| Background:According to epidemiological investigation,patients with diabetes mellitus(DM)often have multiple co-morbidities and complications,and more and more diabetic patients are hospitalized in non-endocrine departments for reasons other than blood sugar.The phenomenon of both chronic obstructive pulmonary disease(COPD)and DM in one patient is more common in inpatients with COPD.When patients coexist with COPD and DM at the same time,they are more likely to be hospitalized repeatedly due to their condition.The higher rate of multiple hospitalization will not only greatly increases the financial burden of patients,but also poses great challenges to the quality of life and life expectancy of patients.At present,there are few studies on the characteristics of blood glucose spectrum of hospitalized patients with COPD complicated with DM and the risk factors of multiple hospitalizations in a short time.Objective:To investigate the characteristics of blood glucose spectrum in COPD patients with DM and their risk factors for short-term multiple hospitalization,and find out the possible intervention links that can reduce patients’ short-term multiple hospitalizations,so as to provide theoretical basis for short-term effective blood glucose management and improvement of clinical prognosis for hospitalized COPD patients with DM.Patient and Methods:Base on the data application platform for respiratory system,patients with COPD,aged ≥18 years old,admitted to the respiratory Department of the First Affiliated Hospital of Guangzhou Medical University from January 2017 to December 2021,with complete clinical data,were enrolled in.They were divided into diabetic group(group A,n=659)and non-diabetic group(group B,n=3597).Propensity score matching(PSM)was used to divided A and B into two groups with 1:1 matching: group Ap(n=652)and group Bp(n=652).Simultaneously,In the study subjects obtained after matching,according to whether multiple hospitalization,they were divided into single hospitalization group after matching(group P1,n=1236)and multiple hospitalization group after matching(group P2,n=68).Group A was divided into a single hospitalization group(group A1,n=598)and a multiple hospitalization group(group A2,n=61)according to whether the patients were hospitalized several times within 90 days after discharge.We applied an automated natural language processing(NLP)system to extract clinical information from the electronic health records(EHRs),and the data of bedside point of care glucose testing(POCT)were collected at the same time.The Glucometrics model,including Population model,Patient model,Patient-Day model,was used to calculate the incidence of adverse events(hypoglycemia,hyperglycemia,and severe hyperglycemia)during hospitalization.SPSS 26.0 software was used for statistical analysis.Measurement data conforming to normal distribution were expressed as mean ± standard((?)±s)deviation,and comparison of mean was performed by t test.The measurement data of non-normal distribution were expressed as median and quartile,and the rank sum test was used to compare the data between the two groups.Counting data is expressed as a percentage,and the comparison of counting data is performed by chi-square test.The above methods were used to compare the differences in demographic and clinical data indicators between groups A and B,Ap and Bp,P1 and P2,A1 and A2,and Chi-square test was used to compare the differences in the incidence of hypoglycemia,hyperglycemia and severe hyperglycemia between A1 and A2.Univariate COX regression analysis was performed for P1 and P2 groups and A1 and A2 groups,respectively.The indicators with statistical differences in univariate analysis were taken as covariates,and multiple hospitalizations were taken as dependent variables.Multivariate COX regression analysis was used for statistical analysis of risk of multiple hospitalizations,and the HR value and 95%CI were calculated.P<0.05 indicates statistical difference.Result:1.Comparison of demographic and clinical data between group A and Group BBefore PSM,there were statistical differences in age and COPD course distribution between group A and group B(P<0.05).The proportions of group A complicated with kidney disease,hypertension,coronary heart disease,cerebrovascular accident and sequelae were significantly higher than those of group B(P<0.05).There was no statistical significance in gender,pulmonary heart disease,smoking history,proportion of multiple hospitalizations and the number of days between multiple hospitalizations between the two groups(P>0.05).After 1:1 PSM was performed on group A and group B,1304 patients were successfully matched,including:1.1 652 patients in Ap group and 652 patients in Bp group,and the distribution of confounding factors between the two groups was balanced(P>0.05).The proportion of multiple hospitalizations within 90 days in the Ap group was significantly higher than that in the Bp group,and the number of hospitalization days and hospitalization costs in the Ap group were significantly lower than those in the Bp group.There was no significant difference in the number of days between multiple hospitalizations between Ap and Bp groups(P>0.05).1.2 Compared with P1 group,the proportion of the duration of COPD ≥ 20 years,complicated with DM,hypertension and coronary heart disease in P2 group was significantly higher than that in P1 group(P<0.05).Hospitalization days and hospitalization costs in group P2 were significantly higher than those in group P1(P<0.05).There was no statistical significance in age,sex,complications of kidney disease,cerebrovascular accident and sequelae,pulmonary heart disease and smoking history between the two groups(P>0.05).2.Comparison of demographic and clinical data between group A1 and Group A2The proportion of COPD course ≥20 years,the proportion of Hb A1c>7.0%,the proportion of combined hypertension,fasting blood glucose at admission and hospitalization cost in group A2 were significantly higher than those in A1 group(P<0.05).Age,gender,previous history of DM,course of DM,BMI,complicated kidney disease,CHD,cerebrovascular accident,pulmonary heart disease,smoking history,serious illness,admission mode,clinical outcome,hypoglycemic regimen,changes in hypoglycemic regimen during hospitalization,use of systemic sex hormones,SDBG,LAGE,CV,average daily blood glucose,and blood output There was no statistical significance in fasting blood glucose and length of stay in hospital(P>0.05).3.COX regression analysis of multiple hospitalizationsIn the comparison between P1 group and P2 group after matching propensity scores,univariate COX regression analysis variables with P<0.05(combined with DM,hypertension,coronary heart disease and COPD)were used as covariates,and multiple hospitalizations were used as dependent variables to conduct multivariate COX regression analysis.Results: The risk of multiple hospitalizations was 5.949 times higher in COPD patients with DM than in those without DM(HR=5.949,95%CI 3.039-11.646,P<0.001).COPD patients with hypertension had 2.142 times the risk of multiple hospitalizations compared with non-hypertensive COPD patients(HR=2.142,95%CI 1.215-3.777,P=0.008).The risk of multiple hospitalizations in patients with COPD ≥20 years was 1.899 times higher than that in patients with COPD 0-9 years(HR=1.899,95%CI 1.041-3.463,P=0.039).In the comparison between group A1 and Group A2,univariate COX regression analysis(P<0.05)variables(COPD course,hypertension,admission fasting blood glucose,Hb A1c)and study observation variables CV,SDBG,LAGE were taken as covariables,and multiple hospitalization was taken as dependent variables to perform multivariate COX regression analysis.Results: Patients with Hb A1c≥7% had 2.223 times the risk of multiple hospitalizations compared with patients with Hb A1c<7%(HR=2.223,95%CI 1.276-3.873,P=0.005).For every 0.1mmol/L increase in admitted fasting blood glucose,the mean risk of multiple hospitalizations increased by 0.104 times,and re-hospitalizations occurred within a shorter time(HR=1.104,95%CI 1.025-1.188,P=0.009).The risk of multiple hospitalization was 2.201 times higher in patients with COPD ≥20 years than in patients with COPD <20 years(HR=2.201,95%CI 1.283-3.778,P=0.004).The risk of multiple hospitalization in patients with hypertension was1.929 times higher than that in patients without hypertension(HR=1.929,95%CI 1.085-3.431,P=0.025).There was no statistical significance in analysis of CV,SDBG and LAGE between the two groups.4.Overall situation of adverse glucose events in group A during hospitalizationBedside glucose data of 155,940 times were obtained for 4827 days in 659 patients.Three types of Glucometrics models were used to evaluate the incidence of adverse events in blood glucose:4.1 In the Patient model,the incidence of hypoglycemia,hyperglycemia and severe hyperglycemia in group A1 or A2 was the highest among the three models,which were 16.91/14.75%,79.40/80.32% and 58.49/52.46%,respectively.4.2 The Population models were 1.19/1.61%,21.89/24.94%,11.60/13.16%,respectively.Patient-day model was 3.39%/4.59%,39.85%/40.64%,22.92/22.45%,respectively.4.3 In the Population model,the incidence of hyperglycemia in group A2 was significantly higher than that in group A1(P=0.005).There was no significant difference in the incidence of hypoglycemia,hyperglycemia and severe hyperglycemia among the other models.Conclusions:1.Compared with COPD patients without diabetes,patients with COPD and diabetes were more likely to have multiple hospitalizations within 90 days.Among these patients,patients with higher admission fasting glucose,higher Hb A1 c,longer duration of COPD,and combined hypertension also had a higher risk of multiple hospitalizations.2.There was a higher incidence of hyperglycemic adverse events in COPD patients with DM who had multiple hospitalizations,and attention should also be paid to the adverse effects of hypoglycemic events. |
PDF Full Text Request