Pediatric Research Week - Stony Brook University Hospital

ABSTRACT 1. Resident Effect of Closed Hybrid Loop Insulin Pump Therapy on Glycemic Control in Children with Type 1 Diabetes Aged Less than 7 years Stergiani Agorastos, DO1; Jennifer Osipoff, MD2 1 Department of Pediatrics, Stony Brook Children’s Hospital; 2Division of Pediatric Endocrinology, Department of Pediatrics, Stony Brook Children’s Hospital, Stony Brook, NY Background: Effective glycemic management in children with type 1 diabetes (T1D) remains a significant challenge for families and diabetes care teams. This can be even more difficult in young children who cannot verbalize when they are hypoglycemic and may eat unpredictably. The Medtronic MiniMedTM 670G system with SmartGuardTM technology (670G) was the first insulin pump to automate insulin delivery. It is a closed hybrid looped system that monitors interstitial glucose levels and adjusts insulin requirements in real-time. This system relies on the user wearing a continuous glucose monitor that transmits information to the insulin pump, three to four daily calibrations via capillary finger stick, and insulin boluses for carbohydrates consumed. Should these parameters be met, the user will stay in “auto mode” which will adjust the basal insulin rate depending upon if the sensor glucose is above or below target glucose level. The FDA approved 670G for those 7 years and older and using 8 units of insulin or more. Objectives: 1. To demonstrate the efficacy and safety of 670G in children 7 years old with T1D. 2. To investigate sensor wear, time in auto mode, time in range, episodes of diabetic ketoacidosis (DKA), episodes of severe hypoglycemia (SHo), changes in hemoglobin a1C (HbA1C), and insulin dose in Auto mode. Results: Thirteen children (average age 3.9 years) were included; one was lost over time to relocation. In those individuals that remained in Auto mode 50% of the time, average sensor wear was 59.5% of the time, average time in target range was 49.6% of the time, average time in severe glucose range was 0.93%, and average HbA1C was 8.8. In those individuals that remained in Auto mode 50% of the time, average sensor wear was 88.7% (p 0.010) of the time, average time in range was 56.7% (p 0.22) of the time, average time in severe low glucose range was 0.06% (p 0.18), and average HbA1C was 7.6 (p 0.002). Average daily units of insulin used at time of 670G initiation was 14.1 units (ranging 6-48 units). Throughout the study there were no episodes of DKA. Conclusions: Use of 670G in this population is safe as no children experienced DKA, and on average 1% of the time was spent in severely low glycemic states. Young children were able to wear the continuous glucose monitor regularly. Those using less than 8 units a day were still able to have success with Auto mode. Higher percent time in Auto mode led to reductions of HbA1C. Off label use of automated insulin delivery via 670G should be considered in children with T1D. Methods: Retrospective study spanning January 1, 2015 to October 1, 2019. Inclusion criteria: patient of Stony Brook Pediatric Endocrinology Clinic, diagnosis of T1D, and 7 years old when first placed on 670G. 7

ABSTRACT 2. Resident QT Prolongation: A Key Finding in Metabolic Hypoglycemia Mallory Carson, DO1; Brecken Esper, BS2; Kimberly Tafuri, DO3; Andrew Lane, MD3; Charlene Fox, MS, RDN, CDCES, CDN4; Jody Weiss-Burns, MS, CGC4; Patricia Galvin-Parton, MD4 1 Department of Pediatrics, Stony Brook Children’s Hospital; 2Stony Brook Medicine; 3Division of Pediatric Endocrinology, Department of Pediatrics, Stony Brook Children’s Hospital; 4Division of Medical Genetics, Department of Pediatrics, Stony Brook Children’s Hospital, Stony Brook, NY Background: The TANGO2 gene encodes a transport and Golgi organization protein, whose function is not well understood at this time. Case Description: A 14 month old female presented to the emergency department with altered mental status and a glucose level of 26 mg/dl. Upon arrival, she was unresponsive, intubated, and admitted to the pediatric intensive care unit. Her initial brain MRI showed hypoxic ischemic encephalopathy. The differential diagnosis for her presentation included disorders of fructose metabolism, as juice was recently introduced. Initial labs revealed elevated lactic acid and unremarkable amino acid and acylcarnitine profiles. Normal critical samples ruled out endocrine etiologies. A metabolic hypoglycemia panel was ordered to evaluate for additional causes of hypoglycemia. During admission, she required dextrose containing fluids to maintain her glucose levels. An echocardiogram showed mildly reduced global left ventricular systolic function. She had persistently elevated troponins and a prolonged QT interval, so she was treated for myocarditis. Her troponin levels down trended and her EKG stabilized. She was weaned off of intravenous fluids and maintained her glucose level with strict regular feeding. A long QT syndrome panel was negative. The hypoglycemia panel demonstrated mutations in TANGO2: a pathogenic 31.8 kb deletion of exon 3 to 9 and a suspected pathogenic hemizygous c.569 592dup, p.Ile190 Leu197dup. Individuals with TANGO2 mutations present in acute metabolic crisis with muscle weakness, ataxia, rhabdomyolysis, or altered mental status, which is typically triggered by acute illnesses or fasting, and have laboratory findings of hypoglycemia, lactic acidosis, and hyperammonemia. Transient EKG and echocardiogram changes are seen during the acute crisis, usually QT prolongation and episodic systolic dysfunction. Hypothyroidism and adrenal insufficiency can also develop. The hypoglycemic panel also excluded diagnoses of fructose 1,6-bisphosphatase deficiency and hereditary fructose intolerance. Upon discharge, she was given no diet restrictions but instructed to avoid fasting. She presented to the emergency room 3 months later with another episode of altered mental status and hypoglycemia, so decision was made to use a continuous glucose monitor to prevent further hypoglycemic episodes. Conclusion: A TANGO2 gene mutation is a newly described diagnosis that should be considered in patients with acute metabolic crisis, hypoglycemia, and cardiac arrhythmias. This patient’s presentation and genetic testing is consistent with this diagnosis. 8

ABSTRACT 3. Resident Comparison of Diagnostic Group and Severity of Diagnosis on Caregiver Burden of Children with Known Neurological Disease Mitchell P. Creed, MD, MA1; Ruth A. Reinsel, PhD2; Louis Manganas, MD, PhD2 1 Department of Pediatrics, Stony Brook Children’s Hospital; 2Division of Pediatric Neurology, Department of Neurology, Stony Brook University Hospital, Stony Brook, NY Background: The clinical features of pediatric neurologic and epileptiform disorders are well-described. However, little is known about the impact on the family of a child with a neurologic or epileptiform diagnosis. calls. 44 responses were received from primary caregivers (47% response rate). Respondents were female (96%) and well-educated (91% with college degree or greater) with health insurance coverage (80% private insurance, 18% Medicare). Respondents reported on children with a variety of neurological diagnoses (Epilepsy n 12, ADHD/ ADD n 15, Autism/Developmental Delay n 5, Tics/Movement Disorder n 6, Headaches n 6). ZBI scores were 7.6 8.2 (mean sd) in the mild DS group (n 20) and 14.0 11.2 in the Moderate/ Severe group (n 23, P 0.037). Overall, the ZBI scores for the 3 diagnostic groups did not differ on one-way ANOVA (P 0.214), but the subset of moderate/severe ADHD patients posed significantly more caregiver burden than moderate/severe Epilepsy patients (P 0.014 by Mann-Whitney U Test, see Figure). Objective: To demonstrate that severity of a patient’s neurologic or epileptiform disorder is directly correlated to perception of burden by the caregiver. We compared caregiver burden of patients with epilepsy to patients with other neurological conditions, considering disease severity (DS). Results: 106 fliers were returned, of which 13 were considered ineligible. 93 participants were sent personalized survey links on the secure Qualtrics platform. Initial non-respondents were sent reminder emails and up to 3 follow-up phone Conclusion: Caregiver burden is more influenced by severity of diagnosis than diagnostic group. Additionally, moderate/severe ADHD/ADD may present with caregiver burden greater than that of moderate/severe epilepsy. 40 Zarit Burden Scale, Total Score Methods/Design: After IRB approval, a nonrandomized, observational study was performed via an internet survey. Recruitment fliers were distributed to caregivers of children aged 2 to 17 years with diagnosed neurological conditions during routine clinic visits. Parent/caregiver selfreport data was collected using Peds QL (Quality of Life) 3.0 Epilepsy Module, Peds QL Core questionnaires and the Zarit Burden Interview (ZBI) (12-item short form). DS was rated by the parent/caregiver for children with non-epileptic diagnoses. For epileptic patients, DS was estimated by number of antiseizure medications ( 2 moderate/severe) and number of seizures, ER visits and hospitalizations in the past year ( 1 moderate/severe). DS was classed as mild (n 20) versus moderate/severe (n 23) and ZBI scores were compared by t-test. ZBI scores were grouped by diagnosis (Epilepsy, ADD/ADHD and Other) and compared using nonparametric tests. Disease Severity Mild Moderate/Severe * 30 20 74 10 0 N 7 N 8 ADD/ADHA N 5 N 7 Epilepsy N 8 N 8 All Other Dx Epilepsy vs. ADHA vs. All Other Diagnoses 9

ABSTRACT 4. Resident Pain Perception and Reduction Modalities in Pediatric Rheumatic Patients Receiving Biologic Injections and Infusions Ommul Fatmi, DO1; Farzana Nuruzzaman, MD2; Héctor Alcalá, PhD, MPH, CPH3; Julie Cherian, MD2 1 Department of Pediatrics, Stony Brook Children’s Hospital; 2Division of Pediatric Rheumatology, Department of Pediatrics, Stony Brook Children’s Hospital; 3Department of Family, Population and Preventive Medicine, Program in Public Health, Stony Brook University, Stony Brook, NY Background: The advent of biologic therapies has revolutionized long term outcomes and quality of life for children with rheumatic diseases. However, these treatments, which may involve frequent painful injections or infusions, also pose a challenge. They may cause psychological trauma and aversion to necessary treatments, which may result in medication noncompliance and worsening outcomes. Current studies have focused on evaluating pain associated with intravenous line (IV) placement and injections and the benefits of pain control methods in children. However, these studies have evaluated short term injections/immunizations, having IVs placed, or phlebotomy. These studies are difficult to extrapolate to children receiving chronic, frequent, biologic medications for long term disease control. Objectives: The primary aim of this study was to identify what form of pain control, if any, patients and parents use to aid in administering these medications. A second aim is to identify if presence of anxiety, fear and/or pain during medication administration is associated with treatment non-adherence, with the possibility of poor clinical outcomes. Methods: This is an observational, cross sectional study conducted using online surveys completed by parents and/or patients. Inclusion criteria were patients 2-21 years of age currently receiving biologic medications via IV infusions or subcutaneous injections to treat rheumatic disease. In addition to parent surveys, children above the age of 7 were asked to complete the online survey. Statistical analysis was performed using the Fischer’s exact test to compare parent and patient responses about pain control methods and evaluate the variability of pain perception among different biologic medications. Results: A total of 35 surveys were completed and 74% of respondents were Caucasian. Most participants had JIA and the majority took weekly subcutaneous injections. Approximately 63% of participants reported using varying pain reducing techniques to aid in medication administration. Most children did not report fear of getting their medications, though their parents perceived them to have anxiety. Patients getting medications more frequently were more likely to report anxiety with the process. Conclusion: Though limited, the survey indicates that children use a variety of pain reducing techniques to take their biologic medications and still have anxiety with receiving them. There may also be parental factors that may increase anxiety during the medication administration. The implications of this study may be extrapolated to other specialties using biologic therapies in pediatric patients. 10

ABSTRACT 5. Resident Severe Hypovolemic Shock and Methemoglobinemia Secondary to Food Protein-Induced Enterocolitis Syndrome in an Infant Annamarie Fernandes, MD1; Catherine Urban, MD2 1 Department of Combined Internal Medicine-Pediatrics, Stony Brook University Hospital; 2Division of Pediatric Critical Care, Department of Pediatrics, Stony Brook Children’s Hospital, Stony Brook, NY Background: Food protein-induced enterocolitis syndrome (FPIES) is an uncommon non-IgE mediated gastrointestinal illness in infants, most commonly triggered by cow’s milk, soy and rice. The range of symptoms include failure to thrive, vomiting, diarrhea, hypovolemic shock and rarely methemoglobinemia. Case Description: A 4-week-old full term male presented with one week of profuse non-bloody and nonbilious vomiting, diarrhea, decreased oral intake, and weight loss. Prenatal course was unremarkable. In the emergency room, vital signs were normal and weight decreased from 4.13kg to 3.11kg over one week. Labs were notable for sodium 141, potassium 4.7, chloride 119, bicarbonate 6, otherwise normal chemistry and hepatic panel. WBC 27.9, platelet count 742, and positive fecal occult blood test. Upon arrival to the pediatric intensive care unit, the infant was lethargic and hypoxemic (normalized with supplemental oxygen), with weak peripheral pulses and tone. Initial arterial blood gas as follows: pH 7.13, pCO2 18, paO2 161, HCO3 6, base deficit 23, metHg 30. The patient was aggressively fluid resuscitated but continued to have significant acidosis so opted to treat methemoglobinemia with methylene blue. A sepsis and metabolic workup were initiated and empiric ceftriaxone was started. CSF studies, cultures, rotavirus resulted normal. Upper GI series showed no malrotation, KUB demonstrated a non-obstructive bowel pattern, abdominal ultrasound was normal. Further history revealed that the patient was primarily fed breast milk, plus recent supplementation with cow’s milk formula. Mother denied sick contacts, use of additional medications, creams or oral numbing gel to account for the methemoglobinemia. No concern for exposure to outdoor chemicals or carbon monoxide. Patient was stabilized and was restarted on breast milk with adequate weight gain. Genetic inborn errors of metabolism studies showed a low concentration of several amino acids thought to be related to severe malnutrition. With infectious and metabolic diagnoses ruled out, FPIES as a diagnosis of exclusion was considered and diagnostic criteria was fulfilled (American Academy of Allergy, Asthma & Immunology 2017 Guidelines). Treatment included continuing breastmilk, supplementing with elemental formula, and avoidance of cow’s milk by mother. Conclusion: FPIES is an uncommon severe diarrheal illness in infants. Severity can vary significantly and can be associated with lifethreatening methemoglobinemia. Literature, especially relevant to the more severe form, is limited. Early recognition is important to prevent excessive testing, prolonged hospital stay, and failure to thrive. 11

ABSTRACT 6. Resident An Interdisciplinary Quality Improvement Initiative to Standardize the Inpatient Pediatric Medication Reconciliation Process Joshua Glass, MD1; Jessica Kr

Welcome to Stony Brook Children's Hospital's second annual Department of Pediatrics Virtual Research Week. This is our department's 13th year recognizing the scholarship of our resident . patient of Stony Brook Pediatric Endocrinology Clinic, diagnosis of T1D, and 7 years old when first placed on 670G. Results: Thirteen children .