. Certain legumes (eg, lentils, split peas) are permitted, but others (eg, chickpeas, soybeans) are not. Neither are grains. Saccharin and honey are allowed as sweeteners. Canned fruits and vegetables are not allowed because of possible added sugars and starches. Unprocessed meats are permitted without limitation, but processed, canned, and most smoked meats are restricted because of possible sugars and starches used in additives. Milk is not permitted because of its lactose content, but certain cheeses with minimal lactose content are allowed, as is homemade yogurt fermented for 24 hours.
To our knowledge, the SCD has not been thoroughly assessed for the other health benefits that have been demonstrated with the MD; however, several small studies have provided preliminary evidence that the SCD, as described in the Gottschall book, and some modified versions of this diet (such as allowing oats or rice) improve symptoms and reduce bowel inflammation.14-17 Suskind et al17 reported the effectiveness of the SCD in 7 children with CD who were not receiving any immunosuppressive therapies. All had clinical improvement by 3 months, and in those whose inflammatory markers were measured, there was notable improvement. In a small randomized trial of SCD, modified SCD, and a whole-food diet for children with CD, there was evidence of improved CD symptoms and markers of systemic inflammation within 12 weeks after following both the SCD and the modified SCD; however, fecal calprotectin (FC) concentration did not improve in the SCD group by week 12.16 Cohen et al14 studied 10 children who followed the SCD as primary therapy for CD. Video capsule endoscopy was completed at baseline and after 12 weeks of diet therapy. Four of 10 patients achieved complete endoscopic healing, as measured by video capsule endoscopy, and 6 of 10 achieved clinical remission.
Patients with CD may wish to receive dietary advice, but physicians, nurses, and dietitians lack high-quality data to guide their recommendations. For example, a Cochrane systematic review of diet for induction of remission in CD concluded that all studies provided low- or very low-quality evidence and that they were too heterogeneous for formal meta-analysis.18 Additionally, major medical society guidelines generally lack any guidance on the use of therapeutic whole-food diets as treatment for CD.19 Thus, there is a major knowledge gap regarding dietary recommendations for patients with CD.
This RCT, the Diet to INducE Remission in Crohn's Disease (DINE-CD) study, was inspired in part by a survey of members of the IBD Partners participant-powered research network (PPRN) (https://www.pcori.org/sites/default/files/PCORI-RI-Kappelman009-English-Abstract.pdf), who identified research on the role of diet to manage IBD as a top priority for patient members. Thus, in conjunction with the PPRN, the Crohn's & Colitis Foundation and study investigators sought to compare the efficacy of the MD and the SCD for treatment of patients with CD who have mild to moderate symptoms, via an RCT. We tested the hypothesis that the SCD would be superior to the MD in achieving symptomatic remission, clinical remission, and reduction in biomarkers of inflammation among patients with CD who have mild to moderate symptoms.
Patients and other stakeholders were engaged in the research at several levels. One key stakeholder, the Crohn's & Colitis Foundation, is the leading global support organization for patients with IBD. Angela Dobes, MPH, a foundation employee, was the administrative principal investigator (PI) of the study and worked in collaboration with James D. Lewis, MD, MSCE, the scientific PI. The foundation provided financial support for the study by paying for the prepared meals and snacks that were supplied to study participants. In addition, the foundation's Clinical Research Alliance, which is a federation of academic and community-based clinical investigators who provide care to patients with CD and conduct CD-related clinical research, played an important role. All but 4 of the participating sites were Clinical Research Alliance members, and the PI routinely provided updates on study progress at biannual Clinical Research Alliance meetings. In addition, the Alliance's Data Coordinating Center was used for this study. The foundation provided in-kind support through direct-to-patient advertising; by coordinating a media day about the trial; and through access to its IBD Plexus Biosample Repository, where plasma samples were stored.
The DINE-CD study benefited from the expertise of a steering committee composed of physician-scientists with expertise in CD and clinical investigation and patients with CD (Appendix A). Two of the patients were identified from the IBD Partner's PPRN. Some of the physician-scientists were also patients with CD or the related form of IBD, ulcerative colitis. The steering committee assisted the PI and the Data Coordinating Center in making decisions about study design, protocol deviations, protocol amendments, data analysis, data interpretation, and report and presentation preparation.
The IBD Partners Patient Governance Council (PGC) served as an external advisory board for the DINE-CD steering committee (Appendix B). The PGC conducts monthly meetings, and while the study was being conducted, the PI or project manager provided updates on study progress and sought feedback on topics such as promoting the study to patients, amending protocols, and placing the results into a context that was meaningful to patients. The PGC played an important role in the decision to broaden the inclusion criteria to those with and without elevated markers of inflammation.
The relationships among the steering committee, external advisory committee, and study team are displayed in Appendix Figure A.
Note: The material presented in this section is adapted from the following peer-reviewed publication: Lewis JD, Sandler R, Brotherton C, et al; DINE-CD Study Group. A randomized trial comparing the specific carbohydrate diet to a Mediterranean diet in adults with Crohn's disease. Gastroenterology. 161(3):837-852.e9. doi:10.1053/j.gastro.2021.05.047 [PMC free article: PMC8396394] [PubMed: 34052278] [CrossRef]
The DINE-CD study was a parallel-group RCT comparing the effectiveness of the MD and the SCD to resolve symptoms and inflammation among patients with CD.
The DINE-CD study was conducted in 33 sites across the United States (Appendix C) as a PCORnet PPRN demonstration project. Enrollment in the trial occurred between September 29, 2017, and October 8, 2019.
Potential study participants were identified and recruited by local site investigators during routine clinical care by referral from other providers, or through local and national advertisements (eg, clinicaltrials.gov). The local study team introduced potentially eligible participants to the DINE-CD study and provided them with the opportunity to receive answers to their questions. Subsequently, informed consent was provided by all participants. No study procedures were conducted before obtaining informed consent. Following the informed consent process, participants were screened for eligibility by the local study team through a combination of medical record review and patient interviews.
Study participants were adult patients (aged >18 years) with CD, established by usual endoscopic, histologic, and radiologic criteria and with mild to moderate symptoms, defined as a short Crohn's Disease Activity Index (sCDAI) score >175 and <400. The sCDAI is a weighted index derived from the original CDAI, which is the most commonly used disease-activity assessment employed in CD clinical trials.20 The sCDAI includes 3 patient-reported symptoms: (1) number of mostly liquid bowel movements per day, (2) severity of abdominal pain (assessed from none to severe), and (3) general well-being (assessed from generally well to terrible). Participants were required to have internet access to complete daily online symptom surveys. Exclusion criteria were as follows:
A 4-week washout period was required for patients who had recently discontinued thiopurine or methotrexate, and an 8-week washout period was required for those who had discontinued natalizumab, vedolizumab, anti-TNF-α therapy, or ustekinumab.
Participants were not compensated for participation in the study other than through the provision of food, as described below.
A computer-generated randomization order, stratified by whether participants were currently receiving therapy with a biologic drug, was used to assign them in a 1:1 ratio to the MD or SCD via random block sizes of 2 and 4. Investigators and participants were not aware of which diet would be assigned next (ie, blinded allocation).
Because the SCD employs a strict list of foods that are permitted or not permitted, while the MD is built more around relative amounts of foods (eg, small amounts of red meats or sweets vs large amounts of fruits and vegetables), the SCD may be more difficult for some patients to follow. In contrast, because the MD does not have a strict list of foods that one should or should not eat, this may represent a challenge for other patients. Similarly, some patients might have less access to the necessary ingredients or less experience preparing foods consistent with the diets. To address these challenges, the DINE-CD trial included provision of meals for the first 6 weeks of the trial. During this phase, participants received a weekly delivery of prepared meals consistent with their assigned diet (breakfast, lunch, dinner, and 2 snacks per day). Meals were prepared by Healthy Chef Creations (Orlando, FL) according to menus developed by the food vendor in consultation with study dietitians. Meals provided to women totaled 2500 kcal per day, while those for men totaled 3000 kcal per day. Participants were not required to consume all the food. Participants assigned to the SCD received a 3-day starter diet as recommended in Breaking the Vicious Cycle.13 Meals were designed to be heated in an oven or microwave. No other preparation was required. Example menus are included in Appendix D.
After the first 6 weeks, participants received information about how to purchase and prepare their own meals and snacks. They were provided with meal planning guidance via the DINE-CD study website. Separate websites were created for each study group. A dietitian was available to answer questions. Participants had the option of purchasing some or all of their meals from Healthy Chef Creations.
The primary and key secondary outcomes were measured 6 and 12 weeks after participants started the diet. The primary outcome measure was symptomatic remission at week 6, defined as sCDAI score <150 in the absence of initiation or increase of any CD medications. Key secondary outcomes were clinical remission (CDAI score <150), FC response (reduction of FC to <250 μg/g and by >50% from screening among those with screening FC >250 μg/g), and C-reactive protein (CRP) response (reduction in high-sensitivity CRP [hsCRP] to <5 mg/L and >50% reduction from screening among those with screening hsCRP >5 mg/L). FC and hsCRP assays were conducted by LabCorp.
The LabCorp normal range or FC is <16 to 50 μg/g. For hsCRP, LabCorp notes low risk at <1.00 mg/L, average risk at 1.00 mg/L to 3.00 mg/L, and high risk at >3.00 mg/L. There is no single concentration for FC or hsCRP that has been consistently determined to best reflect active inflammation in CD.21,22 The steering committee selected FC >250 μg/g and hsCRP >5 mg/L to be sufficiently above the upper limit of normal such that a 50% reduction would reflect a meaningful change but not so high that reduction to this level would not be clinically meaningful.
Additional dichotomous outcome measures at weeks 6 and 12 included a 2-component patient-reported outcome measure (PRO2 remission), defined as mean daily liquid or very soft stool frequency ≤1.5 and abdominal pain ≤1, neither of which being worse than at baseline; combined symptomatic remission and FC response; combined symptomatic remission and CRP response; combined PRO2 remission and FC response; and combined PRO2 remission and CRP response. Additional continuous outcome measures included sCDAI; CDAI; the Short Inflammatory Bowel Disease Questionnaire (SIBDQ);23 Patient-Reported Outcomes Measurement Information System (PROMIS) measures for fatigue (Short Form [SF] 7a, Item Bank v1.0), social isolation (SF-4a, Item Bank v2.0), pain interference (SF- 6a, Item Bank v1.0), and sleep (SF-8a, Item Bank v1.0);24 Routine Assessment of Patient Index Data (RAPID3; a measure of arthritis symptoms); and the Bath Ankylosing Spondylitis Functional Index (BASFI) among those with Inflammatory Back Pain Screening score of 4 or 5 out of 5, per the guidance of the Assessment of SpondyloArthritis International Society.25 PROMIS measures are scaled to T scores, where higher scores reflect more of the condition, 50 is comparable to the average for the US general population, and the SD is 10. The primary and secondary outcomes were assessed again at week 12, although the week 6 outcomes were considered the primary analysis.
For all dichotomous outcomes, participants who withdrew before the time of assessment or were lost to follow-up were categorized as having treatment failure. For continuous outcomes, missing data were imputed using baseline observation carried forward.
Confirmed inflammation at screening was defined as hsCRP >5 mg/L, FC >250 μg/g, or presence of inflammation on endoscopy in the 3 months before screening, based on retrospective review of procedure reports by an investigator at the clinical site. All procedure reports were also reviewed by an investigator (A.S.) from the Data Coordinating Center; any disagreement was resolved by the study PI (J.D.L.) without knowledge of treatment group or outcomes. Colonoscopy was not a required screening procedure. Rather, we included data from endoscopic procedures performed for clinical purposes in the 3 months before screening.
Adherence was measured at weeks 3, 6, 9, and 12 via participant self-report on adherence to the study diet during the previous week, according to the following scale: followed the diet all the time, some of the time, or none of the time. Participants completed a 24-hour dietary recall at baseline, during week 6, and during week 12. These data were used to compute the Alternate Mediterranean Diet (aMED) score.26 The aMED score was adapted for use in an American population from an index developed for a Greek population.27 The aMED score is based on consumption of 9 food types, where 1 point is given for intake below the sex-specific median for red and processed meat and for intake above the sex-specific median for whole grains other than potatoes, vegetables, fruits, nuts, legumes, fish, and the ratio of monounsaturated to saturated lipids. One point is given for alcohol intake between 10 and 25 g per day for men and 5 to 15 g per day for women. Thus, a higher aMED score is consistent with greater adherence to an MD and has been linked to lower overall, cancer-related, and coronary heart disease-related mortality27 and plasma biomarkers of inflammation.26
Adverse events (AEs) were assessed at screening, week 6, and week 12 or the final study visit in the case of early withdrawal. AEs were categorized according to the Common Toxicity Criteria for Adverse Events. Serious AEs (SAEs) included death, life-threatening AEs, inpatient hospitalization or prolongation of stay, persistent or significant disability, congenital anomaly or birth defect, or other medically significant event deemed as such by the site investigator.
Each clinical site underwent monitoring when it had randomly assigned 4 participants and again when it had randomly assigned at least 10 participants. Monitoring focused on regulatory documentation, consent process, and eligibility, because the primary and key secondary outcomes were based on either self-reported symptoms or laboratory test values. To minimize missing data, we collected the components of the sCDAI through a text-message link to an electronic symptom diary. In addition, we sent an email reminder the following morning if the sCDAI had not been completed.
Early in the trial, an error was discovered in the data management system that resulted in 7 participants being randomly assigned who had an sCDAI score of ≤175. Specifically, the abdominal pain component of the sCDAI was mistakenly coded from 1 to 4 rather than 0 to 3, resulting in an inflated sCDAI score. Before analyzing any data, we elected to include these 7 participants in the safety analysis, as well as the 4 participants who had sCDAI scores between 151 and 175 in the efficacy analyses.
With 97 participants per group, the study had 80% to 90% power, with a type I error of 5% to detect a 20% absolute difference in the effectiveness of the 2 diets depending on the success rate in the MD group. Our PPRN PGC opined that a smaller difference was unlikely to justify the challenges of following a strict restriction diet.
We established the time period for the intervention based on interviews with patients during the project's design phase. These patients noted that they would be unlikely to continue with a highly restrictive diet, such as the SCD, if they did not experience symptomatic benefit within 6 weeks.
Except where specified, all analyses adhered to the principle of intention to treat such that participants' outcomes were considered in the group to which they were randomly assigned regardless of whether they followed the diet. P < .05 was considered statistically significant for the primary and key secondary outcomes. Descriptive data are reported as median and interquartile range (IQR) or counts and percentage. Comparisons between the treatment groups employed standardized mean differences (SMDs), where an SMD of 0.20 to 0.49 is considered small, 0.50 to 0.79 is medium, and >0.79 is large.28 Within-group analysis of change in continuous outcome measures used a paired t test, and comparison of magnitude of change between groups used unpaired t tests for symptoms and PROMIS measures. Wilcoxon sign rank and rank sum tests were used for within- and between-group comparisons of change in concentration of FC and CRP, respectively. Dichotomous outcomes were compared between treatment groups using the Cochran-Mantel-Haenszel χ2 test. Continuous variables were compared between groups using linear regression, adjusted for the stratification factor. Although this was an RCT, we prespecified that we would test for residual confounding by the following variables for the primary and key secondary outcomes using logistic regression: age; sex; smoking status; time since diagnosis with CD; evidence of ongoing inflammation at screening, defined as hsCRP >5 mg/L; FC >250 μg/g or colonoscopy demonstrating mucosal breaks within 3 months before screening; duration of CD; presence of colonic and/or rectal disease; use of corticosteroids during the trial; and use of immunomodulator drugs during the trial. After learning that most patients had previous biologic drug exposure, we conducted a post hoc analysis that stratified whether the participant had ever used anti-TNF-α drugs. As an exploratory analysis, we examined the proportion of participants with a rise in FC >250 μg/g and hsCRP >5 mg/L at weeks 6 and 12.
Prespecified tests for HTE evaluated sex, presence or absence of evidence of ongoing inflammation at screening, duration of CD, presence of colonic and/or rectal disease, use of corticosteroids during the trial, use of biologic therapy during the trial, number of previous biologic therapies, and previous surgery for CD.
We hypothesized that the diets might differ in efficacy in patients with and without evidence of ongoing inflammation at baseline. Specifically, we hypothesized that the SCD might be more effective in patients without markers of inflammation who would be more likely to have IBS in addition to CD due to some similarities between the SCD and a diet low in fermentable oligo-, di-, and monosaccharides and fermented polyols (low FODMAP) and a gluten-free diet.29-32 We hypothesized that both diets, but SCD in particular, might be more effective in more recently diagnosed patients and those with isolated ileal disease.33-35 Similarly, we hypothesized that patients treated with corticosteroids or biologics might be less likely to respond to both diets than those who did not require these therapies. Analyses for HTE used conditional logistic regression to account for stratified randomization and an interaction term between treatment and the potential effect modifier.
A prespecified per-protocol analysis was limited to participants who reported that they attempted to follow the diet all the time in the week before the week 6 visit. The efficacy in this group was compared with that in those who reported following the diet less than all the time and in those with missing data for adherence to the diet.
Although it was not possible to blind the study participants or the evaluators to the treatment assignment, all primary analyses were finalized before we revealed the treatment assignment to the steering committee.
To extract DNA from samples, we used the DNeasy PowerSoil kit (QIAGEN). We employed the Nextera XT DNA library prep kit (Illumina) to prepare libraries and the HiSeq 2500 System (Illumina), using 2 × 125-bp chemistry, to sequence them. We analyzed shotgun metagenomic data with Sunbeam, a user-extendable bioinformatics pipeline that we developed for this purpose.36 Quality control steps were performed by the default workflows in Sunbeam, which are optimized to remove host-derived sequences and reads of low sequence complexity. Kraken was used to estimate the abundance of bacteria.37 Sample similarity was assessed by Bray-Curtis and Jaccard distances, and community-level differences between groups were assessed using permutational multivariate analysis of variance (PERMANOVA). The abundance of genes and taxa were analyzed at a community level using pairwise distance between samples and visualized with principal coordinates analysis. Linear mixed-effects models were used to detect differences in taxon abundance between sample groups. P values from multiple testing procedures were corrected to control for a specified false-discovery rate or Bonferroni method. CRP and FC values were log transformed for inclusion in linear mixed-effects models.
As required in the initial inclusion criteria, participants had to have active inflammation documented by any of the following: FC >250 μg/g, hsCRP >5 mg/L, or ulceration of the small bowel and/or colon consistent with a Simple Endoscopic Score for Crohn's Disease38 ≥6 (or, if disease was limited to the ileum and/or right colon, a combined score ≥4 in these 2 segments), as documented in routine clinical practice within 4 weeks of screening. Six months after recruitment began, the study steering committee decided to broaden the inclusion criteria to all patients with CD who had mild to moderate symptoms, regardless of whether there was evidence of active inflammation. Additionally, to shorten the time from screening to randomization, we decided to stratify randomization based only on use of biologic medications rather than to stratify randomization based on use of biologic medications, FC level, and hsCRP level. The decision to modify the inclusion criteria was made after consultation with the PGC and PCORI leadership. At the time of protocol amendment, 44 patients had been screened, of whom 30 had screen failure, 8 were in screening, and 6 had been randomly assigned.
Note: The material presented in this section is adapted from the following peer-reviewed publication: Lewis JD, Sandler R, Brotherton C, et al; DINE-CD Study Group. A randomized trial comparing the specific carbohydrate diet to a Mediterranean diet in adults with Crohn's disease. Gastroenterology. 161(3):837-852.e9. doi:10.1053/j.gastro.2021.05.047 [PMC free article: PMC8396394] [PubMed: 34052278] [CrossRef]
We screened 460 individuals, of whom 263 failed to meet 1 or more of the eligibility criteria and 3 elected not to participate in the study after meeting the eligibility criteria but before being assigned to a diet. Thus, the study population comprised 194 individuals, all of whom were included in the safety analyses. Because of baseline sCDAI scores <150, 1 participant who followed the MD and 2 who followed the SCD were excluded from the efficacy analyses. Thus, 191 individuals were included in the efficacy analyses. Thirty-three participants withdrew from the study before week 6, and 37 participants withdrew between weeks 6 and 12. Figure 1 summarizes recruitment, diet allocation, and early discontinuation. Ninety-three participants were randomly assigned to follow an MD and 101 an SCD.
Study Enrollment, Allocation of Diets, and Follow-up.
Overall, 63% of participants were female, 91% were White, and 4% were of Hispanic ethnicity. The median BMI at screening was 25.1 (IQR, 21.8-29.4). Nonstricturing and nonpenetrating disease behavior was reported for 61% of participants. At screening, 57% of participants were taking a biologic medication, 67% had used 1 or more anti-TNF-α medications in the past, and 11% had used 3 or more anti-TNF-α medications. Oral corticosteroids were used by 7% of participants and rectal corticosteroids by 2% at screening. Confirmed evidence of inflammation at baseline by hsCRP, FC, or colonoscopy was documented in 47% of participants.
Table 1 describes the characteristics of the 2 cohorts in the efficacy analyses. Covariates were generally well balanced between treatment groups, although participants in the MD group were slightly older (37.0 years vs 36.0 years; SMD, 0.21) and more likely to be women (68.5% vs 58.6%; SMD, 0.21). In contrast, the participants in the MD group were less likely to have FC >250 μg/g (14.6% vs 24.0%; SMD, 0.24) and to be currently taking oral mesalamine (5-ASA) (9.8% vs 17.2%; SMD, 0.22). Only 3 participants were current smokers, all in the MD group. During screening, MD participants reported slightly greater sleep disturbance (57.3 vs 54.5; SMD, 0.25), and the SCD participants reported slightly less social isolation (47.2 vs 48.4; SMD, 0.24).
Baseline Characteristics of Participants in the Efficacy Analyses (N = 191).
The percentage of participants who achieved the primary outcome—symptomatic remission at week 6—was 43.5% in those assigned to the MD and 46.5% in those assigned to the SCD (P = .77) (Figure 2).
Primary and Key Secondary Outcomes After 6 Weeks of Diet Therapy.
Within the MD and SCD groups, from week 0 to week 6, there was improvement in sCDAI, CDAI, and SIBDQ scores and in fatigue, sleep interference, pain, and social isolation as measured by PROMIS scales (P < .02 for all outcomes in both groups). However, the magnitude of change did not differ between the treatment groups (P > .2 for all comparisons) (Table 2). Mean (SD) percentage change in weight was −2.6% (3.3%) on the MD and −2.6% (3.7%) on the SCD (P = .91).
Change in Symptoms From Screening to Week 6 by Treatment Group.
The level of hsCRP did not change significantly in either group from screening to week 6. Reduction in FC concentration was significant only in the SCD group (median reduction [IQR], −140 [−321 to 0] μg/g; P = .0001 within group, P = .44 between groups) (Table 3).
Change in CRP and FC Concentration From Screening to Week 6 Among Those With Elevated CRP or FC at Screening.
The percentage of patients achieving each key secondary outcome did not differ between treatment groups at week 6 (Figure 2). Clinical remission was achieved in 47.8% of participants on the MD and 48.5% on the SCD (P = .92). Among those with elevated FC at screening (MD, n = 13; SCD, n = 23), FC response was achieved in 4 participants (30.8%) on the MD and 8 participants (34.8%) on the SCD (P = .83). Among those with elevated hsCRP at screening (MD, n = 28; SCD, n = 37), CRP response was achieved in only 1 participant (3.6%) on the MD and 2 participants (5.4%) on the SCD (P = .68). Because of the slight imbalance of several baseline characteristics, we used regression models to assess for residual confounding despite the randomized design. None of the factors tested altered the odds ratio (OR) for symptomatic remission or clinical remission by 10% or more. For FC response, the final model included duration of CD of greater or less than 10 years as a covariate. The results were comparable to the overall analysis, with an adjusted OR of 1.55 (95% CI, 0.34-7.01; P = .57). Too few patients achieved 50% reduction in hsCRP to allow for adjusted analyses.
The percentage of participants achieving other secondary outcomes at week 6 did not differ between the 2 treatment groups (Table 4). Among the participants achieving FC response at week 6, 6 of 8 on the SCD (75%) also achieved symptomatic remission at week 6 compared with 1 of 4 (25%) on the MD. Overall, combined symptomatic remission and FC response rates were 7.7% on the MD and 26.1% on the SCD (P = .25).
Rates of Achieving Secondary Outcomes.
At week 12, the percentage of participants who achieved symptomatic remission was 40.2% with the MD and 42.4% with the SCD (P = .87) (Figure 3). Clinical remission at week 12 was achieved in 46.7% on the MD and 40.4% on the SCD (P = .28). Among those with elevated FC at screening, FC response was achieved in 1 participant (7.7%) on the MD and 6 participants (26.1%) on the SCD (P = 0.20). Among those with elevated hsCRP at screening, CRP response was achieved in 2 participants (7.1%) on the MD and 4 participants (10.8%) on the SCD (P = .55). Among the participants achieving FC response at week 12, 3 of 6 on the SCD (50%) achieved symptomatic remission compared with 0 of 1 (0%) on the MD. Combined symptomatic remission and FC response rates at week 12 were 0% on the MD and 13.0% on the SCD (P = .23). Similarly, combined symptomatic remission and CRP response rates at week 12 were 3.6% on the MD and 10.8% on the SCD (P = .27) (Table 4). Changes in CRP and FC between screening and week 12 among those with elevated concentrations at screening are summarized in Table 5.
Primary and Key Secondary Outcomes After 12 Weeks of Diet Therapy.
Change in CRP and FC Concentration From Screening to Week 12 Among Those With Elevated CRP or FC at Screening.
A post hoc analysis was performed to assess change in the individual items of the sCDAI within and between groups at week 6 and week 12. Liquid stool frequency, abdominal pain, and general well-being each improved with both diets from screening to week 6 and screening to week 12 (P < .001 for all comparisons). The magnitude of change in these symptoms was not statistically significantly different between the diet groups (P > .1 for all comparisons).
Exploratory analyses of worsening symptoms or increased inflammation based on a rise in FC or hsCRP levels by week 6 did not show significant differences between treatment groups. Fifteen participants in the MD group and 18 in the SCD group withdrew from the study before week 6. Among participants with data available at week 6, only 2 had a rise in sCDAI score of 100 points, both in the SCD group (P = .17) (Table 6). A total of 3 MD participants (6.3%) and 2 SCD participants (4.3%) experienced a rise in FC level above 250 μg/g (P = .67), while 8 MD participants (16.7%) and 9 SCD participants (17.3%) had a rise in hsCRP level above 5 mg/L (P = .88).
Worsening of Symptoms and Inflammation Over the First 6 Weeks of the Clinical Trial.
We used a post hoc analysis to examine the proportion of patients starting corticosteroids by week 6. This change occurred in 4 patients in the MD group and 2 in the SCD group (P = .43). By week 12, 7 patients on the MD and 3 patients on the SCD started steroids (P = .20).
Stratified results to assess for HTE are summarized in Figure 4. At screening, 88 participants had confirmed inflammation, defined as an elevated hsCRP or FC concentration or a colonoscopy demonstrating mucosal breaks (MD, n = 38; SCD, n = 50). Among these, symptomatic remission was achieved in 50% of participants on the MD and 48% on the SCD (P = .86), while among those without documented inflammation, symptomatic remission was achieved in 40% of participants on the MD and 46% on the SCD (P = .68; test for interaction, P = .68). No significant HTE was detected for other variables, including duration of CD (P = .08), colonic involvement (P = .33), use of a biologic medication during the trial (P = .49), number of previous biologic medications used (P = .09), previous bowel resection surgery (P = .17), sex (P = .20), or use of oral steroids (P = .36).
Stratified Analyses to Assess for HTE.
The SIBDQ and PROMIS measures for fatigue, pain interference, social isolation, and sleep disturbance improved during the course of the trial regardless of the assigned diet (Table 2). However, there was no significant difference between treatment groups at week 6 or 12 in any of these QOL measures (P > .3 for all comparisons) (Table 7).
Comparison of Patient-Reported Outcomes and QOL Measures at Week 6 and Week 12.
Most patients did not have severe back or joint symptoms at enrollment (Table 1). Among the patients with an Inflammatory Back Pain Screen25 score >3, the median BASFI scores at weeks 6 and 12 were comparable between groups (Table 8). A sensitivity analysis comparing median BASFI scores among patients with an Inflammatory Back Pain Screen score ≥3 produced similar results (week 6, P = .16; week 12, P = .23). Similarly, among the patients with a RAPID3 score >12 at screening, there was no difference in the median RAPID3 score at week 6 or 12 between groups (Table 8).
Back and Joint Pain at Weeks 6 and 12 Among Those With Symptoms at Screening.
Adherence to the study diet was based on self-report. Adherence all the time in the sixth week of the study was reported by 59 (64%) of those on the MD and 67 (68%) of those on the SCD. Among those reporting full adherence during the sixth week, symptomatic remission was achieved in 49% of the MD participants and 52% of the SCD participants (P = .70). Similar comparative effectiveness was observed for those reporting adherence to the diet less than all the time and those with missing adherence data (test for interaction, P = .98); however, absolute remission rates were lower in those missing adherence data (3 of 16 on the SCD and 2 of 14 on the MD) (Figure 4). By week 12, 39 participants(42%) and 40 participants (40%) reported adherence all the time to the MD and SCD, respectively. Among participants on the MD, 14 (15%) reported purchasing food from Healthy Chef Creations during the 12th week of the study compared with 8 participants (8%) on the SCD.
We used 24-hour dietary recalls to further characterize change in diet based on the aMED. At screening, participants had relatively similar aMED scores (MD: median [IQR], 3 [2-4]; SCD: median [IQR], 3 [2-4]; SMD, 0.27). The aMED score increased in both groups (P < .0001 for MD and SCD) such that the median aMED score was comparable between groups at week 6 (MD: median [IQR], 5 [4-5]; SCD: median [IQR], 4 [3-5]; P = .93). In both groups, there was a significant increase in consumption of fruits (MD: P = .002; SCD: P < .0001) and vegetables (MD: P = .03; SCD: P = .0002) from screening to week 6.
The diets were relatively well tolerated over the course of the study. SAEs were reported by 2 participants in both groups (P > .99) in the first 6 weeks. These AEs included 1 participant having small bowel obstruction; 1 participant experiencing worsening of CD symptoms; 1 participant with abdominal pain, nausea, vomiting, and diarrhea; and 1 participant with abdominal pain. By week 12, SAEs were reported by 5 participants on the MD and 3 on the SCD (P = .48), with 1 small bowel obstruction in each diet group. Having any AE was reported by 22 participants (24%) on the MD and 29 participants (29%) on the SCD in the first 6 weeks (P = .43) and by 34 participants (37%) on the MD and 40 participants (40%) on the SCD by 12 weeks (P = .77). Most AEs were gastrointestinal (GI) related, with abdominal pain being the most common GI complaint in both groups during the first 6 weeks (Table 9).
Adverse Events.
Fecal microbiome was characterized in terms of α and β diversity, with α diversity measures including richness and Shannon diversity index. Richness is a measure of the number of species present. The Shannon diversity index accounts for both the number of species present and the evenness of the abundance of the different species. Richness was assessed as the number of species at a rarefying level of 1000 reads. The 2 diet groups had comparable richness and Shannon diversity indices, and there was no significant change in diversity over the course of the study (Figure 5). A measure of the variability of the taxa between samples, β diversity is characterized in Figure 6 and demonstrates distinct gradients on principal component 1 (PC1) and PC2 that were characterized by the increasing relative abundance of Bacteroides vulgatus on PC1 and Proteobacteria Enterobacteriaceae on PC2. There was decreasing relative abundance of Firmicutes on PC2, including Faecalibacterium prausnitzii, Eubacterium eligens, and Eubacterium rectale. Over the course of the study, β diversity changed slightly, but this change was independent of diet. Previous intestinal surgery was included in the PERMANOVA model because it was strongly related to β diversity (PERMANOVA P values: time in study, P = .02; diet, P = .22; interaction of time in study and diet, P = .99; previous surgery, P = .001).
Microbiome α and β Diversity Stratified by Diet and Time in Study.
β Diversity Over the Course of the Study.
Neither α nor β diversity at screening was associated with achieving symptomatic remission after adjusting for assigned diet (richness, P = .88; Shannon diversity, P = .90; β diversity, P = .94). Similarly, at week 6, neither α nor β diversity was associated with symptomatic remission (richness, P = .35; Shannon diversity, P = .20; β diversity, P = .46) or assigned diet (richness, P = .10; Shannon diversity, P = .26; β diversity, P = .26). In analysis limited to those with elevated CRP at screening, β diversity was not associated with log CRP at any time point (P > .3 for all comparisons). In analysis limited to those with elevated FC concentration at screening, week 6 log FC concentrations were associated with β diversity after adjusting for diet (PERMANOVA P = .02); significant associations between log FC concentrations and β diversity were not evident at screening or at week 12 (P = .3; P = .12). The week 6 association was not significant after Bonferroni correction for 6 total comparisons.
Note: The material presented in this section is adapted from the following peer-reviewed publication: Lewis JD, Sandler R, Brotherton C, et al; DINE-CD Study Group. A randomized trial comparing the specific carbohydrate diet to a Mediterranean diet in adults with Crohn's disease. Gastroenterology. 161(3):837-852.e9. doi:10.1053/j.gastro.2021.05.047 [PMC free article: PMC8396394] [PubMed: 34052278] [CrossRef]
Patients with CD and their physicians could benefit from high-quality RCTs to guide dietary recommendations. This RCT is the largest trial to date to compare the effectiveness of the SCD and the MD in treating symptoms and inflammation in patients with CD. Although there was significant improvement in symptoms among patients on both diets, we observed lack of superiority of either diet to achieve symptomatic remission at 6 or 12 weeks and no evidence of HTE based on the presence or absence of confirmed inflammation during screening. Because symptoms do not always correlate strongly with inflammation in patients with CD,39 we also examined biomarkers of inflammation. Among the subgroup with elevated FC at screening, FC response by week 6 was achieved by >30% of patients following both diets and did not differ between the diets. CRP response at week 6 or week 12 was achieved by <11% of patients following both diets. Thus, although symptomatic remission was common, few patients achieved the treatment goal of combined symptomatic remission and resolution of inflammation. The results of this trial, the greater ease of following the MD, and other health benefits associated with the MD6-10 suggest that MD is as good as or a better dietary option than SCD for most patients with CD who have mild to moderate symptoms.
Given the high rates of symptomatic remission, it is not surprising that participants in both treatment groups demonstrated improvement in QOL measures, including IBD-specific QOL and more general symptoms, such as fatigue. We are not able to determine from this study whether the improvement represents regression to the mean, the impact of participating in a clinical trial, or a true biological effect related to the diets. If it is the last, it may relate to the consumption of a diet sourced from fresh ingredients.
Both the SCD and the MD can be difficult to follow. An important feature of the trial was the provision of prepared food for the first 6 weeks that had only to be heated. This inclusion resulted in relatively high self-reported adherence to the study diets, which improves the internal validity of the study. Whether similar results can be achieved without providing meals to participants will need to be assessed in future studies. The lower self-reported adherence in the second 6 weeks of the study suggests that adherence may decline significantly when patients receive instructions about the diet but are expected to buy ingredients and prepare meals themselves. Alternatively, this change in adherence could reflect the challenge of maintaining a diet pattern that differs from one's usual diet. Both diets investigated in this trial had some similarities, which included an emphasis on fresh fruits and vegetables. The result was that both groups had an increase in aMED score during the trial. The lack of difference in symptomatic outcomes and QOL between treatment groups could be the result of this or other shared similarities, such as both diets being prepared with fresh ingredients.
Given the comparative effectiveness design, it is not possible to state definitively that either diet is better than one's usual diet. The 2 study diets differed from participants' usual diets as evidenced by the increase in the aMED score in both groups. The increase in aMED score was primarily in response to increased fruit and vegetable intake and the ratio of monounsaturated to polyunsaturated lipid intake. The high percentage of participants who achieved symptomatic remission suggests that it may not be necessary to recommend a low-residue diet for patients with mild to moderate symptoms of CD.
In designing the DINE-CD trial, we considered including a group in which the participants were assigned to follow their usual diet. We opted not to use this design to avoid bias that might result from participants knowing that they were being assigned to no treatment for their symptoms for up to 12 weeks. Specifically, we anticipated that participants randomly assigned to follow their usual diet with no additional intervention would be more likely to withdraw from the trial before the end of the study, which would have biased the study to showing a benefit of the study diets. This is one of the many challenges of studying dietary interventions for symptomatic diseases.
Overall, the study diets appeared safe, with few SAEs. The most common AEs related to both diets were GI events. Because clinicians may be concerned that delaying changes to medical therapy while attempting diet-based therapy may predispose their patients to significant worsening of inflammation, we conducted an exploratory analysis of the rise in CRP or FC levels. Over the first 6 weeks, a rise in CRP >5 mg/L occurred in approximately 17% of participants on both diets. A rise in FC >250 μg/g was uncommon, occurring in 6.3% of the MD group and 4.3% of the SCD group. Thus, there was little evidence of significant harm from a 6-week trial of diet-based therapy for patients with CD who had mild to moderate symptoms.
Modest weight loss was observed in both treatment groups. Given that participants were consuming fewer prepackaged, ultraprocessed foods and more fruits and vegetables, this is not surprising. We encourage providers and patients to monitor for excess weight loss when using these or any other therapeutic diet to treat CD.
Another important feature of this trial was the inclusion of patients with and without documented inflammation. The presence of inflammation was based on elevated CRP or FC or on documentation of inflammation via an endoscopic procedure that was performed in the weeks leading up to randomization. Because colonoscopy was not part of screening, we relied on the results of colonoscopies performed for clinical indications. We then categorized the presence or absence of inflammation based on procedure reports abstracted both by the local investigator and a central reader. In previous studies, the Crohn's Disease Exclusion Diet (CDED) demonstrated significant benefit in patients with documented inflammation,40 while the low FODMAP diet improved symptoms in patients without evidence of ongoing inflammation.30 The MD and CDED share some features, such as an emphasis on consumption of chicken and fresh fruits and vegetables and limited intake of dairy and confectionaries. Some aspects of the SCD resemble a low FODMAP diet, such as avoidance of dairy, bread, and grains and unlimited consumption of unprocessed meats. In this trial, when stratifying the analysis based on the presence or absence of documented inflammation at baseline, we observed no evidence of HTE for the outcome of symptomatic remission.
This study was not designed to assess endoscopic healing, a target of some treatment guidelines.41 Exclusive enteral nutrition has been demonstrated to induce endoscopic healing.42,43 It is less clear whether diet therapies other than exclusive enteral nutrition can effectively induce endoscopic healing in individuals with CD. An uncontrolled study of the CDED documented endoscopic healing in >70% of patients who underwent colonoscopy, although not all patients had a colonoscopy.44 In contrast, a small study of 7 patients following a modified SCD demonstrated that none had complete endoscopic healing.45 Whether the lack of endoscopic healing was the result of the modified version of the SCD, insufficient adherence to the diet, lack of therapeutic benefit of the SCD for endoscopic healing, or other factors could not be determined from that study. Although this study demonstrated a significant reduction in FC levels among those with elevated FC at screening who followed the SCD, larger studies using colonoscopy-based outcomes are needed to further define the extent of endoscopic and histologic healing that can be expected.
We established the time period for the intervention based on interviews with patients during the project's design phase. These patients noted that they would be unlikely to continue with a highly restrictive diet such as the SCD if they did not experience symptomatic benefit within 6 weeks. Based on a survey conducted with patients who had undertaken an SCD to reduce CD and ulcerative colitis symptoms, this might not have been enough time to receive a benefit: 34% of these patients said it took >3 months to achieve resolution of symptoms (ie, clinical remission).46 However, our analyses of outcomes at 12 weeks were similar to those at 6 weeks, suggesting that trial duration was not a significant limitation.
The study was powered to detect a difference in rates of symptomatic response. We had less power to examine difference in effectiveness for CRP response and FC response. The magnitude of difference in FC response at week 6 was only 4%; however, by week 12, the difference was 18.4% in favor of SCD, albeit the increased difference in FC response rates was mostly because of low rates of FC response at week 12 in the MD group. We cannot exclude the possibility that in a larger trial, SCD would be associated with greater FC response rates than MD. A similar pattern was not observed for CRP response.
As with medications, one can expect that adherence to a diet is necessary to achieve optimal results. For example, we previously observed greater reduction in FC among patients receiving exclusive enteral nutrition than among those receiving partial enteral nutrition.47 A strong commitment is required to strictly follow either the MD or the SCD. To facilitate adherence, we provided all the necessary food for the first 6 weeks. We documented comparably high rates of self-reported dietary adherence in the first 6 weeks. To explore the potential impact of adherence, our subgroup analysis among patients reporting consumption of the diet all the time demonstrated results similar to those of the primary analysis. Thus, although provision of food to participants likely optimized adherence in this trial and improved the internal validity, such provision may result in an overestimate of the effectiveness that would be expected when implementing these diets in general practice.
Incomplete follow-up has the potential to bias study results. The proportion of participants who withdrew from the study was comparable between treatment groups. For the primary outcome measure (symptomatic remission) and key secondary outcomes (clinical remission, CRP response, and FC response), those who withdrew from the study were considered as having treatment failure. This approach minimizes missing data and the risk of bias for these key outcome measures. We cannot exclude the possibility that the characteristics of the participants who had incomplete follow-up differed between study groups. In theory, this could have biased the results of some analyses, particularly those of continuous outcome measures, such as QOL measures. Because the results for these secondary outcomes were comparable to the primary and key secondary outcomes, such bias is less likely to have occurred.
Participants were required to have internet access so that they could complete study survey tools. This requirement may limit the generalizability of the results. An additional limitation of the trial was the high percentage of White women participating in the study. Generalizability to Americans of other races is unknown at this time. Our stratified analyses and test for HTE did not find significant differences in the comparative effectiveness of the 2 diets between men and women, although numerically there was some evidence of reduced effectiveness of the MD in men. Finally, this study was conducted exclusively in the United States. Given different dietary patterns around the world, the generalizability of the results beyond the United States is unknown.
The mechanisms by which therapeutic diets may improve CD symptoms and inflammation remain to be fully elucidated. Among the hypothesized mechanisms are altering the gut microbiome or the metabolome, changing the ratio of omega-3 and omega-6 fatty acids, promoting short chain fatty acid production by intestinal bacterial fermentation, and eliminating additives.1,12 Healthy Chef Creations used fresh ingredients in preparing meals for both study diets; thus, there was a similar lack of food additives. The lack of reduction in inflammatory markers, particularly CRP, challenges the hypothesis that highly processed foods contribute to the ongoing inflammation associated with CD. Future studies addressing the role of highly processed foods and food additives in the inflammatory process for other diseases are needed. Because CD has been associated with alteration in the gut microbiome (dysbiosis)48,49 and some researchers have hypothesized that effectiveness of SCD is mediated through alteration of the gut microbiome, we used whole-genome sequencing to compare the fecal microbiome among those following the 2 diets. The richness and Shannon diversity were comparable between groups and remained stable throughout the study; β diversity changed slightly over the course of the study. The small changes in β diversity are consistent with previous research on the effect of diet and inflammation on composition of the gut microbiome.48,50 This finding was not related to the diet or symptomatic remission but was weakly associated with FC concentration. Future research is needed to define the mechanisms of diet in the management of CD and the mechanisms that drive CD-related dysbiosis and to understand the interaction of diet, dysbiosis, inflammation, and symptoms among patients with CD.
As noted previously, the comparative effectiveness design of our study precludes assessment of whether either of these diets is superior to continuing one's usual diet. Future research should compare these diets with other diets recommended in the treatment of CD. Similarly, future research could assess whether these diets, when used in conjunction with a new medication, could increase the effectiveness over either intervention alone.
To the extent that diet becomes a mainstay of the CD therapeutic armamentarium, additional research is needed on methods to optimize adherence to diets. This study relied on provision of prepared foods as an introduction to the diets, which allowed time for participants to adapt to the diet and plan for how they would provide their own meals in the future. However, this approach may not be practical or economically feasible, and other strategies should be developed.
It is possible that the effectiveness of either or both of the diets could be improved by slightly modifying the composition. The challenge in identifying individual foods that have favorable or harmful effects is the need to control all of the other food that people consume. Observational research designs could address this question, but they are subject to confounding and other forms of bias. Randomized trials to determine whether individual foods are beneficial or harmful could be considered while holding the base diet constant or within a fixed range, but the challenge and cost of such studies is substantial. Additionally, this consideration assumes that the base diet is not having an effect on the underlying disease.
The future of nutrition-based therapies may lie in personalized nutrition. It is certainly possible that different patients have different dietary needs. Tools are needed to predict which patient will respond to which diet. Such tools would minimize the need to try multiple different diets until the correct one is found and could dramatically improve providers' ability to provide dietary advice to patients with CD. Similarly, future research should focus on how an individual's change in diet is related to outcomes.
Note: The material presented in this section is adapted from the following peer-reviewed publication: Lewis JD, Sandler R, Brotherton C, et al; DINE-CD Study Group. A randomized trial comparing the specific carbohydrate diet to a Mediterranean diet in adults with Crohn's disease. Gastroenterology. 161(3):837-852.e9. doi:10.1053/j.gastro.2021.05.047 [PMC free article: PMC8396394] [PubMed: 34052278] [CrossRef]
The MD and the SCD were well tolerated among patients with CD who had mild to moderate symptoms. Symptomatic remission was common with both diets and did not appear to be influenced by the presence or absence of confirmed inflammation before randomization. Neither diet was associated with normalization of CRP concentration. Given the lack of improvement in CRP, additional RCTs are needed to validate the reduction in FC levels observed after 6 weeks of following the SCD. For patients with CD who had mild to moderate symptoms, the MD may be preferred to the SCD because of its well-documented general health benefits6-10 and relative ease of implementation.
We thank Ariel Myatt, MS, for assistance in creating manuscript figures.
Research reported in this report was funded through a Patient-Centered Outcomes Research Institute® (PCORI®) Award (PPRND-1507-31465). Further information available at: https://www.pcori.org/research-results/2016/comparing-two-diets-decrease-symptoms-crohns-disease-dine-cd-study
Steering Committee Members (PDF, 121K)
External Advisory Committee Members (PDF, 98K)
Participating Sites and Local PIs (PDF, 101K)
Example Menus (PDF, 162K)
Lewis JD, Sandler R, Brotherton C, et al. (2021). Comparing Two Diets to Decrease Symptoms from Crohn's Disease – The DINE-CD Study. Patient-Centered Outcomes Research Institute (PCORI). http://doi.org/10.25302/10.2021.PPRND.150731465
The [views, statements, opinions] presented in this report are solely the responsibility of the author(s) and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute® (PCORI®), its Board of Governors or Methodology Committee.
This book is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License which permits noncommercial use and distribution provided the original author(s) and source are credited. (See https://creativecommons.org/licenses/by-nc-nd/4.0/
Your browsing activity is empty.
Activity recording is turned off.
See more...
