Sepsis is a highly lethal syndrome with heterogeneous clinical manifestation that can be hard to identify and treat. Early diagnosis and appropriate treatment are critical to reduce mortality and promote survival in suspected cases and improve the outcomes. Several screening prediction systems have been proposed for evaluating the early detection of patient deterioration, but the efficacy is still limited at individual level. The increasing amount and the versatility of healthcare data suggest implementing machine learning techniques to develop models for predicting sepsis. This work presents an experimental study of some machine-learning-based models for sepsis prediction considering vital signs, laboratory test results, and demographics using Medical Information Mart for Intensive Care III (MIMIC-III) (v1.4), a publicly available dataset. The experimental results demonstrate an overall higher performance of machine learning models over the commonly used Sequential Organ Failure Assessment (SOFA) and Quick SOFA (qSOFA) scoring systems at the time of sepsis onset.

Anemic syndrome is a frequent problem in intensive care units. The most probable etiology is the suppression of the erythropoietin response due to the direct effects of cytokines, as well as frequent blood sampling. Transfusions are not free of complications, therefore transfusion reactions are estimated to occur in 2% of the total packed red blood cells (pRBCs) transfused. In the past several years, several trials had tried to compare the restrictive with the more liberal use of transfusions, and they were found to be equally effective. Nosocomial pneumonia is the most common nosocomial infection in intensive care units; the prevalence is 47% with an attributive mortality of 33%. There are multiple risk factors for the development of nosocomial pneumonia. Colonization of the upper airways is the most important pathophysiological factor but there are other factors implicated like, sedation techniques, inappropriate use of antibiotics and recumbent positioning. A secondary analysis of the CRIT study describes transfusion therapy and its practices in the United States. They found that transfusion practice is an independent risk factor for the development of nosocomial pneumonia. This is a multicenter, prospective cohort study in different intensive care units in Colombia. A total of 474 patients were selected who had more than 48 hours of mechanical ventilation. The primary objective is to try to demonstrate the hypothetical relationship between the use of transfusions and nosocomial pneumonia. Secondly, we will try to determine which other factors are implicated in the development of pneumonia in intensive care units and describe the incidence of pneumonia and transfusion practices. Ventilator associated pneumonia is a primary problem in the intensive care unit, multiple factors have been associated with its presence in this study we try to explore the possible association between pneumonia and transfusion, describe all other factors associated with this, and the possible association with other nosocomial infections.

Background: Despite hybrid and vaccine-induced immunity, SARS-CoV-2 continues to cause disease. The characterization of humoral and cellular immune responses is essential for guiding prevention strategies and booster dose policies; Methods: A prospective cohort study was conducted with 131 hospitalized patients with confirmed COVID-19 in the Aburrá Metropolitan Valley, Colombia. Clinical and immunological data were evaluated on days 1–3, days 5–7, days 8–12, and 4–5 months after diagnosis. Humoral immunity was assessed by enzyme-linked immunosorbent assay (ELISA), chemiluminescent microparticle immunoassay (CMIA), and neutralization testing, and cellular immunity by CD4⁺/CD8 T-cell responses. Results: vaccinated patients had higher baseline levels of IgG and neutralizing antibody positivity than unvaccinated patients (ELISA 89.1% vs. 60.0%; CMIA 86.4% vs. 50.0%; neutralizing antibodies 88.2% vs. 65.0%), but cases of severe disease occurred in both groups. Adults aged ≥65 years had higher antibody positivity, but severe disease persisted. Mortality at 28 days was 7.6%, mainly among critically ill patients with comorbidities. Antibodies persisted at 4–5 months but were lower in those with severe acute disease. Those who received the booster dose showed stronger CD4⁺/CD8⁺ activation (notably against the Omicron variant) than unvaccinated/partially vaccinated patients. Conclusions: Vaccination improved humoral and cellular responses, but severe breakthrough infections still occurred, particularly in high-risk patients.