Baseline characteristics of participants
Of the 62,604 participants included, 32,584 were men and 30,020 were women. There were 323 CV events and 114 CV deaths during the follow-up period, with a mean follow-up period of 3.3 ± 2.1 years (maximum: 9.0 years). The mean values of LDL-C of the 1st, 2nd, 3rd, and 4th quartiles were 73.2 ± 15.1 mg/dL, 103.5 ± 6.5 mg/dL, 125.6 ± 6.9 mg/dL, and 161.6 ± 25.0 mg/dL and those of HDL-C of the 1st, 2nd, 3rd, and 4th quartiles were 38.7 ± 4.5 mg/dL, 48.5 ± 2.3 mg/dL, 56.7 ± 2.6 mg/dL, and 73.8 ± 30.2 mg/dL, respectively. The mean BMI of the participants was 24.1 ± 3.0 kg/m2, with 22,633 (36.2%) being classified as obese (BMI ≥ 25.0 kg/m2). Of the total number of participants, 11.9% were current smokers, and 59.2% and 17.9% had been diagnosed with hypertension and diabetes, respectively. Each basic characteristic of participants had a significant sex-difference (Table 1).
CV events and CV mortality
During the observation period of 204,025.6 person-years, the incidence rates of ischemic heart disease and ischemic brain disease were 0.97 and 0.61 per 1,000 person-years, respectively. The mortalities from these diseases were 0.22 and 0.34 per 1,000 person-years, during the observation period of 204,058.7 person-years (Table 2).
Association between quartiles of low-density lipoprotein cholesterol and CVD
Proportions of CV event and CV death according to the quartiles of LDL-C are presented in Figure 2. The proportion of CV event was the highest in participants with LDL-C from the 1st quartile, and it decreased as the quartile of LDL-C increased. However, CV death rate showed no specific pattern in accordance with the quartile of LDL-C.
In the crude model (Table 2), the incidence of ischemic brain disease was 20% lower in participants with LDL-C from the 4th quartile (HR: 0.80, 95% CI: 0.68–0.94) compared to those with LDL-C from the 1st quartile, while its mortality was not significantly related to the LDL-C levels (HR: 0.90, 95% CI: 0.73–1.12). However, neither ischemic heart disease event nor its mortality was significantly associated with high levels of LDL-C. The total CV events were 17% lower in participants with LDL-C from the 4th quartile (HR: 0.83, 95% CI: 0.75–0.92) than in those with LDL-C from the 1st quartile, while there was no significant association between the total CV mortality and high levels of LDL-C (HR: 0.89, 95% CI: 0.76–1.06). The incidences of ischemic heart disease (β= -4.17×10-4, p= 0.041), ischemic brain disease (β= -4.02×10-4, p= 0.013), and total CV event (β= -8.19×10-4, p= 0.002) had negative correlations with LDL-C levels in the crude model by linear regression analysis.
In Model 1 and Model 2, the incidence of ischemic brain disease was 17% lower in participants with LDL-C from the 4th quartile (HR: 0.83, 95% CI: 0.70–0.98 in both models) compared to those with LDL-C from the 1st quartile, while its mortality was not significantly associated with LDL-C levels (HR: 0.93, 95% CI: 0.75–1.16 in both models). However, neither ischemic heart disease event nor its mortality was significantly related to high levels of LDL-C. In Model 1, the total CV events were 13% lower in participants with LDL-C from the 4th quartile (HR: 0.87, 95% CI: 0.79–0.97) than in those with LDL-C from the 1st quartile, while the relationship between them was not significant in Model 2 (HR: 0.90, 95% CI: 0.81–1.01). Total CV mortality was not associated with high levels of LDL-C (HR: 0.95, 95% CI: 0.80–1.13 in both models). The incidences of ischemic brain disease and total CV event had negative correlations with LDL-C levels in Model 1 and Model 2 by linear regression analyses.
In Model 3, the incidence of ischemic brain disease was 16% lower in participants with LDL-C from the 4th quartile (HR: 0.84, 95% CI: 0.70–1.00) than in those with LDL-C from the 1st quartile, while its mortality was not significantly associated with LDL-C levels (HR: 0.95, 95% CI: 0.74–1.22). There was no significant association between the incidence of an ischemic heart disease event or its mortality and high levels of LDL-C and total CV event or its mortality and high levels of LDL-C. The total number of participants who died was 202 and CV deaths were 22 within 1 year from the study. The results from the analysis of Model 3, excluding the elderly who died within 1 year from the time of study, were not different; both total CV event and CV death were not significantly associated with LDL-C levels.
Association between quartiles of high-density lipoprotein cholesterol and CVD
Proportions of CV event and CV death according to quartiles of HDL-C are presented in Figure 2. The proportion of CV event was the lowest in participants with HDL-C from the 4th quartile, while U-shaped association was observed between the proportions of CV death and HDL-C levels.
In the crude model (Table 3), ischemic heart disease events were 12% lower in participants with HDL-C from the 4th quartile (HR: 0.88, 95% CI: 0.77–1.00) compared to those with HDL-C from the 1st quartile, while its mortality was not significantly lowered (HR: 0.79, 95% CI: 0.61–1.04). However, neither an ischemic brain disease event (HR: 0.95, 95% CI: 0.82–1.12) nor its mortality (HR: 1.04, 95% CI: 0.84–1.28) was significantly associated with high levels of HDL-C. Total CV events were 9% lower in participants with HDL-C from the 4th quartile (HR: 0.91, 95% CI: 0.82–1.00) than in those with HDL-C from the 1st quartile, while there was no significant association between total CV mortality and high levels of HDL-C (HR: 0.93, 95% CI: 0.79–1.10). The incidence of total CV event had negative correlation with HDL-C level (β= -4.57×10-4, p= 0.041) in the crude model by linear regression analysis.
In Models 1, 2, and 3, the incidence of CV events and CV mortality were not significantly associated with high levels of HDL-C. The result from the analysis of Model 3, excluding the elderly who died within 1 year from the time of study, was not different; both total CV event and CV death were not significantly associated with HDL-C levels.
Stratified analysis
In a stratified analysis of the association between LDL-C and CV events, the risk of CV events in non-smokers with LDL-C from the 4th quartile was significantly reduced (adjusted hazard ratio, aHR: 0.84, 95% CI: 0.73–0.96) (Figure 3 a). In a stratified analysis of the association between LDL-C and CV death, CV mortality in diabetic patients with LDL-C from the 4th quartile increased significantly (aHR: 1.47, 95% CI: 1.05–2.05) (Figure 3 b). Stratified analysis could not find any strata with a significant association between HDL-C and CV event or CV death (Figure 3 c, d).