气象因素与急性闭角型青光眼发作关系的研究

Study on the Relationship Between Meteorological Factors and Acute Angle-closure Glaucoma

目的研究南昌市气象因素与急性闭角型青光眼(AACG)发病的关联。方法收集2015年1月-2019年10月期内南昌市大中型综合型三级甲等医院门诊接诊的AACG病例资料及其当期的气象材料(日平均温度Ta、日最高温度Th、日最低温度T1、日均值气压Pa、日最大气压Ph、日最小气压P1、日均值空气湿度Ha、日最低湿度H1)。采用圆分布剖析AACG病发的集中化趋势;用单因素相关及多元逐步回归分析8项气象因子与AACG关联性。选用时间序列分析,以2015年1月1日-2018年10月31日(共46个月)的数据监测为模型数据信息,以2018年11月1日-2019年10月31日(共12个月)的数据监测为验证数据信息,建立模拟气象因素与AACG逐月就诊人次数的自回归移动平均模型(ARIMA),预测分析AACG的就诊人数。结果共收集AACG就诊患者226例,冬天较多发且每年均有显著的集中化趋势。圆分布分析结果显示4年累计病发的集中趋势均值夹角341.6771°,即12月12日;单因素相关性分析结果显示AACG与Ta、Th、T1呈负相关(r=-0.216、-0.167、-0.171,P<0.05),与Pa、Ph、P1、Ha、Hl呈正相关(r=0.048、0.049、0.041、0.079、0.074;P=0.157、0.034、0.077、0.022、0.052);多元逐步回归分析显示,Ta、Th、T1、Hl与AACG就诊人次数有关(P<0.05);建立气象因素与AACG月就诊人次数预测模型:ARIMA(1,0,0)(2,0,0)_(12),标准化BIC=0.651、平稳R^(2)=0.658、均方根误差=1.124、Ljung-BoxQ=9.762,P=0.834,残差序列为白噪声,将该模型进一步预测分析运用,预测值的动态趋势与具体情况较为一致,各月具体值均落入预测值的可信区间内,模型对AACG就诊的预测分析实际效果比较理想。结论 AACG冬季多发,其病发与温度、气湿有关,气象因素之间协同作用影响AACG的发生。温度剧变造成AACG发生的风险升高。ARIMA(1,0,0)(2,0,0)_(12)对AACG的发病预测效果比较理想,能用来预测分析AACG月就诊状况,实现警示预报功效。

Objective To study the relationship between meteorological factors and acute angle-closure glaucoma(AACG)in Nanchang.Methods Collected AACG case data and current meteorological data from the outpatient clinics of large and medium-sized comprehensive tertiary first-class hospitals in Nanchang from January 2015 to October 2019(Daily average temperature Ta,daily maximum temperature Th,daily minimum temperature Tl,daily average atmospheric pressure Pa,daily maximum atmospheric pressure Ph,daily minimum atmospheric pressure Pl,daily average air humidity Ha,daily minimum humidity Hl).The circular distribution was used to analyze the centralization trend of AACG;single factor correlation and multiple stepwise regression were used to analyze the correlation between 8 meteorological factors and AACG.Time series analysis is selected,and the data monitoring from January 1,2015 to October 31,2018(46 months in total)is used as the model data information.Using the data monitoring from November 1,2018 to October 31,2019(12 months in total)as the verification data information,establishing an autoregressive moving average model(ARIMA)that simulates meteorological factors and the monthly number of visits by AACG to predict and analyze the number of visits to AACG.Results A total of 226 patients treated by AACG were collected,which occurred frequently in winter and showed a significant trend of centralization every year.The results of circle distribution analysis showed that the mean angle of the central tendency of the cumulative incidence of illness in 4 years was 341.6771°,that is,December 12;the results of single factor correlation analysis showed that AACG was negatively correlated with Ta,Th,and Tl(r=-0.216,-0.167,-0.171,P<0.05),positive correlation with Pa,Ph,Pl,Ha,Hl(r=0.048,0.049,0.041,0.079,0.074;P=0.157,0.034,0.077,0.022,0.052);Multiple stepwise regression analysis showed that Ta,Th,T1,Hl were related to the number of AACG visits(P<0.05);Establish a prediction model for meteorological factors and the number of AACG monthly visits:ARIMA(1,0,0)(2,0,0)12,standardized BIC=0.651,stable R2=0.658,root mean square error=1.124,Ljung-BoxQ=9.762,P=0.834,the residual sequence was white noise,and the model was further used for prediction and analysis,the dynamic trend of the predicted value was more consistent with the specific situation,and the specific value of each month fell within the confidence interval of the predicted value.The actual effect of the model on the prediction and analysis of AACG visits was relatively ideal.Conclusion AACG occurs frequently in winter,and its occurrence is related to temperature and humidity.The synergistic effect of meteorological factors affects the occurrence of AACG.The sudden change in temperature increases the risk of AACG.ARIMA(1,0,0)(2,0,0)12 has an ideal predictive effect on the incidence of AACG,and can be used to predict and analyze the monthly medical visits of AACG,and realize the warning forecast function.