Angurana et al. Rapid measurement of B- type natriuretic peptide in the diagnosis of congestive heart failure in patients presenting to the emergency department with acute shortness of breath. Int. J. Med. Med. Sci. 2011

An acutely ill patient presented to the emergency department needs to be diagnosed correctly and promptly because misdiagnosis could place such a patient at risk of increased morbidity and mortality (Wurez and Meador, 1992; Bales and Sorrentino, 1997). It is imperative to institute appropriate treatment early in the course of disease to derive maximum survival benefit (Packer and Cohn, 1999). The signs and symptoms of heart failure are often nonspecific (Wei et al., 1993). A helpful history is not readily obtainable in an acutely ill patient. Moreover dyspnea, a key symptom of heart failure may be a nonspecific finding in elderly or obese patients in whom respiratory disease and physical deconditioning are common (Nakagawa et al., 1995). Routine laboratory tests like electrocardiogram and X-ray chest are also not accurate enough to always make the appropriate diagnosis (Hunt et al., 2001; Davie et al., 1996). Thus, it becomes difficult for clinicians to differentiate patients with heart failure from other diseases, such as pulmonary disease, on the basis of routinely available laboratory tests.

Echocardiography, although currently the gold standard for diagnosing left ventricular dysfunction, is relatively costly and has limited availability in urgent care settings. Severely dyspneic patients may be unable to stay still long enough for an echocardiographic study, while in others imaging may be suboptimal due to co-morbid conditions like obesity or lung disease. Therefore, even if emergency echocardiographic assessment is available, a rapid estimation of BNP can prove to be a useful cost effective investigation for the diagnosis of heart failure, saving valuable time. The source of plasma BNP is cardiac ventricles and is indicative of ventricular dys-function. Its release appears to be directly proportional to the ventricular volume and pressure overload (Nakagawa et al., 1995; Dickstein, 1998; Maeda et al., 1998; Luchner et al., 1998).

BNP is an independent predictor of high LV end diastolic pressure and correlates well with grade of NYHA class (Maeda et al., 1998; Clerico et al., 1998; Ceyhan et al., 2008). NT-proBNP, a peptide similar to BNP, is also being used for diagnosing CHF, however it has been shown that its plasma levels are vulnerable in patients with renal insufficiency as compared to BNP. The investigators in the PRIDE study evaluated the relationship between renal function, NT-proBNP and CHF. They found an inverse relationship between renal function and NT-proBNP and also refuted the claimthat, it loses its specificity for the diagnosis of acute CHF due to impaired renal functions, although patients with serum creatinine of more than 2.5 mg/dl were excluded in their study. They also observed that NT-proBNP was the strongest risk factor of death regardless of renal function (Anwaruddin et al., 2006).

However, in our study we used BNP which has been shown to be more stable and predictable, irrespective of renal functions, in the diagnosis of CHF. In our study BNP assay was able to distinguish heart failure from pulmonary disease with a high degree of sensitivity, specificity, and accuracy. In our study, plasma BNP levels of heart failure patients were significantly higher as compared to those with no heart failure (399.6+289.2 pg/ml vs 84.9+42.2 pg/ml. p< 0.001). Heart failure patients were categorized into 3 subgroups. Subgroup analysis revealed that, BNP levels were equally useful in differentiating systolic heart failure, diastolic heart failure, and the miscellaneous category. In patients with predominantly systolic heart failure (n = 28) the BNP level was significantly higher than other groups with range of 196 to 1125, mean 652+345 pg/ml. In patients having diastolic heart failure (n = 14), the BNP levels were high but less than those with systolic heart failure (98 to 697, mean 250 + 154 pg/ml). The miscellaneous group (n = 2) which included patients with atrial fibrillation and fast ventricular rate presented with flash pulmonary edema. Their BNP levels were 187 and 196 pg/ml. Similar observations in various forms of heart failure have been reported in earlier studies (Yamamoto et al., 1996; Logeart et al., 2002; Hammerer et al., 2004).

In our study, the plasma level of BNP correlated well with the severity of LV dysfunction. Similar results have been reported by other investigators (Ogawa et al., 1998; Cowie et al., 1997). Univariate analysis of plasma BNP assay showed that, it is the most accurate variable at 175 pg/ml for the diagnosis of heart failure with accuracy of 87.5%, sensitivity 81.8%, specificity 96.4% and positive predictive value of 77.1%. The results of our study are consistent with other reported studies (Maisel et al., 2002; Morrison et al., 2002; Koulori et al., 2004).

Using multiple variable logistic regression analysis of various factors used for differentiating patients with and without heart failure with significant p value, we observed that addition of BNP at cut off level of 175 pg/ml increased the combined explanatory power of the symptoms, signs, radiological study and other laboratory findings. Our data showed that, plasma BNP level of 175 pg/ml was the strongest independent predictor of the heart failure diagnosis with odds ratio of 41.9, followed by history of paroxysmal nocturnal dyspnea (odds ratio of 37.3), decreased ejection fraction (odds ratio 29.0), cardiomegaly on chest x-ray and pulmonary edema (odds ratio of 27.0). These findings of multiple variable analysis are almost consistent with other reported studies with some variations which can be explained on the basis of differences in population studied and sample size (Logeart et al., 2002; Hammerer et al., 2004; Maisel et al., 2002; Morrison et al., 2002 ).