Non-HDL-C as a dynamic prediction marker of carotid intima-media thickening in health checkup people

Cox regression models were fitted using either baseline nHDL-C values or the longitudinal nHDL-C trajectory patterns. The predictive accuracy of the models was evaluated using area under the curve (AUC) and concordance index (C-index).Results The study included 5,741 participants aged 18–60 years(mean age: 34.80 ± 8.27 years), comprising 2,487 males(mean age: 35.41 ± 8.45 years) and 3,252 females(mean age: 36.48 ± 8.11 years).

Three nHDL-C trajectories were identified in males and four in females, each associated with different risks of carotid intima-media thickening. BMI was significantly associated with increased nHDL-C levels in both males(β= 2.952, P< 0.001) and females (β= 2.327, P< 0.001). In the male population, the Cox model with baseline nHDL-C values yielded a C-index of 0.749, while the model based on nHDL-C trajectory had a C-index of 0.826.

The combined model, using both baseline values and trajectory, achieved a C-index of 0.832. In females, the Cox model with baseline nHDL-C values had a C-index of 0.796, and the trajectory-based model had a C-index of 0.822. The combined model reached a C-index of 0.824. nHDL-C trajectory provided superior predictive accuracy for carotid intima-media thicknening compared to baseline nHDL-C values.

Compared to the low-stable group, males in the slow-rising and rapid-rising groups had hazard ratios(HRs) of 10.51(95% CI: 7.90–13.98) and 23.25(95% CI: 10.40–51.98), respectively, for carotid intima-media thickening. In females, the low-stable rising group had an HR of 3.69(95% CI: 2.27–5.99), the medium-stable group had an HR of 15.48(95% CI: 10.56–22.70), and the rapid-rising group had an HR of 13.93(95% CI: 5.44–35.69).Conclusion Distinct nHDL-C trajectories in the health checkup population are linked to varying risks of carotid intima-media thickening, with trajectory-based classifications providing more accurate risk predictions compared to baseline nHDL-C values.