In a recent study, published in The Lancet Diabetes and Endocrinology, researchers evaluated trends in diabetic ketoacidosis (DKA) prevalence at pediatric type 1 diabetes mellitus (T1D) diagnosis prior to and post-coronavirus disease 2019 (COVID-19).
The researchers also investigated potential estimators of DKA prevalence alterations during the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic.
The unprecedented increase in COVID-19 case counts has impacted human lives across the globe with considerable morbidities and deaths. In response, health policies and strategies to limit SARS-CoV-2 spread were executed in most nations in 2020.
SARS-CoV-2-targeted diagnostic efforts delayed several other diagnoses, and the resultant diagnosis at advanced stages was associated with increased mortality rates. An increased DKA prevalence at T1D diagnosis has been observed among children in the initial SARS-CoV-2 wave at several diabetes health centers worldwide.
About the study
In the present international multicentre study, researchers explored the pre-pandemic trends of DKA prevalence at pediatric T1D diagnosis between 2006 and 2019 and compared the DKA prevalence during the SARS-CoV-2 pandemic from 2020 to 2021 with estimates calculated from the pre-pandemic years between 2006 and 2019. They also identified potential estimators of DKA prevalence changes during the COVID-19 pandemic, like COVID-19 severity or SARS-CoV-2 containment measures.
Data were obtained from several national registries [n=13, from Austria, Australia, Denmark, Czechia, Italy, Germany, New Zealand, Luxembourg, Norway, Sweden, the United States of America (USA, Colorado), Wales, and Slovenia]. The study cohort comprised children between six months and 18 years diagnosed with T1D between 1 January 2006 and 31 December 2021.
DKA prevalence rates during 2020 and the consecutive year were compared to estimations based on the pre-SARS-CoV-2 pandemic year trends (between 2006 and 2019). Relationships between DKA prevalence changes and COVID-19 severity and measures for containment were investigated based on excessive all-cause deaths in the entire study cohort and the SI (stringency index) values obtained from the OxCGRT (Oxford COVID-19 government response tracker) database.
In addition, the PubMed database was searched on 22 July 2022, for relevant articles without any language restrictions, as a result of which, 78 results were retrieved, of which 53 articles had relevant topics, including 10, 15, 25, one, and two case reports, single-centre studies, multicentre studies, review, and meta-analyses, respectively. The multicentre studies were conducted largely in single nations and were limited to the initial COVID-19 wave until 28 February 2021.
The IPSAD (international society for pediatric and adolescent diabetes) criteria were followed for DKA detection. The variables assessed included demographic variables (age, sex, month, and year of T1D diagnosis) and DKA variables such as pH of venous blood, serological bicarbonate levels, or clinical DKA diagnosis. Log-binomial and joinpoint regression modeling were used for the analysis, and the estimates were presented as RRs (risk ratios) of DKA development.
The final study population comprised 104,290 children, of which 87,228, 8209, and 8853 children were identified with T1D between 2006 and 2019, in 2020, and 2021, respectively. DKA prevalence was greater among females (28%) and among children aged <6 years (30%). DKA prevalence at T1D diagnosis differed remarkably among countries, ranging between 20% (Germany) and 48% (Colorado).
The average yearly rise in DKA prevalence increased with age at T1D, from 1·1% among children aged <6 years to 2·2% among those aged between 12 years and 18 years, and was most prominent in Australia, Germany, Wales, Slovenia, and Colorado. Only Italy showed a significant average yearly reduction of –1·1%.in DKA at T1D diagnosis from 2006 to 2019.
The highest number of annual DKA cases across the national registries in the pre-pandemic period was 2338, in 2019, which increased to 3005 cases (by 29%) during 2020 and further to 3266 cases (by 40%) in 2021. In 20202, DKA prevalence ranged between 24% in Denmark and 55% in Australia, and in 2021, the prevalence ranged between 28% in Sweden and Denmark to 53% in Australia.
DKA rates over the previous year increased by six percent yearly post-2015, compared to 0.9% yearly prior to 2015. Between 2006 and 2019, DKA at T1D diagnosis was identified in 27% (n=23775) of children. The average yearly increase in DKA prevalence in the entire cohort between 2006 and 2019 was reversed by two percent but subsequently increased by three percent between 2010 and 2019.
The observed DKA prevalence at T1D diagnosis during 2020 and 2021 was 39%, significantly greater than the estimated prevalence of 33% for the two years, with no significant differences by sex or age. DKA likelihood increased with higher SI values, with a RR estimate of having DKA for every 10-unit SI increase of 1.0 for the two years. Contrastingly, DKA likelihood at T1D diagnosis in 2020 and 2021 was not associated significantly with excessive all-cause deaths in the entire population.
Overall, the study findings showed heightened pre-existing increases in DKA prevalence at T1D diagnosis among children during the SARS-CoV-2 pandemic and that the increase was mainly associated with existing increasing trends in DKA prevalence before the pandemic. The findings underscore the importance of prompt diagnosis of T1D among children and that universal efforts are required for reversing the increasing DKA trends.