Substantial intraindividual variability of BNP concentrations in patients with hypertension

Abstract

Plasma concentrations of B-type natriuretic peptides (BNP) independently predict the risk of death and cardiovascular events. In the present study, we investigated the intraindividual variability of BNP concentrations, a potential confounder of risk prediction. Consecutive outpatients with blood pressure (BP) values of at least 140/90 mm Hg and not taking BP lowering therapy were asked to participate. Exclusion criteria were renal insufficiency, structural heart disease on echocardiography, except left ventricular hypertrophy and any other severe concomitant illness. Plasma BNP levels were determined on two different days using the same assay. In total, 77 patients were included. Mean age was 54712 years, 55% were male and mean systolic/diastolic BP was 163716/ 9678 mm Hg. Mean creatinine was 70714 lmol/l. The median interval between the two BNP assays was 10 days (interquartile range 1-23 days). Median BNP concentrations were 17 and 16 pg/ml for the first and second visit, respectively (P ¼ 0.48). However, there was a wide range of differences in BNP values among individual patients, 34 patients (44%) having an absolute difference of at least 10 pg/ml. When patients were categorized according to tertiles of BNP levels, 25 (32%) changed from one tertile at the first visit to another at the second visit. In conclusion, these data indicate that BNP levels may be used on a population level. However, the high intraindividual variability seems to preclude useful risk stratification in the individual patient. Care should be taken in the interpretation of single BNP values below the currently accepted thresholds for heart failure.

The association of brain natriuretic peptide and insulin resistance in obesity-related hypertension

“…On the other hand, the high intraindividual variability of BNP was defined in several studies. 20,21 This variability seems to preclude the assessment of its concentration in the individual patient with either hypertension or obesity and/or insulin resistance. The assessment of the changes in BNP concentration in longitudinal studies may enlight this issue.…”

Intraindividual temporal variance of biomarkers and the definition of individualized prognosis in cardiovascular patients

“…The article by Conen et al 1 also allows us to extend the principle that he illustrates. This is that repeated measures of biomarkers may bring more information than an insight into the variability of the biomarker and its assay.…”

“…If repeated measures of biomarkers is an adequate marker of disease stability and or response to treatment, then is it not perfectly logical to extend the study from Conen et al 1 to use repeated measures of sensitive biomarkers within individuals to define prognosis and treatment efficacy. With respect to BNP, there is already a large body of opinion and some data suggesting tailored treatment on the basis of BNP levels.…”

“…There is good reason to draw attention to the paper by Conen et al 1 in this edition of the Journal of Human Hypertension. In the first instance, this simple study reaffirms the often forgotten principle that an isolated measure is simply just that, a mere approximation of any biological variable subject to both measurement and biological variance.…”

“…Of course it is equally well understood that the normal range of any point assay is variable and that at different ranges of concentration the variance associated with a laboratory measurement differs, as illustrated by Conen et al 1 With the application of BNP measurement in mind, there are significant issues as to the interpretation of absolute levels (which are generally made on point estimates in clinical practice) on no more than one estimate at any given time point. In addition, the time constant of response for BNP measurements in any setting is poorly defined, and significant changes in cardiac function might outstrip the ability of BNP to respond.…”