Characteristics of participants
This study included 3,097 participants, comprising 1,471 females and 1,626 males, with a median age of 29 years (interquartile range: 20–38), and the median (25%-75%) of remnant cholesterol was 0.47(interquartile range: 0.34–0.69) mmol/L, in the racial distribution, the majority are non -Hispanic white, with a total of 1,117 participants. Three-fifths of the participants were never smokers. 1,989 participants have impaired cardiovascular fitness (e-table 2). According to the quartiles of the remnant cholesterol within the study, the baseline characteristics of 3,097 respondents were shown in Table 1. All variables were statistically significant in the four remnant cholesterol groups. Participants in the RC Q4 group, as opposed to those in the lower RC group, were predominantly male, older, Non-Hispanic White, and exhibited never smoking habits. They presented with elevated value of BMI, SBP, DBP, CRP, LDL-C, Hb, PLT, FBG, and presented with lower levels of HDL-C, VO2max, physical activity. Due to the strong correlation between gender and age with CVF, it is essential for us to consider their impact on CVF. Therefore, the following sections will further analyze the relationship between RC and CVF, adjusted for gender and age.
Table 1
Characteristics of the participants according to the quartiles of the RC
Characteristic | Quartile1, n = 848a | Quartile 2, n = 707a | Quartile 3, n = 781a | Quartile 4, n = 761a | Pa |
Age years | 24 [17, 35] | 28 [18, 37] | 29 [20, 38] | 33 [24, 40] | < 0.001 |
Sex | | | | | < 0.001 |
female | 429 (52.4) | 377 (59.9) | 368 (47.7) | 297 (39.2) |
male | 419 (47.6) | 330 (40.1) | 413(52.3) | 464(60.8) |
Race | | | | | < 0.001 |
Mexican American | 214 (8.2) | 196 (8.2) | 263 (10.5) | 263 (12.0) |
Non-Hispanic White | 245 (64.4) | 231 (63.5) | 304 (70.3) | 337 (70.8) |
Non-Hispanic Black | 336 (19.9) | 210 (16.0) | 160 (9.8) | 94 (6.1) |
Other Race | 53 (7.5) | 70 (12.2) | 54 (9.4) | 67(11.1) |
Physical activity | | | | | 0.007 |
level 0 | 67 (8.5) | 57 (6.5) | 74 (9.6) | 88 (11.8) |
level 1 | 124 (13.7) | 121(19.7) | 125 (17.9) | 123 (15.6) |
level 2 | 159 (16.7) | 135 (17.1) | 155 (17.6) | 159 (21.0) |
level 3 | 276 (37.2) | 229 (36.9) | 263 (37.9) | 277 (39.0) |
level 4 | 31 (1.5) | 25 (1.6) | 31 (2.1) | 15 (1.0) |
level 5 | 57 (7.0) | 54 (6.6) | 44 (4.2) | 28 (2.4) |
level 6 | 134 (15.4) | 86 (11.5) | 89 (10.7) | 71 (9.1) |
Smoking status | | | | | 0.001 |
Former smoking | 103 (13.7) | 87 (10.2) | 130 (19.2) | 129 (17.7) |
Current smoking | 162 (22.7) | 122 (19.8) | 175 (25.5) | 202 (26.2) |
Never smoking | 583 (63.6) | 498(70.0) | 476 (55.3) | 430 (56.1) |
BMI kg/m2 | 22.67 [20.41, 25.75] | 23.69 [20.87, 27.13] | 24.62 [21.77, 29.30] | 27.63 [24.27, 30.95] | < 0.001 |
Waist circumference, cm | 80.20 [73.10, 88.90] | 81.89 [74.66, 91.50] | 86.60 [77.50, 98.40] | 95.50 [86.30, 104.33] | < 0.001 |
SBP, mmHg | 109 [10, 118] | 110 [103, 118] | 112 [106, 120] | 114 [108, 122] | < 0.001 |
DBP, mmHg | 66 [60, 73] | 68 [61, 74] | 70 [63, 76] | 72 [65, 78] | < 0.001 |
CRP, mg/dL | 0.05 [0.03, 0.16] | 0.07 [0.03, 0.24] | 0.10 [0.04, 0.29] | 0.18 [0.08, 0.42] | < 0.001 |
PLT,109/L | 249 [213, 294] | 260 [226, 297] | 259 [222, 304] | 271 [234, 314] | 0.001 |
Hb, g/dL | 14.2 [13.3, 15.2] | 14.2 [13.4, 15.2] | 14.7 [13.6, 15.7] | 15.1 [13.9, 16.0] | < 0.001 |
TC, mmol/L | 3.92 [3.49, 4.40] | 4.29 [3.76, 4.81] | 4.47 [3.96, 5.04] | 5.07 [4.40, 5.79] | < 0.001 |
HDL-C, mmol/L | 1.45 [1.22, 1.73] | 1.36 [1.17, 1.65] | 1.27 [1.07, 1.50] | 1.11 [0.93, 1.29] | < 0.001 |
LDL-C, mmol/L | 2.28 [1.86, 2.77] | 2.69 [2.17, 3.15] | 2.77 [2.33, 3.34] | 3.15 [2.53, 3.75] | < 0.001 |
RC, mmol/L | 0.26 [0.23, 0.29] | 0.36 [0.34, 0.39] | 0.51 [0.46, 0.56] | 0.83 [0.72, 1.05] | < 0.001 |
FBG, mmol/L | 5.01 [4.75, 5.31] | 5.01 [4.75, 5.33] | 5.15 [4.91, 5.44] | 5.19 [4.90, 5.56] | < 0.001 |
VO2max, ml/kg/min | 41.01 [35.08, 47.60] | 39.49 [33.75, 46.06] | 38.99 [33.76, 45.46] | 39.40 [34.19, 45.00] | 0.009 |
Impaired CVF | 547 (50.9) | 458 (54.0) | 514 (58.3) | 470 (55.1) | 0.194 |
a Median (25%-75%); n (unweighted) (%) |
b Kruskal-Wallis rank-sum test for complex survey samples; chi-squared test with Rao & Scott's second-order correction |
BMI = body mass index; CVF = cardiovascular fitness; CRP = C-creative protein; DBP = diastolic blood pressure; FBG = Fast blood glucose; PLT = platelet; Hb = hemoglobin; LDL-c = Low-Density Lipoprotein Cholesterol; RC = remnant cholesterol; SBP = systolic blood pressure; TC = Total cholesterol. |
Remnant cholesterol and impaired cardiovascular fitness
The relationship between various covariates and impaired CVF was first analyzed using weighted logistic regression. Age and gender, being the two major influencing factors of CVF, were included in analysis. As shown in Table 2, impaired CVF is associated with BMI, WC, SBP, DBP, Hb, non-Hispanic black, and RC. Among them, RC is significantly associated with impaired CVF, with each unit (1mmol/L) increase associated with an 83% increase in the risk of impaired CVF 1.83 (95%CI 1.29–2.61). In weighted multivariable logistic regression, the RC showed a significant positive correlation with impaired CVF. After adjusting for age, gender, race, DBP, SBP, BMI, WC, Hb, LDL-c, HDL-c, Glu, CRP, PLT and physical activity, each unit increase in remnant cholesterol was associated with 1.65-fold higher risk of impaired CVF (OR:1.65; 95%CI 1.06–2.57). Converting RC from a numerical variable to categorical quartiles, the multivariable-adjusted odds ratios (OR) and 95% confidence intervals (CI) for increasing RC quartiles were as follows: 1.00 (reference), 1.25 (95% CI: 0.96–1.64), 1.57 (95% CI: 1.22–2.02), and 1.59 (95% CI: 1.17–2.18), respectively (Table 3,4). To achieve a comprehensive understanding of the link between the RC and impaired CVF, RCS curves were used to fit the dose-response relationship between RC and impaired CVF. As shown in Fig. 2, after adjusting all covariates, it demonstrates a linear trend between RC and impaired CVF, with a p-value for non-linearity of 0.052. In e-Figure1,2, RCS curves for two additional adjustment models were also presented.
Table 2
After adjusting age and gender, the association of different variables and impaired CVF.
Variables | HR (95CI) | P-value |
Race (%) | | <0.001 |
Mexican American | reference | |
Non-Hispanic White | 0.84(0.64–1.10) | |
Non-Hispanic Black | 1.44(1.09–1.90) | |
Other Race | 1.43(0.74–1.76) | |
Physical activity level | | <0.001 |
0 | reference | |
1 | 1.15(0.77–1.72) | |
2 | 1.25(0.79–1.99) | |
3 | 1.15(0.77–1.72) | |
4 | 0.99(0.44–2.24) | |
5 | 0.59(0.33–1.07) | |
6 | 0.43(0.77–1.21) | |
Smoking status | | 0.125 |
Never smoking | reference | |
Former smoking | 1.10(0.81–1.48) | |
Current smoking | 0.79(0.62–1.01) | |
BMI, kg/m2 | 1.04(1.02–1.06) | <0.001 |
Waist Circumference, cm | 1.02(1.01–1.02) | <0.001 |
SBP, mmHg | 1.01(1.00-1.02) | 0.018 |
DBP, mmHg | 1.02(1.00-1.03) | <0.001 |
CRP, mg/dL | 1.04(0.83–1.30) | 0.765 |
PLT, 109/L | 1.00(1.00–1.00) | 0.083 |
HGB, g/dL | 0.86(0.79–0.92) | <0.001 |
TC, mmol/L | 1.15(1.02–1.29) | 0.031 |
HDL-C, mmol/L | 0.87(0.69–1.08) | 0.217 |
LDL-C, mmol/L | 1.12(0.97–1.29) | 0.126 |
FBG, mmol/L | 1.20(0.97–1.47) | 0.102 |
RC, mmol/L | 1.83(1.29–2.61) | 0.002 |
As age and gender are the two major influencing factors of CVF, we included these two covariates in a weighted logistic regression analysis and analyzed the relationship between various covariates and impaired CVF. |
BMI = body mass index; CVF = cardiovascular fitness; CRP = C-creative protein; DBP = diastolic blood pressure; FBG = Fast blood glucose; PLT = platelet; Hb = hemoglobin; LDL-c = Low-Density Lipoprotein Cholesterol; RC = remnant cholesterol; SBP = systolic blood pressure; TC = Total cholesterol. |
Table 3
After adjusting age and gender, the association of quartile RC and impaired CVF.
The quartile RC HR (95%CI) |
Q 1 | Q 2 | Q 3 | Q 4 | P for trend |
Ref | 1.25(0.96–1.64) | 1.57(1.22–2.02) ** | 1.59(1.17–2.18) * | 0.003 |
Statistical significance is indicated in bold |
OR = odds ratio, 95%CI = 95% Confidence Interval, RC = remnant cholesterol, Ref = References. |
*P < 0.05 |
**P <0.001 |
Table 4
Multivariate adjusted ORs (95%CIs) of RC and impaired CVF in different models
| RC |
Per 1 SD increase | Per 1 unit increase |
Model 1 | 1.19(1.08–1.32) * | 1.84(1.28–2.64) * |
Model 2 | 1.14(1.01–1.27) * | 1.55(1.05–2.30) * |
Model 3 | 1.16(1.02–1.32) * | 1.65(1.06–2.57) * |
Model 1: Adjusted for age, gender |
Model 2: included adjustments from Model 1 plus race, DBP, SBP, BMI, WC, Hb, and physical activity |
Model 3: included adjustments from Model 2 plus LDL, HDL, GLU, CRP, PLT |
OR = odds ratio, 95%CI = 95% Confidence Interval, RC = remnant cholesterol. |
*P < 0.05 |
**P <0.001 |
Stratified analyses of the associations between remnant cholesterol and impaired CVF
As shown in Fig. 3, subgroup analysis and interaction tests were conducted to explore the consistency of the relationship between RC and impaired CVF. These subgroups included gender, age, body mass index, and race. Stratified analyses identified a more significant association between males and impaired CVF 1.96 (95% CI: 1.33–2.88) compared to females 1.05 (95% CI: 0.66–1.67). In terms of race, the association with impaired CVF is stronger in non-Hispanic black individuals 2.07 (95% CI: 0.97–4.44) and other Hispanic individuals 3.80 (95% CI: 1.43–10.08), particularly with a significant risk elevation observed in the other Hispanic subgroup. In age groups, the association between impaired CVF and the younger population aged 12 to 19 years 2.02 (95% CI: 1.20–3.41) is more significant compared to the adult population aged 20 to 49 years 1.36 (95% CI: 0.93–1.97). The association reached statistical significance in the normal weight group (BMI 18.5–25) 1.86 (95% CI: 1.09–3.18). Interaction terms were utilized to evaluate variations among each subgroup, demonstrating that the positive relationship between RC and the likelihood of impaired CVF remained consistent across populations with differing statuses, suggesting that this association may be applicable in diverse population contexts.