| Objective:In this study,type 2 diabetes model of Guangxi bama mini pig was established to investigate the pathological and biochemical changes in the early stage of diabetic nephropathy and to investigate the value of diffusion kurtosis imaging(DKI)and intravoxel incoherent motion-diffusion weighted imaging(IVIM-DWI)in diagnosing early diabetic nephropathy.Methods:1?Twelve pigs were divided into the experimental group(7 pigs)and the control group(5 pigs),used the random number table method.The first 5 months all pigs were fed on a free feeding diet using normal fodder.Before modeling,fasting blood glucose and body weight were measured in all mini pigs to obtain baseline data.From the 6th month,the experimental group was fed with high-fat high-sugar diet,and then repeatedly injected small doses(50 mg/kg)of Streptozotocin(STZ)through the ear vein.Meanwhile,the control group was fed with normal diet and injected with the same dose of citric acid-sodium citrate buffer solution.After modeling,fasting blood glucose(GLU)and insulin(INS)were monitored regularly and homeostasis model assessment-insulin resistance index(HOMA-IR)was calculated.The type 2 diabetes was considered successful when the fasting blood glucose was greater than 7.0 mmol/L and HOMA-IR was higher than the control group.2?After the type 2 diabetes was established successfully,T1 WI?T2WI?DKI and IVIM sequence imaging were performed every month for 2 pigs from the experimental group and the control group,respectively.Mean kurtosis(MK)?axial kurtosis(K//)?radial kurtosis(K?)?fractional anisotropy(FA)?mean diffusivity(MD)?Standard ADC?Slow ADC?fast ADC and f values were measured on the pseudo-color map of the post-processing workstation by 2 senior radiologists.The T2WI image was used as a reference,and the region of interest(ROI)was manually placed in the renal cortex and medulla,ROI area from 30 mm2 to 50 mm2.The ROI was measured for 3 times,and the average was taken as the final result.Fasting Blood glucose(GLU),insulin(INS),renal function,urine routines and random albumin creatinine ratio(RACR)were measured before MRI scan.Specimens from bilateral kidneys were taken for pathological examination after MRI scan.3?Statistical analysis was performed by SPSS 22.0 software,Kolmogorov-Smirnov(K-S)method was used to tested whether the samples conformed with the normal distribution.The consistency of the parameter values measured by two physicians was evaluated and described by Intraclass correlation coefficient(ICC).The paired t test was used to compare the parameter values of the cortex and medulla.Independent sample t test(for normal distribution)or Mann-Whitney U test(unsatisfied with the normal distribution)was used to compare the parameter values of the experimental group and the control group?biochemical test and weight.Analysed the diagnostic efficiency of DKI and IVIM-DWI parameters between the experimental group and control group used the ROC curve(MedCalc 15.8 software).Used the Pearson(for normal distribution)or Spearman(unsatisfied with the normal distribution)analysis of the correlation between imaging parameters and laboratory results.The difference of P<0.05 was statistically significant.Results:1?Modeling result:In the experimental group,type 2 diabetes were successfully modeled in all pigs.After modeling,the weight?GLU and insulin resistance index of experimental group were higher than those in the control group,and the difference were statistically significant(all P<0.05).2?Laboratory results:In the experimental group,3 pigs were positive for urine protein,3 pigs were positive for urine sugar,and 2 for RACR greater than 30 mg/g;In the control group,1 pig was weakly positive for urine protein,1 pig was positive for urine protein,and all pigs were negative for urine sugar and RACR.Renal function was normal in both experimental group and control group.3?Pathological results:the experimental group had a total of 12 kidneys from 6 pigs accord with early diabetic nephropathy(?-? grade)change,12 kidneys were included in the analysis.4?Imaging results:DKI sequence:The DKI parameters of cortex/medulla in the experimental group were:MK0.60±0.06?0.66±0.07,K//0.59±0.06?0.66±0.08,K? 0.53±0.10?0.62±0.09,FA0.16±0.03?0.19±0.04,MD 2.15±0.23(10-3mm2/s)?2.07±0.22(10-3 mm2/s).The DKI parameters of cortex/medulla in the control group were:MK0.53±0.03?0.59±0.03,K//0.60±0.05?0.66±0.04,K? 0.44±0.06?0.52+0.06,FA0.17±0.04?0.20±0.04,MD 2.29±0.15(10-3 mm2/s)?2.20±0.12(10-3 mm2/s).Except that there was no statistically significant difference(P>0.05)between the cortex and medulla of MD in the experimental group,the MK?K//?K ??FA values of medulla were increased compared with the cortex and the difference were statistically significant(P<0.01).The MK?K? values of cortex and medulla were increased in the experimental group compared with the control group and the difference were statistically significant(P<0.05).There were no significant differences in the K//?FA and MD values between two groups(both cortex and medulla,P>0.05).IVIM-DWI sequence:The IVIM-DWI parameters of cortex/medulla in the experimental group were:Standard ADC(2.64±0.29)X 10-3(mm2/s)?(2.37±0.24)× 10-3(mm2/s),Slow ADC(2.08±0.28)×10-3(mm2/s)?(2.07±0.25)×10-3(mm2/s),f38.83±13.96(%)?32.58±11.46(%).The IVIM-DWI parameters of cortex/medulla in the control group were:Standard ADC(2.30±0.19)× 10-3(mm2/s)?(2.00±0.19)× 10-3(mm2/s),Slow ADC(1.97±0.20)× 10-3(mm2/s)?(1.85±0.13)× 10-3(mm2/s),f 32.58±11.46(%)?24.77±8.42(%).The standard ADC values of cortex in experimental group and control group were higher than those of medulla,and the difference was statistically significant(P<0.05);Cortical slow ADC values was slightly higher than medulla in both groups,but the difference was not statistically significant(P>0.05).The standard ADC values of the experimental group and the control group were all higher than those of the corresponding slow ADC values(both cortex and medulla),the differences were statistically significant(P<0.05).The standard ADC values of cortex and medulla in the experimental group were higher than those in the control group(P<0.05).The cortical slow ADC value in the experimental group was slightly higher than that in the control group,but the difference was not statistically significant(P>0.05).The medulla slow ADC value in the experimental group was higher than that in the control group,and the difference was statistically significant(P<0.05).Diagnostic efficacy of DKI and IVIM parameters:standard ADC was the best IVIM parameter in differentiating the experimental group and the control group.While MK and K ? were the best DKI parameters in differentiating the experimental group and the control group.5?Correlation of various examination indexes:The weight has certain correlation with most laboratory indexes,while most DKI and IVIM parameters lack correlation with laboratory indexes.Conclusion:1?The early stage of diabetic nephropathy may have pathological changes in the kidney,but laboratory tests are often negative and lack clinical symptoms.2?DKI sequence can reflect the tissue structure differences of normal renal cortex and medulla,and reflect the complex microstructure;DKI sequence is feasible for non-invasive early diagnosis of diabetic nephropathy,and to some extent reveals the pathological change in early diabetic nephropathy.3?IVIM sequence accurately reflects the true diffusion of water molecules in the tissue by isolating the effects of microvascular perfusion.It can also indicate the high blood perfusion state and diffusion limitation degree in the early stage of diabetes nephropathy.4?This study suggests that there was a certain correlation among the laboratory tests such as weight,fasting blood glucose,insulin resistance index,urinary protein and RACR,which indicates that the influencing factors of diabetic nephropathy were complicated and mutually promoted. |
PDF Full Text Request