The Study of Spectral Analysis of Heart Rate Variability in Different Blood Pressure Types in Euvolemic Peritoneal Dialysis Patients

Abstract

Background: Heart rate variability (HRV) is an useful noninvasive tool to assess autonomic nervous system (ANS) function and may provide an insight into the understanding of the role of ANS in the pathogenesis of blood pressure (BP) abnormality in euvolemic continuous ambulatory peritoneal dialysis (CAPD) patients. Methods: In this cross-sectional study, 62 CAPD patients with normal hydration values (assessed by a bioimpedance spectroscopy device) were enrolled from our peritoneal dialysis (PD) unit. Patients were divided into three groups according to their BP: normotension (NT), hypertension (HT), and hypotension (HyT). Spectral analysis of HRV was measured by examination of the average heart rate using standard electrocardiogram in 5 min. Results: The HyT group had the lowest serum urea nitrogen among the three groups (all p < 0.05), and a lower serum potassium and albumin as compared with the NT group (p < 0.05). The HyT group also had the highest total power (TP) and highest frequency power (HF) among the three groups (all p < 0.05), and a lower normalized low frequency power (LF), LF/HF ratio, and a higher normalized HF as compared with the HT group (all p < 0.05). Conclusion: Our study suggested that autonomic insufficiency was present in euvolemic PD patients, and hypotensives had a relatively higher parasympathetic activity and blunted sympathetic activity.

• heart rate variability
• peritoneal dialysis
• blood pressure
• hypertension
• hypotension
• autonomic nervous system

INTRODUCTION

The prevalence of hypertension (HT) in peritoneal dialysis (PD) patients is high and ranges from 29% to 88%,^1,2 and the prevalence of hypotension (HyT) is about 12%.³ Both of them have been shown to be associated with poor clinical outcomes in end-stage renal disease (ESRD) patients,^4–6 and many factors have been shown to be associated with blood pressure (BP) abnormality in ESRD. Volume overload or depletion was considered to play a key role in BP regulation in patients with ESRD on chronic dialysis.^3,7–9 However, in our experience, it is not uncommon that HT or HyT occurred in dialysis patients with normal volume status. Thus, other factors associated with BP regulation should be investigated in this patient population.

Chronic imbalance of the autonomic nervous system (ANS) had been found to be a common phenomenon in uremic patients and might have a number of clinical sequelae.¹⁰ Till now, many researches were focused on the ANS and BP regulation.^11–13 Spectral analysis of heart rate variability (HRV) can partially distinguish parasympathetic and sympathetic influences and may provide important insights into the role of the ANS in the pathogenesis of BP abnormality.¹⁴ Although previous studies have indicated that lower HRV contributed to the generation and maintenance of high BP in essential HT^15–17 and abnormal HRV has also been suggested as a cause of hemodialysis (HD)-induced HyT;^18–21 the association between HRV and different BP types in PD patients, especially in euvolemic continuous ambulatory peritoneal dialysis (CAPD) patients have never been clearly elucidated. So in this study, we have compared the measures of HRV in euvolemic CAPD patients with HyT, normotension (NT), and HT.

PARTICIPANTS AND METHODS

Study Population

It is a cross-sectional study. From January to May 2010, CAPD patients from our PD unit were enrolled. The exclusion criteria were (a) dialysis duration of less than 3 months, (b) unwillingness to participate in the study, (c) a cardiac event that occurred less than 2 months before the study, (d) hemodynamically significant valvular or congestive heart failure, (e) atrial fibrillation or flutter, (f) high grade heart block or a permanent pacemaker, and (g) did not achieve normal volume status (see below). According to the above criteria, 62 patients were included in this study.

Measurement of BP

BP was measured with a standard mercury sphygmomanometer with the cuffs adapted to arm circumference after the subject had rested in the sitting position for at least 10 min. The systolic and diastolic BP levels were taken as the points of appearance and disappearance of Korotkoff sounds, respectively. The average of three measurements was used for analysis. We used The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High BP (JNC-7) classification as a guide²² and divided the patients into three groups according to the condition of BP: HT group, systolic BP ≥ 140 mmHg and/or diastolic BP ≥ 90 mmHg despite the use of antihypertensive medication; NT group, systolic BP range from 91 to 139 mmHg and diastolic BP range from 61 to 89 mm Hg, irrespective of the use of antihypertensive medication; and HyT group, systolic BP ≤ 90 mmHg and diastolic BP ≤ 60 mmHg. No antihypertensive medication was used in HyT group.

Measurement of Volume Status

The volume status was assessed by a bioimpedance spectroscopy device called BCM (7BJA0202, Fresenius Medical Care, Bad Homburg, Germany). After patients had drained their dialysate and had been lying in a supine position for at least 10 min, standard tetrapolar electrodes were placed on the dorsum of the wrist and anterior aspect of the ankle on the left side of the body. Measurement was performed during a 2-min period, whereby value for overhydration was recorded. Normohydration range for Chinese CAPD patients was defined as −1 to 2.0 L.^23,24

Spectral Analysis of HRV

A dedicated doctor was in charge of all HRV measurements with standardized method. Briefly, all patients were studied in a quiet, comfortable room. Before recording, we asked all patients to breathe at normal frequency and the respiratory rate was determined. The patients were then lying in a supine position for at least 10 min before starting a baseline. The patients underwent ambulatory electrocardiogram (ECG) recording for 5 min. All recordings were analyzed using ECG Explorer 500A pc electrocardiograph (Multiparameter Analysis and Review System).

Frequency domain measures of R–R variability were assessed with power spectral analysis of ECG recordings by a nonparametric (fast Fourier transformation) method for the following frequency bands: the low-frequency energy in the power spectrum of 0.0400–0.1500 Hz (LF, ms²); the high-frequency energy in the power spectrum of 0.1500–0.400 Hz (HF, ms²); the total power spectrum of 0.0033–1.7070 Hz (TP, ms²); and the ratio of low- to high-frequency power (LF/HF ratio). The LF and HF bands were expressed in normalized units (LF normalized units = LF/[TP – VLF]; HF normalized units = HF/[TP – VLF]).¹⁴

Statistical Analysis

Statistical analysis was performed using SPSS for Windows software, version 12.0 (SPSS Inc., Chicago, IL, USA). Measures of HRV were analyzed after logarithmic transformation. All data were expressed as mean ± SD for continuous variables or percentages for categorical variables. To compare different groups, one-way analysis of variance (ANOVA) test was used for continuous variables as appropriate, and the chi-square test was used for categorical variables. All tests were two sided, and a p-value of below 0.05 was considered as significant.

RESULTS

Demographic Data of the Study Population

The demographic characteristics of the study population are shown in Table 1. Of the 62 patients studied, there were 24 male and 38 female patients with an average age of 63 ± 15 years. The mean time on dialysis was 33 ± 21 months. Primary kidney disease included chronic glomerulonephritis (20), diabetes mellitus (9), HT (11), interstitial nephritis (6), polycystic kidney disease (2), other, or unknown causes (14).

Table 1.  Demographic characteristics of the study population.

Number of patients	62
Male/female	24/38
Age (years)	63 ± 15
Weight (kg)	60 ± 11
Height (cm)	158 ± 9
Body mass index (kg/m^2)	24 ± 3
Dialysis duration (months)	33 ± 21
Etiology of uremia (n, %)
Chronic glomerulonephritis	20 (32)
Diabetes mellitus	9 (15)
Hypertension	11 (18)
Interstitial nephritis	6 (10)
Polycystic kidney disease	2 (3)
Other or unknown causes	14 (23)

Comparisons of Demographic Characteristics and Biochemical Variables among the Different BP Groups

The comparisons of demographic characteristics and biochemical variables among the different BP groups are shown in Table 2. There were 20 (32%), 35 (56%), and 7 (11%) patients in the HT, NT, and HyT groups, respectively. The HyT group had a lower serum urea nitrogen as compared with other two groups (p < 0.05), but the differences between the NT group and the HT group were not significant. The HyT group had a lower serum potassium and albumin as compared with the NT group (p < 0.05), but the difference of serum potassium between the other groups was not significant. The gender distribution, age, weight, body mass index, prevalence of diabetics, and serum creatinine among the three groups were not significantly different. No significant difference in the indices of dialysis adequacy [renal Kt/V (rKt/V), peritoneal Kt/V (pKt/V), and total Kt/V (tKt/V)] was observed among the three groups.

Table 2.  Comparisons of demographic characteristics and biochemical variables among the different BP groups.

	HyT group	NT group	HT group	p-Value
Number of patients	7	35	20	NS
Gender (male, %)	3 (43)	14 (40)	7 (35)	NS
Age (years)	66.67 ± 18.17	61.93 ± 12.87	57.08 ± 16.75	NS
Weight (kg)	53.32 ± 10.75	61.01 ± 10.98	59.68 ± 10.24	NS
Height (cm)	153.50 ± 13.31	159.69 ± 7.80	156.46 ± 7.70	NS
BMI (kg/m^2)	22.45 ± 2.10	23.88 ± 3.53	24.30 ± 3.17	NS
Time on PD (months)	32.94 ± 11.24	35.54 ± 24.07	29.50 ± 19.01	NS
Diabetic status (n, %)	0	8 (23)	6 (30)	NS
Serum urea nitrogen (mmol/L)	16.66 ± 5.45,**	22.67 ± 5.18	22.25 ± 6.71	0.021
Creatinine (μmol/L)	749.17 ± 257.96	817.22 ± 332.42	959.83 ± 326.72	NS
Hemoglobin (g/L)	124.17 ± 12.53	120.25 ± 11.62	120.25 ± 13.71	NS
Potassium (mmol/L)	3.70 ± 0.51	4.23 ± 0.57	4.07 ± 0.63	0.043
Albumin (g/L)	37.29 ± 4.79	41.12 ± 3.479	39.85 ± 2.44	0.020
OH (L)	0.62 ± 0.64	1.07 ± 0.73	1.12 ± 0.76	NS
rKT/V	0.32 ± 0.51	0.20 ± 0.36	0.39 ± 0.51	NS
pKT/V	1.39 ± 0.56	1.55 ± 0.27	1.53 ± 0.74	NS
tKT/V	1.71 ± 0.83	1.75 ± 0.45	1.92 ± 0.70	NS
Beta-blockers (n, %)	0	13 (38)	7 (35)	NS
ACE inhibitors (n, %)	0	6 (18)	6 (30)	NS

Notes: BMI, body mass index; OH, overhydration; pKt/V, peritoneal Kt/V; rKt/V, renal Kt/V; tKt/V, total Kt/V (rKt/V + pKt/V); NS, nonsignificant.

*Significantly differed as compared with the NT group.

**Significantly differed as compared with the HT group.

Table 3.  Comparison of HRV indices among the different BP groups.

HRV indices	HyT group	NT group	HT group	p-Value
TP (ms^2)^a	7.82 ± 0.97,**	6.67 ± 1.16	6.23 ± 1.10	<0.05
LF (ms^2)^a	6.26 ± 1.02	5.21 ± 1.35	4.92 ± 1.45	NS
LF [normalized units (nu)]	28.33 ± 17.02**	44.45 ± 20.23	55.69 ± 14.48	0.004
HF (ms^2)^a	6.91 ± 2.02,**	5.17 ± 1.70	4.34 ± 1.84	<0.05
HF [normalized units (nu)]	48.33 ± 8.29**	41.55 ± 14.81	31.23 ±12.03	0.014
LF/HF ratio	0.66 ± 0.55**	1.40 ± 1.36	2.06 ± 0.92	0.021

Notes: ^aBecause the indices of HRV were not normally distributed, log transformation was performed, and the antilog of the mean is presented.

*Significantly differed as compared with the NT group.

**Significantly differed as compared with the HT group.

Comparisons of HRV among the Different BP Groups

The TP and HF in the HyT group were significantly higher than that of the other groups (p < 0.05), but there was no significant difference between the NT and the HT groups (Table 3). The normalized HF was significantly higher in the HyT group as compared with the HT group (p < 0.05), but the difference in the normalized HF between the other groups was not significant. The normalized LF and LF/HF ratio were significantly lower in the HyT group as compared with the HT group (p < 0.05), but the differences in the normalized LF and LF/HF ratio between the other groups were not significant. The LF among the three groups was not significantly different.

DISCUSSION

HRV is influenced by the ANS. Spectral analysis of HRV can partially distinguish parasympathetic from sympathetic influences on the heart. The HF mostly reflects fast breathing-related beat-to-beat changes and is a measure of parasympathetic involvement in circulatory control. The LF is affected by the baroreceptor reflex and is thought to reflect sympathetic and parasympathetic involvement. The TP can be estimated with the sum of the frequencies, and the LF/HF ratio is considered by some investigators to mirror sympathovagal balance or to reflect sympathetic modulations. When expressed in normalized units, LF is considered to be a quantitative index of sympathetic activity.¹⁴

Some research groups demonstrated reduced LF and HF in PD and HD patients as compared with healthy controls.^25,26 However, other groups have reported reduced HF, increased LF, and LF/HF in HD patients.²⁷ Indeed, Giordano et al.²⁷ have suggested that this increased sympathetic activity may predispose to sudden cardiac death in waiting list renal transplant patients. So, early studies suggested ANS abnormality is not uncommon in chronic renal failure. As ANS plays an important role in BP regulation and BP abnormality is very common in dialysis patients, we hypothesized that ANS dysfunction in PD patients has lots of modality and patients in different BP status have different sympathovagal balance.

The most interesting result in our research was that the hypotensive euvolemic patients had a significantly higher HF, normalized HF and lower normalized LF, and LF/HF ratio, which suggested the presence of disturbed equilibrium, with a predominance of an increased parasympathetic activity and blunted sympathetic activity. Pelosi et al.¹⁸ reported in late dialysis that a decreased LF/HF ratio in hypotensive patients which indicated the activation of sympathetic tone discovered in normal blood pressure HD patients was impaired in hypotensive HD patients, therefore contributing to the onset of the hypotensive crisis in these HD patients especially in late stage of dialysis. Sato et al.¹⁹ also reported that persistent HD-induced HyT patients had a reduced LF, HF, and LF/HF ratio as compared with the normotensive group, which suggested that HD-induced HyT might be related to autonomic insufficiency. Takahashi et al.²⁰ showed that HD patients with persistent HyT had a continuously decreased LF/HF even after standing as compared with HT or normotensive patients, which indicated that impairment of the ANS exists in persistent hypotensive HD patients. Those studies which investigated the relationship between HRV and HyT in ESRD have been largely restricted to studies in HD patients. Our study found that PD patients also have an impaired sympathetic function, which was consistent with the result that normalized LF and LF/HF ratio were usually low in hypotensive patients as compared with non-hypotensive patients. HRV changes observed in this study are in line with what might be expected from normal physiology. As BP is a function of vascular resistance and cardiac output, which are controlled by the ANS, we speculated that decreased cardiac output, vascular resistance, and venous return volume resulted from a decreased heart rate and myocardial contractility arising from abnormal ANS with blunted sympathetic activity may contribute to the occurrence of HyT in these patients.

Sympathetic overactivity has long been considered as the pathogenesis of HT in patients with renal impairment. Several data provided evidence that increase in BP might be due to sympathetic overactivity, which is triggered by afferent signals emanating from the kidney and resetting sympathetic tone by stimulation of hypothalamic centers.¹⁷ In Mohamed’ study, reduced vagal activity (HF) and higher sympathetic function (LF/HF ratio) were associated with higher values of BPs putting subjects with such a trend at a higher risk of developing HT. But in our research, although normalized LF and LF/HF ratio were higher in the HT group as compared with the HyT group, there were no significant differences between the HT group and the NT group; this lack of statistical significance might be due to our small sample size and warrants further studies.

In the Alexander study, a prospective study in chronic PD, a strong association between HyT and death from all causes, including cancer, was observed. This suggests that HyT is an indicator of poor general health. In our study, we found serum urea nitrogen, potassium, and albumin level were lower in HyT patients that may reflect poor general health in hypotensives.

There were 14 diabetic patients in the research (9 had diabetic nephropathy as primary kidney disease, and the other 5 developed diabetes in the course of PD). Subjects with diabetes are known to be at high risk of developing autonomic dysfunction.²⁸ But in our study, there were no diabetes in HyT group and the diabetes distribution between the other two groups was not significantly different. Thus, this confirms that autonomic dysfunction is already present in different BP subtypes of CAPD patients with or without diabetes.

Antihypertensive medications have known effects on HRV: beta-blockers and angiotensin converting enzyme inhibitors (ACEIs) suppress sympathetic activation that leads to HF increase.^29,30 Although the HyT group was not receiving antihypertensive treatment, the HF of the HyT group was still higher than the other groups. Thus, the effect of antihypertensive treatment could be excluded.

There are several limitations in this study. This is a cross-sectional study. Inferences regarding associations should be evaluated cautiously, owing to the possibility of reverse causality. However, the interpretation that the HRV imbalances in different BP subtypes and hypotensives have an impaired sympathetic function and a relatively higher parasympathetic activity is logically made in this study, because it is in line with what might be expected from normal physiology. It is also recognized that the sample size of this study is relatively small, resulting in some wide confidence intervals. Large-scale studies of HRV and BP in the dialysis population will be needed in the future, particularly cohort studies and randomized controlled trials of medications that alter HRV indices.

In conclusion, this study showed that autonomic nerve function assessed by HRV in euvolemic CAPD patients was associated with different status of BP. Patients with HyT had a blunted sympathetic activity and a relatively higher parasympathetic activity.

ACKNOWLEDGMENTS

This study was partly supported by grants from the National Natural Science Foundation of China (Project 30900681), Beijing Municipal Science & Technology Commission (D09050704310905), and the Fund of Peking University Third Hospital (YZZ 08-5-28 and 76496-02).

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.