Risk Factors of Mortality in Hospitalized Patients With COVID-19 Applying a Machine Learning Algorithm Factores de riesgo de mortalidad en pacientes hospitalizados con COVID-19 aplicando un algoritmo de aprendizaje automático

Introduction Risk stratification of patients with COVID-19 can be fundamental to support clinical decision-making and optimize resources. The objective of our study is to identify among the routinely tested clinical and analytical parameters those that would allow us to determine patients with the highest risk of dying from COVID-19. Material and methods We carried out a retrospective cohort multicentric study by consecutively, including hospitalized patients with COVID-19 admitted in any of the 11 hospitals in the healthcare network of HM Hospitals-Spain. We collected the clinical, demographic, analytical, and radiological data from the patient's medical records. To assess each of the biomarkers’ predictive impact and measure the statistical significance of the variables involved in the analysis, we applied a random forest with a permutation method. We used the similarity measure induced by a previously classification model and adjusted the k-groups clustering algorithm based on the energy distance to stratify patients into a high and low-risk group. Finally, we adjusted two optimal classification trees to have a schematic representation of the cut-off points. Results We included 1246 patients (average age of 65.36 years, 62% males). During the study one hundred sixty-eight patients (13%) died. High values of age, D-Dimer, White Blood Cell, Na, CRP, and creatinine represent the factors that identify high-risk patients who would die. Conclusions Age seems to be the primary predictor of mortality in patients with SARS-CoV-2 infection, while the impact of acute phase reactants and blood cellularity is also highly relevant.

Effect of high versus low dose of dexamethasone on clinical worsening in patients hospitalised with moderate or severe COVID-19 Pneumonia: an open-label, randomised clinical trial.

BACKGROUND: Low dose dexamethasone demonstrated clinical improvement in patients with coronavirus disease 2019 (COVID-19) needing oxygen therapy; however, evidence on the efficacy of high dose of dexamethasone is limited. METHODS: We performed a randomised, open-label, controlled trial involving hospitalised patients with confirmed COVID-19 pneumonia needing oxygen therapy. Patients were randomly assigned in a 1:1 ratio to receive low dose dexamethasone (6 mg once daily for 10 days) or high dose dexamethasone (20 mg once daily for 5 days, followed by 10 mg once daily for additional 5 days). The primary outcome was clinical worsening within 11 days since randomisation. Secondary outcomes included 28-day mortality, time to recovery, and clinical status at day 5, 11, 14 and 28 on an ordinal scale ranging from 1 (discharged) to 7 (death). RESULTS: A total of 200 patients (mean (sd) age, 64 (14) years; 62% male) were enrolled. Thirty-two patients of 102 (31.4%) enrolled in the low dose group and 16 of 98 (16.3%) in the high dose group showed clinical worsening within 11 days since randomisation (rate ratio, 0.427; 95% CI, 0.216-0.842; p=0.014). The 28-day mortality was 5.9% in the low dose group and 6.1% in the high dose group (p=0.844). There was no significant difference in time to recovery, and in the 7-point ordinal scale at day 5, 11, 14 and 28. CONCLUSIONS: Among hospitalised COVID-19 patients needing oxygen therapy, high dose of dexamethasone reduced clinical worsening within 11 days after randomisation as compared with low dose.

Development and validation of a clinical score to estimate progression to severe or critical state in COVID-19 pneumonia hospitalized patients.

The prognosis of a patient with COVID-19 pneumonia is uncertain. Our objective was to establish a predictive model of disease progression to facilitate early decision-making. A retrospective study was performed of patients admitted with COVID-19 pneumonia, classified as severe (admission to the intensive care unit, mechanic invasive ventilation, or death) or non-severe. A predictive model based on clinical, laboratory, and radiological parameters was built. The probability of progression to severe disease was estimated by logistic regression analysis. Calibration and discrimination (receiver operating characteristics curves and AUC) were assessed to determine model performance. During the study period 1152 patients presented with SARS-CoV-2 infection, of whom 229 (19.9%) were admitted for pneumonia. During hospitalization, 51 (22.3%) progressed to severe disease, of whom 26 required ICU care (11.4); 17 (7.4%) underwent invasive mechanical ventilation, and 32 (14%) died of any cause. Five predictors determined within 24 h of admission were identified: Diabetes, Age, Lymphocyte count, SaO2, and pH (DALSH score). The prediction model showed a good clinical performance, including discrimination (AUC 0.87 CI 0.81, 0.92) and calibration (Brier score = 0.11). In total, 0%, 12%, and 50% of patients with severity risk scores ≤ 5%, 6-25%, and > 25% exhibited disease progression, respectively. A risk score based on five factors predicts disease progression and facilitates early decision-making according to prognosis.