Related Experiment Video
Updated: Sep 2, 2026

Drug-Induced Sleep Endoscopy (DISE) with Target Controlled Infusion (TCI) and Bispectral Analysis in Obstructive Sleep Apnea
Published on: December 6, 2016
Inverse association between serum sLRP-1 levels and obstructive sleep apnea in snoring individuals: a case-control
Zeli Sun1, Zhifang Lv1, Qiong Huang1
1Department of Respiratory and Critical Care Medicine, Anqing Municipal Hospital, Anqing, 246000, China.
Background:
Obstructive sleep apnea (OSA) is a common but underdiagnosed sleep disorder characterized by intermittent hypoxia and systemic inflammation. Soluble low-density lipoprotein receptor-related protein-1 (sLRP-1) is a multifunctional receptor involved in lipid metabolism, inflammation, and vascular regulation, potentially linking metabolic and vascular pathways to OSA pathophysiology.
Methods:
This case-control study included 180 adults presenting with snoring symptoms. Participants underwent overnight polysomnography (PSG) and were categorized into OSA (AHI ≥ 5/h, n = 100) and non-OSA (AHI < 5/h, n = 80) groups. Serum soluble low-density lipoprotein receptor-related protein 1 (sLRP-1) levels were measured using ELISA. Multivariable logistic regression and subgroup analyses were conducted to assess associations. Machine learning models, including XGBoost, were developed to evaluate classification performance and identify important features associated with OSA.
Results:
Serum sLRP-1 levels were significantly lower in the OSA group compared to the non-OSA group (P < 0.001). Logistic regression revealed that sLRP-1 was independently and inversely associated with OSA risk (fully adjusted OR = 0.35; 95% CI, 0.22-0.56). Subgroup analyses confirmed consistency across age and C-reactive protein (CRP) strata. Naive Bayes achieved an AUC of 0.94 in distinguishing OSA from non-OSA cases, with sLRP-1 ranked among the most influential features based on SHAP analysis.
Conclusion:
Lower serum sLRP-1 levels were associated with OSA status in adults with snoring symptoms referred for polysomnography. Its contribution to internally validated machine-learning models suggests potential exploratory value, but external validation is required before clinical application.