Impact of Covid-19 on the Rate of Diabetic Ketoacidosis in Pediatric New-Onset Type 1 Diabetes

Introduction: Several demographic and clinical characteristics, including age, low socioeconomic status, and misdiagnosis at initial clinical presentation were previously associated with increased risk of diabetic ketoacidosis (DKA) at diagnosis of type 1 diabetes (T1D) in the pediatric population. However, it is unclear whether the coronavirus (COVID-19) pandemic and subsequent lockdown influenced the rate of DKA in children newly diagnosed with T1D. We undertook this study to identify the impact of the COVID-19 pandemic on the rate of DKA in children newly diagnosed with T1D in a single tertiary care referral center in central Pennsylvania. Methods: We performed an extension of a retrospective analysis of all pediatric patients (age ≤18) newly diagnosed with T1D within a tertiary care referral center between 01/01/2017-09/14/2020. Demographics, insurance coverage, and all clinical documents 30 days before their T1D diagnosis were abstracted to assess for symptoms at diagnosis (polyuria, polydipsia, nocturia, weight loss, nausea, vomiting, altered mental status, infection, vision changes, and autism spectrum disorder), lab values (blood glucose, HbA(1c), venous pH, and bicarbonate), and any healthcare encounters within 30 days of their diagnosis of T1D. We performed descriptive statistics and univariate analyses [evaluating children diagnosed with T1D during the pre-COVID-19 era (diagnosed between 1/1/2017-2/28/2020) and post-COVID-19 era (diagnosed between 03/01/2020-09/14/2020) associated with the incidence of DKA], followed by logistic regression analysis (incorporating key clinical factors previously associated with DKA and the pre- or post- COVID-19 era classification). Results: 412 pediatric patients with T1D [171 F:241 M; 370 pre-COVID-19 era:42 post-COVID-19 era] were included. The percentages of DKA diagnoses at admission were very similar between the pre-COVID-19 and post-COVID-19 groups (47% vs. 48%), as were the severity (13% vs. 14% mild DKA; 33% vs. 31% moderate or severe DKA). There were no temporal associations with the rate of DKA in respect to COVID-19, however, age (0-3 and 9-13 years), misdiagnosis during a preceding healthcare encounter, presenting to the emergency department directly, elevated HbA(1c) (>10.0%/13.4mmol/L), and altered mental status were associated with increased risk of DKA on multivariable analysis. Conclusion: There were no fluctuations in the rate of DKA among pediatric patients newly diagnosed with T1D throughout the COVID-19 pandemic in central Pennsylvania. Interestingly, some geographic locations observed an increased frequency of DKA in children newly diagnosed with T1D, while others noted a decreased rate. Regardless, our findings suggest previously described predictors of DKA in the pediatric population persist, even in the setting of the COVID-19 pandemic.