Correlation Between Diabetic Peripheral Neuropathy And Body Composition In Adult Type 1 Diabetes Mellitus Patients

IntroductionType 1 diabetes mellitus(T1DM)is a chronic autoimmune endocrine disease caused by insufficient insulin-secretion which results in hyperglycemia and metabolism disorder of fat and protein.The incidence and prevalence of T1DM has been increasing all over the world.Although the latest research shows that although China has the lowest incidence of T1DM in the world(1.01/100,000 per year),the total number of newly diagnosed children in China is the fourth largest in the world.On the other hand,diabetic complications are the main causes of death and disability in patients with diabetes,including cardiovascular disease,stroke,peripheral artery disease,nephropathy,retinopathy,neuropathy,as well as the reduced ability to fight infection.Diabetic neuropathy is the commonest complication(about 28%in patient of T1DM),causing a heavy public health burden of disease.The most common type of diabetic neuropathy is distal symmetric polyneuropathy(approximately 75%),or diabetic peripheral neuropathy(DPN).The clinical presentation of DPN is chronic,symmetrical,length-dependent sensorimotor polyneuropathy followed a "stocking and glove" distribution.The most distal extremities are usually involved first,and then slowly progress proximally,accompanied by tingling,pain and numbness of skin.Nerve conduction studies are the current golden standard for the diagnosis of DPN.Unfortunately,once DPN has reached a stage at which it is detectable by conventional bedside tools,it might be irreversible at that point.Because treatment typically focuses on pain control,along with modification of risk factors(such as good glycaemic control),and no treatment is available for reversion or reliable prevention of the disease progression.The pathophysiology of DPN has not been well illuminated yet,and its etiology is most likely multifactorial.Persistent hyperglycemia leads to an increase in the production of cytosolic and mitochondrial reactive oxygen species(ROS)and reactive nitrogen species(RNS),which cause damage not only to axons of peripheral nerves and its microvasculature,but also to its mitochondrial DNA and nuclear DNA.It is believed that derangements of normal metabolic homeostasis,autoimmunity,and microvascular insufficiency also play an important role in occurrence of DPN.Current research has illustrated that hyperglycemia,metabolic syndrome(obesity),hypertension,hypertriglyceridemia,age,smoking and particular gene are the risk factors of DPN for T1DM patients.It also reported that the odds of DPN increased with worse renal function(OR1.96[1.03-3.74]for estimated glomerular filtration rate levels<30 vs.≥90 mL/min/1.73 m2)by a population representative study enrolled 5558 T1DM patients from Scotland.Bioelectrical impedance analysis(BIA)is a newly developing examination measuring resistance of human body under alternating voltage of different frequency.Hence the cell membrane is an insulator at low frequency,extracellular fluid is responsible for the body resistance.While the cell membrane acts as a perfect capacitor at very high frequency and therefore total body resistance reflect the combined of both extracellular and intracellular fluid.Using BIA equations and established procedures,BIA allows determination of fat-free mass,total body water(TBW),extracellular water(ECW)and other relevant indexes.On the basis of theoretical principles,BIA is most appropriately reserved for the assessment of body hydration.Because the equipment is portable and safe,the procedure is simple and noninvasive,and the results are reproducible and rapidly obtained,BIA has become a very prevalent method among hospitalized patients as well as people having health examination in China,which can be used in monitoring of body fluid volumes,assessment of obesity,and status of nutrition.Although quite many studies have been focused on BIA and body composition analyze in type 2 diabetes mellitus(T2DM),few studies have been conducted on type 1 diabetes mellitus.Hence,we carried out this research in order to find out association between BIA parameters and DPN among T1DM patients.PurposeTo investigate the result of body impedance analysis(BIA)in type 1 diabetes mellitus patients with diabetic peripheral neuropathy(DPN).Materials and MethodsThe current study was a retrospective case-control study which included all adult T1DM patients hospitalized in our hospital who had performed BIA from January 2016 to June 2021.Diagnosis of type 1 diabetes was based upon clinical presentations and positive islet cell autoantibodies.In order to eliminate potential influence to the fluid status and lipid metabolism,patients with dysfunction of other endocrine organs,including pituitary,thyroid,parathyroid and gonad were excluded from the study according to blood hormonal levels and symptoms.Following participants were excluded from the study:with heart failure,with cachexia,with hepatitis or liver cirrhosis,with renal failure(serum creatinine>442umol/L)and pregnancy.Finally,the study population comprised 125 T1DM patients aged 18-73.Mean body mass index of 125 patients was 20.89(±2.94)Kg/m2.All the patients were on the diabetic diet and did not take extra exercise beside activity of daily life during hospitalization.No patient with diabetic foot ulcers or foot deformities was included in the study.Informed consent from each patient was not required due to the retrospective casecontrol study design and the use of anonymized data.Every patient took part in diabetic neuropathy Screen in our hospital.Five examinations were included in the screen:ankle jerk,vibration sensation,pressure sensation,temperature sensation and needling pain sensation.All these examinations were conducted by trained nurses with a vibration perception threshold determinator(Sensiometer A200,laxons,China).According to the Chinese guideline on diagnosis of diabetic neuropathy of 2021 edition,the diagnostic criteria of DPN were:1.A clear history of diabetes;2.Neuropathy at or after the diagnosis of diabetes;3.Clinical symptoms of neuropathy,such as pain,numbness,paresthesia,etc.,and abnormality of any one of the five examinations;If there are no clinical symptoms,any 2 abnormalities of the 5 tests can be diagnosed.4.Except for neuropathy caused by other causes.According to whether DPN was occurred,the patients were divided into the control group(N=95)and the DPN(N=30)groups.This study was approved by the Ethics Committee of Shandong Provincial Hospital affiliated to Shandong University and its protocol was in line with the Declaration of Helsinki(as revised in Brazil in 2013).All data was retrieved from previous hospital records of patients admitted to our hospital.Venous blood was collected after fasting overnight.High-performance liquid chromatography(HPLC)with a hemoglobin A1c analyzer(TOSOH Corporation,Japan)was used to determine Glycated hemoglobin(HbA1c).total cholesterol(TC),triglycerides(TGs),low-density lipoprotein cholesterol(LDL-c),high-density lipoprotein cholesterol(HDL-c),fasting plasma glucose(FPG),uric acid(UA),serum creatinine(SCr)and blood urea nitrogen(BUN)were analyzed using an automatic biochemistry analyzer(Beckman Coulter Analyzer AU58 Series,USA).Estimated Glomerular filtration rate(eGFR)was calculated based on serum creatinine concentration using the Xiangya equation.The novel equation provides more accurate GFR estimates in C hinese adults than other formulas.The values of height and weight were taken with the subjects wearing light clothing and without shoes.The height was determined to the nearest 1 cm and the weight was determined to the nearest 0.1 kg.We use BIA(InBody720 body composition analyzer,InBody,Seoul,Korea)with an established protocol to assess fluid status parameters including ECW and TBW as well as body composition parameters including fat mass,body muscle,visceral fat area(VFA)and so on.The normality of all Characteristics was tested by the P-P Plot before performing parametric tests.Normally distributed continuous data are showed as the mean±SD,otherwise they are expressed as the medians with interquartile ranges.Categorical variables are represented as percentages.To analyze differences between the two groups,variables with a normal distribution were tested by t-test,while nonparametric test was used to test skewed ones,and Chi-square test to the data of ratio.Fisher’s precision probability test was used when any of the four theoretical frequencies is less than five.Parameters with statistical differences between the DPN group and the control group were constituted the co-variates and whether complicated with DPN was constituted the dependent variable.Binary logistic regression analysis was used to determine which parameters were risk factors of DPN.What’s more,the receiver operating characteristic(ROC)curve of meaningful parameter of DPN had been portrayed to access its diagnostic significance.Spearman and Pearson correlation analysis were used to explore the correlation between BIA indexes and other study variables,and multiple stepwise regression analysis was used for further inspection.All P values were two-tailed,which less than 0.05 were considered statistically significant.SPSS version 22.0(SPSS,Chicago,IL,USA)was used for all Statistical analysis.ResultsData were collected from 125 Chinese patients including 71 males and 54 females with an age ranging from 18-73.Anthropometric and clinical characteristics of the patients were illustrated in Table 1.There was no significant difference between the DNP group and the control group in BIA parameters except ECW/TCW.The DPN group had higher ratio of ECW/TBW than the control group(0.3969±0.0097vs0.3886±0.0086,P<0.001).The DPN group had older age(45.6±15.6vs35.8±14.8 years,P=0.002)and longer duration of diabetes(10.22±5.2vs5.5±6.4 years,P<0.001).The DNP group also has larger proportions of diabetic peripheral vascular disease(DPVD)(60.0%vs24.2%,P<0.001)and diabetic nephropathy(DN)(36.7%vs8.4%,P=0.001).Besides,the DPN group has higher BUN(5.7±3.1vs4.7±2.2,P=0.043)concentration and lower eGFR(94.3±21.0vs104.9±19.5,P=0.012)than the control group.According to the explanation of result from the Inbody company,a normal ECW/TBW is 0.36 to 0.39.The value of ECW/TBW between 0.39 to 0.40 is determined as mild dropsy,and the value larger than 0.40 is defined as dropsy.ECW/TBW in the DPN group is 0.3969±0.0097,which means that most patients in the DPN group were suffering from mild dropsy.Age,duration of diabetes,DPVD,DN,ECW/TBW,BUN and GFR were selected as possible risk factors in to binary logistic regression analysis.We chose 0.395 as cut off value of ECW/TBW because it is slightly less than the average of the DPN patients.Longer duration of diabetes(OR=1.099 95%CI:1.011～1.196,P=0.027),ECW/TBW≥0.395(OR=4.589 95%CI:1.609～13.091,P=0.004),DPVD(OR=3.84195%CI 1.072～13.765,P=0.039)and DN(OR=6.182 95%CI 1.429～26.738,P=0.015)may be risk factors of DPN for T1DM patients.On the other hand,the author had drawn the ROC curve for ECW/TBW in predicting DPN in T1DM patients.The Area under curve(AUC)of the ROC Curve is 0.752(95%CI 0.646-0.857),and corresponding cut off to the maximum of youden Index(41.9%)is 0.3965,of which the sensitivity is 0.567,and the specificity is 0.853.Further statistical analysis in this study had indicated that ECW/TBW is relevant to age(Pearson Correlation Coefficient=0.192,P=0.032),sex(Spearman Correlation Coefficient=-0.193,P=0.031),DPVD(Spearman Correlation Coefficient=0.195,P=0.029),DN(Spearman Correlation Coefficient=0.231,P=0.010)DPN(Spearman Correlation Coefficient=0.231,P=0.010)and BMI(Pearson Correlation Coefficient=-0.246,P=0.006).To our surprise,the BIA parameter is not correlated to the SCr,BUN and eGFR.In order to explore the relationship between ECW/TBW and other indicators,multiple stepwise linear regression analysis was conducted with ECW/TBW as the dependent variable(Y)and age,gender,DPVD,DN,DPN and BMI as the independent variables.Finally,DPN,gender and BMI were entered into the equation(F=12.850,P<0.001,R2=0.242):Y=0.4066+0.0082*DPN-0.0044*gender0.00074*BMI.If the DPN is complicated,the value of DPN is 1.Otherwise,the value of DPN is 0.The sex value is 1 for males and 0 for females.According to the standardized partial regression coefficient,DPN has the greatest impact on ECW/TBW among all the parameters included in the study.According to R2,DPN,gender,and BMI could explain 24.2%of the variation in Y,suggesting that there may be other factors influencing the result of ECW/TBW.ConclusionThe increase of ECW/TBW may be a new risk factor of DPN.T1 DM patients with ECW/TBW>0.395 had 4.589 times higher risk of developing DPN than those with ECW/TBW<0.395.ECW/TBW could help doctors predict the risk of DPN in T1DM patients.The best cut off value is 0.3965(corresponding youden Index=41.9%),of which the sensitivity is 0.567,and the specificity is 0.853.ECW/TBW is relevant to age,sex,DPVD,DN,DPN and BMI.Multiple stepwise linear regression analysis shows that ECW/TBW(Y)=0.4066+0.0082*DPN0.0044*gender-0.00074*BMI.DPN has the greatest impact on ECW/TBW than other parameters.