Can HbA1c predict quality of life in children with type 1 diabetes mellitus?

The aim of this study was to examine a possible association of Hb A1c , quality of life (QoL), tness, and electrophysiological parameters in children with type 1 diabetes mellitus (T1DM). The study population (n = 34) consisted of patients with T1DM (n = 17) and an age- and BMI-matched healthy control group (n = 17). Hb A1c was obtained from patients with T1DM at time of diagnosis (T0), at 6 months (T6), at 12 months (T12), and at time of study inclusion (Tstudy). QoL was determined with a standardized questionnaire (KINDL-R). All children completed a 6-minute walk test (6MWT) to evaluate their tness level. Electrodiagnostic studies established upper and lower limb motor and sensory nerve conduction velocities (NCV).


Abstract Background
The aim of this study was to examine a possible association of Hb A1c , quality of life (QoL), tness, and electrophysiological parameters in children with type 1 diabetes mellitus (T1DM).

Methods
The study population (n = 34) consisted of patients with T1DM (n = 17) and an age-and BMI-matched healthy control group (n = 17). Hb A1c was obtained from patients with T1DM at time of diagnosis (T0), at 6 months (T6), at 12 months (T12), and at time of study inclusion (Tstudy). QoL was determined with a standardized questionnaire (KINDL-R). All children completed a 6-minute walk test (6MWT) to evaluate their tness level. Electrodiagnostic studies established upper and lower limb motor and sensory nerve conduction velocities (NCV).

Results
Higher Hb A1c (Tstudy) was associated with lower QoL showing in the subscales self-esteem, friends and school. Higher Hb A1c (T6) and (T12) was associated with lower QoL in the subscale self-esteem. Based on various subscales, perceived problem areas differed signi cantly between children and their parents.
No differences in tness level and NCV were found between patients and controls except for a signi cantly slower median motor NCV in patients. Hb A1c was not associated with NCVs at this early stage of disease.

Conclusions
Good metabolic control re ected by adequate Hb A1c values seems to be important for a good QoL in children with T1DM. Early Hb A1c analysis serves as predictor for QoL during follow-up.
Trial registration: Retrospectively registered Background One important goal of diabetes management in children and adolescents is to achieve psychological well-being and a high level of quality of life (QoL) despite chronic disease burden (1). The association of metabolic control and QoL in patients with type 1 diabetes mellitus (T1DM) has been already shown in several studies (2, 3). Hoey et al. found that good metabolic control -indicated by lower Hb A1c values -is related to better QoL in adolescents with T1DM (3). Additionally, girls showed a poorer overall QoL than boys (3). In the assessment of QoL it seems important to separate ratings of children and their parents (4). Ratings of adolescent QoL and burden perceived by parents were different (3).
Findings are controversial concerning the association of metabolic to nerve conduction abnormalities (5).
Peripheral neuropathy is one possible complication of T1DM as it occurs more often with increasing duration of disease and is mainly found in adult patients associated with increased morbidity and mortality (6). Symptomatic peripheral neuropathy is uncommon in children, but nerve conduction studies demonstrated subclinical neuropathy in 28-58% of children with T1DM (7-10). The progression of subclinical peripheral nerve conduction abnormalities is predicted by poor metabolic control and is associated with body height and enduring hypoglycemia (5,11). Despite modern multiple insulin injection therapy enabling good metabolic control, children and adolescents with insulin-dependent diabetes may still show subclinical nerve dysfunction (12). There is evidence that early de cits in nerve conduction predict the progress of diabetic neuropathy (13) enforcing the focus both on motivating children for metabolic control and on the early detection of children with nervous system abnormalities (14).
The tness of children with T1DM is controversially discussed in the literature. Some studies suggest the tness of children with T1DM is reduced compared to healthy controls (15). Chronic hypoglycemia in patients with T1DM might lead to alterations in aerobic and anaerobic muscle functions, as assessed by maximal isometric grip strength, and an incremental cycling test until exhaustion, respectively. Impaired muscle function was found in children with poor glycemic control, whereas children with good metabolic control did not show reduced tness (16).
Findings in QoL, subclinical neuropathy and tness of children with T1DM are still under debate. Data is sparse concerning the association of long-term Hb A1c and QoL later on. Therefore, the goal of this study was to evaluate a possible association of metabolic control from the onset of disease over time, QoL, nerve conduction, and tness in patients with T1DM compared to a healthy age-, sex-and BMI-matched control group.

Methods
This single-center study was conducted at the Department of Pediatrics, Saint Vinzenz Hospital, Zams, Austria.
Patients with T1DM were recruited during outpatient visits. Children with other chronic diseases, genetic syndromes or neurological disorders were excluded from the study. All participants had no abnormalities in motor and cognitive development. The subjects were ambulatory, normally physically active and on no additional medication. Healthy children matched for age, sex and body mass index [BMI] seen as outpatients for routine or preoperative investigations were the control group.
In patients, Hb A1c values were obtained during outpatient visits. Both patients and healthy controls, and their parents, lled out separately a standardized QoL questionnaire. Each participant completed a 6minute walk test (6MWT) and underwent assessment of nerve conduction velocities (NCVs). In addition, weight and height of all participants were measured using a wall-mounted stadiometer and a calibrated weight scale. Afterwards, the BMI was computed and converted to standardized BMI using the national BMI reference (17). The standardized follow-up is presented in Fig. 1.
The regional university ethics committee (Ethikkommission der Medizinischen Universität Innsbruck) approved the study. Written informed consent was obtained from participants and/or a parent prior to participation in the study.

Hb A1c measurements
Hb A1c measurements of each patient were obtained at time of diagnosis of T1DM (T0), at 6 months (T6), at 12 months (T12), and at time of study (Tstudy).

Quality of Life Questionnaire
The questionnaire "Kinder Lebensqualität Fragebogen" measuring QoL in children and adolescents (revised version KINDL-R) (18,19) was lled out by the children and one of their parents (proxy version). The questionnaire consists of 24 items equally divided into six subscales: physical wellbeing, emotional wellbeing, self-esteem, family, friends, and school. The items measure the average feelings and experiences during the past week and are rated on a ve-point scale (from 1 = never to 5 = always). Mean item scores of all subscales and the total QoL score were calculated and transformed to a scale ranging from 0 to 100 with 100 representing the highest QoL.

6-minute walk test (6MWT)
Each subject completed a 6MWT to determine the personal level of tness according to the guidelines of the American Thoracic Society as previously published and modi ed for children (20). Before and after the walk, heart rate was measured with a nger pulse oximeter (Nonin Flight Stat, Aeromedix, Jackson, USA).

Nerve conduction velocity (NCV)
Objective, sensitive and validated measure of nerve function is the assessment of NCV (21). Surface electrodes were used for assessing nerve conduction with standard technique. Motor conduction velocities were measured unilaterally in the median, ulnar, peroneal and tibial nerves. Sensory conduction velocities were measured unilaterally in median, ulnar and sural nerves.
The electrophysiological recordings were evaluated by two independent raters.

Statistical Analysis
Statistical Package for Social Sciences for Windows (SPSS Inc., Version 15.0) was used for the statistical analysis.
Due to the small sample size nonparametric tests were chosen. Group differences were assessed using the Mann-Whitney-U-test and correlation of metric variables was analyzed with Spearman correlation.
Data presented are the mean and standard deviation (SD). Statistical tests were performed two-tailed with an alpha level of < 0.05 indicating statistical signi cance.

Results
Thirty-four participants were eligible, and all agreed to participate in the study. All patients were included in statistical analyses. Subjects were grouped in 17 patients with T1DM (6 girls, 11 boys) and 17 controls (6 girls, 11 boys).
Demographic data and clinical characteristics are presented in Table 1. Table 1 Baseline characteristics of the participants. There were no group differences between patient and control groups for the total QoL score or with any of the child-rated and parent-rated subscales. No sex differences were found. Total QoL and subscale values are presented in Table 2. Hb A1c (T0) was neither correlated with the total score of QoL, nor with any subscales child-rated or parentrated.

Anthropometric parameters and 6-minute walk test (6MWT)
Patients and controls did not differ signi cantly in anthropometric parameters, walking distance (6MWD) and heart rate (pre/post walking) of 6MWT as presented in Table 3. The 6-minute walk distance (6MWD) was not associated with QoL and any subscales. 6MWD and heart rate were not correlated with Hb A1c (Tstudy) in the patient group.

Nerve conduction velocity (NCV)
NCVs are presented in Table 4. Patients and controls did not differ signi cantly except for a signi cantly slower median motor NCV in patients. Correlation between Hb A1c (Tstudy) and NCVs did not reach statistical signi cance. Overall, there were no signi cant differences in patients with T1DM using an insulin pump versus patients using no insulin pump. Additionally, there were no sex differences.

Discussion
The most important nding of the present study is that Hb A1c obtained during the rst year after diagnosis of disease is inversely correlated to certain subscales of QoL of patients with T1DM at Tstudy, i.e. some 5 years after disease onset. Our results are therefore in line with previous studies in which good metabolic control was shown to be associated with better QoL (3). The ratings of parents and children differed from each other signi cantly as already found by Hoey et al. (3,4) enforcing the importance of separate ratings.
In previous studies, Hb A1c was measured at time of study inclusion, e.g. more than 5 years after diagnosis (3), whereas in the present study, Hb A1c was obtained during the rst year after diagnosis of disease. The development of Hb A1c from onset of T1DM over a year was found to be associated with QoL at Tstudy, i.e. 4.9 (3.6) years after onset. When Hb A1c was higher in the rst year, patients reported signi cantly lower QoL on average 4 years later. These results concur with Hb A1c being a potential predictor of QoL. The adjustment of metabolic control re ected by Hb A1c from onset of T1DM obviously has an impact on the well-being of the children later on. Importantly, patients indicated lower self-esteem. As low self-esteem is associated with psychiatric disorders such as depression or substance use (22,23), children with T1DM might need close follow-up.
The development of Hb A1c after disease onset varied during the observational period. Hb A1c was highest at T0. At T6 the decline in Hb A1c is probably due to a more rigorous adjustment of metabolic control.
Later on, the motivation of children is likely to be reduced as indicated by an increase in Hb A1c at T12.
This underlines the importance to keep the focus on good metabolic control and on the acceptance of the disease in order to enhance QoL also later on.
With respect to Hb A1c (Tstudy), the QoL ratings of parents and patients differed. Children with higher Hb A1c values rated themselves lower on overall QoL especially on the subscale of self-esteem whereas parents perceived lower QoL of their child on the subscales of friends and school. This is an important issue for diabetes management because the different perceptions of patients and parents may call for the need of tailored support in order to discover problem elds and to maximize QoL.
The nal part of the study was to measure tness level and electrophysiological abnormalities in children with T1DM. As patients and controls did not differ signi cantly in the results of the 6MWT the subjects were presumably on average at the same tness level. This is in contrast to previously published ndings of reduced tness in children with T1DM (15). Notably, only children with poor metabolic control showed alterations in aerobic an anaerobic muscle functions (16).
For electrodiagnostic parameters, a detailed neurophysiological examination of children with T1DM compared to healthy children was performed. There were no electrophysiological abnormalities in patients with T1DM compared to control group, except for a signi cantly slowed median motor NCV. Additionally, no correlation between NCVs and Hb A1c was found at any point in time. This is in contrast to other studies reporting frequent subclinical neuropathy in diabetic children (13,14). Notably, mean disease duration was considerably longer in previous studies, exceeding seven years (14). Diabetic polyneuropathy did therefore likely not occur at this early stage of disease in the present study. Nerve conduction studies are the gold standard for the detection of subclinical neuropathy and determining neurophysiological measurements (15). Measuring NCVs in children is a big challenge, which renders it often di cult to nd differences in T1DM patients. Due to artefacts and limited compliance of the children, a supramaximal stimulation was not always ensured in our study. Potentially, more suitable screening tools such as vibration sensation thresholds and thermal discrimination thresholds that are quicker and easier in the implementation might be more appropriate for use in studies of children (14).

Conclusion
In this comprehensive study, children with T1DM showed no clinical or subclinical differences to healthy controls as they were on the same tness level and showed no neurophysiological abnormalities. However, there was an association of Hb A1c at T(6), T (12), and T(study) with subscales of the "Kinder Lebensqualität Fragebogen" measuring QoL in children and adolescents (revised version KINDL-R), as assessed 4.9 (3.6) years after establishing the diagnosis. It seems to be important to teach patients with T1DM from onset of disease about the importance of continuously emphasizing good metabolic control in order to avoid psychological impairment and to facilitate a better QoL during follow-up.

Declarations
Ethics approval and consent to participate: The regional university ethics committee (Ethikkommission der Medizinischen Universität Innsbruck) approved the study. Written informed consent was obtained from participants and/or a parent prior to participation in the study.
Availability of data and materials: The datasets during and/or analyzed during the current study available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no competing interests.