Study design and data collection
This study had a cross-sectional design and was conducted in 10 Brazilian cities, from all geographic regions of the country, with patients followed in 14 public clinics between 2011/2014.
All patients received free health care (NPH and regular insulins, syringes, needles, glucometers and strips for blood glucose monitoring) from the National Brazilian Health Care System (NBHCS). Each clinic provided data for at least 50 T1D outpatients that were treated by an endocrinologist in secondary or tertiary care settings. Included patients were those with the diagnosis of T1D done by a physician and the need for continuous insulin use since the diagnosis, at least 13 years of age, and followed at each diabetes center for at least 6 months. Pregnant or lactating women, patients who had an acute infection or ketoacidosis in the three preceding months or had a history of renal transplantation were excluded [12].
The total sample was composed of 1,760 patients. They were diagnosed as having T1D between 1960 and 2014. Three hundred and sixty-seven patients (20.9%) were adolescents, according to the World Health Organization criteria [13] (13 to 19 years old) and formed the sample of this study. Each center had a local ethics committee that approved the study. Patients and/or their parents where necessary, signed a written informed consent agreeing with the participation in the study.
The collected data were: current age, age at diagnosis, self-reported color-race (White, Black, Brown (“parda”), Asian (“amarela”) and Indigenous (“indígena”)) [14], diabetes duration, years of school attendance, frequency of self-monitoring of blood glucose (SMBG), smoking status, type of prescribed insulin therapeutic regimens (ITR), self-reported adherence to diet (following at least 80% of the time the prescribed diet) [15] and to prescribed ITRs [12], BMI and self-reported frequency of physical activity (at least three times a week). Family history of diabetes, obesity, hypertension and coronary diseases in first degree relatives were also assessed. The coexistence of another health care insurance, besides that offered by the BNHCS was also assessed.
Adequate glycemic control was defined as the presence of HbA1c levels < 7.5% (58 mmol/mol) [16], and inadequate glycemic control was defined as HbA1c levels being ≥ 7.5% (58 mmol/mol). HbA1c was measured using high-performance liquid chromatography (HPLC, Bio-Rad Laboratories, Hercules, California, USA. The last value of HbA1c in the previous year was obtained from the medical records. Fasting triglycerides, HDL cholesterol, total cholesterol, alanine aminotransferase (ALT), aspartate aminotransferase (AST) and gamma-glutamyl transferase were measured using enzymatic techniques and serum uric acid by an uricase-based commercial (mg/dl). Creatinine was measured using a colorimetric assay kit, corrected for standardized creatinine assay by mass spectrometry. All the above measurements were performed with BioSystem, model A25; Barcelona, Spain. Friedewald’s equation was used to calculate LDL cholesterol values [17]. ITRs were stratified as follows: exclusive use of intermediate insulin (NPH) or regular insulin, long-acting insulin analogs plus short acting insulin or the use of continuous subcutaneous insulin infusion (CSII). Overweight was defined as a BMI of ≥ 85th percentile, and obesity as a BMI of ≥ 95th percentile according to age and gender [18] and current smoking as the use of more than one cigarette per day. Pre-hypertension and hypertension were defined as the presence of a systolic blood pressure (sBP) and/or a diastolic blood pressure (dBP) 120/<80 to 129/<80 mm Hg and ≥ 130/80 mm Hg, respectively [19].
Sample Calculation and Economic Status Evaluation
Sample calculation of the study has been previously described [12, 20]. The sample represented the distribution of T1D cases across four geographic regions of Brazil, estimated using the overall population distribution reported in the 2000 Brazilian Institute of Geography and Statistics Population Census (IBGE) [21], combined with national estimates of diabetes prevalence, to determine the minimum number of patients to be studied in each region [22]. Economic status was defined according to the Brazilian Economic Classification Criteria that takes in account the education level [23]. The following economic status categories were considered: high, middle, low, and very low.
Diabetes-Related Chronic Complications Assessment
Evaluation of Renal Function
Renal function was estimated by the CKD-EPI equation [24] in patients with age ≥ 16 years, by the Schwartz formula in patients younger than 16 years [25] and was expressed as estimated glomerular filtration rate (eGFR) in milliliters per minute per 1.73m2 (mL/min/1.73m2). Albuminuria concentration (immunoturbidimetry, detection limit: 0.01mg/dl) was measured at least twice from a morning urine sample. The presence of albuminuria was defined as an albuminuria ≥ 30mg/dl. Patients with normal renal function had an eGFR ≥ 60 mL/min/1.73m2 and the absence of albuminuria. Chronic kidney disease (CKD) was defined as an eGFR < 60 mL/min/1.73m2, with or without the presence of albuminuria and an eGFR ≥ 60 mL/min/1.73m2 with the presence of albuminuria [26, 27].
Evaluation of Retinopathy
The screening for DR was performed by mydriatic binocular indirect ophthalmoscopy (BIO), by a retinal specialist. The classification of DR was assessed in the eye that was the most compromised. Each eye was classified based on whether DR was present. Patients were then classified according to the international classification as: absent, non-proliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR) and macular edema [28].
Metabolic Syndrome Assessment
The definition of MS was done according to the International Diabetes Federation criteria [29]. Adolescents aged 16 years or older were classified according to the same criteria adopted for adults as follows: central obesity: waist circumference (WC) ≥ 90 cm for South American men or ≥ 80 cm in South American women; triglycerides ≥ 150 mg/dL (1.7 mmol/L) or on drug therapy for elevated triglycerides; HDL < 40 mg/dL (1.03 mmol/L) in men or < 50 mg/dL (1.29 mmol/L) in women or on drug therapy for low HDL; elevated BP ≥ 130×85 mmHg or using antihypertensive drugs.
Adolescents aged 13 to 15 years old were classified according to the following criteria: (1) abdominal obesity: WC > 90th percentile for age and gender; (2) triglycerides ≥ 150 mg/dL (1.7 mmol/L); (3) HDL-c < 40 mg/dL (1.03 mmol/L); (4) elevated BP ≥ 130 × 85 mmHg [20]. As there are no reference values of percentiles on abdominal waist in the Brazilian population, we used the 90th percentile for each age and gender group of our sample. Considering that all participants have T1D, central obesity plus an additional factor was necessary for diagnosing MS [30].
Statistical Analysis
For the purpose of statistical analysis overweight and obesity were considered together. An exploratory analysis was initially performed, and the data were presented as mean (± SD) or median, interquartile range [IQR] for continuous variables and percentage for discrete variables. Parametric and non-parametric tests were used for comparison between the groups as indicated. Pearson’s correlation coefficient was calculated when applicable.
We have done backward Wald logistic multivariate analysis with overweight/obesity as a dependent variable (outcome variable); for the independent variables, those with p < 0.2 in exploratory analysis, or those which presented relevance, mainly related to demographic and social data, such as gender, age, diabetes duration, years of school attendance, self-reported color-race, level of care, family history of hypertension, sBP and dBP, use of anti-hypertensive drugs, insulin dose/kg, proportion of basal/bolus, geographic region of the country and GFR. Adjustments for social economic status, self-reported color-race and age at diabetes diagnosis were performed. All analyses were performed using the Statistical Package for the Social Sciences (SPSS version 17.0, SPSS, Inc., Chicago, Illinois, USA). Odds ratios with 95% confidence intervals (CIs) were calculated where indicated. A two-sided p value less than 0.05 was considered to be significant.