基于可穿戴系统的体位/体动信息量化及其与心力衰竭患者NYHA分级关联分析

Quantitative analysis of body position/physical activity information and its correlation with NYHA class in patients with heart failure based on wearable system

背景日常活动量减少和运动功能受限是心力衰竭患者的特征性表现之一,体位/体动信息与心衰患者疾病严重程度和预后密切相关。通过可穿戴生理监测系统量化体位/体动信息或可作为一种潜在的心衰病情严重程度定量评价手段,其与纽约心脏病协会(New York Heart Association,NYHA)心功能分级的关系需进一步研究。目的探讨心衰患者体位/体动信息定量分析结果与NYHA分级的相关性。方法纳入2021年5月—2022年11月在四川大学华西医院心内科住院的心衰患者,通过可穿戴生理监测系统采集患者入院当天和出院前1 d各24 h的连续生理监测数据,同步收集临床数据。通过对可穿戴生理监测系统内的三轴加速传感器信息进行处理分析,计算卧床时间、活动时间、步数、睡眠翻身次数4个体位/体动指标。基于患者入院时NYHA分级、入院和出院情况、出院时NYHA分级改善与否进行分组,分析体位/体动指标与NYHA分级的关联性。结果纳入心衰患者69例,平均年龄(60.90±14.24)岁,其中男性40例,NYHAⅡ、Ⅲ、Ⅳ级的患者分别有9例、24例、36例。随着NYHA分级的升高,心衰患者全天的卧床时间占比逐渐增多,而全天的活动时间占比、平均每小时步数逐渐降低,以上3个指标在NYHAⅡ、Ⅲ、Ⅳ级间均有统计学差异(P均<0.05);其中卧床时间占比(r_(s)=0.319,P=0.008)与NYHA分级呈正相关,活动时间占比(r_(s)=-0.312,P=0.009)、平均每小时步数(r_(s)=-0.309,P=0.010)与NYHA分级存在负相关。出院时的卧床时间占比显著低于入院时(96.25%vs 97.63%,P=0.026);出院时的活动时间占比显著高于入院时(3.32%vs 1.78%,P<0.001);出院时的平均每小时步数显著高于入院时(97.17步/h vs 35.58步/h,P<0.001);其中出院时NYHA改善组患者的体位/体动指标变化趋势同上,未改善组仅出院时的平均每小时步数显著高于入院时,NYHA改善组的出入院平均每小时步数变化值显著高于未改善组(71.21步/h vs 21.31步/h,P=0.003)。结论可穿戴生理监测系统能够对心衰患者的体位/体动信息进行客观长程的监测,心衰患者的卧床时间与NYHA分级呈正相关关系;活动时间、步数与NYHA分级呈负相关关系,这些体位/体动指标或可作为心衰患者疾病严重程度分级和状态监测评估的有用指标,未来可进一步延伸到对患者的居家和长程监测。

Background Heart failure patients typically exhibit decreased daily activity and limited motor function,body position/physical activity information is closely related to disease severity and prognosis of patients with heart failure.Quantifying body position/physical activity information through wearable physiological monitoring systems can be used as a potential quantitative evaluation method for the severity of heart failure,and its relationship with New York Heart Association(NYHA)heart function class requires further research.Objective To investigate the correlation between body position/physical activity information and NYHA class in patients with heart failure.Methods This study included heart failure patients hospitalized in the Department of Cardiology,West China Hospital of Sichuan University from May 2021 to November 2022.The 24-hour continuous physiological monitoring data were collected by the wearable physiological monitoring system on the day of admission and one day before discharge,and clinical data were also collected simultaneously.Four body position/physical activity indicators were calculated by processing and analyzing the information of the three-axis acceleration sensor in the wearable physiological monitoring system,including bed rest time,activity time,step number,and the number of toss-and-turn events.Patients were grouped based on the NYHA class at admission,admission and discharge,and whether the NYHA class improved at discharge.Subsequently,the correlation between body position/physical activity indicators and NYHA class was analyzed.Results A total of 69 patients with heart failure were involved,including 40 males,with an average age of(60.90±14.24)years old and there were 9 patients in the NYHA classⅡgroup,24 patients in classⅢ,and 36 patients in classⅣ.With the increase of NYHA class,the proportion of bed rest time throughout the day gradually increased,while the proportion of activity time and the average number of steps per hour gradually decreased,the above three indicators showed significant differences between NYHAⅡ,Ⅲ,andⅣclass(P<0.05).The proportion of bed rest time(r_(s)=0.319,P=0.008)was positively correlated with NYHA class,while the proportion of activity time(r_(s)=-0.312,P=0.009)and the average number of steps per hour(r_(s)=-0.309,P=0.010)were negatively correlated with NYHA class.The proportion of bed rest time at discharge was significantly lower than that at admission(96.25%vs 97.63%,P=0.026),while the proportion of activity time at discharge was significantly higher than that at admission(3.32%vs 1.78%,P<0.001),the average number of steps per hour at discharge was significantly higher than at admission(97.17 steps/h vs 35.58 steps/h,P<0.001).The trend of changes in body position/physical activity indicators in the improved group was the same as above.In patients with no improvement,only the average number of steps per hour at discharge was significantly higher than that at admission,but the enlargement was greater in the NYHA-improved group(71.21 steps/h vs 21.31 steps/h,P=0.003).Conclusion The wearable physiological monitoring system can objectively monitor the body position/physical activity information of patients with heart failure for a long time.The bed rest time is positively correlated with NYHA class,while activity time and step number are negatively correlated with NYHA class.These body position/physical activity indicators may be used as useful indicators for disease severity classification and state monitoring evaluation of patients with heart failure,which can be further extended to home and longterm monitoring of patients in the future.