OBJECTIVE: Early prevention strategies are needed to mitigate the high risk of cardiovascular disease in adolescents with type 1 diabetes (T1D). Residential neighbourhood features can promote healthy lifestyle behaviours and reduce cardiovascular risk, but less is known about their role in lifestyle behaviours in adolescents with T1D, and no studies used comparisons to healthy controls.
METHODS: We examined associations between residential neighbourhood features and lifestyle behaviours in adolescents with T1D and healthy controls. Data were analyzed from the CARdiovascular Disease risk factors in pEdiatric type 1 diAbetes (CARDEA) study, a cross-sectional investigation of 100 adolescents with T1D (14 to 18 years) from a pediatric diabetes clinic in Montréal, Canada, and 97 healthy controls. Outcomes included physical activity and sedentary behaviour (accelerometry), screen time and sleep duration (questionnaires), and dietary habits (24-hour recalls). Cluster analysis of selected neighbourhood indicators computed for participants\' postal codes resulted in 2 neighbourhood types: central urban and peri-urban. Central urban neighbourhoods were characterized by very high population density, high active living index, numerous points of interest, higher social deprivation, higher residential mobility, and lower median household income compared with peri-urban neighbourhoods. Associations of neighbourhood type with lifestyle behaviours were estimated with multiple linear regressions and interactions by T1D status were tested.
RESULTS: Living in central urban neighbourhoods was associated with greater daily minutes of moderate-to-vigourous physical activity (beta = 8.61, 95% confidence interval 1.79 to 15.44) compared with living in peri-urban neighbourhoods. No associations were observed for other lifestyle behaviours, and no statistically significant interactions were found between neighbourhood type and T1D status.
CONCLUSIONS: Features that characterize central urban built environments appear to promote physical activity in adolescents, regardless of T1D status.

OBJECTIVE: Diabetic ketoacidosis (DKA) occurring after diabetes diagnosis is often associated with risk factors for other diabetes-related complications. In this study we aimed to determine the prognostic implications of DKA on all-cause mortality and complications in type 1 diabetes (T1D).
METHODS: Previously collected data from the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) study were obtained through the the National Institute of Diabetes and Digestive and Kidney Diseases Central Repository. Using Cox proportional hazards models with time-dependent covariates, we examined age- and sex-adjusted, glycated hemoglobin-adjusted, and fully adjusted associations of DKA with all-cause mortality, cardiovascular disease, microvascular, and acute complications over 34 years.
RESULTS: Of the 1,441 study participants, 297 had 488 DKA events. Prior DKA was associated with a higher risk of age- and sex-adjusted all-cause mortality (hazard ratio [HR] 8.28, 95% confidence interval [CI] 3.74 to 18.32, p<0.001), major adverse cardiovascular events (MACEs) (HR 2.05, 95% CI 1.34 to 3.13, p<0.001), and all advanced microvascular and acute complications compared with no prior DKA. Most associations except retinopathy were significant even after adjustment for covariates. In our fully adjusted analysis, prior DKA was associated with a significantly higher risk of subsequent all-cause mortality (HR 9.13, 95% CI 3.87 to 21.50, p<0.001), MACEs (HR 1.66, 95% CI 1.07 to 2.59, p=0.03), advanced kidney disease (HR 2.10, 95% CI 1.00 to 4.22, p=0.049), advanced neuropathy (HR 1.49, 95% CI 1.05 to 2.13, p=0.03), severe hypoglycemia (HR 1.53, 95% CI 1.28 to 1.81, p<0.001), and recurrent DKA (HR 3.24, 95% CI 2.41 to 4.36, p<0.001) compared with person-time without DKA.
CONCLUSIONS: DKA is a prognostic marker for diabetes complications, including excess all-cause mortality. Intensified clinical interventions, such as cardiovascular prevention strategies, may be warranted after diagnosis of DKA.

OBJECTIVE: Evidence suggests that glucose levels in menstruating females with type 1 diabetes change throughout the menstrual cycle, reaching a peak during the luteal phase. The Type 1 Diabetes Exercise Initiative (T1DEXI) study provided the opportunity to assess glycemic metrics between early and late phases of the menstrual cycle, and whether differences could be explained by exercise, insulin, and carbohydrate intake.
METHODS: One hundred seventy-nine women were included in our analysis. Glycemic metrics, carbohydrate intake, insulin requirements, and exercise habits during the early vs late phases of their menstrual cycles (i.e. 2 to 4 days after vs 2 to 4 days before reported menstruation start date) were compared.
RESULTS: Mean glucose increased from 8.2±1.5 mmol/L (148±27 mg/dL) during the early follicular phase to 8.6±1.6 mmol/L (155±29 mg/dL) during the late luteal phase (p<0.001). Mean percent time-in-range (3.9 to 10.0 mmol/L [70 to 180 mg/dL]) decreased from 73±17% to 70±18% (p=0.002), and median percent time >10.0 mmol/L (>180 mg/dL) increased from 21% to 23% (p<0.001). Median total daily insulin requirements increased from 37.4 units during the early follicular phase to 38.5 units during the late luteal phase (p=0.02) and mean daily carbohydrate consumption increased slightly from 127±47 g to 133±47 g (p=0.05); however, the difference in mean glucose during early follicular vs late luteal phase was not explained by differences in exercise duration, total daily insulin units, or reported carbohydrate intake.
CONCLUSIONS: Glucose levels during the late luteal phase were higher than those of the early follicular phase of the menstrual cycle. These glycemic changes suggest that glucose management for women with type 1 diabetes may need to be fine-tuned within the context of their menstrual cycles.

OBJECTIVE: Exercise is a recommended component of type 1 diabetes (T1D) treatment because high physical activity levels improve health outcomes. However, many people with T1D do not meet physical activity recommendations. Our aim in this study was to identify factors influencing physical activity levels in people with T1D.
METHODS: This questionnaire-based study included adults with T1D from 1 outpatient clinic in the United Kingdom and 2 clinics in Denmark. Exercise characteristics, motivators, and barriers were assessed. Physical activity level was measured using the Saltin-Grimby Physical Activity Level Scale. Respondents were categorized into 3 activity groups: inactive, light active, and moderate-to-vigourous active.
RESULTS: Of the 332 respondents, 8.4% rated themselves as inactive, 48% as light active, and 43% as moderate-to-vigourous active. Seventy-eight percent of inactive and light active repondents expressed a desire to become more physically active. Fifty-three percent of respondents had received guidance concerning exercise/physical activity from their diabetes team. Being male and having received guidance were associated with a higher physical activity level. The major motivators for exercising/being physically active were improved mental and physical health and glycemic control, whereas the most frequent barriers were busyness with work/private life and lack of motivation. Worries about glucose excursions, costs, lack of knowledge, and health-related reasons were more prevalent barriers in the least active groups.
CONCLUSIONS: This study demonstrated that 78% of inactive and light active respondents reported wishing to become more physically active. Receiving guidance about exercise/physical activity was associated with a higher physical activity level, but only 53% of respondents had received support from their diabetes team.

OBJECTIVE: In this study we explore the impact of postprandial exercise timing (morning vs evening) on glycemia in individuals with type 1 diabetes (T1D) during short all-out sprints on a cycle ergometer.
METHODS: Ten healthy physically sedentary male (n=7) and female (n=3) volunteers with type 1 diabetes, 22.8±2.8 years of age, and with a diabetes duration of 9.7±5.5 years and glycated hemoglobin level of 8.6±1.2%, underwent comprehensive screening and assessment of their physical health and fitness status before study participation, under the guidance of a physician. Each participant underwent 2 postprandial exercise sessions on separate days: the first in the morning at 8:00 AM and second in the evening at 8:00 PM, both conducted 60 minutes after a standardized meal.
RESULTS: Morning exercise showed a less pronounced reduction in plasma glucose (PG) levels compared with evening exercise (-2.01±1.24 vs -3.56±1.6 mmol/L, p=0.03). In addition, higher cortisol levels were observed in the morning vs evening (128.59±34 vs 67.79±26 ng/mL, p<0.001).
CONCLUSIONS: Morning repeated sprint exercise conducted in the postprandial state consistent with the protective effect of higher cortisol levels resulted in a smaller reduction in PG levels compared with evening exercise. This highlights the potential influence of exercise timing on glycemic responses and cortisol secretion in the management of T1D.

OBJECTIVE: Current exercise recommendations for people with type 1 diabetes (T1D) are based on research involving primarily young, fit male participants. Recent studies have shown possible differences between male and female blood glucose response to exercise, but little is known about whether these differences are sex-related (due to physiological differences between male and female participants) or gender-related (behavioural differences between men and women).
METHODS: To better understand gender-based behavioural differences surrounding physical activity (PA), we asked men and women (n=10 each) with T1D to participate in semistructured interviews. Topics discussed included motivation and barriers to exercise, diabetes management strategies, and PA preferences (type, frequency, duration of exercise, etc). Interview transcripts were coded by 2 analysts before being grouped into themes.
RESULTS: Six themes were identified impacting participants\' PA experience: motivation, fear of hypoglycemia, time lost to T1D management, medical support for PA, the role of technology in PA accessibility, and desire for more community. Gender differences were found in motivations, medical support, and desire for more community. Women were more motivated by directional weight dissatisfaction, and men were more motivated to stay in shape. Men felt less supported by their health-care providers than women. Women more often preferred to exercise in groups, and sought more community surrounding T1D and PA.
CONCLUSIONS: Although men and women with T1D experience similar barriers around PA, there are differences in motivation, desire for community, and perceived support from medical providers.

OBJECTIVE: Expert guidelines recommend an aerobic cooldown to lower blood glucose for the management of post-exercise hyperglycemia. This strategy has never been empirically tested. Our aim in this study was to compare the glycemic effects of performing an aerobic cooldown vs not performing a cooldown after a fasted resistance exercise session. We hypothesized that the cooldown would lower blood glucose in the 30 minutes after exercise and would result in less time in hyperglycemia in the 6 hours after exercise.
METHODS: Participants completed 2 identical resistance exercise sessions. One was followed by a low-intensity (30% of peak oxygen consumption) 10-minute cycle ergometer cooldown, and the other was followed by 10 minutes of sitting. We compared the changes in capillary glucose concentration during these sessions and continuous glucose monitoring (CGM) outcomes over 24 hours post-exercise.
RESULTS: Sixteen participants completed the trial. Capillary glucose was similar between conditions at the start of exercise (p=0.07). Capillary glucose concentration decreased by 0.6±1.0 mmol/L during the 10-minute cooldown, but it increased by 0.7±1.3 mmol/L during the same time in the no-cooldown condition. The resulting difference in glucose trajectory led to a significant interaction (p=0.02), with no effect from treatment (p=0.7). Capillary glucose values at the end of recovery were similar between conditions (p>0.05). There were no significant differences in CGM outcomes.
CONCLUSIONS: An aerobic cooldown reduces glucose concentration in the post-exercise period, but the small and brief nature of this reduction makes this strategy unlikely to be an effective treatment for hyperglycemia occurring after fasted exercise.

OBJECTIVE: Our aim in this study was to determine the correlation between serum fructosamine and average blood glucose, as measured by continuous glucose monitoring (CGM) in children with type 1 diabetes.
METHODS: Ninety-seven blood samples were collected from 70 participants in the Timing of Initiation of continuous glucose Monitoring in Established pediatric diabetes (CGM TIME) Trial. Each eligible participant had 3 weeks of CGM data with at least 60% CGM adherence before blood collection. Ordinary least-squares linear regression incorporating restricted cubic splines was used to determine the association between fructosamine levels and mean blood glucose.
RESULTS: An association was found between fructosamine and mean blood glucose, with an F statistic of 9.543 (p<0.001). Data were used to create a formula and conversion chart for calculating mean blood glucose from fructosamine levels for clinical use.
CONCLUSIONS: There is a complex relationship between average blood glucose, as determined by CGM and fructosamine. Fructosamine levels may be clinically useful for assessing short-term glycemic management when CGM is not available.

OBJECTIVE: Diabetes distress (DD) has been understudied in the pregnancy population. Pregnancy is known to be a complex, highly stressful time for women with diabetes because of medical risks and the high burden of diabetes management. Our aim in this study was to explain and understand DD in women with pre-existing diabetes in pregnancy.
METHODS: An explanatory, sequential mixed-methods study was undertaken. The first strand consisted of a cross-sectional study of 76 women with type 1 and type 2 diabetes. A nested sampling approach was used to re-recruit 18 women back into the second strand for qualitative interviews using an interpretive description approach.
RESULTS: DD was measured by the validated Problem Area in Diabetes (PAID) scale. A PAID score of ≥40 was positive for distress. DD prevalence was 22.4% in the cross-sectional cohort and the average PAID score was 27.75 (standard deviation 16.08). In the qualitative strand, women with a range of PAID scores (10.0 to 60.0) were sampled for interviews. The majority of these participants described themes of DD in their interviews. Of the 15 women who described DD thematically, only 6 had positive PAID scores.
CONCLUSIONS: Integration of the mixed-methods data underscores important meta-inferences about DD in pregnancy, namely that DD was present to a greater degree than the PAID tool is sensitive to. DD was present qualitatively in most of the qualitative sample, despite interviewing women with a range of PAID scores. Future research on a pregnancy-specific DD scale is needed.

OBJECTIVE: Glycoprotein acetyls (GlycA\'s) are biomarkers of systemic inflammation and cardiovascular disease, yet little is known about their role in type 1 diabetes (T1D). In this study we examined the associations among GlycA\'s, central adiposity, insulin resistance, and early kidney injury in youth with T1D.
METHODS: Glomerular filtration rate and renal plasma flow by iohexol and p-aminohippurate clearance, urine albumin-to-creatinine ratio (UACR), central adiposity by dual-energy x-ray absorptiometry, and estimated insulin sensitivity were assessed in 50 youth with T1D (16±3.0 years of age, 50% female, glycated hemoglobin 8.7%±1.3%, T1D duration 5.7±2.6 years). Concentrations of GlycA were quantified by targeted nuclear magnetic resonance spectroscopy. Correlation and multivariable linear regression analyses were performed.
RESULTS: GlycA\'s were higher in girls vs boys (1.05±0.26 vs 0.84±0.15 mmol/L, p=0.001) and in participants living with overweight/obesity vs normal weight (1.12±0.23 vs 0.87±0.20 mmol/L, p=0.0004). GlycA\'s correlated positively with estimated intraglomerular pressure (r=0.52, p=0.001), UACR (r=0.53, p<0.0001), and trunk mass (r=0.45, p=0.001), and inversely with estimated insulin sensitivity (r=-0.36, p=0.01). All relationships remained significant after adjustment for age, sex, and glycated hemoglobin.
CONCLUSIONS: As biomarkers of inflammation, GlycA\'s were higher in girls and those with overweight or obese body habitus in T1D. GlycA\'s associated with parameters of early kidney dysfunction, central adiposity, and insulin resistance.