Coeliac disease

Autoimmune process triggered by gluten, distinct from type 1 IgE mediated wheat allergy –  leads to subtotal villous atrophy in terminal ileum. Multifactorial, genetics important – at least 90% positive for DQA1*05 and DQB1*02 alleles that code (confusingly) for DQ2 and/or the DQB1*03:02 allele that codes for DQ8 molecules.  The risk of having coeliac disease without these variants is less than 0.1% so good negative predictive value, but very little positive predictive value – you may not have it now, but you may well get it in the future.

1 in 200, mostly undiagnosed/adult, decreasing in children. More girls than boys, 10% risk to 1st degree relatives.

Randomized trial of early introduction of gluten (4 months, rather than UK recommended 6 months) in 1300 babies found not a single case of coeliac disease at age 3, cf 7 cases in later introduction group (JAMA Ped 2020). Similar result to peanut in EAT study of course.


Become less convincing the older you are! Can present even in infancy, although diagnosis may be delayed for a year or more.

  • Poor appetite
  • Short stature, growth faltering, weight loss
  • Abdominal pain (dyspepsia in adults), nausea (not vomiting, interestingly), abdominal distension (bloating)
  • Chronic diarrhoea (but can present with constipation in adults!)
  • Chronic lethargy
  • Poor motor skills (muscle wasting)

Should also be considered in patients with:

  • Idiopathic short stature!
  • Constipation!
  • IBS (in adults)!
  • Dermatitis herpetiformis (chronic papular or blistering rash)
  • iron, vitamin B12 or folate deficiency anaemia resistant to treatment, or unexplained
  • chronic unexplained elevations of aminotransferases,
  • recurrent aphthous ulcers
  • Dental enamel defects
  • Osteoporosis, Pathological fractures
  • Delayed menarche

Should also be considered for:

  • unexplained arthritis,
  • Unexplained neurological problem esp neuropathy, palsy, ataxia
  • Subfertility and recurrent miscarriage
  • Epilepsy with associated intracranial calcification!
  • Migraine (headache quite a common symptom in older children)!

[BSPGHAN guidelines 2013, NICE 2015]


For symptomatic patients, measure IgA anti-tTG antibody, plus total IgA (because IgA deficiency is associated with coeliac disease and gives false negatives). Also good to check FBC, ferritin, folate, LFTs, Calcium, prothrombin time.

  • Positive tTG alone is insufficient for diagnosis, and should not be a reason for gluten free diet.
  • If tTG is high but less than 10x upper limit, then proceed to duodenal biopsy.
  • If tTG more than 10x upper limit of normal, alternative to biopsy is to check IgA EMA and do HLA DQ2/8 screening. If BOTH positive, diagnosis is confirmed. If EMA not available, a second strongly positive tTG is an acceptable alternative (but save serum for EMA testing at future date).
  • If IgA deficient, use IgG tTG or EMA, but less specific so low threshold for biopsy.

Note that antibodies may be false negative if the diet is low in gluten at the time of testing. So normalize gluten intake (at least 2 meals per day, for at least 6 weeks) but discuss with specialist if clinical condition so poor that treatment cannot be safely delayed.

False positives also occur, so diagnosis must be confirmed by endoscopy (or another diagnosis may become apparent). Endoscopy for high risk symptoms (diarrhoea, wt loss, anaemia) and serology is 100% sensitive for coeliac. But a proportion with high risk symptoms and positive serology do NOT have positive biopsies! Cause? Perhaps atypical disease (ie no GI symptoms) is actually more common than typical! Consider the incidences of anaemia, infertility, short stature, unfavourable pregnancy outcomes…


Coeliac disease is associated with other conditions, which may justify screening:

  • Type 1 diabetes (8%)
  • IgA deficiency (up to 7%)
  • Downs, Williams and Turners syndromes!
  • Autoimmune thyroiditis
  • Autoimmune liver disease
  • Unexplained abnormal LFTs
  • Relatives of coeliac disease patient (10% in 1st degree relative)

Before testing, discuss relative risks of untreated coeliac disease (osteoporosis, infertility, small bower cancer), and the need for biopsy and gluten free diet if tests are positive. Then test:

BSPGHAN say check HLA DQ status and IgA tTG.  NICE says do not use HLA DQ2/8 in non-specialist settings, but consider if not proceeding to biopsy or those on low gluten.

  • If HLA DQ2/8 neg, then v unlikely.
  • If DQ pos but tTG neg, then unlikely (unless IgA deficient or on low gluten diet). Still potential to get disease later in life, so test again in 3yrs or if symptomatic.
  • If DQ/tTG pos but tTG less than 3x upper limit of normal, then do IgA endomysial antibody (EMA) and proceed to biopsy if positive. If EMA negative, continue to monitor.
  • If DQ pos and tTG higher than 3x, biopsy. (Although option is given to retest in 3-6 months)
  • If IgA deficient (<0.07mg/L), consider using IgG EMA, deamidated gliadin peptide (DGP), or tTG
  • Have a low threshold for retesting if signs/symptoms consistent!


Villous atrophy on biopsy (do 4) is characteristic. Marsh grading system: infiltrative changes with crypt hyperplasia is compatible, positive serology strengthens diagnosis. If diagnosis still uncertain, consider HLA testing as above, repeat biopsy after increasing gluten intake or else go the other way, and biopsy after gluten free diet!

Biopsy if gluten excluded and persistent high titres at 12 months or persistent symptoms.  Serological tests alone not adequate to determine if gluten has been excluded!


The benefits of treatment, in symptomatic children are:

  • resolution of symptoms,
  • reversal of bone demineralisation,
  • Resolution of micronutrient deficiencies and probably better height gain
  • less delayed puberty, menstrual problems, subfertility, spontaneous abortions and LBW babies
  • decreased risk of some intestinal cancers, to normal levels (T cell lymphoma, Ca oes/pharynx).  T cell lymphoma risk falls to baseline after 10 years gluten free diet
  • Possible prevention of associated autoimmune conditions (evidence conflicting) – see below

In asymptomatic patients, there are no long term studies as yet to show benefits so need to discuss with family.


Food labelling allows up to 200ppm gluten in gluten-free foods: some people may be sensitive at lower levels than this (at least, their mucosa may be sensitive – they may remain asymptomatic). Gluten free foods are available on prescription.

Oats are safe for 95% of patients (besides contamination issues and the above food labelling problem). Ideally exclude initially and consider reintroduction (products marked gluten free) when well eg at 1yr; normalization of tTG and continued low titres reassuring!  NICE says can be included at ANY stage, but review immunological/clinical/histological response.

Most coeliac patients can tolerate wheat starch and malt extract but not all.

Follow up

If compliant on diet, antibodies should go negative within a year. The mild transaminitis commonly seen at diagnosis should resolve – if not, and on diet, then consider Autoimmune Hepatitis. Complete mucosal recovery can take years.

Monitor symptoms, growth, puberty, antibodies (eg 6 monthly, then 12-24 monthly). Poor clinical response = poor compliance! TTG, FBC, Ca/Phos/ALP, Ferritin, Vit B12, folate, Vit D, PTH, TFTs.

Consider bone scan – see osteoporosis

Discuss oats, as above.

Pneumococcal vaccine recommended.


Understanding of diagnosis, antibodies.

  • Encourage maturation of communication and decision-making skills.
  • Allow patients to take responsibility for medical self-management.
  • Help the patient develop healthy habits and self-care skills that encourage autonomy eg smoking, exercise, alcohol/drugs
  • Education and counselling re: gluten-free diet (food on prescription, nutritional completeness, healthy weight) and consequences of non-adherence.
  • Recognition and treatment of psychological/emotional issues eg : discouragement, feeling overwhelmed, anxiety about the future and complications such as depression and eating disorders.
  • Familiarize young person with healthcare system

For those patients with significant pubertal delay, paediatric provider may be better suited to provide guidance until transition to an adult provider at the completion of puberty.

Topics for discussion –

  • How there can be a long interval between gluten exposure and the return of symptomatic disease.
  • How coeliac disease can interfere with school, education and work (eg military service)
  • Sexuality and fertility (women with untreated CD are more likely to suffer an adverse pregnancy outcome) – lifetime fertility however is similar in individuals with and without coeliac
  • Adolescents report lower adherence than younger children, esp at social events. Dietary non-adherence is associated with poorer QoL and increased physical symptoms. Most young people with CD think that avoiding cancer is the most important reason to adhere to diet, but the risk for cancer is much lower than previously presumed, so osteoporosis and adverse pregnancy outcome may be bigger issues.
  • Risk of passing it on to children.
  • Associations with diabetes, thyroid disease, autoimmune liver disease.

Primary care may be a suitable care provider.

[Prague consensus, Ludvigsson Gut 2016;65:1242-1251.]

Coeliac UK support group