key: cord-1025304-9fwdunx9
authors: Armeni, Eleni; Aziz, Umaira; Qamar, Sulmaaz; Nasir, Sadia; Nethaji, Chidambaram; Negus, Rupert; Murch, Nicholas; Beynon, Huw Clarke; Bouloux, Pierre; Rosenthal, Miranda; Khan, Sidrah; Yousseif, Ahmed; Menon, Ravi; Karra, Efthimia
title: Protracted ketonaemia in hyperglycaemic emergencies in COVID-19: a retrospective case series
date: 2020-07-01
journal: Lancet Diabetes Endocrinol
DOI: 10.1016/s2213-8587(20)30221-7
sha: 7592bf443b6e027680309f4851354279e1c80072
doc_id: 1025304
cord_uid: 9fwdunx9

nan

Data collection was performed by clinical staff in participating centres who systematically reviewed the hand-written and electronic medical files of all COVID-19 inpatients. All patients fulfilling the inclusion criteria were selected, as eligible for this study. Patients with DKA and/or HHS were treated with intravenous insulin, starting dose of 0.05-0.1 Units per kg of body weight per hour, whereas those with hyperglycemic ketosis received intravenous insulin 1-4 units/hour, based on their blood glucose levels and local guidelines.

We collected clinical/biochemical and anthropometric parameters at the time of presentation, details on admission to the intensive care unit as well as the outcome of the admission at the time of the study. Moreover, details were retrieved related with the classification of diabetes, glycaemic control prior to admission (i.e. HbA1c on at least two occasions, within the last 6 months before admission), as well as medical treatment for control of diabetes, if any. For this purpose, relative general and / or specialist practitioner and/or regular pharmacist or biomedical laboratory were contacted. Data on comorbidities and overall medication history were also included.

Data were analyzed with GraphPad Prism 8 (GraphPad) and the Statistical Package for the Social Sciences (SPSS) version 25. Quantitative data were expressed as median and interquartile range (IQR). Due to the small sample size, the within group analysis for HHS patients (n=2) was performed by presenting median values and range. Categorical variables were given as number (percentage) of participants. Correlation was assessed using Pearson correlation coefficient. The following categorizations were used to describe the strenght of correlations, according to the absolute value of r-coefficient, as follows: <0.20 very weak, 0.20-0.39 weak, 0.40-0.59 moderate, 0.60-0.79 strong, 0.80+ very strong. Kaplan-Meier curves were used to estimate the median time until intensive care admission and the median time to discharge. For all statistical analyses, P < 0.05 was considered statistically significant.

Supplementary Table 1 : Correlation analysis between serum levels of ketones and time to ketone resolution with blood gas parameters using Pearson's correlation coefficient. Statistical significance was set at the level of P < 0.05 

Contributors EK and EA had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis

Concept and design: EK, AY and EA conceived the study. EK and EA established collaborative links

Acquisition, analysis, or interpretation of data: All authors Patient recruitment: All authors Statistical analysis: EA, EK Drafting the manuscript: EK, EA. Critical revision of the manuscript for important intellectual content: all co-authors

All authors have approved the final version of the manuscript

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors