Background

Coronary artery disease (CAD) is a major cause of mortality in COPD but may be overlooked due to overlap in symptoms. There is a need to define the extent of subclinical CAD in COPD, improve diagnosis and manage future risk.

Aims

To characterise the prevalence of CAD on CT in a COPD cohort, correlate to symptoms and explore non-invasive means of detection.

Methods

50 subjects from the London COPD Exacerbation Cohort underwent CT coronary angiogram, spirometry, symptom scores, 6-minute walk(6MW), QRISK®3 and cardiac biomarkers. CTs were split into 3 groups: no CAD, non-obstructive CAD (<50% stenosis in ?1 vessel) or obstructive CAD (?50% stenosis in ?1 vessel).

Results

49 CTs were analysed. 6(12%) had no CAD, 23(47%) had non-obstructive CAD and 20(41%) had obstructive CAD. 75% with CAD, and 60% with obstructive CAD on CT did not have a previous diagnosis or chest pain. Those with CAD did not report more dyspnoea but did have a reduced 6MW distance, correcting for age, sex, FEV1 and BMI (P=0.02, linear regression).

Troponin I was associated with CAD in univariate (no CAD: median 2.35ng/L (IQR 2.3-2.5), non-obstructive CAD: 3.3ng/L (2.3-4.7), obstructive CAD: 4.2ng/L (2.55-6.05); P=0.03, Kruskal-Wallis test), but not multivariate analysis. QRISK®3 did not differ between groups (median 20% (IQR 17-27), 25% (18-32), 25% (19-41); P=0.6, Kruskal-Wallis test). Comparing ROC curves, troponin performed best (AUC 0.68 (95%CI 0.52-0.83) for obstructive CAD and 0.78 (0.63-0.93) for any CAD).

Conclusions

CAD is common in COPD, frequently occurring without classic symptoms. QRISK®3 did not differentiate those with CAD. Troponin I and 6MW were more useful. Identifying screening methods should be a priority for future studies.