The Efficacy And Safety Of Short-term CSII Intensive Treatment In Different Durations Of T2DM

Objective: To evaluate the efficacy and safety of short-term intensive CSIItreatment in different durations of T2DM.Methods:1. To evaluate627T2DM inpatients in the endocrinology department in DalianCentral Hospital, from March2009to March2011. According to patient’s duration ofT2DM, we placed214patients with T2DM less than3years in group A,87patientswith T2DM3~5years in group B,149patients with T2DM5~10years in group C and177patients with T2DM more than10years in group D. All the patients were undershort-term intensive CSII treatment.2. To study on the efficacy and safety of short-term intensive CSII treatment in thefour groups of patients. The indices of curative effect were blood sugar success rate,blood sugar standard deviation, number of days patients wearing pump and insulindosage. The index of Safety was incidence of hypoglycemia.3. To evaluate the correlation between blood sugar standard deviation and insulindosage. The influencing factors were age, duration of diabetes, height, waist, hipcircumference, waist-hip ratio, weight, BMI, HbA1c,24MA, C-P,2h C-P, number ofdays patient wearing pump. Evaluating the influencing factor of blood sugar successrate, which were gender, age, duration of diabetes, height, waist, hip circumference,waist-hip ratio, weight, BMI, HbA1c,24MA, C-P,2h C-P, incidence of hypoglycemiaand insulin dosage.Results:1. Criteria which measure the safety and efficacy of short-term intensive CSIItreatment.(1) Blood sugar success rate: Blood sugar success rate for group A, B, C and Dwere74.77%,63.22%,65.77%and62.71%respectively. Group A’s blood sugar success rate is significantly higher than other three groups. Difference from comparingevery two groups among the four groups have statistical significance with χ2=9.013,P<0.05.(2) Blood sugar standard deviation: Blood sugar standard deviation for group A, B,C and D were (1.79±0.84) mmol/L,(2.05±0.86) mmol/L,(2.25±0.98) mmol/L, and(2.03±0.96) mmol/L respectively. Group A’s blood sugar standard deviation issignificantly lower than other groups. The difference between group A and the othergroups has statistical significance with P<0.05, P<0.01, P<0.01respectively.(3) Number of days patients wearing pump: Number of days patients wearingpump for group A, B, C and D were (7.31±2.56),(7.58±3.16),(7.95±3.19) and(8.12±2.68) days respectively. Group A has smaller number of days patients wearingpump than those of group C and D. The difference between group A and group C, andthe difference between group A and group D have statistical significance, P<0.05andP<0.01respectively.(4) Insulin dosage: Insulin dosage for group A, B, C and D were (42.13±14.70)U/d,(42.58±13.19) U/d,(45.26±14.56) U/d and (54.63±16.53) U/d respectively.Group A’s Insulin dosage is lower than that of group C and D. Group B’s Insulindosage is lower than that of group C. Group C’s Insulin dosage is lower than that ofgroup D. The difference between group A and C, between group a and D, betweengroup B and C and between group C and D have statistical significance, P<0.01, P<0.01,P<0.01and P<0.01respectively.(5) Incidence of hypoglycemia: Incidence of hypoglycemia dosage for group A, B,C and D were10.20%,9.52%,9.40%and10.17%respectively. The differencesbetween each of the two groups have no statistical significance, P>0.05.2. Correlation between blood sugar success rate, blood sugar standard deviationand other influencing factors:(1) Multivariate non-conditioned logistic regression analysis. Using blood sugarsuccess rate (1=positive,0=negative) as independent variable, and using gender(males=1, female=2), age, duration of diabetes, height, waist, hip circumference,WHR, weight, BMI, HbA1c,24MA, C-P,2h C-P, blood sugar standard deviation,incidence low blood sugar (hypoglycemia=1, not occurring=0) and insulin dosage asdependent variable. The result shows that the blood sugar standard deviation（OR=0.699,95%CI0.544-0.903, P<0.01）, Insulin dosage （OR=0.985,95%CI0.970-1.000, P<0.05）and hip circumference （OR=0.963,95%CI0.932-0.996, P<0.05）have negative correlation with blood sugar success rate.(2) Multivariate non-conditioned logistic regression analysis with blood sugarstandard deviation as independent variable. Using age, patient’s duration of diabetes,height, waist, hip circumference, WHR, weight, BMI, HbA1c,24MA, C-P,2h C-P,number of days patient wearing pump and insulin dosage as dependent variable. Age (β=0.175, P<0.01) and HbA1c (β=0.185, P<0.01) had significantly positive correlationwith blood sugar standard deviation; BMI (β=0.128, P<0.05) and the pump wearingdays (β=0.147, P<0.05) had significantly negative correlation with blood sugarstandard deviation.(3) Multivariate stepwise regression analysis with insulin dosage as independentvariable. Using age, duration of diabetes, height, waist, hip circumference, WHR,weight, BMI, HbA1c,24MA, C-P,2h C-P, number of days patient wearing pump andblood sugar standard deviation as dependent variable. Duration of diabetes（β=0.141,P<0.5), BMI (β=1.501, P<0.01), HbA1c (β=0.121, P<0.05), weight (β=2.130,P<0.01) had significantly positive correlation with insulin dosage; height (β=-1.083,P<0.01) had significantly negative correlation with insulin dosage.Conclusion: When using CSII to treat T2DM patients, comparing to long durationof T2DM group, short duration of T2DM group had higher blood sugar success rate,fewer days of patient wearing pump, lower insulin dosage, and no increase of incidenceof hypoglycemia.