Article Text

Download PDFPDF

Should we screen adults with osteoporotic fractures for coeliac disease?
  1. A A Fisher1,
  2. M W Davis1,
  3. M M Budge2
  1. 1Department of Geriatric Medicine, Canberra Hospital, ACT, Australia
  2. 2Department of Geriatric Medicine, Canberra Hospital, ACT, Australia and Australian National University Medical School, Woden, AC, Australia
  1. Correspondence to:
    Associate Professor M M Budge
    PO Box 11, Woden, AC, Australia;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

In the recently published debate in Gut regarding the utility of mass screening of European and North American populations for coeliac disease (CD), divergent conclusions were presented (Gut 2003;52:168–9 and 170–1). In this context, the increased utility of screening adults for CD in those presenting with concomitant morbidity (for example, metabolic bone disease and fracture) was raised. To support such an hypothesis, evidence of either an increased fracture rate in those with CD or, alternatively, an increased incidence of CD in those presenting with fracture would be required.

Thomason and colleagues,1 in a study of 244 patients with CD and 161 age and sex matched controls, addressed the first of these possibilities. They found that patients with CD “as a whole do not represent a population of particularly high risk of osteoporotic fracture”.

Available data regarding the prevalence of CD in older people with osteoporosis are limited and controversial. Several reports2,3 have suggested an increased prevalence of CD among patients with idiopathic osteoporosis leading the authors to recommend screening for CD of all osteoporotic patients. However, these findings are not supported by other studies.4,5 In a study aimed at determining the prevalence of previously undetected secondary contributors to osteoporosis in otherwise healthy older women (mean age 65.5 years), the incidence of CD was 1.7%.6

Consequently, it seems important to know whether in older adults screening for CD in those presenting with osteoporotic fractures would yield a significant number of unsuspected cases. Osteoporotic hip fracture, a dramatic consequence of osteoporosis and a leading cause of morbidity and mortality in older people, has been reported in association with clinically silent CD.7,8 However, to our knowledge, serological screening tests for CD have not been systematically studied in older adults with hip fracture.

We screened the serum of 347 consecutive older patients (>60 years of age) with hip fracture (74% females; age range 60–101 years, mean age 81.5 (SD 7.3) years) for the presence of IgA endomysial antibodies (EMA), IgA and IgG gliadin antibodies (IgA-AGA and IgG-AGA), and total IgA. In 13% of patients, the IgA-AGA test was positive (above 34 ELISA units) while in 11% of patients the titre of IgG-AGA was slightly elevated (above 46 ELISA units). However, none of the patients had a positive anti-EMA test which is known to have a high specificity (98–100%).9 This negative finding is particularly noteworthy given that 86% of the screened population had a low body weight (<60 kg), 79.1% had low serum 25-hydroxyvitamin D concentrations (<50 nmol/l), 69% had secondary hyperparathyroidism (serum PTH >5.5 pmol/l), and 21.6% had anaemia (haemoglobin <110 g/l). Such abnormalities are often associated with CD and are believed to contribute to the development of osteoporosis in CD. Therefore, one might expect that investigation of a cohort of older adults with osteoporosis presenting with a hip fracture might yield a moderate number of people with subclinical CD. However, this was not the case in this analysis. Our findings indicate that CD appears not to be an important contributing pathogenic factor in an older hip fracture population with osteoporosis. It further suggests that routine screening for CD in a similar population, or even in those individuals with a hip fracture and accompanying hypovitaminosis D and/or secondary hyperparathyroidism, would have a low yield and not be cost effective.

Despite these findings, we would continue to encourage physicians evaluating older adults to consider, but not routinely screen for, CD when unexplained metabolic bone disease presents even in the absence of gastrointestinal complaints and/or dermatitis herpetiformis.


Linked Articles