Photoelectric Plethysmography In Predicting Clinical Risk And Clustering Of Cardiovascular Risk Models For Obstructive Sleep Apnea Hypopnea Syndrome

Objective: We propose a multi-channel photoplethysmography（PPG）to evaluate the feasibility of cardiovascular risk in patients with obstructive sleep apnea syndrome（OSAHS）.We aimed to establish a clinical subgroups of OSAHS based on the PPG technology using cluster analysis and assess each subgroup,s characteristics.Methods: Patients were selected according to the OSAHS criteria and the American College of Sleep Medicine（AASM）Sleep and Respiratory Events Interpretation Manual.Our research focuses on the pulse wave signals using multi-channel PPGs.We also describe in details the experimental protocol and data acquisiton parameters derived from the PPG technology.The patients with OSAHS who meet the standards of automatic ventilator pressure（APAP）and agree to use APAP were recruited.The APAP pressure-regulating treatment was studied in the patients of OSAHS.The patients of OSAHS could be divided into two groups of high-risk and low-risk based on the cardiac risk cut-off point of 0.5.A Systemic clustering and K-means clustering analysis were performed using parameter of CRI to determine the clinical subtype of OSAHS.Result:1)A total of 150 cases in OSAHS group（52 mild cases,39 moderate cases and 59 severe cases）were studied,and 68 cases in control group.There was no significant difference in sex,age,height,SBP,smoking history,snoring history and comorbidity history among the four groups（P>0.05）.Body weight,BMI,neck circumference,waist circumference and DBP increased in moderate and severe OSAHS group（P<0.05）,and LaSO₂ decreased in moderate and severe OSAHS group（P<0.001）.CRI increased in OSAHS group（P<0.001）.Smoking history,diabetes history,cardiovascular history,age,BMI,neck circumference,waist circumference,AHI,LaSO₂ and CRI can affect the occurrence of hypertension.CRI was the most influential factor（P<0.001,regression coefficient 1.971）.CRI was negatively correlated with height（r=-0.274 P<0.001）,age（r=0.466 P<0.001）,SBP（r=0.254 P<0.001）,LaSO₂（r=-0.200 P=0.003）,AHI（r=0.246 P<0.001）,ODI（r=0.176 P=0.009）.The sensitivity and specificity of CRI were 80.8% and 59.1% respectively,P<0.001.The area under the AUC curve is 0.72,（P<0.001）.2)There was no significant difference in PWA_Ⅰ and PPT among the four groups（P>0.05）.RCRD,PASD,SpO_2Ⅰ and timebelow90 increased in severe OSAHS group（P<0.001）,while PR_Ⅰ decreased in moderate and severe OSAHS group（P<0.001）.CRI was positively correlated with RCRD,PR_Ⅰ,PWA_Ⅰ,PPT,PASD,SpO_2Ⅰ and timebelow90.CRI was not related to LaSO₂ and irregular pulse rate（P>0.05）.AHI was positively correlated with and with the lowest pulse oxygen,ODI,RCRD,PR_Ⅰ,PWA_Ⅰ,PASD,SpO_2Ⅰ and timebelow90.AHI had no correlation with gender,age and PTT（P>0.05）.PWA,PTT and PASD were independent influencing factors of CRI（regression coefficient 0.280,P<0.001;regression coefficient 0.464,P<0.001;regression coefficient 0.332,P<0.005）.3)After APAP treatment,AHI decreased（P=0.001）,LaSO₂ increased（P<0.062）,CRI decreased（P<0.001）.There were no difference in PWA_Ⅰ and Arrihyth.RCRD,PTT,PASD,SpO_2Ⅰ and timebelow90 decreased after APAP treatment（P<0.05）.4)（CRI>0.5 group）wad defined by higher in smoking history,hypertension history,diabetes history,age,waist circumference and SBP（P<0.05）;AHI and ODI were lower than those in the low risk group（CRI<0.5 group）（P<0.05）.The five subgroups that emerged from the cluster analysis as follows: cluster 1 is OSAHS physiological and sleep indicators;cluster 2 is hypertension complications;cluster 3 is bad hobbies;cluster 4 is OSAHS complications;cluster 5 is night blood oxygen indicators.A model with four phenotype clusters provided the best fit: the first cluster of young people with high cardiovascular risk;the second cluster without cardiovascular risk;the third cluster of hypertension and high cardiovascular risk;the fourth cluster of old people with high cardiovascular risk.Conclusion:1)PPG technology can be used to evaluate the cardiovascular risk model of OSAHS patients,because CRI can reflect the severity of OSAHS disease and predict the occurrence of hypertension.2)CRI and derived parameters are superior to AHI in assessing the cardiovascular risk model of OSAHS.They are mainly related to age,autonomic nerve,arterial stiffness,central sleep breathing events,and multiple mechanisms of hypoxia.The risk compass is composed of parameters Objectively predict cardiovascular risk.3)Both CRI and AHI can be improved after the APAP intervention of OSAHS,The parameters of PWA_Ⅰ and Arrihyth can not be changed,which may be the reason why the cardiovascular complications of OSAHS cannot be improved.4)We were established a cardiovascular risk model which CRI was be incorporated,and four major clinical subtypes of cardiovascular complications of OSAHS were established based on the objective indicators of CRI.