The impact of B-type natriuretic peptide in addition to troponin I, creatine kinase-MB, and myoglobin on the risk stratification of emergency department chest pain patients with potential acute coronary syndrome

Study objective: The emergency department (ED) evaluation of chest pain patients with potential acute coronary syndrome is limited by the initial sensitivity of cell injury markers. BNP is increased during myocardial ischemia and is associated with adverse outcomes. We determine whether the addition of B-type natriuretic peptide (BNP) to troponin I, creatine kinase-MB (CK-MB), and myoglobin increases the sensitivity and negative predictive value (NPV) for acute myocardial infarction, acute coronary syndrome, and 30-day adverse events among chest pain patients with potential acute coronary syndrome. Methods: A convenience sample of patients aged 30 years or older and presenting to an urban academic ED with nontraumatic chest pain, thus prompting an ECG, was enrolled, and consent was obtained. Blood samples were drawn at 0 and 90 minutes. Thirty-day follow-up was performed for all enrolled patients. Main outcomes were acute myocardial infarction, acute coronary syndrome, and 30-day events (death, acute myocardial infarction, or revascularization). BNP cutoffs were derived from receiver operator characteristics curves. The sensitivity, specificity, positive predictive value (PPV), and NPV with 95% confidence intervals (CIs) were calculated with and without BNP. Differences in sensitivity and specificity with the addition of BNP were calculated with 95% CIs, and McNemar's test was performed to compare sensitivities and specificities. Results: Four hundred twenty-six patients were enrolled and analyzed. The cohort was 54.7 +/- 13.9 years old, 47.7% men, and 63.5% black. The outcomes were acute myocardial infarction, 39 (9.2%), acute coronary syndrome, 101 (23.7%), and 30-day adverse cardiovascular events 52 (12.2%). BNP cutoffs derived were 51, 31, and 31 pg/mL for acute myocardial infarction, acute coronary syndrome, and 30-day events, respectively. The addition of BNP showed increased sensitivity at the cost of decreased specificity for all 3 outcomes, as follows: (1) acute myocardial infarction: sensitivity: 87.2% (95% CI 72.6% to 95.7%) to 97.4% (95% CI 86.5% to 100%), difference 10.3% (95% CI-0.2% to 24.6%), P = .125; specificity: 62.3% (95% CI 57.2% to 67.1%) to 47.8%.(95% CI 42.7% to 52.9%), difference 14.5% (95% CI 11.1% to %18.4), P < .0001; (2) acute coronary syndrome: sensitivity: 75.2% (95% CI 65.7% to 83.3%) to 88.1% (95% CI 80.2% to 93.7%), difference 12.9% (95% CI 7.0% to 21.0%), P = .0002; specificity: 68.0% (95% CI 62.6% to 73.0%) to 48.6% (95% CI 43.1% to 54.2%), difference 19.4% (95% CI 15.2% to 24.1%), P < .0001; (3) 30-day events: sensitivity: 71.2% (95% CI 56.9% to 82.9%) to 88.5% (95% CI 76.6% to 95.7%), difference 17.3% (95% CI 7.7% to 30.3%), P = .004; specificity: 61.8% (95% CI 56.6% to 66.7%) to 43.9% (95% CI 38.8% to 49.0%), difference 17.9% (95% CI 14.2% to 22.2%), P < .0001. There were trends toward increased NPV and decreased PPV for all outcomes, and the addition of BNP achieved a NPV of 99.5% (95% CI 97.0% to 100%) compared with 98.0% (95% CI 95.3% to 99.3%) for acute myocardial infarction. Conclusion: The addition of BNP as a dichotomous test to troponin I, CK-MB, and myoglobin produces increased sensitivity at a cost of decreased specificity for acute myocardial infarction, acute coronary syndrome, and 30-day adverse events. Because of this tradeoff, BNP cannot be recommended for use among all ED chest pain patients. However, the improved sensitivity may make this test useful in selected cohorts when the decreased specificity is less important.