Het glutenvrij dieet is de enige mogelijke behandelmethode voor patiënten met glutengerelateerde aandoeningen. Door behandeling van de ziekte verdwijnen klachten veelal of grotendeels, waardoor patiënten zich beter voelen en hun kwaliteit van leven toeneemt. Anderzijds wordt er in de praktijk gemerkt dat er veel patiënten zijn die, ondanks glutenvrij dieet, toch klachten blijven ervaren. Dit staat nog los van het feit dat het volgen van een glutenvrij dieet de nodige zorgen en beperkingen met zich meebrengt. Het is dan ook niet ongebruikelijk dat dit verlies van kwaliteit van leven met zich meebrengt.

Description of studies

Studies in children

Chellan (2019) performed a prospective follow-up study to evaluate the impact of coeliac disease (CD) on psychosocial parameters in children. A total of 60 pediatric patients were enrolled in this study (mean age 6.03 ± 0.42 years, 32% girls). Children with symptoms suggestive of CD and elevated tissue transglutaminase were included. CD was diagnosed on the basis of standard European Society for Paediatric Gastroenterology Hepatology and Nutrition (ESPGHAN) criteria. Follow-up was conducted after 6 months of gluten-free diet (GFD). Both the generic (PSC) and disease-specific questionnaires were used to assess the QoL in children with CD. Baseline assessment of a child’s psychosocial parameters was performed with the help of the PSC questionnaire. Each PSC item was rated never: 0, sometimes: 1, and often: 2. The total score was calculated and recorded as a dichotomous variable. PSC scores of 28 or higher for children aged >6 years and 24 or higher for children between 2 and 6 years of age were considered to be emotional and psychosocial impairment. The CD-specific questionnaire included questions on barriers to dietary compliance and parental behavior and perceptions. The effect of CD and GFD on the feelings of children with respect to friends, school, or teachers; feelings about following special GFD; and difficulty in determining if food available at various places like schools, supermarkets, or restaurants was gluten free were also included in the questionnaire. The feelings of the child and the difficulty faced by him or her to determine whether food available was free from gluten was expressed as all the time, most of the time, sometimes, or never, and children falling in each category were expressed numerically and by percentages. A total of 16 children were lost to follow-up, resulting in a sample of 44 children at 6 months follow-up.

Nordyke (2013) performed a nested case-referent study to investigate the health-related quality of life (HRQoL) of adolescents with screening detected CD. A total of 144 adolescents with screening-detected CD and 828 referents without CD who participated in the same screening were included in this study. Adolescents were about 12 years old at the time of the screening, and 54% of the patients were girls. The adolescents reported their HRQoL at the time of the screening and 1 year after the cases received their diagnosis. HRQoL was assessed with a Swedish child-friendly pilot version of the EQ-5D instrument (EQ-5D). The EQ-5D is a generic tool used to measure health status and report HRQoL. It is comprised of two parts, the EQ-5D descriptive system where respondents classify their health status in five dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression and on level of severity (no problems, moderate problems, or severe problems). The second part of the EQ-5D is a thermometer like visual analogue scale (VAS) where respondents score their health today from worst to best imaginable (0–100). A total of 103 adolescents had data at baseline and follow-up.

Studies in adults

Burger (2017) performed a systematic review and meta-analysis to assess the effect of gluten free diet on HRQoL in CD patients. The databases PubMed, EMBASE, CINAHL, PsycINFO and The Cochrane Library were searched from January 1960 to 30 September 2015. Studies that met the following inclusion criteria were included: 1) assessed HRQoL using validated questionnaires; 2) participants were above 16 years of age; 3) coeliac disease was confirmed by biopsy; 4) compared HRQoL in patients and non-coeliac controls, before and after gluten-free diet initiation or between subgroups of treated patients (disease-specific questionnaires only). A total of 18 studies reporting on HRQoL in coeliac disease were included in the review and ultimately 16 articles were suitable for meta-analysis. We are only interested in studies comparing HRQoL before and after the introduction of GFD. A total of eight studies with 998 patients were included, providing prospective data on the effect of a gluten-free diet.

Choung (2018) performed a longitudinal study to compare quality of life before and after institution of a GFD. A total of 130 patients with CD were included in the study, 8 patients were lost to follow-up resulting in a sample of 122 patients (mean age 45.0 ± 14.9 years, 67% female). Of these patients, 62 had classical CD (mean age 44.7 ± 15.5 years, 69.4% female) and 60 had nonclassical CD (mean age 45.4 ± 14.4 years, 65.0% female). To classify the classical and nonclassical phenotypes of CD, clinical evaluations and the self-report bowel disease questionnaire were used. Classical phenotype was defined by patients who had watery diarrhea, steatorrhea, or weight loss at the time of diagnosis. Those who did not have these classical symptoms were categorized as having nonclassical CD. Confirming diagnosis of CD was based on the positive serology and intestinal mucosal histology —increased intraepithelial lymphocytes and/or mucosal villous atrophy in the proximal small intestine. Quality of life was assessed with the 36-item Short Form Survey Instrument at the time of diagnosis and after following a GFD. The mean duration of GFD was 6.7 ± 3.7 years.

Deepak (2018) performed a single center prospective follow-up study to assess quality of life in patients with CD. A total of 60 patients with CD were included in this study (mean age 28.85 ± 12.43 years, 44% female). CD was diagnosed according to the standard ESPGHAN Guidelines. Quality of life of the patients was assessed using SF-12 and CD-QOL questionnaires. The SF-12 questionnaire comprises composite scores in eight domains: general health, physical functioning, social functioning, physical and emotional role functioning, vitality, bodily pain and mental health, and yields a physical health composite score (PCS) and a mental health composite score (MCS). The scale scores were calculated by summing the responses and then transformed into a 0–100 scale, with a higher score indicating better health. The CD-QOL questionnaire consists of 20 items across four clinically relevant subscales that are to be answered using a Likert scale. These include the health-concern subscale, the inadequate treatment subscale, the dysphoria subscale, and the limitations subscale. The overall score is expressed on a scale of 0–100, with a higher score indicating poorer health. Quality of life was assessed before starting GFD and after 6 months of GFD. one patient was excluded as his endoscopic and histopathological features were not consistent with CD.

Results

Studies in children

Two studies described quality of life (QoL) in children with CD before and after following a GFD (Chellan, 2019; Nordyke, 2013). Chellan (2019) used both the generic (PSC) and disease-specific questionnaires to assess QoL and Nordyke (2013) used a child-friendly pilot version of the EQ-5D.

Chellan (2019) reported that median PSC scores improved from 11.5 (2 to 35) before starting GFD to 2.5 (0 to 34) after 6 months of GFD. 

Nordyke (2013) reported that the cases for all dimensions of the EQ-5D were lower after 1 year GFD than at baseline, except for the anxiety/depression dimension. For the mobility dimension, 3/103 (2.9%) patients had problems at baseline, compared to 1/103 (1.0%) at follow up. For the self-care dimensions, there were no problems at both baseline and follow-up. A total of 3/103 (2.9%) had problems with activity at baseline, compared to 2/102 (1.9%) at follow-up. For the pain dimension, 21/102 (20.4%) patients had problems at baseline, compared to 13/103 (12.6%) patients at follow-up. Finally, 13/103 (12.6%) patients had problems with anxiety/depression at baseline, compared to 15 (14.6%) patients at follow-up. In total, 40/103 (38.8%) children had problems for one or more of the dimensions at baseline, compared to 31/103 (30.1%) at follow-up. The RR is 1.29 (95%CI 0.88 to 1.89). This is a clinically relevant difference.

Studies in adults

Three studies described QoL in adults with CD before and after following a GFD, of which one systematic review (Burger, 2017) and two additional studies (Choung, 2018; Deepak, 2019). The systematic review by Burger (2017) included eight studies, of which four studies used the Psychological General Well-Being (PGWB) questionnaire and four other studies used the 36-Item Short Form Survey (SF-36) to assess QoL. Choung (2018) also applied the SF-36 survey and Deepak (2019) used the 12-Item Short Form Survey (SF-12) and a CD-QOL questionnaire.

Burger (2017) reported that QoL improved after 1 year GFD. They reported a pooled effect from four studies (n=846) with a mean difference of 7.34 (95%CI 1.96 to 12.72) between baseline and 1 year follow-up for PGWB total. For SF-36 MCS (n=152), the mean difference was 7.37 (95%CI 1.84 to 12.90) and the mean difference for SF-36 PCS (n=152) was 5.72 (95%CI 1.50 to 9.95).

Choung (2018) reported that QoL improved after 1 year GFD in both patients with classical CD and non-classical CD. At baseline, mean SF-36 PSC was 43.5 ± 10.8 for the classical group of patients and 46.5 ± 11.1 for the nonclassical group, compared to 48.0 ± 9.8 and 48.0 ± 10.1 after 1 year GFD. Mean SF-36 MSC was 44.3 ± 11.2 for the classical group and 49.8 ± 9.4 for the nonclassical group at baseline, compared to 48.2 ± 12.1 (classical) and 53.3 ± 8.5 (nonclassical) after 1 year GFD.

Deepak (2018) reported a mean SF-12 PCS of 37.2 ± 11.1 at baseline, compared to 50.3 ± 9.9 after 6 months GFD. The mean difference in score before and after initiation of GFD was 8.30 (95%CI 5.16 to 11.44). Mean SF-12 MSC was 41.9 ± 8.4 at baseline, compared to 50.2 ± 9.0 after 6 months GFD. The mean difference was 13.10 (95%CI 9.30 to 16.90). These differences are clinically relevant.

Median QoL as measured by the CD-QoL was 54.5 (range 26 – 73) at baseline, compared to 48 (24 – 48) after 6 months GFD.

The pooled effect from four studies for QoL as assessed by the PGWB questionnaire before and after initiation of GFD was 7.34 (95%CI 1.96 to 12.72). QoL was better after initiation of GFD (Figure 1), this difference in not clinically relevant.

The pooled effect from five studies for QoL as assessed by the SF-36 questionnaire before and after initiation of GFD was 6.57 (95%CI 2.12 to 11.03) for the mental health composite score, and 5.09 (95%CI 1.72 to 8.45) for the physical health composite score. QoL was better after initiation of GFD (Figure 1). This difference is clinically relevant.

Figure 1. Pooled mean difference in health-related QoL before and after initiation of GFD.

Level of evidence of the literature

The evidence came from observational studies and starts at a low quality. The level of evidence was downgraded to very low because of risk of bias (loss-to-follow-up) and imprecision (crossing the border of clinical relevance).

A systematic review of the literature was performed to answer the following question:

What are the benefits and harms of following a gluten-free diet on the quality of life in patients with coeliac disease, compared to no gluten-free diet?

P (patients) Children and adults with celiac disease

I (intervention) Gluten-free diet

C (control) No gluten-free diet

O (outcome) Quality of life

Relevant outcome measures

The guideline working group considered quality of life (QoL) as a critical outcome measure for decision making.

A priori, the working group did not define the outcome measures listed above but used the definitions used in the studies.

The working group defined the GRADE standard limit of 25% difference for dichotomous outcomes (RR <0.8 or >1.25) and 0.5 SD for continuous outcomes as a minimal clinically (patient) important difference.

Search and select (Methods)

The databases Medline (via OVID) and Embase (via Embase.com) were searched with relevant search terms from 1 January, 2000 until 9 March, 2020. The detailed search strategy is depicted under the tab Methods. The systematic literature search resulted in 297 hits. Studies assessing quality of life before and after initiation of a gluten-free diet within the same person were included. A total of 39 studies were initially selected based on title and abstract screening. After reading the full text, 11 studies were included, and 28 studies were excluded (see the table with reasons for exclusion under the tab Methods). Of the included studies, two studies were performed in children and nine studies in adults. One of the studies in adults was a systematic review (Burger, 2017), which included six of the studies that we initially included. Therefore, our final inclusion consisted of five studies: two studies in children and three studies in adults (one systematic review plus two additional studies).

Results

A total of five of studies were included in the analysis of the literature, of which one systematic review and three observational studies. Important study characteristics and results are summarized in the evidence tables. The assessment of the risk of bias is summarized in the risk of bias tables.