Alimentary Tract
Impact of gluten-free diet on cardiovascular risk factors. A retrospective analysis in a large cohort of coeliac patients

https://doi.org/10.1016/j.dld.2013.04.001Get rights and content

Abstract

Background

Concerns have been raised on whether a gluten-free diet affects the cardiovascular risk profile of coeliac patients.

Aims

To assess changes of multiple cardiovascular risk factors in coeliac patients evaluated before and during a gluten-free diet.

Methods

Retrospective analysis of the effects of 1–5 years of gluten-free diet on indicators of cardiovascular risk and on distribution in cardiovascular risk categories in 715 coeliac patients.

Results

Compared to baseline, significant increases were found in body mass index (21.4 ± 3.4 vs. 22.5 ± 3.5; p < 0.0001), total cholesterol (171.2 ± 37.4 mg/dL vs. 181.4 ± 35.1 mg/dL; p < 0.0001), and γ-glutamyl transpeptidase (16.5 ± 14.9 vs. 19.5 ± 19.2 U/L; p < 0.0001). Significant reductions were found in serum triglycerides (87.9 ± 49.5 vs. 80.2 ± 42.8 mg/dL; p < 0.0001) and homocysteine (16.9 ± 9.6 vs. 13.3 ± 8.0 μmol/L; p = 0.018) during gluten-free diet. The proportion of patients included in an arbitrarily defined category of “lowest cardiovascular risk profile” decreased from 58% at baseline to 47% during gluten-free diet.

Conclusions

A gluten-free diet significantly affects cardiovascular risk factors in coeliac patients, but changes do not consistently point towards worse or better risk profiles, thus suggesting that the diet is unlikely to be atherogenic.

Introduction

Coeliac disease (CD) is an autoimmune enteropathy that develops in genetically predisposed subjects and is triggered by ingestion of gluten, found in wheat, and of related prolamines, found in barley and rye [1]. The histological hallmark of the disease is villous atrophy. Clinical manifestations are heterogeneous and largely related to intestinal malabsorption that, in most cases, results in selective deficiency of dietary components such as iron, vitamins, calcium or cholesterol, and in a minority of cases results in a classical malabsorption syndrome. Due to the disease characteristics the majority of CD patients has been reported to have normal or low BMI in population based [2] and outpatient based studies [3], [4], [5]. Furthermore, when compared to controls cholesterolemia has been reported to be lower in undetected celiacs in the general population [6] and in selected coeliac patients referred to a CD clinic [7], [8]. These latter observations suggest not only that untreated CD patients may have a favourable cardiovascular (CV) risk profile, but also that CD may confer protection from ischaemic heart disease [6], [8]. However, results from epidemiological studies are rather contradictory on this point [9], [10], [11], [12], [13], [14], [15].

The only treatment for CD is lifelong adherence to a gluten-free diet (GFD) [1], [16], which is expected to result in reconstitution of the small intestinal villous structure, symptoms disappearance and improvement of nutritional status and general well being. Most of the emphasis on GFD focuses on strict avoidance of gluten [1], [16], [17], but available information suggests that this diet may not be nutritionally adequate because it is rich in simple carbohydrates [18], [19] and saturated fatty acids [20] and poor in fibres, micronutrients, vitamins and complex carbohydrates [18], [19], [21], [22], [23].

Taken together the observations reported above have raised concerns that GFD may alter the favourable CV risk profile of untreated CD patients, and thus the concept of active identification and treatment of asymptomatic undiagnosed coeliac patients has been questioned [24]. Available evidence is scarce, limited to the assessment of a single risk factor [2], [5], [8], [24], [25], [26] and, with few exceptions [3], [4], mostly based on small numbers of patients. In some studies, the BMI has been reported to increase during GFD [3], [27], especially in underweight patients [4], and to be higher in treated CD patients than in untreated patients [27]. Other studies assessing changes in lipid profile and homocysteine levels, reported increased high-density lipoprotein (HDL) cholesterol levels [8], [24], [27], with increased [24], [27] or unchanged [8] total cholesterol, and decreased homocysteine levels [5], [25], [26]. These results do not provide clear-cut evidence on whether GFD determines a better or a worse CV risk profile in CD patients.

The aim of our retrospective study was to assess changes in multiple factors associated with CV risk in a large cohort of CD patients studied before and during GFD.

Section snippets

Materials and methods

Clinical, serological and histological data, collected before and during GFD, were retrospectively collected from a prospectively maintained database of all CD patients referred to our CD Clinic from January 1990 to July 2011. CD was diagnosed on the basis of typical duodenal histopathology and positive CD-related serology, either tissue transglutaminase (tTG) or anti-endomysial antibodies (EMA). The information extracted from the database included: demographic characteristics, CD-related

Statistical methods

Continuous variables are expressed as means ± standard deviation (SD) and categorical variables are expressed as frequencies and proportions. Differences between results obtained before and during GFD were tested for statistical significance using the t-test for paired observations of continuous variables, and the χ2-test or Fisher's exact test for categorical variables, as appropriate. Statistical analysis was performed using SAS® 9.2 (Cary, NC, USA) and GraphPad Prism® 5.0 (La Jolla, CA, USA)

Patients’ characteristics

Of the 1740 CD patients recorded in our CD database, 923 had laboratory tests available both at baseline and between 1 and 5 years during GFD, of which 759 had no associated disease and 715 were older than 15 years of age, thus fulfilling our inclusion criteria. Anthropometric, clinical and histological characteristics of patients at time of CD diagnosis are detailed in Table 1. Mean age at diagnosis was 35 ± 13 years and 70.5% were female. In 82% the reason for first referral was presence of

Discussion

Our study shows that at baseline the majority of CD patients were in the low-risk category for CV risk, and that GFD significantly alters these factors causing one third of patients to switch to different CV risk categories.

When considering BMI, 17% of our patients was underweight and 14% overweight/obese at baseline. This distribution is remarkably similar to previous reports [2], [4], although a lower prevalence of underweight patients [3], [9] and a higher prevalence of overweight patients

Conflict of interest statement

None declared.

Acknowledgments

We are grateful to Nurse Elvira Spreafichi and to Mrs Anna Tomasini for their assistance in our clinical practice.

References (41)

  • O. Olen et al.

    Coeliac disease and body mass index: a study of two Swedish general population-based registers

    Scandinavian Journal of Gastroenterology

    (2009)
  • W. Dickey et al.

    Overweight in coeliac disease: prevalence, clinical characteristics, and effect of a gluten-free diet

    American Journal of Gastroenterology

    (2006)
  • J. Cheng et al.

    Body mass index in coeliac disease: beneficial effect of a gluten-free diet

    Journal of Clinical Gastroenterology

    (2010)
  • W. Dickey et al.

    Homocysteine and related B-vitamin status in coeliac disease: effects of gluten exclusion and histological recovery

    Scandinavian Journal of Gastroenterology

    (2008)
  • J. West et al.

    Seroprevalence, correlates, and characteristics of undetected coeliac disease in England

    Gut

    (2003)
  • N.R. Lewis et al.

    Cholesterol profile in people with newly diagnosed coeliac disease: a comparison with the general population and changes following treatment

    British Journal of Nutrition

    (2009)
  • J. West et al.

    Risk of vascular disease in adults with diagnosed coeliac disease: a population-based study

    Alimentary Pharmacology and Therapeutics

    (2004)
  • L.A. Anderson et al.

    Malignancy and mortality in a population-based cohort of patients with coeliac disease or “gluten sensitivity”

    World Journal of Gastroenterology

    (2007)
  • C. Canavan et al.

    No difference in mortality in undetected coeliac disease compared with the general population: a UK cohort study

    Alimentary Pharmacology and Therapeutics

    (2011)
  • J.F. Ludvigsson et al.

    Small-intestinal histopathology and mortality risk in coeliac disease

    Journal of the American Medical Association

    (2009)
  • Cited by (0)

    View full text