European Society for the Study of Coeliac Disease (ESsCD) guideline for coeliac disease and other gluten‐related disorders

This is a clinical practice guideline, not a single experiment. The authors systematically reviewed the published literature from 1990 to 2018 on gluten-related disorders, then formulated recommendations using the GRADE system (Grading of Recommendations, Assessment, Development and Evaluations). The guideline covers:

**Diagnostic pathways**: Which serological tests (tissue transglutaminase IgA, endomysium antibodies, deamidated gliadin peptides) to use, when to perform duodenal biopsy, and the specific criteria for diagnosing coeliac disease without biopsy in children.

**Dietary management**: The safe threshold for gluten intake in coeliac disease (less than 10 mg per day), the role of dietitians, and monitoring for nutritional deficiencies on a gluten-free diet.

**Management of complications**: How to evaluate slow-responders (those who do not improve after 6–12 months on a gluten-free diet) and refractory coeliac disease (persistent villous atrophy despite strict diet for >12 months).

**Non-coeliac gluten sensitivity (NCGS)**: Diagnostic criteria requiring exclusion of coeliac disease and wheat allergy, with symptoms that appear within hours to days of gluten ingestion and resolve on a gluten-free diet.

**Extra-intestinal manifestations**: Dermatitis herpetiformis (skin lesions with granular IgA deposits), gluten ataxia (idiopathic sporadic ataxia with positive coeliac serology), and other neurological conditions.

The comparator in each case is either a gluten-containing diet (for diagnostic testing) or a gluten-free diet (for treatment response). Outcome measures include symptom resolution, serological normalisation (tissue transglutaminase antibodies returning to normal), histological healing on biopsy, and quality of life.

This guideline does not report a single study population. Instead, it draws on:

**Population-level data**: Prevalence studies from Western countries showing ~0.6% histologically confirmed coeliac disease and ~1% seropositive in general population screening.

**Clinical cohorts**: Multiple case-control and cohort studies of adults and children with coeliac disease, dermatitis herpetiformis, gluten ataxia, and NCGS. Specific numbers are not aggregated in the guideline itself.

**Genetic studies**: Populations of HLA-DQ2 and/or HLA-DQ8 carriers (approximately 30–40% of the general population carry these genes, but only ~3% develop coeliac disease).

**Age ranges**: Both paediatric (including diagnostic criteria for children under 2 years) and adult populations (including elderly patients diagnosed later in life).

**Special populations**: Patients with IgA deficiency (more common in coeliac disease, affecting ~2–3% of patients), first-degree relatives (10–15% risk of developing coeliac disease), and patients with other autoimmune conditions (type 1 diabetes, autoimmune thyroiditis, Down syndrome).

The guideline does not use a single measurement instrument but synthesises findings from studies using:

**Serological assays**: IgA anti-tissue transglutaminase (TG2) antibodies (normal range varies by assay, but >10 times upper limit of normal is considered highly predictive), IgA endomysium antibodies (EMA, highly specific, ~99%), and deamidated gliadin peptides (DGP, used especially in children under 2 years and in IgA-deficient patients).

**Histopathological grading**: Modified Marsh classification (Marsh 0 = normal, Marsh 1 = increased intraepithelial lymphocytes, Marsh 2 = crypt hyperplasia, Marsh 3a/b/c = partial/subtotal/total villous atrophy). Diagnosis requires at least Marsh 2 or 3.

**HLA typing**: Detection of HLA-DQ2 and/or HLA-DQ8 haplotypes (negative predictive value >99% — if you don't have these genes, you almost certainly don't have coeliac disease).

**Clinical symptom scales**: Not standardised across studies, but includes gastrointestinal symptoms (diarrhoea, bloating, abdominal pain), extra-intestinal symptoms (fatigue, anaemia, osteoporosis, neurological symptoms), and quality of life questionnaires.

**Dietary adherence assessment**: Combination of dietitian interview, serological monitoring (persistently elevated antibodies suggest ongoing gluten exposure), and in some cases, follow-up biopsy.

**Skin biopsy for dermatitis herpetiformis**: Direct immunofluorescence showing granular IgA deposits in dermal papillae (gold standard).

**Study design**: This is a clinical practice guideline developed by the European Society for the Study of Coeliac Disease (ESsCD), a transnational multidisciplinary group of paediatric and adult gastroenterologists. Two board representatives were nominated by the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN).

**Literature search**: The authors searched PubMed for articles published from 1 January 1990 to August 2018 using the terms "celiac", "coeliac", "non-tropical sprue", "gluten", "dermatitis herpetiformis", "enteropathy", and "ataxia", with no language restrictions. They emphasised publications since 2000 but included older landmark studies. They mostly selected cohort and case-control studies, the few randomised trials available, and smaller non-controlled clinical studies of particular relevance.

**Evidence grading**: The GRADE system was used. "Strong" recommendations mean benefits clearly outweigh negatives. "Conditional" recommendations mean some uncertainty remains. Evidence quality is graded from high (further research unlikely to change confidence) to low (further study would likely change the estimate).

**Consensus process**: Board members reviewed each section critically, communicated via email, then met in Vienna, Austria (October 2018, parallel to UEG Week) to finalise the draft.

**What this design can and cannot prove**: A clinical guideline is not a primary study — it cannot prove anything new. It can only synthesise existing evidence and highlight gaps. The strength of a guideline is that it provides standardised, evidence-based recommendations for clinicians. The weakness is that the quality of recommendations depends entirely on the quality of underlying studies. Many recommendations in this guideline are based on low-quality evidence (observational studies, expert opinion) because randomised controlled trials are rare in coeliac disease research — it would be unethical to randomise a diagnosed patient to continue eating gluten. The guideline explicitly notes areas of uncertainty and calls for future research.

**Major methodological weaknesses**:

The literature search ended in August 2018, so studies published in the last 6+ years are not included.

The guideline does not provide a systematic review protocol (e.g., PRISMA checklist) or a formal assessment of risk of bias across included studies.

Many recommendations are based on expert consensus rather than high-quality randomised trials.

The section on NCGS is particularly weak — the authors acknowledge that the condition is poorly defined and that many symptoms attributed to gluten may actually be due to FODMAPs (fermentable carbohydrates) or amylase-trypsin inhibitors.

**Prevalence and epidemiology:**

Coeliac disease prevalence in Western countries: ~0.6% histologically confirmed, ~1% by serological screening.

Female-to-male ratio: 1.3:1 to 3:1 depending on the study.

Prevalence has increased substantially over the past 50 years, partly due to better diagnostics and increased awareness.

The majority of coeliac disease patients worldwide remain undiagnosed (the "iceberg" phenomenon).

**Diagnostic criteria:**

Diagnosis requires positive serology (TG2 IgA or EMA) PLUS duodenal biopsy showing Marsh 2 or 3 changes, while the patient is on a gluten-containing diet.

In children with symptoms, TG2 IgA >10 times the upper limit of normal, positive EMA, and positive HLA-DQ2/DQ8 can diagnose coeliac disease without biopsy (conditional recommendation, moderate-quality evidence).

HLA-DQ2/DQ8 typing has a negative predictive value >99% — a negative result essentially rules out coeliac disease.

IgA deficiency should be ruled out in all patients being tested for coeliac disease (total IgA measurement is recommended alongside TG2 IgA).

**Gluten-free diet:**

Safe gluten intake: less than 10 mg per day (strong recommendation, moderate-quality evidence). For context, one slice of wheat bread contains approximately 2,000–3,000 mg of gluten.

Nutritional deficiencies are common at diagnosis: iron deficiency (30–50%), folate deficiency (20–30%), vitamin B12 deficiency (10–20%), vitamin D deficiency (40–60%), and zinc deficiency (30–50%).

A gluten-free diet can lead to weight gain and metabolic syndrome — up to 20–30% of patients become overweight or obese after diagnosis.

**Slow-responders and refractory disease:**

Slow-responders (persistent symptoms or villous atrophy after 6–12 months on a strict gluten-free diet) occur in 10–30% of adults diagnosed with coeliac disease.

Refractory coeliac disease (RCD) type I: persistent villous atrophy despite >12 months of strict gluten-free diet, with normal intraepithelial lymphocyte phenotype. 5-year survival >90%.

Refractory coeliac disease type II: abnormal intraepithelial lymphocyte phenotype (clonal expansion). 5-year survival ~50%, with high risk of developing enteropathy-associated T-cell lymphoma (EATL).

**Non-coeliac gluten sensitivity:**

No reliable biomarkers exist. Diagnosis is based on symptom improvement on a gluten-free diet and relapse on gluten challenge, after excluding coeliac disease and wheat allergy.

The guideline notes that many patients diagnosed with NCGS may actually be reacting to FODMAPs or ATIs rather than gluten itself.

**Dermatitis herpetiformis:**

Granular IgA deposits in dermal papillae on direct immunofluorescence are diagnostic (gold standard).

Skin lesions clear with gluten withdrawal, but may take months to years in adults.

Dapsone can be used for rapid symptom control (24–48 hours), but does not treat the underlying enteropathy.

**Gluten ataxia:**

Defined as idiopathic sporadic ataxia with positive coeliac serology, with or without enteropathy.

Other causes of ataxia (genetic, ischaemic, paraneoplastic) must be excluded.

Gluten-free diet may improve or stabilise neurological symptoms, but evidence is limited to observational studies.

Because this is a guideline, effect sizes are not reported as single numbers. However, translating key findings:

**Diagnostic accuracy**: TG2 IgA antibodies have a sensitivity of ~93% and specificity of ~96% for coeliac disease. EMA has specificity of ~99% but slightly lower sensitivity (~90%). This means that if 100 people with coeliac disease are tested, about 93 will have a positive TG2 IgA, and about 7 will be missed. If 100 people without coeliac disease are tested, about 4 will have a false positive.

**HLA negative predictive value**: If you test negative for HLA-DQ2 and HLA-DQ8, your chance of having coeliac disease is less than 1 in 100. This is the most powerful rule-out test available.

**Dietary response**: Among adults with coeliac disease who strictly adhere to a gluten-free diet, ~70–80% will have normalised serology within 6–12 months, but only ~30–50% will have complete histological healing (normal villous architecture) at 1 year. In children, histological healing occurs in ~80–90% within 1–2 years.

**Risk of lymphoma**: Coeliac disease carries a ~2–4 fold increased risk of non-Hodgkin lymphoma compared to the general population. A strict gluten-free diet reduces this risk to near-baseline levels after 5 years of adherence.

**NCGS prevalence**: In population-based studies, ~10–15% of people report gluten-related symptoms, but when tested with double-blind placebo-controlled challenges, only ~1–3% actually react to gluten specifically (the rest react to FODMAPs or have nocebo effects).

**What the authors acknowledge:**

Many recommendations are based on low-quality evidence (observational studies, expert opinion) because randomised controlled trials are often unethical or impractical in coeliac disease.

The definition and diagnostic criteria for NCGS remain controversial and poorly validated.

The guideline does not address the management of coeliac disease in resource-limited settings where serology and biopsy may not be available.

Long-term outcomes of a gluten-free diet (beyond 5–10 years) are not well studied.

**What a critical reader would note:**

**Industry funding**: The guideline does not declare specific funding sources for the development process. Several authors have ties to pharmaceutical companies developing novel therapies for coeliac disease (e.g., ImmusanT, Alvine Pharmaceuticals), which could introduce bias in recommendations about non-dietary treatments.

**Publication bias**: The literature search may have missed negative studies or studies published in non-English journals (despite stating "no language restrictions," PubMed indexing is English-biased).

**Lack of patient involvement**: The guideline was developed by clinicians and researchers only. No patient representatives were included in the consensus process.

**Outdated search**: The literature search ended in August 2018. Since then, several important studies have been published, including large randomised trials of novel therapies (e.g., latiglutenase, ALV003) and updated epidemiological data.

**Confounding in NCGS diagnosis**: The guideline recommends diagnosing NCGS based on symptom improvement on a gluten-free diet and relapse on gluten challenge, but does not adequately address the high rate of nocebo effects (up to 40% in some double-blind studies) or the confounding role of FODMAPs.

**Population limits**: The guideline focuses on Western populations. Coeliac disease prevalence and presentation may differ in Asian, African, and Middle Eastern populations, where wheat consumption patterns and genetic backgrounds vary.

For someone running their own n=1 experiment:

### What to test

**If you suspect coeliac disease**: Do NOT start a gluten-free diet before getting tested. You need to be eating gluten (at least 1 slice of bread per day for 6–8 weeks) for accurate serology and biopsy. Test: TG2 IgA + total IgA (to rule out IgA deficiency). If positive, confirm with upper endoscopy and duodenal biopsy.

**If you suspect non-coeliac gluten sensitivity**: First, rule out coeliac disease (serology + biopsy while on gluten). Then, try a strict gluten-free diet for 2–4 weeks. If symptoms improve, do a double-blind placebo-controlled gluten challenge (e.g., 5–10 g of gluten powder vs. placebo in capsules, 1 week each, with a 1-week washout). This is the only way to confirm NCGS and rule out nocebo effects.

**If you have dermatitis herpetiformis**: See a dermatologist for a skin biopsy with direct immunofluorescence. Do not start a gluten-free diet before the biopsy.

### Minimum meaningful duration

**For coeliac disease diagnosis**: 6–8 weeks of gluten-containing diet before testing (at least 1 slice of bread per day).

**For gluten-free diet trial**: 2–4 weeks for symptom improvement in NCGS; 6–12 months for serological normalisation in coeliac disease.

**For gluten challenge**: 1–2 weeks per challenge phase, with a 1-week washout between phases.

### What to measure (specific metrics)

**Primary outcome**: Symptom severity (use a daily symptom diary: abdominal pain, bloating, diarrhoea, fatigue, brain fog, skin rash — rate 0–10 each day).

**Secondary outcomes**:

- Stool frequency and consistency (Bristol Stool Chart)

- Energy levels (1–10 scale)

- Skin lesions (photograph and count)

- Weight (weekly)

**Objective biomarkers** (if accessible):