engleski

Cluster analysis of specific sleep apnoea phenotypes in women: results from the esada cohort

Objectives/Introduction: Cluster analysis in Obstructive Sleep Apnoea (OSA) patients are conducted in order to reveal the diversity and heterogeneity of the disease and allow personalized management. The current study aimed to identify specific OSA phenotypes in women in the ESADA pan European database. Methods: Latent class analysis was applied to data from 9, 710 females among the 32, 700 ESADA participants. Clusters were built and identified by including the following variables: age, Body Mass Index (BMI), Apnea Hypopnea Index (AHI), Epworth Sleepiness Scale (ESS), co-morbidities (cardiovascular, pulmonary, psychiatric, metabolic, other). Comparisons between clusters were performed by using non-parametric Kruskal– Wallis tests for quantitative variables and chi- square or Fisher's exact test for qualitative variables. Results: Four clusters were identified: Cluster 1 “Females with ischemic heart disease” (38.3%): the largest cluster characterized by ischemic heart disease (56%), middle aged women, overweight, with moderate OSA, median AHI: [IQR] 22.9/h [17.4 ; 30], non-sleepy(Epworth Sleepiness Scale (ESS): 9 [5 ; 12]). Cluster 2 “Elderly, females with co-morbidities” (23%): elevated AHI 46/h [30 ; 60.1], nocturnal hypoxia with time spent at SaO2 < 90%: 40.6 min [10.1 ; 104.7], non-sleepy (ESS: 9 [6 ; 13]) and with high degree of co-morbidities (hypertension, diabetes, hyperlipidemia, COPD, neurologic, others). Cluster 3: “Sleepy obese females” (16.2 %): middle aged, presenting the highest BMI: 43 [37.6 ; 48.9] kg/m2 and AHI: 53.3/h [32 ; 80.5], most sleepy (ESS: 12 [8 ; 16]), with essentially associated psychiatric disease (22.7%) and asthma (13.9%). Cluster 4: “Females with insomnia and mild OSA” (22.5 %): middle aged, overweight with low AHI, with insomnia and non-sleepy (ESS: 9 [5 ; 13]), presenting low frequency of co-morbidities. Positive airway pressure (PAP) was prescribed to the majority of patients in Cluster 1 (77.3%), Cluster 2 (92.1%) and Cluster 3 (91.7%). In Cluster 4, 23.8% were treated with PAP and 15% with mandibular advancement devices (MAD). CPAP adherence > 4 h was highest in Cluster 1 (85.5%) and Cluster 2 (87%). Conclusion: Four distinct clinical OSA phenotypes were identified in the female patients of a large pan-European database highlighting the importance of gender-based phenotypes and their impact on patient's risk stratification and individualized treatment.

sleep apnoea phenotypes ; women ; cluster analysis

nije evidentirano