Cardiovascular mortality is high in African Americans, and those with normal results on stress echocardiography remain at increased risk. The aim of this study was to develop a risk scoring system to improve the prediction of cardiovascular events in African Americans with normal results on stress echocardiography. Clinical data and rest echocardiographic measurements were obtained in 548 consecutive African Americans with normal results on rest and stress echocardiography and ejection fractions ≥50%. Patients were followed for myocardial infarction and death for 3 years. Predictors of cardiovascular events were determined with Cox regression, and hazard ratios were used to determine the number of points in the risk score attributed to each independent predictor. During follow-up of 3 years, 47 patients (8.6%) had events. Five variables—age (≥45 years in men, ≥55 years in women), history of coronary disease, history of smoking, left ventricular hypertrophy, and exercise intolerance (<7 METs in men, <5 METs in women, or need for dobutamine stress)—were independent predictors of events. A risk score was derived for each patient (ranging from 0 to 8 risk points). The area under the curve for the risk score was 0.82 with the optimum cut-off risk score of 6. Among patients with risk scores ≥6, 30% had events, compared with 3% with risk score <6 (p <0.001). In conclusion, African Americans with normal results on stress echocardiography remain at significant risk for cardiovascular events. A risk score can be derived from clinical and echocardiographic variables, which can accurately distinguish high- and low-risk patients.
Cardiovascular stress testing is recommended for the diagnosis and risk stratification of patients with suspected or known coronary artery disease (CAD). After normal results on exercise echocardiography, the yearly event rate is <1%, which is similar to the expected event rate in the age-adjusted general population. Similarly, patients with normal results on dobutamine stress echocardiography have low cardiovascular risk, although the yearly event rate is more modest at 1.3%, because of the higher risk nature of patients referred for dobutamine stress studies. However, published reports suggest a higher event rate in African Americans. After normal results on dobutamine stress echocardiography, 9% of African Americans had myocardial infarctions (MIs) or died at 3 years compared with 5% of European Americans. Among patients with low-risk findings on single photon-emission computed topography, African Americans had a 3% annualized cardiovascular death rate compared with <1% for European Americans. Given the wide disparity in event rates between European American and African American patients, we sought to identify characteristics associated with worse outcomes in African Americans after normal results on stress echocardiography and to create a clinical model to aid in further risk-stratifying this population.
Methods
The Indiana University Institutional Review Board approved this study. From January 2003 to November 2005, a consecutive series of 548 patients who underwent stress echocardiography at an urban public hospital who self-identified as African American or black and had normal results on stress echocardiography were included in this study. Clinical characteristics and results of stress echocardiography were recorded with data taken from the electronic medical record, with all baseline characteristics referring to variables known at the time of stress echocardiography.
Patients were considered to have smoking histories if they were current or former smokers, and hyperlipidemia was defined as total cholesterol >240 mg/dl or the use of a cholesterol-lowering agent. Hypertension was defined as systolic blood pressure >140 mm Hg and/or diastolic blood pressure >90 mm Hg. Family history of CAD was considered if there was ≥1 first-degree relative with the onset of coronary disease before 55 years of age in men and 65 years of age in women. Risk factors for CAD were considered to be hypertension, current tobacco use, family history of CAD, age ≥45 years for men and ≥55 years for women, and high-density lipoprotein <40 mg/dl. Chronic kidney disease was defined as a glomerular filtration rate <60 ml/min. Obesity was defined as a body mass index ≥30 kg/m 2 . Patients were considered to have histories of CAD if they had previous MIs, revascularization, or coronary angiography demonstrating ≥1 vessels with ≥50% diameter stenosis.
Exercise echocardiography was performed using protocols individualized on the basis of the patient’s age and exercise capacity using standard end points. Images were obtained in the apical 4- and 2-chamber views and the parasternal long- and short-axis views at baseline and immediately after stress. The dobutamine protocol was conducted with a stepwise infusion using previously described methods and end points. Images were obtained in the same views at rest, low dose, and peak dose. Stress echocardiograms were interpreted by experienced echocardiographers blinded to the clinical data and the results of follow-up. Measurements were initially performed by experienced sonographers and revised as needed by the interpreting echocardiographer. Normal results on stress echocardiography were defined by normal wall motion at rest and with stress and an ejection fraction ≥50%. Stress electrocardiography was considered to indicate ischemia in the presence of ≥1 mm of horizontal or downsloping ST-segment depression but was not used in defining normal or abnormal stress echocardiographic results. Exercise intolerance was defined as either the clinical need for a dobutamine stress study or the inability to achieve 5 METs of exercise in women or 7 METs of exercise in men.
Left ventricular (LV) dimensions were measured using 2-dimensional echocardiography in the parasternal long- or short-axis view at the level of the mitral leaflet tips. LV mass was calculated using linear dimensions. LV hypertrophy was defined by LV mass ≥96 g/m 2 in women and ≥116 g/m 2 in men when indexed to body surface area. Relative wall thickness was calculated as 2 × posterior wall thickness/internal LV diameter. Concentric remodeling was defined as normal LV mass with relative wall thickness ≥0.43. Left atrial systolic diameter was measured as the maximum anteroposterior diameter in the parasternal long-axis view and indexed to body surface area, with a value ≥2.4 cm/m 2 considered abnormal.
Follow-up was obtained primarily by retrospective review of the Regenstrief Medical Record System, which is an electronic medical record system coordinating inpatient and outpatient records of 18 Indianapolis-area hospitals. Patients were followed for the combined end point of MI or death for 3 years from the time of stress echocardiography. This follow-up period was chosen because the yearly event rate after normal results on stress echocardiography increases after a 3-year period. Follow-up could not be obtained in 5 patients (0.9%). All statistics were performed on an intention-to-treat principle. The end point of death was verified through the electronic medical record and the Social Security Death Index database. Patients were considered to have reached the end point of MI if the clinical impression at the time of the event was ischemia due to a primary coronary event on the basis of typical chest pain, electrocardiographic changes, cardiac enzyme elevations, and/or cardiac catheterization data.
All statistical analyses were performed assuming a 2-sided p value <0.05 as significant. Cut points for creating categorical variables were chosen using previously defined values. The Cox proportional-hazard model was used to identify univariate predictors of MI and death. A multivariate analysis (forward selection method) was conducted using all variables significant on univariate analysis. Each independent predictor identified by multivariate analysis was assigned risk points by calculating its hazard ratio relative to the predictor with the lowest hazard ratio and then rounding to the nearest integer, allowing ease of use in a clinical setting. Risk points were subsequently totaled to establish a total risk score. A receiver-operating characteristic curve was created by plotting 1 minus specificity against sensitivity of the risk score for MI or death. The optimal cut point was determined as the value that maximized sensitivity and specificity for predicting future events. The true error rate of our statistical model was calculated by 10-fold cross validation by separating the cohort into 10 equally sized and randomly determined groups. Each subgroup was used as a validation set, with the other 9 groups used as the training set using the previously described statistical method. The true error rate was then determined by averaging the error rate of the 10 validation sets. Annualized event rates were calculated by dividing events by the total number of person-years in follow-up. Kaplan-Meier curves were used to compare event-free survival. All statistical analysis was performed using SPSS version 17 (SPSS, Inc., Chicago, Illinois).
Results
Table 1 lists the clinical and stress echocardiographic characteristics of the patient population. The combined end point of MI or death occurred in 3.6% and 8.6% at 1 and 3 years, respectively. Table 2 lists the clinical univariate predictors and Table 3 the echocardiographic univariate predictors. On multivariate analysis, there were 5 predictive variables ( Table 4 ): age (≥45 years in men, ≥55 years in women), smoking history, CAD history, LV hypertrophy, and exercise intolerance (dobutamine stress study or low exercise capacity on treadmill testing). After calculating each predictor’s relative hazard ratio on multivariate analysis, age, smoking history, and exercise intolerance were each assigned 2 risk points, and CAD and LV hypertrophy were each assigned 1 risk point. A risk score was calculated for each patient by totaling the number of risk points (range 0 to 8), using the following formula: 2 (if age ≥45 years in men or ≥55 years in women) + 2 (if history of smoking) + 2 (if dobutamine stress protocol or <7 METs in men or <5 METs in women) + 1 (if history of CAD) + 1 (if LV hypertrophy on stress echocardiography).
| Clinical Variable | Value | Stress or Echocardiographic Variable | Value |
|---|---|---|---|
| Age (yrs) | 56 ± 12 | Stress electrocardiography | |
| Men | 31% | Normal | 57% |
| Hypertension | 83% | Ischemia | 10% |
| Family history of CAD | 34% | Nondiagnostic | 33% |
| Hyperlipidemia | 41% | <85% of predicted heart rate | 25% |
| Smoking history | 57% | Dobutamine stress study | 32% |
| ≥3 cardiac risk factors | 61% | Peak METs ∗ | 7.6 ± 5.8 |
| Diabetes mellitus | 32% | Exercise intolerance † | 49% |
| Obesity | 45% | LV internal diameter in diastole (cm) | 4.3 ± 0.5 |
| Chronic kidney disease | 14% | LV internal diameter in systole (cm) | 2.9 ± 0.5 |
| CAD | 12% | Fractional shortening | 0.34 ± 0.15 |
| MI | 6% | LV posterior wall thickness (cm) | 1.06 ± 0.20 |
| Heart failure | 5% | Interventricular septal thickness (cm) | 1.10 ± 0.22 |
| Medication use at time of study | LV mass index (g/m 2 ) | 83 ± 25 | |
| Aspirin | 45% | Relative wall thickness | 0.51 ± 0.01 |
| Statins | 33% | LV remodeling pattern | |
| Hypertension therapy | 77% | Normal | 24% |
| Angiotensin-converting enzyme inhibitors | 48% | Concentric remodeling | 57% |
| β blocker | 44% | LV hypertrophy | 19% |
| Indication | Left atrial diameter index (cm/m 2 ) | 1.9 ± 0.3 | |
| Atypical chest pain | 55% | Left atrial enlargement | 6% |
| Typical chest pain | 16% | Ejection fraction (%) | 61 ± 6 |
| Dyspnea | 10% | ||
| Preoperative | 11% | ||
| Other | 7% |
∗ Includes only patients who underwent exercise stress testing.
† Exercise intolerance refers to either a dobutamine stress study or low workload on exercise testing (<7 METs in men, <5 METs in women).
| Characteristic | Hazard Ratio (95% Confidence Interval) | p Value |
|---|---|---|
| Age ∗ | 4.82 (2.05–11.34) | <0.001 |
| Men | 1.73 (0.97–3.09) | 0.063 |
| Hypertension | 1.93 (0.77–4.89) | 0.163 |
| Family history CAD | 1.08 (0.59–1.96) | 0.803 |
| Hyperlipidemia | 1.20 (0.61–2.35) | 0.605 |
| Smoker | 5.32 (2.26–12.53) | <0.001 |
| ≥3 cardiac risk factors | 3.63 (1.63–8.10) | 0.002 |
| Diabetes mellitus | 1.51 (0.84–2.70) | 0.169 |
| Obesity | 0.35 (0.18–0.69) | 0.002 |
| Chronic kidney disease | 2.32 (1.20–4.46) | 0.012 |
| CAD | 2.93 (1.55–5.55) | 0.001 |
| MI | 3.12 (1.46–6.68) | 0.003 |
| Heart failure | 1.38 (0.43–4.45) | 0.589 |
| Aspirin | 1.83 (1.02–3.28) | 0.042 |
| Statin | 1.12 (0.62–2.03) | 0.715 |
| Hypertension therapy | 1.74 (0.78–3.88) | 0.177 |
| Angiotensin-converting enzyme inhibitor | 1.10 (0.62–1.95) | 0.756 |
| β blocker | 1.95 (1.10–4.45) | 0.023 |
| Indication † | 2.72 (1.51–4.90) | 0.001 |
∗ Age ≥45 years in men and ≥55 years in women.
† Test indication of typical angina, dyspnea, or preoperative evaluation.
| Characteristic | Hazard Ratio (95% Confidence Interval) | p Value |
|---|---|---|
| Stress electrocardiographic ischemia | 1.72 (0.73–4.06) | 0.213 |
| <85% of predicted heart rate | 1.90 (1.06–3.42) | 0.033 |
| Exercise intolerance ∗ | 4.69 (2.27–9.69) | <0.001 |
| LV internal diameter in diastole | 1.16 (0.67–2.00) | 0.591 |
| LV internal diameter in systole | 1.28 (0.69–2.37) | 0.438 |
| Fractional shortening | 0.44 (0.01–19.57) | 0.673 |
| LV posterior wall thickness | 5.05 (1.91–13.38) | 0.001 |
| Interventricular septum thickness | 3.83 (1.44–10.20) | 0.007 |
| Concentric remodeling | 0.673 (0.38–1.20) | 0.177 |
| LV hypertrophy | 2.71 (1.51–4.88) | 0.001 |
| Left atrial enlargement | 3.33 (1.55–7.12) | 0.002 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
