Diagnosis

Signs and Symptoms. The major symptoms of ADHF, shortness of breath, congestion, and fatigue, are not specific for cardiac and circulatory failure.8 They may be caused by other conditions which mimic HF, complicating the identification of patients with this syndrome. Various forms of pulmonary disease, including pneumonia, reactive airway disease and pulmonary embolus, may be especially difficult to differentiate clinically from HF.9

Diagnostic Utility of Natriuretic Peptides. Two forms of natriuretic peptides, BNP and NT-proBNP, have been studied as aids to establish the diagnosis, estimate prognosis and monitor the response to therapy of patients with ADHF.10

Measurement of these peptides has been proposed in cases where the diagnosis of HF is uncertain. A large, multicenter investigation, The Breathing Not Properly Study provides important evidence supporting the clinical utility of plasma BNP in the assessment of patients presenting with possible HF.11,12 This study evaluated 1586 patients seen in the emergency department with the complaint of acute dyspnea who had prospective determination of BNP by bedside assay. Patients were assigned a probability of HF by physicians in the emergency department who were blinded to the results of the BNP assay. The final determination of whether or not HF was present was based on a review of the clinical data by 2 cardiologists also blinded to the BNP assay results. The sensitivity and specificity of BNP measurements for the diagnosis of HF were compared with the accuracy of an assessment based on standard clinical examination.

The diagnostic accuracy of BNP, using a cutoff value of 100 pg/mL, was 83% relative to the assessment made by the independent cardiologists, whereas the negative predictive value of BNP for HF when levels were <50 pg/mL was 96%. As expected, measurement of BNP appeared to be most useful in patients with an intermediate probability of HF. In these patients, a BNP cutoff value of 100 pg/mL resulted in the correct classification 74% of the time. BNP was found to be predictive of HF when left ventricular (LV) function was depressed or preserved.13 Although BNP levels were lower in patients with HF associated with preserved LVEF, the cutoff value of 100 pg/mL still had a sensitivity of 86% and a negative predictive value of 96%. BNP levels increase with age, more so in older women, so that cutoff of 100 pg/mL may not provide the same degree of specificity for the diagnosis of HF, especially in elderly women with dyspnea.14,15

The clinical utility of NT-proBNP in the diagnosis of HF was reported in the N-terminal Pro-BNP Investigation of Dyspnea in the Emergency Department (PRIDE) study. This study used NT-proBNP measurements in the emergency department to rule out acute HF in 600 patients who presented with dyspnea.16 NT-proBNP results were correlated with a clinical diagnosis of acute HF as determined by study physicians blinded to these measurements. The median NT-proBNP level among the 209 patients who had acute HF (35%) was 4054 versus 131 pg/mL among 390 patients who did not (65%, P<.001). NT-proBNP levels increase with age so that the study investigators recommend NT-proBNP cut points of >450 pg/mL for patients younger than 50 years of age and >900 pg/mL for patients age 50 years or older, both of which were highly sensitive and specific for HF in this study. For patients 75 years or older, 1800 pg/mL is the recommended cutpoint for NT-proBNP.17,18

Prognostic Role of Natriuretic Peptides. Although baseline BNP levels may correlate only modestly with pulmonary capillary wedge pressure (PCWP), changes in PCWP do correlate directly with changes in BNP concentration during hospitalization.19,20 The predischarge BNP after treatment for acute HF appears to predict patients at risk of early readmission or death following hospitalization for HF.21,22 Although specific discharge cutoff values are still being defined, patients whose BNP increases during hospitalization are at very high risk, as are patients with levels >700 pg/mL at discharge. Patients with levels <350 pg/mL at discharge appear to be at relatively low risk of readmission and death after discharge. Two recent studies have demonstrated that discharge BNP and change in BNP from admission to discharge provide independent predictive value for poor outcomes after an episode of ADHF.22,23

Triage Value of Natriuretic Peptides. The value of BNP determination in the triage of patients seen in the emergency department has been evaluated in a prospective, randomized, controlled, single-blind study in which 452 patients presenting with acute dyspnea were randomized to assessment with routine clinical evaluation or routine clinical evaluation plus the measurement of BNP. The diagnosis of HF was considered ruled out when BNP levels were <100 pg/mL, whereas levels of >500 pg/mL were considered diagnostic of ADHF.

Fewer patients were hospitalized or admitted to intensive care units in the BNP aided group compared with those evaluated by standard clinical evaluation alone. The median time to discharge was 8 days in the group with BNP measured versus 11 days in the control group (P = .001). Although the data on outcomes from this study are not definitive and the hospital lengths of stay are not reflective of practice patterns in the United States, making generalizability problematic, they do not suggest that triage using BNP resulted in the under-treatment of patients truly at risk. The readmission rate for HF was similar in the 2 study groups and the mortality rate, while not reduced statistically, was lower in those patients with BNP determined. Larger randomized trials of this strategy are needed to assess the impact of this approach on adverse outcomes associated with admission for ADHF.

Use of Natriuretic Peptides to Guide Therapy. A small number of studies have evaluated the use of BNP or NT-proBNP to guide HF therapy. In the initial study, Troughton et al24 randomized 69 patients with symptomatic HF and LVEF <40% to a clinically guided treatment group and a group for whom therapy was increased to drive the aminoterminal portion of BNP (N-BNP) level to <200 pg/mL. In the N-BNP guided group there were fewer total cardiovascular events (death, hospital admission, or HF decompensation than in the clinical group (19 vs 54, P=0.02). At 6 months, 27% of patients in the N-BNP group and 53% in the clinical group had experienced a first cardiovascular event (P=0.034). Changes in LV function, quality of life, renal function, and adverse events were similar in both groups. In the Systolic Heart Failure Treatment Supported by BNP (STARS-BNP) study25 220 patients with New York Heart Association (NYHA) Class II-III HF symptoms on evidence-based medical therapy with angiotensin converting enzyme (ACE) inhibitors and beta blockers were randomized to a clinical care group and a group for whom the goal was a BNP of <100 pg/mL. The primary endpoint of HF hospitalization or HF death was significantly lower in the BNP group (24% vs 52%, P, 0.001). All-cause hospital stays were not different in the two groups (60 in the control group vs 52 in the BNP group) while HF hospital stays were significantly different favoring the BNP group (48 in control group vs 22 in BNP group, P<0.0001) Thus there were 30 non-HF hospital stays in the BNP group vs only 12 in the control group raising the concern that targeting therapy to BNP might lead to hospitalizations for hypotension, renal insufficiency, or hyperkalemia although the specific reasons for non-HF hospitalizations were not mentioned.

The randomized controlled multicenter Trial of Intensified vs Standard Medical Therapy in Elderly Patients With Congestive Heart Failure (TIME-CHF) enrolled 499 patients aged 60 years or older with systolic HF (LVEF <=45%), NYHA class of II or greater, prior hospitalization for HF within 1 year, and N-terminal BNP level of 2 or more times the upper limit of normal.26 The primary endpoints were survival free of all cause hospitalizations and quality of life. There were similar rates of survival free of all-cause hospitalizations (41% vs 40%, respectively; hazard ratio [HR], 0.91 [95% CI, 0.72-1.14]; P=.39) in both groups over 18 months of follow-up. Quality-of life metrics improved but these improvements were similar in both the N-terminal BNP-guided and symptom-guided strategies. Survival free of hospitalization for HF, a secondary end point, was higher among those in the N-terminal BNP-guided group (72% vs 62%, respectively; HR, 0.68 [95% CI, 0.50-0.92]; P=.01).

The "Can Pro-Brain-Natriuretic Peptide Guided Therapy of Chronic Heart Failure Improve Heart Failure Morbidity and Mortality?" (PRIMA) study presented at the American College of Cardiology 2009 Scientific Sessions enrolled 345 HF patients who were hospitalized with elevated NT-proBNP levels (>=1700 pg/mL).27 After NT-proBNP levels dropped by more than 10% (to 850 pg/mL or less), patients were randomized to receive NT-proBNP-guided treatment (n=174) or clinically guided treatment (n=171). Serum levels of NT-proBNP were measured at discharge and again at the first follow-up period (two weeks post-discharge). The lesser of the two values was deemed the target value. If the NT-proBNP levels in patients in the guided-treatment group showed any increase at any subsequent follow-up, more intensive heart-failure therapy was immediately instituted. At a median follow-up of 702 days (range 488-730) there was a small but non-significant increase in the trial's primary end point--number of days alive outside the hospital--among patients in the NT-proBNP-guided group. Survival free of HF hospitalizations, a secondary endpoint, was significantly lower in the NT-proBNP group.

Based on all these results, it is not yet possible to recommend the use of natriuretic peptides to guide HF therapy, in either the outpatient or inpatient setting. Larger trials using HF hospitalization and mortality are being planned.

Limitations of Natriuretic Peptides. There are limitations concerning the utility of natriuretic peptides in the diagnosis of HF that need to be considered to gain maximum benefit from this testing.28 Some patients with obvious ADHF by clinical criteria may not have BNP levels typically considered to be diagnostic. In contrast, there may be patients, especially those with chronic LV systolic function, who have persistently elevated BNP levels despite clinical compensation and adequate volume status.29 Single measurements of BNP or NT-Pro BNP may not correlate well with measures of PCWP in patients in the intensive care unit, especially in patients with renal dysfunction.13 In addition, the biologic variability of the assays for BNP is high making interpretation of day-to-day measurements problematic.30

Interpretation of natriuretic hormone levels can be problematic in patients with pulmonary disease. BNP and NT-proBNP may be increased in patients with pulmonary embolus or cor pulmonale resulting from right HF in the absence of congestion.31 Some patients with HF without LV dysfunction may require treatment for peripheral edema despite having low BNP levels, indicating that BNP determination cannot always identify patients who need diuretic therapy. Patients with pulmonary disease may have concomitant LV dysfunction which may become more symptomatic during a primary respiratory illness, further complicating the interpretation of BNP levels.

The ranges of BNP for patients with and without a final diagnosis of HF overlap, which makes the test potentially less valuable in an individual patient with intermediate levels of BNP. Because many conditions can increase BNP levels, low values of BNP are most useful because they make the diagnosis of decompensated HF very unlikely as an explanation for dyspnea. Decision analysis indicates that BNP testing is generally most useful in patients who have an intermediate probability of HF. BNP levels rarely alter the diagnosis in patients who are very likely or unlikely to have HF based on usual clinical evaluation. ADHF remains a clinical phenomenon of symptoms due to circulatory dysfunction whose identification as yet cannot be reduced to a single laboratory measurement. Results of BNP testing must be interpreted in the context of the overall clinical evaluation, and such testing must augment rather than supersede careful clinical reasoning.32