Relationships between Brain Natriuretic Peptide, Troponin I and QT Dispersion in Asymptomatic Dialysis Patients

Abstract

Objectives. The relationships between increased wall stress, myocyte death, and ventricular repolarization instability in patients with heart failure were reported. Design and Methods. The relationships between brain natriuretic peptide (BNP), a predictor of increased wall stress of hearth; troponin I (cTnI), a predictor of myocyte death; and QT dispersion (QT_d), a reflection of ventricular repolarization instability were evaluated in age- and sex-matched asymptomatic 29 hemodialysis (HD) patients and 26 peritoneal dialysis (PD) patients, and the finding were compared. Results. Serum BNP and cTnI levels in HD patients (722.9 ± 907.9 pg/mL, 0.05 ± 0.07 μg/L, respectively), just before HD, were significantly higher than those of PD patients (255.4 ± 463.7 pg/mL, 0.02 ± 0.02 μg/L, respectively; p < 0.05). There was no significant difference between groups with regard to corrected QT_d and maximum and minimum QT intervals (p > 0.05). Serum cTnI levels were significantly and positively correlated with serum BNP levels in both dialysis groups (r = 0.447, p = 0.048). No relationship was found between plasma BNP and ECG parameters studied in both groups (p > 0.05). Conclusion. Increased serum cTnI levels were associated with elevated BNP levels in both dialysis groups. The increases in BNP and troponin I are more likely to reflect hypervolemia. Although CAPD patients were receiving dialysis daily and HD patients were more hypervolemic, CAPD patients have similar QT_dc and accordingly a similar tendency toward arrhythmias. This suggests that factors other than electromechanical interaction may be important in determining the QT interval length in patients on dialysis.

Keywords:

• brain natriuretic peptides
• peritoneal dialysis
• hemodialysis
• QT dispersion
• troponin

INTRODUCTION

Elevated concentrations of cardiac biomarkers, such as troponin I (cTnI) and brain natriuretic peptide (BNP), and prolonged QT interval dispersion (QT_d) have been shown to be predictive of poorer long-term cardiovascular outcomes in stable patients with end-stage renal disease (ESRD).[1–6]

Brain natriuretic peptide (BNP) synthesis, occurring primarily in the ventricle, is influenced by increases in ventricular filling pressures and increased afterload. BNP has been shown to be a good biochemical marker of left ventricular hypertrophy, volume overload, and all-cause cardiovascular mortality.[4],[5],[7–9]

The presence of cardiac troponin in the serum indicates myocardial injury or loss of cell membrane integrity, and it is useful in diagnosing and assessing prognosis in acute coronary syndromes. However, several small studies have reported elevated cardiac troponin levels in patients with decompensated heart failure in the absence of acute coronary syndromes, and furthermore a correlation between troponin elevation and poor prognosis.[10–13] Therefore, Horwich et al.[14] reported that the release of cTnI into the serum was strongly correlated with elevation of cardiac filling pressures and BNP in the absence of acute coronary syndromes.

Dispersion of the QT interval (QT_d), which is the difference between the longest and the shortest QT interval in all electrocardiographic leads, is a reflection of regional variation in ventricular repolarization and a predictor of arrhythmia and cardiovascular mortality in conditions such as cardiomyopathies and chronic heart failure. It was reported that myocardial stretch caused by increased intracardiac pressure slows conduction, enhances refractoriness, and triggers after depolarizations and ventricular ectopic beats.[15–17] These changes may lead to ventricular action potential prolongation and ventricular repolarization instability in patients with heart failure.[15–17]

The purposes of this study were to compare serum BNP and cTnI levels and QT_d in asymptomatic chronic hemodialysis (HD) patients, when they are on peak volume status just before HD with those of continuous ambulatory peritoneal dialysis (CAPD) patients, and to investigate the relationship between serum BNP levels and cTnI and QT_d.

MATERIAL AND METHODS

Age- and sex-matched 29 HD patients (16 men, 13 women; mean age, 55.6 ± 14.8 years) and 26 PD patients (11 men, 15 women; mean age, 48.6 ± 15.3 years) on dialysis for longer than two months were studied.

Patients with cardiovascular diseases such as myocardial infarction, angina pectoris, valvular heart disease, and hypertrophic and dilated cardiomyopathy were excluded in this study by history, physical examination, electrocardiography, and echocardiography.

All HD patients received regular dialysis using hemophane hollow-fiber dialyzer three times per week in sessions lasting four hours. The dialysate was bicarbonate buffered and contained 140 mmol/L sodium. All patients on CAPD received four exchanges per day using standard dialysis bags (8 L/day).

A 12 lead-standard ECG, in which all leads were recorded simultaneously at a speed of 25 mm/sec, was used. During the entire study period, no patient was taking potentially QT-prolonging agents or developed serious hypocalcemia or hyperkalemia, factors known to cause QT prolongation. All electrocardiograms were visually analyzed by a single experienced medical doctor without knowledge of clinical information, to the nearest 10 ms with the aid of a caliper and a magnifying device with a grid as described.[18] QT_d was defined as the difference between the maximum (QT_max) and minimum (QT_min) QT interval in any two leads. QT intervals and QT_d were corrected for heart rate with Bazett's formula:where R-R is the RR interval in seconds.[19] The derived ECG indices—corrected QT_max (Qt_maxc), corrected QT_min (QT_minc), and corrected QT_d (QT_dc)—were correlated with the demographic findings and serum BNP levels.

In HD patients, ECG recordings, serum BNP, cTnI, and other biochemical measurements were obtained just before HD session.

Cardiac troponin I and BNP levels were measured with the Abbott AxSYM microparticle enzyme immunoassay methods (Abbott Laboratories, Chicago, Illinois, USA). Blood samples were collected without an anticoagulant for the measurement of cardiac enzyme level. The samples were allowed to coagulate for at least 30 min and then centrifuged at 520 g for 10 min. The resulting serum was divided into aliquots, frozen, and stored at −70°C until the completion of cardiac marker concentration measurements.

After 20–30 min of quiet resting in a semirecumbent position, samples for BNP were taken into chilled ethylenediaminetetraacetic acid vacutainers, placed immediately on ice, and centrifuged within 30 min at −4°C.

Routine clinical blood chemistry variables, including blood urea nitrogen (BUN), albumin, and creatinine, were analyzed using standard methods.

Statistical Analyses

All results are expressed as mean ± SD. Associations with variables were examined with Mann-Whitney U tests. Pearson's linear correlation coefficient (or a Spearman's test if any variable was ordinal and/or not normally distributed) for correlation between variables were used.

RESULTS

Causes of renal failure in CAPD patients were diabetic nephropathy (n = 7, 26.9%), chronic glomerulonephritis (n = 3, 11.5%), hypertensive nephrosclerosis (n = 2, 7.7%), others (n = 2, 7.7%), and unknown (n = 12, 46.2%).

Causes of renal failure in HD patients included diabetic nephropathy (n = 10, 34.5%), hypertensive nephrosclerosis (n = 3, 10.3%), others (n = 7, 24.1%), and unknown (n = 9, 31.0%).

There were no significant differences between HD and CAPD patients with regards to age, sex, or dialysis duration (p > 0.05). Demographic and laboratory characteristics of the all patients were shown in Table 1. Serum BNP levels and cTnI levels were markedly greater in HD patients than those of PD patients (p < 0.05).

Table 1 Demographic and laboratory characteristics of all patients

	PD	HD
	Mean ± SD	Mean ± SD
Age (years)	48.6 ± 15.3	55.6 ± 14.8	NS
Duration of dialysis (years)	3.7 ± 2.5	3.4 ± 2.6	NS
BUN (mg/dL)	50.5 ± 17.6	72.1 ± 18.9	p < 0.05
Cr (mg/dL)	8.1 ± 2.1	7.6 ± 2.6	p < 0.05
Alb (mg/dL)	3.7 ± 0.3	3.6 ± 0.3	NS
BNP (pg/mL)	255.4 ± 463.7	722.9 ± 907.9	p < 0.05
cTnI (μg/L)	0.02 ± 0.02	0.05 ± 0.07	p < 0.05
QTmaxc	449.3 ± 58.5	452.4 ± 39.8	NS
QTminc	393.0 ± 41.8	401.7 ± 40.1	NS
QTdc	56.4 ± 36.2	50.8 ± 28.6	NS

Abbreviations: NS = nonsignificant, BUN = blood urea nitrogen, Cr = creatinine, Alb = serum albumin, BNP = brain natriuretic peptide, cTnI = cardiac troponin I, QT_maxc = maximum corrected QT interval, QT_minc = minimum corrected QT interval, QT_dc = corrected QT dispersion.

There was no difference with regard to QT_maxc, QT_minc, and QT_dc between the two groups. The cTnI concentration was below the assay detection limit (0.01μg/L) in eight (30.7%) patients, but the remaining 18 (69.2%) patients had cTnI concentrations of 0.01–0.09 μg/L in the CAPD group. The cTnI concentration was below the assay detection limit in four (13.7%) patients, but the remaining 25 (86.2%) patients had cTnI concentrations of 0.01–0.37 μg/L in the HD group.

Serum cTnI levels were positively correlated with elevated BNP levels in both dialysis groups (p < 0.05). There was no significant correlation between BNP concentrations and ECG parameters studied in either HD or CAPD patients (p > 0.05; see Tables 2 and 3).

Table 2 Correlations between clinical parameters, QT dispersion, cTnI, and BNP in HD patients

	cTnI		QTmaxc		QTminc		QTdc		BNP
	r	p	r	p	r	p	r	p	r	p
Age (years)		NS		NS		NS		NS		NS
Duration (months)		NS		NS		NS		NS		NS
BNP (pg/mL)	0.405	0.040				NS		NS		NS

Abbreviations: NS = nonsignificant, BNP = brain natriuretic peptide, cTnI = cardiac troponin I, QT_maxc = maximum corrected QT interval, QT_minc = minimum corrected QT interval, QT_dc = corrected QT dispersion.

Table 3 Correlations between clinical parameters, QT dispersion, cTnI, and BNP in CAPD patients

	cTnI		QTmax		QTmin		QTdc		BNP
	r	p	r	p	r	p	r	p	r	p
Age (years)				NS		NS		NS	0.396	0.045
Duration (months)		NS		NS		NS		NS		NS
BNP (pg/mL)	0.551	0.005				NS		NS		NS

Abbreviations: NS = nonsignificant, BNP = brain natriuretic peptid, cTnI = cardiac troponin I, QT_maxc = maximum corrected QT interval, QT_minc = minimum corrected QT interval, QT_dc = corrected QT dispersion.

DISCUSSION

In patients on HD, the peaks and valleys of total body volume are seen. PD has more consistent volume removal than HD. Here, serum BNP levels, and cTnI and QT_dc in patients on HD when they are on peak volume status, were compared just before HD with those of CAPD patients. BNP levels were markedly greater in HD patients just before HD (722.9 ± 907.9 pg/mL) than those of PD patients (255.4 ± 463.7 pg/mL). In both dialysis groups, BNP levels were upper limit of the normal range (<100 pg/mL). The HD patients might have more volume load than PD patients. A greater expansion of extra-cellular volume causing myocardial stretching and increased left ventricular pressures may be the principal cause of higher BNP levels in HD patients.

The cTnI concentration was below the assay detection limit in eight (30.7%) patients, but the remaining 18 (69.2%) patients had cTnI concentrations of 0.01–0.09 μg/L in the CAPD group. The cTnI concentration was below the assay detection limit in four (13.7%) patients, but the remaining 25 (86.2%) patients had cTnI concentrations of 0.01–0.37 μg/L in the HD group. In this study, serum cTnI levels were positively correlated with serum BNP levels, a predictor of increased ventricular filling pressures and afterload, in both dialysis groups.

Cardiac troponin levels are commonly elevated in asymptomatic end stage renal disease (ESRD) patients even in the absence of clinically suspected acute myocardial ischaemia and with an increase in mortality. In a large study that included 733 asymptomatic patients with ESRD, an extremely high percentage of the patients had elevated concentrations of troponins despite the use of the latest assay technology.[1] The causes of these elevated concentrations are unclear, but may include injury to skeletal muscle, heart failure, left lenticular hypertrophy, apoptosis, or impaired renal clearance.[1],[2]

In vitro experiments with cardiac muscle cells have identified a link between myocardial wall stretch and myocyte functional injury and cell death.[20],[21] Increased troponin proteolysis has been identified in volume-overloaded rat hearts.[22] At the cellular level, multiple intracellular signaling cascades are activated in the heart in response to changes in mechanical loading. The association between increased wall stress and myocyte death was explained by multiple mechanisms. It was reported that increased wall stress may directly activate intracellular signaling cascades. It has also been hypothesized that increased myocardial wall stress leads to decreased subendocardial perfusion, even in the absence of coronary artery disease, resulting in a decline in systolic function.[23–25] Recently, Wang et al.[26] reported that troponin T elevation was positively associated with left ventricular mass and negatively associated with left ventricular systolic function, and it was predictive of subsequent risk of cardiovascular congestion in peritoneal dialysis patients.

In the literature, there are few studies that compared PD and HD patients according to QT dispersion. In one study, patients on HD were reported to have a higher rate of increased QT dispersion than PD patients, although this observation was not consistent elsewhere.[27],[28] In these studies, QT dispersion was evaluated at the post-hemodialysis period in HD patients. No difference was found with regard to QT_dc between two groups. QT_d >40 ms has 88% sensitivity and 57% specificity for the prediction of inducibility of sustained ventricular tachycardia during and electrophysiology study.[29] We found that QT_dc was (50.8 ± 28.6 ms) in HD and (56.4 ± 36.2 ms) in PD patients. The causes of the prolongation of QT dispersion in dialysis patients are multifactorial, including fibrosis and hypertrophy of the heart, changes of cellular, or interstitial fluid composition. However, no significant correlation was found between plasma BNP and ECG parameters, even though HD patients are more hypervolemic. This suggests that factors other than electromechanical interaction may be important in determining the QT_c interval length in patients on dialysis.

In conclusion, increased serum cTnI levels was associated with elevated BNP levels in both dialysis groups. The increases in BNP and troponin I are more likely to reflect hypervolemia. Although CAPD patients are receiving dialysis daily, they also have similar QT dispersions and accordingly a similar tendency to arrhythmias.