| Objective To explore the relationship between type 2 diabetes and cerebral microbleeds(CMBs)in patients with first-ever ischemic stroke.Methods A prospective data was collected from patients with acute high-risk nondisabling ischemic cerebrovascular events in PRINCE study in Tangshan Gongren Hospital from October 2015 to February 2017.General clinical data,biochemical indicators and imaging characteristics were recorded.The susceptibility weighted imaging(SWI)was preformed on admission and after 90 days.New-onset CMBs were recorded after follow-up.Subjects were divided into two groups: type 2 diabetes group and non-diabetes group,in order to compare the difference of general clinical data,biochemical indicators and imaging characteristics.Whether type 2 diabetes is the cause of the increase in CMBs is analyzed with calculating odds ratio(OR)and 95% confidence interval(CI)based on single and multiple Logistic regression model.At the same time,the relationship of diabetes,diabetic duration,as well as the increase in CMBs after 90 days is discussed.Results 41 patients were involved in this study,including 11 patients(26.8%)in type 2 diabetes group,and 30 patients without type 2 diabetes(73.2%).Compared with patients in non-diabetes group,higher rates of CMBs on admission were showed in type 2 diabetes group [91%(10/11)and 73%(23/30),P = 0.15],a much higher proportion of hypertension [72.7%(8/11)and 43.3%(13/30),P = 0.1],a lower value of creatinine [53.2±15.2 and 59.3±10.4,P=0.15].Baseline data from the two groups(P<0.2)were used as confounding factors.The remaining baseline was not statistically significant.The single Logistic regression model showed that type 2 diabetes is associated with the increase in CMBs after 90 days [(7/11,63.6%)and(6/30,20%),OR=7.00,95%CI=1.53~32.00;P=0.01].After correcting the P<0.2 confounding factors(the incidence of CMBs,hypertension,hypertension duration and creatinine on admission),the analysis based on multiple Logistic regression model between the two groups showed that compared with patients in non-diabetes group,type 2 diabetes is related to the increase in CMBs after 90 days [63.6%(7/11)and 20%(6/30),OR = 11.45,95% CI = 1.50~87.43,P = 0.02],which is suggested that type 2 diabetes may be correlated with the increase in CMBs after 90 days.The total number of increased CMBs after 90 days was 13,including 7(63.6%)in diabetes group,among which there was 1 with 7 increased CMBs,1 with 4 increased CMBs,3 with 2 increased CMBs and 2 with 1 increased CMBs;and 6(20.0%)in nondiabetes group,among which 1 with 4 increased CMBs,2 with 2 increased CMBs,3 with 1 increased CMBs.The growth rate of the increasing number of CMBs in the diabetic group(7/11,63.6%)was higher than that in the non-diabetic group(6/30,20%).The increasing number of CMBs in the diabetic group(1.73±1.30)is more than that in the non-diabetic group(0.43±0.44).Mann-Whitney U test: Z=-2.679,P=0.006,which indicated that the increasing number of CMBs after 90 days is more than that in the nondiabetic group.Diabetes Phase 0(non-diabetes group): 30 cases,of which 3 with 1 increased CMBs,2 with 2 increased CMBs,1 with 4 increased CMBs.Diabetes Phase 1(0-5 years): 3 cases,of which 2 with 1 increased CMBs,1 with 2 increased CMBs.Diabetes Phase 2(5-10 years): 7 cases,of which 1 with 2 increased CMBs,1 with 4 increased CMBs,1 with 7 increased CMBs.Diabetes Phase 3(10-15 years): 1 case,of which 1 with 7 increased CMBs.According to Sperman rank correlation analysis: rs=0.430,P=0.005,which is statistically significant,the increased CMBs is positively correlated with the diabetic duration.With the duraion of diabetes,the number of CMBs tends to increase.Conclusions Type 2 diabetes may be an independent risk factor for the increase of short-term CMBs in patients with first ischemic stroke.Patients with both ischemic stroke and type 2 diabetes were more likely to have increased CMBs in the short and medium term than those without diabetes.The increasing number of CMBs in patients with first ischemic stroke in short and medium term has a positive correlation with the duration of diabetes. |
PDF Full Text Request