How do you differentiate CHF from other causes of dyspnea in the ED? What will be the most effective way to do this in the next year or two?

by
W. Frank Peacock, IV, MD, FACEP
The Cleveland Clinic
Emergency Department
Director of Clinical Operations Associate Professor,
Ohio State University

Heart failure (HF) is the end-result of a cascade initiated by sub-clinical myocardial injury. With a fall in left ventricular performance, activation of compensatory neurohormonal systems occurs (e.g., renin-angiostenosin system, sympathetic nervous system) and there are increases in vasoconstrictor hormones (endothelin, arginine vasopressin). These effects preserve systemic circulation, but at the cost of increased systemic vascular resistance. Many of these hormones demonstrate myocardial toxicity. The end-result is LV dysfunction and eventual cardiac remodeling. This may progress silently until the patient ultimately presents to the ED. We can measure these hormones to help in the diagnosis of HF, although only BNP can be done in the ED environment.

In the ED, HF diagnosis is difficult. The signs and symptoms of HF are non-specific and insensitive. Dyspnea is the most common presenting symptom, but it has a sensitivity and specificity in the 50-percent range. Orthopnea is marginally better, with a specificity of 88%, but no better sensitivity. No other finding has an adequate combination of both an sensitivity and specificity. The available ED tests offer little additional improvement in diagnostic accuracy for HF. Chest radiographs miss up to 20% of proven cardiomegaly1. And the detection of pleural effusion by CXR has a sensitivity and specificity of approximately 70%. When done portably, the performance is even worse2. Consequently, ED misdiagnosis occurs in as many as 12% of patients, equally divided between over and under diagnosis3.

Brain Natriuretic Peptide (BNP) measurement improves diagnostic accuracy. Natriuretic peptides (NP) are released as a result of volume stimulus. BNP functions as the counterpoise to neurohormonal activation by causing vasodilation, decreased aldosterone levels, inhibition of the renin-angiotensin-aldosterone system, and decreases in sympathetic nervous system activity.

BNP levels correlates well with the diagnosis of HF and functional status. In non-HF patients, BNP levels averaged 38 pg/mL, compared to HF patients with a mean level was 1,076 pg/ml4. And, when HF is divided by class, BNP levels vary directly with severity of class5.

In the ED, BNP is helpful in those patients most difficult to diagnose e.g. those with the combination of COPD and HF. COPD patients will have levels less than 100 pg/mL whereas, when dyspnea is from CHF, levels were >1000 pg/ml. Similar results are seen in patients with edema from HF and non-HF causes4. But, there are confounders to BNP. It is increased in the elderly, women, possibly those on hormone replacement therapy, and probably in pulmonary embolus. Finally, this test has a co-efficient of variation of about 10%. In the ranges that BNP occurs clinically, this is insignificant

The diagnosis of HF is suggested by the proper scenario and a BNP >100 pg/ml. At this level, BNP demonstrates both a sensitivity and specificity of 94%. The positive predicted value is 92%, and the negative predictive value is 96%, for an overall accuracy of 94%. Thus, this is an excellent test for ED physicians to determine the presence or absence of congestive heart failure2. Few guidelines for HF management include BNP6. The recent literature suggests it is most useful to exclude HF in those conditions where it would normally be diagnosed. In a patient whose clinical presentation would suggest HF, a BNP in the normal range should result in the strong consideration of an alternative diagnosis. Because non-HF conditions can result in an elevated BNP, the clinical context of a positive BNP must be considered. A positive BNP should prompt the performance of routine tests confirming the diagnosis, as well as to evaluate the cause and define the type of heart failure (e.g., ECG, chest x-ray, and echocardiogram).

Send our editor your comments on this question with your name, title and affiliation.

References

1. Kono T, Suwa M, Hanada H, Hirota Y, Kawamura K: Clinical Significance of Normal Cardiac Silhouette in Dilatd Cardiomyopathy - Evaluation Based Upon Echocardiography and Magnetic Resonance Imaging. Jap Circ 56(4): 359-65, 1992.

2. Ruskin JA, Gurney JW, Thorsen MK, Goodman LR: Detection of Pleural Effusions on Supine Chest Radiographs. AJR 148: 681-83, 1987

3. Dao, Q., Maisel, A. et al: J. American College of Cardiology, Vol 37, No. 2, 2001

4. Maisel A, et al: J Am Coll Cardiol 2001; 37(2): 379-85

5. Triage® BNP Test Package Insert

6 Remme WJ, Swedberg K. Guidelines for the Diagnosis and Treatment of Chronic Heart Failure. Eur Heart J, 22 (17); 1527-60, 2001).