Multivariate models in sample with anatomic, Pcrit and activated EMGgg measures
| Variable | OR (95% CI)* | p Value† |
|---|---|---|
| Model 0: Covariates only | ||
| N OSAS=32, N Controls=23, AUC=0.680 | ||
| Age | 1.33 (0.71 to 2.50) | 0.368 |
| Male | 2.72 (0.81 to 9.07) | 0.105 |
| African American | 0.85 (0.20 to 3.63) | 0.822 |
| BMI Z-score | 1.34 (0.77 to 2.33) | 0.301 |
| Model 1: Anatomy adjusted for covariates‡ | ||
| N OSAS=32, N Controls=23, AUC=0.844 | ||
| ATV | 4.57 (1.53 to 13.6) | 0.006 |
| NPAV | 0.45 (0.20 to 1.02) | 0.054 |
| Model 2: Pcrit adjusted for covariates‡ | ||
| N OSAS=32, N Controls=23, AUC=0.841 | ||
| Activated Pcrit | 2.04 (0.93 to 4.47) | 0.077 |
| Hypotonic Pcrit | 2.26 (1.04 to 4.90) | 0.039 |
| Model 3: Activated EMG adjusted for covariates‡ | ||
| N OSAS=32, N Controls=23, AUC=0.747 | ||
| EMGgg slope (activated) | 6.74 (1.05 to 43.1) | 0.044 |
| Model 4: Anatomy+Pcrit+activated EMGgg adjusted for covariates‡ | ||
| N OSAS=32, N Controls=23, AUC=0.958 | ||
| ATV | 9.05 (1.56 to 52.5) | 0.014 |
| NPAV | 0.04 (0.004 to 0.45) | 0.009 |
| Activated Pcrit | 7.72 (1.40 to 42.6) | 0.190 |
| Hypotonic Pcrit | 1.92 (0.66 to 5.60) | 0.231 |
| EMGgg slope (activated) | 33.8 (1.17 to 980.7) | 0.040 |
| Final model: significant variables adjusted for covariates‡ | ||
| N OSAS=32, N Controls=23, AUC=0.951 | ||
| ATV | 10.0 (1.78 to 56.4) | 0.009 |
| NPAV | 0.05 (0.01 to 0.43) | 0.007 |
| Activated Pcrit | 8.79 (1.70 to 45.6) | 0.010 |
| EMGgg slope (activated) | 24.0 (1.10 to 521.1) | 0.043 |
*OR associated with a 1 SD increase in continuous variables or between groups for categorical variables.
†p Value from logistic regression model.
‡Model includes listed predictors and covariates shown in Model 0. Significant estimates (p<0.05) are shown in bold.
ATV, adenotonsillar volume; AUC, area under the receiver operating characteristic curve; BMI, body mass index; EMGgg, genioglossal electromyogram; NPAV, nasopharyngeal airway volume; OSAS, obstructive sleep apnoea syndrome; Pcrit, critical closing pressure.