Journal of Physical Activity and Health, 2015, 12, 232 -237 http://dx.doi.org/10.1123/jpah.2013-0024 © 2015 Human Kinetics, Inc.
Official Journal of ISPAH www.JPAH-Journal.com ORIGINAL RESEARCH
Correlation Between Glycemic Control and Physical Activity Level in Adolescents and Children With Type 1 Diabetes Cristiane Petra Miculis, Wagner De Campos, and Margaret Cristina da Silva Boguszewski Background: The aim of this study was to correlate glycemic control (GC) and variables of physical activity levels (PAL) in children with type 1 diabetes mellitus (T1DM). Methods: Fifty children and adolescents with T1DM were selected. Personal and medical data for the patients were collected. Physical evaluations of body weight and sexual maturation were undertaken. Bouchard’s questionnaire was applied to evaluate PAL as well as for time spent on physical activities. Results: Sixty-four percent of the subjects were sexually mature. Differences were observed between females and males in insulin dose, duration of light physical activity, and sleeping time (P < .05). Ninety percent presented poor GC and 80% had a low PAL. Fasting blood glucose (FBG) was significantly correlated with PAL, with sedentary time, and with sleeping time. Glycated hemoglobin (HbA1c) was significantly correlated with sedentary time and sleeping time. Among the three groups of PAL (insufficient × moderate × active) there were differences in HbA1c (%), FBG (mg/dL), duration of disease (years), and insulin dose (UI/kg/day) (P < 0.001). Conclusion: GC was significantly correlated with PAL. Among the three groups of physical activity level, the most active group was seen to have the best GC. Keywords: special needs population, adolescent, evaluation, exercise, metabolic health, physical activity A fundamental treatment parameter for type 1 diabetes mellitus (T1DM) is adequate glycemic control (GC).1 Poor GC raises the risk of microvascular complications and correlates with risk factors for macroangiopathies.1,2 Adequate GC can be reached through medical treatment (multiple insulin doses) as well as a healthy lifestyle, such as a balanced diet and regular physical activity.3 Regular physical activity is insufficient in the general pediatric population4,5 and in T1DM children and adolescents. T1DM children and adolescents have a low regular physical activity index for the reasons of fear of hypoglycemia, inadequate time available for physical activities, poor GC, and lower physical fitness levels.6–8 The effect of regular physical activity on GC is variable9–14 depending on factors such as frequency of physical activity per week, remaining active for 12 weeks or more,10,15 and reducing sedentary behaviors such as watching television and playing computer games.12,16 These factors demonstrate a significant and positive effect on GC including a lower glycated hemoglobin (HbA1c) level in T1DM children and adolescents, independent of the type of activity (ie, aerobic versus strength and endurance).9–11,15 Studies using lower weekly exercise frequency combined, and not combined, with shorter intervention duration report an insignificant effect on GC.12–14 These studies report other positive outcomes such as a better sense of well-being,12,14 augmented physical fitness,13 reduced insulin dose,13,14 and improved waist circumference.14 We hypothesize that lower physical activity levels (PAL) are associated with poorer GC in T1DM children and adolescents. We explored the relationship among GC, PAL, and time spent on a range of physical activity intensities in children and adolescents with T1DM. Miculis ([email protected]) and da Silva Boguszweski are with the Dept of Pediatrics, Federal University of Paraná, Curitiba, Brazil. De Campos is with the Dept of Physical Education, Federal University of Paraná, Curitiba, Brazil. 232
Methods Fifty children and adolescents with T1DM were studied (n = 21 boys and n = 29 girls). All subjects were 9 to17 years old, living in Curitiba, Brazil or its metropolitan areas, and free of diabetes-related complications (ie, retinopathy, neuropathy, or nephropathy). The sample size (n) was calculated for a significance level of 5% and a power of 80% for mean difference and correlations.17 We used a one-tailed test of the hypothesis and mean difference (difference = 3, SD = 8.5) for this cross-sectional study. Parents or legal guardians of the children were informed of all procedures and the objective of this research, including risks and benefits. Participation was free of any financial compensation, performed according to the criteria of 196/96 National Council Resolution,18 and approved by the local ethics committee of the Hospital de Clinicas, Federal University of Paraná, Brazil, under CEP/SD 666.001.09.01 and CAAE 0104.0.000.091 to 09 on February 20, 2009. Written informed consent was obtained from all participants or their parents. A review of the medical records was performed in search of clinical and laboratory data indicating the fasting blood glucose (FBG) and HbA1c measured the week before the medical appointment at the Diabetes Outpatient Clinic. The hexokinases method was used for FBG while high-performance liquid chromatography (HPLC) was used for HbA1c. All analyses were conducted using Architect c8000 (Abbott, Chicago, IL). Children and adolescents, with the assistance of their parents, replied to a questionnaire seeking personal and medical data such as actual insulin dose, number of self-monitored blood glucose levels, diabetes duration, and physical activity level. The dose of each insulin type was summed and divided by body weight (kg), resulting in daily insulin dose (UI/kg/day). Diabetes duration was recorded in years. A physical evaluation was conducted to determine body weight (kg) with minimal clothing (ie, shorts and a light t-shirt) on an
Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 2
Glycemic Control and Physical Activity 233
electronic scale (TOLEDO, model 2096 PP, 0.1 kg of precision; São Bernardo do Campo, Brazil). Pubertal stages were classified according to the Tanner scale;19 they were self-identified using figures in the privacy of a separate room in the presence of the parents and one evaluator of the same sex as the child. Physical activity level was measured through Bouchard’s physical activity questionnaire,20 which is validated for children as young as 10 years old, also in Brazil.21 This questionnaire collects data on the physical activity of the last 3 days, ending in Saturday or Sunday. The day is divided into 96 periods of 15 minutes. For each 15-minute period, energy expenditure is classified using a scale from 1 to 9, corresponding to the primary activity of that period (see below for description). The median energy expenditure (kcal/kg/15 min) of each category is used to calculate daily energy expenditure. Cut-off points for PAL were based on the sample’s percentile of energy expenditure (kcal/kg/day).22 Girls were classified as: insufficiently active (PAL < 37.06 kcal/kg/day), moderately active (PAL ≥ 40.18, < 44.01 kcal/kg/day), or active (PAL ≥ 44.01 kcal/ kg/day). Boys were classified as insufficiently active (PAL < 36.45 kcal/kg/day), moderately active (PAL ≥ 39.72, < 42.74 kcal/kg/day), or active (PAL ≥ 42.74 kcal/kg/day). Bouchard’s questionnaire20 allows estimation of the time spent, in minutes, on each type of activity. Bouchard’s energy table20 describes the 1 to 9 scale as follows: 1 = Sleeping Time (S), corresponds to 0.26 kcal/kg/15 min; 2 = Sedentary Behavior Time (SB), corresponds to 0.38 kcal/kg/15 min; 3 to 5 = Light Physical Activity Time (LPA), corresponds to 0.57 to 0.84 kcal/kg/15 min; and 6 to 9 = Moderate to Vigorous Physical Activity Time (MVPA), corresponds to 1.2 to 2 kcal/kg/15 min.
Statistics To test for normality, Kolmogorov-Smirnov and Shapiro-Wilk tests were conducted, considering parametric data for P > .05. Only 2 variables reached that criteria (daily insulin dose and sedentary behavior time), therefore we chose to use nonparametric tests. Spearman’s rho (r) bivariate correlation was used to compare metabolic and physical activity variables. For differences between sexes, the Mann-Whitney U test was applied. Friedman’s test was chosen to detect differences among 3 groups of PAL and GC variables.
Friedman’s test is similar to ANOVA, however it does not locate where the differences are among the groups. For these reasons, the Wilcoxon test was conducted, followed by Tukey’s post hoc test. Variability and central tendency measures were used (mean, standard deviation for normally distributed data, and median with minimum-maximum for nonnormally distributed data). SPSS (version 13.0, IBM, Chicago, IL) was used when P < .05.
Results Table 1 shows clinical characteristics for each sex. A higher daily insulin dose was observed among girls (P = .017) as opposed to boys. According to the Tanner classification, 12% were prepubertal, 64% pubertal, and 24% postpubertal. In general, the sample had poor GC: 89.4% had elevated values of FBG (> 210 mg/dL) and 91.5% had HbA1c > 7.5%. In the analyses of PAL and the time spent in a range of activity intensities (Table 2), a significant difference was observed between girls and boys for the time spent in LPA and S among the median of the 3 days and for days 1 and 2 (P < .05). Girls were more active in each of these cases. We found that 28% of the sample was insufficiently active, 52% were considered moderately active, and 20% were active. The majority of the sample (n = 33/50) chose Saturday as the third day for their Bouchard’s questionnaire. When the FBG was used as an indication of GC, a significant relationship was found between PAL (median of the 3 days), SB (day 3), and S (median of the 3 days and day 3) (Figure 1a–e). Using HbA1c as an indication of GC, a relationship was found between SB (day 1) and S (day 3) (P < .05) (Figure 1f–g). The duration of T1DM was directly correlated with daily insulin dose (P = .025, r = .318) and SB on day 3 (P = .031, r = .306). Similarly, daily insulin dose was correlated with S and day 1 (P = .006, r = .383), day 2 (P = .012, r = .351), and the median of the 3 days (P = .017, r = .336). The number of self-monitored blood glucose levels was significantly related to total PAL on day 2 (P = .019, r = –.334), LPA on day 2 (P = .016, r = –.340), SB on day 1 (P = .032, r = .304), and the median of the 3 days (P = .026, r = .314). The sample was divided into 3 groups according to PAL percentiles in both sexes, resulting in significant differences of GC between the groups (P < .0001) (Table 3).
Table 1 General Characteristics of the Sample Girls (n = 29)
Boys (n = 21)
Mean
Standard Deviation
Mean
Age (years)
11.86
1.57
12.67
2.29
Weight (kg)
43.68
9.97
43.46
12.15
T1DMD (years)
4.60
3.23
4.60
3.21
Capillary glycemia (times/day)
3.55
1.02
3.86
.57
Insulin dose (UI/kg/day)
1.06*
.34
.83
.28
FBG (ml/dL)
233.44
93.42
287.6
150.52
10.5
2.72
10.42
2.29
HbA1c (%)
Note. T1DMD = type 1 diabetes mellitus duration; FBG = fasting blood glucose. *P < .05, female different from male.
Standard Deviation
Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 2
Figure 1 — Correlation between glycemic control variables, physical activity level, and time spent on different activities intensities. (a) FBG vs PAL day 3 (p = .021; r = –.335); (b) FBG vs PAL mean (p = .027; r = .324); (c) FBG vs sedentary behavior day 3 (p = .022; r = –.334); (d) FBG vs sedentary behavior median (p = .045; r = –.294); (e) FBG vs sleeping time day 3 (p = .044; r = .295); (f) HbA1c vs sedentary behavior day 1 (p = .046; r = .292); (g) HbA1c vs sleeping time day 3 (p = .018; r = .343). FBG = fasting blood glucose; PAL = physical activity level. 234
Glycemic Control and Physical Activity 235
Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 2
Table 2 Physical Activity Level and Time Spent on a Range of Activity Intensities
PAL 1 (kcal/kg/day) PAL 2 (kcal/kg/day) PAL 3 (kcal/kg/day) PAL mean (kcal/kg/day) SB 1 (min) SB 2 (min) SB 3 (min) SB mean (min) LPA 1(min) LPA 2 (min) LPA 3 (min) LPA mean (min) MVPA 1 (min) MVPA 2 (min) MVPA 3 (min) MVPA mean (min) S 1 (min) S 2 (min) S 3 (min) S mean (min)
Girls (n = 29) Median Min–Max 39.06 32.29–55.14 38.80 31.62–57.14 42.04 33.44–56.58 40.18 33.58–55 615 405–825 615 330–810 480 255–780 553.33 370–755 165 0–375 195* 0–420 195 0–540 175 0–425 60 0–285 45 0–390 150 0–435 100 0–345 540* 135–720 540* 435–705 600 480–750 575* 490–705
Boys (n = 21) Median Min–Max 37.94 32.08–49.35 38.75 33.78–58.88 40.3 31.94–54.37 39.72 35.15–53.92 645 450–855 675 390–885 510 180–810 625 345–835 135 15–450 120 30–450 165 45–525 140 60–445 75 0–315 90 0–360 90 0–330 95 10–290 480 405–690 480 405–690 615 420–720 530 470–700
Note. PAL = physical activity level; SB = sedentary behavior; LPA = light physical activity; MVPA = moderate to vigorous physical activity; S = sleeping time; numbers 1, 2, and 3 represent different days in physical activity questionnaire. *P < .05, female different from male.
Table 3 Glycemic Control by Physical Activity Level Groups Mean ± Standard Deviation HbA1c (%) FBG (mg/dL) T1DMD (years) Insulin (UI/kg/day)
Insufficiently Active 10.78 ± 2.46 277 ± 157.86 5.16 ± 4.21 1.10 ± 0.23
Moderately Active 12.12 ± 2.94 257.20 ± 59.57 5.30 ± 2.90 1.42 ± 0.36
Active 9.56 ± 2.31* 240.20 ± 162.49** 2.25 ± 1.91* 0.89 ± 0.38*
Note. T1DMD = type 1 diabetes mellitus duration; FBG = fasting blood glucose. *Active group significantly different from moderately and insufficiently active groups, two-tailed Wilcoxon’s test P < .0001; **active group significantly different from insufficiently active groups, two-tailed Wilcoxon’s test P < .0001, Tukey’s post hoc test.
Discussion Our findings demonstrated a significant association between FBG and PAL, and sedentary and sleeping activities. Likewise, we found an association between HbA1c and sedentary and sleeping activities. Studies in patients with T1DM demonstrate that time spent on certain types of sedentary behaviors (eg, watching television, playing computer games) positively correlates to FBG and HbA1c levels, indicating that increased time on those kinds of activities may reduce GC.9,16 Some authors do not explore a correlation between PAL and GC, however they demonstrate the exercise-induced effect on GC.9–11,15 HbA1c could be reduced by 0.96% compared with basal values in T1DM adolescents after 12 weeks of endurance and aerobic training.15 Shorter or longer training duration interventions also demonstrate a significant and long-lasting effect on HbA1c and glycemia levels.10,11,23,24 Herbst et al6,9 shows that the
most significant factor for GC is regular physical activity; in the same way, moderate or moderate to vigorous physical activities are related to HbA1c after 16 weeks of physical training for adolescents with T1DM.23 HbA1c glycemia significantly correlates to physical activity (minutes per week) with a greater magnitude than other variables of physical fitness.25 Among our sample, most of the T1DM subjects were at puberty (25 girls and 14 boys) with poor GC. This is not uncommon among adolescents with T1DM.3,26 Also, the median insulin dose was higher for girls as compared with boys, however the reasons for this poor GC is multifactorial, influenced by maturation status, and is beyond the scope of our study. Surprisingly, the difference in median 3-day PAL did not reach statistical significance when comparing girls to boys. Most studies6,9,12,27 show higher PAL for males. One study28 found a higher regular physical activity index among females versus males in adults with T1DM. The difference could be assigned to leisure-
Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 2
236 Miculis, De Campos, and da Silva Boguszewski
time physical activities (eg, domestic tasks, personal care, and light physical activity). Among our sample, we found that girls spent more time on LPA compared with boys, suggesting that the lack of difference in PAL could be assigned to the difference in LPA. This is in accordance with recent data23 from adolescents with T1DM. Sleeping time on the majority of days was higher for girls and correlated positively with FBG, HbA1c, and daily insulin dose. This is a novel fact, because most studies explore the role of hypoglycemia associated with sleep deprivation and its pathological alterations29–31 instead of the effect of sleeping duration on the hyperglycemia. Our findings suggested that sleep duration on weekend days for adolescents may be associated with poor GC. One could speculate if this association occurred because the subjects lost their diabetes management routine on the weekends and slept more than on weekdays, thus exceeding the time for self-monitoring and insulin dosing. However, the cross-sectional design of our study does not permit us to deduce causality. As in the general population, 28% of our sample was insufficiently active, 52% moderately active, and 20% active.4,5 We used percentile division of the sample data because there is no Brazilian reference for the PAL of children and adolescents with T1DM. We are concerned that children with T1DM do not engage in an adequate amount of physical activity as recommended for the prevention of diabetes-related diseases, since they are at higher risk than their healthy peers.26,32 The most active group of our study showed the lowest mean values of HbA1c, FBG, diabetes duration, and daily insulin dose when compared with the moderately active and insufficiently active subgroups. In a sample of 23,000 T1DM patients, Herbst et al6 found differences between the number of days of regular physical activity and GC variables; the most active were associated with better HbA1c levels. The higher one’s PAL, the better their GC, depending on the frequency and regularity of the physical activity.9,10,15 In conclusion, our findings show that GC is associated with physical activities and sleeping time and reinforce that the regular practice of physical activity in children and adolescents with T1DM is highly recommended, whether it is aerobic or endurance exercise.
References 1. Diabetes Control and Complication Trial (DCCT). The effect of intensive treatment of diabetes on the development and progression of longterm complications in insulin-dependent diabetes mellitus. The Diabetes Control and Complications Trial Research Group. N Engl J Med. 1993;329:977–986. PubMed doi:10.1056/NEJM199309303291401 2. Nathan DM, Cleary PA, Backlund JY, et al. Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. N Engl J Med. 2005;353:2643–2653. PubMed doi:10.1056/ NEJMoa052187 3. Teles SA, Fornés NS. Food consumption and metabolic control in children and adolescents with type 1 diabetes mellitus. Rev Paulista Pediatr. 2011;29(3):378–384. doi:10.1590/S0103-05822011000300012 4. Ribeiro RQ, Lotufo PA, Lamounier JA, Oliveira RG, Soares JF, Botter DA. Additional cardiovascular risk factors associated with excess weight in children and adolescents: the Belo Horizonte heart study. Arq Bras Cardiol. 2006;86:408–418. PubMed doi:10.1590/S0066782X2006000600002 5. Fakhouri THI, Hughes JP, Brody DJ, Kit BK, Ogden CL. Physical activity and screen-time viewing among elementary school–aged children in the united states from 2009 to 2010. JAMA Pediatr. 2013;167(3):223–229. PubMed doi:10.1001/2013.jamapediatrics.122
6. Herbst A, Kordonouri O, Schwab KO, Schmidt F, Holl RW. Impact of physical activity on cardiovascular risk factors in children with type 1 diabetes: a multicenter study of 23,251 patients. Diabetes Care. 2007;30:2098–2100. PubMed doi:10.2337/dc06-2636 7. Valerio G, Spagnuolo MI, Lombardi F, Spadaro R, Siano M, Franzese A. Physical activity and sports participation in children and adolescents with type 1 diabetes mellitus. Nutr Metab Cardiovasc Dis. 2007;17:376–382. PubMed doi:10.1016/j.numecd.2005.10.012 8. Brazeau AS, Rabasa-Lhoret R, Strychar I, Mircescu H. Barriers to physical activity among patients with type 1 diabetes. Diabetes Care. 2008;31:2108–2109. PubMed doi:10.2337/dc08-0720 9. Herbst A, Bachran R, Kapellen T, Holl RW. Effects of regular physical activity on control of glycemia in pediatric patients with type 1 diabetes mellitus. Arch Pediatr Adolesc Med. 2006;160:573–577. PubMed doi:10.1001/archpedi.160.6.573 10. Sideraviciute S, Gailiuniene A, Visagurskiene K, Vizbaraite D. The effect of long-term swimming program on glycemia control in 14-19-year aged healthy girls and girls with type 1 diabetes mellitus. Medicina (Kaunas). 2006;42:513–518. PubMed 11. Wang YC, Stewart S, Tuli E, White P. Improved glycemic control in adolescents with type 1 diabetes mellitus who attend diabetes camp. Pediatr Diabetes Denmark; 2008:29–34. 12. Aman J, Skinner TC, de Beaufort CE, Swift PG, Aanstoot HJ, Cameron F. Associations between physical activity, sedentary behavior, and glycemic control in a large cohort of adolescents with type 1 diabetes: the Hvidoere Study Group on Childhood Diabetes. Pediatr Diabetes. 2009;10:234–239. PubMed doi:10.1111/j.1399-5448.2008.00495.x 13. D’hooge R, Hellinckx T, Van Laethem C, et al. Influence of combined aerobic and resistance training on metabolic control, cardiovascular fitness and quality of life in adolescents with type 1 diabetes: a randomized controlled trial. Clin Rehabil. 2011;25:349–359. PubMed doi:10.1177/0269215510386254 14. Ramalho AC, de Lourdes Lima M, Nunes F, et al. The effect of resistance versus aerobic training on metabolic control in patients with type-1 diabetes mellitus. Diabetes Res Clin Pract. 2006;72:271–276. PubMed doi:10.1016/j.diabres.2005.11.011 15. Mosher PE, Nash MS, Perry AC, LaPerriere AR, Goldberg RB. Aerobic circuit exercise training: effect on adolescents with well-controlled insulin-dependent diabetes mellitus. Arch Phys Med Rehabil. 1998;79:652–657. PubMed doi:10.1016/S0003-9993(98)90039-9 16. Margeirsdottir HD, Larsen JR, Brunborg C, Sandvik L, Dahl-Jorgensen K. Strong association between time watching television and blood glucose control in children and adolescents with type 1 diabetes. Diabetes Care. 2007;30:1567–1570. PubMed doi:10.2337/dc06-2112 17. Wallymahmed ME, Morgan C, Gill GV, MacFarlane IA. Aerobic fitness and hand grip strength in Type 1 diabetes: relationship to glycaemic control and body composition. Diabet Med. 2007;24:1296–1299. PubMed doi:10.1111/j.1464-5491.2007.02257.x 18. Ministry of Health (MH). 196 Resolution of october 10th 1996, National Health Council, Brazil. Disclosed by: http://conselho.saude. gov.br/resolucoes/reso_96.htm. Date: 14/06/2013. 19. Tanner J. Growth at adolescence. 2nd ed. Oxford, UK: Blackwell; 1962. 20. Bouchard C, Tremblay A, Leblanc C, Lortie G, Savard R, Theriault G. A method to assess energy expenditure in children and adults. Am J Clin Nutr. 1983;37:461–467. PubMed 21. Guedes DP, Guedes JERP, Barbosa DS, Oliveira JA. Levels of regular physical activity in adolescents. Revista Brasileira de Medicina do Esporte. 2001;7(6):187–199. doi:10.1590/S151786922001000600002 22. Franks PW, Ekelund U, Brage S, Wong MY, Wareham NJ. Does the association of habitual physical activity with the metabolic syn-
Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 2
Glycemic Control and Physical Activity 237 drome differ by level of cardiorespiratory fitness? Diabetes Care. 2004;27:1187–1193. PubMed doi:10.2337/diacare.27.5.1187 23. Michaliszyn SF, Faulkner MS. Physical activity and sedentary behavior in adolescents with type 1 diabetes. Res Nurs Health. 2010;33:441– 449. PubMed doi:10.1002/nur.20393 24. Marrone S, Plume JW, Kerr P, et al. The role of free-play physical activity in healthy blood glucose maintenance in children with type 1 diabetes mellitus. Psychol Health Med. 2009;14:48–52. PubMed doi:10.1080/13548500801983066 25. Lehmann R, Kaplan V, Bingisser R, Bloch KE, Spinas GA. Impact of physical activity on cardiovascular risk factors in IDDM. Diabetes Care 1997;20:160–-1611. 26. Schwab KO, Doerfer J, Marg W, Schober E, Holl RW. Characterization of 33 488 children and adolescents with type 1 diabetes based on the gender-specific increase of cardiovascular risk factors. Pediatr Diabetes. 2010;11:357–363. PubMed doi:10.1111/j.13995448.2010.00665.x 27. Bernardini AL, Vanelli M, Chiari G, et al. Adherence to physical activity in young people with type 1 diabetes. Acta Biomed. 2004;75:153–157. PubMed
28. Ligtenberg PC, Blans M, Hoekstra JB, van der Tweel I, Erkelens DW. No effect of long-term physical activity on the glycemic control in type 1 diabetes patients: a cross-sectional study. Neth J Med Netherlands; 1999:59–63. 29. Donga E, van Dijk M, van Dijk JG, et al. Partial sleep restriction decreases insulin sensitivity in type 1 diabetes. Diabetes Care. 2010;33:1573–1577. PubMed doi:10.2337/dc09-2317 30. Happe S, Treptau N, Ziegler R, Harms E. Restless legs syndrome and sleep problems in children and adolescents with insulin-dependent diabetes mellitus type 1. Neuropediatrics. 2005;36:98–103. PubMed doi:10.1055/s-2005-837685 31. Perfect MM, Patel PG, Scott RE, et al. Sleep, glucose, and daytime functioning in youth with type 1 diabetes. Sleep. 2012;35:81–88. PubMed 32. Ce GV, Rohde LE, da Silva AM, Punales MK, de Castro AC, Bertoluci MC. Endothelial dysfunction is related to poor glycemic control in adolescents with type 1 diabetes under 5 years of disease: evidence of metabolic memory. J Clin Endocrinol Metab. 2011;96:1493–1499. PubMed doi:10.1210/jc.2010-2363
Official Journal of ISPAH www.JPAH-Journal.com ORIGINAL RESEARCH
Correlation Between Glycemic Control and Physical Activity Level in Adolescents and Children With Type 1 Diabetes Cristiane Petra Miculis, Wagner De Campos, and Margaret Cristina da Silva Boguszewski Background: The aim of this study was to correlate glycemic control (GC) and variables of physical activity levels (PAL) in children with type 1 diabetes mellitus (T1DM). Methods: Fifty children and adolescents with T1DM were selected. Personal and medical data for the patients were collected. Physical evaluations of body weight and sexual maturation were undertaken. Bouchard’s questionnaire was applied to evaluate PAL as well as for time spent on physical activities. Results: Sixty-four percent of the subjects were sexually mature. Differences were observed between females and males in insulin dose, duration of light physical activity, and sleeping time (P < .05). Ninety percent presented poor GC and 80% had a low PAL. Fasting blood glucose (FBG) was significantly correlated with PAL, with sedentary time, and with sleeping time. Glycated hemoglobin (HbA1c) was significantly correlated with sedentary time and sleeping time. Among the three groups of PAL (insufficient × moderate × active) there were differences in HbA1c (%), FBG (mg/dL), duration of disease (years), and insulin dose (UI/kg/day) (P < 0.001). Conclusion: GC was significantly correlated with PAL. Among the three groups of physical activity level, the most active group was seen to have the best GC. Keywords: special needs population, adolescent, evaluation, exercise, metabolic health, physical activity A fundamental treatment parameter for type 1 diabetes mellitus (T1DM) is adequate glycemic control (GC).1 Poor GC raises the risk of microvascular complications and correlates with risk factors for macroangiopathies.1,2 Adequate GC can be reached through medical treatment (multiple insulin doses) as well as a healthy lifestyle, such as a balanced diet and regular physical activity.3 Regular physical activity is insufficient in the general pediatric population4,5 and in T1DM children and adolescents. T1DM children and adolescents have a low regular physical activity index for the reasons of fear of hypoglycemia, inadequate time available for physical activities, poor GC, and lower physical fitness levels.6–8 The effect of regular physical activity on GC is variable9–14 depending on factors such as frequency of physical activity per week, remaining active for 12 weeks or more,10,15 and reducing sedentary behaviors such as watching television and playing computer games.12,16 These factors demonstrate a significant and positive effect on GC including a lower glycated hemoglobin (HbA1c) level in T1DM children and adolescents, independent of the type of activity (ie, aerobic versus strength and endurance).9–11,15 Studies using lower weekly exercise frequency combined, and not combined, with shorter intervention duration report an insignificant effect on GC.12–14 These studies report other positive outcomes such as a better sense of well-being,12,14 augmented physical fitness,13 reduced insulin dose,13,14 and improved waist circumference.14 We hypothesize that lower physical activity levels (PAL) are associated with poorer GC in T1DM children and adolescents. We explored the relationship among GC, PAL, and time spent on a range of physical activity intensities in children and adolescents with T1DM. Miculis ([email protected]) and da Silva Boguszweski are with the Dept of Pediatrics, Federal University of Paraná, Curitiba, Brazil. De Campos is with the Dept of Physical Education, Federal University of Paraná, Curitiba, Brazil. 232
Methods Fifty children and adolescents with T1DM were studied (n = 21 boys and n = 29 girls). All subjects were 9 to17 years old, living in Curitiba, Brazil or its metropolitan areas, and free of diabetes-related complications (ie, retinopathy, neuropathy, or nephropathy). The sample size (n) was calculated for a significance level of 5% and a power of 80% for mean difference and correlations.17 We used a one-tailed test of the hypothesis and mean difference (difference = 3, SD = 8.5) for this cross-sectional study. Parents or legal guardians of the children were informed of all procedures and the objective of this research, including risks and benefits. Participation was free of any financial compensation, performed according to the criteria of 196/96 National Council Resolution,18 and approved by the local ethics committee of the Hospital de Clinicas, Federal University of Paraná, Brazil, under CEP/SD 666.001.09.01 and CAAE 0104.0.000.091 to 09 on February 20, 2009. Written informed consent was obtained from all participants or their parents. A review of the medical records was performed in search of clinical and laboratory data indicating the fasting blood glucose (FBG) and HbA1c measured the week before the medical appointment at the Diabetes Outpatient Clinic. The hexokinases method was used for FBG while high-performance liquid chromatography (HPLC) was used for HbA1c. All analyses were conducted using Architect c8000 (Abbott, Chicago, IL). Children and adolescents, with the assistance of their parents, replied to a questionnaire seeking personal and medical data such as actual insulin dose, number of self-monitored blood glucose levels, diabetes duration, and physical activity level. The dose of each insulin type was summed and divided by body weight (kg), resulting in daily insulin dose (UI/kg/day). Diabetes duration was recorded in years. A physical evaluation was conducted to determine body weight (kg) with minimal clothing (ie, shorts and a light t-shirt) on an
Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 2
Glycemic Control and Physical Activity 233
electronic scale (TOLEDO, model 2096 PP, 0.1 kg of precision; São Bernardo do Campo, Brazil). Pubertal stages were classified according to the Tanner scale;19 they were self-identified using figures in the privacy of a separate room in the presence of the parents and one evaluator of the same sex as the child. Physical activity level was measured through Bouchard’s physical activity questionnaire,20 which is validated for children as young as 10 years old, also in Brazil.21 This questionnaire collects data on the physical activity of the last 3 days, ending in Saturday or Sunday. The day is divided into 96 periods of 15 minutes. For each 15-minute period, energy expenditure is classified using a scale from 1 to 9, corresponding to the primary activity of that period (see below for description). The median energy expenditure (kcal/kg/15 min) of each category is used to calculate daily energy expenditure. Cut-off points for PAL were based on the sample’s percentile of energy expenditure (kcal/kg/day).22 Girls were classified as: insufficiently active (PAL < 37.06 kcal/kg/day), moderately active (PAL ≥ 40.18, < 44.01 kcal/kg/day), or active (PAL ≥ 44.01 kcal/ kg/day). Boys were classified as insufficiently active (PAL < 36.45 kcal/kg/day), moderately active (PAL ≥ 39.72, < 42.74 kcal/kg/day), or active (PAL ≥ 42.74 kcal/kg/day). Bouchard’s questionnaire20 allows estimation of the time spent, in minutes, on each type of activity. Bouchard’s energy table20 describes the 1 to 9 scale as follows: 1 = Sleeping Time (S), corresponds to 0.26 kcal/kg/15 min; 2 = Sedentary Behavior Time (SB), corresponds to 0.38 kcal/kg/15 min; 3 to 5 = Light Physical Activity Time (LPA), corresponds to 0.57 to 0.84 kcal/kg/15 min; and 6 to 9 = Moderate to Vigorous Physical Activity Time (MVPA), corresponds to 1.2 to 2 kcal/kg/15 min.
Statistics To test for normality, Kolmogorov-Smirnov and Shapiro-Wilk tests were conducted, considering parametric data for P > .05. Only 2 variables reached that criteria (daily insulin dose and sedentary behavior time), therefore we chose to use nonparametric tests. Spearman’s rho (r) bivariate correlation was used to compare metabolic and physical activity variables. For differences between sexes, the Mann-Whitney U test was applied. Friedman’s test was chosen to detect differences among 3 groups of PAL and GC variables.
Friedman’s test is similar to ANOVA, however it does not locate where the differences are among the groups. For these reasons, the Wilcoxon test was conducted, followed by Tukey’s post hoc test. Variability and central tendency measures were used (mean, standard deviation for normally distributed data, and median with minimum-maximum for nonnormally distributed data). SPSS (version 13.0, IBM, Chicago, IL) was used when P < .05.
Results Table 1 shows clinical characteristics for each sex. A higher daily insulin dose was observed among girls (P = .017) as opposed to boys. According to the Tanner classification, 12% were prepubertal, 64% pubertal, and 24% postpubertal. In general, the sample had poor GC: 89.4% had elevated values of FBG (> 210 mg/dL) and 91.5% had HbA1c > 7.5%. In the analyses of PAL and the time spent in a range of activity intensities (Table 2), a significant difference was observed between girls and boys for the time spent in LPA and S among the median of the 3 days and for days 1 and 2 (P < .05). Girls were more active in each of these cases. We found that 28% of the sample was insufficiently active, 52% were considered moderately active, and 20% were active. The majority of the sample (n = 33/50) chose Saturday as the third day for their Bouchard’s questionnaire. When the FBG was used as an indication of GC, a significant relationship was found between PAL (median of the 3 days), SB (day 3), and S (median of the 3 days and day 3) (Figure 1a–e). Using HbA1c as an indication of GC, a relationship was found between SB (day 1) and S (day 3) (P < .05) (Figure 1f–g). The duration of T1DM was directly correlated with daily insulin dose (P = .025, r = .318) and SB on day 3 (P = .031, r = .306). Similarly, daily insulin dose was correlated with S and day 1 (P = .006, r = .383), day 2 (P = .012, r = .351), and the median of the 3 days (P = .017, r = .336). The number of self-monitored blood glucose levels was significantly related to total PAL on day 2 (P = .019, r = –.334), LPA on day 2 (P = .016, r = –.340), SB on day 1 (P = .032, r = .304), and the median of the 3 days (P = .026, r = .314). The sample was divided into 3 groups according to PAL percentiles in both sexes, resulting in significant differences of GC between the groups (P < .0001) (Table 3).
Table 1 General Characteristics of the Sample Girls (n = 29)
Boys (n = 21)
Mean
Standard Deviation
Mean
Age (years)
11.86
1.57
12.67
2.29
Weight (kg)
43.68
9.97
43.46
12.15
T1DMD (years)
4.60
3.23
4.60
3.21
Capillary glycemia (times/day)
3.55
1.02
3.86
.57
Insulin dose (UI/kg/day)
1.06*
.34
.83
.28
FBG (ml/dL)
233.44
93.42
287.6
150.52
10.5
2.72
10.42
2.29
HbA1c (%)
Note. T1DMD = type 1 diabetes mellitus duration; FBG = fasting blood glucose. *P < .05, female different from male.
Standard Deviation
Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 2
Figure 1 — Correlation between glycemic control variables, physical activity level, and time spent on different activities intensities. (a) FBG vs PAL day 3 (p = .021; r = –.335); (b) FBG vs PAL mean (p = .027; r = .324); (c) FBG vs sedentary behavior day 3 (p = .022; r = –.334); (d) FBG vs sedentary behavior median (p = .045; r = –.294); (e) FBG vs sleeping time day 3 (p = .044; r = .295); (f) HbA1c vs sedentary behavior day 1 (p = .046; r = .292); (g) HbA1c vs sleeping time day 3 (p = .018; r = .343). FBG = fasting blood glucose; PAL = physical activity level. 234
Glycemic Control and Physical Activity 235
Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 2
Table 2 Physical Activity Level and Time Spent on a Range of Activity Intensities
PAL 1 (kcal/kg/day) PAL 2 (kcal/kg/day) PAL 3 (kcal/kg/day) PAL mean (kcal/kg/day) SB 1 (min) SB 2 (min) SB 3 (min) SB mean (min) LPA 1(min) LPA 2 (min) LPA 3 (min) LPA mean (min) MVPA 1 (min) MVPA 2 (min) MVPA 3 (min) MVPA mean (min) S 1 (min) S 2 (min) S 3 (min) S mean (min)
Girls (n = 29) Median Min–Max 39.06 32.29–55.14 38.80 31.62–57.14 42.04 33.44–56.58 40.18 33.58–55 615 405–825 615 330–810 480 255–780 553.33 370–755 165 0–375 195* 0–420 195 0–540 175 0–425 60 0–285 45 0–390 150 0–435 100 0–345 540* 135–720 540* 435–705 600 480–750 575* 490–705
Boys (n = 21) Median Min–Max 37.94 32.08–49.35 38.75 33.78–58.88 40.3 31.94–54.37 39.72 35.15–53.92 645 450–855 675 390–885 510 180–810 625 345–835 135 15–450 120 30–450 165 45–525 140 60–445 75 0–315 90 0–360 90 0–330 95 10–290 480 405–690 480 405–690 615 420–720 530 470–700
Note. PAL = physical activity level; SB = sedentary behavior; LPA = light physical activity; MVPA = moderate to vigorous physical activity; S = sleeping time; numbers 1, 2, and 3 represent different days in physical activity questionnaire. *P < .05, female different from male.
Table 3 Glycemic Control by Physical Activity Level Groups Mean ± Standard Deviation HbA1c (%) FBG (mg/dL) T1DMD (years) Insulin (UI/kg/day)
Insufficiently Active 10.78 ± 2.46 277 ± 157.86 5.16 ± 4.21 1.10 ± 0.23
Moderately Active 12.12 ± 2.94 257.20 ± 59.57 5.30 ± 2.90 1.42 ± 0.36
Active 9.56 ± 2.31* 240.20 ± 162.49** 2.25 ± 1.91* 0.89 ± 0.38*
Note. T1DMD = type 1 diabetes mellitus duration; FBG = fasting blood glucose. *Active group significantly different from moderately and insufficiently active groups, two-tailed Wilcoxon’s test P < .0001; **active group significantly different from insufficiently active groups, two-tailed Wilcoxon’s test P < .0001, Tukey’s post hoc test.
Discussion Our findings demonstrated a significant association between FBG and PAL, and sedentary and sleeping activities. Likewise, we found an association between HbA1c and sedentary and sleeping activities. Studies in patients with T1DM demonstrate that time spent on certain types of sedentary behaviors (eg, watching television, playing computer games) positively correlates to FBG and HbA1c levels, indicating that increased time on those kinds of activities may reduce GC.9,16 Some authors do not explore a correlation between PAL and GC, however they demonstrate the exercise-induced effect on GC.9–11,15 HbA1c could be reduced by 0.96% compared with basal values in T1DM adolescents after 12 weeks of endurance and aerobic training.15 Shorter or longer training duration interventions also demonstrate a significant and long-lasting effect on HbA1c and glycemia levels.10,11,23,24 Herbst et al6,9 shows that the
most significant factor for GC is regular physical activity; in the same way, moderate or moderate to vigorous physical activities are related to HbA1c after 16 weeks of physical training for adolescents with T1DM.23 HbA1c glycemia significantly correlates to physical activity (minutes per week) with a greater magnitude than other variables of physical fitness.25 Among our sample, most of the T1DM subjects were at puberty (25 girls and 14 boys) with poor GC. This is not uncommon among adolescents with T1DM.3,26 Also, the median insulin dose was higher for girls as compared with boys, however the reasons for this poor GC is multifactorial, influenced by maturation status, and is beyond the scope of our study. Surprisingly, the difference in median 3-day PAL did not reach statistical significance when comparing girls to boys. Most studies6,9,12,27 show higher PAL for males. One study28 found a higher regular physical activity index among females versus males in adults with T1DM. The difference could be assigned to leisure-
Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 2
236 Miculis, De Campos, and da Silva Boguszewski
time physical activities (eg, domestic tasks, personal care, and light physical activity). Among our sample, we found that girls spent more time on LPA compared with boys, suggesting that the lack of difference in PAL could be assigned to the difference in LPA. This is in accordance with recent data23 from adolescents with T1DM. Sleeping time on the majority of days was higher for girls and correlated positively with FBG, HbA1c, and daily insulin dose. This is a novel fact, because most studies explore the role of hypoglycemia associated with sleep deprivation and its pathological alterations29–31 instead of the effect of sleeping duration on the hyperglycemia. Our findings suggested that sleep duration on weekend days for adolescents may be associated with poor GC. One could speculate if this association occurred because the subjects lost their diabetes management routine on the weekends and slept more than on weekdays, thus exceeding the time for self-monitoring and insulin dosing. However, the cross-sectional design of our study does not permit us to deduce causality. As in the general population, 28% of our sample was insufficiently active, 52% moderately active, and 20% active.4,5 We used percentile division of the sample data because there is no Brazilian reference for the PAL of children and adolescents with T1DM. We are concerned that children with T1DM do not engage in an adequate amount of physical activity as recommended for the prevention of diabetes-related diseases, since they are at higher risk than their healthy peers.26,32 The most active group of our study showed the lowest mean values of HbA1c, FBG, diabetes duration, and daily insulin dose when compared with the moderately active and insufficiently active subgroups. In a sample of 23,000 T1DM patients, Herbst et al6 found differences between the number of days of regular physical activity and GC variables; the most active were associated with better HbA1c levels. The higher one’s PAL, the better their GC, depending on the frequency and regularity of the physical activity.9,10,15 In conclusion, our findings show that GC is associated with physical activities and sleeping time and reinforce that the regular practice of physical activity in children and adolescents with T1DM is highly recommended, whether it is aerobic or endurance exercise.
References 1. Diabetes Control and Complication Trial (DCCT). The effect of intensive treatment of diabetes on the development and progression of longterm complications in insulin-dependent diabetes mellitus. The Diabetes Control and Complications Trial Research Group. N Engl J Med. 1993;329:977–986. PubMed doi:10.1056/NEJM199309303291401 2. Nathan DM, Cleary PA, Backlund JY, et al. Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. N Engl J Med. 2005;353:2643–2653. PubMed doi:10.1056/ NEJMoa052187 3. Teles SA, Fornés NS. Food consumption and metabolic control in children and adolescents with type 1 diabetes mellitus. Rev Paulista Pediatr. 2011;29(3):378–384. doi:10.1590/S0103-05822011000300012 4. Ribeiro RQ, Lotufo PA, Lamounier JA, Oliveira RG, Soares JF, Botter DA. Additional cardiovascular risk factors associated with excess weight in children and adolescents: the Belo Horizonte heart study. Arq Bras Cardiol. 2006;86:408–418. PubMed doi:10.1590/S0066782X2006000600002 5. Fakhouri THI, Hughes JP, Brody DJ, Kit BK, Ogden CL. Physical activity and screen-time viewing among elementary school–aged children in the united states from 2009 to 2010. JAMA Pediatr. 2013;167(3):223–229. PubMed doi:10.1001/2013.jamapediatrics.122
6. Herbst A, Kordonouri O, Schwab KO, Schmidt F, Holl RW. Impact of physical activity on cardiovascular risk factors in children with type 1 diabetes: a multicenter study of 23,251 patients. Diabetes Care. 2007;30:2098–2100. PubMed doi:10.2337/dc06-2636 7. Valerio G, Spagnuolo MI, Lombardi F, Spadaro R, Siano M, Franzese A. Physical activity and sports participation in children and adolescents with type 1 diabetes mellitus. Nutr Metab Cardiovasc Dis. 2007;17:376–382. PubMed doi:10.1016/j.numecd.2005.10.012 8. Brazeau AS, Rabasa-Lhoret R, Strychar I, Mircescu H. Barriers to physical activity among patients with type 1 diabetes. Diabetes Care. 2008;31:2108–2109. PubMed doi:10.2337/dc08-0720 9. Herbst A, Bachran R, Kapellen T, Holl RW. Effects of regular physical activity on control of glycemia in pediatric patients with type 1 diabetes mellitus. Arch Pediatr Adolesc Med. 2006;160:573–577. PubMed doi:10.1001/archpedi.160.6.573 10. Sideraviciute S, Gailiuniene A, Visagurskiene K, Vizbaraite D. The effect of long-term swimming program on glycemia control in 14-19-year aged healthy girls and girls with type 1 diabetes mellitus. Medicina (Kaunas). 2006;42:513–518. PubMed 11. Wang YC, Stewart S, Tuli E, White P. Improved glycemic control in adolescents with type 1 diabetes mellitus who attend diabetes camp. Pediatr Diabetes Denmark; 2008:29–34. 12. Aman J, Skinner TC, de Beaufort CE, Swift PG, Aanstoot HJ, Cameron F. Associations between physical activity, sedentary behavior, and glycemic control in a large cohort of adolescents with type 1 diabetes: the Hvidoere Study Group on Childhood Diabetes. Pediatr Diabetes. 2009;10:234–239. PubMed doi:10.1111/j.1399-5448.2008.00495.x 13. D’hooge R, Hellinckx T, Van Laethem C, et al. Influence of combined aerobic and resistance training on metabolic control, cardiovascular fitness and quality of life in adolescents with type 1 diabetes: a randomized controlled trial. Clin Rehabil. 2011;25:349–359. PubMed doi:10.1177/0269215510386254 14. Ramalho AC, de Lourdes Lima M, Nunes F, et al. The effect of resistance versus aerobic training on metabolic control in patients with type-1 diabetes mellitus. Diabetes Res Clin Pract. 2006;72:271–276. PubMed doi:10.1016/j.diabres.2005.11.011 15. Mosher PE, Nash MS, Perry AC, LaPerriere AR, Goldberg RB. Aerobic circuit exercise training: effect on adolescents with well-controlled insulin-dependent diabetes mellitus. Arch Phys Med Rehabil. 1998;79:652–657. PubMed doi:10.1016/S0003-9993(98)90039-9 16. Margeirsdottir HD, Larsen JR, Brunborg C, Sandvik L, Dahl-Jorgensen K. Strong association between time watching television and blood glucose control in children and adolescents with type 1 diabetes. Diabetes Care. 2007;30:1567–1570. PubMed doi:10.2337/dc06-2112 17. Wallymahmed ME, Morgan C, Gill GV, MacFarlane IA. Aerobic fitness and hand grip strength in Type 1 diabetes: relationship to glycaemic control and body composition. Diabet Med. 2007;24:1296–1299. PubMed doi:10.1111/j.1464-5491.2007.02257.x 18. Ministry of Health (MH). 196 Resolution of october 10th 1996, National Health Council, Brazil. Disclosed by: http://conselho.saude. gov.br/resolucoes/reso_96.htm. Date: 14/06/2013. 19. Tanner J. Growth at adolescence. 2nd ed. Oxford, UK: Blackwell; 1962. 20. Bouchard C, Tremblay A, Leblanc C, Lortie G, Savard R, Theriault G. A method to assess energy expenditure in children and adults. Am J Clin Nutr. 1983;37:461–467. PubMed 21. Guedes DP, Guedes JERP, Barbosa DS, Oliveira JA. Levels of regular physical activity in adolescents. Revista Brasileira de Medicina do Esporte. 2001;7(6):187–199. doi:10.1590/S151786922001000600002 22. Franks PW, Ekelund U, Brage S, Wong MY, Wareham NJ. Does the association of habitual physical activity with the metabolic syn-
Downloaded by University of South Carolina on 09/16/16, Volume 12, Article Number 2
Glycemic Control and Physical Activity 237 drome differ by level of cardiorespiratory fitness? Diabetes Care. 2004;27:1187–1193. PubMed doi:10.2337/diacare.27.5.1187 23. Michaliszyn SF, Faulkner MS. Physical activity and sedentary behavior in adolescents with type 1 diabetes. Res Nurs Health. 2010;33:441– 449. PubMed doi:10.1002/nur.20393 24. Marrone S, Plume JW, Kerr P, et al. The role of free-play physical activity in healthy blood glucose maintenance in children with type 1 diabetes mellitus. Psychol Health Med. 2009;14:48–52. PubMed doi:10.1080/13548500801983066 25. Lehmann R, Kaplan V, Bingisser R, Bloch KE, Spinas GA. Impact of physical activity on cardiovascular risk factors in IDDM. Diabetes Care 1997;20:160–-1611. 26. Schwab KO, Doerfer J, Marg W, Schober E, Holl RW. Characterization of 33 488 children and adolescents with type 1 diabetes based on the gender-specific increase of cardiovascular risk factors. Pediatr Diabetes. 2010;11:357–363. PubMed doi:10.1111/j.13995448.2010.00665.x 27. Bernardini AL, Vanelli M, Chiari G, et al. Adherence to physical activity in young people with type 1 diabetes. Acta Biomed. 2004;75:153–157. PubMed
28. Ligtenberg PC, Blans M, Hoekstra JB, van der Tweel I, Erkelens DW. No effect of long-term physical activity on the glycemic control in type 1 diabetes patients: a cross-sectional study. Neth J Med Netherlands; 1999:59–63. 29. Donga E, van Dijk M, van Dijk JG, et al. Partial sleep restriction decreases insulin sensitivity in type 1 diabetes. Diabetes Care. 2010;33:1573–1577. PubMed doi:10.2337/dc09-2317 30. Happe S, Treptau N, Ziegler R, Harms E. Restless legs syndrome and sleep problems in children and adolescents with insulin-dependent diabetes mellitus type 1. Neuropediatrics. 2005;36:98–103. PubMed doi:10.1055/s-2005-837685 31. Perfect MM, Patel PG, Scott RE, et al. Sleep, glucose, and daytime functioning in youth with type 1 diabetes. Sleep. 2012;35:81–88. PubMed 32. Ce GV, Rohde LE, da Silva AM, Punales MK, de Castro AC, Bertoluci MC. Endothelial dysfunction is related to poor glycemic control in adolescents with type 1 diabetes under 5 years of disease: evidence of metabolic memory. J Clin Endocrinol Metab. 2011;96:1493–1499. PubMed doi:10.1210/jc.2010-2363
