Search

Background: Several risk factors have been identified to predict worse outcomes in patients affected by SARS-CoV-2 infection. Machine learning algorithms represent a novel approach to identifying a prediction model with a good discriminatory capacity to be easily used in clinical practice. The aim of this study was to obtain a risk score for in-hospital mortality in patients with coronavirus disease infection (COVID-19) based on a limited number of features collected at hospital admission., Methods and Results: We studied an Italian cohort of consecutive adult Caucasian patients with laboratory-confirmed COVID-19 who were hospitalized in 13 cardiology units during Spring 2020. The Lasso procedure was used to select the most relevant covariates. The dataset was randomly divided into a training set containing 80% of the data, used for estimating the model, and a test set with the remaining 20%. A Random Forest modeled in-hospital mortality with the selected set of covariates: its accuracy was measured by means of the ROC curve, obtaining AUC, sensitivity, specificity and related 95% confidence interval (CI). This model was then compared with the one obtained by the Gradient Boosting Machine (GBM) and with logistic regression. Finally, to understand if each model has the same performance in the training and test set, the two AUCs were compared using the DeLong's test. Among 701 patients enrolled (mean age 67.2 ± 13.2 years, 69.5% male individuals), 165 (23.5%) died during a median hospitalization of 15 (IQR, 9-24) days. Variables selected by the Lasso procedure were: age, oxygen saturation, PaO2/FiO2, creatinine clearance and elevated troponin. Compared with those who survived, deceased patients were older, had a lower blood oxygenation, lower creatinine clearance levels and higher prevalence of elevated troponin (all P < 0.001). The best performance out of the samples was provided by Random Forest with an AUC of 0.78 (95% CI: 0.68-0.88) and a sensitivity of 0.88 (95% CI: 0.58-1.00). Moreover, Random Forest was the unique model that provided similar performance in sample and out of sample (DeLong test P = 0.78)., Conclusion: In a large COVID-19 population, we showed that a customizable machine learning-based score derived from clinical variables is feasible and effective for the prediction of in-hospital mortality., (Copyright © 2022 Italian Federation of Cardiology - I.F.C. All rights reserved.)

Introduction: The role of sex compared to comorbidities and other prognostic variables in patients with coronavirus disease (COVID-19) is unclear., Methods: This is a retrospective observational study on patients with COVID-19 infection, referred to 13 cardiology units. The primary objective was to assess the difference in risk of death between the sexes. The secondary objective was to explore sex-based heterogeneity in the association between demographic, clinical and laboratory variables, and patients' risk of death., Results: Seven hundred and one patients were included: 214 (30.5%) women and 487 (69.5%) men. During a median follow-up of 15 days, deaths occurred in 39 (18.2%) women and 126 (25.9%) men. In a multivariable Cox regression model, men had a nonsignificantly higher risk of death vs. women (P = 0.07).The risk of death was more than double in men with a low lymphocytes count as compared with men with a high lymphocytes count [overall survival hazard ratio (OS-HR) 2.56, 95% confidence interval (CI) 1.72-3.81]. In contrast, lymphocytes count was not related to death in women (P = 0.03).Platelets count was associated with better outcome in men (OS-HR for increase of 50 × 103 units: 0.88 95% CI 0.78-1.00) but not in women. The strength of association between higher PaO2/FiO2 ratio and lower risk of death was larger in women (OS-HR for increase of 50 mmHg/%: 0.72, 95% CI 0.59-0.89) vs. men (OS-HR: 0.88, 95% CI 0.80-0.98; P = 0.05)., Conclusions: Patients' sex is a relevant variable that should be taken into account when evaluating risk of death from COVID-19. There is a sex-based heterogeneity in the association between baseline variables and patients' risk of death., (Copyright © 2022 Italian Federation of Cardiology - I.F.C. All rights reserved.)

Aims: Patients with cancer are at higher risk of atrial fibrillation, thromboembolic complications and bleeding events compared with the general population. The aim of the present meta-analysis was to compare the efficacy and safety of direct oral Xa inhibitor anticoagulants versus warfarin in patients with cancer and atrial fibrillation., Methods: We searched electronic databases for randomized controlled trials comparing direct oral Xa inhibitor anticoagulants and warfarin in cancer patients. The primary efficacy outcome was stroke or systemic embolism. The primary safety outcome was major bleeding. A subgroup analysis was performed to explore the outcome differences between patients with active cancer or history of cancer., Results: Three trials with a total of 3029 cancer patients were included in the analysis. There was no statistically significant difference in the risk of stroke or systemic embolism [risk ratio (RR) 0.76; 95% confidence interval (CI) 0.52-1.10] between the two therapeutic strategies. Direct oral Xa inhibitors significantly reduced the incidence of major bleeding compared with warfarin (RR 0.79; 95% CI 0.63-0.99; P = 0.04; number needed to treat = 113). These results were consistent both in patients with active cancer and in those with history of cancer., Conclusion: In patients with cancer and atrial fibrillation, direct oral Xa inhibitors have a similar efficacy and may be safer compared with warfarin. These results are consistent both in patients with active cancer and history of cancer.