ABSTRACT
Since the World Health Organization (WHO) has declared Artificial Intelligence (AI) as a powerful tool in the fight against COVID-19, multiple studies have been launched aiming to shed light into risk factors for ICU admission and mortality. None of the existing studies, however, have captured the dynamic trajectories of hospitalized COVID-19 patients who receive steroids nor have explored trajectory-based mortality indicators. In this work, we present a novel, hybrid approach to address this need. Latent Growth Mixture Modelling (LGMM) was used to analyze the trajectories of patients who received steroids. The patients were then grouped into clusters based on the similarity of their dynamic trajectories. State-of-the art machine learning classifiers are trained on the original dataset with and without dynamic trajectories to assess whether their inclusion can enhance the prediction of mortality. Our results highlight the importance of trajectories for predicting mortality in patients who receive steroids yielding 4% and 5% increase in the sensitivity (0.84) and specificity (0.85). The FiO2 and percentage of neutrophils at day 5, along with the percentage of lymphocytes at day 7, were identified as the main causes for mortality in patients who receive steroids, where the SatO2 levels showed significant alterations in the dynamic trajectories. © 2022 IEEE.
ABSTRACT
Background: Coronavirus Disease-19 (COVID-19) caused by the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is characterized by a wide range of clinical manifestations 1. Although COVID-19 was initially considered a respiratory infection, it was shortly recognized as a multisystemic disorder associated with heightened infammatory responses, including autoimmune phenomena 1. The presence of autoantibodies (AAbs) has been described in COVID-19 patients, highlighting the state of immune dysregulation in COVID-19 1. The clinical signifcance of AAbs, however, is still elusive. Objectives: To assess the prevalence of AAbs in critically ill, mechanically ventilated COVID-19 patients admitted to the intensive care unit (ICU) and investigate whether AAbs influence the clinical outcome of these patients. Methods: The current study evaluated prospectively from March 8th, 2021 to May 10th, 2021 the presence of AAbs against nuclear antigens (ANA), extracta-ble nuclear antigens (ENA), neutrophil cytoplasmic antigens (ANCA), cyclic cit-rullinated peptides (anti-CCP), double stranded-DNA (anti-dsDNA), cardiolipin (anti-CL), β2-glycoprotein-I (anti-β2-GPI), thyroid peroxidase (anti-TPO), and thyroglobulin (anti-TG) in critically ill COVID-19 patients upon admission in the ICU (n=217). Samples from 60 COVID-19 patients that were available 15 days after ICU admission were further analyzed for the evaluation of de novo AAbs production. Serum samples of age and sex matched healthy individuals before the COVID-19 pandemic were used as a control group (n=117). Results: COVID-19 patients treated in ICU had more commonly at least one AAb compared to age and sex matched controls (174/217, 80.2% vs 73/217, 62,4%, p< 0,001). More specifcally, COVID-19 patients expressed more frequently ANAs (48.4% vs 21.4%, p<0.001), anti-dsDNA (5.1% vs 0%, p=0.01), anti-CCP (8.3% vs 1.7%, p=0.014) and anti-CL IgM AAbs (21.7% vs 9.4%, p=0.005) than controls. The majority of critical COVID-19 patients who were positive for AAbs (144, 82.8%) expressed reactivity in up to three autoantigens with the most prominent being ANA, anti-phospholipid, ANCA and anti-TPO AAbs. AAbs-positive patients demonstrated more robust anti-SARS-CoV-2 humoral responses compared to AAbs-negative patients [detectable anti-SARS-CoV-2 S1-protein IgG antibodies: 150 (86.2%) vs 28 (65.1%), p=0.001;adequate neutralizing activity: 159 (91.4%) vs. 33 (76.7%), p=0.007]. The two groups, however, did not differ in terms of clinicoepidemiologic characteristics or the incidence of death in the ICU. Convalescent COVID-19 patients (n=111) compared to those who died (n=106), did not differ in the prevalence of serum AAbs or antibody responses against SARS-CoV-2. Differences were only shown in clinicolaboratory parameters including patients' age, comorbidities, O2 saturation, infammatory markers, and in-hospi-tal prognostic scores as expected. Paired samples testing (n=60) revealed that 45 patients had at least one newly induced AAb, 28 patients lost at least one reactivity and only 6 patients didn't show any seroconversion. The most common new-onset AAb reactivity was against anti-CL (IgG isotype) (n=21) followed by ANA (n=20), anti-β2-GPI (IgG isotype) (n=11), myositis-related antigens (n=13) and ENAs (n=9);nevertheless, no associations with clinicoepidemiologic features or COVID-19 outcome were revealed. Conclusion: Patients with severe COVID-19 express AAbs more commonly than age and sex matched controls, suggesting that SARS-COV-2 infection may induce a hitherto unknown B-cell autoreactivity. The presence of autoantibodies does not play a role in the outcome of SARS-COV-2 infection. However, further studies are needed to defne their role in future development of systemic autoimmune disorders or the long-COVID syndrome.