Unattended sleep studies in pediatric OSA: A systematic review and meta‐analysis

The objective of this study was to systematically assess the evidence for the diagnostic accuracy of unattended type 2 or type 3 multichannel studies for predicting and monitoring pediatric obstructive sleep apnea (OSA). Four medical databases were searched for eligible studies. Only studies that evaluated the ability of unattended multichannel devices to accurately diagnose or monitor OSA in children were included. Those with appropriate data were selected for meta-analysis. Study quality was assessed by using the Quality Assessment Tool for Diagnostic Accuracy Studies tool. Summary estimates of diagnostic accuracy were determined by using sensitivity, specificity, positive and negative likelihood ratios, diagnostic odds ratio, summary receiver operating characteristics, and the area under the curve (AUC). Ten diagnostic studies with 724 patients were included in the systematic review, which was followed by a meta-analysis using studies that provided sufficient information for the calculation of diagnostic parameters. The overall analysis revealed a moderate sensitivity of 76% (95% confidence interval [CI]: 64%-85%), a moderate specificity of 76% (95% CI: 60%-88%), and a pooled diagnostic odds ratio of 15.18 (95% CI: 3.52-65.43). The AUC (0.88) indicated that unattended multichannel devices showed good diagnostic performance for predicting pediatric OSA. Using a cutoff of AHI > 1 for the diagnosis of OSA, the results appeared to be more useful, with a higher sensitivity (88%) while maintaining a moderate specificity (71%). These findings suggest that unattended sleep studies are a good tool for predicting both the presence and severity of OSA in children, especially in those with mild-to-moderate disease.