Utility of N-terminal pro-brain natriuretic peptide in detecting diastolic dysfunction in asymptomatic hypertensive patients : comparison with echocardiography

Introduction Studies have provided evidence of a direct relationship between blood pressure and cardiovascular disease. Blood pressure is a major contributor of heart failure (HF) as 90% of new cases of HF in the Framingham Heart Study had earlier history of hypertension.1 Diastolic HF is responsible for as many as 74% of cases of HF among patients with hypertension.2 Despite this, this entity is rarely diagnosed because these patients have normal systolic function on echocardiogram. Detection of diastolic dysfunction in hypertensive patients is a major problem. Assessment of diastolic function is difficult and ideally requires cardiac catheterization, although non-invasive measure like echocardiogram is commonly


Introduction
Studies have provided evidence of a direct relationship between blood pressure and cardiovascular disease.Blood pressure is a major contributor of heart failure (HF) as 90% of new cases of HF in the Framingham Heart Study had earlier history of hypertension. 1 Diastolic HF is responsible for as many as 74% of cases of HF among patients with hypertension. 2Despite this, this entity is rarely diagnosed because these patients have normal systolic function on echocardiogram.Detection of diastolic dysfunction in hypertensive patients is a major problem.Assessment of diastolic function is difficult and ideally requires cardiac catheterization, although non-invasive measure like echocardiogram is commonly used, they are non-specific.Although Doppler assessment of mitral inflow velocity pattern by echocardiogram 3 has been in use to assess left ventricular (LV) diastolic function, there are some limitations of this technique which suggest the need for other objective measures to look for diastolic abnormalities. 4 Addition of pulsed wave tissue Doppler imaging (TDI) has significantly enhanced the noninvasive assessment of diastolic function.This test is not easily available for routine diagnostic screening in resources limited areas.Therefore, alternative diagnostic procedures which will be helpful to detect diastolic dysfunction is necessary.Natriuretic peptides are natriuretic hormones released primarily from ventricles of the heart.The serum concentration of natriuretic peptides gets raised in both asymptomatic and symptomatic patients with LV dysfunction, allowing their use in diagnosis. 5Data suggest that serum brain natriuretic peptide (BNP) and N terminal pro-BNP (NT pro-BNP) partly reflects ventricular pressure, so we hypothesized that BNP and NT pro-BNP levels could have a role in the early detection of diastolic abnormalities in hypertensive patients.Therefore, we performed this study to assess the role of NT pro-BNP for early detection of diastolic abnormalities in asymptomatic hypertensive patients and to correlate its serum concentration level with echocardiographic measures of diastolic dysfunction.

Materials and Methods
This is a comparative cross-sectional study which included 100 asymptomatic hypertensive patients.Inclusion criteria was systolic blood pressure >160 mm Hg or diastolic blood pressure >100 mm Hg on 2 or more measurements taken at least 60 minutes apart.Enrolled patients had normal LV systolic function, defined as LV ejection fraction >55%, with no major wall motion abnormalities, except for LV hypertrophy on echocardiogram, no clinical symptoms or signs of heart failure or other acute or chronic target organ damage related to hypertension.Also one-hundred age and sex-matched apparently healthy subjects (controls) whom we selected among the accompanying attendants of patients in cardiology clinic of B.P. Koirala Institute of Health Sciences (BPKIHS), Dharan, Nepal from July 2015 to June 2016 were selected.We included controls in this study to look for the prevalence of diastolic dysfunction among apparently healthy people detected by echocardiogram and to get the reference value of serum NT pro-BNP in our healthy population so that we make a comparison with hypertensive patients who are known to have a higher prevalence of diastolic dysfunction.There was no cut off value of serum NT-pro BNP for diagnosis of diastolic dysfunction.

Echocardiography
All hypertensive patients and controls underwent M-Mode, 2 dimensional (2D), color flow and pulsed wave Doppler transthoracic echocardiogram by Philips IE33 echo machine.Qualified registered cardiologist performed the echocardiogram and again reconfirmed by another cardiologist.We measured pulsed wave Doppler recordings in the apical four chamber view from a 4 x 4 mm sample volume placed at the tips of mitral leaflets.The transmitral pulsed Doppler velocity readings from three consecutive cardiac cycles used to derive peak velocities in the early part of diastole (E) and contraction of the atrium (A).We measured deceleration time (DT) as the interval from the E wave to identify declining velocity regarding to baseline.We obtained LV isovolumic relaxation time (IVRT) in milliseconds from the apical 5-chamber view with a pulsed Doppler sample volume positioned to straddle the LV outflow tract and mitral orifice to get signals from aortic valve closure.If diastolic abnormalities were present with any of these three measures i.e. conventional Doppler recordings, the effects of the Valsalva maneuver on transmitral flow velocities or TDI based on criteria for the diagnosis of diastolic dysfunction, 6 the subjects were considered to have LV diastolic dysfunction.

Measurement of serum NT-proBNP
After a resting period for 20 minutes, we drew 10 ml blood in a plain tube from an antecubital vein from the subjects under study for serum NT-proBNP measurement.After centrifugation at 2500 rpm for 5 minutes, we took serum for NT-proBNP measurement immediately or was kept at -40 o C if delayed.We used the VIDAS instrument using Enzyme-linked Fluorescent assay (ELFA) technique for serum NT-proBNP measurement.VIDAS NT-proBNP is an automated quantitative test for the determination of human serum or plasma NT-proBNP.We entered collected data in Microsoft Excel 2007 and converted into SPSS 20 version.We applied Independent t-test to find out the significant differences between groups (cases and controls) and within the group at a 95% confidence interval.Pearson correlation statistics measured the degree of relationship between serum NT-proBNP values and echocardiographic parameters.

Results
Table 1 shows the baseline characteristics of all 200 patients (100 cases and 100 controls).Systolic, diastolic and mean arterial blood pressures were significantly higher (hypertensive range) in cases compared to controls.Echocardiographic parameters of diastolic dysfunction like E/A ratio, E septal and lateral velocity, E/E' , DT, IVRT showed the significant difference between cases and controls as shown in Table 2.The level of mean serum NT-proBNP was significantly higher in patients with hypertension compared to healthy controls.There was no significant difference in mean serum NT-proBNP levels between cases with no left ventricular diastolic dysfunction (n=14) and controls (n=100) but it shows the statistically significant difference between cases with left ventricular diastolic dysfunction of all grades (n=86) and controls (n=100).Table 3 represents the mean serum NT-proBNP levels in the normal and abnormal diastolic function groups.The difference between groups was significant as patients diagnosed with abnormal diastolic function (n=102) had a mean serum NT-proBNP concentration of 187.6 ± 117.9 pg/mL whereas the patients with normal diastolic function (n=98) had a mean concentration of 60.1 ± 22.08 pg/mL.NT pro-BNP levels were significantly higher in all groups of patients with E/E' ratios <8, 8-12, >12 (197 ± 121.25 pg/mL, 263.12 ± 122.52 pg/mL and 180 ± 106.56 pg/mL respectively) as compared to healthy controls (58.51 ± 11.01 pg/mL) (P value <0.015) as shown in Figure 1.However, the difference was not significant among patient groups with E/E'<8, 8-12 and >12.The non-significant higher values in cases with E/E' <8 compared to cases with >12 could be due to by chance or small sample size.Mean serum NT-proBNP levels were significantly higher in patients with IVRT >90 ms (270.43 ± 121.3 pg/mL) as compared to healthy controls (58.51 ± 11.01 pg/mL) (P=0.03)but the difference was not significant between cases with IVRT of 60-90 ms (152.02 ± 102.2 pg/mL) and controls (58.51 ± 11.01 pg/mL) (P =0.09).NT-proBNP levels were higher in both group of patients with E/A ratios <0.9 and 0.9-1.5 (245.72 ± 166.7 pg/mL and 210.69 ± 143.53 pg/mL respectively) as compared to healthy controls (58.51 ± 11.01 pg/mL).Figure 2 represents the mean serum NT-proBNP levels in the normal and abnormal diastolic function groups.The difference between groups was significant.Patients diagnosed with abnormal diastolic function (n=102) had mean serum NT-proBNP concentration of 187.6 ± 113.9 pg/mL whereas the patients with normal diastolic function (n=98) had a mean concentration of 60.1 ± 22.08 pg/mL.Table 4 shows the correlation of serum NT-proBNP level with blood pressure and echocardiographic parameters of hypertensive patients.

Discussion
The results of this study show that raised serum concentration of NT-proBNP represents a serum marker providing positive evidence of the presence of a diastolic impairment in patients with asymptomatic   Diastolic dysfunction in hypertension is common and usually accompanied by left ventricular hypertrophy (LVH).Our study showed that 75% of cases of hypertension had echocardiographic evidence of LVH and 40% met electrocardiographic criteria of LVH which is more or less similar to the results of a previous study. 7Eightysix percent of our hypertensive patients had diastolic abnormalities in contrast to 16% of healthy controls.This higher prevalence of LVH and LV diastolic dysfunction in our asymptomatic hypertensive population could be due to long-standing undiagnosed hypertension, lack of awareness and screening program in our community.Doppler trans-mitral velocity measurements have become the foundation for the diagnosis of diastolic dysfunction. 8owever, pitfalls in the echo-Doppler assessment of diastolic dysfunction exist.A simple, rapid blood test that reflects diastolic dysfunction would be of significant clinical benefit.
A study has demonstrated that BNP can predict the absence or presence of LV dysfunction as seen in echocardiogram and might be used as a screening tool for the detection of LV dysfunction and which may, in fact, obviate the need for echocardiogram. 9Our study reveals significantly higher serum level of mean serum NT-proBNP in patients with LV diastolic dysfunction of all grades (n=86) compared to controls (n=100) indicating that NT-proBNP could be a valuable test to easily rule out the diastolic dysfunction in asymptomatic hypertensive patients with early mild diastolic dysfunction.Studies suggest that age and gender may influence circulating natriuretic peptide levels. 10,11Our study didn't show the significant difference in serum NT-proBNP concentration between genders and among different age groups in both cases and healthy controls.This could be due to the small sample size.Likewise, our study did not show the significant difference in serum NT-proBNP concentration among patients with different body mass index since the obese population has lower circulating NT-proBNP and BNP levels. 12 study has shown an association between plasma BNP levels and blood pressure variability and found that BNP level correlated with mean blood pressure. 13Similarly, in our study, serum NT-proBNP levels positively correlated with mean arterial blood pressure among hypertensive patients (r =0.22), (P value = 0.02).This could be explained by the release of an increased amount of BNP by pressureoverloaded LV due to higher mean blood pressure.With increasing age, the mitral E velocity and E/A ratio decrease, whereas DT and A velocity increase.This can lead to over-diagnosis of diastolic dysfunction without increased LV filling pressure in elderly patients.NT-proBNP could be of value to mitigate this problem.In our study, NT-proBNP levels were higher in both group of patients with E/A ratios <0.9 and 0.9-1.5 as compared to healthy controls.This indicates that serum NT-proBNP level correctly diagnoses the impaired diastolic dysfunction with increased LV filling pressure.E/E' ratio has long been considered as the best parameter for the diagnosis of diastolic HF and it could be used for estimation of LV filling pressures. 14Lubian et al 15 found increased serum concentration of NT-proBNP levels in patients with diastolic dysfunction but differences between hypertensive and non-hypertensive patients without diastolic dysfunction was not significant.This indicates that, in our patients with hypertension, the level of NT-proBNP increased because of changes in diastolic function.In other words, NT-proBNP levels in  hypertensive patients probably increase due to diastolic dysfunction.
In this study, serum concentration of NT-proBNP levels was significantly higher in all groups of patients with E/E' ratios <8, 8-12 and >12 as compared to healthy controls.The increased concentration in cases with E/E' <8 in this study could be due to that hypertensive patients with normal filling pressure may have higher serum concentration of NT-proBNP levels or due to by chance owing to small sample size which needs further study.
In patients with diastolic dysfunction, LV pressure falls slowly which leads to delayed opening of the mitral valve and prolongation of IVRT. 6In this study, mean serum concentration of NT-proBNP levels were significantly higher in patients with IVRT >90 ms as compared to healthy controls (P = 0.03) which indicate that these hypertensive patients have an echocardiographic measure of impaired myocardial relaxation who had increased serum concentration of NT-proBNP.

Limitations
This is a comparative cross-sectional study in a limited number of asymptomatic patients of hypertension.Moreover, we enrolled cases and controls in a consecutive manner rather than randomization.We measured serum NT-proBNP at a point of time when patients presented with severe hypertension and did not look at the effect when blood pressure is well controlled.

Conclusion
Serum concentration of NT-proBNP in the setting of normal LV systolic function correlates with the absence or presence of diastolic abnormalities detected by echocardiogram.A low NT-proBNP level may obviate the need for an echocardiogram in some patients.An elevated serum NT-proBNP level is an indication of the presence of LV dysfunction warranting further cardiac workup.This study demonstrated that NT-proBNP could be of help for the detection of LV diastolic dysfunction in asymptomatic patients with hypertension.Future large studies would be needed to decide whether serum NT-proBNP levels might be part of one of the diagnostic modalities for the diagnosis of diastolic dysfunction in patients with hypertension.

Figure 1 .
Figure 1.Serum NT-proBNP expressed as reflection of E/E' ratios.Values are expressed as mean.NT-proBNP is higher in cases (E/E'<8, 8-12 and >12 groups) as compared to controls (P value <0.015)

Table 1 .
Baseline characteristics and echocardiographic measures of patients with asymptomatic hypertensive patients vs healthy controls BMI: body mass index; BPM: beat per minute; SBP: systolic blood pressure; DBP: diastolic blood pressure; LVH: left ventricular hypertrophy.

Table 3 .
Comparison of Serum NT-proBNP between cases and controls