Data-Driven Phenotyping of Central Disorders of Hypersomnolence With Unsupervised Clustering

Gool, Jari K.; Zhang, Zhongxing; Oei, M. S. S. L.; Mathias, Stephanie; Dauvilliers, Yves; Mayer, Geert; Plazzi, Giuseppe; Rio-Villegas, Rafael del; Cano, Joan; Šonka, Karel; Partinen, Markku; Overeem, Sebastiaan; Peraita-Adrados, Rosa; Heinzer, Raphaël; Silva, António Martins da; Högl, Birgit; Wierzbicka, Aleksandra; Heidbreder, Anna; Feketeová, Eva; Manconi, Mauro; Bušková, Jitka; Cañellas, Francesca; Barateau, Lucie; Pizza, Fabio; Schmidt, Markus H.; Fronczek, Rolf; Khatami, Ramin; Lammers, Gert Jan

doi:10.1212/wnl.0000000000200519

Cited by 29 publications

(20 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cluster 3 was separated from cluster 4 mostly by sex, but weaker differences were also found for REM latency, mean REM latency, and naps. These results suggest that our approach was able to discriminate, in a data-driven way, NT1 patients from patients with NT2 and IH, confirming previous reports based on different cohorts [14].…”

Section: Re Sults Cdh Cohortsupporting

confidence: 89%

See 1 more Smart Citation

Disentangling the complex landscape of sleep–wake disorders with data‐driven phenotyping: A study of the Bernese center

Aellen

Meer

Dietmann

et al. 2023

Euro J of Neurology

View full text Add to dashboard Cite

Background and purposeThe diagnosis of sleep–wake disorders (SWDs) is challenging because of the existence of only few accurate biomarkers and the frequent coexistence of multiple SWDs and/or other comorbidities. The aim of this study was to assess in a large cohort of well‐characterized SWD patients the potential of a data‐driven approach for the identification of SWDs.MethodsWe included 6958 patients from the Bernese Sleep Registry and 300 variables/biomarkers including questionnaires, results of polysomnography/vigilance tests, and final clinical diagnoses. A pipeline, based on machine learning, was created to extract and cluster the clinical data. Our analysis was performed on three cohorts: patients with central disorders of hypersomnolence (CDHs), a full cohort of patients with SWDs, and a clean cohort without coexisting SWDs.ResultsA first analysis focused on the cohort of patients with CDHs and revealed four patient clusters: two clusters for narcolepsy type 1 (NT1) but not for narcolepsy type 2 or idiopathic hypersomnia. In the full cohort of SWDs, nine clusters were found: four contained patients with obstructive and central sleep apnea syndrome, one with NT1, and four with intermixed SWDs. In the cohort of patients without coexisting SWDs, an additional cluster of patients with chronic insomnia disorder was identified.ConclusionsThis study confirms the existence of clear clusters of NT1 in CDHs, but mainly intermixed groups in the full spectrum of SWDs, with the exception of sleep apnea syndromes and NT1. New biomarkers are needed for better phenotyping and diagnosis of SWDs.

show abstract

Section: Re Sults Cdh Cohortsupporting

confidence: 89%

“…After establishing that our clustering and marker selection pipeline reproduced previously reported findings [14], we proceeded with a larger patient cohort, including the full spectrum of SWDs. Patients The remaining two thirds of cluster 7 were mixed, and cluster 8 contained 28% other CDH patients.…”

Section: Full Cohort Of Swdsmentioning

confidence: 96%

Disentangling the complex landscape of sleep–wake disorders with data‐driven phenotyping: A study of the Bernese center

Aellen

Meer

Dietmann

et al. 2023

Euro J of Neurology

View full text Add to dashboard Cite

show abstract

“…Delta attacks in NT1 mice, likely analogous to sleep attacks in patients with NT1, as well as frequency of cataplectic attacks were more frequent in female mice. Recently, Gool et al [ 17 ] published a cluster analysis of 1078 unmedicated adolescents and adults with central disorders of hypersomnolence from the EU-NN database and found a striking female preponderance within cluster 4 in combination with mild cataplexy, hypnagogic hallucinations, and sleep paralysis. The authors hypothesized as to the existence of a female NT1 subtype.…”

Section: Introductionmentioning

confidence: 99%

Central Disorders of Hypersomnolence: Association with Fatigue, Depression and Sleep Inertia Prevailing in Women

et al. 2022

Self Cite

View full text Add to dashboard Cite

Fatigue, depression, and sleep inertia are frequently underdiagnosed manifestations in narcolepsy and idiopathic hypersomnia. Our cross-sectional study design included diagnostic interview accompanied by assessment instruments and aimed to explore how these factors influence disease severity as well as to elucidate any sex predisposition. One hundred and forty-eight subjects (female 63%) were divided into narcolepsy type 1 (NT1; n = 87, female = 61%), narcolepsy type 2 (NT2; n = 22, female = 59%), and idiopathic hypersomnia (IH; n = 39, female = 69%). All subjects completed a set of questionnaires: Epworth Sleepiness Scale (ESS), Hospital Anxiety and Depression Scales (HADS), Fatigue Severity Scale (FSS), and Sleep Inertia Questionnaire (SIQ). In narcoleptic subjects, questionnaire data were correlated with the Narcolepsy Severity Scale (NSS), and in subjects with idiopathic hypersomnia, with the Idiopathic Hypersomnia Severity Scale (IHSS). The highest correlation in narcoleptic subjects was found between NSS and ESS (r = 0.658; p < 0.0001), as well as FSS (r = 0.506; p < 0.0001), while in subjects with idiopathic hypersomnia, the most prominent positive correlations were found between IHSS and SIQ (r = 0.894; p < 0.0001), FSS (r = 0.812; p < 0.0001), HADS depression scale (r = 0.649; p = 0.0005), and HADS anxiety scale (r = 0.528; p < 0.0001). ESS showed an analogic correlation with disease severity (r = 0.606; p < 0.0001). HADS anxiety and depression scores were higher in females (p < 0.05 and p < 0.01), with similar results for FSS and SIQ scales (p < 0.05 for both), and a trend toward higher ESS values in females (p = 0.057). Our study illustrates that more attention should be focused on pathophysiological mechanisms and associations of fatigue, depression, as well as sleep inertia in these diseases; they influence the course of both illnesses, particularly in women.

show abstract

“…In the Research Article entitled “Data-Driven Phenotyping of Central Disorders of Hypersomnolence With Unsupervised Clustering” by Gool et al, 1 the top-to-bottom order of the values of the color bar in Figure 2 should be as follows: 6σ (dark red), 3σ, 0, -3σ, and -6σ (dark blue). Figure 2 has been replaced by a corrected version.…”

mentioning

confidence: 99%