Effects of nutritional parameters on nocturnal blood pressure in patients undergoing hemodialysis

Abstract

Background: Malnutrition is a common problem in uremic patients. It is unclear whether there is an association between the degree of malnutrition and 24-h ambulatory blood pressure patterns in patients undergoing hemodialysis. In the present study, we observed the relationship between the degree of malnutrition and deterioration of the rhythm of diurnal blood pressure, which are both risk factors for cardiovascular morbidity--mortality and associated with hypervolemia. Method: We observed 148 patients undergoing hemodialysis in the Nephrology Department of Celal Bayar University Hospital. All cases were assessed for body weight alterations, dietary food intake, gastrointestinal symptoms, loss of subcutaneous fat and muscle tissue, presence and severity of comorbidities, functional capacity (subjective global assessment), and anthropometric indices. Ambulatory blood pressure measurements were performed for all cases on the day between the two hemodialysis sessions. Results: We found that the circadian blood pressure rhythm deteriorated in patients with a high-malnutrition score, and that malnutrition was more common and severe in those subjects with the non-dipper and reverse-dipper blood pressure patterns. Malnutrition score was positively correlated with the nighttime systolic and nighttime mean blood pressures and mean 24-h arterial blood pressure (all p ≤ 0.01). We identified a positive correlation between malnutrition score and the reduction in serum albumin and anthropometric indices. Conclusion: This is the first study to demonstrate an association between malnutrition and deterioration in the circadian blood pressure rhythm in a hemodialysis population. Nutritional disturbance is associated with an increase in night-time blood pressure. Low serum albumin levels and hypervolemia may contribute this situation.

• Hemodialysis
• hypertension
• hypervolemia
• malnutrition
• nocturnal blood pressure

Introduction

Malnutrition is a common problem in uremic patients.1 Many studies have shown that malnutrition and related factors increase morbidity and mortality rates in patients with chronic renal failure.2–4 Malnutrition is a nontraditional risk factor that increases cardiovascular mortality.5

Disappearance of the difference between the mean day and night blood pressures in patients with essential hypertension is associated with target organ damage and worsening of cardiovascular parameters.6 Furthermore, the nondipper (ND) blood pressure pattern is an independent risk factor for hypertensive target organ damage and left ventricular hypertrophy (LVH) in patients undergoing hemodialysis (HD).7 Accumulating evidence indicates that malnutrition affects the development of LVH.8 Malnutrition is closely related with volume overload in patients with end stage renal disease.9,10 We previously reported that hypervolemic cardiovascular markers (increased left atrial volume and decreased vena cava inferior collapsibility index) and LVH (increased left ventricular mass index) were associated with deterioration of nutritional status in HD patients.11 Hypervolemia is considered as the main causative factor for increased blood pressures at nighttime.7 It is unclear whether there is an association between the degree of malnutrition and 24 h ambulatory blood pressure patterns in patients undergoing HD.

In the present study, we observed the relationship between the degree of malnutrition and deterioration of the rhythm of diurnal blood pressure, which are both risk factors for cardiovascular morbidity--mortality and associated with hypervolemia. This is the first study to investigate the possible association between malnutrition and deterioration of the circadian rhythm of the blood pressure in a HD population.

Materials and methods

Patients and study design

We observed 167 patients undergoing HD in the Nephrology Department of Celal Bayar University Hospital (Manisa, Turkey). All patients were receiving bicarbonate HD treatment 3 times/week. The exclusion criteria were a history of chronic liver disease, malignant disease, myocardial infarction, acute disease in the last 3 months, and HD treatment for <1 year. Fifteen patients were excluded based on these criteria, and one patient died from a cerebrovascular event. Three patients were excluded because they would not adapt to the study protocol. Thus, 148 patients were enrolled. This study was approved by the local ethics committee of Celal Bayar University.

Malnutrition evaluation

All cases were assessed for body weight alterations, dietary food intake, gastrointestinal symptoms, loss of subcutaneous fat and muscle tissue, presence and severity of comorbidities, and functional capacity. Each parameter was evaluated between 1 and 5 points; thus, a total malnutrition score was calculated for each case (subjective global assessment).12 The anthropometric characteristics of the patients were recorded using standard techniques. Body weight and height were measured, and the body mass index was calculated as body weight divided by the square of the height. Mid-arm circumference (MAC) was measured on the nondominant arm without a fistula. Triceps and biceps skinfold thicknesses (TSF and BSF, respectively) were measured with calipers. Mid-arm muscle circumference (MAMC) values were calculated as follows: MAC – (3.1415 × BSF).12 Basic hematological and biochemical parameters were assessed at the second dialysis session the week before HD.

Ambulatory blood pressure monitoring (ABPM)

ABPM was performed for all cases on the day between the two HD sessions. The same device (Cardio Navigator, Tracker 2 NIBP; Raynaulds Medical Ltd., Hertford, England, UK) was used to monitor all subjects. Daytime (08:00–20:00) blood pressure measurements were performed every 20 min, and nighttime (20:00–08:00) blood pressure measurements were performed every 30 min. Physical activity was not restricted during ABPM. Daytime and nighttime mean systolic and diastolic blood pressures and mean 24-h systolic and diastolic blood pressures were recorded. The nighttime/daytime mean blood pressure ratio was calculated. The patients were grouped as dippers (Ds), NDs, or reverse-dippers (RDs) according to the ratios (Ds, 0.8–0.9; NDs, 0.9–1.0; and RDs, >1).13 Standard deviations of the daytime and nighttime systolic and diastolic blood pressures were estimated. The ratio of the standard deviation of systolic and diastolic blood pressure within a time period and mean systolic and diastolic blood pressures during the same time period was defined as blood pressure variability. We did not change the medications of the study population during this study.

Statistical analysis

A statistical analysis was performed with SPSS 19.0 (Chicago, IL). All data are expressed as the mean ± standard deviation. Bivariate Pearson’s correlation tests and logistic regression analyses were used to assess the association between the malnutrition index and blood pressure measurements. For the multivariate analysis, the possible factors identified with univariate analyses were further analyzed into the logistic regression model to determine independent predictors of malnutrition status. Hosmer--Lemeshow goodness of fit statistics were used to access model fit. A 5% type 1 error level was used to infer statistical significance. Differences between the ABPM subgroups were assessed with one-way analysis of variance and Kruskal–Wallis tests. The Mann–Whitney U test was performed to determine differences between the groups. A p value <0.05 was considered significant.

Results

The malnutrition scores of the patients in the RD group were significantly higher than those of the patients in the ND (p = 0.021) and D (p = 0.003) groups. Table 1 shows the demographic and anthropometric characteristics and the hematological and biochemical parameters of the ABPM subgroup cases, as well as the bivariate and multivariate associations between the clinical parameters and malnutrition index. Combined univariate and multivariate analyses revealed a negative correlation between malnutrition score and TSF, MAMC, and serum albumin levels and a positive correlation between malnutrition score and cardiothoracic index.

Table 1. Characteristics of the cases, analysis of differences between subgroups and association between clinical parameters and malnutrition scores.

	Dipper (n = 53)	Non-dipper (n = 49)	Reverse-dipper (n = 46)	p^a	Mean ± SD	p^b	r^2	p^c	RR^c (%95 CI)
Age (years)	59 ± 18	56 ± 16	60 ± 12	>0.05	58 ± 15	<0.01	0.26	>0.05	1.21 (0.73–3.14)
BMI (kg/m^2)	23.6 ± 4.0	23.4 ± 4.7	24.7 ± 4.5	>0.05	23.8 ± 4.5	>0.05	−0.08	>0.05	0.94 (0.84–1.06)
BSF (cm)	0.39 ± 0.2	0.38 ± 0.2	0.41 ± 0.2	>0.05	0.38 ± 0.2	>0.05	−0.09	>0.05	0.69 (0.29–1.63)
TSF (cm)	1.02 ± 0.4*	0.67 ± 0.4	0.63 ± 0.3	<0.01	0.73 ± 0.4	<0.01	−0.36	0.03	0.47 (0.25–0.90)
MAC (cm)	26.7 ± 3.4	25.7 ± 3.3	26.1 ± 3.1	>0.05	26 ± 3.2	0.04	−0.21	>0.05	0.55 (0.22–1.39)
MAMC (cm)	23.4 ± 2.9	23.6 ± 3.2	23.1 ± 2.8	>0.05	22.4 ± 3	0.03	−0.31	0.01	0.39 (0.19–0.82)
Hemoglobin (g/dL)	11.6 ± 1.3*	10.8 ± 1.5	10.9 ± 1.7	0.02	11.1 ± 9.3	>0.05	−0.18	>0.05	0.94 (0.84–1.06)
Creatinine (mg/dL)	6.4 ± 3.0	6.5 ± 2.7	5.9 ± 2.8	>0.05	6.2 ± 2.8	>0.05	−0.15	>0.05	0.79 (0.44–1.92)
Albumin (g/dL)	3.4 ± 0.6	3.3 ± 0.7	3.0 ± 0.5*	0.02	3.2 ± 0.7	<0.01	−0.29	<0.01	0.43 (0.19–0.77)
Total protein (g/dL)	6.4 ± 0.9	6.4 ± 0.8	6.1 ± 0.7	>0.05	6.4 ± 0.8	>0.05	−0.19	>0.05	0.64 (0.51–1.84)
Cholesterol (mg/dL)	196 ± 71	171 ± 47	159 ± 37	>0.05	172 ± 51	>0.05	−0.11	>0.05	0.91 (0.62–3.12)
Triglyceride (mg/dL)	178 ± 105	187 ± 68	162 ± 64	>0.05	156 ± 77	>0.05	−0.13	>0.05	0.88 (0.59–2.87)
Cardiothoracic index	0.47 ± 0.1*	0.54 ± 0.1	0.55 ± 0.1	0.01	0.52 ± 0.1	>0.05	0.18	0.04	1.82 (1.23–3.52)
HD duration (months)	28 ± 12	32 ± 21	31 ± 24	>0.05	31 ± 16	>0.05	0.09	>0.05	1.29 (0.97–1.48)
Malnutrition score (mean)	16.2 ± 6.5	20.1 ± 5.4	24.9 ± 5.6	0.01	–	–	–	--	--

^aOne-way ANOVA.

^bPearson correlation test.

^cLogistic regression analysis.

*Subgroup that make the difference statistically (Mann–Whitney test).

BMI, Body mass index; BSF, Biceps skinfold; TSF, Triceps skinfold; MAC, Mid-arm circumferences; MAMC, Mid-arm muscle circumferences.

Table 2 shows the ambulatory blood pressure and multivariate and univariate associations between the malnutrition scores and mean ambulatory blood pressures of the subjects. Serum albumin was significantly lower in the RD group compared with the D and ND groups. Serum albumin was negatively correlated with HD duration (p = 0.04, r = −0.21), nighttime systolic blood pressure (p = 0.005, r = −0.28), nighttime mean arterial blood pressure (p = 0.04, r = −0.21), nighttime/daytime mean systolic blood pressure ratio (p = 0.016, r = −0.36), nighttime/daytime mean diastolic blood pressure (p = 0.027, r = −0.32), and nighttime/daytime-mean blood pressure (p = 0.012, r = −0.41). Albumin was not correlated with other ABPM parameters.

Table 2. Blood pressure measurements for 24 h of the cases and their univariate and multivariate relationship with malnutrition score.

	Mean ± SD	p^a	r^a	p^b	RR^2 (%95 CI)
Daytime-mean systolic blood pressure (mm Hg)	127 ± 20	>0.05	0.16	>0.05	1.19 (0.84–1.46)
Daytime-mean diastolic blood pressure (mm Hg)	70 ± 17	> 0.05	0.17	>0.05	1.06 (0.47–2.18)
Daytime-mean arterial blood pressure (mm Hg)	91 ± 16	>0.05	0.12	>0.05	1.21 (0.67–1.94)
Nighttime-mean systolic blood pressure (mm Hg)	129 ± 25	0.01	0.25	<0.001	2.14 (1.34–4.36)
Nighttime-mean diastolic blood pressure (mm Hg)	72 ± 18	>0.05	0.19	0.03	1.69 (1.22–3.82)
Nighttime-mean arterial blood pressure (mm Hg)	89 ± 18	0.03	0.22	0.01	1.92 (0.92–3.79)
Mean systolic blood pressure of 24 h (mm Hg)	128 ± 21	>0.05	0.17	0.04	1.47 (0.88–2.14)
Mean diastolic blood pressure of 24 h (mm Hg)	74 ± 21	>0.05	0.18	>0.05	1.03 (0.59–1.73)
Mean arterial blood pressure of 24 h (mm Hg)	93 ± 19	0.03	0.21	<0.001	1.81 (1.24–3.95)
Variability of daytime-systolic blood pressure	9.8 ± 3.0	>0.05	0.14	>0.05	1.28 (0.60–2.14)
Variability of daytime-diastolic blood pressure	11.7 ± 4.2	>0.05	0.05	>0.05	1.15 (0.54–2.63)
Variability of nighttime-systolic blood pressure	7.7 ± 2.1	<0.001	0.41	0.01	2.03 (1.54–4.23)
Variability of nighttime-diastolic blood pressure	9.4 ± 2.9	>0.05	0.19	>0.05	1.12 (1.03–2.02)

^aPearson correlation test.

^bLogistic regression analysis.

The malnutrition index was significantly correlated with the mean nighttime/daytime systolic blood pressure ratio (p < 0.001, r = 0.42) and mean nighttime/daytime diastolic blood pressure ratio (p = 0.002, r = 0.40). Patients were grouped either into the high malnutrition score group (malnutrition score, >20) or low malnutrition score group (malnutrition score, ≤20). A total of 13 of the 67 cases with a high malnutrition score were in the D subgroup, whereas 23 were in the ND subgroup and 31 were in the RD subgroup. A total of 81 cases were in the low malnutrition score group, while the D, ND, and RD subgroups consisted of 40, 26, and 15 cases, respectively. The D pattern was significantly more frequent in the low malnutrition score group (p = 0.02). RD was the most frequent pattern in the high malnutrition score group. Mean nighttime systolic blood pressure, mean nighttime blood pressure, and mean 24-h blood pressure among the cases in the high malnutrition score group increased significantly compared to subjects with a low malnutrition score (Table 3). The malnutrition score increased in those patients whose ratios of nighttime/daytime mean systolic blood pressure (p = 0.001) and nighttime/daytime mean blood pressure (p = 0.023) were >1.

Table 3. Comparison of ABPM in subgroups with high and low malnutrition scores.

	High score (>20)	Low score (≤20)	p^a
Daytime-mean systolic blood pressure (mm Hg)	131 ± 23	127 ± 18	>0.05
Daytime-mean diastolic blood pressure (mm Hg)	73 ± 18	68 ± 14	>0.05
Daytime-mean arterial blood pressure (mm Hg)	92 ± 19	90 ± 16	>0.05
Nighttime-mean systolic blood pressure (mm Hg)	136 ± 27	121 ± 22	0.02
Nighttime-mean diastolic blood pressure (mm Hg)	75 ± 19	69 ± 16	>0.05
Nighttime-mean arterial blood pressure (mm Hg)	96 ± 20	79 ± 15	<0.01
Mean systolic blood pressure of 24 h (mm Hg)	131 ± 23	126 ± 19	>0.05
Mean diastolic blood pressure of 24 h (mm Hg)	77 ± 26	71 ± 15	>0.05
Mean arterial blood pressure of 24 h (mm Hg)	97 ± 21	86 ± 16	0.02

^aStudent’s t-test.

Discussion

We found that the circadian blood pressure rhythm deteriorated in patients with a high malnutrition score, and that malnutrition was more common and severe in those subjects with the ND and RD blood pressure patterns. Malnutrition score was positively and strongly correlated with the nighttime systolic and nighttime mean blood pressures and mean 24 h arterial blood pressure.

Hypertension is common in patients undergoing HD. Blood pressure is usually suboptimal,14,15 which contributes to cardiovascular morbidity in these patients.16 The issue of which measurement time period better reflects the burden of hypertension and correlates with cardiovascular outcomes is unclear. Autonomic dysfunction and hypervolemia can cause a loss of the circadian rhythm of blood pressure in patients undergoing HD.13,17 These patients lack the 15–20% decrease in blood pressure seen in normal individuals at nighttime. Therefore, it is hypothesized that these patients are exposed to higher blood pressures over 24 h and more frequent target organ damage. Related studies have revealed that nighttime blood pressure is associated with LVH 18–21 and is a predictor of negative cardiovascular outcomes in patients undergoing HD.22,23 One study reported that cardiovascular survival was worse in patients with the ND blood pressure pattern compared to patients with the D blood pressure pattern, and it stated that an elevated 24 h pulse pressure and nighttime systolic blood pressure were associated with cardiovascular mortality.22 Another prospective study found higher rates of cardiovascular mortality and coronary artery stenosis in ND patients compared to D patients undergoing HD.23 Based on these studies and our results, we suggest that a deteriorated diurnal blood pressure rhythm parallels the malnutrition status in patients undergoing HD. Interestingly, which is an indicator of hypervolemia, cardiothoracic index of the patients with ND and RD pattern was significantly higher than patients with D pattern.

In our study, variability in nighttime systolic blood pressure was associated with the malnutrition index. This conclusion is important, as it shows the relationship between two entities that are both associated with mortality and morbidity. In one study, Fisher’s rats fed a low protein diet had an elevated mean arterial pressure, mean arterial pressure variability, heart rate, and heart rate variability compared to control rats on a normal diet.24 Tozawa et al. reported that blood pressure variability was a poor prognostic risk factor for cardiovascular disease in patients with end-stage renal failure.25 Twenty-four-hour blood pressure variability is an independent risk factor for cardiovascular mortality26 and is associated with end organ damage.27

Several studies have reported that anthropometric measurements and serum albumin levels decrease in parallel with malnutrition.12,28 This suggests that malnutrition reduces the TSF, MAC, and MAMC, as well as serum albumin levels and lymphocyte counts. We identified a positive correlation between malnutrition score and the reduction in serum albumin and anthropometric indices (TSF and MAMC). Reduced serum albumin levels were also associated with an elevated nighttime blood pressure, suggesting that deterioration in the circadian blood pressure rhythm is associated with malnutrition.

We found that subjects with the RD blood pressure pattern had greater malnutrition scores compared to ND subjects. Furthermore, the malnutrition scores of those subjects with the ND blood pressure pattern were significantly higher than those with the D pattern. Interestingly, we found that ND and RD patterns were more frequent in patients with a high malnutrition index and nighttime systolic and nighttime mean blood pressures, and that the mean 24 h blood pressure of these patients were significantly elevated compared to patients with a lower malnutrition score. Although malnutrition increases cardiovascular morbidity and mortality in patients undergoing HD, its association with ambulatory blood pressure parameters has not been reported. Previously protein and energy intake in patients undergoing peritoneal dialysis with volume overload was significantly reduced in a report by Wang et al., indicating that malnutrition scores are associated with volume overload.29 Another study of peritoneal dialysis reported that nutrition improved significantly following regular reductions in volume overload over a 9-month period but further deteriorated in patients whom volume control failed.9

Clinical measurements related to malnutrition, inflammation, and atherosclerosis (e.g., serum albumin, C-reactive protein [CRP], and carotid intima media thickness) were associated with symptoms of volume overload, as detected by echocardiography and bioimpedance analyses in a study of 95 PD patients.10 Pro-brain natriuretic peptide, an indicator of volume overload, is correlated with troponin T, homocysteine, and CRP levels and malnutrition inflammation scores in patients undergoing HD.30 These findings are important, as they indicate an association between hypervolemia and malnutrition. Besides autonomic dysfunction, hypervolemia also contributes to deterioration of the diurnal blood pressure balance. Therefore, hypervolemia may explain the underlying pathophysiological mechanism of deterioration in both nutritional status and blood pressure circadian rhythm.

In conclusion, this is the first study demonstrated an association between malnutrition and deterioration in the circadian blood pressure rhythm in an HD population. Still, larger prospective studies using advanced biochemical markers and which consider the effects of the renin--angiotensin--aldosterone system and sympathetic overactivity on ambulatory blood pressure measurements are needed.

Declaration of interest

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

Acknowledgements

The English in this document has been checked by at least two professional editors, both native speakers of English. For a certificate, please see: http://www.textcheck.com/certificate/20CD1L.