Determinants of Bone Mineral Density in Transgender Youth.

Purpose: We aimed to study determinants of bone health in transgender youth in anticipation of or shortly after initiating puberty suppression and/or gender-affirming hormone therapy. Methods: This was a retrospective review of records of transgender adolescents in our institution between June 2014 and June 2019. Dual energy X-ray absorptiometry was used to assess bone mineral density (BMD). Baseline characteristics were collected and included in a multilinear regression model to assess determinants of lumbar spine (LS) BMD Z-scores adjusted for age and height and accounting for race. Welch's t-test was used to compare characteristics across genders. Results: One hundred nineteen patient records were analyzed. Forty-six patients (38.7%) were assigned male at birth (AMAB) and 73 patients (61.3%) were assigned female at birth (AFAB). Mean (±standard deviation [SD]) age (years) was 14.7±2.6 for AMAB and 15.0±2.2 for AFAB. The adjusted LS BMD Z-score was lower in the AMAB population with a mean (+SD) of -0.605±1.42 compared with 0.043±1.09 in AFAB (p=0.010). In a multivariate model, AMAB gender, vitamin D deficiency, and lower body mass index (BMI) z-scores were determinants of lower LS BMD Z-scores (R 2=0.206). Age, race, ethnicity, insurance status, and Tanner stage were not determinants of BMD. However, post hoc analysis did show that pubertal status modified the results. Conclusion: AMAB transgender adolescents have lower BMD compared with AFAB patients, before or shortly after starting puberty suppression and/or gender-affirming hormone therapy. Lower BMI and vitamin D deficiency were determinants of lower BMD. Further studies are needed to explore etiology for bone health discrepancy in this population.

Assessment of Microvascular Function in Children and Adolescents with Diabetes and Obesity.

BACKGROUND: Endothelial dysfunction (ED) is a marker of vascular damage. Glycated hemoglobin (A1C) predicts vascular complications. The EndoPAT (peripheral arterial tonometry) device calculates the reactive hyperemic index (RHI), a measure of endothelial function. The greater the vasodilation, the higher the RHI. We hypothesized that children with poorly-controlled diabetes mellitus (DM) and non-diabetes mellitus (NDM) obese children have ED. METHODS: A cross-sectional study using the EndoPAT device was performed on children with poorly-controlled DM and NDM children. ANOVA, t-test, Mann-Whitney U test, multiple linear regression and Spearman correlation were used. RESULTS: Of 58 children that completed the study (aged 13.1 ± 3.42 years), 33 with type 1 diabetes (T1DM), 8 with type 2 diabetes (T2DM) and 17 were NDM obese children. Eighty-five percent were African-American, 60% were female and 79% entered puberty. The RHI of children with DM (1.42 ± 0.48) versus NDM obese group (1.40 ± 0.34) was not different (P = 0.86) regardless of the type of DM or body mass index. In the DM group, for every 1% increase in latest A1C, the RHI decreased by 0.097 (P = 0.01) after adjusting for age, gender, and type of DM. The RHI of DM patients with latest A1C of < 10% (1.70 ± 0.58) versus those with A1C ≥10% (1.21 ± 0.19) was statistically different (P = 0.02). In the total study population, males had significantly lower RHI (1.28 ± 0.36) when compared to females (1.51 ± 0.46), P = 0.04 but this difference disappeared when considering pubertal status and type of diabetes. CONCLUSIONS: Our data showed that patients with poorly-controlled DM as reflected by latest A1C of ≥ 10% had worse endothelial function as reflected by lower RHI score.

Role of Social Factors in Glycemic Control Among African American Children and Adolescents with Type 1 Diabetes.

OBJECTIVE: With the rising incidence of Type 1 diabetes (T1DM), it is important to recognize deficiencies in care and areas of improvement to provide better access to resources and education for T1DM patients. The objective of this study was to recognize social factors and compliance barriers affecting glycated hemoglobin (A1c) level in T1D patients among the minority population. METHODS: A total of 84 T1DM patients, ages 3 to 21 years, 49% males, 87% African American participated in the study. Study questionnaires assessing patient knowledge and other variables were distributed and patient charts were reviewed retrospectively to obtain relevant clinical data. T-tests, one-way ANOVA and spearman correlation were used for analysis. RESULTS: Mean A1c in our study was 10.5% and mean knowledge score was 10.1 out of 14. There was no significant correlation (r = 0.12, p = 0.26) between A1c and patients' knowledge scores. Patients with more frequent blood sugar (BS) monitoring (3-4 times/day) had 2 points lower A1c (9.6 vs 11.6 %, 95% CI 0.2-3.7, p = 0.03) than those with 2 or less times/day. No significant difference in A1c between 3-4 checks/day vs >4 checks/day BS checks. Most patients reported 'forgetfulness' (19%) followed by 'too time consuming' (17.9%) as barriers to daily BS monitoring. There was no significant difference in A1c between pen or pump users (10.5 vs 10.2 %, p = 0.55). Surprisingly, those with home supervision had higher A1c than those without (10.7 vs 9.6 %, p = 0.04) while there was no significant difference between those with or without nurse supervision at school (10.6 vs 9.8 %, p = 0.33). Those reporting happy mood interestingly had higher A1c than those with sad/depressed mood (10.7 vs 9.4 %, p = 0.04). On multiple linear regression analysis, frequency of BS checks, home supervision and mood were the most significant predictors of A1c and altogether explained 20% of the variability in A1c. CONCLUSION: Frequent BS monitoring is associated with lower A1c. Supervision at home and school did not improve A1c, but it was self-reported information. Mood did not affect A1c contrary to that reported in other studies.

Screening for autoimmune thyroiditis and celiac disease in minority children with type 1 diabetes.

BACKGROUND: Hashimoto's thyroiditis (HT) and celiac disease (CD) are commonly associated with type 1 diabetes (T1DM). There is no consensus on screening, however, the American Diabetes Association (ADA) and the International Society for Pediatric and Adolescent Diabetes (ISPAD) recommend testing for thyroid function (TFT), thyroid antibodies and anti-tissue transglutaminase antibodies (TTG) IgA soon after diagnosis. TFT should be repeated every 1-2 years while TTG IgA should be tested for within 2 and 5 years. We hypothesize that the rate of HT and CD in our T1DM children is lower, so screening may need to be revised to reflect their underlying risk. METHODS: An Institutional Review Board (IRB)-approved retrospective chart review was conducted on children with T1DM in the past 10 years. Age, sex, race, A1C, TFT, thyroid and celiac antibodies were obtained. t-Tests, the Wilcoxon-Mann-Whitney test and stepwise regression were performed. RESULTS: Of 222 children with T1DM, with a mean age of 15.8±5.53 years, followed for 6.1±4.0 years, 53% female, mean A1C 11.1±1.9% and 87% African American (AA). Three had Graves' disease (1.3%), three had HT (1.3%) and 97% were euthyroid. TFT were assessed on average every 1.3 years and thyroid antibodies every 2.5 years. Positive thyroid antibody was found in 11%, negative in 57% and unknown in 32%. The positive antibody group had higher mean A1C and TSH. No biopsy confirmed cases of CD (0%) were found when screened every 2.3 years. CONCLUSIONS: The number of individuals who screened positive for hypothyroid HT and CD was lower than expected in our population. Further studies are needed to assess the optimal screening frequency for HT and CD in minority children with T1DM.

Improving paediatric residents' knowledge and perspectives regarding the insulin pump using a novel educational workshop.

OBJECTIVES: As a pilot study, we aimed to investigate the knowledge and perceptions of categorical paediatric residents (RES) at our institution regarding insulin pumps (IPs) and the impact following a targeted workshop. METHODS: All RES at our institution in attendance at a routine noon conference participated in a workshop, completing an anonymous survey before and right after the intervention to evaluate knowledge, attitudes and self-reported comfort regarding IPs. The workshop consisted of a didactic lecture followed by an insulin pump (IP) device demonstration of three commonly available brands. Knowledge score (KS) was calculated for each RES based on the total correct responses. Attitudes were assessed via 5-point Likert scale. Frequencies, t-test and McNemar tests were used to analyse data. RESULTS: Thirty four completed surveys were analysed out of 49 RES (69.3%) who attended the workshop. Among them, there were 19 first-year, 8 second-year and 7 third-year residents. Following the intervention, KS increased significantly (p<0.001) with progression in residents' attitudes. Overall, more RES reported being comfortable with handling the IP, including looking up and changing the settings (p<0.001). CONCLUSION: There is scope for improvement in the knowledge and perceptions of RES regarding IPs. Educational interventions like ours are needed to familiarise our future physicians with IPs to allow hospitals to provide their systematic and safe inpatient use.