ABSTRACT
Objective:
Cardiovascular diseases (CVD) are the most important cause of morbidity and mortality in patients with type 1 diabetes (T1D). Children with T1D have a similar or higher prevalence of being overweight (OW) or obese (Ob) compared to healthy peers. The aim of this study was to determine the prevalence of CVD risk factors in children and adolescents with T1D and the impact of obesity and sex differences on these factors.
Methods:
Data of patients aged 10-21 years and who had been using intensive insulin therapy with a diagnosis of T1D for at least three years were evaluated. Patients were divided into normal weight (NW), OW and Ob groups based on body mass index percentiles. Risk factors for CVD (obesity, dyslipidemia, hypertension) were compared between groups, and impact of gender was also analyzed.
Results:
Data of 365 patients (200 girls, 54.8%), were evaluated. Prevalence of OW/Ob was 25.9% and was significantly higher in girls (30.6% vs 20.1%, p<0.001). Rate of hypertension was highest in OW/Ob girls followed by OW/Ob boys, and similar in NW girls and boys (p=0.003). Mean low density lipoprotein cholesterol (LDL-c) and triglyceride (TG) levels were highest in OW/Ob girls, followed by OW/Ob boys, NW girls and NW boys, respectively (p<0.001 and p<0.001, respectively). Mean high density lipoprotein-cholesterol (HDL-c) levels were similar among groups. Rates of high LDL-c and TG were similar between OW/Ob girls and boys and higher than NW girls, followed by NW boys (p<0.001 and p<0.001, respectively). The rate of low HDL-c was similar in OW/Ob girls and boys, and higher than NW girls, followed by NW boys (p<0.001). Overall, girls were 1.9 times more likely than boys to have two or more risk factors for CVD. Factors associated with risk for CVD in multiple logistic regression analyses were being a girl, followed by higher daily insulin dose, higher hemoglobin A1c, and longer duration of diabetes (r=0.856; p<0.001).
Conclusion:
In spite of the increased prevalence for obesity in both sexes, the trend for CVD risk factors was greater in Ob girls, followed by Ob boys and NW girls. Girls with T1D are more likely to be OW/Ob and to have CVD risk than boys, highlighting the need for early intervention and additional studies to elucidate the causes.
What is already known on this topic?
Cardiovascular diseases (CVD) are the most important cause of morbidity and mortality in patients with type 1 diabetes (T1D). Children with T1D had similar or higher prevalence of being overweight (OW) or obese (Ob) compared to their healthy peers.
What this study adds?
Girls with T1D are more likely to be OW and Ob and to have CVD risk than boys. Interventions to reduce the risk of CVD in adults with T1D should begin from childhood and be tailored to compensate for gender variations.
Introduction
Cardiovascular diseases (CVD) are the most important cause of morbidity and mortality in patients with type 1 diabetes (T1D) (1). CVD is more common, occurs earlier, and has a higher mortality rate in patients with T1D than individuals without diabetes (2). It has been reported that the pathogenesis of CVD accelerates in patients with T1D and although CVD rarely presents in childhood, subclinical damage to the cardiovascular system begins in the pediatric age group (3). Preventing CVD by alleviating cardiovascular risk factors is an important treatment goal for patients with T1D.
Obesity in the childhood period is also considered to be an important risk factor for CVD in adulthood (4). Being overweight (OW) and having abdominal adiposity have been associated with atherosclerosis and dyslipidemia (5). Children with T1D had similar or higher prevalence of being OW or obese (Ob) compared to their healthy peers, with rates reaching 25-35% (6,7,8,9,10,11). The risk of CVD is already high in T1D, and this risk may be increased with Ob. It is observed that Ob-related comorbidities increasingly affect individuals with T1D. Ob individuals with T1D have lower insulin sensitivity than non-Ob patients with T1D, and it has been shown recently that CVD risk factors are increased in children with T1D who are OW or Ob (12). A report of the Exchange study showed a higher prevalence of hypertension and dyslipidemia in Ob patients with T1D than in non-Ob patients (13). A recent study involving children and young adults with T1D showed a higher rate of metabolic syndrome (MS) in OW and Ob individuals (14). Unhealthy behavioral habits contribute to increased cardiovascular risk. Smoking and a sedentary lifestyle are associated with major CVD morbidity and mortality in individuals with T1D (15).
In this study we aimed to determine the prevalence of CVD risk factors in children and adolescents with T1D and the impact of OW/Ob and sex differences on these factors.
Methods
The data of patients aged 10-21 years and who had been using intensive insulin therapy (89% multiple-dose insulin therapy, 11% insulin pump) with a diagnosis of T1D for at least three years were evaluated in this descriptive, cross-sectional study. The diagnosis of T1D was confirmed by the presence of anti-GAD, anti-islet, or anti-insulin antibodies.
Body weights were measured in kilograms using a digital weighing scale (SECA, 769) with a sensitivity of 0.1 kg. Heights were measured in meters with the patients in the standing position using a stadiometer (Harpenden, Holtain Ltd., Crymych, Dyfed, UK) with a sensitivity of 0.1 cm. All the measurements were made by a nurse trained in auxology. Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared and converted to BMI percentiles for age and sex using the Centers for Disease Control and Prevention (CDC) growth charts from 2000 (CDC: BMI for children and teens, http://www.cdc.gov/nccdphp/dnpa/bmi/bmi-for-age.htm). Those with a BMI <10 percentile were considered as underweight (UW), BMI ≥10 and BMI <85 percentile as normal weight (NW), BMI ≥85 and BMI <95 percentile as OW, and BMI ≥95 percentile as Ob. There were only two patients who were UW and we did not include the underweight group in further analysis in the study, since the number was very small and as we aimed to compare NW versus OW and Ob cases. Since the number of patients in the Ob group was low, when comparing in terms of CVD risk factors regarding genders, the OW and Ob groups were combined and the cases were divided into two groups as NW and combined OW/Ob. At the time of evaluation, the chronological age, body weight, height, BMI, blood pressure, hemoglobin A1c (HbA1c), lipid profile, and daily insulin dose per body weight (IU/kg/day) were noted. The age at diagnosis and duration of diabetes were also recorded. All cases had HbA1c values checked every three months and the HbA1c value used in the study was the average of the four HbA1c values examined in the last year before the evaluation was taken. Insulin dose was defined as the total daily units of insulin divided by the body weight in kilograms and total daily insulin dosage was obtained by randomly selecting three days from the patient records.
Blood pressure was measured with a standardized automatic sphygmomanometer in the right arm with an appropriately sized cuff in the sitting position after 10 minutes of rest, and the average of three measurements was recorded for analysis. For the definition of high blood pressure, the definitions determined by the National High Blood Pressure Education Program study group for children and adolescents were used (16). A mean systolic or diastolic blood pressure above the 95th percentile for age, sex, and height was considered high.
Lipid profiles including triglyceride (TG), total cholesterol (TC), high density lipoprotein-cholesterol (HDL-c), and low-density lipoprotein-cholesterol (LDL-c) were determined in fasting blood samples. Serum lipid levels were measured by an autoanalyzer (Roche Diagnostics). Lipid values were evaluated according to the recommendations of the American Diabetes Association and the International Diabetes Federation (17,18). LDL-c ≥100 mg/dL, TC ≥200 mg/dL, and TG ≥150 mg/dL were defined as high. HDL-c <40 mg/dL for 10-16 years old and <40 mg/dL for boys >16 years old and <50 mg/dL for girls >16 years old was considered low. Dyslipidemia was considered to be an elevation of one or more lipid or lipoprotein levels, or for HDL-c, reduced levels. The presence of at least two risk factors for CVD, such as the presence of OW/Ob (BMI ≥85th percentile), high TG levels, low HDL-c levels, and hypertension were considered as increased cardiovascular risk.
The study was approved by the Hacettepe University Institutional Local Ethics Committee (approval number: 16969557-2202, date: 03.12.2019).
Statistical Analysis
Data analyses were performed using Statistical Package for the Social Sciences (SPSS) for Windows, version 22.0 (SPSS Inc., Chicago, IL, United States). The normality of the continuous variables was tested by the Kolmogrov-Smirnow test. Continuous data are described as mean±standard deviation (SD) and categorical data were described as the number of cases (%). Statistical differences between two independent groups were compared by Student’s t-test. The differences among more than two independent groups were analyzed by one-way ANOVA. When the p-value from one-way ANOVA was statistically significant, post hoc Bonferroni’s corrections were used with either a p-value of 0.0167 or 0.0125 to know which group differ from others.
Clinical and laboratory parameters associated with having two or more risk factors for CVD were first evaluated using univariate analysis. The factors that were significant in the univariate analysis were then evaluated with multivariate logistic regression analysis. Independent variables used in the regression analysis were: gender; age at evaluation; HbA1c level; daily insulin dose; duration of diabetes; and presence of type 2 diabetes in the family. A p-value of less than 0.05 was considered statistically significant.
Results
Three hundred and sixty-five patients (200 girls, 54.8%) who were followed for at least three years with the diagnosis of T1D were included in the study. The mean age of the patients at the time of evaluation was 16.4±3.7 years, and ranged from 10 to 21 years. Mean age at diagnosis was 9.6±3.6 years (range 5-17 years), and mean diabetes duration was 7.0±3.1 years (range 3-14 years). Most (89%) of the patients were using multiple-dose insulin therapy and 11% were using insulin pump therapy. Mean HbA1c was 8.8±2.6%, and mean total daily insulin dose was 1.0±0.3 IU/kg. There was no difference between boys and girls in terms of age at evaluation, HbA1c level, duration of diabetes and daily dose of insulin.
The prevalence of OW, and Ob was 19.3%, and 6.6%, respectively, in the study population (22.6%, and 8% in girls and 15.2%, and 4.9% in boys). The rate of OW and Ob was significantly higher in girls than boys (p<0.001). The three groups (NW, OW and Ob) were at similar ages and had similar HbA1c levels at the time of the evaluation. BMI-SD score was positively correlated to the duration of diabetes (r=0.768, p<0.001). Daily dose of insulin per kg body weight was the highest in the Ob group, followed by the OW and NW group (1.7±0.3 IU/kg/d, 1.3±0.3 IU/kg/d, and 0.9±0.2 IU/kg/d, respectively, p<0.001). The Ob group had the highest rate of hypertension followed by the OW and NW groups (p<0.001). Moreover, the Ob group had the highest mean LDL-c, TG and TC levels and the lowest HDL-c levels, followed by the OW and NW groups (p<0.001, p<0.001, p<0.001 and p<0.001, respectively) (Table 1).
Overall, hypertension prevalence was 9.4% in the whole population. The rate of hypertension was highest in OW/Ob girls, followed by OW/Ob boys, and similar in NW girls and boys (p=0.003) (Table 2). Mean LDL-c and TG levels were highest in OW/Ob girls, followed by OW/Ob boys, NW girls and NW boys, respectively (p<0.001 and p<0.001 respectively) (Figure 1, Panel 2, 3 and Supplementary Table 1). The mean HDL-c levels were similar among groups (Figure 1, Panel 1 and Supplementary Table 1). Rates of high LDL-c and TG were similar between OW/Ob girls and boys and higher than in NW girls, followed by NW boys (p<0.001 and p<0.001, respectively) (Table 2). The rate of low HDL-c was similar in OW/Ob girls and boys, and higher than in NW girls, followed by NW boys (p<0.001) (Table 2). The proportion having at least two risk factors for CVD was highest in the OW/Ob girls, followed by OW/Ob boys, NW girls, and lowest in NW boys (p<0.001) (Table 2). Overall, girls were 1.9 times more likely than boys to have two or more risk factors for CVD (37/199 vs 16/164).
The patients with two or more risk factors for CVD were older, had higher HbA1c levels, were using a higher dose of daily insulin per kg of body weight, and had longer duration of diabetes in comparison to those with a low risk for CVD which persisted in gender specific analyses (Table 3).
Factors associated with risk for CVD in multiple logistic regression analyses were being a girl, followed by higher daily insulin dose, higher HbA1c, and longer diabetes duration (r=0.856; p<0.001) (Table 4).
Discussion
The prevalence of obesity in children and adolescents with T1D is increasing, in parallel with the general population (6,19,20). One in four T1D patients in this study had a BMI over the 85th percentile, which is a higher prevalence than that of healthy Turkish children (21,22,23). Although there are no nationwide statistics on the frequency of Ob in children and adolescents with T1D, a few regional studies also found a similar prevalence (24,25). Previous research from various regions of the world revealed that 25-38.5% of children with T1D were either OW or Ob (6,8,9). According to the Diabetes Control and Complications and The Epidemiology of Diabetes Interventions and Complications Trials (EDIC), intensive insulin treatment, that is multiple insulin injections and insulin pumps, may be responsible for the increased prevalence of Ob in T1D patients compared to the non-diabetic population (26).
In the present study, girls were 1.5 times more likely to be OW/Ob than boys (30.6% vs 20.1%), whereas this is more common in boys in the non-diabetic pediatric population (21,27). There are similar studies showing that girls with T1D are particularly vulnerable to become OW/Ob, particularly during puberty (11,28). It is not known exactly why the obesity rate is higher in girls with T1D although insulin resistance, gender-specific hormonal differences, changes in body composition and energy metabolism during puberty in girls and boys may all contribute to the observed sex-related disparities in prevalence of OW/Ob in children and adolescents with T1D (14,29).
The most common cause of death in T1D is CVD. The prevalence of CVD-related and all-cause mortality is ten times higher in people with T1D than in the general population, despite recent advances in glycemic control and CVD risk management (30). The cumulative incidence of coronary artery disease by age 55 is as high as 35% in T1D compared to <10% in the non-diabetic population. Risk of stroke is four times higher in T1D compared to non-diabetic population. Similarly, people with T1D have a five-fold higher risk and incidence of peripheral vascular disease than people without diabetes. Atherosclerosis begins in childhood and is more common in children and adolescents with T1D, as shown in autopsy and epidemiological studies (31,32). Obesity may have a further impact on the risk of CVD associated with T1D. The EDIC study reported a significant increase in lipid levels and blood pressure in subjects with excessive weight gain while using intensive insulin therapy, similar to insulin resistance syndromes (26). The results of the present study showed that CVD risk factors in OW/Ob diabetic adolescents were more prevalent than that of normal-weight adolescents with T1D. We also showed that girls are more likely to have risk factors for CVD in comparison to boys in a T1D population.
Hypertension is also more common in patients with T1D than in the general population (33). Risk factors for hypertension in adolescents with T1D include obesity and hyperglycemia (34). In the current study the overall prevalence of hypertension was similar to the figures reported in previous studies (35). However, we showed that hypertension was more common in Ob girls with T1D; this is in contrast to the prevalence in the Ob non-diabetic population where hypertension is more common in boys than in girls (36,37). However, it should be noted that this difference (21% vs 15%, p=0.008) was less remarkable than the difference between NW and the Ob diabetic population (5-6% vs 15-21%, p<0.001).
Dyslipidemia is recognized as one of the most important CVD risk factor in patients with diabetes (38). Poor metabolic control causes diabetic dyslipidemia, which is characterized by increased LDL-c and TG levels and low HDL-c levels (39,40). The EDIC study showed that elevated LDL-c and low HDL-c lasting more than 10 years were associated with a higher CVD risk in the T1D population (41). Diabetes is a disease that causes accelerated atherosclerosis and therefore requires regular lipid monitoring and early intervention (42). Most studies showed that 26-58% of children with T1D have lipid levels above these defined values (43,44,45). Adolescent girls with T1D were reported to have higher mean TG, LDL-c, and Apolipoprotein B levels than boys, although HbA1c levels were not different from boys, suggesting that girls with T1D are at risk for CVD starting from younger ages (45). In the current study, the mean LDL-c and TG were greater in Ob girls than in Ob boys, however the prevalence of increased LDL-c and TG as CVD risk factors was similar in both Ob boys and girls with T1D and was significantly higher than that of NW girls with T1D followed by that of NW boys with T1D. In addition, similar to other studies, there was no difference in HbA1c values between the four groups. Despite comparable mean levels across all groups, prevalence of low HDL-c as a risk factor was associated with similar variation in the current study. These findings suggest that obesity plays a significant role in inducing dyslipidemia, and that female sex may also be a secondary risk factor.
A recent study showed that the rate of MS was higher in OW (8.1%) and Ob (35.3%) cases than NW (4.9%) in a population of children, adolescents and young adults with T1D (14). These rates were higher than those in the non-diabetic population where it is reported to have a median prevalence of 11.9% (range 2.8-29.3%) and 29.2% (range 10-66%) in OW and Ob children, respectively (46). The prevalence of MS (0-1%) is even lower in NW, non-diabetic children (46). In children and adolescents, a limited number of studies report a gender predilection for MS, which is mostly in favor of boys (36). However, this predilection does not seem to be preserved in patients with T1D (47,48).
The presence of at least two risk factors for CVD, such as the presence of obesity (BMI ≥85 percentile), high TG levels, low HDL-c levels, and hypertension were considered as increased cardiovascular risk in the current study. In the current study, a high proportion of OW/Ob adolescents with T1D also had other CVD risk factors including elevated TGs (31.9%), elevated blood pressure (19.1%) and low HDL-c (20.2%). Similar to our study, different studies have shown that the risk of hypertension and dyslipidemia increases with an increase in BMI and the majority of Ob children with T1D had at least two CVD risk factors (13,49). This is important because the coexistence of multiple CVD risk factors increases the risk of morbidity and mortality. Having a higher number of CVD risk factors is associated with more progressive early atherosclerotic lesions (50,51).
In the current study, approximately 20% of the girls and 10% of the boys with T1D had at least two CVD risk factors. Girls were twice as likely to have a CVD risk factor as boys. In addition to obesity, being female appears to be a risk factor in terms of CVD. In the regression analysis the most important factor increasing the CVD risk factor was being a girl. In Ob girls, the presence of CVD risk factors was approximately three times more common than in normal-weight girls and 1.3 times more common than in Ob boys (34.4% vs 11.6%, and 34.4% vs 27.3%, respectively). Brown et al. (52) showed that TC, LDL-c, BMI and blood pressure were higher in adolescent girls than boys, similar to our findings. The prevalence of co-existence of several CVD risk factors has been reported to be more common in girls than in boys (40,53,54,55).
Given the later onset of CVD in women in the general population, premenopausal women appear to be protected from CVD compared to men of similar age. These differences among genders have been attributed to biological differences, including sex chromosomes and hormonal status, as well as to gender differences in behavioral and sociocultural aspects (56). Considering that the risk of CVD increases as estrogen levels decrease after menopause, estrogens appear to be protective against CVD. By increasing the amount of HDL-c and inhibiting LDL-c oxidation, estrogen in the blood causes a lipoprotein profile change, which reduces the accumulation of oxidized LDL-c in the arterial wall (57). Gender-related protection from CVD is lost in females with diabetes. Females with T1D have a roughly 40% greater risk of all-cause death and twice the risk of fatal and non-fatal vascular events compared to males with T1D (58). It is not fully explained why the CVD risk is higher in females with T1D compared to males. The mechanisms proposed to date argue for gender differences in both biological and psychosocial factors, as well as in the management of diabetes and CVD risk factors (35). Hormonal changes seen in women with T1D may contribute to a greater atherogenic lipid profile, insulin resistance, higher inflammation, and loss of vasoprotective effect (59). This is a mechanism that warrants further investigation. It was also thought that the higher prevalence of OW/Ob and CVD risk in adolescent girls with T1D might be related to the difficulties encountered in the management of T1D in this population. Sedentary lifestyle and physical inactivity, which are among the leading modifiable risk factors for CVD, are more common in girls with T1D compared to boys (59).
There is also substantial evidence that insulin resistance in T1D may diminish the female protection against CVD. The CACTI study has shown that adults with T1D have increased insulin resistance compared to nondiabetic controls (60) and, more recently, that this effect of T1D on insulin resistance appears to be greater in women (61). The reasons for greater insulin resistance in women with T1D compared to women without diabetes are unclear, but it is thought that differences in estrogen levels may play a role. It has been also shown that healthy girls are less insulin sensitive than boys, but compensate for their decreased sensitivity with increased insulin secretion (62). However, this compensation does not exist in diabetic children and adolescents.
It is also not clear when excess CVD risk begins in women with T1D. We have shown that sex differences in the prevalence of CVD risk factors emerge in the early course of T1D, beginning with adolescence. In particular, adolescent girls with T1D have higher CVD risk factors than boys. Therefore, early diagnosis and gender-specific intervention in girls and young women with T1D will facilitate earlier reduction of the risk of CVD with its attendant morbidity and mortality later in life.
Study Limitations
One of the strengths of this study was the large sample size of children and adolescents with a long follow-up period from a single center. Participants in the study were exposed to comparable management strategies, which is an strength. However, this study also has some limitations. First, since data were gathered retrospectively, measurements such as waist circumference were absent, despite the fact that it is optimal to evaluate obesity using waist circumference in children and adolescents. Second, some potential confounders of obesity and CVD risk, including levels of physical activity, dietary practice, and socioeconomic status, were not accounted for. Thirdly, this was not a longitudinal study, nor was it a population-based study and thus the prevalence rates are not applicable to the entire population of children and adolescents with T1D. The last limitation was that we did not compare our patients to a control group of non-diabetic children.
Conclusion
This study showed that more than a quarter of children and adolescents with T1D were OW or Ob. In spite of the increased prevalence for obesity in both sexes, the trend for CVD risk factors was increased most in Ob girls, followed by Ob boys and then girls who were of NW.
Girls with T1D are more likely to be OW and Ob and to have CVD risk factors than boys, highlighting the need for early intervention and additional studies to elucidate the causes. Interventions to reduce the risk of CVD in adults with T1D should begin from childhood and be tailored to compensate for gender variations. Individuals with T1D may experience less CVD morbidity and mortality as a result of early identification of CVD risk factors and possible gender-specific treatments, which may ultimately lead to better long-term outcomes.
Click the link to access Supplementary Table 1: http://glns.co/b4d13