The Association Between Insulin Resistance and Proliferative Retinopathy in Type 1 Diabetes

The association between insulin resistance and proliferative retinopathy in type 1 diabetes

IRINA DUŢĂ1,2, SIMONA FICA2,3, DANIELA ADRIANA ION2 1“Prof. N. C. Paulescu” National Instititute of Diabetes, Nutrition and Metabolic Diseases, Bucharest, Romania 2“Carol Davila” University of Medicine and Pharmacy, Bucharest, Romania 3“Elias” Emergency Hospital, Bucharest, Romania

Introduction. Little is known about the relationship between insulin resistance and proliferative diabetic retinopathy in type 1 diabetes. The aim of this article is to explore the relationship between sight-threatening proliferative diabetic retinopathy and insulin resistance. Methods. This was a cross-sectional study that included 167 type 1 diabetes patients. Insulin resistance was assessed using eGDR (estimated glucose disposal rate) formula. Diabetic retinopathy was assessed by ophthalmoscopy using Early Treatment Diabetic Retinopathy classification. The association between eGDR and proliferative diabetic retinopathy was assessed in uni- and multivariate models using stepwise logistic regression of covariates. The contribution of individual predictors in the final regresion model was examined using Wald statistic. Results. Significantly lower eGDR’s values were observed in patients with proliferative diabetic retinopathy: 5.5 vs. 7 (p = 0.002). The results remained significant (p < 0.001) after adjusting for multiple covariates (sex, diabetes duration, body mass index, HDL cholesterol, LDL cholesterol, triglycerides, smoking). eGDR variable was retained in the final model of stepwise logistic regression (p < 0.001) and showed the strongest association with proliferative diabetic retinopathy (Wald = 12.73). Conclusions. In type 1 diabetes patients insulin resistance was the most important independent risk factor associated with diabetic proliferative retinopathy. Key words: type 1 diabetes, insulin resistance, estimated glucose disposal rate, proliferative diabetic retinopathy.

INTRODUCTION micro- (nephropathy, neuropathy and retinopathy) and macro- (coronary artery disease and peripheral With 70-100% of patients being affected, vascular disease) vascular complications [11-15]. retinopathy is the most common type 1 diabetes The purpose of this article was to assess insulin (T1D) complication [1]. resistance in T1D patients stratified according to Diabetic retinopathy lesions are divided into the presence or absence of PDR. For our secondary two large categories: nonproliferative (NPDR) and aim we hypothesized that in T1D patients insulin proliferative (PDR). Almost all T1D patients have resistance, quantified using the estimated glucose NPDR. The more severe form of retinopathy, PDR, disposal rate (eGDR) formula, would be associated affects 15-50% of T1D patients over 15-20 years of with PDR. disease duration [2, 3]. It usually progresses to visual impairment. In developed countries diabetic retinopathy is the main cause of blindness in MATERIALS AND METHODS working age group [4]. Identifying patients at risk of ocular impairment is a concern for clinicians. Study design was cross-sectional. We included Elements of metabolic syndrome such as patients with type 1 diabetes who underwent medical hyperglycemia [5-7], hypertension [8], dyslipidemia checkups in “N.C. Paulescu” Institute during a one [9, 10] and abdominal obesity [10] are known to be year period (from January to December 2012). A risk factors for the occurrence of diabetic retino- total of 167 patients with type 1 diabetes were pathy. Insulin resistance is the physio-pathological included in the study. substrate for the metabolic syndrome. Insulin Inclusion criteria. Type 1 diabetes mellitus resistance is a key element in the pathogenesis of was defined as: diabetes onset before the age of T2D. It has been suggested that insulin resistance is 35 years with permanent insulin treatment initiated also present in T1D. It appears that in T1D patients within the first year after diagnosis. Eligibility for insulin resistance is an independent risk factor for inclusion in the study was based on the clinical

ROM. J. INTERN. MED., 2015, 53, 3, 261–266 262 Irina Duţă et al. 2 diagnosis of type 1 diabetes. Confirmation, by Assessment of diabetic retinopathy. Retino- determining serum C peptide level (< 0.3 nmol/l), pathy was assessed following ophthalmological was available only for the subset of patients in examination. The eye with the most severe which there were doubts regarding T1D diagnosis. involvement was used for categorization of ocular The study included either patients with chronic involvement. Early Treatment Diabetic Retinopathy kidney disease stages 4 (urinary albumin excretion Study classification was used to assess the severity rate - AER > 300 mg/24h) and 5 (end stage renal of retinopathy. NPDR was diagnosed if any of disease – predialysis/dialysis/renal transplant) microaneurysms, hemorrhages, hard exudates, venous according to Mogensen classification or patients congestion, cotton wool spots or intraretinal mico- with at least 15 years duration of diabetes, eGFR vascular abnormalities were present. PDR was (estimated glomerular filtration rate) between 90- 2 registered when new vessels, glial proliferation, 130 ml/min/1.73 m and AER < 30 mg/24h. T1D preretinal hemorrhage, vitreous hemorrhage, scars patients that had AER = 30-300 mg/24 h and any of photocoagulation (known to have been directed other condition that might interfere with albumin at new vessels) and/or retinal detachment were excretion (hyperglycaemia, uncontrolled hyper- found. Patients with none of these abnormalities tension, heart failure or urinary tract infection) were classified as not having retinopathy. were re-examined. Ethical considerations. Patient enrollment Exclusion criteria. Patients were excluded complied with the Declaration of Helsinki. The from the study in the case of any doubt about type 1 study was approved by the local ethics committee. diabetes diagnosis (even after determining the Statistical analysis.Continuous variables with serum C peptide level) or if any evidence of non- normal distribution are presented as mean ± SD. diabetic renal disease. Continuous variables with non-normal distribution Patients were stratified according to the are presented as median and interquartile range.The presence or absence of PDR. distribution of continuous variables was tested for Clinical parameters. Registration of: diabetes- normality using Kolmogorov-Smirnov and Shapiro- related data, anthropometric indices, blood pressure, smoking status, alcohol intake, family history, Wilk tests. Categorical variables are presented as diabetes-related complications data was done. Supine frequencies (%). Between-group comparisons were systolic (SBP) and diastolic (DBP) blood pressure made using Student’s t test (for normally distributed was measured in both arms after 10 minutes of rest. variables), Mann-Whitney U test (for non-normally We used the arithmetic mean of the values obtained. distributed variables) and χ2 test (for categorical Hypertension was defined as a BP> 140/90 mm Hg variables). and/or use of antihypertensive treatment. Patients were stratified according to the presence Laboratory parameters. Fasting blood samples or absence of PDR. were collected for assessment of HbA1c (glycated The association between insulin resistance hemoglobin), C-peptide, fasting glucose, lipids (total (quantified using eGDR) and PDR was assessed in cholesterol, HDL cholesterol, triglycerides) and uni and multivariate logistic regression models. serum creatinine. HbA1c level was measured by a Initially, we analyzed the association between standardized DCCT (Diabetes Control and Compli- eGDR and PDR in univariate logistic regression cations Trial) method. C-peptide levels were measured models and then regression was adjusted for duration using an ELISA technique. Glucose, lipids and of diabetes and sex. creatinine and were assessed using enzymatic methods. To assess the contribution of insulin resistance The glomerular filtration rate was estimated using (quantified using eGDR) and potential confounders the CKD-EPI formula (Chronic Kidney Disease to PDR we conducted a multivariate analysis (logistic Epidemiology Collaboration). Urine samples were regression) where the dependent variable was the collected in order to determine AER/24 h. appearance of PDR (noted as presence / absence) Assessment of insulin resistance. Insulin and the independent variables are the risk factors resistance was assessed using the eGDR formula. studied (eGDR, sex, duration of diabetes, BMI, HDL eGDR is an equation derived from hyper- cholesterol, LDL cholesterol, triglycerides and smoking insulinemic euglycemic clamp studies [16]. eGDR status). Backward stepwise procedure was used. (ml/kg/min) is based on clinical parameters and is Statistical analyses were performed using SPSS calculated as follows: eGDR = 24.31 - (12.22 x version 20.0. P-values < 0.05 were considered WHR) - (3.29 X BP) - (0.57 X HbA1C), where statistically significant. Variables that did not meet WHR – waist to hip ratio, BP - history of hyper- this criterion were removed from the model. The tension (yes = 1, no = 0), HbA1c - glycated hemo- contribution of individual predictors in the final globin A1C. regression model was examined using Wald statistic. 3 Insulin resistance and proliferative retinopathy in type 1 diabetes 263

RESULTS significance. HbA1c values (p = 0.44), fasting plasma glucose (p = 0.80) and BMI (p = 0.11) had Characteristics of patients. The study included similar values between the two groups. 167 patients with type 1 diabetes (108 men and PDR group had significantly lower eGDR’s 59 women). General characteristics of the studied value (5.5 vs 7 mg/kg/min; p = 0.002). Characte- population are presented elsewhere [17]. A total of ristics of patients with PDR are presented in Table 1. 43.11% (n = 72) of patients from the targeted group eGDR: univariate and multivariate analysis. In univariate analysis PDR was significantly were free of ocular damage; 18.57% (n = 31) had associated with eGDR (OR = 0.78; p = 0.002). The NPDR and 38.32% (n = 64) had PDR. The association remained significant after adjustment prevalence of hypertension was significantly higher for sex and duration of diabetes (OR = 0.78; p = in the group with PDR (75% vs 48.5%; p = 0.001). 0.003). There were no differences between groups regarding In the multivariate analysis eGDR was one of the prevalence of dyslipidemia (p = 0.25) or the variables retained in the model (OR = 0.73; smoking (p = 0.58). p <0.001) and it was the parameter most strongly Patients with PDR had significant differences associated with PDR (Wald = 12.73). BMI was regarding parameters that assess kidney function: also retained in the regression model (OR= 0.88; AER (358.5 vs 3.7 mg/24 h; p < 0.001) and eGFR p = 0.013 ). 2 (66.7 vs 93.5 ml/min/1.73 m ; p < 0.001). The following variables were rejected from Supine SBP (p = 0.036) values were sig- the model (p > 0.05): sex, duration of diabetes, nificantly higher in the PDR group. Supine DBP LDL cholesterol, HDL cholesterol, triglycerides (p = 0.26) did not differ between groups. WHR (p = and smoking. The uni and multivariate analysis 0.08) and total cholesterol (p = 0.06) had borderline results are presented in Table 2.

Table 1 Characteristics of patients according to proliferative diabetic retinopathy Variable Without PDR With PDR p n 103 64 Men/Women 65/38 43/21 0.62 Age (years) 46.1 ± 13.9 46.8 ± 12.6 0.25 Age at diagnosis (years) 21.2 ± 10.8 22.5 ± 11.4 0.76 Duration of diabetes (years) * 24.9 ± 8 24.1 ± 8.5 0.22 Weight (kg) 71 ± 14.2 72.3 ± 16.4 0.28 Height (cm) 170.1 ± 8.5 168.8 ± 9.4 0.88 BMI (kg/m2) 24.4 ± 4 25 ± 3.7 0.11 Waist circumference (cm) 89.1 ± 12.5 92.9 ± 13.3 0.44 WHR 0.9 ± 0.08 0.93 ± 0.08 0.08 Smokers (%) 58.3 62.5 0.58 Supine SBP (mm Hg) 128.4 ± 20.8 135.6 ± 23.7 0.036 Supine DBP (mm Hg) 74.6 ± 13 79.8 ± 9.8 0.26 Hypertension (%) 48.5 75 0.001 Fasting plasma glucose (mg/dl) 202.2 ± 110.7 186.6 ± 96.5 0.8 HbA1c (%) 8.6 ± 1.5 9.1 ± 1.6 0.74 eGDR (mg/kg/ min) 7 ± 2.1 5.5 ± 2.5 0.002 Family history of diabetes (%) 43.7 43.8 0.99 Total cholesterol (mg/dl) 190.7 ± 45.6 193.6 ± 57.8 0.06 LDL cholesterol (mg/dl) 115.8 ± 41.1 121.1 ± 55.4 0.52 HDL cholesterol (mg/dl) 57.7 ± 17 53.6 ± 13.7 0.79 Triglycerides (mg/dl) * 88.5 ± 51 92 ± 68 0.74 Dyslipidemia (%) 58.3 67.2 0.25 Lipid-lowering treatment (%) 44.7 45.3 0.93 AER (mg/24h) * 3.7 ± 16 358.5 ± 628 <0.001 Creatinine (mg/dl) * 0.8 ± 0.2 1.1 ± 1.2 <0.001 eGFR (ml/min/1.73 m2) 93.5 ± 29.5 66.7 ± 37.2 <0.001 Diabetic kidney disease (%) 32 82.8 <0.001 Normally distributed variables: mean ± SD; * Non-normally distributed variables: median ± interquartile range; Categorical variables: percents. PDR - proliferative diabetic retinopathy; BMI - Body Mass Index; WHR - waist-hip ratio; SBP - systolic blood pressure; DBP - diastolic blood pressure; HbA1c - glycated hemoglobin; eGDR - estimated glucose disposal rate; LDL - low density lipoprotein cholesterol; HDL - high density lipoprotein; AER-urinary albumin excretion rate; eGFR - estimated glomerular filtration rate. 264 Irina Duţă et al. 4

Table 2 Association between eGDR and proliferative diabetic nephropathy

Variable Wald OR 95%CI p Unadjusted eGDR 9.39 0.78 0.66-0.91 0.002 Adjusted eGDR 8.9 0.78 0.66-0.92 0.003 Sex 0.01 1.02 0.51-2.05 0.94 Duration of diabetes 0.29 1.01 0.97-1.05 0.58 Final regression model** eGDR 12.73 0.73 0.62-0.87 <0.001 BMI 6.15 0.88 0.80-0.97 0.013 ** Stepwise logistic regression analysis for eGDR and proliferative diabetic nephropathy. Variables rejected from the model were: sex, duration of diabetes, serum triglycerides, LDL cholesterol, HDL cholesterol and smoking.

DISCUSSION metabolic syndrome, was an independent risk factor (of age, sex, duration of diabetes, glycemic The main finding of the study is that insulin control and prevention cohort) for the development resistance, assessed using eGDR, is a characteristic and progression of retinopathy [14]. In the EURO- of patients with PDR (OR = 0.78; p = 0.002). DIAB Prospective Complications Study markers A low eGDR’s value, suggestive of high insulin such as triglycerides or WHR, suggestive for a low resistance, was independently (of sex, duration of insulin-sensitivity, have been associated with diabetes, BMI, LDL cholesterol, HDL cholesterol, increased incidence of retinopathy and proliferative triglycerides, smoking status) associated with PDR retinopathy [10]. In a recent analysis Pop et al. (OR = 0.73; p < 0.001). found that a low eGDR’s value was significantly Although in T2D some studies have found an associated with PDR [24]. association between decreased insulin sensitivity These studies suggest that an elevated eGDR and diabetic retinopathy [18, 19] in T1DM additional value, suggestive of insulin sensitivity, is associated studies would be needed to clarify this issue. In with a lower risk of diabetic retinopathy.Our results type 1 diabetes the study of the association insulin confirm and extend previous observations, showing sensitivity – diabetic retinopathy has been based on in addition that insulin resistance (assessed using two observations: adolescents with type 1 diabetes eGDR) would be the most important risk factor for are more insulin-resistant than non-diabetic controls PDR. [20] and puberty is characterized by an accelerated BMI was one of the variables retained in the evolution towards retinopathy [21-23]. In the Finn final regression model (OR = 0.88; p = 0.013) – a Diane study the risk of PDR onset was, after low BMI was associated with PDR. It is known adjustment for metabolic control (HbA1c) and BP, that the severity of the retinopathy is associated 1.8 times higher in those with diabetes onset before with the severity of renal morphological changes the age of 15 compared to those post-pubertal onset [25] and also with the severity of clinical nephro- of disease (15-40 years) [21]. After four years of pathy [26]. This data suggest that PDR would be follow-up in WESDR study (The Wisconsin Epide- associated with wasting syndrome due to nephropathy. miologic of Diabetic Retinopathy), the highest These findings are in accordance with other studies incidence of retinopathy was in the 10-12 years old [27-29]. These observations are side findings and age group of patients [22]. Other studies described were not the purpose of current study. a long time from diabetes diagnosis to com- Sex, duration of diabetes, lipid profile complications in adolescents diagnosed before puberty ponents and smoking were among the risk factors compared with those diagnosed after this period excluded from the models after the selection of the [23]. most important covariates for PDR. A possible These evidence suggest that in type 1 diabetes explanation of these observations could be a cohort elevated insulin sensitivity is associated with lower with an insufficient number of patients in order to risk of diabetic retinopathy and proliferative retino- achieve statistical significance for the variables pathy. mentioned above. A good management of modifiable Moreover in the DCCT study insulin resistance, risk factors such as lipid profile and smoking could quantified using eGDR, but neither insulin dose nor be another explanation for the decline in the 5 Insulin resistance and proliferative retinopathy in type 1 diabetes 265 strength of these risk factors. A third possibility factor asociated with PDR. Early screening for would be that eGDR’s formula captures other risk insulin resistance may identify high risk patients factors such as lipid profile components. Another who may benefit the most from preventive and possibility would be that insulin resistance (measured therapeutic interventions. using eGDR) would be the most important risk factor, independent of sex, duration of diabetes Acknowledgement. This paper is supported by the Sectoral lipid profile or smoking. Operational Programme Human Resources Development (SOP HRD) 2007-2013, financed from the European Social Fund and by the Romanian Government under the contract number CONCLUSIONS POSDRU/107/1.5/S/82839.

Our findings suggest that in type 1 diabetes Conflict of interest. The authors declare that there is no insulin resistance was the most important risk duality of interests associated with this manuscript.

Introducere. În diabetul zaharat tip 1, relaţia dintre insulino-rezistenţă şi retinopatia diabetică proliferativă este mai puţin studiată. Scopul acestui articol este de a examina relaţia dintre retinopatia diabetică proliferativă (ce poate determina cecitate) şi rezistenţa la insulină. Metodă. Design-ul studiului este cross-secţional. Analiza a inclus 167 pacienţi cu diabet zaharat tip 1. Rezistenţa la insulină a fost cuantificată cu ajutorul ecuaţiei eGDR (estimated glucose disposal rate). Retinopatia diabetică a fost evaluată oftalmoscopic folosind clasificarea Early Treatment Diabetic Retinopathy. Asocierea dintre eGDR şi retinopatia diabetică proliferativă a fost evaluată în modele uni şi multivariate utilizând regresia logistică stepwise a covariatelor. Contribuţia individuală a predictorilor din modelul final de regresie a fost examinată folosind Wald statistic. Rezultate. Pacienţii cu retinopatie diabetică proliferativă au prezentat valori semnificativ mai mici ale eGDR: 5,5 vs. 7 (p = 0.002). Rezultatele au rămas semnificative (p <0.001) ulterior ajustării pentru mai multe variabile (sex, durata diabetului, indicele de masă corporală, HDLc, LDLc, trigliceride, statutul de fumător). eGDR a fost variabila reţinută în modelul final al regresiei logistice stepwise (p <0.001) şi a prezentat cea mai puternică asociere cu retinopatia diabetică proliferativă (Wald = 12.73). Concluzii. La pacienţii cu diabet zaharat tip 1, insulino-rezistenţa a reprezentat cel mai important factor de risc asociat independent cu retinopatia proliferativă.

Correspondence to: Irina Duţă, Griviţei Street, No. 68, Câmpina Town, Postal code 105600, Prahova District, Romania, Tel.: +40743.019.182 E-mail: [email protected]

REFERENCES

1. SJOLIE AK, STEPHENSON J, ALDINGTON S, KOHNER E, JANKA H, STEVENS L, et al., THE EURODIAB IDDM COMPLICATIONS STUDY GROUP. Retinopathy and vision loss in insulin-dependent diabetes in Europe. Ophthalmology 1997; 104:252–260. 2. KLEIN R, KLEIN BE, MOSS SE, DAVIS MD, DEMETS DL. The Wisconsin epidemiologic study of diabetic retinopathy. II. Prevalence and risk of diabetic retinopathy when age at diagnosis is less than 30 years. Arch Ophthalmol 1984; 102:520-526. 3. ROSSING K, JACOBSEN P, ROSSING P, LAURITZEN E, LUND-ANDERSEN H, PARVING HH. Improved visual function in IDDM patients with unchanged cumulative incidence of sight-threatening diabetic retinopathy. Diabetes Care 1998; 21:2007-2015. 4. MOORE DJ, GREGORY JM, KUMAH-CRYSTAL YA, SIMMONS JH. Mitigating micro-and macro-vascular complications of diabetes beginning in adolescence. Vasc Health Risk Manag 2009; 5:1015-31. 5. AIELLO LP, DCCT/EDIC RESEARCH GROUP. Diabetic retinopathy and other ocular findings in the diabetes control and complications trial/epidemiology of diabetes interventions and complications study. Diabetes Care 2014; 37(1):17-23. 266 Irina Duţă et al. 6

6. 6.THE DIABETES CONTROLAND COMPLICATIONS TRIALRESEARCH GROUP. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993; 329(14):977-86. 7. THE DIABETES CONTROL AND COMPLICATIONS TRIAL/EPIDEMIOLOGY OF DIABETES INTERVENTIONS AND COMPLICATIONS RESEARCH GROUP. Retinopathy and nephropathy in patients with type 1 diabetes four years after a trial of intensive therapy. N Engl J Med 2000; 342:381-389. 8. TICIANADA COSTA RODRIGUES, MIRIAM PECIS, MIRELA JOBIMDE AZEVEDO, JORGE FREITAS ESTEVES, JORGE LUIZ GROSS. Ambulatory blood pressure monitoring and progression of retinopathy in normotensive, normoalbuminuric type 1 diabetic patients: A 6-year follow-up study. Diabetes Res Clin Pract 2006; 74:135-40. 9. HADJADJ S, DULY-BOUHANICK B, BEKHERRAZ A, BRLDOUX F, GALLOIS Y, MAUCO G, et al. Serum triglycerides are a predictive factor for the development and the progression of renal and retinal complications in patients with type 1 diabetes. Diabetes Metab 2004; 30(1):43-51. 10. CHATURVEDI N, SJOELIE AK, PORTA M, ALDINGTON SJ, FULLER JH, SONGINI M, et al., EURODIAB PROSPECTIVE COMPLICATIONS STUDY. Markers of insulin resistance are strong risk factors for retinopathy incidence in type 1 diabetes. Diabetes Care 2001; 24(2):284-9. 11. ORCHARD TJ, CHANG YF, FERRELL RE, PETRO N, ELLIS DE. Nephropathy in type 1 diabetes: a manifestation of insulin resistance and multiple genetic susceptibilities? Further evidence from the Pittsburgh Epidemiology of Diabetes Complication Study. Kidney Int 2002; 62(3):963-70. 12. ORCHARD TJ, OLSON JC, ERBEY JR, WILLIAMS K, FORREST KY, SMITHLINE KINDER L, et al. Insulin resistance- related factors, but not glycemia, predict coronary artery disease in type 1 diabetes: 10-year follow-up data from the Pittsburgh Epidemiology of Diabetes Complications Study. Diabetes Care 2003; 26(5):1374-9. 13. OLSON JC, ERBEY JR, FORREST KY, WILLIAMS K, BECKER DJ, ORCHARD TJ. Glycemia (or, in women, estimated glucose disposal rate) predict lower extremity arterial disease events in type 1 diabetes. Metabolism 2002; 51(2):248-54. 14. KILPATRICK ES, RIGBY AS, ATKIN SL. Insulin resistance, the metabolic syndrome, and complication risk in type 1 diabetes: “double diabetes” in the Diabetes Control and Complications Trial. Diabetes Care 2007; 30(3):707-12. 15. THORN LM, FORSBLOM C, FAGERUDD J, THOMAS MC, PETTERSSON-FERNHOLM K, SARAHEIMO M, et al, FINN DIANESTUDY GROUP. Metabolic syndrome in type 1 diabetes: association with diabetic nephropathy and glycemic control (the FinnDiane study). Diabetes Care 2005; 28(8):2019-24. 16. WILLIAMS KV, ERBEY JR, BECKER D, ARSLANIAN S, ORCHARD TJ. Can clinical factors estimate insulin resistance in type 1 diabetes? Diabetes 2000; 49(4):626-32. 17. DUTA I, RUSU E, COSTACHE A, RADULIAN G, ION DA. The association between insulin resistance and advanced renal disease in Type 1 diabetes. Romanian Journal of Diabetes Nutrition and Metabolic Diseases 2015; 22(2): 167-173. 18. PARVANOVA A, ILIEV I, FILIPPONI M, DIMITROV BD, VEDOVATO M, TIENGO A, et al. Insulin resistance and proliferative retinopathy: a cross-sectional, case-control study in 115 patients with type 2 diabetes. J Clin Endocrinol Metab 2004; 89: 4371-4376. 19. ANAN F, TAKAYUKI M, TAKAHASHI N, NAKAGAWA M, ESHIMA N, SAIKAWA T, et al. Diabetic retinopathy is associated with insulin resistance and cardiovascular autonomic dysfunction in type 2 diabetic patients. Hypertens Res 2009; 32: 299-305 . 20. STEPHANIE A. AMIEL, ROBERT S. SHERRWIN, DONALD C. SIMONSON, ALBERT A. LAURITANO, WILLIAM V. TAMBORLANE. Impaired Insulin Action in Puberty. N Engl J Med 1986; 315:215-219. 21. HIETALA K, HARJUTSALO V, FORSBLOM C, SUMMANEN P, GROOP PH, FINN DIANE STUDY GROUP. Age at Onset and the Risk of Proliferative Retinopathy in Type 1 Diabetes. Diabetes Care 2010; 33(6):1315-9. 22. GRAUR M, HANCU N, POPA AR. Retinopatia diabetica. In: Complicatiile cronice ale diabetului zaharat. Farma Media, Tg. Mures, 2008: 126. 23. JILL NORRIS KOSTRABA, JANICE S DORMAN, TREVOR J ORCHARD, DOROTHY J BECKER, YUKASHI OHKI, DEMETRIUS ELLIS, et al. Contribution of Diabetes Duration Before Puberty to Development of Microvascular Complications in IDDM Subjects. Diabetes Care 1989; 12:10 686-693. 24. POP A, CLENCIU D, ANGHEL M, RADU S, SOCEA B, MOTA E, et al., ROMDIANESTUDY GROUP. Insulin resistance is associated with all chronic complications in type 1 diabetes. J Diabetes 2015; doi:10.1111/1753-0407.12283. 25. KLEIN R, ZINMAN B, GARDINER R, SUISSA S, DONNELLY SM, SINAIKO AR, et al. The relationship of diabetic retinopathy to preclinical diabetic glomerulopathy lesions in type 1 diabetic. Diabetes 2005; 54:527-533. 26. HIETALA K, FORSBLOM C, SUMMANEN P, GROOP PH. Heritability of proliferative diabetic retinopathy. Diabetes 2008; 57:2176-2180. 27. LEE YJ, CHO S, KIM SR. The association between serum adiponectin levels and nutritional status of hemodialysis patients. Ren Fail 2011; 33(5):506-11. 28. QI Y, TAKAHASHI N, HILEMAN SM, PATEL HR, BERG AH, PAJVANI UB, et al. Adiponectin acts in the brain to decrease body weight. Nat Med 2004; 10(5):524-9. 29. PANDURU NM, SARAHEIMO M, FORSBLOM C, THORN LM, GORDIN D, WADEN J, et al., ON BEHALF OF THE FINNDIANE STUDY GROUP.Urinary Adiponectin Is an Independent Predictor of Progression to End-Stage Renal Disease in Patients With Type 1 Diabetes and Diabetic Nephropathy. Diabetes Care 2015.

Received May 30 2015