The incidence of diabetes, both type 1 and type 2, is increasing. Health outcomes in pediatric diabetes are currently poor, with trends indicating that they are worsening. Minority racial/ethnic groups are disproportionately affected by suboptimal glucose control and have a higher risk of acute and chronic complications of diabetes. Correct clinical management starts with timely and accurate classification of diabetes, but in children this is becoming increasingly challenging due to high prevalence of obesity and shifting demographic composition. The growing obesity epidemic complicates classification by obesity's effects on diabetes. Since the prevalence and clinical characteristics of diabetes vary among racial/ethnic groups, migration between countries leads to changes in the distribution of diabetes types in a certain geographical area, challenging the clinician's ability to classify diabetes. These challenges must be addressed to correctly classify diabetes and establish an appropriate treatment strategy early in the course of disease for all. This may be the first step in improving diabetes outcomes across racial/ethnic groups. This review will discuss the pitfalls in the current diabetes classification scheme that is leading to increasing overlap between diabetes types and heterogeneity within each type. It will also present proposed alternative classification schemes and approaches to understanding diabetes type that may improve the timely and accurate classification of pediatric diabetes type. K E Y W O R D S delayed diagnosis, diabetes complications, diabetes mellitus, MODY, pediatrics 1 | INTRODUCTION Diabetes is one of the most common chronic diseases in pediatric populations, and, if poorly managed, can result in acute and long-term complications that can significantly affect childhood and reduce productivity in adult years. 1 The incidences of pediatric type 1 diabetes (T1D) and type 2 diabetes (T2D) are increasing for reasons not completely understood, but this may in part be due to the increasing incidence of obesity. 1 The incidences of pediatric T1D and T2D have increased globally. 2-7 In the United States, at the current rate, the number of T1D and T2D cases could triple and quadruple to almost 600 000 and 85 000 US children, respectively, by 2050. 8 Current trends suggest that this increase will disproportionately affect non-Caucasian groups. 9 In the United Kingdom, South Asians are the group with the most dramatic increase. 4 While lower socioeconomic status is a risk factor for pediatric T2D in developed countries, in emerging countries, children from the most affluent families are at highest risk. 6,10 We and others demonstrated that elevated body mass index (BMI), in addition to its well-known association with T2D, is associated with accelerated progression to T1D in children 11,12 and adults. 13 Obesity-induced insulin resistance may render insufficient the already compromised insulin secretion in individuals with islet autoimmunity and thus, trigger clinical diabetes. In addition, obesi...

Current practice guidelines recommend considering a diagnosis of MODY in patients with a strong family history but without typical features of type 2 diabetes (T2D) (i.e., obesity, insulin resistance, high-risk ethnicity) or type 1 diabetes (T1D) (i.e., islet autoantibodies). We aimed to test if these criteria apply to a racially diverse pediatric patient population. Using EPIC Population Health, we retrospectively reviewed electronic medical records (IRB H-41512) of youth diagnosed with MODY between 2013-2017 in a large academic pediatric hospital in Southwestern U.S. We found that, out of 46 patients with MODY, 50% (n=23) had MODY 2, 17.3% (n=8) MODY 3, 17.3% (n=8) MODY 5, 8.6% (n=4) MODY 1, 2.1% (n=1) MODY 4, and 8.6% of patients had a clinical diagnosis of MODY due to inability to perform molecular testing. The mean (SD, range) age of diabetes diagnosis was 10.1 years (4.4, 2-17); 52% of patients were females; 50% were non-Hispanic white, 32.5% Hispanic, 6.5% African American, 6.6% Lebanese, 2.2% Asian and 2.2% others/unknown. Notably, 15.2% (n=7, all with a molecular diagnosis of MODY) had positive islet antibodies, 8.6% (n=4) Hashimoto’s thyroiditis and 2.1% (n=1) vitiligo. Obesity/overweight was present in 23.8%, 15.2% had hypertension / prehypertension and 21.7% had dyslipidemia. Diabetes was absent in parents and siblings of 26% (n=12) of patients, although in 5 patients, the mother had gestational diabetes. Persistent microalbuminuria was found in 2 youth with MODY 2 and diabetic neuropathy in 1 patient with MODY 2. The mean (SD, range) time between the diagnoses of diabetes and MODY was 14.3 months (22.3, 1-120). In conclusion, youth with MODY often were of ethnic minorities, lacked a family history of diabetes and had characteristics typical of T1D or T2D, suggesting that the current guidelines to consider a diagnosis of MODY may not apply to a racially diverse U.S. pediatric population. Furthermore, in contrast to previous studies, we identified microvascular complications in youth with MODY 2. Disclosure S. Relan: None. F.C. Hessel: None. K.D. Timmons: None. R. Sonabend: None. M.J. Redondo: None.

Background: Adolescents with type 1 diabetes mellitus (T1D) are at increased risk for depressive symptoms relative to youth without diabetes, which is associated with poor diabetes outcomes (Hood et al, 2014). As part of ongoing Quality Improvement initiatives at Texas Children’s Hospital (TCH), we evaluated two validated depression screening measures, the PHQ-2 and PHQ-9: Modified for Teens, to standardize our depression screening process. Objectives: We sought to (1) review screening outcomes from youth with T1D who completed both the PHQ-2 and PHQ-9, (2) determine which measure provided the greatest sensitivity to identifying depressive symptoms, and (3) create a standardized process for implementation of the more sensitive screen. Methods: Data was collected by a chart review of PHQ-2 and PHQ-9 scores for all patients with T1D age 12-17 seen at TCH Endocrine clinic sites between January 1 –December 31, 2017. A positive screen was defined as PHQ-2 score ≥ 3 and PHQ-9 score ≥ 5. We then created an algorithm based on the more sensitive screen for referrals to mental health services. Results: Of 961 eligible patients, 84% were screened using PHQ-2 and 15% using both the PHQ-2 and PHQ-9. Of the patients who completed both measures, the positive screening rate was 20% in PHQ-9 vs. 4% in PHQ-2. None of the patients with “moderate” PHQ-9 (≥5-9) screened positive on the PHQ-2, and only 1 patient with “severe” PHQ-9 (≥10) score screened positive on PHQ-2. Of the 3 patients who endorsed suicidality on the PHQ-9, none had a positive PHQ-2. 80% of patients who received both screens completed the screens within 2 months of each other. Discussion: PHQ-2 depression screen may be less sensitive than PHQ-9 for capturing depressive symptoms in youth with T1D. We have created a standardized algorithm implementing PHQ-9 into the clinic flow and looking at optimizing mental health referrals. Hood KK, Lawrence JM, Anderson A, Bell R, et al. Metabolic and inflammatory links to depression in youth with diabetes. Diabetes Care 2012;35:2443-2446. Disclosure G.K. Kim: None. S. Relan: None. M. Tosur: None. S. Agarwal: None. I. Jindal: None. T. Patel Moorjani: None. K. Fegan-Bohm: None. K.A. Gallagher: None. K. Hendrix: None.

Background Hemihypertrophy is a rare clinical presentation of Cushing's disease. It has been described in three newborns with massive enlargement of adrenal glands in presence of other congenital abnormalities and in a 17-year-old female with pituitary adenoma with a genetic diagnosis of Beckwith Weidman syndrome. We describe a case of Cushing's disease with lower extremity hemihypertrophy in the absence of features suggestive of an underlying syndrome. 1,2 Clinical Case: A 12-year-old male with no significant past medical history presented with poor growth and rapid weight gain. In addition, he had hemihypertrophy of left lower extremity interfering with his physical activity. Physical exam showed Tanner Stage 1 for pubic hair and testicular volume of 4 ml bilaterally, with no signs of hirsutism, bruising or purplish striae. Initial work up showed normal thyroid hormone and growth factor levels. Midnight salivary cortisol levels were 238 ng/dl and 51 ng/dl respectively (range <100 ng/dl). One mg dexamethasone suppression test showed suppressed cortisol level of 1 mcg/dl. At the subsequent visit seven months later, he had gained 13.8 kg and grown 0.9 cm in height. Physical exam showed facial plethora and buffalo hump, which had not been present at the initial visit. Left and right leg circumference above the knee were 22 inches and 20 inches respectively. Repeat midnight cortisol levels on two separate occasions were 251 ng/dl and 441 ng/dl respectively. Repeat 1 mg dexamethasone suppression test showed suppressed cortisol levels of 1.2 mcg/dl. Because of clinical picture consistent with Cushing syndrome and elevated salivary cortisol levels, 24-hour urine cortisol levels were ordered that were 145 mcg and 276 mcg/24h respectively (range 1-45 mcg/24hours). ACTH was elevated at 65 pg/ml. CT scan of abdomen and pelvis was normal. Left leg ultrasound was normal. MRI of brain showed a 4 mm hypoenhancing lesion in the pituitary gland to the left of midline. A peripheral DDAVP stimulation test showed elevated ACTH and cortisol levels suggestive of pituitary origin. MRI of legs showed asymmetric fat deposition causing hemihypertrophy. Patient underwent transsphenoidal resection of pituitary adenoma with postoperative remission (postop cortisol nadir <1 mcg/dl). At follow up visit after surgery, patient had lost 0.2 kg, buffalo hump had disappeared, and left leg circumference had slightly improved. Genetic testing ispending. Conclusion This case demonstrates that hemihypertrophy can be a presenting symptom of Cushing's disease. References: (1)Brioude F, Nicolas C, Marey I, et al. Hypercortisolism due to a pituitary adenoma associated with beckwith-wiedemann syndrome. Horm Res Paediatr. 2016;86(3): 206-211. (2)Carney JA, Ho J, Kitsuda K, Young WF, Stratakis CA. Massive neonatal adrenal enlargement due to cytomegaly, persistence of the transient cortex, and hyperplasia of the permanent cortex: Findings in cushing syndrome associated with hemihypertrophy. Am J Surg Pathol. 2012;36(10): 1452-1463. Presentation: No date and time listed