3.1 Baseline characteristics of NHANES participants
A total of 22,654 subjects aged 18 years and above, with biological age and ACO data, participated in the analysis. Table 1 lists the general characteristics of the participants, including demographic factors, biological age and ACO prevalence. There were 1326 participants with ACO. Among them, 11033 were males and 11621 were females. The calendar ages of the ACO group and the non-ACO group were ( 53.77 ± 16.67 ) years and ( 49.46 ± 17.90 ) years, respectively. There were significant differences in age, gender, race, education level, marital status, poverty level, alcohol using, smoking, obesity, hypertension, hyperlipidemia and diabetes between ACO patients and non-ACO patients. In addition, compared with non-ACO patients, ACO patients had a higher biological age ( HD : 2.41 vs.2.20 ; kDM : 48.97 vs. 44.65 ; phenoAge : 52.69 vs. 46.95 ).Table 1.
Variables
|
ACO(N=1326)
|
NON-ACO(N=21328)
|
P-Value
|
Age ,mean(SD)
|
53.77(16.67)
|
49.46(17.90)
|
<0.001
|
Age ,N(%)
|
|
|
|
<40
|
302(22.8)
|
7308(34.3)
|
<0.001
|
40-60
|
473(35.7)
|
6876(32.2)
|
|
>60
|
551(41.6)
|
7144(33.5)
|
|
hd ,mean(SD)
|
2.41(1.02)
|
2.20(0.98)
|
<0.001
|
kdm ,mean(SD)
|
48.97(20.18)
|
44.65(20.40)
|
<0.001
|
phenoage ,mean(SD)
|
52.69(17.59)
|
46.95(18.84)
|
<0.001
|
BioAgeAccel ,mean(SD)
|
-4.80(15.95)
|
-4.81(15.05)
|
0.978
|
BioAgeAccel ,N(%)
|
|
|
|
YES
|
474(35.7)
|
7418(34.8)
|
0.492
|
NO
|
852(64.3)
|
13910(65.2)
|
|
PhenoAgeAccel ,mean(SD)
|
-1.08(4.99)
|
-2.52(4.73)
|
<0.001
|
PhenoAgeAccel,N(%)
|
|
|
|
YES
|
509(38.4)
|
5816(27.3)
|
<0.001
|
NO
|
817(61.6)
|
15512(72.7)
|
|
Gender ,N(%)
|
|
|
|
male
|
590(44.5)
|
10443(49.0)
|
0.002
|
female
|
736(55.5)
|
10885(51.0)
|
|
Race ,N(%)
|
|
|
|
Mexican American
|
86(6.5)
|
3898(18.3)
|
<0.001
|
Non-Hispanic White
|
800(60.3)
|
9282(43.5)
|
|
Non-Hispanic Black
|
241(18.2)
|
4200(19.7)
|
|
Other
|
199(15.0)
|
3948(18.5)
|
|
EDUcation ,N(%)
|
|
|
|
Below high school
|
381(28.7)
|
5334(25.0)
|
0.003
|
High School or above
|
945(71.3)
|
15994(75.0)
|
|
Marital ,N(%)
|
|
|
|
Yes
|
722(54.4)
|
13155(61.7)
|
<0.001
|
No
|
604(45.6)
|
8173(38.3)
|
|
Poverty ,N(%)
|
|
|
|
Poor
|
358(27.0)
|
4110(19.3)
|
<0.001
|
Not poor
|
968(73.0)
|
17218(80.7)
|
|
Obesity ,N(%)
|
|
|
|
Obesity
|
587(44.3)
|
8174(38.3)
|
<0.001
|
Overweight
|
396(29.9)
|
7228(33.9)
|
|
Normal
|
343(25.9)
|
5926(27.8)
|
|
Drink ,N(%)
|
|
|
|
Yes
|
989(74.6)
|
15050(70.6)
|
0.002
|
No
|
336(25.4)
|
6264(29.4)
|
|
Smoke,N(%)
|
|
|
|
Yes
|
530(40.0)
|
4111(19.3)
|
<0.001
|
No
|
796(60.0)
|
17217(80.7)
|
|
Diabetes ,N(%)
|
|
|
|
YES
|
310(23.4)
|
3479(16.3)
|
<0.001
|
NO
|
1016(76.6)
|
17849(83.7)
|
|
Hyptersion ,N(%)
|
|
|
|
Yes
|
730(55.1)
|
8812(41.3)
|
<0.001
|
No
|
596(44.9)
|
12516(58.7)
|
|
Hyperlipidemia ,N(%)
|
|
|
|
YES
|
1125(84.8)
|
17048(79.9)
|
<0.001
|
NO
|
201(15.2)
|
4280(20.1)
|
|
Table 1. Baseline characteristics of the NHANES participants. Continuous variables were presented as mean (SE); Categorical variables were presented as N (%); means (SE) and % were weight-adjusted. KDM: KlemeraDoubal method; hd: homeostatic dysregulation.
* Patients who are married and cohabiting with their partners are defined as married; patients with a family income index below 1 are defined as poor; those with a body mass index over 30 are defined as obese, between 25 and 30 are defined as overweight, and below 25 are classified as normal; individuals meeting any of the following criteria are defined as diabetic patients: history of diabetes, insulin injection, oral hypoglycaemic medication, Glycohemoglobin level >=6.5, Glucose level >=126; individuals meeting any of the following conditions are defined as hypertensive patients: currently diagnosed with hypertension, SBP >=140, DBP >=90; individuals meeting any of the following conditions are diagnosed as hyperlipidemia patients, with TC >200, TG >150, male HDL <40, female HDL <50, LDL >=130, or with hypercholesterolaemia.
3.2 Logistic regression analysis of the association between biological aging and ACO
Table 2 shows the relationship between ACO and biological age. After adjusting for age, gender and race ( model 1 ), we found that there was a significant association between biological aging and ACO. Adjusted educational level, marital status, ratio of family income to poverty, obesity, alcohol consumption, and cigarette use, the association between BioAgeAccel and the incidence of ACO became no longer significant, and the remaining results were consistent with the results of model 1 in the multivariate regression results of model 2. After further adjustment of the disease status in Model 3, it was found that HD, kdm, and bioageaccel were not significantly associated with the incidence of ACO. In model 3, for every 1 year increase in phenotypic age, the risk of ACO increased by 1 % ( OR 1.01,95 % CI 1.01-1.02, P<0.001 ). Acceleration of biological aging is defined as the residual error of the regression of biological age to calendar age, including BioAgeAccel and PhenoAgeAccel, where Phenoageaccel is associated with the onset of ACO. The acceleration of biological aging indicates that the biological age is older. Compared with PhenoAgeAccel-negative individuals, PhenoAgeAccel-positive individuals were 4 % more likely to develop ACO ( OR 1.04,95 % CI 1.02-1.05, p < 0.001 ).
Biological aging
|
Model 1
|
Model 2
|
Model 3
|
OR (95% CI)
|
P
|
OR (95% CI)
|
P
|
OR (95% CI)
|
P
|
HD
|
1.17 (1.10-1.24)
|
<0.001
|
1.13 (1.07-1.20)
|
<0.001
|
1.01 (0.94-1.08)
|
0.734
|
KDM
|
1.00 (1.00-1.01)
|
0.022
|
1.00 (1.00-1.01)
|
0.019
|
1.00 (0.99-1.00)
|
0.357
|
PhenoAge
|
1.01 (1.01-1.02)
|
<0.001
|
1.02 (1.01-1.03)
|
<0.001
|
1.01(1.01-1.02)
|
<0.001
|
BioAgeAccel
|
1.00 (1.00-1.01)
|
0.018
|
1.00 (1.00-1.00)
|
0.744
|
1.00 (0.99-1.00)
|
0.209
|
PhenoAgeAccel
|
1.04 (1.02-1.05)
|
<0.001
|
1.04 (1.02-1.05)
|
<0.001
|
1.04(1.02-1.05)
|
<0.001
|
Table2 Associations of biological ages and accelerated biological aging with ACO. Model 1: Adjusting for age, gender, race; Model 2: Model 1+adjusted for educational level, marital status, ratio of family income to poverty, obesity, alcohol consumption, and cigarette use. Model 3: Model 2+adjusted for Diabetes, hypertension, and hyperlipidemia.HD: homeostatic dysregulation; KDM: Klemera–Doubal method; OR: Odds ratio; CI: Confidence interval;
The exposure was further grouped into categorical variables according to the quartile, and the P value of the trend was estimated. In model3, higher phenotypic age ( Q4 ) and higher degree of phenotypic aging ( Q4 ) were more likely to cause ACO, with OR values of 2.33 ( 1.59-3.42,95 % CI ) and 1.64 ( 1.36-1.98,95 % CI ), respectively, and P trend was less than 0.001.In addition, lower kdm biological age acceleration ( Q2 ) could reduce the risk of ACO ( OR : 0.79 ( 0.67-0.93 ), 95 % CI ).Table 3
|
OR (95% CI)
|
P value
|
PhenoAgeAccel, (continuous)
|
1.04 (1.02-1.05)
|
<0.001
|
PhenoAgeAccel, (quartile)
|
|
|
Quartile 1
|
1(REF)
|
|
Quartile 2
|
1.32 (1.09-1.59)
|
0.004
|
Quartile 3
|
1.40 (1.16-1.69)
|
<0.001
|
Quartile 4
|
1.64 (1.36-1.98)
|
<0.001
|
P trend
|
<0.001
|
KdmAgeAccel, (continuous)
|
1.00 (0.99-1.00)
|
0.209
|
KdmAgeAccel, (quartile)
|
|
|
Quartile 1
|
1(REF)
|
|
Quartile 2
|
0.79 (0.67-0.93)
|
0.006
|
Quartile 3
|
0.97 (0.82-1.14)
|
0.714
|
Quartile 4
|
0.88 (0.74-1.04)
|
0.144
|
P trend
|
0.49
|
PhenoAge, (continuous)
|
1.01 (1.01-1.02)
|
0.001
|
PhenoAge, (quartile)
|
|
|
Quartile 1
|
1(REF)
|
|
Quartile 2
|
1.44 (1.14-1.83)
|
0.002
|
Quartile 3
|
1.79 (1.30-2.46)
|
<0.001
|
Quartile 4
|
2.33 (1.59-3.42)
|
<0.001
|
P trend
|
<0.001
|
KdmAge, (continuous)
|
1.00 (0.99-1.00)
|
0.357
|
KdmAge, (quartile)
|
|
|
Quartile 1
|
1(REF)
|
|
Quartile 2
|
1.04 (0.85-1.27)
|
0.713
|
Quartile 3
|
1.06 (0.86-1.30)
|
0.599
|
Quartile 4
|
1.00 (0.80-1.26)
|
0.978
|
P trend
|
0.964
|
Table3 Associations of biological ages and accelerated biological aging with ACO (Quartile divided groups)
3.3 Restricted cubic spline regression analysis of the association between biological aging and ACO
Figure2 shows that there is a significant nonlinear positive correlation between the age of the three organisms and ACO ( nonlinear P < 0.001 ). The odds ratio ( OR ) shows that when the HD value exceeds 1.97, it will gradually increase and slowly stabilize. Regarding the non-linear relationship between KDM and ACO, the figure shows that the risk of ACO increases rapidly in the lower range of KDM ( < 42.53 years old ). After KDM reached 42.53 years old, the rate of risk increase slowed down and gradually stabilized. In addition, the risk of ACO increased rapidly before the PhenoAge value reached about 46.39 years old, and then began to increase slowly and gradually entered the plateau period. The results of rcs analysis of BioAgeAccel accelerated biological aging were generally meaningless ( P-overall = 0.186 ). In addition, the results of rcs analysis of PhenoAgeAccel showed that there was a linear relationship between exposure and outcome ( P for nonlinear = 0.699 ), and the critical values of two kinds of accelerated biological aging were-5.33 and-2.69, respectively.
Figure 2. Restricted cubic spline model of the odds ratios of ACO with HD (a), KDM (b), PhenoAge (c), BioAgeAccel (d), and PhenoAgeAccel (e). HD: homeostatic dysregulation; KDM: Klemera–Doubal method; OR: Odds ratio; CI: Confidence interval.