AHA Scientific Sessions 2026 Abstract · Nov 2, 2026

Nighttime Hemodynamic Phenomapping Identifies Distinct Prognostic Clusters in Sleep-Disordered Breathing

By Sato Y, Yoshihisa A, Sato T et al.

Background: Sleep-disordered breathing (SDB) is characterized by nighttime blood pressure (BP) variability and is associated with heart failure and cardiovascular events. SDB is a heterogeneous condition that includes obstructive and central sleep apnea, which are typically associated with obesity and impaired cardiac function, respectively. Therefore, comprehensive phenotypic characterization may improve prognostic stratification and facilitate precision medicine. However, nighttime hemodynamic phenotypes in patients with SDB have not yet been fully elucidated. Methods and Results: We measured nighttime hemodynamics in 466 patients with SDB using a level 3 portable sleep test device. Beat-to-beat systolic and diastolic BP (SBP and DBP, respectively) were measured based on pulse transit time using plethysmography. Very short-term variability of BP and heart rate (HR) was assessed using the coefficient of variation. Hierarchical cluster analysis was performed using age, apnea-hypopnea index, mean peripheral oxygen saturation, nighttime SBP, DBP, HR, very short-term variability of SBP, DBP, and HR. Based on the dendrogram, patients were classified into three clusters. Phenotypic characteristics are shown in a heatmap ( Figure 1 ). Cluster 3 showed the highest age (clusters 1, 2, and 3: 68.0, 69.0, and 75.0 years, P < 0.001) and apnea-hypopnea index (16.5, 20.3, and 36.0, P < 0.001), and the lowest mean peripheral oxygen saturation (94%, 95%, and 91%, P < 0.001). Regarding nighttime vital signs, cluster 2 exhibited the lowest SBP and DBP, and the highest very short-term variability of SBP and DBP (P < 0.001, respectively). Although the prevalence of heart failure and B-type natriuretic peptide levels did not differ among the three clusters, cluster 2 showed the lowest left ventricular ejection fraction (60.0%, 50.0%, and 62.0%, P < 0.001). Characteristics associated with each cluster are shown in Figure 2 . During the follow-up period of median 1,365 days, there were 65 heart failure hospitalizations and 48 major adverse cardiovascular and cerebrovascular events. Kaplan-Meier analysis demonstrated that event-free survival for both outcomes was lowest in cluster 2 ( Figure 3 , log-rank P = 0.001 and 0.008, respectively). Conclusion: In patients with SDB, phenomapping based on nighttime hemodynamics identified three distinct clusters with different clinical characteristics and prognoses.

doi.org/10.1161/circ.154.suppl_1.5777Biomarkers

AgingResearch.News