Measuring NT-proBNP in Hospitalized Heart Failure Patients
Change in Values Could Guide Treatment
By Genna Rollins
B-type natriuretic peptide (BNP) and N-terminal pro BNP (NT-proBNP) have been studied extensively in the diagnosis, management, and treatment of heart failure (HF), with a variety of cutoffs suggested as being predictive of worsening outcomes. Substantial biological variation also has been reported in both biomarkers, and prior research has suggested that serial monitoring of NT-proBNP is clinically significant in HF management. Now a study has prospectively evaluated changes in NT-proBNP during hospitalization for HF, findings reported on in this issue of Strategies.
BNP, NT-proBNP and other precursor amino acid peptides that make up the natriuretic peptide system are released by the cardiac muscle in response to various stimuli, such as increased cardiac wall tension from pressure and volume overload in the heart. The system has a strong diuretic effect, promotes vasodilation, and facilitates cardiovascular remodeling and response to ischemia. For these reasons, BNP and NT-proBNP have been studied extensively in HF. However, evidence has yet to consistently identify a specific cutoff useful in managing the condition, with studies suggesting changes of <30% to <60% as being clinically significant. Other research indicates that there is a strong biological variation in BNP and NT-proBNP, and that a weekly change of 47% in NT-proBNP levels would be important in managing HF. These findings led Johns Hopkins University researchers to conduct a prospective study that evaluated whether a <50% decrease in NT-proBNP levels among patients hospitalized for HF would be associated with increased risk of readmission and mortality (Am J Cardiol doi:10.1016/j.amjcard.2010.I2.018).
“We were interested in NT-proBNP because it’s usually a marker measured on admission for congestive heart failure patients, either to help distinguish heart failure from pulmonary disease, which sometimes can be difficult to distinguish, or to get the baseline status of patients with heart failure. It’s also been used in the outpatient setting to monitor heart failure and response to treatment,” explained lead author Henry Michtalik, MD, MPH, a research and clinical fellow in Hopkins’ Division of Internal Medicine. “The next obvious questions are, what happens to that marker during hospitalization, and can we correlate changes in it with future hospitalizations or adverse events.”
Michtalik and his colleagues consecutively enrolled 241 patients admitted to Johns Hopkins Hospital in Baltimore with a primary admission diagnosis of HF who received intravenous diuretics. NT-proBNP levels were measured at admission and discharge, but physicians made the decision to discharge patients based on clinical judgment alone without knowing the discharge NT-proBNP level. The final analysis included data from 217 patients.
The study’s primary outcome was time to readmission or death within 1 year of discharge. The researchers prospectively grouped participants into two categories based on changes in NT-proBNP levels, including decreases <50% or ≥50%. The analysis revealed that patients with decreases <50% were 40% more likely to be readmitted or die within 1 year, a risk that increased to 57% after adjustment for demographics, admission creatinine level, and admission NT-proBNP level. Additional adjustment for co-morbidities, left ventricular ejection fraction, and length of stay did not significantly change this risk.
“We went into the study with an a priori cutoff change of 50 percent that was not based on variations within our own population, but on other studies that have looked at what’s a clinically significant variation in this marker,” Michtalik said. “The prior literature had evaluated this marker’s correlation with future hospital or adverse events but typically compared patients either to the median in the group and divided them based on that, or evaluated them based on quartiles or quintiles. However, those kinds of cutoffs are based on the data of the population you’re working with, and are not necessarily a reflection of what’s clinically significant with NT-proBNP changes.”
The researchers also looked at the risk of 30-day readmission, and found that 13% of all participants were readmitted during this window, with nearly two-thirds occurring in patients whose NT-proBNP levels decreased <50%. In a multivariate model adjusting for several factors, a <50% decrease in NT-proBNP correlated with a statistically insignificant 42% increase in the risk of readmission within 30 days. However, the researchers cautioned that the reason this result was not statistically significant might have to do with the small number of readmitted patients.
Having relatively small study cohorts is a challenge researchers often face in evaluating HF, and in this case could have masked subtle differences in the patient population, according to Allan Jaffe, MD, professor of medicine in the cardiovascular division at Mayo Clinic and director of clinical core laboratory services at the Mayo Clinic College of Medicine in Rochester, Minn. “One is precluded, because of the size of most heart failure studies, from looking at relative subsets in the study populations,” he said. “It could turn out—and there’s no way to tell from the authors’ data—that there are differences based on the chronicity of heart failure between the groups. So you could argue that patients with more acute, that is to say, more recent, heart failure are much more prone to have a rapid response than those who have chronic heart failure and are therefore further along in the process.”
Jaffe and Michtalik agreed that the study raises broader questions about how best to manage patients hospitalized for HF. “Those individuals whose NT-proBNP levels are not going down as aggressively are probably not as a group doing as well as those whose values are going down aggressively, but what should we do about these patients?” said Jaffe. “The half-life of natriuretic peptides is short, but once you stimulate the system, it gets revved up. So the actual functional half-life of the system, in my view, is longer. That begs the question, once the system’s revved it up, should you not discharge patients until you’ve revved it down?” Jaffe also cautioned about interpreting percentage changes in NT-proBNP. “Perhaps the values seen to be going down will continue to decrease after discharge and thus a 50 percent change is a good metric for response, but not the proper percentage criteria to insist on in other situations,” he added.
Though Michtalik emphasized that follow-up studies are needed, he suggested that these results point to several practical uses for serial NT-proBNP measurements in patients hospitalized for HF. “In inpatient treatment, you could look to achieve a certain level of change in the marker, or if this not possible for patient or hospital reasons, the next step would be to risk-stratify the patients. That would identify the ones who needed more intensive monitoring or follow-up,” he suggested. “What you’d do with that information is establish an early referral system to either primary care providers or cardiologists, to identify these patients as being at higher risk for readmission or relapse and who therefore need more intensive follow-up and treatment.”
Michtalik and his colleagues are considering the possibility of a randomized controlled trial to compare outcomes for HF patients with normal care versus those treated to achieve specific NT-proBNP targets.
Disclosures: Dr. Jaffe reports that he has or does consult for most of the major in vitro diagnostic companies.
Join us for a discussion of this article, and interact with your peers, on LinkedIn.