Influence Of Geographical Environment On Reference Value Of Time Domain Analysis Index Of Heart Rate Variability

At present, the heart rate variability has become one of effective prognosis of cardiovascular disease. Time domain analysis as a long-term detection method of heart rate variability including some commonly used indicators, such as standard deviation of the RR-intervals (SDNN), mean standard deviation of the RR-intervals (SDANN), root mean square of successive differences (RMSSD), and proportion of interval differences of successive intervals greater than 50 ms (pNNso). However, in clinical application, there has not yet been established suitable standard for each time domain index of healthy Chinese, but quoted from foreign standards, which seriously affected the veracity of diagnosis and treatment of the disease for Chinese. In order to establish criteria of heart rate variability time domain index reference values, which are more suitable for physique of healthy Chinese, the relationship between factors, such as age, gender, weight, breathing, emotions, circadian rhythm, smoking and alcoholism, and heart rate variability time domain index reference values has been studied and reported. Just taking Tibet, Qinghai as examples, the influence of altitude on heart rate variability also has been preliminarily explored. These studies indicated that there is a certain relationship between geographical environment and heart rate variability. However, the combined effects of geographical factors on middle-aged and elderly Chinese have not been studied. Therefore, this paper explored the influence of Chinese geographical environment on middle-aged and elderly people which were made as research object, so as to provide the basis for establishing the medical standard reference value of each index.This paper collected and collated samples through the method of literature retrieval with the keywords "Age", "Sample size", "Reference value", "Province", "City", and grouped by age in order to strictly select healthy subjects more than 50 years old from 118 cities of China as subjects. Besides,9 geographical factors of these cities were collected accurately, including topography indicator:the altitude; climate indicators:the average annual temperature, annual sunshine hours, annual precipitation, annual average relative humidity, annual average wind speed, annual range of air temperature; soil indicators:the topsoil organic matter and the topsoil pH. Firstly, the study made linear and nonlinear correlation analysis between geographical factors and each heart rate variability time domain index by SPSS19.0 software to choose geographical factors which have correlation with time domain index. Then predictive models between the relevant geographical factors and each time domain index were established by using multiple linear regression analysis and curve estimation analysis. In addition, by using the combination forecasting, the study got combination curve models, which contains more than one geographical factor. And then, the paper selected the best predictive model of each time domain index by comparing multiple linear regression prediction model with combination curve model. Finally, the heart rate variability time domain reference values of middle-aged and elderly Chinese of 2322 cities could be predicted by the optimal prediction models. The distribution map of time domain reference values also could be fitted out by using disjunctive kriging interpolation of ArcGIS 10.0 software.The results of correlation analysis showed that there exist significant negative correlation between every heart rate variability time domain index and altitude in both the linear and nonlinear correlation analysis, which implied that altitude is the main influential factor. And the average annual temperature, annual precipitation and annual average relative humidity also affect the SDNN and pNN50 reference value. The comparison of two kinds of predictive models displayed that multivariate linear regression model was more suitable for SDNN and pNN50 reference, combination curve forecast model was the optimal model for SDANN and RMSSD reference value. Seen from the distribution map, heart rate variability time domain indexes showed a kind of gradually reduced trend from east to west. That is to say, the indexes decreased with the increase of altitude, significant difference showed from east to west and non-significance existed from north to south.This paper established combination curve forecast model by using the combination forecast theory to better reflect the complex influence of geographical factors on time domain reference value. The accuracy of predictive model was also improved. Understanding the effect of geographical factors on heart rate variability time domain indexes of middle-aged and elderly Chinese could facilitate medical workers to clinical application.