10 year hard outcomes in stable chest pain patients following stress echocardiography vs computed tomography coronary angiography: a retrospective propensity matched analysis

Background Computed tomography coronary angiography (CTCA) is advocated as first line in the assessment of chronic coronary syndrome in patients with a lower pre-test probability of obstructive coronary artery disease (CAD). However, recent PROMISE data showed no difference in long term mortality between CTCA or functional assessment. Therefore, tests requiring no iodinated contrast or ionising radiation, such as stress echocardiography (SE) may be a safer alternative to patients with a broader spectrum of pre-test probability scores. Long term outcomes specifically for SE vs CTCA have not yet been explored. Purpose To assess long term hard outcomes (combined all-cause death and myocardial infarction) and revascularisation rates in patients undergoing either CTCA or SE and whether these are affected by pre-test probability score. Method A single centre retrospective database search of patients presenting for assessment between 2013-2014 with no history of CAD who subsequently underwent CTCA (including either coronary calcium score alone or combined with CT angiography) or SE was undertaken. These results were propensity matched in a 1:1 ratio using 11 predisposing CAD risk factors. Patients' pre-test probabilities for obstructive CAD were scored using the risk factor clinical likelihood (RF-CL) model and NICE chest pain typicality groups. Results Over a median follow up of 10.4 years (interquartile range 9.9 to 10.9 years) 1964 patients (984 CTCA and 980 SE) were analysed (age 59.3 ± 12.3, 47.8% males). There was no difference in hard outcomes (HR 0.92; 95% CI, 0.73-1.17; p=0.5) between modalities but a significant increase in revascularisations in the CTCA arm (HR 1.47; 95% CI 1.06-2.05, p=0.02). Furthermore, there was no differences in hard events between modalities for any RF-CL group (A: ≤5%, B: >5-≤15%, C: >15%) but revascularisations were higher in the low (p=0.03) and moderate (p=0.04) RF-CL groups in the CTCA arm. Patients grouped according to NICE chest pain typicality also showed no differences between modalities for any of the groups (angina equivalent, non-anginal, atypical and typical symptoms). A secondary propensity matched sub analysis, excluding those with calcium scoring only, compared patients that underwent full CTCA (n=638) vs SE (n=633). No difference in hard outcomes (HR 0.85; 95% CI 0.63-1.17, p=0.32) or revascularisations (HR 1.39; 95% CI 0.9-2.16, p=0.14) between modalities were noted, including the RF-CL groups, with the exception of the >15% group where there was an increase in revascularisations in the CTCA arm (HR 2.3; 95% CI 1.11-4.78, p=0.03). Conclusion In patients requiring investigation for chronic coronary syndrome, there were no differences in combined all-cause death and MI when using SE or CTCA as a first line test at any patient risk level. There was an increase in revascularisation in the CTCA arm but this did not translate into a difference in hard endpoints.