OBJECTIVE: To elucidate the relationship between different cephalometric measurements and the severity of childhood OSAHS, then analyze the value of cephalometric measurements in predicting the severity of OSAHS in children. METHOD: Sixty-one patients received PSG examination and standard lateral neck roentgenography including cervical trachea. Cephalometric measurements include adenoid size, adenoidal-nasopharyngeal ratio, narrowest posterior airway space of nasopharynx, ratio of the narrowest posterior airway space of nasopharynx and trachea diameter. Statistical analysis was made to study the relationship between the data of Cephalometric measurements and PSG. RESULT: Cephalometric measurements were all associated with AHI and AI (P < 0.01). Only PAS/TD variable was entered in the stepwise linear regression equation. ROC curve analysis revealed that the optimal probability cut-off in predicting moderate and severe OSA using PAS/TD ratio was 0. 658, with the area under the curve being 0.799. The corresponding sensitivity and specificity were 81.8% and 76.5% respectively. There was no correlation between cephalometric measurements and LSaO2 (P > 0.01). CONCLUSION: PAS/TD is useful and simple in predicting the severity of OSAHS in children, but no use in predicting the severity of arterial oxygen saturation.
OBJECTIVE: To elucidate the relationship between different cephalometric measurements and the severity of childhood OSAHS, then analyze the value of cephalometric measurements in predicting the severity of OSAHS in children. METHOD: Sixty-one patients received PSG examination and standard lateral neck roentgenography including cervical trachea. Cephalometric measurements include adenoid size, adenoidal-nasopharyngeal ratio, narrowest posterior airway space of nasopharynx, ratio of the narrowest posterior airway space of nasopharynx and trachea diameter. Statistical analysis was made to study the relationship between the data of Cephalometric measurements and PSG. RESULT: Cephalometric measurements were all associated with AHI and AI (P < 0.01). Only PAS/TD variable was entered in the stepwise linear regression equation. ROC curve analysis revealed that the optimal probability cut-off in predicting moderate and severe OSA using PAS/TD ratio was 0. 658, with the area under the curve being 0.799. The corresponding sensitivity and specificity were 81.8% and 76.5% respectively. There was no correlation between cephalometric measurements and LSaO2 (P > 0.01). CONCLUSION:PAS/TD is useful and simple in predicting the severity of OSAHS in children, but no use in predicting the severity of arterial oxygen saturation.