Children diagnosed with obstructive sleep apnea (OSA) appear to face a significantly higher risk of influenza and COVID-19 in the years following diagnosis, according to a large retrospective cohort study published in the Journal of Clinical Sleep Medicine.
Using the TriNetX Global Collaborative Network, investigators analyzed electronic health record data from more than half a million children aged 2 to 18 years with newly diagnosed OSA and an equal number of matched controls without OSA. Over a five-year follow-up period, children with OSA were about twice as likely to be diagnosed with influenza or COVID-19 compared with their peers.
After propensity score matching for age and obesity status, each cohort included 539,127 children. Influenza was diagnosed in 5.1% of children with OSA compared with 2.8% of controls, corresponding to a risk ratio (RR) of 1.80. Five-year influenza-free survival was lower in the OSA group (90.3% vs 93%), with a hazard ratio (HR) of 1.45.
For COVID-19, 2.5% of children with OSA were diagnosed during follow-up compared with 1.0% of controls (RR, 2.50). Five-year COVID-19–free survival was 95.0% in the OSA group versus 97.5% in controls.
The association persisted across developmental age strata (2–6, 6–12, and 12–18 years), with similar relative increases in risk for both influenza and COVID-19. Sensitivity analyses excluding children with comorbid conditions known to increase infection risk, such as trisomy 21, asthma, and Prader-Willi syndrome, continued to show a significant association, although the magnitude of risk was somewhat attenuated.
Importantly, the increased risk was also observed in a pre–COVID-19 era influenza analysis, suggesting that pandemic-era testing practices alone do not explain the findings.
When the analysis was restricted to more severe outcomes —specifically pneumonia due to influenza or COVID-19 —the relative risks were even higher. Influenza-related pneumonia occurred in approximately 0.2% of children with OSA versus 0.1% of controls, while COVID-19 pneumonia was markedly more frequent in the OSA cohort, with a risk ratio of 25.96, though absolute event numbers were small.
The investigators also examined whether adenotonsillectomy modified infection risk. In a matched subanalysis of 96,004 treated and untreated children with OSA, adenotonsillectomy was not associated with lower 5-year risk of influenza or COVID-19 diagnoses. Influenza risk was slightly higher in the surgery cohort, whereas COVID-19 risk was similar; time-to-event estimates did not demonstrate a protective effect.
The study authors suggest that persistent immune dysregulation in pediatric OSA may underlie the findings. Prior research has documented alterations in both innate and adaptive immune pathways in children with OSA, including changes in T-cell populations and inflammatory signaling, which could compromise host antiviral responses. Additionally, residual OSA after adenotonsillectomy is common, potentially limiting the procedure’s impact on immune function.
Although the absolute risk of viral infection remained relatively low, the consistent relative increase across analyses has practical implications. The authors note that OSA may serve as a clinical “risk marker” for viral susceptibility, and argue that clinicians should prioritize seasonal influenza vaccination and ensure up-to-date COVID-19 immunization in this population.
As the study is retrospective, causality cannot be established. Vaccination status was not available, and children with OSA may have had greater health care contact, increasing the likelihood of diagnostic coding. Nonetheless, the large sample size, longitudinal follow-up, and multiple sensitivity analyses strengthen the observed association.
In summary, children with OSA were at significantly increased risk of influenza and COVID-19 diagnoses—and related pneumonia—over the five-year period, and this elevated susceptibility was not mitigated by adenotonsillectomy alone.
