Introduction Although stress echocardiography (SE) is invaluable in younger populations, its prognostic value may be attenuated in the elderly due to shorter life expectancy and the frequent presence of severe comorbidities. This study sought to evaluate the clinical effectiveness of SE in octogenarians, particularly its prognostic value over clinical variables, in predicting hard events.

Methods A total of 374 consecutive octogenarians who underwent SE for evaluation of coronary artery disease (CAD) were assessed for feasibility, diagnostic accuracy and safety of the test, and followed up for hard outcomes (all-cause mortality, cardiovascular (CV) deaths and non-fatal myocardial infarction (NFMI)). Cox regression analysis was performed to identify predictors of outcome.

Results Of the 374 tests, 360 (96.3%) were diagnostic. Of the 50 patients with inducible ischaemia, 33 patients (66%) proceeded to angiography of which 27 (82%) patients had significant CAD. During long-term follow-up of 4.0±2.0 years, there were 127 deaths and 36 NFMIs. The annualised mortality, NFMI and combined mortality /NFMI rates were 8.1%, 1.8% and 9.4% for patients with a normal SE and 12.1%, 5.5% and 14.1% for those with an abnormal SE, respectively. Predictors of NFMI on multivariate analysis were prior CAD (HR 2.89, CI 1.03 to 8.15, p=0.045), peripheral vascular disease (HR 3.33, CI 1.18 to 9.45, p=0.02), and inducible ischaemia (HR 3.97, CI 1.49 to 10.55, p=0.006). In patients without prior history of CAD, inducible ischaemia was the only independent predictor of NFMI (HR 8.72, CI 1.46 to 52.2, p=0.018). The larger the extent of ischaemia, the greater the incidence of NFMI. The independent predictors of CV events (NFMI or CV mortality) were PAD (HR 2.81, CI 1.21 to 6.52, p=0.016) and peak wall motion score index (HR 5.71, CI 1.67 to 19.6, p=0.006). Although inducible ischaemia predicted all-cause mortality on unadjusted analysis, it did not on multivariate analysis.

Conclusions In octogenarians, SE demonstrated excellent feasibility, safety and diagnostic accuracy. SE parameters were independent predictors of NFMI and CV events, and the presence of inducible ischaemia was associated with a 50% increase in all-cause mortality.