Background: Obstructive sleep apnea (OSA) is a prevalent and consequential comorbidity in sickle cell disease (SCD). The nocturnal hypoxemia characteristic of OSA may trigger red blood cell sickling, endothelial activation, and pro-inflammatory cascades, theoretically increasing risk for vaso-occlusive crises and end-organ damage. While this pathophysiological link is established, large-scale contemporary data examining the impact on hospital readmissions and resource utilization are lacking, particularly in homozygous sickle cell (Hb-SS) patients. We investigated the association between OSA and readmissions, mortality, and healthcare utilization in hospitalized adults with Hb-SS disease.

Methods: We conducted a retrospective cohort study using the Healthcare Cost and Utilization Project National Readmissions Database (2018-2022). We included non-elective hospitalizations of adults (≥18 years) with HbSS disease identified using ICD-10 code D57.0, which provides for both HbSS with crisis (D57.00) and HbSS with acute chest syndrome (D57.01). OSA was identified using diagnosis codes G47.33 and G47.39. We performed 1:1 propensity score matching, adjusting for age, sex, hospital characteristics (teaching status, bed size, urban/rural location), illness severity (APR-DRG severity and mortality risk), insurance type, weekend admission, blood transfusion, vaso-occlusive crisis, neighborhood income quartile, and patient location. Primary outcomes were 30-day, 60-day, and 90-day all-cause readmissions. Secondary outcomes included hospital mortality, length of stay (LOS), and hospital charges. We calculated adjusted risk ratios (RR) using survey-weighted Poisson regression and odds ratios (OR) using logistic regression. E-value analysis assessed robustness to unmeasured confounding.

Results: Among 162,549 unweighted HbSS hospitalizations (representing 316,015 weighted admissions), OSA prevalence was 3.0% (n=4,900 unweighted). After propensity matching, 9,582 patients were analyzed (4,791 per group). Baseline characteristics achieved excellent balance, with standardized differences <10% for all covariates (most considerable: age 8.1%, hospital bed size -7.1%, female sex -2.6%, teaching hospital -5.2%).

OSA was associated with significantly increased readmissions: 30-day (23.9% vs 22.0%; adjusted RR 1.10, 95% CI: 1.00-1.21, p=0.040), 60-day (41.1% vs 37.8%; adjusted RR 1.09, 95% CI: 1.03-1.16, p=0.006), and 90-day (50.3% vs 46.7%; adjusted RR 1.08, 95% CI: 1.03-1.13, p=0.003). No statistically significant difference in in-hospital mortality was observed (0.38% vs 0.63%; adjusted OR 0.58, 95% CI: 0.30-1.09, p=0.090). OSA patients experienced significantly longer hospitalizations (6.8 vs. 5.9 days; adjusted ratio, 1.15; 95% CI, 1.09-1.20; p < 0.001) and higher charges per admission ($59,474 vs. $51,027; adjusted ratio, 1.16; 95% CI, 1.08-1.25; p < 0.001). Extrapolating to the weighted population, OSA-associated excess charges totaled approximately $80.1 million annually (9,480 OSA patients × $8,447 excess charge per patient), representing $400.5 million over the 5-year study period. E-values were 1.46 for the point estimate and 1.11 for the lower confidence bound, suggesting moderate robustness to unmeasured confounding. Subgroup analysis revealed stronger readmission associations in patients with minor/moderate illness severity (30-day RR 1.17, 95% CI: 1.05-1.30) versus significant/extreme severity (RR 1.00, 95% CI: 0.85-1.18, p-interaction=0.045).

Conclusions: In this extensive national analysis, OSA was independently associated with an 8-10% increased risk of readmission and a 15-16% higher resource utilization in adults with HbSS disease, without a significant impact on in-hospital mortality. The substantial economic burden of almost $400 million in excess charges over five years. This combined with the disproportionate effect in less severely ill patients, supports systematic OSA screening and treatment as potential strategies to reduce readmissions in SCD. Prospective studies should evaluate whether OSA treatment can mitigate this increased readmission risk and associated healthcare costs.

This content is only available as a PDF.
2025
Sign in via your Institution