Examinando por Autor "del Castillo, Alberto"

Mostrando 1 - 2 de 2
  • Miniatura
    Ítem
    Activity and Outcomes of a Multidisciplinary Sepsis Unit: Fifty Thousand Consultations over Thirteen Years
    (MDPI, 0030-07-26) Borges-Sa, Marcio; Martin-Martinez, Elena; Aranda, Maria; Socias, Antonia; Tejada, Sofia; del Castillo, Alberto; Garau, Margarita; Pérez de Amezaga, Luis; Mena, Joana; Giglio, Andres
    Background. Sepsis is a leading, time-dependent cause of in-hospital death, and guidelines now prioritize organized care such as code-sepsis protocols. Yet little is known about the long-term activity, organization and outcomes of permanent, hospital-scale programs. We characterized the activity of a hospital-wide multidisciplinary sepsis unit (MSU) over thirteen years and evaluated mortality trends. Methods. Retrospective cohort study using the registry of a hospital-wide MSU (2011–2023). The analysis unit was the sepsis event. Activity, referral pathways, interventions, follow-up and mortality were analyzed; temporal trends used multivariable logistic regression and standardized mortality ratios (SMR). Results. The unit attended 10,874 patients across 15,723 events and 50,925 consultations (median age 67; 59.3% men), referred mainly through the sepsis code (45.0%) and, increasingly, proactive early-warning detection (15.2%). An antibiotic change was recommended in 57.2% of events, and 89.8% of all suggested changes were implemented. Follow-up was sustained (median 3 visits; 61.8%). In-hospital mortality was 7.6%, and 11.8% in protocol-confirmed sepsis. Crude mortality did not fall, but severity rose (organ dysfunctions 1.67 to 2.72; septic shock 8 to 13%; both p < 0.001) with stable age; after adjustment the calendar-year effect disappeared (adjusted OR 1.00/year, p = 0.70) and the SMR stayed near 1.0, including pandemic years. Age, ICU admission (OR 2.65) and organ-dysfunction count (OR 1.37) independently predicted death. Conclusions. Over thirteen years, this hospital-wide MSU delivered large-scale, multi-channel, longitudinal care to an increasingly severe population while maintaining stable risk-adjusted mortality, supporting the long-term feasibility of a model integrating early detection, multidisciplinary decision-making, antimicrobial stewardship and follow-up.
  • Miniatura
    Ítem
    Service-Specific Heterogeneity in Sepsis Variable Significance and Machine Learning Model Performance: A Stratified Analysis of the BIAlert Cohort
    (MDPI, 2026-06-24) Borges-Sa, Marcio; Macias-Fassio, Eric; Delgado, Alejandro; Santiago Salas Sosa; Aranda, María; Socias, Antonia; del Castillo, Alberto; Giglio, Andrés
    Background/Objectives: Sepsis detection relies on clinical variables and scoring systems assumed to perform uniformly across hospital settings. However, sepsis phenotype distributions shift between clinical environments, suggesting that variable importance may be setting dependent. This study aimed to quantify service-specific variability in the discriminatory capacity of clinical variables for sepsis detection and to evaluate whether this heterogeneity translates into differential performance of machine learning models compared to traditional clinical scoring systems. Methods: This stratified sub-analysis of the BIAlert Sepsis cohort (203,755 patients; 11,864 sepsis episodes, 2014–2018) evaluated 61 structured quantitative variables across nine hospital services (≥90 sepsis episodes each). Within each service, the Mann–Whitney–Wilcoxon test (p < 0.01, Holm-corrected) assessed differences between septic and non-septic episodes. Five machine learning models (Random Forest/BIAlert, XGBoost, CatBoost, SVM, Neural Network) and three clinical rules (NEWS, SIRS, qSOFA) were evaluated globally and stratified across four clinical environments. Results: The proportion of significant variables ranged from 95.1% in the Emergency Department (58/61) to 37.7% in the Intensive Care Unit (23/61). Lactate was the only universally significant variable (9/9 services). Clinical scoring systems collapsed in Critical Care (qSOFA and NEWS AUC 0.459). BIAlert maintained the highest AUC across all environments (0.975–0.857). The Friedman test confirmed significant differences (χ2 = 28.00, p < 0.001), with BIAlert achieving a mean rank of 1.0. Conclusions: The discriminatory capacity of clinical variables for sepsis detection is not uniform across hospital services. ML models, particularly BIAlert, maintained robust performance where fixed-rule scoring systems failed.