Relationship of Obstructive Sleep Apnea and Sickle Cell Disease Severity in Children

Abstract

Statement of the Problem: Obstructive sleep apnea (OSA) initiates hypoxemia and elevated inflammatory markers, events which contribute to vaso-occlusion in sickle cell disease (SCD) independent of OSA. Children with SCD are at increased risk of developing OSA due to SCD-related adenotonsillar hypertrophy, and OSA may increase vaso-occlusion and SCD severity through hypoxemia and inflammation. Identification of OSA-related factors influencing SCD severity is important to reduce SCD-related complications. Purpose: The purpose of this study was to test the hypotheses that SCD severity is associated with OSA, is associated with OSA-related changes in polysomnography parameters and treatment of OSA with adenotonsillectomy decreases SCD severity. Methods: A case series study was conducted collecting data from the medical record. Subjects were children aged 2-18 years with SCD, referred to a sleep laboratory for evaluation of OSA. Outcome measures included rates of help-seeking behaviors (medical contacts and days of care) for vaso-occlusive crises at 2 time points, and a Sickle Cell Disease Severity Index (DSI) score. Analyses were descriptive and nonparametric, with exploratory regression modeling. Results: Younger children had more severe OSA (r=-.314, p=.017), and older children had more severe SCD (medical contacts, r=.303, p=.021; days of care, r=.369, p=.007). Underweight adolescents had the highest obstructive apnea-hypopnea indexes (OAHI). Medical contacts decreased as OSA severity increased, H(2)=7.85, p<.05; and days of care showed a decreasing trend with increasing OSA severity, r=-.26, p<.01. Medical contacts were negatively associated with OAHI (p<.01), respiratory arousal index (p<.01) and peak end-tidal CO2 (p<.05); days of care were negatively associated with OAHI (p<.05); and DSI scores were negatively associated with mean sleep oxyhemoglobin saturation (p<.01) and sleep efficiency (p<.05). Children with OSA having adenotonsillectomy increased help-seeking during the year following surgery (NS; medical contacts, r=-.32; days of care, r=-.35). Conclusion: Contrary to hypothesized relationships, OSA severity was negatively associated with help-seeking during the year preceding polysomnography, and help-seeking increased following adenotonsillectomy. The relationship between OSA and SCD severity is complex, with certain subgroups more severely affected. Screening children with SCD for OSA should be routine, and threshold for polysomnography should be low. Post-adenotonsillectomy, children with SCD should be followed closely.