Cardio-Ankle Vascular Index is Associated with Prevalence and New-Appearance of Atrial Fibrillation in Japanese Urban Residents: A Retrospective Cross-Sectional and Cohort Study

Figures & data

Table 1 Comparison of Demographic and Clinical Characteristics in Subjects with or without Atrial Fibrillation (Cross-Sectional Study)

Variables	Before PSM		After PSM
	AF	Non-AF	AF	Non-AF
	(N = 164)	(N = 47,523)	(N = 163)	(N = 652)
Male (%)	139 (84.8)	19,729 (41.5)*	138 (84.7)	560 (85.9)
Age (years)	68 (61–74)	47 (37–59)*	68 (61–74)	68 (61–74)
BMI (kg/m^2)	24.3 (22.6–26.1)	22.0 (19.0–24.4)*	24.3 (22.6–26.2)	23.1 (21.5–25.0)*
SBP (mmHg)	131 (121–140)	122 (112–132)*	131 (121–140)	134 (124–144)*
DBP (mmHg)	78 (72–86)	72 (64–80)*	78 (72–86)	78 (71–84)
CAVI	9.3 (8.8–9.8)	7.5 (7.0–8.3)*	9.3 (8.8–9.8)	9.0 (8.4–9.5)*
FPG (mg/dL)	92 (86–100)	84 (79–90)*	92 (86–100)	91 (85–100)
LDL-C (mg/dL)	117 (103–142)	122 (101–145)	117 (103–142)	128 (106–147)
HDL-C (mg/dL)	57 (50–69)	67 (55–79)*	57 (50–69)	60 (49–70)
TG (mg/dL)	88 (72–126)	79 (56–118)*	98 (72–139)	88 (72–126)
Creatinine (mg/dL)	0.91 (0.79–1.04)	0.70 (0.60–0.83)*	0.91 (0.79–1.04)	0.84 (0.75–0.94)*
eGFR (mL/min/1.73m^2)	64.1 (51.6–74.0)	79.8 (70.6–90.2)*	64.0 (51.5–74.1)	69.1 (60.3–51.5)*
Current smoking (%)	24 (14.6)	7433 (15.6)	24 (14.7)	80 (12.3)
Habitual alcohol drinking (%)	37 (22.6)	5796 (12.2)*	37 (22.7)	138 (21.1)
BMI ≥ 25 kg/m^2 (%)	64 (39.0)	9925 (20.9)*	64 (39.3)	167 (25.6)*
SBP ≥140 and/or 90 mmHg (%)	55 (33.5)	7548 (15.9)	55 (33.7)	230 (35.3)
Treatment of hypertension (%)	66 (40.2)	4262 (9.0)*	66 (40.5)	195 (29.9)*
Treatment of diabetes (%)	11 (6.7)	897 (1.9)*	11 (6.7)	51 (7.8)
Treatment of dyslipidemia (%)	25 (15.2)	2862 (6.0)*	138 (84.7)	93 (14.3)
eGFR <60 mL/min/1.73m^2 (%)	71 (43.3)	3204 (6.7)*	71 (43.6)	160 (24.5)*

Notes: Data are presented as median (interquartile range) or percent of subjects. *p < 0.05; Mann–Whitney U-test for continuous variables, or Fisher’s exact test for dichotomous variables among two groups.

Abbreviations: AF, atrial fibrillation; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; CAVI, cardio-ankle vascular index; FPG, fasting plasma glucose; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; TG, triglyceride; eGFR, estimated glomerular filtration rate.

Table 2 Logistic Regression Model for the Presence of Atrial Fibrillation After Propensity Score Matching (Cross-Sectional Study)

Variables	Odds Ratio	95% Confidence Interval	p value
BMI (every 1 kg/m^2)	1.15	1.08–1.22	<0.001
CAVI (every 1 index)	1.37	1.08–1.22	0.008
eGFR (every 10 mL/min/1.73m^2)	0.807	0.705–0.924	0.002
Hypertension treatment (Yes; 1, No; 0)	1.14	0.781–1.66	0.497

Note: Akaike’s information criterion: 777.2, residual deviance: 767.2, p < 0.001.

Abbreviations: BMI, body mass index; CAVI, cardio-ankle vascular index; eGFR, estimated glomerular filtration rate.

Table 3 Comparison of Demographic and Clinical Characteristics in Subjects with or without the New-Appearance of Atrial Fibrillation During the 4-Year Period (Cohort Study)

Variables	Subjects with New-Appearance of AF (N=22)	Subjects without New-Appearance of AF (N=5396)	p value
Male sex (%)	16 (72.7)	2331 (43.2)	<0.001*
Age (years)	62.5 (58–74.3)	48 (40–58)	<0.001
BMI (kg/m^2)	23.0 (20.9–25.2)	21.9 (20.0–23.9)	0.152
SBP (mmHg)	127 (117–139)	115 (107–126)	0.005
DBP (mmHg)	76 (72–86)	72 (66–80)	0.058
CAVI	8.7 (8.0–9.4)	7.6 (7.1–8.3)	<0.001
FPG (mg/dL)	89 (85–100)	85 (80–90)	0.011
LDL-C (mg/dL)	124 (114–142)	125 (105–147)	0.929
HDL-C (mg/dL)	66 (51–75)	70 (58–83)	0.114
TG (mg/dL)	94 (69–128)	77 (56–111)	0.090
Creatinine (mg/dL)	0.78 (0.69–0.90)	0.71 (0.60–0.84)	0.030
eGFR (mL/min/1.73m^2)	69.2 (62.2–81.5)	78.3 (69.6–87.9)	0.014
Age ≥ 65 years (%)	10 (45.5)	741 (13.7)	<0.001*
Current smoking (%)	5 (22.7)	726 (13.5)	0.207*
Habitual alcohol consumption (%)	4 (18.2)	835 (15.5)	0.766*
Treatment of hypertension (%)	10 (45.5)	453 (8.4)	<0.001*
Treatment of diabetes (%)	1 (4.5)	58 (1.1)	0.214*
Treatment of dyslipidemia (%)	1 (4.5)	104 (1.9)	0.073*

Notes: Data are presented as median (interquartile range) or percent of subjects. Mann–Whitney U-test or *Fisher’s exact test was used to compare subjects in two groups.

Abbreviations: AF, atrial fibrillation; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; CAVI, cardio-ankle vascular index; FPG, fasting plasma glucose; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; TG, triglyceride; eGFR, estimated glomerular filtration rate.

Table 4 Cox-Proportional Hazard Model for the Association of the New-Appearance of Atrial Fibrillation with Clinical Variables (Cohort Study)

Variables	Hazard Ratio	95% Confidence Interval	p value
Sex (male; 1, female; 0)	3.29	1.29–8.40	0.013
Older age (age ≥ 65 y; 1, < 65 y; 0)	2.17	0.863–5.46	0.099
CAVI ≥ 8.0 (Yes; 1, No; 0)	5.27	1.60–17.3	0.006

Abbreviation: CAVI, cardio-ankle vascular index.

Figure 1 Discriminatory power of CAVI for the prediction of the new-appearance of atrial fibrillation (cohort study). The receiver-operating-characteristic curve shows the performance of CAVI to discriminate new-appearance of AF. The Youden Index was used in to select the optimal cut-off point of CAVI.

Figure 2 Kaplan–Meier curves for the rates of the new-appearance of atrial fibrillation when subjects were stratified by baseline CAVI (cohort study). Participants were divided into two groups: those with baseline CAVI ≥ 8.0 and those with baseline CAVI < 8.0.