(254) Impact of COVID-19 Pandemic on ICU Admission and Outcomes in Diabetic Ketoacidosis Patients

Introduction: The surge in cases during the COVID-19 Pandemic placed immense pressure on healthcare facilities, impacting the care of patients with other medical conditions like Diabetic Ketoacidosis (DKA). The population-level effect of the COVID-19 pandemic on Intensive Care Unit (ICU) utilization among patients with DKA and outcomes of those admitted to the ICU has not yet been determined.

Methods: We used a statewide dataset to identify hospitalizations ages ≥ 18 years in acute care hospitals with DKA in Texas during April 1-December 31 for years 2016 through 2020 (each year is restricted to correspond with the of introduction of ICD-10 code U071 for COVID-19 in 2020). Ketoacidosis was defined using selected ICD-10-CM codes chosen from the Clinical Classifications Software Refined Categories END003, END005, END006, and NVS013. Multilevel negative binomial regression was used on the 2016-2019 data to predict the number of ICU admissions in 2020, with change over time expressed as Annual Percent Change (APC). Multilevel logistic regression was used to estimate the change in short-term mortality (in-hospital mortality or discharge to hospice) between 2019 and 2020 among both groups - all ICU admissions with DKA and ICU admissions with DKA without COVID-19.

Results: There were 77,475 hospitalizations with DKA during 2016-2019, of which 83.8% were admitted to the ICU. The rate of ICU admission did not change between 2016 and 2019 (APC +0.6%, 95% CI -0.1 to +1.3%). For 2020 the observed and predicted ICU admissions were 20,459 (83.2% of hospitalizations) vs 20,923 (95% CI 20,539 – 21,308), including 2,832 (13.8%) with COVID-19. Short-term mortality among all ICU admissions with ketoacidosis in 2019 and 2020 was 3.6% vs 8.1% and for ICU admissions without COVID-19 3.6% vs 5.1%. The risk adjusted odds of short-term mortality among ICU admission was higher in 2020 vs 2019 overall (adjusted odds ratio [aOR] 2.59 (95% CI 2.28 - 2.94) and among those without COVID-19 (aOR 1.63 [95% CI 1.42 – 1.87]).

Conclusions: Approximately eight in ten hospitalizations with DKA were admitted to the ICU, with rates of ICU admission not affected by the pandemic. Short-term mortality for ICU admissions with DKA increased significantly during the first year of the COVID-19 pandemic.