A Machine Learning Model for Predicting Hospitalization in Patients with Respiratory Symptoms during the COVID-19 Pandemic.

A machine learning approach is a useful tool for risk-stratifying patients with respiratory symptoms during the COVID-19 pandemic, as it is still evolving. We aimed to verify the predictive capacity of a gradient boosting decision trees (XGboost) algorithm to select the most important predictors including clinical and demographic parameters in patients who sought medical support due to respiratory signs and symptoms (RAPID RISK COVID-19). A total of 7336 patients were enrolled in the study, including 6596 patients that did not require hospitalization and 740 that required hospitalization. We identified that patients with respiratory signs and symptoms, in particular, lower oxyhemoglobin saturation by pulse oximetry (SpO2) and higher respiratory rate, fever, higher heart rate, and lower levels of blood pressure, associated with age, male sex, and the underlying conditions of diabetes mellitus and hypertension, required hospitalization more often. The predictive model yielded a ROC curve with an area under the curve (AUC) of 0.9181 (95% CI, 0.9001 to 0.9361). In conclusion, our model had a high discriminatory value which enabled the identification of a clinical and demographic profile predictive, preventive, and personalized of COVID-19 severity symptoms.

Relation between red blood cell distribution width and acute kidney injury in patients with sepsis.

OBJECTIVE: The objective of the present study is to evaluate the association of red blood cell distribution width with acute kidney injury in sepsis. METHODS: This is a retrospective study of 849 critically ill patients with sepsis in intensive care unit. Demographic data, renal function, inflammation, complete blood count, and acid-base parameters were compared between acute kidney injury and non-acute kidney injury groups. Therefore, a multivariate analysis was performed to observe independent predictive factors. RESULTS: Comparatively, higher levels of C-reactive protein, lactate, red blood cell distribution width, and Simplified Acute Physiology Score 3 were found in the acute kidney injury group. The study showed a higher frequency of women, hemoglobin (Hgb) concentration, platelets, bicarbonate and PaO2/FiO2 ratio in the non-acute kidney injury group. In addition, there was an independent association of comorbidity-chronic kidney disease [OR 3.549, 95%CI: 1.627-7.743; p<0.001], urea [OR 1.047, 95%CI: 1.036-1.058; p<0.001] and RDW [OR 1.158, 95%CI: 1.045-1.283; p=0.005] with acute kidney injury in sepsis patients. CONCLUSION: As an elective risk factor, red blood cell distribution width was independently associated with sepsis-related acute kidney injury. Thus, red blood cell distribution width acts like a predictive factor for sepsis-induced acute kidney injury in intensive care unit admission.

Relation between red blood cell distribution width and acute kidney injury in patients with sepsis

ABSTRACT Objective The objective of the present study is to evaluate the association of red blood cell distribution width with acute kidney injury in sepsis. Methods This is a retrospective study of 849 critically ill patients with sepsis in intensive care unit. Demographic data, renal function, inflammation, complete blood count, and acid-base parameters were compared between acute kidney injury and non-acute kidney injury groups. Therefore, a multivariate analysis was performed to observe independent predictive factors. Results Comparatively, higher levels of C-reactive protein, lactate, red blood cell distribution width, and Simplified Acute Physiology Score 3 were found in the acute kidney injury group. The study showed a higher frequency of women, hemoglobin (Hgb) concentration, platelets, bicarbonate and PaO2/FiO2 ratio in the non-acute kidney injury group. In addition, there was an independent association of comorbidity-chronic kidney disease [OR 3.549, 95%CI: 1.627-7.743; p<0.001], urea [OR 1.047, 95%CI: 1.036-1.058; p<0.001] and RDW [OR 1.158, 95%CI: 1.045-1.283; p=0.005] with acute kidney injury in sepsis patients. Conclusion As an elective risk factor, red blood cell distribution width was independently associated with sepsis-related acute kidney injury. Thus, red blood cell distribution width acts like a predictive factor for sepsis-induced acute kidney injury in intensive care unit admission.

Soluble Fas affects erythropoiesis in vitro and acts as a potential predictor of erythropoiesis-stimulating agent therapy in patients with chronic kidney disease.

Serum soluble Fas (sFas) levels are associated with erythropoietin (Epo) hyporesponsiveness in patients with chronic kidney disease (CKD). Whether sFas could predict the need for erythropoiesis-stimulating agent (ESA) usage and its influence in erythropoiesis remain unclear. We evaluated the relation between sFas and ESA therapy in patients with CKD with anemia and its effect on erythropoiesis in vitro. First, we performed a retrospective cohort study with 77 anemic patients with nondialysis CKD. We performed in vitro experiments to investigate whether sFas could interfere with the behavior of hematopoietic stem cells (HSCs). HSCs were isolated from umbilical cord blood and incubated with recombinant sFas protein in a dose-dependent manner. Serum sFas positively correlated with Epo levels (r = 0.30, P = 0.001) but negatively with hemoglobin (r = -0.55, P < 0.001) and glomerular filtration rate (r = -0.58, P < 0.001) in patients with CKD at baseline. Elevated sFas serum levels (4,316 ± 897 vs. 2,776 ± 749, P < 0.001) with lower estimated glomerular filtration rate (26.2 ± 10.1 vs. 33.5 ± 14.3, P = 0.01) and reduced hemoglobin concentration (11.1 ± 0.9 vs. 12.5 ± 1.2, P < 0.001) were identified in patients who required ESA therapy compared with patients with non-ESA. Afterward, we detected that the sFas level was slight correlated with a necessity of ESA therapy in patients with nondialysis CKD and anemia. In vitro assays demonstrated that the erythroid progenitor cell frequency negatively correlated with sFas concentration (r = -0.72, P < 0.001). There was decreased erythroid colony formation in vitro when CD34+ HSCs were incubated with a higher concentration of sFas protein (1.56 ± 0.29, 4.33 ± 0.53, P < 0.001). Our findings suggest that sFas is a potential predictor for ESA therapy in patients with nondialysis CKD and that elevated sFas could affect erythropoiesis in vitro.