单核细胞与高密度脂蛋白胆固醇比值与听力损失之间的关联：基于NHANES的横断面研究

文献信息

中文摘要

背景： 本研究旨在揭示美国成年人中单核细胞与高密度脂蛋白胆固醇（HDL-C）比值（MHR）与不同听力损失（HL）类型之间的关联，并评估MHR作为预测HL生物标志物的潜力。
方法： 分析了2001-2006年、2009-2012年和2015-2020年NHANES数据。利用加权单因素和多因素logistic回归（LR）模型揭示MHR与言语频率听力损失（SFHL）和高频听力损失（HFHL）的可能关系。利用限制性立方样条（RCS）回归探索MHR与SFHL和HFHL之间可能的非线性关系。通过亚组分析和交互检验考察不同人群特征间的异质性。最后，通过受试者工作特征（ROC）曲线分析评估完全校正模型对SFHL和HFHL的判别性能。
结果： 我们的研究纳入12,030名参与者，其中3387名（24.11%）患有SFHL，5817名（44.66%）患有HFHL。LR分析显示，在调整所有协变量后，MHR与HFHL[OR（95% CI）：1.66（1.26，2.19）]和SFHL[OR（95% CI）：1.73（1.26，2.38）]呈显著正相关。RCS分析表明MHR与HFHL和SFHL之间均呈线性关系（非线性P>0.05）。交互分析显示，种族（交互作用P=0.026）和BMI（交互作用P=0.029）在MHR-HFHL关系中存在显著交互作用。基于模型3的ROC曲线分析显示对HFHL和SFHL具有良好的判别能力，曲线下面积（AUC）分别为0.908（95% CI：0.903-0.913）和0.883（95% CI：0.876-0.890）。
结论： MHR与HL风险呈正相关。作为一种易于获取的炎症相关生物标志物，MHR可能有助于反映HL的存在，但其在风险评估中的价值尚需在前瞻性研究中进一步验证。

英文摘要

BACKGROUND: This study aimed to unravel the association of the monocyte-to-high-density lipoprotein cholesterol (HDL-C) ratio (MHR) with different hearing loss (HL) types among American adults and assess the potential of MHR as a predictive HL biomarker.
METHODS: 2001-2006, 2009-2012, and 2015-2020 NHANES data were analyzed. The possible relations of MHR to speech-frequency hearing loss (SFHL) and high-frequency hearing loss (HFHL) were unveiled leveraging weighted univariate and multivariate logistic regression (LR) models. Possible nonlinear relations of MHR to SFHL and HFHL were explored utilizing restricted cubic spline (RCS) regression. Heterogeneity across varied population characteristics was examined via subgroup analyses and interaction tests. Finally, the discriminative performance of the fully adjusted model for SFHL and HFHL was evaluated through receiver operating characteristic (ROC) curve analysis.
RESULTS: Our study encompassed 12,030 participants, among whom 3387 (24.11%) had SFHL and 5817 (44.66%) had HFHL. LR analysis showed that, after adjusting for all covariates, MHR was markedly positively related to HFHL [OR (95% CI): 1.66 (1.26, 2.19)] and SFHL [OR (95% CI): 1.73 (1.26, 2.38)]. RCS analysis indicated linear relationships between MHR and both HFHL and SFHL (P for nonlinear>0.05). Interaction analyses revealed significant interactions of race (P for interaction = 0.026) and BMI (P for interaction = 0.029) in the MHR-HFHL relation. ROC curve analysis based on Model 3 demonstrated good discriminative ability for HFHL and SFHL, with the areas under the curve (AUCs) of 0.908 (95% CI: 0.903-0.913) and 0.883 (95% CI: 0.876-0.890), respectively.
CONCLUSION: MHR is positively associated with HL risk. As an easily accessible inflammation-related biomarker, MHR may help reflect the presence of HL, but its value in risk assessment requires further validation in prospective studies.