Abstract
Aims: The phase angle (PhA) derived from a bioelectrical impedance analysis (BIA) is a risk factor for cardiovascular disease (CVD). The present study explored the relationship between PhA and the progression of subclinical atherosclerosis in asymptomatic adults.
Methods: Two cross-sectional studies were performed on 15579 participants who underwent carotid ultrasound testing and a BIA as well as 8228 participants who underwent brachial ankle pulse wave velocity (baPWV) testing and a BIA. We also conducted a longitudinal study in participants without CVD and carotid atherosclerosis (CAS) at baseline who underwent carotid ultrasound ≥ 2 times (n = 2680) or baPWV testing [≥ 2 times] (n = 1775). CAS and the brachial ankle pulse wave velocity (baPWV) were selected as the subclinical atherosclerosis markers.
Results: In the cross-sectional studies, participants with CAS (5.43±0.60° vs. 5.73±0.61°,P＜0.001) or elevated baPWV (5.38±0.62° vs. 5.74±0.59°,P＜0.001) had lower PhA values than controls. Furthermore, the PhA value was independently and inversely correlated with CAS (adjusted odds ratio [OR] = 0.41, 95% confidence interval [CI] 0.37-0.46,P＜0.001) and elevated baPWV (adjusted OR = 0.45, 95% CI 0.39-0.52,P＜0.001). Restricted cubic spline curve analyses indicated dose-response associations of PhA values with subclinical atherosclerosis. In the longitudinal study, high PhA values at baseline decreased the risk of incident CAS (adjusted hazard ratio = 0.44, 95% CI 0.36-0.54,P＜0.001). Multivariate linear regression analyses showed that the PhA was negatively associated with absolute or relative annual changes in baPWV.
Conclusion: The PhA value is significantly associated with the progression of subclinical atherosclerosis, indicating that PhA may serve as a noninvasive marker for monitoring subclinical atherosclerosis in a primary prevention setting.
Keywords: Phase angle, Subclinical atherosclerosis, Carotid atherosclerosis, Brachial ankle pulse wave velocity
Introduction
Cardiovascular disease (CVD) is one of the four major chronic noncommunicable diseases, representing a large burden of morbidity and mortality worldwide, and this burden has increased over the past few decades. Between 1990 and 2016, the number of deaths for which CVD was the underlying cause increased from 2.51 million to 3.97 million 1) . Subclinical atherosclerosis occurring prior to the clinical manifestation of CVD and can be assessed by measuring the carotid intima-media thickness (CIMT), carotid plaque, and brachial-ankle pulse wave velocity (baPWV) 2 , 3) . Early detection and intervention of risk factors for subclinical atherosclerosis seem to be key to initiating timely preventive or delayed progression of CVD events in the primary prevention setting.
The phase angle (PhA) is a simple and noninvasive method for evaluating cellular health, particularly the cell membrane integrity and function 4) , serving as a marker of nutritional status 5) . Recently, the correlation between PhA and CVD has attracted widespread attention in clinical settings. Data from an apparently healthy Danish population showed that a low PhA increased the risk of incident CVD, particularly among women 6) . A cross-sectional study indicated that PhA values were negatively correlated with the first cardiovascular event in men 7) . In addition to CVD, the relationship between the PhA and subclinical atherosclerosis has been the focus of several studies. A cross-sectional study including 344 patients after kidney transplantation reported that the whole-body PhA is significantly associated with arterial stiffness parameters, including the brachial-ankle and carotid-femoral pulse wave velocities (baPWV and cfPWV, respectively) 8) . Furthermore, another cross-sectional study including 67 subjects with type 1 diabetes mellitus, 232 subjects with type 2 diabetes mellitus, and 153 subjects without diabetes mellitus found that the PhA was negatively associated with CIMT and cfPWV but was not significantly associated with CIMT after adjusting for confounding factors 9) . Nevertheless, the results of these studies may not be applicable to the general population. To our knowledge, no previous study has investigated the risk of progression of subclinical atherosclerosis with the PhA.
Considering the limitations of previous studies, the present study 1) investigated the association between the PhA and subclinical atherosclerosis in Chinese adults free from clinical CVD using two cross-sectional studies and 2) explored the effect of the PhA on the risk of the progression of subclinical atherosclerosis.
Materials and Methods
Study Population
This study was designed to incorporate both cross-sectional and longitudinal analyses.
First, 16171 subjects who had undergone checkups including carotid ultrasound and bioelectrical impedance testing at the Health Promotion Center of the First Affiliated Hospital of Nanjing Medical University for health examinations between March 2018 and April 2024 were enrolled in a cross-sectional study. A total of 8556 subjects who had undergone checkups, including baPWV and bioelectrical impedance testing, at the same center for health examinations between March 2018 and April 2024 were enrolled in another cross-sectional study. We excluded participants with a history of CVD, cerebrovascular accident, cancer, or chronic renal failure. We further excluded participants with missing body mass index (BMI) and lipid profile data. A total of 15579 participants in a cross-sectional study and 8228 participants in another cross-sectional study were included in the final analysis.
Second, we analyzed longitudinal study data from 3756 subjects who underwent at least two health checkup examinations, including carotid ultrasound, from the same center as in the cross-sectional study from July 2018 to April 2023. We excluded participants who had carotid atherosclerosis on ultrasound imaging at baseline. Participants who underwent at least two screening examinations within six months or had related missing data were also excluded. Other exclusion criteria were similar to those used in the cross-sectional study. Ultimately, 2680 participants were included in the final analysis.
This study was approved by the Human Research Ethics Committee of the First Affiliated Hospital of Nanjing Medical University. Our study was designed to retrospectively collect available data from participants’ physical examination results. The requirement for informed consent was waived by the ethics committee. The flowchart is presented in Supplementary Fig.1 .
Supplementary Fig.1. The flowchart of the study population.
(A) the cross-sectional studies; (B) the longitudinal study
Data Collection
Participant information, including age, sex, smoking, and medical history, was collected by physicians in the health promotion center. The BMI was calculated using height and weight. We used a digital sphygmomanometer to measure the blood pressure of all participants while they were in a sitting position. Lipid profiles, blood glucose, and serum uric acid levels were determined in fasting blood samples using an automatic biochemical analyzer (AU5800; Beckman-Coulter Co., Brea, CA, USA). Routine blood examinations were performed using an automated blood cell counter (Sysmex XN-10; Sysmex Corporation, Kobe, Japan). Glycosylated hemoglobin A1c (HbA1c) was measured using high-performance liquid chromatography (Bio-Rad D-100; Bio-Rad, Hercules ,USA).
Measurement of the PhA
We conducted a bioelectrical impedance analysis (BIA) to measure the PhA using an InBody 770 body composition analyzer (Biospace, Seoul, Korea) after overnight fasting. The analyzer used an alternating current of 80 mA and assessed the PhA at 50 kHz. The formula for calculating PhA was as follows: PhA = arctan(Xc/R) × 180/π (where Xc is the reactance, and R is the body resistance).
Assessing Carotid Atherosclerosis (CAS)
A carotid B-mode ultrasonographic examination was performed by experienced ultrasonographers (Logiq 7; GE Medical System, Milwaukee, WI, USA). Details of the measurement and definitions of the thickened CIMT and carotid plaque have been described in previous studies 10) . CAS was defined in cases with a CIMT of ≥ 1.0 mm or plaque formation.
BaPWV Measurement
Specially trained nurses used the VP-1000 automated PWV/ABI analyzer to measure the baPWV of participants, according to the manufacturer’s recommendations. After resting for at least 5 min, the participants were measured in the supine position and remained quiet during the measurement. Details of the measurement and definition of elevated baPWV have been provided in previous studies 11) . The mean values of the left and right baPWV were used for further analyses. A baPWV greater than the 75th percentile (1,492 cm/s for men and 1515 cm/s for women in the cross-sectional study) was defined as an elevated baPWV. In the longitudinal study, the baPWV was measured at baseline and at the follow-up visit in 1775 participants. The progression of the baPWV was assessed using relative annual change and absolute annual change-based data from previous studies 12 , 13) . The absolute baPWV change was calculated as the baPWV value at the last follow-up visit minus the baPWV value at baseline. Relative baPWV change was calculated as the absolute baPWV change divided by the baseline baPWV value. The absolute annual change in baPWV was calculated as the absolute baPWV change divided by the time between the baseline and last visit. The relative annual change in baPWV was calculated as the relative baPWV change divided by the time between the baseline and last visit.
Statistical Analyses
Statistical analyses were conducted using the SPSS 23.0 statistics (IBM SPSS Inc., Chicago, IL, USA) and R version 3.3.0 software programs (R Core Team, Vienna, Austria). Data were expressed as the means with standard deviations or numbers (n) with percentages (%) for continuous and categorical variables, respectively. Comparisons between two groups were made using Student’s t-test, while a one-way analysis of variance was used for differences among three or more groups, and Pearson’s chi-squared test was applied where appropriate. A logistic regression analysis was used to evaluate the association of the PhA with subclinical atherosclerosis, adjusted for the age, sex, smoking, blood cell count (WBCC), neutrophil count, monocyte count, BMI, diabetes, hypertension, dyslipidemia, and serum uric acid levels. RCS curves adjusted for the same variables as in the adjusted model were constructed to evaluate the potential linear or nonlinear dose-response associations. The association of the PhA with the progression of subclinical atherosclerosis was investigated using Cox proportional hazards regression models. Multivariate linear regression models were used to estimate the association between the PhA and the relative annual change and absolute annual change in the baPWV. A 2-tailed P value of ＜0.05 was considered statistically significant.
Results
Baseline Characteristics in the Cross-Sectional Studies
Table 1 presents a detailed comparison between participants with and without CAS. Compared with participants without CAS, participants with CAS were more often older and men and had a higher prevalence of smoking, diabetes, and hypertension as well as a higher WBCC. The levels of total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), fasting plasma glucose (FPG), HbA1c, and uric acid were significantly increased in participants with CAS. There were no marked differences in the BMI, monocyte count, neutrophil count, or uric acid levels between the two groups. Notably, the PhA was lower in participants with CAS than in those without CAS (5.43±0.60° vs. 5.73±0.61°, P＜0.001).
Table 1. Baseline patient characteristics based on CAS presence in the cross-sectional study.
| Without CAS (n = 9728) | CAS (n = 5851) | P value | |
|---|---|---|---|
| Age (years) | 42.47±8.89 | 54.38±9.27 | ＜0.001 |
| Sex (male, n, %) | 6167(63.4) | 3863(66.0) | 0.001 |
| Smoking (n, %) | 1706(17.5) | 1354(23.1) | ＜0.001 |
| Diabetes (n, %) | 484(5.0) | 855(14.6) | ＜0.001 |
| Hypertension (n, %) | 816(8.4) | 1395(23.8) | ＜0.001 |
| Dyslipidemia (n, %) | 3185(32.7) | 2326(39.8) | ＜0.001 |
| BMI (kg/m2) | 24.89±3.49 | 24.89±3.04 | 0.906 |
| SBP (mmHg) | 123.15±15.84 | 130.93±18.00 | ＜0.001 |
| DBP (mmHg) | 77.00±11.46 | 79.89±11.27 | ＜0.001 |
| WBCC (×109/L) | 5.75±1.57 | 5.81±1.43 | 0.011 |
| Monocyte count (×109/L) | 0.40±0.12 | 0.40±0.13 | 0.202 |
| Neutrophil count (×109/L) | 3.32±1.17 | 3.34±1.09 | 0.210 |
| FPG (mmol/L) | 5.24±1.02 | 5.66±1.47 | ＜0.001 |
| HbA1c (%) | 5.42±0.59 | 5.75±0.87 | ＜0.001 |
| TC (mmol/L) | 5.13±0.95 | 5.29±1.06 | ＜0.001 |
| TG (mmol/L) | 1.71±1.39 | 1.76±1.32 | 0.013 |
| LDL-C (mmol/L) | 3.18±0.68 | 3.30±0.76 | ＜0.001 |
| HDL-C (mmol/L) | 1.32±0.30 | 1.32±0.29 | 0.858 |
| Uric acid (mmol/l) | 348.41±83.87 | 354.03±90.48 | ＜0.001 |
| PhA (°) | 5.73±0.61 | 5.43±0.60 | ＜0.001 |
Values are presented as the mean±standard deviation.
BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; WBCC, while blood cell count; FPG, fasting plasma glucose; TC, total cholesterol; TG, triglyceride; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; PhA, phase angle; CAS, carotid atherosclerosis
Detailed characteristics stratified by presence of an elevated baPWV are presented in Supplementary Table 1 . The PhA was also lower in participants with an elevated baPWV than in those without an elevated baPWV (5.38±0.62° vs. 5.74±0.59°, P＜0.001).
Supplementary Table 1. characteristics of individuals with or without elevated baPWV in the cross-sectional study.
| Normal (n = 6144) | elevated baPWV (n = 2084) | P value | |
|---|---|---|---|
| Age (years) | 46.30±8.81 | 58.17±8.94 | ＜0.001 |
| Sex (male, n, %) | 1986 (32.3) | 693 (33.3) | 0.433 |
| Smoking (n, %) | 1439 (23.4) | 482 (23.1) | 0.811 |
| Diabetes (n, %) | 401 (6.5) | 479 (23) | ＜0.001 |
| Hypertension (n, %) | 778 (12.7) | 775 (37.2) | ＜0.001 |
| Dyslipidemia (n, %) | 2302 (37.5) | 894 (42.9) | ＜0.001 |
| BMI (kg/m2) | 24.94±3.19 | 25.33±3.14 | ＜0.001 |
| SBP (mmHg) | 122.80±14.37 | 141.55±16.32 | ＜0.001 |
| DBP (mmHg) | 76.69±10.54 | 84.58±11.00 | ＜0.001 |
| WBCC (×109/L) | 5.72±1.44 | 5.90±1.81 | ＜0.001 |
| Monocyte count (×109/L) | 0.39±0.13 | 0.41±0.15 | ＜0.001 |
| Neutrophil count (×109/L) | 3.27±1.08 | 3.46±1.27 | ＜0.001 |
| FBG (mmol/L) | 5.3±1.02 | 6.07±1.88 | ＜0.001 |
| HbA1c (%) | 5.51±0.61 | 5.95±1.04 | ＜0.001 |
| TC (mmol/L) | 5.23±0.99 | 5.25±1.07 | 0.774 |
| TG (mmol/L) | 1.78±1.41 | 1.93±1.44 | ＜0.001 |
| LDL-C (mmol/L) | 3.25±0.73 | 3.25±0.76 | 0.835 |
| HDL-C (mmol/L) | 1.32±0.30 | 1.30±0.29 | 0.047 |
| Uric acid (mmol/l) | 354.49±85.23 | 353.27±83.51 | 0.572 |
| PhA (°) | 5.74±0.59 | 5.38±0.62 | ＜0.001 |
Values are presented as mean±standard deviation.
BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; WBCC, while blood cell count; FPG, fasting plasma glucose; TC, total cholesterol; TG, triglyceride; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; PhA, Phase angle; baPWV, brachial ankle pulse wave velocity
Associations of the PhA Value with Subclinical Atherosclerosis in the Cross-Sectional Studies
The results are summarized in Table 2 . There was a negative relationship between the PhA and risk of CAS in the multivariate adjusted model in all cases. The odds ratios (ORs) decreased linearly across tertiles 2 (OR = 0.80, 95% confidence interval [CI] 0.71-0.89, P＜0.001) and 3 (OR = 0.61, 95% CI 0.54-0.70, P＜0.001) compared to tertile 1 of the PhA value. These associations remained statistically significant when the PhA was used as a continuous variable (adjusted OR = 0.41, 95% CI 0.37-0.46, P＜0.001). In addition, the PhA was positively correlated with the BMI (r = 0.397, P＜0.001), systolic blood pressure (SBP) (r = 0.059, P＜0.001), diastolic blood pressure (DBP) (r = 0.216, P＜0.001), WBCC (r = 0.155, P＜0.001), monocyte count (r = 0.197, P＜0.001), neutrophil count (r = 0.090, P＜0.001), TG (r = 0.225, P＜0.001), and uric acid (r = 0.451, P＜0.001) while negatively correlated with age (r = − 0.295, P＜0.001) and HDL-C (r = − 0.294, P＜0.001) in the Pearson’s correlation analysis. There were no marked associations between the PhA and TC, LDL-C, FPG, or HbA1c levels.
Table 2. Association between the PhA and the risk of carotid atherosclerosis.
| Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|
| OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | |
| Tertile 1 | Reference | Reference | ||||
| Tertile 2 | 0.77 (0.69-0.86) | ＜0.001 | 0.77 (0.67-0.86) | ＜0.001 | 0.80 (0.71-0.89) | ＜0.001 |
| Tertile 3 | 0.58 (0.51-0.65) | ＜0.001 | 0.57 (0.50-0.65) | ＜0.001 | 0.61 (0.54-0.70) | ＜0.001 |
| Per 1-SD increase | 0.42 (0.38-0.46) | ＜0.001 | 0.42 (0.38-0.46) | ＜0.001 | 0.41 (0.37-0.46) | ＜0.001 |
Model 1: Adjusted for the age, sex, smoking
Model 2: Adjusted for the age, sex, smoking, WBCC count, neutrophil count, and monocyte count.
Model 3: Adjusted for the age, sex, smoking, WBCC, neutrophil count, monocyte count, BMI, diabetes, hypertension, dyslipidemia, and serum uric acid
BMI, body mass index; WBCC, while blood cell count; PhA, phase angle
A Pearson’s correlation analysis further indicated that the PhA was inversely associated with baPWV (r = -0.208, P＜0.001). The association between the PhA and baPWV remained significant (β = -0.117, P＜0.001) when age, sex, smoking, WBCC, neutrophil count, monocyte count, BMI, diabetes, hypertension, dyslipidemia, and serum uric acid were further adjusted by multiple linear regression. A significant association between the PhA and elevated baPWV is shown in Table 3 . Each 1-standard deviation increase in the PhA was associated with a 55% decreased risk (OR = 0.45, 95% CI 0.39-0.52, P＜0.001) of an elevated baPWV. The association between the PhA and an elevated baPWV was further investigated using the PhA as a categorical variable. Compared with the lowest tertiles, the OR (95% CI) of an elevated baPWV was 0.79 (0.67-0.9) and 0.67 (0.55-0.81) for tertiles 2 and 3 of the PhA ( Table 3 ) .
Table 3. Association between the PhA and the risk of elevated baPWV.
| Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|
| OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | |
| Tertile 1 | Reference | Reference | ||||
| Tertile 2 | 0.82 (0.70-0.97) | 0.017 | 0.80 (0.68-0.95) | 0.008 | 0.79 (0.67-0.93) | ＜0.001 |
| Tertile 3 | 0.72 (0.60-0.86) | ＜0.001 | 0.70 (0.59-0.85) | ＜0.001 | 0.67 (0.55-0.81) | ＜0.001 |
| Per 1-SD increase | 0.51 (0.44-0.58) | ＜0.001 | 0.50 (0.43-0.57) | ＜0.001 | 0.45 (0.39-0.52) | ＜0.001 |
Model 1: Adjusted for age, sex, smoking
Model 2: Adjusted for age, sex, smoking, WBCC count, neutrophil count, and monocyte count.
Model 3: Adjusted for age, sex, smoking, WBCC, neutrophil count, monocyte count, BMI, diabetes, hypertension, dyslipidemia, and serum uric acid
BMI, body mass index; WBCC, while blood cell count; PhA, phase angle; baPWV, brachial ankle pulse wave velocity
Findings of an RCS Analysis Exploring the Associations of the PhA with Subclinical Atherosclerosis in the Cross-Sectional Studies
We conducted an RCS analysis to flexibly model and visualize the associations between the PhA and subclinical atherosclerosis. After adjusting for traditional CVD risk factors, we observed a non-linear association between the PhA and CAS (P for nonlinear = 0.004) ( Fig.1A ) . However, the PhA exhibited a linear association with an elevated baPWV (P for nonlinear = 0.240) ( Fig.1B ) .
Fig.1. Dose-effect associations of the PhA with subclinical atherosclerosis.
(A) The association between the PhA and carotid atherosclerosis; (B) the association between the PhA value and brachial ankle pulse wave velocity.
Baseline Characteristics of the Longitudinal Study
For all participants, the median of the 25th and 75th percentiles for the follow-up duration was 24 months (13 and 36 months, respectively). In total, 582 (21.72%) of the 2680 participants progressed to CAS. The baseline characteristics of our study population stratified by the incidence of CAS are summarized in Table 4 . Participants with incident CAS were more likely to be older and have hypertension, diabetes, and higher values for SBP, DBP, TC, LDL-C, FPG, and HbA1c than those who remained free of CAS. There were no significant differences between the two groups in the BMI, sex, smoking, dyslipidemia, levels of WBCC, monocyte count, neutrophil count, TG, HDL-C, or uric acid levels. Furthermore, participants with incident CAS had lower baseline PhAs levels than those without CAS (5.71±0.59° vs. 5.39±0.58°, P＜0.001).
Table 4. Baseline characteristics of participants in the longitudinal study.
| Developed carotid atherosclerosis over follow-up | |||
|---|---|---|---|
| No (n = 2098) | Yes (n = 582) | P value | |
| Age (years) | 41.10±7.77 | 48.26±8.90 | ＜0.001 |
| Male (n, %) | 818 (39.0) | 244 (41.9) | 0.514 |
| Smoke (n, %) | 347 (16.5) | 107 (18.4) | 0.289 |
| Hypertension (n, %) | 158 (7.5) | 64 (11.0) | 0.011 |
| Diabetes (n, %) | 83 (4.0) | 36 (6.2) | 0.030 |
| Dyslipidemia (n, %) | 648 (30.9) | 182 (31.3) | 0.127 |
| BMI (kg/m2) | 24.35±3.44 | 24.17±3.12 | 0.252 |
| SBP (mmHg) | 121.14±15.22 | 124.73±16.34 | ＜0.001 |
| DBP (mmHg) | 75.70±11.15 | 77.24±11.53 | 0.004 |
| WBCC (×109/L) | 5.73±1.40 | 5.56±1.36 | 0.056 |
| Monocyte count (×109/L) | 0.39±0.12 | 0.38±0.16 | 0.282 |
| Neutrophil count (×109/L) | 3.31±1.06 | 3.23±1.04 | 0.146 |
| FPG (mmol/L) | 5.16±0.96 | 5.31±1.12 | 0.002 |
| HbA1c (%) | 5.39±0.54 | 5.49±0.63 | ＜0.001 |
| TC (mmol/L) | 5.03±0.91 | 5.22±0.96 | ＜0.001 |
| TG (mmol/L) | 1.65±1.38 | 1.56±1.05 | 0.124 |
| LDL-C (mmol/L) | 3.06±0.66 | 3.22±0.69 | ＜0.001 |
| HDL-C (mmol/L) | 1.32±0.32 | 1.34±0.29 | 0.231 |
| Uric acid (mmol/L) | 353.07±90.41 | 340.72±86.26 | 0.096 |
| PhA (°) | 5.71±0.59 | 5.39±0.58 | ＜0.001 |
Values are presented as the mean±standard deviation.
BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; WBCC, while blood cell count; FPG, fasting plasma glucose;
TC, total cholesterol; TG, triglyceride; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; PhA, phase angle
The Association of the PhA with Incident CAS in the Longitudinal Study
We fitted three Cox regression models to investigate the independent correlations between the PhA and CAS risk ( Fig.2 ) . After fully adjusting for the age, sex, smoking, WBCC, neutrophil count, monocyte count, BMI, diabetes, hypertension, dyslipidemia, and serum uric acid levels, the PhA was found to be significantly associated with incident CAS, with an adjusted hazard ratio (HR) and 95% CI of 0.44 (0.36-0.54). Compared with the lowest tertiles, the HR (95% CI) of incident CAS was 0.65 (0.48-0.87) for the highest tertiles of the PhA.
Fig.2.
Cox proportional hazard analyses of the association between the PhA and incident carotid atherosclerosis
The Association of the PhA with the Relative Annual Change and Absolute Annual Change of baPWV
The associations of the baseline PhA with the relative annual change and absolute annual change in baPWV are presented in Table 5 . A 1-unit increase in the PhA of the right arm decreased the absolute annual change in the baPWV by 20.374 (95% CI, -31.938 to -8.809) cm/s/year, corresponding to a 1.463% (95% CI, -2.267% to -0.659%) decrease in the baPWV annually with a 1-unit increase in the PhA in the fully adjusted model.
Table 5. Association of the PhA with the relative annual change and absolute annual change of the baPWV.
| Outcome | Model | β | B(95% CI) | P |
|---|---|---|---|---|
| Whole body | ||||
|
Absolute annual change in the baPWV (cm/s/year) |
Model 1 | -0.125 | -22.205 (-33.636~ -10.773) | ＜0.001 |
| Model 2 | -0.120 | -21.359 (-32.812 ~ -9.905) | ＜0.001 | |
| Model 3 | -0.115 | -20.374 (-31.938 ~ -8.809) | 0.001 | |
|
Relative annual change in the baPWV (%/year) |
Model 1 | -0.128 | -1.584 (-2.378~ -0.790) | ＜0.001 |
| Model 2 | -0.125 | -1.537 (-2.334~ -0.741) | ＜0.001 | |
| Model 3 | -0.119 | -1.463 (-2.267 ~ -0.659) | ＜0.001 | |
Model 1: Adjusted for the age, sex, smoking
Model 2: Adjusted for the age, sex, smoking, WBCC count, neutrophil count, and monocyte count.
Model 3: Adjusted for the age, sex, smoking, WBCC, neutrophil count, monocyte count, BMI, diabetes, hypertension, dyslipidemia, and serum uric acid
BMI, body mass index; WBCC, while blood cell count; PhA, phase angle; baPWV, brachial ankle pulse wave velocity
Discussion
In the present study, we conducted a comprehensive investigation of the association between the PhA and subclinical atherosclerosis. Our cross-sectional study demonstrated that the PhA was negatively correlated with subclinical atherosclerosis in the general population. The innovation of our study was the longitudinal assessment of the association between the PhA and the progression of subclinical atherosclerosis. We found that a smaller PhA was significantly associated with a higher risk of developing CAS and was positively associated with absolute or relative annual changes in baPWV. Thus, our study suggests that the PhA deserves more attention in clinical practice for early detection and intervention of CVD.
The PhA, which can be obtained by a BIA, is a simple, reliable, and non-invasive marker for evaluating cellular health, particularly reflecting cell membrane integrity and cell function 14) . Currently, the PhA is considered a reliable marker of nutritional status 15) . CVD has been the leading cause of mortality worldwide for many years. The nutritional status is crucial for the prevention and treatment of CVD 16 , 17) . Many studies have been performed to investigate the association between the PhA and CVD, with mixed results. For instance, a systematic review, meta-analysis, and several studies found that a small PhA was an independent risk factor for CVD 6 , 18 , 19) . Furthermore, an 18-year follow-up study among people from the greater Copenhagen area reported that a reduction in the PhA was an independent predictor of higher CVD morbidity and mortality rates 20) . However, another prospective cohort study including 153 patients with acute myocardial infarction found no association between the PhA and mortality or major adverse cardiac events 21) . Furthermore, the PhA in both chronic heart failure and acute heart failure was shown to be smaller than the control group 22 , 23) . Subclinical atherosclerosis is an important pathological process in CVD, and timely recognition can slow or prevent progression to CVD. CAS, including CIMT and carotid plaque, is a reliable surrogate marker for subclinical vascular disease. However, data on the association between the PhA and CAS are relatively limited. Two cross-sectional studies conducted in patients with diabetes mellitus or peritoneal dialysis found that the PhA were negatively associated with CIMT in an unadjusted model. However, this association was attenuated to nonsignificant after further adjusting for confounding factors 9 , 24) . In contrast, our study found that PhA was negatively associated with CAS. When adjusting for confounding factors, including metabolic parameters, these associations were attenuated but remained significant. We also observed a dose-response relationship between the PhA and CAS.
Compared to prior studies, our study has a number of distinct advantages. First, it boasts the largest sample size, thereby strengthening its persuasive power. Second, the studies were conducted in selected populations, such as those with diabetes mellitus or peritoneal dialysis. Our study was conducted in a general population without preexisting CVD or cerebrovascular diseases. Thus, these inconsistent results might be attributed to the relatively small sample size and selection of enrolled patients. Furthermore, to our knowledge, no previous study has investigated the association between the PhA and CAS progression. Therefore, we conducted a longitudinal study to explore this association. We found that each 1-standard deviation increase in the PhA was associated with a 56% decrease in the risk of incident CAS during the follow-up period. Our findings suggests that the PhA may serve as a noninvasive indicator of CAS risk, although more multicenter, multi-ethnic cohort studies are needed to confirm our results.
The PWV, including the baPWV and cfPWV, which measure arterial stiffness, is a good indicator of subclinical atherosclerosis. The relationship between the PhA and PWV has attracted close attention; however, conclusions are inconsistent, varying from a strong association to no association. Two cross-sectional studies reported a negative relationship between PhA and PWV 8 , 9) . Nevertheless, Sobhani et al. found a positive association between the whole-body PhA and PWV 25) . Furthermore, Heleniak et al. reported no significant differences in the PWV and PhA in renal transplant recipients 26) . However, the above-mentioned studies were limited by their small sample sizes or were not conducted in a general population free from clinical CVD or by exploring the relationship between the PhA and elevated baPWV. The findings from the present study revealed a significant inverse relationship between the PhA and baPWV. Furthermore, a small PhA increased the risk of an elevated baPWV even after adjusting for traditional CVD risk factors. We further conducted a longitudinal study to assess the longitudinal association between the PhA and the risk of arterial stiffness progression and discovered that a small PhA was significantly associated with an increased risk of arterial stiffness progression. These findings may be useful for identifying individuals at high risk of arterial stiffness, thereby enabling the early implementation of preventive strategies to prevent CVD.
As the precise mechanisms underlying these findings remain incompletely understood, there are potential mechanisms linking the PhA and subclinical atherosclerosis. First, a reduction in PhA suggests that cellular membranes might be more susceptible to damage, resulting in disruption of body water distribution and potential accumulation in extracellular water. The accumulation of extracellular fluid (ECF) exerts abnormal mechanical forces on vascular endothelial and smooth muscle cells, stimulating angiotensin II secretion, enhancing superoxide generation, and impairing NO bioavailability 27) . These mechanisms contribute to atherosclerosis development. Consistent with this, previous studies have also revealed that extracellular water is an independent predictor of incident CVD or adverse outcomes of CVD 28 , 29) . Second, the PhA has been increasingly frequently applied in nutrition status monitoring. Nutrition plays an important role in atherosclerosis development of atherosclerosis 30) . A decrease in the PhA is an indicator of malnutrition, which increases the risk of CVD 31 , 32) . We speculated that a low PhA, which represents the malnutrition status, increases the risk of subclinical atherosclerosis. Third, atherosclerosis is a chronic inflammatory condition, and pro-inflammatory cytokines play important roles in the pathogenesis of CVD 33) . The PhA has been considered a potential tool for evaluating the chronic inflammatory state in some diseases 4 , 34) . The WBCC and neutrophil and monocyte counts are markers of low-grade inflammation in clinical practice. In the present study, we found that the PhA was positively correlated with the WBCC, neutrophil count, and monocyte count. We speculated that chronic inflammation might mediate the association between the PhA and subclinical atherosclerosis. Fourth, several studies demonstrated that the PhA were significantly correlated with metabolic factors 35 , 36) . In the present study, we showed that the association between the PhA and subclinical atherosclerosis was attenuated after adjusting for metabolic factors. In addition, the PhA was associated with metabolic factors, including the BMI, blood pressure, TG, HDL-C, and uric acid. We speculated that metabolic factors might mediate this association.
The strengths of our study merit further consideration. A large cross-sectional study and longitudinal study were conducted in the present study, which can provide more statistical power to detect associations than previous studies. However, our study also had some limitations. First, several factors that may affect subclinical atherosclerosis, such as dietary and exercise habits, were unavailable in this study, although we adjusted for various confounders. Second, we did not measure other chronic inflammation markers, such as IL-6 and CRP. Therefore, we cannot exclude the role of chronic inflammation in the relationship between the PhA and subclinical atherosclerosis. Finally, our study included only Chinese individuals; therefore, extrapolating our findings to other populations requires further validation.
Conclusion
We demonstrated that the PhA was negatively and independently correlated with subclinical atherosclerosis in Chinese adults without pre-existing cardiovascular and cerebrovascular diseases. Furthermore, a low PhA increased the risk of subclinical atherosclerosis. Based on these findings, the PhA may be a useful noninvasive and practical tool for identifying individuals at a high risk of CVD. Monitoring PhA deserves more attention in clinical practice for the early prevention of subclinical atherosclerosis and reduction in the incidence of CVD. Our findings may inspire future research to explore the potential role of the PhA as an extra parameter in primary cardiovascular prevention.
Acknowledgements
We appreciate the help and support provided by all participants in this study.
Author’s Contributions
Wen Guo and Qun Zhang participated in this study. Guo, Lin, Lu, Yu, and Li conducted the study and collected data. Wen Guo and Fei Lin participated in the data analysis. Wen Guo, Chengxiao Yu, and Qun Zhang wrote and modified the manuscript accordingly. All authors have read and approved the final manuscript.
Funding
This study was supported by the Jiangsu Province Capability Improvement Project through Science, Technology, and Education (ZDXK202248), Jiangsu Medical Association Medical Research Project of Health Management (SYH-32099-0119), and Young Scholars Fostering Fund of the First Affiliated Hospital of Nanjing Medical University (No. PY2021008).
Data Availability Statement
The original contributions presented in this study are included in the article, and further inquiries can be directed to the corresponding authors.
Conflicts of Interest
The authors have no conflicts of interest to disclose.
References
- 1).Liu S, Li Y, Zeng X, Wang H, Yin P, Wang L, Liu Y, Liu J, Qi J, Ran S, Yang S, Zhou M. Burden of Cardiovascular Diseases in China, 1990-2016: Findings From the 2016 Global Burden of Disease Study. JAMA Cardiol, 2019; 4: 342-352 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2).Yang CW, Guo YC, Li CI, Liu CS, Lin CH, Liu CH, Wang MC, Yang SY, Li TC, Lin CC. Subclinical Atherosclerosis Markers of Carotid Intima-Media Thickness, Carotid Plaques, Carotid Stenosis, and Mortality in Community-Dwelling Adults. Int J Environ Res Public Health, 2020; 17: 4745 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3).Lu Y, Zhu Y, Ma Y, Li C, Hua R, Zhong B, Wang H, Xie W. Association of subclinical atherosclerosis and cognitive decline: a community-based cross-sectional study. BMJ Open, 2022; 12: e059024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4).da Silva BR, Orsso CE, Gonzalez MC, Sicchieri JMF, Mialich MS, Jordao AA, Prado CM. Phase angle and cellular health: inflammation and oxidative damage. Rev Endocr Metab Disord, 2023; 24: 543-562 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5).Victoria-Montesinos D, García-Muñoz AM, Navarro-Marroco J, Lucas-Abellán C, Mercader-Ros MT, Serrano-Martínez A, Abellán-Aynés O, Barcina-Pérez P, Hernández-Sánchez P. Phase Angle, Handgrip Strength, and Other Indicators of Nutritional Status in Cancer Patients Undergoing Different Nutritional Strategies: A Systematic Review and Meta-Analysis. Nutrients, 2023; 15: 1790 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6).Langer RD, Larsen SC, Ward LC, Heitmann BL. Phase angle measured by bioelectrical impedance analysis and the risk of cardiovascular disease among adult Danes. Nutrition, 2021; 89: 111280 [DOI] [PubMed] [Google Scholar]
- 7).Portugal MRC, Canella DS, Curioni CC, Bezerra FF, Faerstein E, Neves MF, Koury JC. Bioelectrical impedance analysis-derived phase angle is related to risk scores of a first cardiovascular event in adults. Nutrition, 2020; 78: 110865 [DOI] [PubMed] [Google Scholar]
- 8).Heleniak Z, Illersperger S, Malgorzewicz S, Debska-Slizien A, Budde K, Halleck F. Arterial Stiffness as a Cardiovascular Risk Factor After Successful Kidney Transplantation in Diabetic and Nondiabetic Patients. Transplant Proc, 2022; 54: 2205-2211 [DOI] [PubMed] [Google Scholar]
- 9).Tsilingiris D, Schimpfle L, Kender Z, Sulaj A, von Rauchhaupt E, Herzig S, Szendroedi J, Kopf S. Utility of bioelectrical phase angle for cardiovascular risk assessment among individuals with and without diabetes mellitus. Diab Vasc Dis Res, 2024; 21: 14791641231223701 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10).Jun JE, Kang H, Hwang YC, Ahn KJ, Chung HY, Jeong IK. The association between lipoprotein (a) and carotid atherosclerosis in patients with type 2 diabetes without pre-existing cardiovascular disease: A cross-sectional study. Diabetes Res Clin Pract, 2021; 171: 108622 [DOI] [PubMed] [Google Scholar]
- 11).Guo W, Li X, Wu J, Zhu W, Lu J, Qin P, Diao Q, Xu N, Zhang Q. Serum alkaline phosphatase is associated with arterial stiffness and 10-year cardiovascular disease risk in a Chinese population. Eur J Clin Invest, 2021; 51: e13560 [DOI] [PubMed] [Google Scholar]
- 12).Wu S, Xu L, Wu M, Chen S, Wang Y, Tian Y. Association between triglyceride-glucose index and risk of arterial stiffness: a cohort study. Cardiovasc Diabetol, 2021; 20: 146 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13).Herder M, Arntzen KA, Johnsen SH, Mathiesen EB. The metabolic syndrome and progression of carotid atherosclerosis over 13 years. The Tromso study. Cardiovasc Diabetol, 2012; 11: 77 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14).Norman K, Stobaus N, Pirlich M, Bosy-Westphal A. Bioelectrical phase angle and impedance vector analysis--clinical relevance and applicability of impedance parameters. Clin Nutr, 2012; 31: 854-861 [DOI] [PubMed] [Google Scholar]
- 15).Peng Z, Xu D, Li Y, Peng Y, Liu X. Phase Angle as a Comprehensive Tool for Nutritional Monitoring and Management in Patients with Crohn’s Disease. Nutrients, 2022; 14: 2260 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16).Ravera A, Carubelli V, Sciatti E, Bonadei I, Gorga E, Cani D, Vizzardi E, Metra M, Lombardi C. Nutrition and Cardiovascular Disease: Finding the Perfect Recipe for Cardiovascular Health. Nutrients, 2016; 8: 363 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17).Popiolek-Kalisz J, Blaszczak P. Nutritional Status of Coronary Artery Disease Patients-Preliminary Results. Int J Environ Res Public Health, 2023; 20: 3464 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18).de Borba EL, Ceolin J, Ziegelmann PK, Bodanese LC, Goncalves MR, Canon-Montanez W, Mattiello R. Phase angle of bioimpedance at 50 kHz is associated with cardiovascular diseases: systematic review and meta-analysis. Eur J Clin Nutr, 2022; 76: 1366-1373 [DOI] [PubMed] [Google Scholar]
- 19).Saad MA, Jorge AJ, de Andrade Martins W, Cardoso GP, Dos Santos MM, Rosa ML, Lima GA, de Moraes RQ, da Cruz Filho RA. Phase angle measured by electrical bioimpedance and global cardiovascular risk in older adults. Geriatr Gerontol Int, 2018; 18: 732-737 [DOI] [PubMed] [Google Scholar]
- 20).Langer RD, Ward LC, Larsen SC, Heitmann BL. Can change in phase angle predict the risk of morbidity and mortality during an 18-year follow-up period? A cohort study among adults. Front Nutr, 2023; 10: 1157531 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21).Queiroz SA, Gonzalez MC, da Silva AMB, Costa JKA, de Oliveira CDR, de Sousa IM, Fayh APT. Is the standardized phase angle a predictor of short- and long-term adverse cardiovascular events in patients with acute myocardial infarction? A cohort study. Nutrition, 2022; 103-104: 111774 [DOI] [PubMed] [Google Scholar]
- 22).Scicchitano P, Ciccone MM, Iacoviello M, Guida P, De Palo M, Potenza A, Basile M, Sasanelli P, Trotta F, Sanasi M, Caldarola P, Massari F. Respiratory failure and bioelectrical phase angle are independent predictors for long-term survival in acute heart failure. Scand Cardiovasc J, 2022; 56: 28-34 [DOI] [PubMed] [Google Scholar]
- 23).Scicchitano P, Massari F. The role of bioelectrical phase angle in patients with heart failure. Rev Endocr Metab Disord, 2023; 24: 465-477 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24).Demirci MS, Demirci C, Ozdogan O, Kircelli F, Akcicek F, Basci A, Ok E, Ozkahya M. Relations between malnutrition-inflammation-atherosclerosis and volume status. The usefulness of bioimpedance analysis in peritoneal dialysis patients. Nephrol Dial Transplant, 2011; 26: 1708-1716 [DOI] [PubMed] [Google Scholar]
- 25).Sobhani S, Vakili S, Javid Jam D, Aryan R, Khadem-Rezaiyan M, Eslami S, Alinezhad-Namaghi M. Relationship between anthropometric indices and arterial stiffness: Insights from an epidemiologic study. Obes Sci Pract, 2022; 8: 494-499 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26).Heleniak Z, Illersperger S, Naik MG, Osmanodja B, Ronicke S, Eleftheriadis G, Halleck F, Budde K. Influence of Belatacept- vs. CNI-Based Immunosuppression on Vascular Stiffness and Body Composition. J Clin Med, 2022; 11: 1219 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27).Park S, Lee CJ, Jhee JH, Yun HR, Kim H, Jung SY, Kee YK, Yoon CY, Park JT, Kim HC, Han SH, Kang SW, Park S, Yoo TH. Extracellular Fluid Excess Is Significantly Associated With Coronary Artery Calcification in Patients With Chronic Kidney Disease. J Am Heart Assoc, 2018; 7: e008935 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28).Knudsen NN, Kjaerulff TM, Ward LC, Saebye D, Holst C, Heitmann BL. Body water distribution and risk of cardiovascular morbidity and mortality in a healthy population: a prospective cohort study. PLoS One, 2014; 9: e87466 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29).Tsai YC, Chiu YW, Tsai JC, Kuo HT, Hung CC, Hwang SJ, Chen TH, Kuo MC, Chen HC. Association of fluid overload with cardiovascular morbidity and all-cause mortality in stages 4 and 5 CKD. Clin J Am Soc Nephrol, 2015; 10: 39-46 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30).Torres N, Guevara-Cruz M, Velazquez-Villegas LA, Tovar AR. Nutrition and Atherosclerosis. Arch Med Res, 2015; 46: 408-426 [DOI] [PubMed] [Google Scholar]
- 31).Arikawa R, Kanda D, Ikeda Y, Tokushige A, Sonoda T, Anzaki K, Ohishi M. Prognostic impact of malnutrition on cardiovascular events in coronary artery disease patients with myocardial damage. BMC Cardiovasc Disord, 2021; 21: 479 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32).Chien SC, Chandramouli C, Lo CI, Lin CF, Sung KT, Huang WH, Lai YH, Yun CH, Su CH, Yeh HI, Hung TC, Hung CL, Lam CSP. Associations of obesity and malnutrition with cardiac remodeling and cardiovascular outcomes in Asian adults: A cohort study. PLoS Med, 2021; 18: e1003661 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33).Zhang H, Dhalla NS. The Role of Pro-Inflammatory Cytokines in the Pathogenesis of Cardiovascular Disease. Int J Mol Sci, 2024; 25: 1082 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34).Fajardo-Espinoza FS, Cabrera-Nieto SA, Espada-Vargas AL, Perez-Camargo DA, Mohar A, Cruz-Ramos M. Phase angle as a potential tool to evaluate chronic inflammatory state and predict quality of life deterioration in women with breast cancer and obesity: A narrative review. Nutrition, 2024; 127: 112524 [DOI] [PubMed] [Google Scholar]
- 35).Mat S, Tan MP, Mohktar MS, Kamaruzzaman SB, Ibrahim F. Phase angle and diabetes in community-dwelling older adults: cross-sectional analysis from the Malaysian elders longitudinal research (MELoR) study. Eur J Clin Nutr, 2022; 76: 680-684 [DOI] [PubMed] [Google Scholar]
- 36).Bucan Nenadic D, Radic J, Kolak E, Vuckovic M, Novak I, Selak M, Radic M. Phase Angle Association with Dietary Habits and Metabolic Syndrome in Diabetic Hypertensive Patients: A Cross-Sectional Study. Nutrients, 2022; 14: 5058 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The original contributions presented in this study are included in the article, and further inquiries can be directed to the corresponding authors.