Dynamic of lipoprotein responses in patients with severe COVID-19

Background and Aims: Hypolipidaemia is a known consequence of sepsis, predominantly from HDL-C (HDL-cholesterol) lowering. The dynamic of lipoprotein responses is in COVID-19 is not yet elucidated. We aim to describe a lipoprotein response pattern in patients with severe COVID-19 admitted to Intensive Care Department (ICU) at TUH during the first wave of the pandemic in Ireland. Methods: A multidisciplinary team extracted the clinical data and laboratory results of all patients diagnosed with COVID-19 by RT-PCR and admitted to the ICU department in March and April 2020. Data are presented as means, apart from laboratory data where patients had more than one set of results in 24 hours, when median results were calculated for each 24-h period. Results: Twenty-five patients were admitted to ICU (table 1). Presenting comorbidities included hypertension in 10, cardiovascular disease in 5 and diabetes mellitus in 8 patients. Lipoprotein median concentrations demonstrated initial reduction at admission to ICU, followed by rise in concentration during ICU stay (table 1 and figure 1). A significant negative correlation was observed between ICU outcome and HDL-C area under the curve (AUC) (R=-0.506, p=0.004) and LDL-AUC (R=-0.575, p=0.003). Delta LDL-AUC had the strongest correlation with ICU length of stay (LOS) (R=0.455, p=0.02), hospital LOS (R=0.484, p=0.02) and ICU outcomes (R=-0.454, p=0.02). Individual lipoprotein parameters did not demonstrate significant correlation. [Formula presented] [Formula presented] Conclusions: Lipoprotein concentrations (HDL-C and LDL-C) upon ICU admission are low in severe COVID-19 pneumonia patients and subsequent changes in concentrations may be associated with patient outcomes.

The potential impact of post-COVID symptoms in the healthcare sector.

BACKGROUND: The phenomenon of post-COVID syndrome (PCS) is evolving from an abstract array of non-specific symptoms to an identifiable clinical entity of variable severity. Its frequency and persistence have implications for service delivery and workforce planning. AIMS: This study was aimed to assess the prevalence of symptoms consistent with PCS and the subjective degree of recovery in a cohort of healthcare workers, focusing on those who have returned to work. METHODS: A study population of 1176 was surveyed when attending for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibody testing. Two sub-groups were identified: those with known (i.e. diagnosed on PCR testing) and assumed (i.e. antibody evidence of previous infection) SARs-CoV-2 infection, at least 12 weeks prior to the study. Each group was asked about their subjective degree of recovery and the nature of their persistent symptoms. Results were analysed via excel and SPSS. RESULTS: In total, 144 employees showed PCR evidence of previous infection, with 139 of these being infected at least 12 weeks prior to the study. Of these 139, only 19% (n = 26) reported feeling 100% recovered, and 71% reported persistent symptoms. Of those with assumed SARS-CoV-2 infection (n = 78), 32 (41%) were truly asymptomatic since the commencement of the pandemic, while 46 (59%) described symptoms suggestive of possible infection at least 12 weeks prior to the study. Of this latter group, 23% (n = 18) also reported residual symptoms. CONCLUSIONS: PCS is prevalent among this group, including those not previously diagnosed with COVID-19. Its' frequency and duration present challenges to employers with regards to the management of work availability and performance.

Machine Learning Techniques for Predicting Outcomes of COVID-19 for Patients with preexisting Chronic Diseases

COVID-19 was officially confirmed during December 2019 in the city of Wuhan, China, while the first case of the disease was recorded on 17 November 2019. The World Health Organization has declared a pandemic due to the rapid spread of this disease. It is believed, due to worldwide population aging, that middle-aged and geriatric patients who suffer from chronic diseases are more prone to respiratory failure and having a poorer outcome caused by COVID-19. This paper presents the association of certain chronic diseases such as diabetes mellitus, COPD, hypertension, asthma, and others with COVID-19. Testing was done on 400 samples who were positive for the virus, 250 samples were sick of some of the listed diseases, while 150 were healthy. Machine learning algorithms use computational methods to “learn” information directly from data without relying on a predetermined equation as a model. The algorithms adaptively improve their performance as the number of samples available for learning increases. Specific parameters for different diseases were used in the paper. Based on the results presented in the paper, we concluded that chronic diseases greatly affect the number of people infected with COVID-19. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.