Effects of Crohn’s disease exclusion diet on remission: a systematic review

Abstract

Background:

Dietary therapy may potentially reduce inflammation and promote mucosal healing in patients with Crohn’s disease and is associated with fewer side effects and lower cost compared to medical therapy. Recently the Crohn’s disease exclusion diet (CDED) has been developed to reduce exposure to individualized dietary components which negatively affect the intestine in patients with Crohn’s disease.

Objectives:

This systematic review aimed to explore the effectiveness of CDED in Crohn’s disease patients.

Design:

A systematic review

Data sources and methods:

A systematic search was performed on the PubMed, EBSCOhost, Cochrane library, OVID, Embase, Scopus, and CINHAL to identify relevant clinical trials published from 1 January 2014 to 31 August 2022.

Results:

A total of 1120 studies were identified and 7 studies were finally included in the analysis. The study was reported according to Preferred Reporting Items for Systematic Reviews and Meta-analysis statement.

Conclusion:

Our findings suggested that the use of CDED seemed to be effective for induction and maintenance of remission in children and adults with mild to moderate Crohn’s disease. However, heterogeneity and limitations existed among the studies included. Further investigation in the form of well-designed randomized clinical trials is needed to validate the present findings.

Registration:

PROSPERO registration number CRD42022335453.

Introduction

Crohn’s disease (CD), a subtype of inflammatory bowel disease (IBD), is a chronic inflammatory disease of the gastrointestinal tract with unclear pathogenesis. To the best of our knowledge, the occurrence and development of CD may be related to the dysregulation of immune response caused by the interaction of complex factors, such as genetic susceptibility factors, environmental triggers, and intestinal flora changes. ¹ The global incidence of CD has increased by 4–15% annually over the past three decades, ² with a relatively high and stable rate in industrialized western countries, ranging from 9.6 to 27.4 per 100,000 population in the past two decades. ³ Besides, CD is becoming increasingly prevalent in newly industrialized countries. ⁴

CD is characterized by recurrent episodes of remission and activity, patients with prolonged inflammation may present with complications such as intestinal obstruction, gastrointestinal perforation, bleeding, and cancer. CD occurs primarily in youth, which may lead to increased risk of growth retardation or stagnation and osteoporosis. ⁵ At present there is no cure for CD. The most common mode of treatment for CD is anti-inflammation therapy. Although it can aid mucosal healing, it may also cause immunosuppression and various side effects, and are associated with high treatment costs.^6–8

Diet is one of the key factors in maintaining normal intestinal microenvironment, and strongly related to inflammation.^9,10 The loss of microorganisms known to have beneficial roles in intestinal homeostasis can promote inflammation of the intestinal membrane. One mechanism for the proinflammatory or anti-inflammatory effects of diet is the intermediate effect of diet on the composition and metabolic activity of the gut microbiome. ¹¹ It has a major impact on the development and treatment of CD. In addition, dietary therapy is less costly than medical therapy and more appropriate for patients with milder CD or can serve as a bridge to medical therapy if medication is delayed and may address the issue of diet as an inflammatory trigger. ¹²

Dietary therapies have high safety. Exclusive enteral nutrition (EEN) is observed with high remission rates, which is recommended as a first-line treatment for pediatric CD. ¹³ However, its low palatability limits patients’ compliance. ¹⁰ In 2014, Professor Arie Levine and colleagues developed the Crohn’s disease exclusion diet (CDED), a dietary therapy specifically designed for CD patients. ¹⁴ CDED is a whole food diet combined with partial enteral nutrition (PEN) and excluding the amount of dietary components that have negative effects on the intestines, ⁹ such as processed foods, emulsifiers, animal fat, red and processed meat, carrageenan, gluten, dairy products, etc. In addition, CDED also includes fresh whole foods such as chicken, eggs, potatoes, rice, as well as fruits and vegetables to be the sources of resistant starch and pectin. Thus, the design of CDED is strict which includes mandatory, allowed and not allowed foods in general.

CDED usually consists of two to three phases that change every 6 weeks, with each stage of the diet allowing access to a wider range of specific foods that gradually become easier for the patient. The first phase is the most restrictive, limiting the amount of certain fruits and vegetables in addition to eliminating foods and dietary components that trigger inflammation. PEN, which generally uses polymeric formula (Modulen^®, Nestlé) is added to meet 50% of energy demand. The second phase allows for a wider variety of foods, such as gluten, oily fish, red meat and legumes, and almost all fruits and vegetables can be allowed in the final weeks of this phase. PEN is reduced to provide 25% of energy demand. After 12 weeks of dietary treatment, the maintenance phase begins, with the diet balanced by free meals, allowing intermittent and controlled exposure to certain foods that were not allowed in the first phase. It is recommended to maintain PEN to cover 25% of energy requirements. If a patient in remission exhibits inflammatory activity throughout the maintenance phase, recovery can be assessed by re-executing the most restrictive dietary phase for an acclimatization period, based on the response obtained. Levine et al. and Sigall et al. indicated that this dietary approach might be as effective as EEN, but with better adherence, and might reduce C-reactive protein and calprotectin concentrations or lead to endoscopic remission.^10,14

Although several studies have investigated CDED, to date, no study has attempted to evaluate and synthesize the impact of CDED on clinical remission of CD. Therefore, in this systematic review, we summarize the current evidence on the effectiveness of CDED in achieving CD remission, with the aim of contributing to the refinement and development of CDED and providing scientific, effective and professional dietary support for CD patients.

Materials and methods

This systematic review adhered to Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines ¹⁵ (Supplemental Appendix A) and was registered with the International Prospective Systems Review Registry (PROSPERO), ¹⁶ registration number CRD42022335453.

Search strategy

A systematic literature search was performed by two researchers (H.Z. and Y.L.) who had undergone training on systematic database searching. Our search on PubMed, EBSCOhost, Cochrane library, OVID, Embase, Scopus, and CINHAL was limited to articles published between 1 January 2014 and 31 August 2022. We selected the start time because CDED was first developed in 2014. The search was limited to papers published in English. Reference lists for all eligible articles and related reviews were also searched manually to find additional studies which were not identified by the search strategy. Search terms were limited to MeSH terms and keywords. For example, PubMed search included the following index terms: [‘Crohn Disease’(Mesh) OR Crohn Disease OR Crohn’s colitis OR CD] AND (Crohn’s disease exclusion diet OR CD exclusion diet OR CDED). Detailed search strategies are provided in Appendix B.

Eligibility criteria

Studies that met the following criteria were included in the analysis: (1) study type: randomized controlled trials (RCTs) or quasi-experimental studies, (2) studies involving patients diagnosed with CD, (3) studies involving patients who underwent CDED and whose remission was assessed using quantification methods, including scales and laboratory indicators, (4) the primary outcome was clinical remission of CD, which was measured by pediatric Crohn’s disease activity index (PCDAI) ⩽ 10 or less than 7.5 without height component for children and adolescents aged up to 18 years, or measured by Harvey-Bradshaw Index (HBI) < 5 or Crohn’s disease activity index (CDAI) < 150 for adults. The secondary outcomes were markers of inflammation, measured by C-reactive protein and fecal calprotectin, (5) studies published in English language.

The following studies were excluded: (1) the studies lacked available data, or the full texts were inaccessible, such as reviews, commentary, case report, editorials, conference abstracts. (2) Duplicate publications.

Study selection

All the retrieved records were imported into Endnote (version X9) to remove duplicate records. Two researchers (H.Z. and Y.L.) independently screened the study titles, abstracts, and full texts based on inclusion and exclusion criteria. Any disagreements were resolved by a third researcher (Z.L.) through discussion and studies that met the inclusion criteria selected for further review.

Quality assessment

The quality of the included studies was assessed using the Joanna Briggs Institute (JBI) Critical Appraisal Tools for RCT and quasi-experimental studies. ¹⁷ The JBI appraisal tool for RCTs consists of 13 appraisal questions regarding study design, randomization, intervention, data collection, and instrumentation, homogeneity of study participants, data analysis, and presentation of results. The tool for the quasi-experimental studies consists of nine questions, including clarity of variables, homogeneity of participant characteristics, use of control groups, similarity of routine care provided and outcome measures, completion of follow-up, reliability of the instrument, and appropriateness of statistical analyses. The two authors (H.Z. and Y.L.) independently rated the quality of each item as ‘yes’, ‘no’, ‘unclear’, or ‘not applicable’. When disagreement occurred, a consensus was reached through discussion.

Data extraction

The following data were separately extracted by two researchers and recorded on a standard form: first author, year of publication, country, patient characteristics (patients’ age, sample size, patients’ gender, and patients’ disease), study design, detailed intervention and control, length of follow-up, and outcomes (measures). We contacted the author/s via emails for essential information which could not be found in the article if necessary. Any disagreements between the researchers were resolved through discussion.

Data analysis

Given the heterogeneous nature of the included studies, we did not statistically integrate results in a meta-analysis but employ an in-depth narrative synthesis to clarify the relationship between CDED and remission among people with CD.

Results

Overview of the study selection

Initially, 1120 studies were identified from 7 databases, of which 587 duplicate records were removed. A total of 533 articles proceeded to screening of the titles and abstracts, leaving 15 records for full text review and 518 articles were ultimately excluded. Among the 15 records, 4 records were identified as abstracts. One abstract was inaccessible, and the other three only reported the number of patients in remission, while data related to patient characteristics, dropouts and study outcomes were not available, so these four abstracts were excluded. Finally, seven studies were included in the analysis. Isakov et al. ¹⁸ examined the effectiveness of CDED on clinical remission in patients with pouch inflammation, which was included in the review because of the similarities between the pathogenesis of pouchitis and CD. Details on included studies and selection process are shown in the PRISMA flowchart on Figure 1.

Figure 1.

Flowchart of literature search and study selection.

Characteristics of the included studies

A total of seven studies that investigated the effectiveness of CDED on CD remission were included in this systematic review. Five of these studies were conducted in Israel^{10,12,14,18,19}, one in Israel and Canada, ¹⁰ and two in Poland.^20,21 With regard to the study design, two of these studies were RCTs^10,12 whereas five were prospective nonrandomized studies.^14,18–21 Two studies targeted children,^10,21, three targeted adults.^12,18,20 Sigall et al.^14,19 included both adults and children. The seven studies involved a total of 277 patients, with sample sizes ranging from 15 to 74 patients.

Patient remission was measured based on disease activity indices. Three studies used the PCDAI which is applied to children and adolescents aged up to 18 years.^10,14,21 Three studies used the HBI,^14,19,20 including one that combined HBI and physician’s global assessment (PGA). ¹⁹ One study used the CDAI to evaluate remission in adults. ¹² One study used clinical perianal disease activity index (cPDAI) and modified perianal disease activity index (mPDAI) to assess disease activity in adults with pouchitis. ¹⁸ Additionally, all studies used normalization of inflammatory markers, referred to C-reactive protein (CRP), and fecal calprotectin (FCP) level. Erythrocyte sedimentation rate (ESR) and simplified endoscopic activity score for Crohn’s disease (SES-CD) were also used as secondary outcome measures of remission. The main characteristics of the included studies are illustrated in Table 1. Interventions and outcomes of the included studies are shown in Table 2.

Table 1.

Summary of main characteristics of included studies.

First author	Year of publication	Country	Patient characteristics					Study design
			Age, year (Range/mean ± standard deviation)	Sample size	Patients’ gender (males %)	Patients’ disease	Disease severity
Levine et al. 10	2019	Israel, Canada	CDED + PEN:13.8 ± 2.8 EEN: 14.5 ± 2.6	CDED + PEN:40 EEN: 34	CDED + PEN: 26 (65) EEN: 20 (59)	CD	Mild to moderate a	RCT
Yanai et al. 12	2022	Israel	CDED + PEN: 23–38 CDED alone: 25–39	CDED + PEN:19 CDED alone: 21	CDED + PEN: 10 (53) CDED alone: 8 (38)	CD	Active non-complicated mild-to-moderate b	RCT
Isakov et al. 18	2022	Israel	19–62	15	3 (20)	Pouchitis	Active pouchitis c on stable medical therapy	prospective nonrandomized studies
Sigall et al. 14	2014	Israel	Total: 13.86 ± 4.71 Children: 11.8 ± 3.1 Adults: 19.3 ± 3.9	47	30 (63.4)	CD	Active CD d	prospective nonrandomized studies
Sigall et al. 19	2017	Israel	22.2 ± 8.9	21	12 (57)	CD	Mild to severe failing biological therapy	prospective nonrandomized studies
Szczubełek et al. 20	2021	Poland	31.42 ± 9.01	32	14 (43.7)	CD	Active CD e	prospective nonrandomized studies
Matuszczyk et al. 21	2022	Poland	4–17	48	27 (56)	CD	active CD f	prospective nonrandomized studies

^aPCDAI ⩾ 10 and⩽40; ^bHBI score of 5–14; ^cclinical PDAI (cPDAI) >2 and modified PDAI (mPDAI) ⩾5; ^dPCDAI ⩾ 10 or HBI > 3; ^eCDAI > 150; ^fFCP level ⩾250 µg/g.

Table 2.

Interventions and outcomes of the included studies.

First author	intervention group	Control group	Length of follow-up	Outcomes		Dropouts (n)
				Primary outcome (s)	Secondary outcome(s)
Levine et al. 10	Received the CDED + PEN a	Received EEN: phase I (first 6 weeks): standard of care EEN phase II (from week 7 to 12): 25%PEN	12 weeks	Patient’s tolerance	PCDAI, CRP, ESR, FCP	n = 4 (for intolerance by 48 h)
Yanai et al. 12	Received the CDED + PEN: phase I (first 6 weeks): CDED+(1000 kcal per day) of PEN phase II (from week 7 to 24): CDED+(600 kcal per day) of PEN + calcium supplement	Received CDED with calcium supplement for 24 weeks	24 weeks	HBI	CRP, FCP level, endoscopic remission, SES-CD score	n = 11 (2 lost to follow-up, 6 worsening of disease, 2 intolerance to CDED, and 1 was admitted to hospital for another condition)
Isakov et al. 18	CDED + 2000 international units of vitamin D and 1000 mg calcium of supplements per day	/	24 weeks	cPDAI, mPDAI, CRP, FCP level, endoscopic PDAI score		n = 7 (4 antibiotics, 2 bowel obstruction and 1 drop-out)
Sigall et al. 14	Received the CDED + PEN a	/	12 weeks	HBI, PCDAI	CRP	n = 0
Sigall et al. 19	Received the CDED + PEN a	/	12 weeks	HBI	CRP, ESR, PGA and calprotectin	n = 0
Szczubełek et al. 20	Received the CDED + PEN a	/	12 weeks	CDAI	CDAI, CRP, WBC, IBDQ, laboratory parameters	n = 4 (for refused to continue dietary restrictions)
Matuszczyk et al. 21	Received the CDED + PEN a	/	12 weeks	PCDAI	CRP, ESR and FCP level	n = 11 (6 lack of therapeutic efficacy, 2 difficulty in adherence to the diet, 1 poor tolerance of the enteral formula, and 1 parent made the decision to discontinue)

^aPhase I (first 6 weeks): CDED + 50%PEN

Phase II (from week 7 to 12): CDED + 25%PEN

CDED, Crohn’s disease exclusion diet; CRP, C-reactive protein; EEN, exclusive enteral nutrition; ESR, erythrocyte sedimentation rate; FCP, fecal calprotectin; WBC, white blood cell; HBI, Harvey-Bradshaw Index; IBDQ, Inflammatory Bowel Disease Questionnaire; PCDAI, pediatric Crohn’s disease activity index ; PEN, partial enteral nutrition; cPDAI, clinical perianal disease activity index ; mPDAI, modified perianal disease activity index; PDAI, perianal disease activity index.

Quality assessment of the included studies

The risk of bias assessment is summarized in Tables A1 and Tables A2 (Appendix C). Two RCTs were rated with 10 ‘yes’ responses. Both reported sufficient random sequences generated by computers or random number tables, and delineated allocation concealment explicitly. Blinding was difficult to implement. Both studies clarified the reasons for dropping out, reported prespecified outcomes, and did not selectively report. The five nonrandomized studies applied self-controlled designs, so the items related to the comparison of the two groups were not applicable for evaluation.

Effect of CDED on clinical response and remission in children

All three studies defined clinical response as a drop in PCDAI of at least 12.5 points for children. In the study of Levine et al., ¹⁰ clinical response on an ITT analysis at week 6 indicated no statistically significant difference [34/40 (85%) CDED + PEN versus 29/34 (85.3%), EEN, p = 0.97; Delta 0.3%; 95% CI ±16%; OR 0.97; 95% CI 0.27–3.53]. Later, when Sigall et al. ²² analyzed unpublished data from the study, they found that of the 73 patients who showed up at week 3, 61 patients out of 73 (83.5%) responded to dietary therapy, and 46 of 73 (63%) obtained remission by week 3, but there was no significant difference between both groups. In addition, the other 12 nonresponders at week 3 were characterized by a high rate of nonadherence, as 9 of the 12 nonresponders (75%) had poor adherence to the randomized diet. Sigall et al. ¹⁴ reported that 37 of the 47 patients have obtained clinical response at week 6. However, the results of adults and children were combined for analysis so that clinical responses in children could not be obtained separately. Matuszczyk et al. ²¹ found that of the 29 children who were not in clinical remission (PCDAI ⩾ 10.00) at baseline, 20/29 (68.97%) achieved clinical response after 12 weeks.

Two studies reported the remission of pediatric CD patients at 6 weeks (phase I).^10,14 Levine et al. ¹⁰ found that corticosteroid-free remission (defined as PCDAI ⩽ 10) was realized in 32 out of 40 (80%) CDED + PEN patients, as well as remission by the more stringent PCDAI < 10 was achieved by 32/40 (80%) CDED + PEN patients. Median (interquartile range) PCDAI improved from 25 (20–35) at week 0 (baseline) to 2.5 (0–7.5) at week 6 for CDED + PEN patients. Further study of unpublished data by Sigall et al. ²² found that of 61 patients with dietary reactions at week 3, 46 (75.4%) subsequently achieved remission at week 6. Of the 73 patients in the entire cohort, 49 (67%) achieved remission at week 6 (PCDAI < 10), 46 (94%) responded at week 3, and 40 out of 49 (81.6%) had remission at week 3. They also found that remission at week 3 was a predictor of remission at week 12. On the other hand, noncompliance was predictive for remission at week 12 only for patients receiving CDED + PEN. In the study by Sigall et al., ¹⁴ 24 out of 34 pediatric patients who used CDED + PEN achieved complete remission and of these, 18 (75%) had mild disease, 5 (71%) had moderate disease, and 1 (33.3%) had severe disease.

Three studies reported the remission of pediatric CD patients,^10,14,21 following the gradual introduction of free oral diet at 12 weeks (phase II). Levine et al. ¹⁰ found that at week 12, more CDED+PEN patients achieved sustained remission [28/37 (75.6%), p = 0.01; Delta 30.5%; 95% CI 10.4–52.6%; OR 3.77; 95% CI 1.34–10.59]. Sigall et al. ¹⁴ also noted that at week 12, PCDAI declined significantly compared with week 0 (p = 0.000) and Matuszczyk et al. ²¹ found that 29 out of 48 CDED+PEN patients were in clinical remission at week 12, which was statistically significant compared with week 0 (p = 0.002). Table 3 summarizes the remission rates in children with CD treated with CDED.

Table 3.

Remission rates in children with Crohn’s Disease treated with CDED.

Study	Response criteria	Response at week 6	Response at week 12	Remission criteria	Remission at week 6	Remission at week 12
Levine et al. 10	a drop in PCDAI of 12.5 points	34/40 (85%)	/	PCDAI ⩽ 10	32/40 (80%)	28/37 (75.6%)
Sigall et al. 14		/	/	PCDAI < 7.5	24/34 (71%)	/
Matuszczyk et al. 21		/	20/29 (68.97%)	PCDAI ⩽ 10	/	29/48 (60%)

Effect of CDED on clinical response and remission in adults

Only one study mentioned the effect of CDED on the clinical process. Szczubełek et al. ²⁰ defined clinical response as CDAI < 100. Results showed that 83.3% and 85.7% of adult patients responded at Week 6 and 12, respectively.

Two studies reported the remission of adult patients at 6 weeks and 12 weeks (phase I + phase II).^12,20 Szczubełek et al. ²⁰ found that based on CDAI, 76.7% (95% CI 57.7–90.1) and 82.1% (95% CI 63.1–93.9) of the patients achieved clinical remission after 6 and 12 weeks of treatment, respectively, when compared with the baseline. After 12 weeks of dietary intervention, the levels of calprotectin reduced significantly when compared with the baseline (p = 0.021) but there was no significant difference between week 0 and 6. ²⁰ Yanai et al. ¹² found that CDED, with or without PEN, effectively induced and maintained CD remission in adult CD patients, without significant differences, but this study was not powered to answer this question.

Additionally, Yanai et al. ¹² noted that median HBI score decreased significantly between baseline and week 6, 12, and 24 (phase III) and scores were not significantly different. Isakov et al. ¹⁸ reported a clinical remission (cPDAI sub-score ⩾ 2) rate of 66.7%, 60%, and 46.7% at week 6, 12, and 24, respectively, whereas clinical and endoscopic remission (mPDAI < 5) was observed in 40% and 46.7% of the patients by week 12 and 24, respectively. Compared to baseline, a significant decrease in mPDAI occurred [8.0 (5.0–10.0) versus 4.5 (1.0–9.0), p = 0.007] and in the cPDAI sub-score [4.0 (3.0–4.0) versus 2.0 (0.0–4.0), p = 0.004]. Table 4 summarizes the remission rates in adults with CD treated with CDED.

Table 4.

Remission rates in adults with Crohn’s Disease treated with CDED.

Study	Response criteria	Response at week 6	Response at week 12	Remission criteria	Remission at Week 6	Remission at week 12	Remission at week 24
Yanai et al. 12	/	/	/	HBI < 5	25/40 (63%)	22/40 (55%)	/
Szczubełek et al. 20	CDAI < 100	83.3%	85.7%	CDAI < 150	76.70%	82.10%	/
Isakov et al. 18	/	/	/	cPDAI⩽2; mPDAI < 5	66.7%; 66.7%	60%; 40%	46.7%; 46.7%

CDED, Crohn’s disease exclusion diet.

Effect of CDED on inflammatory markers (CRP and FCP)

Seven studies reported the effect of CDED on CRP. Levine et al. ¹⁰ found that median CRP improved from 23.6 mg/L (9.8 to 54.2) at week 0, to 5 mg/L (2.7–8.0) at week 6 for pediatric CDED + PEN patients (p < 0.001), and normal CRP occurred in 22 of 29 (75.9%) CDED + PEN patients at week 12. FCP dropped significantly between week 0 and week 6 (p = 0.002). Matuszczyk et al. ²¹ noted that 16 out of 48 (33.33%) pediatric CDED + PEN patients had a normal CRP result (i.e. <0.50 mg/dL) after 12 weeks of nutritional therapy and found a statistically significant decrease in FCP level between weeks 0 and 12. Isakov et al. ¹⁸ found a decreased level of CRP and FCP levels at week 12 compared to baseline [11.2 (4.7–25.3) versus 5.6 mg/dL (1.2–14.8), p = 0.015], and [498.0 (37.0–2551.0) versus 367.0 mg/kg (9.0–2100.0), p = 0.084), respectively]. Yanai et al. ¹² observed that median CRP decreased from 14.5 mg/L (IQR 7.7–37.1) to 8.4 mg/L (5.4–18.5) at week 6 (p = 0.0378), which was sustained at week 24 [8.0 mg/L (5.2–19.9), p = 0.0098]. The median concentration of FCP decreased significantly from baseline to week 12 [262 μg/g (IQR 73–1092) versus 97 μg/g (54–212), p = 0.0123]. Sigall et al. ¹⁴ presented significant improvement compared with baseline for CRP and FCP by week 6 and week 12.

On the contrary, Szczubełek et al. ²⁰ found no significant improvement compared with baseline for CRP by week 6 and week 12 (both p > 0.05).

Discussion

This systematic review evaluated the impact of CDED on achieving CD remission. All the included studies identified the beneficial role of CDED on the remission of CD patients, reporting decreased disease activity and improved CRP and FCP levels.

According to the current 2020 CD Treatment guidelines delivered by European Crohn’s and Colitis Organization (ECCO)/European Society of Pediatric Gastroenterology, Hepatology And Nutrition (ESPGHAN), clinical disease activity should not be used alone as the primary criterion for evaluating treatment effectiveness, as it is not a marker of mucosal inflammation. CRP, as an indicator reflecting systemic inflammation, increases with the occurrence of inflammation in the early stage of IBD, but it cannot accurately reflect the degree of mucosal lesions during treatment. A recognized noninvasive marker of mucosal healing is a FCP level below 250.00 µg/g. In addition, mucosal inflammation was found to persist in approximately half of patients in clinical remission. Therefore, it is recommended that FCP levels should be used as the basis for treatment decisions, even in the absence of clinical symptoms of the disease. ²³ The results of FCP levels are more objective and enable the assessment of the full effect of treatment, which is to achieve deep remission and mucosal healing and reduce the risk of disease progression. All the seven included studies showed that CDED could significantly reduce FCP levels in patients. Isakov et al. ¹⁸ reported a significant decline in FCP, but numerically very modest (498 ug/g versus 367 ug/g). This may indicate that the effect of CDED on relieving mucosal inflammation in pouchitis patients is inadequate, and further experiments are needed to confirm it.

The development of CD has been attributed to an interaction between genetics, the environment, gut microbiota, and the immune system.^24,25 CD is mainly treated using immunomodulators or immunosuppressants, such as corticosteroids, methotrexates, thiopurines, and agents like antitumor necrosis factor alpha (TNFα).^23,26 However, immunosuppressants have high costs and are associated with side effects, including a higher risk of infection and cancer, and stunted growth in children.^6,7 An unmet need exists for safer treatments, especially for patients with milder conditions and lower risk of complicated diseases, or those with immunosuppressive disorders.^27–29

Diet has been found to have a strong impact on gut microbiota composition and ecology, which plays a crucial role in regulating metabolic and immune responses.^9,30,31 Current medical treatments are primarily directed against inflammation. Dietary therapy has been shown to reduce inflammation by altering the microbiota and reducing intestinal permeability.^32,33 It is superior to drugs in terms of reduced side effects because it does not increase immunosuppression, has no side effects, and reduces the need to add or change drugs, while providing an indication of whether diet therapy is clinically beneficial. Dietary therapy has emerged as an effective alternative to immunosuppression in mild or uncomplicated CD in which drug treatment may not be necessary. It may also be used as a monotherapy to induce remission at the onset of disease or in patients who have failed to maintain treatment. ¹⁹ Dietary therapy can also be used as a bridge to biotherapy, since the initiation of biologic agents is usually accompanied by regulatory reasons, medical testing, and vaccinations. ²² In addition, during the COVID-19 pandemic, CD patients’ access to specialized health care services is significantly limited, which makes it necessary to find alternatives to immunosuppressive therapy.

EEN, which is recommended by the ECCO and the ESPGHAN as the first-line treatment for mild-to-moderate, recent onset pediatric CD, has been associated with a remission rate of up to 85%.^34,35 However, many patients do not fully comply with EEN, mainly because of its complete exclusion of a natural diet, feeding only on an enteral diet for 6–8 weeks, practical difficulties, lack of motivation and support to complete the treatment, as well as the unpleasant taste of the formulation.^36–38 Compared with EEN, CDED is a long-term strategy that includes a diet based on regular food products, which improves tolerance and cost effectiveness. ¹³ Levine et al. ¹⁰ noted that compared to EEN, tolerance to CDED + PEN was significantly better. Moreover, the remission rate did not differ significantly between CDED + PEN and EEN, indicating that CDED + PEN and EEN were similarly effective at inducing pediatric CD remission.

Avoidant food intake is common in IBD patients. ³⁹ When IBD patients use an independent, unsupervised diet to manage their disease, they may develop restrictive eating behavior that leads to inadequate nutrient intake, thereby increasing the risk of malnutrition. ⁴⁰ CDED offers such patients with a scientific and comparative, and professional support. ⁴¹ Szczubełek et al. ²⁰ used the Inflammatory Bowel Disease Questionnaire to measure the impact of CDED on the quality of life of CD patients and found that after 6 and 12 weeks of intervention, the patients’ quality of life improved significantly when compared with baseline. In general, CDED is beneficial for patients to develop a long-term sustainable healthy lifestyle, which can effectively overcome the difficulties caused by EEN because it allows access to food in a standardized and controlled manner and can be used not only as an induction strategy, but also as a good option for long-term maintenance of remission. ⁴² In studies involving CDED treatment, the study population is mainly children and young adults with short duration of disease and mild to moderate disease activity.^10,14,21 In adults, the disease becomes more severe over time and leads to complications. Despite this, a lower response rate to CDED in adult patients has not been observed in existing studies. CDED is also effective in inducing and maintaining remission in mild to moderate adult patients.^12,18,20

Yanai et al. ¹² found that no statistically significant difference exists between the effect of CDED + PEN and CDED alone in achieving remission in CD patients, suggesting that CDED may be effective in the treatment of CD and its benefits may be due to the exclusion of certain inflammatory components, rather than the effects of PEN. Thus, PEN may not be necessary for remission but is the main source of calcium.^12,14 However, since this study was not powered to answer this question, this conclusion needs to be further confirmed by more clinical studies. Existing studies on PEN have found that although PEN can provide some moderate benefits, clinical responses are not comparable to EEN in inducing remission. ⁴³ In clinical practice, PEN-induced remission is generally not recommended.^43,44 In addition, compared with patients in the CDED plus PEN group, patients in the CDED alone group gained less weight and consumed fewer calories, suggesting that obese patients may benefit from CDED alone. Since overweight and obesity in IBD patients may be associated with poorer outcomes, such as reduced response to biological therapies,^45,46 therapeutic regimen with CDED alone may invoke better outcomes.

Currently, CDED is mainly used for induced remission in patients with uncomplicated mild to moderate CD and can be used as an effective alternative to EEN induction therapy. The wide variety of diets and cultural differences might be a result of globalization, which makes it difficult to standardize dietary treatments. But due to the great versatility of mandated and permitted foods and the use of cooking condiments, each recipe can be tailored to the tastes of patients with different dietary cultures in different countries. To facilitate the application of CDED, health care providers and dietitians can provide practical tools such as mandatory food checklists, recipe brochures for different stages, and samples of weekly meal plans. Individualized dietary guidance for patients based on food preferences and dietary needs should be considered, considering differences among families. In addition, patients should be encouraged to help each other, and patient exchanges should be organized so that they can understand each other and share their experiences, questions, and tips. It is also important to provide patients with a communication channel such as phone or email to ensure treatment compliance, because an early response to dietary therapy (e.g. at week 3) increases the likelihood that remission will be achieved later (e.g. at week 6). If patients are noncompliant early on, they are less likely to be compliant with subsequent dietary therapy and other treatment options should be considered early on.

This study has several limitations. Firstly, the number of articles included in this study and the sample size involved are small, which limits the generalizability of the observations made. Most of the existing study lacked a control group and the open-label observational clinical study. Thus, more randomized trials are needed to obtain more robust evidence. Secondly, the data in the studies included were not suitable for quantitative synthesis. The results from the qualitative analysis might affect the reliability of these findings. Thirdly, because this systematic review only included articles published in English, thus publication bias is inevitable.

To the best of our knowledge, this is the first systematic review to identify the effect of CDED on CD remission. In all seven studies included in this review, a significant effect of CDED on remission was observed in both adult and pediatric patients with mild-to-moderate Crohn’s disease. CDED is a long-term strategy to maintain remission, acting on mechanisms that influence the emergence of inflammation (diet and microbiota), correcting bacterial dysregulation present in these patients, increasing clinical remission time, and promoting healthy lifestyle habits. CDED is as effective as EEN in achieving clinical and biochemical remission but is superior to EEN in tolerability and adherence. CDED can be used as monotherapy, combination therapy, drug demotion, and as a bridge to the initiation of immune suppression. Our study may contribute to further studies of CDED as an alternative to EEN in patients with mild-to-moderate CD. In addition, of the seven articles listed in this systematic review, five were conducted in Israel, one in Israel and Canada and two in Poland. Therefore, extensive research and validation from other parts of the world is needed. Given the limited number of studies in this field, more clinical trials with rigorous study designs and further evidence are needed to verify the effects of CDED.

Conclusions

Our findings suggest that CDED is associated with effective remission of CD. Given the methodological heterogeneity of the studies included in this review, more high-quality RCTs are needed to further support these results. Future studies are needed to focus more on patient outcomes, including quality of life and adherence. Projects on the effect of CDED on remission in CD patients deserve to be further explored in future studies.