Article Text
Abstract
Introduction Stable angina is diagnosed on the basis of clinical assessment and diagnostic testing. Angina-type pain is defined by: (i) constricting discomfort in the chest, neck, shoulders, jaw, or arms, (ii) precipitated by physical exertion, and (iii) relieved by rest or glyceryl trinitrate (GTN). Typicality of these symptoms, in addition to risk factor burden, is used to estimate the pre-test probability of coronary artery disease, guiding investigation and management. However, it remains unclear whether Rapid Access Chest Pain Clinic (RACPC) assessment provides advantages to that in primary care. We studied patients with suspected stable angina, to evaluate concordance between GP and cardiologist findings, and identify barriers to diagnosis in primary care.
Methods A database of all patients reviewed in a high-volume, cardiologist-led RACPC was prospectively maintained, from Jul 2012–Jan 2015. Standardised GP referral proformas were used to code typicality of angina pain (3 features -– typical angina; 2 features – atypical angina; 0 or 1 feature – non-angina), and missing data identified. Concordance between GP and cardiologist assessment was ascertained using Cohen’s kappa (K) and Bland-Altman plots. Age, sex, typicality of angina, presence of risk factors (smoking, diabetes or hyperlipidaemia) and ECG criteria (presence of Q-waves or ST/T changes) were used to estimate GP and cardiologist-calculated pre-test probabilities, compared using Student’s t-test. Thematic analysis of free-text clinical details was performed, to explore reasons for missing data. All data are presented: mean [SD].
Results Data regarding n = 1 928 referrals were available. After exclusion of cases for which referral was via non-structured or incomplete letter, 1 634 cases were included in the final analysis. Agreement between GP and cardiologist in typicality of angina was poor (723/1 634, 44.2%; K = 0.235 p < 0.001; Table 1, Figure 1). Pre-test probability calculated using GP assessment was greater than that assessed by cardiologist (54.7 [25.0] vs. 40.9% [22.2]; p < 0.001), although a strong positive correlation was observed (R2 = 0.56, p < 0.001; Figure 2). GP referral data regarding character of chest pain were missing in n = 10 (0.6%). Thematic analysis demonstrated this was primarily due to non-pain related symptoms, predominantly breathlessness. No data regarding relationship to exertion were available in n = 57 (3.5%), due to short duration of pain. No data regarding relationship to rest or GTN were available in n = 475 (29.1%), attributable to the referrer only considering GTN, which was often only prescribed at the point of referral.
Conclusions GP and cardiologist-led diagnosis of stable angina is discordant; the pre-test probability is overestimated in primary care, which would have significant implications for subsequent diagnostic investigation. The RACPC remains an essential service for patients with suspected stable angina.
- Stable IHD
- Prevention
- Risk assessment