Clinical response to gluten withdrawal is not an indicator of coeliac disease

Although the diagnosis of coeliac disease requires specific histological and serological findings, patients considered to be affected by coeliac disease only on the basis of clinical improvement after gluten withdrawal are commonly referred to our outpatient clinic. The objective of this study was to investigate whether the clinical response of gastrointestinal symptoms to gluten withdrawal and subsequent dietary re-introduction could be an indicator of the presence of coeliac disease.

From December 1998 to January 2007, 180 patients on a gluten-free diet because of a diagnosis of coeliac disease not based on proper diagnostic criteria came to our out-patient clinic. In 112 of these patients, gluten was re-introduced into their diet. Subsequent duodenal biopsies and endomysial antibodies confirmed the diagnosis of coeliac disease in 51 of them. The relationship between improvement/worsening of symptoms and withdrawal/re-introduction of dietary gluten was analysed.

Gastrointestinal symptoms improved in 64.7% of coeliac patients and 75.0% of non-coeliac patients after gluten withdrawal (chi(2) test, p=NS). Gluten re-introduction was followed by clinical exacerbation in 71.4% of coeliac patients and 54.2% of non-coeliac patients (chi(2) test, p=NS). The positive predictive value for clinical improvement after gluten withdrawal was 36%; the positive predictive value for clinical exacerbation after gluten re-introduction was 28%.

Clinical response to either withdrawal or re-introduction of dietary gluten has no role in the diagnosis of coeliac disease.