The Characterisation and Risk Factors of Ischaemic Heart Disease in Patients With Coeliac Disease

L. Emilsson; R. Carlsson; M. Holmqvist; S. James; J. F. Ludvigsson


Aliment Pharmacol Ther. 2013;37(9):905-914. 

In This Article


Study Design

In this study we conducted a case-only comparison. We compared the clinical presentation of IHD (hospitalised myocardial infarction or death from MI) and IHD risk-factor profiles in patients with biopsy-verified CD to reference individuals from the general population at time of their first myocardial infarction.


In Sweden, small-intestinal biopsy is routine clinical practice for the diagnosis of CD. In fact, 96–100% of Swedish gastroenterologists and paediatricians[11] perform a biopsy before assigning a diagnosis of CD. There is universal access to publicly funded health care, including in-patient care, for all residents. Through the unique national personal identity number issued to all Swedish residents,[12] data from national and virtually complete administrative or clinical registers on demographics, morbidity and mortality can be linked.

CD Patients

In this study, we used a previously identified and defined population-based cohort of CD patients.[11] To put this cohort together data on all pathology reports performed from 1969 to 2008 of small-intestinal biopsies with a morphology code indicating villous atrophy were gathered from all of Sweden's 28 regional pathology departments. A validation study of a subset of these pathology reports (n = 114) was performed using information on clinical CD diagnosis from patient charts as gold standard. The validation study indicated that 95% of all individuals with a pathology report indicating villous atrophy also had CD.[11] Based on this high validity, all individuals with a pathology report from a small-intestinal biopsy indicating villous atrophy (Marsh stage 3) (for a detailed list of relevant morphology codes, see the Appendix) were included in the CD cohort. Positive CD serology was not required to be classified as having CD, but serology was positive at diagnosis in 88% of the CD patients.[11] Detailed information on the CD cohort and the validation study has been published previously.[11] In total, we identified 29 096 individuals with biopsy-verified CD. Each individual entered the study at the date of CD diagnosis.

Reference Population

For each patient with CD, we randomly selected five individuals with no previous record of small-intestinal biopsy from the Total Population Register (which includes all Swedish residents). Each reference individual was matched for age, gender, county and calendar year. In all, 144 522 reference individuals were identified and all entered the study at the date of CD diagnosis of their matched CD patient.

Data Sources

Using the personal identity number, we linked the CD patients and their matched reference individuals to the following data sources for which data were available through 31 December 2008: The Swedish Patient Register, the Total Population Register, the Cause of Death Register and SWEDEHEART. The Swedish Patient Register contains information on in-patient care since 1964, with nationwide 100% coverage since 1987.[13] The register lists date of admission, date of discharge and the discharge diagnosis (primary and secondary diagnoses) as set by the discharging physician and classified according to the calendar year-specific ICD. The Total Population Register includes information on deaths, emigration and immigration for the entire Swedish population.[14] The Cause of Death Register has complete coverage of all deaths occurring and includes information on all causes of death. It was initiated in 1951, is updated yearly and is based on death certificates.[15] The Swedish cardiac care register SWEDEHEART was formed by a merger of earlier quality registers: the Register of Information and Knowledge About Swedish Heart Intensive Care Admissions (RIKSHIA), the Swedish Coronary Angiography and Angioplasty Registry (SCAAR), the Swedish Heart Surgery Registry, the Thoracic Surgery Registry and the National Registry of Secondary Prevention (SEPHIA).[16] SWEDEHEART has a 100% coverage of coronary care units and virtually 100% coverage of all coronary angiographies performed in Sweden.[16] It contains information on traditional cardiovascular risk factors and treatment at the time of admission, laboratory values such as total cholesterol, haemoglobin (Hb) levels, creatinine levels and CRP at admission, as well as presenting symptoms of IHD and other clinical characteristics, indication for angiography, findings from the procedure and what treatment decisions are made based on the findings. There is also information on traditional cardiovascular risk factors and information on previous coronary artery bypass graft (CABG) or percutaneous coronary intervention (PCI). Coronaries were classified as significantly stenosed if a >50% stenosis could be detected and considered clinically important or if the fractional flow reserve was <75%. For detailed information on SWEDEHEART, please be referred to previously published studies.[16,17]

By linking the CD patients and their matched reference individuals to these registers, we identified all admissions listing myocardial infarction or angina pectoris, all individuals included in SWEDEHEART, all deaths from IHD and all emigrations from Sweden during follow-up.

Study Population

All CD patients and their matched reference individuals with an IHD diagnosis before study entry were excluded from the analyses and only first-time myocardial infarctions registered in RIKSHIA after study entry were eligible. After excluding all patients with a prior event, 1075 CD patients and 4142 reference individuals with an ICD code of IHD in the Swedish Patient Register or the Cause of Death Register and who had a record in SWEDEHEART during follow-up (i.e. two independent records of IHD) were identified. Figure 1 presents a flow chart of exclusions. Calendar-specific ICD codes used to define IHD are found in the Appendix.

Figure 1.

Flow chart of study population. CD, coeliac disease; MI, myocardial infarction; PCI, percutaneous coronary intervention; PR, Patient Register (Swedish National). # Patients with a history of prior MI were excluded since we were only interested in individuals with their first MI.

Number of Patients in the Subgroup Analyses

Of the 1075 CD patients and 4142 reference individuals identified, 430 CD patients and 1988 reference individuals had a record of first myocardial infarction in SWEDEHEART. Data were available from angiographies in 742 CD patients and 2901 reference individuals (Figure 1).

Statistical Analyses

We used logistic regression adjusted for age group (0–59, 60–69, 70–79, 80–95) and gender to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for binary outcome variables. We tested statistical significance for continuous variables with the Mann–WhitneyU-test as well as unianova, adjusted for age at SWEDEHEART admission, calendar period and gender. In all our analyses the comparison was made between CD patients with MI/coronary angiography vs. reference individuals with MI/coronary angiography. The statistical analyses were nonstratified and we did not consider any time frame in the analyses since the purpose of this study was to describe the clinical presentation of first MI in CD patients compared with reference individuals at the time of the first infarction. In a sensitivity analysis we restricted our data to patients with an IHD indication for angiography because angiographic evaluation may be undertaken for non-ischaemic reasons. We also excluded individuals with earlier PCI or CABG. This analysis was based on 292 CD patients and 1107 reference individuals with primary angiography that was due to strictly ischaemic indication. In post hoc analyses we evaluated the angiographic findings of 251 CD patients and 1180 reference individuals with an angiography performed within 10 days before and up to 30 days after their first SWEDEHEART admission. In a separate analysis we also examined the proportion of coeliac patients (and MI) who had persistent villous atrophy on second biopsy (0.5–5 years after the first biopsy). Persistent villous atrophy may correlate with chronic inflammation and could potentially contribute to IHD. We defined statistical significance as 95% CIs for risk estimates not including 1.0. We used pss version 20.0 (SPSS Inc, Chicago, IL, USA) for all analyses.

Ethics and Role of Funding Source

This study was approved by the Research Ethics Committee of Karolinska Institutet. The funding sources did not influence the study's design, conduct and reporting.