RELATIONSHIP BETWEEN HEART RATE VARIABILITY AND ELECTROLYTE CONCENTRATION IN CHRONIC RENAL FAILURE PATIENTS UNDER HEMODIALYSIS

A. C. Tsai, ^1,2, H. W. Chiu^1
1 Graduate Institute of Medical Informatics, Taipei Medical University
No. 250, Wushing St., Taipei 110, Taiwan
² Cardiovascular Intensive Care Unit, Far Eastern Memorial Hospital
No. 21, Sec. 2, Nanya S. Rd., Banchiau City, Taipei County 220, Taiwan

Abstract: The aim of this study is to explore the relationship between heart rate variability (HRV) parameters and electrolyte ion concentrations in both pre- and post-dialysis. 5-minute HRV of 20 chronic renal failure patients were analyzed. Results revealed that calcium is negatively correlated to mean of RR intervals and normalized high-frequency (HF) power after hemodialysis. Phosphate is related with low-frequency (LF) power of HRV, normalized LF and LF/HF ratio. Calcium and phosphate are proved to have effects on autonomic modulation.

INTRODUCTION

The mortality of the end stage renal disease is high [1]. An important cause of death in chronic renal failure (CRF)   patients with chronic hemodialysis therapy is cardiovascular disease [2], especially arrhythmia. Acute and chronic body fluid, electrolytes, and pH balance changes are the leading cause of arrhythmia. The electrolytes, including potassium, calcium, and magnesium, are associated with arrhythmia [3].  Autonomic dysfunction is one of uremic neuropathy [4], and adequately recognized in patients with CRF [5]. Heart rate variability (HRV) has been reported to be a useful and noninvasive tool to evaluate cardiac autonomic modulation [6]. There are rare studies mentioned about the relationships among HRV parameters and electrolyte ion concentrations change.

METHODS

20 CRF patients (11 women, 56.2±17.1years, 9 men, 56.4±12.8 years) receiving maintenance hemodialysis therapy 4 hours a time, three times per week at Jen Chi Hospital were included in this study. Patients are measured 5-minute ECG by Megacart (Siemens) at rest.  Calcium, phosphate, sodium, potassium, chloride levels are collected pre- and postdialysis. ECGs were digitized by microcomputer. A computer program was used to detect each normal QRS complex and APCs and VPCs are rejected. To check selected R wave and edit data manually. The next step was calculating the RR interval duration, then measuring MeanRR and SDRR. The short-term frequency domain HRV parameters are computed using Fast Fourier Transform (FFT). Five frequency domain and two time domain parameters of HRV are calculated  [7]: (1) LF : power in LF range ( 0.04~0.15Hz ); (2) LF norm : LF power in normalized units; (3) HF : power in HF range ( 0.15~0.4 Hz); (4) HF norm :  HF power in normalized units; (5) LF/HF : Ratio LF /HF; (6) MeanRR : average of  RR intervals; (7) SDRR : standard deviation of all RR interval.

T test was performed to compare HRV parameters and electrolyte levels pre- and postdialysis. Linear regression was used to correlate HRV parameters with electrolyte ion concentrations before and after hemodialysis. Calcium, phosphate, sodium, potassium, chloride were chosen in this study because of their potential physiologic influences on HRV. p <0.05 were considered significant. The ECG and HRV processing tools as well as statistical analysis were performed with Matlab v.5.3 (The Math Works, Inc., MA) for Microsoft Windows 2000.

RESULTS

Table 1 shows the HRV parameters and electrolyte ion concentrations before and after hemodialysis. Calcium (8.5±0.8mg/dl vs 10.2±1.1mg/dl), phosphate (4.2±1.6 mg/dl vs 2.3±0.7mg/dl), sodium (137±3.4 mEq/L vs 139.6±2.16 mEq/L), potassium (4.3±0.7 mEq/L vs 3±0.4 mEq/L), chloride (100.7±3.5 mEq/L vs 103±2.5 mEq/L) ion concentration changes had significant value between pre- and postdialysis. There are no significant changes on HRV parameters before and after hemodialysis.

TABLE I
Pre- and Post dialysis HRV Parameters and Electrolytes

n=20	Predialysis	Postdialysis	P value
MeanRR (ms)	750.9±129.4	720.0±133.7	0.4755
SDRR (ms)	12.4±6.0	14.3±6.4	0.3338
LF (ms2)	14.3±14.3	19.7±20.4	0.3396
LF norm (nu)	34.8±18.3	38.9±18.3	0.4830
HF (ms2)	8.9±8.1	12.8±12.3	0.2576
HF norm (nu)	28.6±17.3	26.2±15.9	0.6508
LF/HF Ratio	1.8±1.4	2.4±2.2	0.3326
Ca (mg/dl)	8.5±0.8	10.2±1.1	<0.001
P (mg/dl)	4.2±1.6	2.3±0.7	<0.001
Na (mEq/L)	137±3.4	139.6±2.16	<0.05
K (mEq/L)	4.3±0.7	3±0.4	<0.001
Cl (mEq/L)	100.7±3.5	103±2.5	<0.05

Calcium level was negatively proportional to MeanRR and HF norm in postdialysis. Figure 1 shows the relationship between Ca and MeanRR.  Phosphate level was positively proportional to LF, LF norm and LF/HF ratio after hemodialysis. The relationship between phosphate and LF/HF ratio is shown in Figure 2. There is no significant relationship between HRV parameters and electrolyte ion concentrations in predialysis.

Figure 1. Relationship between calcium and MeanRR is negatively proportional. Regression line (y=-63.7523x +1370.6638) is also shown in this figure and r is the correlation coefficient.

Figure 2. Relationship between phosphate and LF/HF ratio is positively proportional. Regression line (y=1.8585x-1.8853) is also shown in this figure and r is the correlation coefficient.

DISCUSSION

HRV reduction on CRF patients was proved [8,9]. The abnormality of electrolytes of CRF patients may influence HRV, but less research discussed about relationship between HRV parameters and variance of electrolyte ion concentrations. Recently, the relationship between calcium ion concentration and LF/HF ratio was confirmed [10]. In this study, statistically significant relationships between calcium ion, MeanRR, and HF norm in postdialysis were negatively proportional. Calcium ion is needed when muscle contraction and nerve pulse conduction occur [3]. Since HF power reflects the activity of vagal nerves [7], our results revealed that calcium ion concentration had effects on vagal nervous modulation in postdialysis. In addition, phosphate ion concentration was positively correlated to LF, LF norm, and LF/HF ratio. Some investigations demonstrated that LF norm and LF/HF ratio are considered to reflect the sympathetic modulations [7,8]. Therefore, an increased phosphate ion concentration will increase sympathetic nervous activity according to the positive correlation between P and LF related parameters.

Acknowledgements: This study was supported by Grant NSC90-2213-E-010-038 from the National Science Council, ROC.

REFERENCES

[1] U. S. Renal Data System. “Patient mortality and survival , ” Am J Kidney Dis , vol. 28, pp.  S79-S92, 1996.

[2] E. Braunwald, A.S. Fauci, K.J. Isselbacher, et al.. Harrison’s Principles of Internal Medicine. NY:The McGraw-Hill Companies, 2001.

[3] S.C. Lin, Advanced Resuscitation.  Taipei: Li Ta, 2001.

[4] A. Yaldiz, M.S. Sever, S. Demirel, et al..”Improvement of uremic autonomic dysfunction after renal transplantation : a heart rate variability study,” Nephron, vol. 80, pp. 57-60, 1998.

[5] K. Nishi, M. Yoshida, Y. Wada, et al.. “Vascular response to vasoactive agents in dialysed patients with chronic renal failure with and without diabetes mellitus,” Nephrol Dial Transplant, vol 9, pp.10-15, 1994.

[6] R.E. Kleiger, J.T. Bigger, M.S. Bosner, et al..  “Stability over time variables measuring heart rate variability in normal subjects,” Am J Cardiol, vol. 68, pp. 626-630, 1991.

[7] Task Force of the European Society of Cardiology the North American Society of Pacing Electrophysiology. “Heart Rate Variability : Standards of Measurement, Physiological Interpretation, and Clinical Use,” Circulation, vol. 93, pp. 1043-1065, 1996.

[8] A.A. Steinberg, R.L. Mars, D.S. Goldman, et al.. “Effect of end-stage renal disease on decreased heart rate variability,” American Journal of Cardiology, vol. 82, pp.1156-8, 1998.

[9] J. Hayano, H. Takahashi, T. Toriyama, et al.. “Prognostic value of heart rate variability during long-term follow-up in chronic haemodialysis patients with end-stage renal disease,” Nephrol Dial Transplant, vol. 14, pp. 1480-1488, 1999.

[10] S. Buda, T. Stompor, W. Sulowicz, et al.. “The impact of changes in levels of calcium, phosphate and magnesium during hemodialysis on autonomic system reactivity as measured by heart rate variability analysis,” Przeglad Lekarski, vol. 57, 340-345, 2000.