C反应蛋白–甘油三酯–葡萄糖指数与帕金森病相关性的加权横断面研究：基于NHANES数据库的分析

Association between the C-Reactive Protein-Triglyceride-Glucose Index and Parkinsonism: A Weighted Cross-Sectional Study Based on the NHANES Database

摘要: 目的：探讨C反应蛋白–甘油三酯–葡萄糖指数(CTI)与帕金森病(PD)之间的关联。方法：基于NHANES数据库，纳入12,291名参与者并进行复杂抽样加权分析。将CTI分别作为连续变量和三分位分组变量纳入加权Logistic回归模型，采用限制性立方样条(RCS)评估剂量–反应关系，同时采用ROC曲线比较CTI、TyG和CRP对PD的区分效能。结果：纳入样本加权后代表1,033,961,743名人群，PD加权患病率为1.15%。随着CTI三分位升高，PD加权患病率由T1组0.72%升至T3组1.71%。在完全调整模型中，CTI作为连续变量时与PD患病可能性升高相关(OR = 1.79, 95% CI: 1.23~2.60, P = 0.003)；与T1 (5.2~8.01)组相比，T3 (8.81~13.11)组PD患病可能性较高(OR = 2.00, 95% CI, 1.01~4.00, P = 0.049)，趋势检验P = 0.003。RCS分析提示CTI与PD之间呈正向线性关联(非线性检验P = 0.118)。ROC分析显示，CTI、TyG和CRP区分PD的AUC分别为0.636、0.558和0.574。亚组分析显示总体关联稳定(OR = 1.87, 95% CI: 1.41~2.50, P < 0.001)，年龄分层存在交互作用(交互作用检验P = 0.003)。结论：在NHANES加权人群中，较高CTI与PD患病可能性升高相关，与TyG和CRP相比，CTI对PD具有相对更好的区分效能。

Abstract: Objective: To investigate the association between the C-reactive protein-triglyceride-glucose index (CTI) and Parkinsonism (PD). Methods: Based on the NHANES database, 12,291 participants were included and analyzed using complex sampling weights. CTI was included in weighted logistic regression models both as a continuous variable and as a tertile-based categorical variable. Restricted cubic spline (RCS) analysis was used to evaluate the dose-response relationship. Receiver operating characteristic (ROC) curves were used to compare the discriminatory performance of CTI, TyG, and CRP for PD. Results: The weighted sample represented 1,033,961,743 individuals, and the weighted prevalence of PD was 1.15%. With increasing CTI tertiles, the weighted prevalence of PD increased from 0.72% in T1 to 1.71% in T3. In the fully adjusted model, CTI as a continuous variable was associated with increased odds of PD (OR = 1.79, 95% CI: 1.23~2.60, P = 0.003). Compared with the T1 group (5.2~8.01), the T3 group (8.81~13.11) had higher odds of PD (OR = 2.00, 95% CI: 1.01~4.00, P = 0.049), with a P value for trend of 0.003. RCS analysis suggested a positive linear association between CTI and PD (P for nonlinearity = 0.118). ROC analysis showed that the AUC values of CTI, TyG, and CRP for discriminating PD were 0.636, 0.558, and 0.574, respectively. Subgroup analysis showed a stable overall association (OR = 1.87, 95% CI: 1.41~2.50, P < 0.001), with an interaction by age stratification (P for interaction = 0.003). Conclusion: In the weighted NHANES population, higher CTI was associated with increased odds of Parkinsonism. Compared with TyG and CRP, CTI showed relatively better discriminatory performance for PD.