Phenotyping comorbidity in obstructive sleep apnea syndrome

Background: Phenotyping OSAS9 comorbidity has only recently been attempted for the first time. Our aim was determine phenotypes of comorbidity in OSAS patients employing a data-driven approach. Methods: Data from 1472 consecutive patient records were recovered from our sleep laboratory9s database. Categorical Principal Component Analysis and TwoStep Clustering were employed to detect distinct clusters in the data. Univariate comparisons included One-Way ANOVA with Bonferroni correction and Chi-Square tests. Predictors of pairwise cluster membership were determined via a binary logistic regression model. Results: Six distinct clusters were identified; A: "Healthy, reporting sleeping related symptoms", B: "Mild OSAS without significant comorbidities", C₁: "Moderate OSAS, obesity, without significant comorbidities", C₂: "moderate OSAS with severe comorbidity, obesity and exclusively including stroke", D₁: "severe OSAS and obesity without comorbidity and a 33.8% prevalence of hypertension" and D₂: "severe OSAS with severe comorbidities, along with the highest ESS score and highest BMI". Clusters differed significantly in AHI, DI, AI, Age, BMI, minimum SaO₂,daytime SaO₂ (one-way ANOVA p<0.0001). Binary Logistic Regression determined that older Age, greater BMI, lower daytime SaO₂ and Hypertension were independently associated with an increased risk of belonging in a comorbid cluster. Conclusion: Five distinct phenotypes of OSAS and its comorbidities were identified. Mapping the heterogeneity of OSAS may help identify at-risk groups early; Finally, determining predictors of comorbidity for the moderate and severe strata of these phenotypes implies a need to evaluate these factors when considering treatment options.