NT-ProBNP Proposed for LVD Screening in Patients at High Risk for Heart Failure

News Author: Steve Stiles
CME Author: Penny Murata, MD

CME Released: 06/25/2009; Valid for credit through 06/25/2010

June 25, 2009 — Plasma levels of N-terminal fragment of B-type natriuretic peptide (NT-proBNP) were significantly increased in association with asymptomatic left ventricular (LV) systolic or diastolic dysfunction in a group of about a 1000 patients with type 2 diabetes or hypertension seen in a primary-care setting, in a report in the June 2009 issue of the Journal of Cardiac Failure [1]. In such asymptomatic patients at increased risk for heart failure, the report suggests, NT-proBNP levels could potentially provide an easy and economical rule-out of LV dysfunction.

An NT-proBNP reading below the threshold of 125 pg/mL had a sensitivity of 98% and negative predictive value (NPV) of 99.8% in identifying echocardiography-defined moderate to severe diastolic dysfunction, report the authors, led by Dr Irene Betti (Ospedale S Maria Annunziata, Florence, Italy). It had about the same sensitivity and NPV for either systolic dysfunction (LVEF <50%) or moderate to severe diastolic dysfunction.

Even though ECG and echocardiography are more precise in identifying LV dysfunction in the absence of symptoms, the group notes, "a NT-proBNP concentration below the cutoff threshold that has been identified in the present study could provide a simplified diagnostic approach with an excellent NPV. Conversely, higher values could force clinical decision making toward a more appropriate and tailored diagnostic and possibly therapeutic strategy." Potentially, they speculate, use of NT-proBNP to rule out LV dysfunction in populations at risk for heart failure, such as diabetics and hypertensives, would be more cost-effective than echocardiography.

Asked to comment on the study, Dr James de Lemos (University of Texas Southwestern, Dallas) said there's no disagreement on whether the identification of asymptomatic systolic dysfunction is worth identifying--there are treatments for that, especially ACE [angiotensin converting enzymes] inhibitors. "But asymptomatic diastolic dysfunction is a bit of a trickier issue," he told heartwire . There aren't any proven therapies for it. And in the current analysis, the patients overwhelmingly had diastolic dysfunction.

A better clinical question to ask, he said, is, "Can we pick out the subgroup that has early LV systolic dysfunction? And they really didn't test that at all, unfortunately."

Echocardiography was performed and NT-proBNP levels obtained from 1012 persons with hypertension, diabetes, or both but no apparent heart failure; 32.4% had mild and 4% had moderate to severe LV diastolic dysfunction, and 1.1% had LV systolic dysfunction. Both forms of LV dysfunction were present in 32.6%. The analysis had excluded persons with a history of heart failure, ACS [acute coronary syndrome] moderate to severe valvular disease, atrial fibrillation, and other heart and lung conditions with influence on LV performance.

In the combined subgroups with either or both moderate to severe diastolic dysfunction and systolic dysfunction, 5.1% of the entire population, 125 pg/mL was identified as the NT-proBNP cut point with the best balance of predictive indicators in an analysis of receiver operating characteristics, overall, and when stratified by age and sex.

Sensitivity, Specificity, and Predictive Values of NT-ProBNP Cut Point of 125 pg/mL for Identifying Systolic or Moderate to Severe Diastolic LV Dysfunction, by Age and Sex

Subgroup	Sensitivity (%)	Specificity (%)	Negative predictive value (%)	Positive predictive value (%)
Male <67 y	87.5	92.7	99.5	33.3
Male >67 y	100	77.1	100	32.5
Female <67 y	100	84.1	100	33.3
Female >67 y	100	59.9	100	23

Levels of NT-proBNP also were increased in persons without LV diastolic or systolic dysfunction but who showed left ventricular or left atrial dilation, according to the authors, "thus reinforcing the importance of left chamber measurements in the echocardiographic screening of asymptomatic LV dysfunction."

"It's a little too good to be true," according to de Lemos. "The data for using NT-proBNP, from our group and from other groups, suggest it has modest value, not the value that was seen here. So this study is an outlier."

The issue explored in the study "has merit," he said. "But my guess is that we're going to need additional biomarkers besides NT-proBNP for it to perform well enough to be an actual routine screening tool."

Reference

1. Betti I, Castelli G, Barchielli A, et al. The role of N-terminal PRO-brain natriuretic peptide and echocardiography for screening asymptomatic left ventricular dysfunction in a population at high risk for heart failure. The PROBE-HF Study. J Cardiac Fail 2009; 15:377-384. Abstract