Evaluation Of The Left Ventricular Function In Patients With Type 2 Diabetes Mellitus By Cardiac Cine Magnetic Resonance Imaging

[Objective] The study evaluates the left ventricular structure and function in patients with type 2 diabetes mellitus(DM) by cardiac cine magnetic resonance imaging . We also discuss the relation between the changes of left ventricular function and the pathologic foundation for earlier diagnosis and prevention.[ Subjects and Methods ] We studied left ventricular function with cardiac cine MRI in 85 diabetes subjects(45 males and 40 females, age from 28-78, mean age 53.87±9.86,duration of diabetes from 1 month to 22 years) and control subjects(20 males and 23 females, age from 34-65, mean age 52.14±8.06) free of hypertension, heart failure and previous myocardial infarction. Cardiac cine MRI was carried out on a superconductive GE 1.5T Twin-Speed Infinity with Excite II system. Imaging was performed using a cardiac phased-array coil with electrocardiogram(ECG)-gating and during patient breath-holding. Following survey scans, breath-hold cine sequences in the sagittal axis, four-chamber planes were acquired to allow planning of the short-axis orientation. Multiphase data sets covering the left ventricle (LV) in 8-12 short axis slices from apex to base were then acquired. Image analysis was performed off-line on a workstation (AW4.0) using commercially available analysis software(MASS analysis). The endocardial and epicardial contours were traced automatically and adjusted manually in contiguous short-axis slices. Parameters of resting LV diastolic and systolic function and diastolic filling time-volume curve were analyzed for all subjects, and comparedthem between diabetes and control subjects.[Results] The data are presented as mean ± SD. Continuous data were compared with a two-tailed Independent-Samples t test. P value less than 0.05 indicated statistical significance.1. Comparison of the DM group and control group in rest: In the DM group, the ESVI (21.75±8.42ml/m2) was smaller than the control (24.54±5.41ml/m2, /=2.265, P=0.025) , but the EF(59.14±8.64%)was bigger than the control (55.92 + 6.00%, /=-2.457, P=0.016). The WTES of SI to S6, WTED of S2 to S6 and WM of S2 to S3 were bigger than the control group(P<0.05). The PFR(282.19±72.63ml/s)in DM group were smaller than the control group (320.84 + 98.98 ml/s, /=2.508, P=0.013). The other parameters showed nonsignificant of difference(P > 0.05).2. In male DM group, the WTES of S2(13.99±2.58mm)and WTES of S3(14.42±2.22mm)were bigger than the control group(/><0.05).3. In female DM group, the EDM was close to the significant of difference(P= 0.051) , EDMI(34.63±8.09 kg/m2) was bigger than the control group (31.11+3.41 kg/m2, /=-2.405. P=0.019) . The WTED of S2 to S5, WTES of S2 to S5 were bigger than the control group(P<0.05). The PFR(269.14±52.26ml/s)was smaller than the control group (310.75+ 73.60ml/s, /=2.614, P=0.011) andTPF(141.03±62.33m s) was bigger than the control (106.52 + 38.92ms, f=-2.395, PO.020) .4. In young male DM group, the WTED of S2(6.93 +0.82) was bigger and than the old male group (6.38±0.92)(P<0.05),but the TPF (161.84 + 39.02) was smaller than the old group(191.81 +60.72). In contrast, the WTED of S4 to S6 and WTES of SI in the old female DM group were bigger and than the young female group(P<0.05),but the PFR(250.76 + 51.20) was smaller than the young group(294.00 +43.70).5. The strength of the link between the continuous variables was tested by Pearson correlation coefficient r. The age of patients with diabetes correlatedV.K J. HT TL it =f- HL V6 X.inversely with the PFR (/- =-0.311, ^=0.004) and duration of diabetes correlated inversely with the PFR {r =-0.295, P =0.006). No significant correlation between the BMI and PFR was observed (P > 0.05). Furthermore, the EDM correlated inversely with the PFR (r =-0.281, P =0.009). The age, duration of diabetes and BMI did not correlate significantly with left ventricle ejection fraction (P > 0.05). The EDM correlated inversely with the EF(r =-0.228, P =0.036). Though the fasting glucose did not correlate significantly with EF and PFR (P > 0.05), multiple logistic regression model showed that fasting glucose was the risk factor of left ventriclefunctional dysfunction (X2=3.742, P=0.053).[Conclusion I Cardiac cine MRI and FIESTA sequence are considered the gold standard for evaluating the left ventricular function because of shorter imaging time, higher image quality and accurate assessment. Diabetic cardiomyopathy, manifested by diastolic dysfunction followed by abnormalities in systolic function, is common in type 2 diabetes subjects without the apparent cardiocomplications. Abnormalities of left ventricular structure and diastolic function can be observed in critique age diabetic female patients because of special physiologic factor. LV mass and wall thickness increased was more striking in women compared with men. High fasting glucose is the risk factor of diabetic cardiomyopathy. In conclusion, Cardiac cine MRI is an effective means to detect and diagnose DC in early stage.