Exercise echocardiography (EE) is the most physiological imaging technique for the detection of ischemia and offers advantages such as its low cost and absence of radiation. Although its limited diagnostic value has been questioned compared with anatomical techniques,1,2 its high prognostic value has been demonstrated in various clinical settings.3–9 Several advances have facilitated its implementation (second harmonic, continuous acquisition, simultaneous comparison of baseline with exercise, or the use of contrast). Indications and training for its use have been established by expert consensus.10 The aim of this study was to identify the predictors of events in patients with EE without inducible ischemia, paying special attention to cardiovascular risk (CVR) and its possible role in modifying the risk of cardiac events.

An observational retrospective cohort study was conducted with 2307 consecutive patients with EE using a Bruce or Naughton treadmill with 12-lead electrocardiogram monitoring. Image acquisition was performed in 4-, 3-, and 2-chamber views and short-axis orientations or in 5-chamber view (depending on the quality of the parasternal window or the use of contrast) at baseline and immediately after exercise or peak exercise with continuous acquisition, side-by-side comparison, and digital storage. Contractile abnormalities were assessed in a 16-segment left ventricular model. Contrast was used in 329 patients (20.1%). Inclusion criteria were age > 18 years and EE without induced ischemia (ie, without new segmental contractile abnormalities with exercise compared with the patient at rest, regardless of baseline abnormality in patients with heart disease). EE testing without inducible ischemia was based on segmental contractility and not on clinical or electrocardiographic abnormalities. Positive, doubtful, or inconclusive echocardiographic studies (noninterpretable due to poor window or defective acquisition) were excluded. There were no fatal complications. Demographic data, clinical variables, and data derived from EE were collected.

Known coronary artery disease (CAD) was defined as a history of myocardial infarction, previous percutaneous or surgical revascularization, or severe coronary lesion on previous angiography. In patients without CAD and without diabetes mellitus (DM), CVR was estimated according to the European SCORE (Systematic Coronary Risk Evaluation) scale calibrated for the Spanish population for a risk of cardiovascular death at 10 years by age, sex, systolic blood pressure, total cholesterol, and smoking.11 Patients without CAD were classified into 3 groups: SCORE < 5% (low-medium risk), 5% to 9% (high risk) and ≥ 10% or DM (very high risk). Follow-up was conducted by electronic and paper history, telephone call, and death register.

The primary outcome variable was the composite of fatal event (sudden death of unexplained origin and cardiac death due to heart failure or acute coronary syndrome [ACS]), nonfatal ACS (due to ACS with or without ST-segment elevation), and late revascularization. Acute coronary syndrome was defined as symptoms or ECG abnormalities compatible with ischemia with increased cardiac markers (troponin T). Late revascularization refers to that performed during follow-up, rather than to early revascularization with cardiac catheterization, despite normal EE. “Short-term prognosis” was arbitrarily defined as 1 year after EE and “long-term prognosis” as 3 years after EE.

A total of 2307 EE studies were conducted between January 2007 and December 2012, leading to the exclusion of 348 patients with a positive EE and 171 with doubtful or inconclusive EE. Of the remaining, we included 1788 patients without ischemia, only 1640 (434 with CAD and 1206 without CAD for whom data were available for European SCORE calculation). Mean age was 63.5 ± 10.7 years and 52.8% were men. Of the 1206 patients without CAD, there were 783 with SCORE < 5% (65%), 173 with SCORE 5% to 9% (14.3%), and 250 with SCORE ≥ 10%, or 250 with DM (20.7%) (Figure 1). Of the 434 CAD patients, 242 had a previous history of myocardial infarction and 356 of percutaneous or surgical revascularization. Demographic and clinical characteristics and pharmacological treatment are shown in Table 1. The EE data are shown in Table 2. Of the patients, 21.6% had contractile abnormalities at baseline. Obviously, the segmental contractility index did not increase during EE without ischemia. Mean left ventricular ejection fraction (LVEF) was 61.3% ± 10.4%. Of the 1640 study patients, 84 (5.12%) had an LVEF < 50% and less than 1% had an LVEF < 35%.

Figure 1.

Flowchart. Characteristics of the study. DM, diabetes mellitus; EE, exercise echocardiography; SCORE, Systematic COronary Risk Evaluation.

(0.25MB).

Prognosis Over Follow-up and Cardiovascular Risk

In patients with EE without ischemia after a median follow-up of 35 [23-54] months, event-free survival (EFS) estimated by Kaplan-Meier curves was 93.1% of the total group at the end of follow-up. Based on CVR, EFS was 95.4%, 98.7%, and 89.8% of patients with SCORE < 5%, 5% to 9%, or ≥ 10% or DM, respectively, with no significant differences at the end of follow-up between the patients with known CAD and those with SCORE ≥ 10 or DM (87.1% vs 89.8%; P = .193). In the 2 groups with higher CVR, cardiac EFS was significantly lower (P < .001) than in the 2 groups with SCORE < 10% (Figure 2).

Figure 2.

Kaplan-Meier curves as a function of cardiovascular risk in patients with stress echocardiography without ischemia. DM, diabetes mellitus; SCORE, Systematic COronary Risk Evaluation.

(0.11MB).

Event-free survival was assessed at different times during follow-up to determine if there had been an increase in risk after EE based on CVR. In patients without CAD, EFS was 99.2%, 98.4%, and 97.6% at 1, 2, and 3 years, respectively. Based on short-term CVR (1 year of follow-up), cardiac EFS was high in all groups (99.4% if SCORE < 5%, 100% if SCORE 5% to 9%, 98% if SCORE ≥ 10% or DM, and 97% with known CAD). At 2 years, cardiac EFS was 98.8%, 100%, 95.9%, and 93.5%, respectively. Based on long-term CVR (3 years of follow-up), cardiac EFS was similar in the 2 low-CVR groups (98.4% if SCORE < 5% and 98.7% if SCORE 5% to 9%; P = .40). However, in the group with SCORE ≥ 10% or DM, cardiac EFS at 3 ≥ years had significantly decreased compared with the group with SCORE < 5% (94.5%, P < .001) and was similar to that of known CAD patients (91.1%; P = .24) (Figure 3).

Figure 3.

Bar chart of event-free survival at 1, 2, and 3 years according to cardiovascular risk. DM, diabetes mellitus; SCORE, Systematic COronary Risk Evaluation.

(0.13MB).

The results show that patients without CAD with EE without ischemia have excellent prognosis. McCully et al.12 estimated that overall survival at 1 year and 3 years was 99.5% and 98.6%, respectively, in 1325 patients with normal EE without CAD and ventricular dysfunction. Prognostic data were very similar (99.2% and 97.6% at 1 year and 3 years, respectively) in the 1206 patients without CAD. Therefore, the favorable prognostic value of EE without ischemia reported in our setting is similar to that of hospitals with extensive experience. Our results are comparable to those reported in a meta-analysis conducted by Metz et al.,13 which included 3021 patients with negative EE and whose cardiac death or myocardial infarction rate was 1.56% at a mean follow-up of 33 months. In the present study, the composite rate of cardiac or sudden death and nonfatal ACS was higher (2.3% after a median follow-up of 35 months). This disparity may be related to the follow-up time and the differential classification of acute myocardial infarction after troponins were included in the study.

The finding of predictors of cardiac events in patients with EE without ischemia is of special interest because such patients are frequently discharged from cardiology clinics. In other studies, most of the predictors of events in patients with negative EE were obtained from the test itself. Thus, a normal but submaximal EE (HR < 85%)14,15 or decreased functional capacity (< 7 MET in men or < 5 MET in women) entails a higher rate of events.8,16 In the univariable analysis, an achieved maximum heart rate < 85% was also a predictor of cardiac events. In line with previous studies,17 electrocardiographic abnormalities during the test were not predictive of events. In the multivariable analysis, chest pain during EE was the only predictor of cardiac events derived from the test. Other predictors in the multivariate analysis were clinical factors (high CVR and kidney function) and baseline LVEF, indicating the importance of patient characteristics in prognosis in EE without induced ischemia.

The influence of the risk profile of CAD on the risk of cardiac events in patients with negative EE was barely addressed. Some studies have shown that patients with negative EE have a cardiac event rate < 1% per year regardless of pretest probability.12,18 The pretest probability of coronary heart disease was not used in this study, because it is estimated using clinical scores that include the type of pain (typical angina, atypical angina, or nonanginal pain). The retrospective nature of the study made it impossible to reliably assess the characteristics of pain for its appropriate classification. Furthermore, only 55% of the patients had been seen for chest pain. Therefore, we chose CVR, which was estimated using CVR factors to determine its influence on cardiac events. To our knowledge, this is the first study to establish the risk of events and their timing in patients with negative EE based on CVR. Despite the excellent prognosis initially indicated by a negative EE in all groups, the data obtained suggest that the number of long-term events is modulated by CVR. Thus, the decrease in long-term EFS was slightly modified if SCORE was < 10%, and was maintained if > 98% at 3 years, but was most striking in the group with SCORE ≥ 10 or DM or patients with CAD (< 95%). These results have clear implications for patient treatment. First, regardless of CVR, the favorable prognostic effect of negative EE is maintained for 1 year then decreases over time in patients with CAD (91.1% at 3 years) or SCORE ≥ 10% or DM (94.5% at 3 years). Second, although we cannot recommend the use of additional diagnostic methods or anti-ischemic treatment, primary prevention measures should definitely be applied in patients with SCORE ≥ 10% or DM and secondary prevention should be applied in patients with coronary disease despite the absence of induced ischemia.

Initial prognosis after an EE without ischemia is favorable but is subsequently modulated by cardiovascular risk.

This study was possible due to the consent of the Ethics Committee of the Hospital of Galdakao and a grant from the Department of Health of the Basque Regional Government: No. 2012111009.

None declared.