Study On The Correlation Between Body Fat Content And Insulin Resistance In T2DM Patients With NAFLD By PDFF Technique

Object ive:The liver and whole body fat content in patients with type 2 diabetes mellitus(T2DM)complicated with non-alcoholic fatty liver disease(NAFLD)was quantitatively measured by MRI whole body fat quantification proton density fat fraction(PDFF)technique,and its correlation with insulin resistance(IR),serum indexes and fatty liver index(FLI)was discussed.Materials and methods:A total of 62 patients with T2DM who underwent quantitative body fat examination by magnetic resonance imaging in the Third Provincial Hospital Affiliated to Shandong University from December 2021 to November 2022 were collected and divided into simple T2DM group(n=30)and T2DM combined with NAFLD group(n=32)according to whether they had NAFLD.At the same time,40 healthy subjects of similar age at the same period were selected as the control group.All subjects ranged in age from 32 to 70 years,with an average age of 52.07±8.12 years old.The uMR588 1.5T magnetic resonance imaging system was used for image acquisition,with head coil and body 12-channel phase-controlled front coil.The sequences of routine MRI scanning included abdominal routine axial T1 weighted image(T1WI)sequences,axial T2 weighted image(T2WI)sequences for lipid suppression,coronal T2WI sequences,axial diffusion weighted imaging sequences,and axial Whole Body sequences.All the scanned images were transferred to the uMR588 1.5T magnetic resonance workstation for post-processing to obtain R2*and fat quantitative parameters.All patients’past medical histories were recorded in detail.Then their height,body mass,waist circumference were measured,and body mass index was calculated.At the same time,blood glucose metabolism indexes,lipid indexes and liver function indexes were recorded,and iron concentration and clinical serum model FLI values were calculated.SPSS 23.0 software was used for statistical analysis of the collected imaging and serum data.KolmogorovSmirnow method was used to test whether the measurement data conform to the normal distribution,and normal distribution measurement data were expressed as mean±standard deviation(x±s).One-Way ANOVA was used for comparison among multiple groups.If the differences were statistically significant,multiple comparisons were conducted.If the variances were homogeneous,Bonferroni method was used to test while if the variances were not uniform,Tamhane’s T2 method was used to test.Non-normal distribution measurement data were described as median(interquartile distance)[M(P25,P75)],and Kruskal-Wallis H test was used for comparison between groups.Count data were expressed as constituent ratio,and chi-square test was used for comparison between groups.Pearson correlation method was used to analyze the correlation between liver PDFF and homeostasis model insulin resistance index(HOMA-IR),clinical serum indexes and serum model FLI.With whether to merge NAFLD as dependent variable,the independent risk factors of T2DM with NAFLD were analyzed by multivariate Logistic regression.P>0.05 means no statistically significant difference,P<0.05 means statistically significant difference,and P<0.01 means statistically significant difference.Results:1.Comparison of general data and fat quantitative results of the three groups of subjectsCompared with the general data of the control group,the simple T2DM group and the T2DM combined with NAFLD group,there were no significant differences in age,course of disease and sex composition ratio among the three groups.Besides,there were no significant differences in HbAlc,body fat percentage,subcutaneous fat percentage,subcutaneous fat/intracellular fat between simple T2DM group and T2DM combined with NAFLD group.In simple T2DM group,FPG,FINS,HOMA-IR,HbAlc,PDFF and the proportions of internal fat were respectively 6.65 ± 1.33mmol/L,8.46 ± 3.75mU/L,2.48 ± 1.15,3.87±0.43%and 33.60±6.39%,which were significantly higher than those in control group[4.90±0.44mmol/L,4.55±0.50mU/L,0.98±0.09,3.40±0.66%,25.00±6.24%,respectively],while subcutaneous fat ratio and subcutaneous fat/intracellular fat ratio were 66.41 ±6.39%and 1.46±0.39,respectively,which were significantly lower than those in control group[75.30±6.64%and 2.54±0.85],with statistical significance(all P<0.05).In T2DM combined with NAFLD group,body mass index,FPG,FINS,HOMA-IR,PDFF,proportion of internal fat and total body fat volume were respectively 27.61 ±2.51kg/m2,9.17 ± 2.67mmol/L,13.57±5.01mU/L,5.3 7 ± 2.12 and 8.65 ± 2.47%,37.85± 6.43%,30186.72±6480.73cm3,which were significantly higher than those in simple T2DM group[24.81 ± 2.47kg/m2,6.65 ± 1.33mmol/L,8.46 ± 3.75mU/L,2.48 ± 1.15,3.87 ± 0.43%,33.60±6.39%,24751.37±5631.79cm3,respectively]and control group[23.74±2.27kg/m2,4.90±0.44mmol/L,4.55±0.50mU/L,0.98±0.09,3.40±0.66%,25.00±6.24%,21555.28±7963.22 cm3,respectively],with statistical significance(all P<0.05).In T2DM combined with NAFLD group,the percentage of body fat was 44.91 ±5.67,which was significantly higher than control group(39.29±9.94),and the percentage of subcutaneous fat and subcutaneous fat/intracellular fat were respectively 64.85±7.75%and 1.43±0.52,which were significantly lower than those in control group(75.30±6.64 and 2.54±0.85,respectively),with statistical significance(all P<0.05).2.Correlation between HOMA-IR and various fat quantitative indexes as well as in T2DM combined with NAFLD groupThere was significant correlation between HOMA-IR and PDFF as well as FLI in T2DM combined with NAFLD group(P<0.05).However,there was no significant correlation with body fat percentage,total body fat volume,subcutaneous fat ratio,internal fat ratio,subcutaneous fat/luminal fat.Furthermore,Pearson analysis showed that,HOMAIR was positively correlated with PDFF and FLI in T2DM combined with NAFLD group(r=0.645,0.449,both P<0.05).3.Comparison of clinical serum indexes among three groups of subjectsTG in simple T2DM group was significantly higher than that in control group,while TP,ALB,GLB,TBiL and HDL_C were significantly lower than those in control group,with statistical significance(all P<0.05).ALT and AST in T2DM combined with NAFLD group were significantly higher than those in simple T2DM group and control group,with statistical significance(all P<0.05).There was no significant difference in LDL_C among the three groups.4.Correlation between PDFF and clinical serum indexes in T2DM combined with NAFLD groupPDFF in T2DM combined with NAFLD group was significantly correlated with clinical serum indexes TP,GLB,A/G,TG,HDL_C and LDL_C(P<0.05),but had no significant correlation with ALT,AST,ALB,GGT,TBiL and TC.PDFF was positively correlated with TG,TP and GLB(r=0.633,0.538,0.557,P<0.05),and negatively correlated with A/G,HDL_C and LDL_C(r=-0.434,-0.584,-0.426,P<0.05).5.Correlation between PDFF and clinical serum model FLI in T2DM combined with NAFLD groupPDFF in T2DM combined with NAFLD group was positively correlated with FLI in clinical serum model(r=0.678,P<0.05),and FLI value increased with the increase of PDFF.6.The influencing factors of T2DM with NAFLD were analyzed by multivariate Logistic regressionHOMA-IR and internal fat ratio are independent risk factors for T2DM with NAFLD.Respectively,the AUC of HOMA-IR and internal fat ratio were 0.8922 and 0.6865,and the optimal cut-off values were 4.18 and 39.05.Conclusion:With the advantages of non-invasive,accurate and high reproducibility,MRI whole body fat quantification PDFF technique could quantitatively evaluate the content and distribution indexes of total body fat in T2DM with NAFLD patients.It could identify mild NAFLD in the early stage in combination with clinical serum indexes,and suggested that attention should be paid to lipid-lowering therapy while hypoglycemic treatment was performed,so as to effectively improve insulin resistance and correct abnormal glucose and lipid metabolism.Furthermore,it could provide a reliable image index for the evaluation of clinical treatment effect.