Peripheral biomarkers to diagnose obstructive sleep apnea in adults: A systematic review and meta-analysis

Obstructive Sleep Apnea (OSA) has been recognized as a major health concern worldwide, given its increasing prevalence, difficulties in diagnosis and treatment, and impact on health, economy, and society. Clinical guidelines highlight the need of biomarkers to guide OSA clinical decision-making, but so far, without success. In this systematic review and meta-analysis, registered on the International Prospective Register of Systematic Reviews database (ID CRD42020132556), we proposed to gather and further explore candidates identified in the literature as potential OSA biomarkers.

Search strategies for eight different databases (PubMed/Medline, Cochrane Library, Biblioteca Virtual da Saúde, Web of Science, EMBASE, World Intellectual Property Organization database, and bioRxiV and medRxiV Preprint Servers) were developed. We identified studies exploring potential biomarkers of OSA, in peripheral samples of adults, with and without OSA, with no comorbidities defined in study inclusion criteria, published after the last systematic review and meta-analysis conducted on OSA biomarkers, until May 31st, 2020. Risk of bias was assessed through the 14-item Quality Assessment Tool for Diagnostic Accuracy Studies. Demographic, clinical, and candidate biomarkers' data were collected and analyzed via random effects meta-analyses.

Among the 1512 unique studies screened, 120 met the inclusion criteria and 16 studies with low risk of bias were selected for meta-analyses. The selected 16 studies enrolled a total of 2156 participants, from which 1369 were diagnosed with OSA and 787 were disease-free controls. The assessed variables showed high heterogeneity. From the 38 biomarker candidates evaluated, only two were evaluated in more than one study. Most studies pinpointed candidates with more potential for OSA prognosis. ADAM29, FLRT2 and SLC18A3 mRNA levels in PBMCs, Endocan and YKL-40 levels in serum, and IL-6 and Vimentin levels in plasma revealed the most promising candidates for OSA diagnosis.

Although the current systematic review and meta-analysis allowed us to identify candidates to further explore as potential biomarkers in future studies, it is evident that OSA biomarkers research is still at an early stage. Most findings derive from small-size single-center study cohorts and single-candidate studies. We point several gaps in current OSA biomarker research that may guide into new directions and approaches towards the identification of OSA biomarkers.