Introduction: Adherence to the Mediterranean diet (MD) was shown to be associated with decreased disease activity in adult patients with Crohn’s disease (CD). Nevertheless, data on its association with fecal calprotectin (FC), particularly in children, remain limited. This study aimed to assess the association between adherence to the MD and FC as an indicator of mucosal healing in patients who are predominantly in remission while undergoing biological therapy. Methods: This was a cross-sectional study among children with CD. Adherence to MD was evaluated using both the KIDMED questionnaire and a food frequency questionnaire (FFQ). Israeli Mediterranean Diet Adherence Screener (I-MEDAS) score was calculated, and FC samples were obtained. Results: Of 103 eligible patients, 99 were included (mean age 14.3 ± 2.6 years; 38.4% females); 88% were in clinical remission, and 30% presented with elevated FC. The mean KIDMED score was higher among patients who had FC <200 μg/g compared to patients with FC >200 μg/g (5.48 ± 2.58 vs. 4.37 ± 2.47, respectively; p = 0.04). A moderate correlation between the KIDMED score and the I-MEDAS score was observed (r = 0.46; p = 0.001). In a multivariate regression analysis, adherence to MD was associated with decreased calprotectin levels, OR 0.75 [95% CI: 0.6–0.95], p = 0.019. Vegetable consumption was found to be inversely associated with elevated FC (0.9 portion/day [0.3–2.9] in FC >200 μg/g vs. 2.2 portions/day [0.87–3.82] in FC <200 μg/g; p = 0.049). Conclusions: In children with CD who are mostly in clinical remission under biological therapy, high adherence to MD is associated with decreased FC levels. Encouraging vegetable consumption, especially during remission, may benefit these patients.

Diet plays an important role in inflammatory bowel diseases (IBDs), particularly in Crohn’s disease (CD) [1]. There has been a marked increase in the prevalence of CD among children over the last decade, especially in newly industrialized countries that are adopting a Western lifestyle [2]. This increase could be driven by exposure to different environmental factors and specifically dietary components characterizing the Western diet, such as processed foods and high-fat, high-sugar foods, which promote pro-inflammatory intestinal immune responses [3].

Mediterranean diet (MD) is considered a healthy dietary approach that is associated with increased life expectancy, reduced risk of major chronic diseases such as cardiovascular disease, obesity, hypertension, and diabetes mellitus and is associated with an improved quality of life (QOL) and well-being [4, 5]. Additionally, recent data in healthy volunteers showed an association with beneficial microbiome-related metabolomic profiles [6]. The MD is characterized by high consumption of fruit and vegetables, olive oil, legumes, whole grains, nuts, and seeds, moderate consumption of fish, poultry, and dairy foods with low consumption of red meat and processed food [7]. With regards to IBD, adherence to MD was shown to be associated with a lower risk of developing CD, but not with ulcerative colitis, in two Swedish large prospective cohort studies [8]. Several studies reported that adherence to MD was low among patients with IBD [9‒11] and that adherence is higher among patients with inactive disease than patients with active disease [7, 12]. In addition, adherence to MD was negatively associated with disease activity, inflammatory markers, and QOL in adult patients with CD [7]. Several groups have shown that in patients with an active form of IBD, adherence to MD was associated with lower fecal calprotectin (FC) ranges using different cutoffs [13, 14]. The role of MD in children with CD is not fully elucidated. Recently, several anecdotal studies were published showing an association with inflammation in children as well [14, 15]. However, none of the studies investigated the role of MD among children who are mostly in clinical remission under biological therapy.

There are several methods to assess adherence to the MD, and these vary between studies as there are different MD types available in different regions and populations. The KIDMED questionnaire is a validated tool to assess compliance with MD in children [16], and the I-MEDAS was designed to assess compliance with MD in adults and was adjusted to the Israeli population [17].

In recent years, the use of FC has been increased in clinical practice to optimize treatment decisions while reducing the burden of repeated endoscopies [18]. FC is an established, noninvasive, easily available, and reliable surrogate marker for assessing mucosal healing (MH) in patients with CD having high sensitivity at a low threshold [18, 19]. According to STRIDE II, decreased FC to an acceptable range is considered an intermediate therapeutic target in IBD [20]. In addition, in patients with clinical remission under biological agents, FC levels might remain above the normal range in some patients [21], which might predict a later relapse [22].

Therefore, the current study aimed to evaluate the association between adherence to MD and FC as an indicator of MH in a cohort of predominantly biologic-treated children with CD, mostly in clinical remission. In addition, we aimed to compare the adherence to MD between children with CD and healthy children in Israel and to assess whether there is an association between KIDMED and I-MEDAS.

Study Population

This prospective cross-sectional study included children and young adolescents aged 6–20 with an established diagnosis of CD who attended the outpatient clinic or came for biologic infusion therapy in two tertiary IBD centers in Israel between January 2020 and April 2021. Patients were eligible if they were on stable doses of any disease-related medications for at least 12 weeks or did not receive any medication. Patients were excluded in case of positive stool culture for enteric pathogens or Clostridioides difficile toxin, current disease-related complications including abscess, fistula, or any active extraintestinal manifestations, surgery during the previous 6 months, symptomatic strictures, or evidence of pre-stenotic dilatation on imaging, penetrating complications, and active perianal disease. Patients were also excluded if they consumed EEN or parenteral nutrition during 6 months prior to the study period.

Data Collection

Demographic parameters (e.g., sociodemographic and environmental data), as well as complete medical history, laboratory results, and anthropometrics, were collected via parental interview and medical records. Weight and height were measured by a nurse and served as the basis for Z-score calculation, according to the World Health Organization (WHO). Disease activity was assessed using the Pediatric Crohn’s Disease Activity Index (PCDAI) [23]. Physician Global Assessment of disease activity was assessed on a Likert scale. An IMPACT questionnaire was used to assess the QOL [24]. In addition, we calculated the Mini Index, a tool developed to assess MH, comprising stool frequency, FC, and CRP or ESR. It was validated previously in several cohorts [25].

Eligible patients were requested to provide a stool sample for calprotectin and to complete three questionnaires to assess QOL and adherence to MD including the KIDMED questionnaire and FFQ. The KIDMED is a questionnaire designed for children to assess adherence to MD containing 14 dichotomic questions about food items that are characteristics of the MD [16]. The FFQ that we used in this study was developed by the Nutrition Department of the Israeli Ministry of Health and is the most frequently used FFQ in studies conducted in Israel, including the Israeli national youth health and nutrition survey named Mabat [26, 27]. Patients were asked how often they consumed each food item with a commonly used portion size (from several times a day to less than once in a month on a 4-point Likert scale). The I-MEDAS score was calculated based on the FFQ, which assesses adherence to MD and adjusted to the Israeli adult population [17]. Further, we calculated the quantitative consumption of food groups that comprise the I-MEDAS score and are characteristics of the MD pattern.

We classified patients with elevated FC based on several cutoff levels according to the current literature (using 100 μg/g, 200 μg/g, and 250 μg/g) [13, 28‒30]. In addition, we used the validated allocation of the KIDMED based on the level of adherence using low <3, moderate 4–7, and high >7 according to the literature [27].

Statistical Analysis

The correlations between adherence to MD (calculated by KIDMED) and constructs of disease activity, including FC, PCDAI, CRP, IMPACT, body mass index (BMI), and I-MEDAS score, were utilized by Pearson or Spearman’s correlations as appropriate according to the data distribution. The association between adherence to MD (the KIDMED) and dichotomous variables (i.e., below or above 100 μg/g/200 μg/g, 250 μg/g, PCDAI below or above 10, or CRP below or above 0.5 mg/dL) was analyzed using Student’s t test or Mann-Whitney U test as appropriate. The χ2 test was used for categorical variables. The correlation between the KIDMED questionnaire and the I-MEDAS score was calculated using the Pearson correlation coefficient test. We further conducted a logistic regression to evaluate the association between adherence to MD and FC >200 μg/g, and we adjusted for potential confounders, including age (in years), gender, disease duration (in years), disease location (L1, L2, L3), the KIDMED score, use of biologic (as a dichotomic variable) together with disease activity variables (PCDAI and CRP). The correlation between food groups and calprotectin, PCDAI, and CRP was analyzed by Spearman’s correlation test.

Data were analyzed using IBM SPSS Statistics for Windows software, Version 25.0. Armonk, NY: IBM Corp. The study was approved by the Local Ethics Committee at each hospital, and written informed consent was obtained accordingly.

Study Participants

Out of 103 eligible patients, four did not provide stool samples, resulting in the inclusion of 99 patients in the study. Table 1 presents the demographic characteristics of the patients. The mean age was 14.38 ± 2.6 years. Most patients had involvement of the small bowel, with 35.4% classified as L1 and 40.4% as L3. The median disease duration was 2 years (interquartile range [IQR] 0.9–4.5). The study population primarily consisted of patients in clinical remission (88%) undergoing biological therapy (85%). The median FC level was 59 μg/g (IQR 18–308).

Table 1.

Patient characteristics – entire cohort

VariableStudy population (n = 99)
Age, mean±SD, years 14.38±2.6 
Age at diagnosis, mean±SD, years 11.84±0.13 
Female, n (%) 38 (38.4) 
Disease duration in years, median [IQR] 2.09 [0.9–4.5] 
Z-score BMI, mean±SD −0.4±1.24 
Paris classification, n (%) L1 35 (35.4) 
L2 21 (21.2) 
L3 40 (40.4) 
Paris classification, n (%) L4a 43 (43.4) 
L4b 4 (4) 
L4a+b 12 (12.1) 
PCDAI, median [IQR] 0 [0–2.5] 
PGA, normal, not at all ill, n (%) 81 (81) 
CRP, mg/dL, median [IQR] 0.2 [0.06–0.5] 
ESR, mm/h, median [IQR] 8 [2–16] 
Albumin, g/dL, mean±SD 4.4±4 
FC, μg/g, median [IQR] 59 [18–308] 
Mini Index, median [IQR] 0 [(−3)−8] 
Concomitant medications 
 Adalimumab, n (%) 54 (54.5) 
 Infliximab, n (%) 24 (24.2) 
 Ustekinumab, n (%) 3 (3) 
 Vedolizumab, n (%) 3 (3) 
 Any biologics, n (%) 84 (85) 
 Injection week among patients on biologics (n = 84), median [IQR] 77.7 [28.5–158.9] 
 Azathioprine, n (%) 6 (6.1) 
 Methotrexate, n (%) 8 (8.1) 
 Budesonide, n (%) 1 (1) 
 Clinical remission, n (%) 87 (88) 
 Biochemical remission, n (%) 82 (83) 
 FC <200 μg/g, n (%) 69 (70) 
VariableStudy population (n = 99)
Age, mean±SD, years 14.38±2.6 
Age at diagnosis, mean±SD, years 11.84±0.13 
Female, n (%) 38 (38.4) 
Disease duration in years, median [IQR] 2.09 [0.9–4.5] 
Z-score BMI, mean±SD −0.4±1.24 
Paris classification, n (%) L1 35 (35.4) 
L2 21 (21.2) 
L3 40 (40.4) 
Paris classification, n (%) L4a 43 (43.4) 
L4b 4 (4) 
L4a+b 12 (12.1) 
PCDAI, median [IQR] 0 [0–2.5] 
PGA, normal, not at all ill, n (%) 81 (81) 
CRP, mg/dL, median [IQR] 0.2 [0.06–0.5] 
ESR, mm/h, median [IQR] 8 [2–16] 
Albumin, g/dL, mean±SD 4.4±4 
FC, μg/g, median [IQR] 59 [18–308] 
Mini Index, median [IQR] 0 [(−3)−8] 
Concomitant medications 
 Adalimumab, n (%) 54 (54.5) 
 Infliximab, n (%) 24 (24.2) 
 Ustekinumab, n (%) 3 (3) 
 Vedolizumab, n (%) 3 (3) 
 Any biologics, n (%) 84 (85) 
 Injection week among patients on biologics (n = 84), median [IQR] 77.7 [28.5–158.9] 
 Azathioprine, n (%) 6 (6.1) 
 Methotrexate, n (%) 8 (8.1) 
 Budesonide, n (%) 1 (1) 
 Clinical remission, n (%) 87 (88) 
 Biochemical remission, n (%) 82 (83) 
 FC <200 μg/g, n (%) 69 (70) 

Variables are presented as mean ± SD or as median [IQR] when variables are not normally distributed.

PGA, Physician Global Assessment; PCDAI, Pediatric Crohn’s Disease Activity Index, normal range PCDAI ≤10; CRP, C-reactive Protein, normal range CRP <0.5; ESR, erythrocyte sedimentation rate.

Association between Adherence to MD and Elevated FC

We first assessed the association between KIDMED and FC. We could not detect a significant correlation between FC as a continuous variable and the KIDMED score, r = −0.12, p = 0.23 (Fig. 1a). The association between KIDMED and different cutoff FC levels was significant for 200 μg/g (p = 0.04), almost significant for 250 μg/g (p = 0.07), and nonsignificant for 100 μg/g (p = 0.2). The mean KIDMED score was higher among patients who had FC <200 μg/g compared to patients with FC >200 μg/g (5.48 ± 2.58 vs. 4.37 ± 2.47, p = 0.04), respectively (Fig. 1b).

Fig. 1.

a Association between the KIDMED score and FC. b Association between adherence to the MD and elevated FC.

Fig. 1.

a Association between the KIDMED score and FC. b Association between adherence to the MD and elevated FC.

Close modal

Further, we assessed the association between FC above and below 200 μg/g and several variables. We found a significant difference in BMI, PCDAI, PGA, CRP, and Mini Index between patients with FC above and below 200 μg/g, as shown in Table 2.

Table 2.

Patient characteristics among those with FC above and below 200 μg/g

VariableCalprotectin below 200 μg/g (n = 69)Calprotectin above 200 μg/g (n = 30)p value Val
Age, mean±SD, years 14.49±2.7 14.13±2.5 0.62 
Female, n (%) 27 (39.1) 11 (36.7) 0.81 
Disease duration, median [IQR] 1.8 [0.9–4.3] 2.9 [0.8–4.9] 0.48 
Z-score BMI, mean±SD −0.22±1.2 −0.81±1.16 0.01 
Z-score weight, mean±SD −0.26±1.19 −0.75±1.23 0.06 
Z-score height, median [IQR] −0.43 [−0.8 to 0.3] −0.45 [−0.9 to 0.27] 0.99 
Paris classification, disease location, n (%) 
 L1 26 (37.7) 9 (30) 0.53 
 L2 14 (20.3) 7 (23.3) 
 L3 26 (37.7) 14 (46.7) 
Disease location upper GI 
 L4a, n (%) 32 (46.4) 11 (36.7) 0.34 
 L4b, n (%) 3 (4.3) 1 (3.3) 
 L4a+b, n (%) 10 (14.5) 2 (6.7) 
 PCDAI, median [IQR] 0 [0–0] 0 [0–10] 0.005 
 PGA, normal, not at all ill, n (%) 63 (91.3) 18 (60) 0.001 
 CRP, mg/dL, median [IQR] 0.1 [0.05–0.39] 0.5 [0.18–1.4] <0.001 
 ESR, mm/h, median [IQR] 7.5 [2–15] 10 [5–22] 0.165 
 Albumin, g/dL, median [IQR] 4.5 [4.3–4.7] 4.2 [4–4.4] <0.001 
 FC, μg/g, median [IQR] 27 [12.4–78.5] 694 [422–1,520] <0.001 
 Mini Index, median [IQR] −2 [(−3)−0.5] 10 [8–14] <0.001 
Concomitant medications 
 Adalimumab, n (%) 37 (53.6) 17 (56.7) 0.21 
 Infliximab, n (%) 20 (29) 4 (13.3) 
 Ustekinumab, n (%) 1 (1.4) 2 (6.7) 
 Vedolizumab, n (%) 1 (1.4) 2 (6.7) 
 Total biologics, n (%) 59 (85.5) 25 (83.3) 0.78 
 Injection/infusion week, median [IQR] 77 [29–148] 76 [20–186] 0.85 
 Azathioprine, n (%) 5 (7.2) 1 (3.3) 0.21 
 Methotrexate, n (%) 4 (5.8) 4 (13.3) 
 Budesonide, n (%) 1 (1.4) 
VariableCalprotectin below 200 μg/g (n = 69)Calprotectin above 200 μg/g (n = 30)p value Val
Age, mean±SD, years 14.49±2.7 14.13±2.5 0.62 
Female, n (%) 27 (39.1) 11 (36.7) 0.81 
Disease duration, median [IQR] 1.8 [0.9–4.3] 2.9 [0.8–4.9] 0.48 
Z-score BMI, mean±SD −0.22±1.2 −0.81±1.16 0.01 
Z-score weight, mean±SD −0.26±1.19 −0.75±1.23 0.06 
Z-score height, median [IQR] −0.43 [−0.8 to 0.3] −0.45 [−0.9 to 0.27] 0.99 
Paris classification, disease location, n (%) 
 L1 26 (37.7) 9 (30) 0.53 
 L2 14 (20.3) 7 (23.3) 
 L3 26 (37.7) 14 (46.7) 
Disease location upper GI 
 L4a, n (%) 32 (46.4) 11 (36.7) 0.34 
 L4b, n (%) 3 (4.3) 1 (3.3) 
 L4a+b, n (%) 10 (14.5) 2 (6.7) 
 PCDAI, median [IQR] 0 [0–0] 0 [0–10] 0.005 
 PGA, normal, not at all ill, n (%) 63 (91.3) 18 (60) 0.001 
 CRP, mg/dL, median [IQR] 0.1 [0.05–0.39] 0.5 [0.18–1.4] <0.001 
 ESR, mm/h, median [IQR] 7.5 [2–15] 10 [5–22] 0.165 
 Albumin, g/dL, median [IQR] 4.5 [4.3–4.7] 4.2 [4–4.4] <0.001 
 FC, μg/g, median [IQR] 27 [12.4–78.5] 694 [422–1,520] <0.001 
 Mini Index, median [IQR] −2 [(−3)−0.5] 10 [8–14] <0.001 
Concomitant medications 
 Adalimumab, n (%) 37 (53.6) 17 (56.7) 0.21 
 Infliximab, n (%) 20 (29) 4 (13.3) 
 Ustekinumab, n (%) 1 (1.4) 2 (6.7) 
 Vedolizumab, n (%) 1 (1.4) 2 (6.7) 
 Total biologics, n (%) 59 (85.5) 25 (83.3) 0.78 
 Injection/infusion week, median [IQR] 77 [29–148] 76 [20–186] 0.85 
 Azathioprine, n (%) 5 (7.2) 1 (3.3) 0.21 
 Methotrexate, n (%) 4 (5.8) 4 (13.3) 
 Budesonide, n (%) 1 (1.4) 

Variables are presented as mean±SD or as median [IQR] when variables are not normally distributed.

PGA, Physician Global Assessment; PCDAI, Pediatric Crohn’s Disease Activity Index, normal range PCDAI ≤10; CRP, C-reactive protein, normal range CRP <0.5; ESR, erythrocyte sedimentation rate.

Association between Adherence to MD, Inflammation, QOL, and Nutritional Status

Since most patients were in clinical remission [88%] and biochemical remission (82% presented normal CRP levels [CRP <0.5 mg/dL]), as expected, we did not observe a significant correlation to adherence to MD. We found that among the entire cohort, 30% presented FC >200 μg/g. Among those in clinical remission, 24% presented FC >200 μg/g. We next assessed the association between the IMPACT QOL questionnaire and adherence to MD and did not find any significant association (correlation of r = 0.086, p = 0.4). Moreover, when classifying patients into high and low adherence (based on KIDMED score above vs. below 7), we did not detect any significant association with QOL (76.2 ± 11 vs. 72.1 ± 14, respectively, p = 0.13). Similarly, we did not find any significant correlation between the KIDMED and BMI Z-score as a marker of nutritional status (r = 0.18, p = 0.08).

We performed a multivariate logistic regression analysis to investigate further the association between elevated FC with a cutoff of 200 μg/g and adherence to MD. An adjusted model was built while choosing potential confounders that are clinically relevant or might impact elevated FC [adjusted to age, gender, disease duration, disease location, the KIDMED score, use of biologics (as a dichotomous variable), PCDAI, and CRP]. The unadjusted model revealed OR 0.84 [95% CI: 0.7–1], p = 0.053. The KIDMED score was associated with lower odds for elevated FC (adjusted OR 0.75 [95% CI: 0.6–0.95], p = 0.019).

We classified patients based on their adherence to MD and assessed whether patients with high adherence to MD (KIDMED >7) were likelier to have a lower FC. Still, we could not detect any significant association among patients with KIDMED <7 FC levels were 51.5 [17–378] compared to 78 [18–252] among patients with KIDMED >7 (p = 0.78).

Adherence to MD among Children with CD Compared to a Healthy Population

Next, we aimed to compare the adherence to MD in children with CD in comparison to the healthy population. Following allocating the KIDMED score into 3 categories, we found that 31.3% of our study population exhibited high adherence, 51.5% demonstrated moderate adherence, and 17.2% displayed poor adherence to MD.

The characteristics of the study population, as categorized by KIDMED allocation, are outlined in Table 3. FC levels were lower in patients with the highest adherence score, although they did not show a dose-response pattern. In the Israeli cohorts, comparable adherence rates were reported among healthy adolescents based on the Mabat Israeli National Youth Health Survey for high, moderate, and poor, with 19.1%, 55%, and 25.5% in the survey conducted in 2003–2004 [27] and 43.1%, 45.3%, and 11.6% in the 2015–2016 survey [31], respectively, as presented in Figure 2.

Table 3.

Patients’ characteristics based on adherence to MD (assessed by the KIDMED)

VariableKIDMED ≤2 Poor MD adherence (n = 17)KIDMED 3–6 moderate MD adherence (n = 51)KIDMED ≥7 High MD adherence (n = 31)p value
Age, mean±SD, years 15.2±2 14.2±2.8 14.2±2.6 0.57 
Female, n (%) 4 (23.5) 26 (51) 8 (25.8) 0.02 
Disease duration, median [IQR] 2.9 [1–4.4] 2.5 [1–4.5] 1.22 [0.7–3] 0.59 
Z-score BMI, mean±SD −0.7±1.2 −0.34±1.1 −0.3±1.3 0.48 
PCDAI, median [IQR] 0 [0–1.25] 0 [0–2.5] 0 [0–5] 0.68 
PGA, normal, not at all ill, n (%) 14 (82.2) 40 (78.4) 27 (87.1) 0.35 
CRP, mg/dL 0.2 [0.05–0.5] 0.2 [0.05–0.5] 0.2 [0.06–0.5] 0.14 
FC, μg/g, median [IQR] 59 [17.5–781.5] 84 [13–454] 41 [20–135] 0.13 
Mini Index, median [IQR] 0 [(−2.5)−11] 2 [(−3)−8] −2 [(−3)−6] 0.62 
IMPACT, mean±SD 75.2±14.1 71±14 76.2±11 0.21 
I-MEDAS, mean± SD 5.76±1.5 6.9±1.5 8.13±1.7 <0.001 
Total biologics, n (%) 15 (88.2) 44 (86.3) 25 (80.6) 0.72 
VariableKIDMED ≤2 Poor MD adherence (n = 17)KIDMED 3–6 moderate MD adherence (n = 51)KIDMED ≥7 High MD adherence (n = 31)p value
Age, mean±SD, years 15.2±2 14.2±2.8 14.2±2.6 0.57 
Female, n (%) 4 (23.5) 26 (51) 8 (25.8) 0.02 
Disease duration, median [IQR] 2.9 [1–4.4] 2.5 [1–4.5] 1.22 [0.7–3] 0.59 
Z-score BMI, mean±SD −0.7±1.2 −0.34±1.1 −0.3±1.3 0.48 
PCDAI, median [IQR] 0 [0–1.25] 0 [0–2.5] 0 [0–5] 0.68 
PGA, normal, not at all ill, n (%) 14 (82.2) 40 (78.4) 27 (87.1) 0.35 
CRP, mg/dL 0.2 [0.05–0.5] 0.2 [0.05–0.5] 0.2 [0.06–0.5] 0.14 
FC, μg/g, median [IQR] 59 [17.5–781.5] 84 [13–454] 41 [20–135] 0.13 
Mini Index, median [IQR] 0 [(−2.5)−11] 2 [(−3)−8] −2 [(−3)−6] 0.62 
IMPACT, mean±SD 75.2±14.1 71±14 76.2±11 0.21 
I-MEDAS, mean± SD 5.76±1.5 6.9±1.5 8.13±1.7 <0.001 
Total biologics, n (%) 15 (88.2) 44 (86.3) 25 (80.6) 0.72 

Variables are presented as mean ± SD or as Median [IQR] when variables are not normally distributed.

PGA, Physician Global Assessment; PCDAI, Pediatric Crohn’s Disease Activity Index, normal range PCDAI ≤10; CRP, C-reactive Protein, normal range CRP <0.5; ESR, erythrocyte sedimentation rate.

Fig. 2.

Comparison of adherence to MD among children with CD and healthy population in two periods of time. Comparison of adherence to MD assessed by the KIDMED questionnaire between healthy adolescents from the Israeli cohort in two different periods between 2003–2004 and 2015–2016 and patients with CD from our cohort.

Fig. 2.

Comparison of adherence to MD among children with CD and healthy population in two periods of time. Comparison of adherence to MD assessed by the KIDMED questionnaire between healthy adolescents from the Israeli cohort in two different periods between 2003–2004 and 2015–2016 and patients with CD from our cohort.

Close modal

Evaluation of the Association between the KIDMED Score and the I-MEDAS Score in Assessing Adherence to the MD

Next, we assessed whether there was a correlation between the two methods evaluating adherence to MD [16, 17]. We found a significant moderate correlation of r = 0.46, p = 0.001, between the KIDMED and I-MEDAS in assessing adherence to MD in Israel.

Association between Elevated FC and Food Groups that Characterize MD

We further investigated whether there is a correlation between the consumption of specific food groups that are characteristic of the MD or Western diet (including fruits, vegetables, red meat, processed meat, legumes, whole grains, nuts, non-sweet dairy products, fish, soft drinks, olive oil, sweets, and snacks) and markers of inflammation (FC, PCDAI, CRP). Our analysis revealed an inverse association between the consumption of nuts and FC levels (as shown in Fig. 3). Additionally, we examined whether elevated FC levels (200 μg/g) were associated with any specific food group. We found that only the consumption of vegetables demonstrated an inverse association with elevated FC levels, with a consumption rate of 0.9 portions/day [0.3–2.9] among patients with FC >200 μg/g compared to 2.2 portions/day [0.87–3.82] among patients with FC <200 μg/g, p = 0.049.

Fig. 3.

Correlation between food groups and markers of inflammation Spearman’s correlations between inflammatory markers and food groups. Green represents negative correlation, red represents positive correlations, and color intensity represents the effect size. Significant and borderline p values are presented.

Fig. 3.

Correlation between food groups and markers of inflammation Spearman’s correlations between inflammatory markers and food groups. Green represents negative correlation, red represents positive correlations, and color intensity represents the effect size. Significant and borderline p values are presented.

Close modal

The key findings of our study reveal that adherence to the MD was inversely associated with elevated FC levels in pediatric CD, who are mostly in remission under biological therapy. This population has not been thoroughly examined, but a nutritional strategy could potentially offer a solution, particularly for those individuals who continue to exhibit elevated FC levels. Additionally, in our study, we observed that for every 1-point increase in the KIDMED score, there was a 24% decrease in the likelihood of having elevated FC levels exceeding 200 μg/g. However, we did not observe a significant correlation between FC, considered as a continuous variable, and the KIDMED score. This lack of correlation could be attributed to the limited variability in FC values within our relatively small cohort.

It is known that diet plays an important role in the reduction of inflammation, especially in children. EEN is considered the first-line therapy for induction of remission in children with luminal CD according to current guidelines due to extensive data showing high remission rates, improvement in growth velocity, and MH [32, 33]. Recently, an additional strategy, the Crohn’s Disease Exclusion Diet with partial enteral nutrition, was shown to be better tolerated than EEN with comparable high remission rates in children with mild to moderate disease [34]. This strategy focused on patients with active mild-to-moderate disease rather than patients in remission. Due to the nature of the included patients in the current study, we investigated a different study population, mainly patients in remission or with mild symptoms under biological therapy, which usually required a less strict diet, and, therefore, we assessed the role of MD among these patients. The aim of the present study was to address a specific gap in the context of patients who attain clinical remission but may still exhibit residual inflammation, falling short of the desired goal of MH, as indicated by elevated FC levels. Given that MH is a concrete therapeutic target in pediatric CD, as outlined in the STRIDE II criteria, and considering our findings that high adherence to MD is inversely associated with reduced FC levels, it is conceivable that adhering to an MD may play a role in bridging this gap and enhancing treatment efficacy. Since the initiation of our study, several groups have reported an association between adherence to MD and reduction of inflammation. Amrousy et al. [15] showed in a randomized controlled trial that adherence to MD was associated with decreased levels of FC in patients with active disease and better remission rates under MD compared to a normal diet. They showed that patients who adhered to the MD improved earlier in both PCDAI and CRP. In our study, due to the different population, which was already mostly in remission, we did not find an association with either PCDAI, CRP, or KIDMED score.

More recently, a randomized controlled trial comparing MD to the Specific Carbohydrate Diet among adult patients with CD showed that both diets led to the same moderate remission rate and concluded that MD is easier to follow with other health benefits and, therefore, might be preferred over Specific Carbohydrate Diet in most CD patients [35]. One of the advantages of MD is that it is a feasible dietary approach rather than a strict diet and therefore might be sustainable in patients in remission under medical care.

We could not detect an association between adherence to MD and QOL as was shown previously by Papada et al. [7] and Fiorindi et al. [36]. Since QOL is most strongly associated with disease activity and since the vast majority of our patients were in clinical remission, the cohort was not powered to show differences in QOL with respect to MD. We did not identify a significant correlation with Z-score BMI, although it appeared borderline; however, when adjusting for elevated FC, this correlation seemed to diminish. Our results indicate that adhering to the MD alone was inadequate to impact BMI scores significantly. This underscores that dietary patterns alone may not be sufficient to be associated with BMI, emphasizing the necessity for consistent support from dietitians.

When dividing our patients based on their adherence to MD as previously reported [27], we found that patients with the highest adherence scores to the MD had lower FC levels. However, it did not follow a strict dose-dependent pattern. Patients with lower adherence to the MD displayed a broader IQR, suggesting higher FC levels in this group. In addition, we found that our study population had a comparable level of adherence to MD compared to the healthy Israeli population in both surveys; 82.8% were with high and moderate adherence in our cohort compared to 74.1% in the first survey and 88.4% in the later survey [27, 31]. Since our study population was mostly in remission, it is unsurprising that the adherence rates were very close to the healthy population. These results align with Fiorindi et al. [12], who showed that patients with inactive disease have a higher adherence to MD. This strengthens the notion that IBD patients in remission have common characteristics with the healthy population, including the adherence rates to MD.

Our study found a moderate correlation between the KIDMED score and the I-MEDAS score. This finding strengthens the complexity of assessing adherence to MD in different populations (children and adults) using different tools but also shows that both tools can be used to assess adherence to MD in the Israeli population.

We found that vegetables were the only food group in the MD associated with elevated FC (above 200 μg/g). In addition, we found that the consumption of nuts as an additional source of fiber was associated with FC as a continuous variable. In that context, Khalili et al. [8] examined the association between individual components of the mMED score and the risk of CD. They found that most of the components were not associated, suggesting that the effect is related to the long-term whole dietary pattern rather than a specific individual component. Of note, while vegetables are an important part of MD, historically, patients with CD were instructed to avoid fruits and vegetables due to mechanical reasons [37]. Due to the importance of vegetables to human health and microbiome, our results support the consumption of fruits and vegetables in CD, especially during remission, based on the patient’s tolerance [38]. Consumption of nuts while patients are in remission and able to tolerate them is also warranted.

Our study has several limitations. The study design was cross-sectional, enabling it to detect associations but not causality. It is unclear whether patients with high adherence to MD are more likely to have a lower FC or if patients with elevated FC are more likely to avoid fruits and vegetables and, therefore, present a lower adherence rate to MD. Since our criterion for participation excluded all patients with complicated disease and/or active perianal disease and/or recent medication changes, this led to a potential selection bias in which children with a milder disease phenotype and/or stable remission were selected. The characteristics outlined in Table 1 reveal a higher-than-expected prevalence of ileal disease (35.4%) and a lower-than-expected occurrence of ileocolonic phenotypes (40.4%). This once again suggests that the inclusion criteria may have resulted in the selection of patients with a milder disease phenotype. While this enabled us to study the role of MD on inflammation, it allowed us to investigate the role of MD in special populations where the evidence is scarce. Additionally, accurate assessment of adherence to MD is always challenging and is based on dichotomous questionnaires. We tried to improve the accuracy by using two questionnaires and an additional FFQ that contained more detailed questions regarding the actual consumption.

Nevertheless, this was a large prospective cohort of patients examining the association of the MD with FC as a measure of MH in special populations using multiple assessment methods and a statistical analysis that dealt with many of the potential confounders. It is worth exploring the impact of adherence to the MD on a more diverse cohort, encompassing a broader spectrum of clinical remission rates, various disease phenotypes, different medication regimens, and distinct disease statuses. This approach is more likely to reveal a distinct association between adherence to the MD and disease activity scores, as well as FC levels.

In conclusion, we found that high adherence to MD might be associated with decreased levels of FC in patients who are mostly in remission under biological therapy. Our results suggest the role of MD in pediatric CD and may serve as a basis to develop a feasible dietary intervention based on the MD principles to decrease elevated FC levels in children in clinical remission under biological therapy but still did not achieve MH, presenting with elevated FC. Given the increasing consumption of processed foods among children in Westernized countries, providers should encourage implementing healthy dietary habits, including the MD, especially in patients with CD.

We would like to thank the patients and their families for their collaboration. In addition, we wish to thank the team at the Institute for Gastroenterology, Nutrition and Liver Disease at Schneider Children’s Medical Center and the team at the Institute for Gastroenterology and Nutrition at Shaare Zedek Medical Center, especially the research units for their support and help with the recruitment of patients.

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Schneider Medical Center number 0728-19 RMC on December 25, 2019, and Shaare Zedek Medical Center number 0005–20 on May 17, 2020. Written informed consent was obtained by the parents of patients under 18 and by the patients themselves if they were above 18 years.

The authors have not received any financial support for this study. The authors declare that they have no conflicts of interest.

The authors declare that no funding was received for this study.

Conceptualization: R.S.B., A.A., and R.S.; methodology: R.S.B., A.A., R.S., and G.C.; analysis and investigation: R.S.B., T.T.P., and G.C.; writing – original draft preparation: R.S.B.; writing – review and editing: R.S.B., A.A., R.L.-Z., D.S., R.S., and G.C.; patients’ recruitment: R.S.B., A.A., R.L.-Z., M.M., D.S., and R.S.; and data collection: R.S.B. and C.S.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.

1.
de Castro MM, Pascoal LB, Steigleder KM, Siqueira BP, Corona LP, Ayrizono MLS, et al. Role of diet and nutrition in inflammatory bowel disease. World J Exp Med. 2021;11:1–16.
2.
Kaplan GG, Ng SC. Understanding and preventing the global increase of inflammatory bowel disease. Gastroenterology. 2017;152(2):313–21.e2.
3.
Wark G, Samocha-Bonet D, Ghaly S, Danta M. The role of diet in the pathogenesis and management of inflammatory bowel disease: a review. Nutrients. 2020;13:135–26.
4.
Martinez-Gonzalez MA, Martin-Calvo N. Mediterranean diet and life expectancy; beyond olive oil, fruits, and vegetables. Curr Opin Clin Nutr Metab Care. 2016;19(6):401–7.
5.
Aridi YS, Walker JL, Roura E, Wright ORL. Adherence to the mediterranean diet and chronic disease in Australia: national nutrition and physical activity survey analysis. Nutrients. 2020;12(5):1251.
6.
de Filippis F, Pellegrini N, Vannini L, Jeffery IB, la Storia A, Laghi L, et al. High-level adherence to a Mediterranean diet beneficially impacts the gut microbiota and associated metabolome. Gut. 2016;65(11):1812–21.
7.
Papada E, Amerikanou C, Forbes A, Kaliora AC. Adherence to Mediterranean diet in Crohn’s disease. Eur J Nutr. 2020;59(3):1115–21.
8.
Khalili H, Håkansson N, Chan SS, Chen Y, Lochhead P, Ludvigsson JF, et al. Adherence to a Mediterranean diet is associated with a lower risk of later-onset Crohn’s disease: results from two large prospective cohort studies. Gut. 2020;69(9):1637–44.
9.
Taylor L, Almutairdi A, Shommu N, Fedorak R, Ghosh S, Reimer R, et al. Cross-sectional analysis of overall dietary intake and mediterranean dietary pattern in patients with crohn’s disease. Nutrients. 2018;10(11):1761.
10.
Marsh A, Radford-Smith G, Banks M, Lord A, Chachay V. Dietary intake of patients with inflammatory bowel disease aligns poorly with traditional Mediterranean diet principles. Nutr Diet. 2022;79(2):229–37.
11.
Vrdoljak J, Vilović M, Živković PM, Tadin Hadjina I, Rušić D, Bukić J, et al. Mediterranean diet adherence and dietary attitudes in patients with inflammatory bowel disease. Nutrients. 2020;12(11):3429–14.
12.
Fiorindi C, Dinu M, Gavazzi E, Scaringi S, Ficari F, Nannoni A, et al. Adherence to mediterranean diet in patients with inflammatory bowel disease. Clin Nutr ESPEN. 2021;46:416–23.
13.
Godny L, Reshef L, Pfeffer-Gik T, Goren I, Yanai H, Tulchinsky H, et al. Adherence to the Mediterranean diet is associated with decreased fecal calprotectin in patients with ulcerative colitis after pouch surgery. Eur J Nutr. 2020;59(7):3183–90.
14.
Strisciuglio C, Cenni S, Serra MR, Dolce P, Martinelli M, Staiano A, et al. Effectiveness of mediterranean diet’s adherence in children with inflammatory bowel diseases. Nutrients. 2020;12(10):3206.
15.
El Amrousy D, Elashry H, Salamah A, Maher S, Abd-Elsalam SM, Hasan S. Adherence to the mediterranean diet improved clinical scores and inflammatory markers in children with active inflammatory bowel disease: a randomized trial. J Inflamm Res. 2022;15:2075–86.
16.
Serra-Majem L, Ribas L, Ngo J, Ortega RM, García A, Pérez-Rodrigo C, et al. Food, youth and the mediterranean diet in Spain. Development of KIDMED, mediterranean diet quality index in children and adolescents. Public Health Nutr. 2004;7:931–5.
17.
Abu-Saad K, Endevelt R, Goldsmith R, Shimony T, Nitsan L, Shahar DR, et al. Adaptation and predictive utility of a Mediterranean diet screener score. Clin Nutr. 2019;38(6):2928–35.
18.
Ayling RM, Kok K. Fecal calprotectin. Adv Clin Chem. 2018;87:161–90.
19.
Kostas A, Siakavellas SI, Kosmidis C, Takou A, Nikou J, Maropoulos G, et al. Fecal calprotectin measurement is a marker of short-term clinical outcome and presence of mucosal healing in patients with inflammatory bowel disease. World J Gastroenterol. 2017;23(41):7387–96.
20.
Turner D, Ricciuto A, Lewis A, D’Amico F, Dhaliwal J, Griffiths AM, et al. STRIDE-II: an update on the selecting therapeutic targets in inflammatory bowel disease (STRIDE) initiative of the international organization for the study of IBD (IOIBD): determining therapeutic goals for treat-to-target strategies in IBD. Gastroenterology. 2021;160(5):1570–83.
21.
Lee YM, Choi S, Choe BH, Jang HJ, Kim S, Koh H, et al. Association between fecal calprotectin and mucosal healing in pediatric patients with crohn’s disease who have achieved sustained clinical remission with anti-tumor necrosis factor agents. Gut Liver. 2022;16(1):62–70.
22.
Ferreiro-Iglesias R, Barreiro-de Acosta M, Otero Santiago M, Lorenzo Gonzalez A, Alonso de la Peña C, Benitez Estevez AJ, et al. Fecal calprotectin as predictor of relapse in patients with inflammatory bowel disease under maintenance infliximab therapy. J Clin Gastroenterol. 2016;50(2):147–51.
23.
Hyams J, Markowitz J, Otley A, Rosh J, Mack D, Bousvaros A, et al. Evaluation of the pediatric Crohn disease activity index: a prospective multicenter experience. J Pediatr Gastroenterol Nutr. 2005;41(4):416–21.
24.
Grant A, MacIntyre B, Kappelman MD, Otley AR. A new domain structure for the IMPACT-III health-related quality of life tool for pediatric inflammatory bowel disease. J Pediatr Gastroenterol Nutr. 2020;71(4):494–500.
25.
Cozijnsen MA, ben Shoham A, Kang B, Choe BH, Choe YH, Jongsma MME, et al. Development and validation of the mucosal inflammation noninvasive index for pediatric crohn’s disease. Clin Gastroenterol Hepatol. 2020;18(1):133–40.e1.
26.
Peng W, Berry EM, Goldsmith R. Adherence to the Mediterranean diet was positively associated with micronutrient adequacy and negatively associated with dietary energy density among adolescents. J Hum Nutr Diet. 2019;32(1):41–52.
27.
Peng W, Goldsmith R, Berry EM. Demographic and lifestyle factors associated with adherence to the Mediterranean diet in relation to overweight/obesity among Israeli adolescents: findings from the Mabat Israeli national youth health and nutrition survey. Public Health Nutr. 2017;20(5):883–92.
28.
Weinstein-Nakar I, Focht G, Church P, Walters TD, Abitbol G, Anupindi S, et al. Associations among mucosal and transmural healing and fecal level of calprotectin in children with crohn’s disease. Clin Gastroenterol Hepatol. 2018;16(7):1089–97.e4.
29.
Jukic A, Bakiri L, Wagner EF, Tilg H, Adolph TE. Calprotectin: from biomarker to biological function. Gut. 2021;70(10):1978–1988.
30.
García-Sánchez V, Iglesias-Flores E, González R, Gisbert JP, Gallardo-Valverde JM, González-Galilea A, et al. Does fecal calprotectin predict relapse in patients with Crohn’s disease and ulcerative colitis?J Crohns Colitis. 2010;4(2):144–52.
31.
Peng W, Goldsmith R, Shimony T, Berry EM, Sinai T. Trends in the adherence to the Mediterranean diet in Israeli adolescents: results from two national health and nutrition surveys, 2003 and 2016. Eur J Nutr. 2021;60(7):3625–38.
32.
van Rheenen PF, Aloi M, Assa A, Bronsky J, Escher JC, Fagerberg UL, et al. The medical management of paediatric crohn’s disease: an ECCO-ESPGHAN guideline update. J Crohns Colitis. 2021;15(2):171–94.
33.
Grover Z, Muir R, Lewindon P. Exclusive enteral nutrition induces early clinical, mucosal and transmural remission in paediatric Crohn’s disease. J Gastroenterol. 2014;49(4):638–45.
34.
Levine A, Wine E, Assa A, Sigall Boneh R, Shaoul R, Kori M, et al. Crohn’s disease exclusion diet plus partial enteral nutrition induces sustained remission in a randomized controlled trial. Gastroenterology. 2019;157(2):440–50.e8.
35.
Lewis JD, Sandler RS, Brotherton C, Brensinger C, Li H, Kappelman MD, et al. A randomized trial comparing the specific carbohydrate diet to a mediterranean diet in adults with crohn’s disease. Gastroenterology. 2021;161(3):837–52.e9.
36.
Fiorindi C, Dinu M, Gavazzi E, Scaringi S, Ficari F, Nannoni A, et al. Adherence to mediterranean diet in patients with inflammatory bowel disease. Clin Nutr ESPEN. 2021;46:416–23.
37.
Brown AC, Rampertab SD, Mullin GE. Existing dietary guidelines for Crohns disease and ulcerative colitis. Expert Rev Gastroenterol Hepatol. 2011;5(3):411–25.
38.
Levine A, Rhodes JM, Lindsay JO, Abreu MT, Kamm MA, Gibson PR, et al. Dietary guidance from the international organization for the study of inflammatory bowel diseases. Clin Gastroenterol Hepatol. 2020;18(6):1381–92.