Association between Serum Cystatin C levels and long-term cardiovascular outcomes and all-cause mortality in older patients with obstructive sleep apnea.

Background and Aims: To investigate the association between obstructive sleep apnea (OSA) severity and baseline serum cystatin C (Cys-C) concentration and to explore the association between baseline serum Cys-C and long-term cardiovascular outcomes and mortality in older patients with OSA. Methods: Between January 2015 and October 2017, a total of 1107 consecutive eligible older patients (≥60 years) with OSA were included in this multicenter, prospective cohort study, and baseline demographics, clinical characteristics, sleep parameters, and follow-up outcomes were collected. Participants were divided into different groups based on baseline serum Cys-C levels. The primary end point was major adverse cardiovascular events (MACE) and the secondary end point was all-cause mortality. The correlation between OSA severity and baseline serum Cys-C was evaluated by Spearman correlation analysis. Multivariate Cox regression was used to analyze the association between Cys-C and the incidence of MACE and mortality. Results: Participants included 672 men and 435 women, with a median age of 66 (range, 60-96) years. At baseline, apnea-hypopnea index (AHI) (r = 0.128, p < 0.05), oxygen desaturation index (ODI) (r = 0.116, p < 0.05), and the lowest pulse oxygen saturation (LSpO2) (r = -0.097, p < 0.05) were correlated with serum Cys-C concentration. During the median follow-up period of 42 months, 97 patients (8.8%) experienced MACE and 40 patients (3.6%) experienced death. The association between serum Cys-C levels and the risk of MACE and all-cause mortality was slow rising shaped. The multivariable Cox regression analysis showed patients with a serum Cys-C concentration of ≥1.14 mg/L had higher risks of MACE (HR = 5.30, 95% CI: 2.28-12.30, p < 0.05) and all-cause mortality (HR = 9.66, 95% CI: 2.09-44.72, p < 0.05) compared with patients with serum Cys-C of ≤0.81 mg/L in older patients with OSA. The receiver-operating characteristic curve showed baseline serum Cys-C levels exhibited moderately capable of identifying patients with a long-term risk of clinical adverse events (MACE and mortality). Conclusion: OSA severity was positively correlated with serum Cys-C concentration. High levels of Cys-C were independently associated with increased risks of MACE and all-cause mortality in older patients with OSA, suggesting that lowering Cys-C levels should be considered as a therapeutic target, and monitoring serum Cys-C may be beneficial to the favorable prognosis of older patients with OSA.