To our knowledge, few papers have focused on assessing the nutritional status of children and adolescents with type 1 diabetes mellitus, and most of them were not cited in our study because of their small sample size, wide age limits or lack of a control group.
Research presented in 2016 at the International Society for Pediatric and Adolescent Diabetes showed that in the years 2010–2014 the incidence of T1DM increased. It is estimated that currently 18.4 per 100,000 people get this disease in the age group up to 18 years. It has been observed that the majority of new patients are between 10 and 14 years of age [17]. Due to the above reports, our study was conducted in a similar age group. The median age of people with type 1 diabetes mellitus was 12.0 years and 11.0 years in the control group.
The most frequently mentioned factors that have a huge impact on the formation of excessive body weight are the difficulty to adequately manage glucose control, an improper diet rich in animal fats and products with a high glycemic index and low levels of daily physical activity [17, 18]. In patients with T1DM, the anabolic effect of insulin has a significant impact [19]. A connection was found between HbA1c and high accumulation of fat in the abdominal area, which was confirmed in our study, as well as in Ingberg’s study [20, 21]. Good metabolic management is essential not only for normal growth and development in pediatric patients with T1DM, but also for reduced or delayed progression of existing complications [22, 23]. Our research showed median HbA1c was 7.4% (57 mmol/mol). Similar metabolic control was presented in the works of Majewska et al. (7.7%; 61 mmol/mol), Särnblad et al. (7.9%; 63 mmol/mol) and Pietrzak et al. (8.0%; 64 mmol/mol) [24,25,26]. We proved that between the two insulin therapies used, with respect to maintaining good metabolic management there is a definite difference in favor of PIP (7.1 vs. 8.0%; 54 vs. 64 mmol/mol; P < 0.001). However, in our study, we found that girls using PIP had higher HbA1c than boys (7.5 vs. 6.6%), but when comparing both insulin therapies girls had similar results (7.5 vs. 7.7%). Samuelsson et al., in their large population study compiled on the basis of data from the national registry, showed a gender difference—girls had poor metabolic control, i.e., higher HbA1c values [27]. The reason may be that girls have worse metabolic control during adolescence than boys [28]. Among the factors influencing this may be the difference in hormonal changes between the respective genders during this period [29]. Some of the studies have proven that both insulin dosages and HbA1c values are significantly higher in girls [30, 31].
Methods for assessing the nutritional status of children and adolescents make it possible to determine the correct nutritional status or detect disorders at an early stage. In pediatric patients, because the anthropometric measurements described below are dependent on age and sex, they should be referred to as developmental norms [14]. According to the IDF, it is important that the patient is assessed according to local percentile meshes [13]. In Poland, percentile meshes proposed by the World Health Organization are recommended up to the age of 3, whereas for children from 3 to 18 years of age, national standards drawn up on the basis of the OLA and OLAF studies are used. Distributions of anthropometric parameters were prepared for body height, weight, BMI, WC and HC [14].
In our study, the BMI value in the T1DM group was 19.2 kg/m2. Published data of other authors were similar—Maffeis et al. (19.3 kg/m2), Ab El Dayem et al. (20.1 kg/m2) and Lipsky et al. (21.3 kg/m2) [32,33,34]. In recent years, the reliability of the BMI indicator in the assessment of nutritional status has been repeatedly denied. In identifying the risk of developing metabolic disorders by estimating visceral fat mass in children and adolescents aged 7–17 years using magnetic resonance imaging, Brambilla et al. proved that WC and WHtR can be a much more sensitive indicator than BMI [35]. WHtR is used to assess the distribution of abdominal fat in people with excessive body weight [36, 37]. In this study, WHtR was used to assess the occurrence of abdominal obesity in children and adolescents with T1DM, which made it possible to compare the results with the research of Nawarycz et al., who examined > 26,000 healthy children in the Łódź region, aged 7–19 years [37]. The authors showed that a WHtR value > 0.5 can be used as a simple and universal criterion for the initial diagnosis of abdominal obesity among adolescents, in both boys and girls. In Szadkowska’s study central obesity was twice as common in children with diabetes as in the general population [37, 38]. Our research showed that in overweight and obese children with T1DM WHR was 1.1; WC and HC were 75.0 and 94.5 cm, respectively. No differences were found according to WHtR. Average WHtR and WC in Maffeis’s study were 0.44 and 67.8 cm, respectively [32]. Completely different results were received as well by Marigliano et al. (0.39 and 57.5 cm) [39]. This is probably because the authors calculated the averages for the above-listed parameters for the whole group, not only for people with excessive body mass.
In our research the percentage of fat and muscle mass in the T1DM group was 19.1% and 36.1%, respectively. Similar results were presented by Maffeis et al. (fat mass: 18.5%) and with a little divergence by Margaliano et al. (fat mass: 15.0%), Majewska et al. (24.5 and 45.5%, respectively) and Lipsky et al. (fat mass: 27.5%) [24, 32, 34, 39]. In our study, despite similar median percentages of fat mass between the two insulin therapies, after subgrouping by gender, the content was lower in people using CSII, but not statistically significantly so. However, the obtained medians are within the normal range. However, a large spread between the results may be related to the fact that the CSII group was characterized by a lower HbA1c, BMI and percentage of people with excessive fat mass compared to the MDI. Unfortunately, it was not possible to compare muscle mass with results from other authors because of our usage of dissimilar parameters (muscle mass instead of fat-free mass). A statistically significant difference in the percentage of muscle tissue could result from the fact that in the group of boys we had a dispersion of results, especially in those with MDIs. Their higher BMI may confirm that this is an unreliable indicator for people with higher muscle mass, possibly related to greater physical activity. The general conclusion from the comparison of this parameter in healthy participants and with T1DM may indicate that adolescents more often prefer spending their time passively than actively. People with T1DM are often more aware of the impact of physical activity on health [40].
The nutritional status of most respondents was normal, however, 13% of the adolescents with T1DM were overweight. Moreover, 10% of those using CSII and 30% of MDI had too much fat mass. Our research showed that BMI, fat mass and parameters indicating abdominal obesity positively correlated with HbA1c.Assessment of anthropometric parameters among young diabetics should be one of the standard periodic tests. Methods for assessing nutritional status are safe and non-invasive, and the results of the study can be used by physicians in diabetic patients, helping them monitor the metabolic control of the disease, which determines the proper somatic development of pediatric patients. This is why it is worth controlling the nutritional status as a whole, and not just height and weight separately.
The recommendation about the number of meals per day for diabetic children is related to insulin dosage. Therefore, careful meal planning is most important [40]. The introduction of validated mobile medical applications for diabetes care would facilitate appropriate therapeutic decisions and positively impact outcomes, including HbA1c levels and hypoglyemia rates [41]. The diabetics had more main meals and fewer sweetened snacks per day compared to healthy children. Nevertheless, the difference between meals and snacks is difficult to assess, because the overall food composition can be similar [42].
There were a few limitations to our study. First was the small number of participants because the children with T1DM were from two voivodeships and randomly included in this research. Second, the disproportion between the number of persons in the groups with different types of insulin therapy occurred because most of the children use CSII rather than pens. When designing the study we were guided by the data about the number of pediatric patients with T1DM (around 22,000) in Poland and estimates of people using CSII among them (16,600, which is 75% of this group) [6]. However, the relationships observed between features show a huge potential to develop studies on a wider scale and on larger groups. Research will be conducted from this perspective by our team.