| BACKGROUND:Most of the follow-up of diabetic patients are performed in the outpatient department,the quality of outpatient glucose management needs to be improved.Measurement of blood glucose levels in diabetic patients is important to assess patient response to treatment.Continuous glucose monitoring(CGM)provides a full-day blood glucose profile that shows glycemic fluctuation and hypoglycemia characteristics unlike traditional blood glucose monitoring methods.For this reason,the clinical benefits of CGM are leveraged in many outpatient units of hospitals.CGM system includes retrospective system and real-time system,the characteristic and applicable people are different.1.The accuracy of outpatient CGM data is correlated with the clinical benefit of CGM.Therefore,accuracy and reliability of CGM data is important for establishment and implementation of individualized diagnosis and treatment programs.Studies show the accuracy rate of outpatient CGM is low.The premise of improving the accuracy rate is to clarify the influencing factors.At present,there is a lack of research on the factors affecting the accuracy of outpatient CGM.2.It is necessary to use CGM parameters to improve the quality of outpatient service.Besides simple and easy to understand,Time in range(TIR)can independently predict the long-term complications of diabetes.TIR has been recommended by many guidelines for outpatient diagnosis and treatment.The correlation and conversion relationship between TIR and mean blood glucose or glycemic fluctuation are still unclear.OBJECTIVE:1.To summarize the application status of CGM in our hospital in recent years.2.To analyze the common factors affecting the accuracy of iPro-2 retrospective monitoring system,and provide a basis for improving the accuracy of this monitoring system.3.To analyze the common factors affecting the accuracy of Guardian real-time(GRT)monitoring system,and provide a basis for improving the accuracy of this monitoring system.4.To analyze the correlation and conversion relationship between TIR and common glucose metabolism indicators.METHODS:1.Patient data concerning CGM use in Changhai hospital from January 2012 to June2019 was retrieved.The monitoring data is divided into the standard data and the substandard data according to the guidelines of the “Chinese Continuous Glucose Monitoring Clinical Application Guide(2012)”.The accuracy rate in different year and different monitoring system was compared.Patients were classified according to gender,age,course of disease,HbA1 c or educational level,and the number of cases monitored was compared.2.From September 2016 to June 2019,570 T2 DM outpatients received iPro-2monitoring,and 532 patients were enrolled for this study.The standard and substandard data of the first monitoring data were identified according to the relevant guidelines described in the “Chinese Continuous Glucose Monitoring Clinical Application Guide(2012)”.The clinical data of 441 patients who met the standard and 91 patients who did not meet the standard were compared.With standard or substandard as dependent variables,clinical data as independent variables,the clinical characteristics of patients with substandard accuracy were analyzed by logistic regression.We then analyzed the factors leading to substandard data and classified them as patient factors,doctor-patient factors and medical factors.In recent years,our hospital updated the iPro-2 outpatient monitoring record table twice,the rate of factors affecting the accuracy before and after the update were compared.3.From September 2016 to June 2019,178 T2 DM outpatients received GRT monitoring,and 155 patients were enrolled for this study.The standard and substandard data of the first monitoring data were identified according to the relevant guidelines described in the “Chinese Continuous Glucose Monitoring Clinical Application Guide(2012)”.The clinical data of 119 patients who met the standard and 36 patients who did not meet the standard were compared.With standard or substandard as dependent variables,clinical data as independent variables,the clinical characteristics of patients with substandard accuracy were analyzed by logistic regression.We then analyzed the factors leading to substandard data and classified them as patient factors,doctor-patient factors and medical factors.The incidence rates of the three factors in iPro-2 patients and GRT patients were compared.4.A total of 441 T2 DM patients who received iPro-2 monitoring were enrolled.The monitoring accuracy met the stipulated standards and all clinical data was available foreach patient.The TIR target range was set to 3.9-10 mmol/L or 3.9-7.8 mmol/L.The simple correlation between TIR and HbA1 c,mean blood glucose(MBG),mean amplitude of glycemic excursions(MAGE)was analyzed by Pearson or Spearman method.The trend and rate of HbA1 c,MBG and MAGE with TIR variation were analyzed by regression method.The corresponding HbA1 c,MBG and MAGE values were calculated by fitting formulas for TIR(3.9-10)and TIR(3.9-7.8)at values of 0%,10%,...,100%.Results:1.Since 2012,CGM monitoring has been employed in our outpatient clinics.The number of cases monitored has been increasing yearly,and the outpatient monitoring in the whole year of 2018 was 337 cases.During the study period,1084 cases of type 2 diabetes patients(T2DM)were monitored,50 cases of type 1 diabetes patients(T1DM)were monitored;30 cases of gestational diabetes patients were monitored;11 cases of special type of diabetes patients were monitored.During the study period,609 males and 563 females received CGM monitoring,the number of patients whose educational level was <high school or ≥ high school was 622 and 550.The average age,course of disease,HbA1 c were 59.72,10.13 and 7.81%.2.A total of 532 patients were enrolled in the study.There were differences in age,duration of disease and HbA1 c between the two groups.Age and HbA1 c entered the logistic regression equation,and the OR values were greater than 1.There were 109 cases of adverse factors,the incidence rate was 20.49%.The incidence rate related to patient was10.34%,the incidence rate related to both doctor and patient was 4.51%,the incidence rate related to medical treatment was 5.64%.Further analysis showed that patients with old age or HbA1 c ≥8% had a higher incidence of patient factors during surveillance.After the update of iPro-2 monitoring record table,the corresponding data validity problem was significantly reduced.3.A total of 155 patients were enrolled in the study.There were differences in age,HbA1 c and educational level between the two groups.HbA1 c and educational level entered the logistic regression equation,and the OR values were 6.326 and 0.372 respectively.There were 45 cases of adverse factors,the incidence rate was 29.03%.The incidence rate related to patient was 16.77%,the incidence rate related to both doctor and patient was 8.39%,the incidence rate related to medical treatment was 3.87%.Further analysis showed that patients with HbA1c≥8% or educational level<high school had a higher incidence of patient factors and doctor-patient factors during surveillance.Compared with iPro-2 monitoring,the incidence rate of patient factors was higher.4.TIR(3.9-10)was highly correlated with HbA1 c and MBG,moderately correlated with MAGE.TIR(3.9-7.8)was moderately correlated with HbA1 c and MAGE,highly correlated with MBG.The fitting lines of TIR(3.9-10),TIR(3.9-7.8)and HbA1 c were double exponential regression lines,the adjusted R squares were 0.69 and 0.47,respectively.The fitting lines of TIR(3.9-10),TIR(3.9-7.8)and MBG were linear regression lines,the adjusted R squares were 0.78 and 0.56,respectively.The fitting lines of TIR(3.9-10),TIR(3.9-7.8)and MAGE were linear regression lines,the adjusted R squares were 0.44 and0.30,respectively.According to the fitting lines,when MAGE is 3.9 mmol/L,the corresponding TIR(3.9-10)and TIR(3.9-7.8)were 90% and 70%,respectively.Conclusion:1.The number and accuracy rate of patients receiving CGM monitoring in the outpatient clinics of our hospital has been increasing yearly.In our hospital,CGM monitoring is mainly applied to T2 DM patients.2.Patients with old age or HbA1c≥8% have lower CGM accuracy,the reason is that patient operation is not standardizedand,and clinicians should strengthen the guidance and education.Effective management of monitoring records can increase data accuracy.3.Patients with HbA1 c ≥8% or educational level <high school have lower CGM accuracy,the reason is patient’s standardized operation and inadequate doctor-patient communication,clinicians should strengthen the guidance and communication.Compared with iPro-2 monitoring,outpatient GRT monitoring requires higher patient operation ability and education level,and clinicians should carefully select monitoring objects.4.TIR is negatively correlated with glucose metabolism indicators and can reflect mean blood glucose and glycemic fluctuation.The relationship between TIR and HbA1 c is nonlinear,whereas the relationship of TIR with MBG and MAGE is linear.For a given TIR value,the predicted values of HbA1 c,MBG,MAGE can be calculated by fitting the regression line.When TIR(3.9-10)is below 90% or TIR(3.9-7.8)is below 70%,the likelihood of high glycemic fluctuation should be concerned. |
PDF Full Text Request