K-Means Clustering for Shock Classification in Pediatric Intensive Care Units.

Shock is described as an inadequate oxygen supply to the tissues and can be classified in multiple ways. In clinical practice still, old methods are used to discriminate these shock types. This article proposes the application of unsupervised classification methods for the stratification of these patients in order to treat them more appropriately. With a cohort of 90 patients admitted in pediatric intensive care units (PICU), the k-means algorithm was applied in the first 24 h data since admission (physiological and analytical variables and the need for devices), obtaining three main groups. Significant differences were found in variables used (e.g., mean diastolic arterial pressure p < 0.001, age p < 0.001) and not used for training (e.g., EtCO2 min p < 0.001, Troponin max p < 0.01), discharge diagnosis (p < 0.001) and outcomes (p < 0.05). Clustering classification equaled classical classification in its association with LOS (p = 0.01) and surpassed it in its association with mortality (p < 0.04 vs. p = 0.16). We have been able to classify shocked pediatric patients with higher outcome correlation than the clinical traditional method. These results support the utility of unsupervised learning algorithms for patient classification in PICU.

Oxigenoterapia de alto flujo: el soporte respiratorio no invasivo sale de la UCIP. ¿Es una alternativa eficiente? / High-flow oxygen therapy: Non-invasive respiratory support goes out of the PICU. Is it an efficient alternative?

No disponible

Identification of a panel of serum protein markers in early stage of sepsis and its validation in a cohort of patients.

BACKGROUND: Sepsis is a life-threatening illness with a challenging diagnosis. Current serum biomarkers are not sensitive enough for diagnosis. With the aim of finding proteins associated with sepsis, serum protein profile was compared between patients and healthy donors and serum classical inflammatory proteins were analyzed in both groups. METHODS: Serum protein profiles were characterized by two-dimensional electrophoresis (2DE). Identification of the proteins was carried out by mass spectrophotometry and their validation was performed by Enzyme-Linked-ImmunoSorbent Assay (ELISA) in a cohort of 85 patients and 67 healthy donors. Seven classical inflammatory proteins were analyzed in the same cohort by ELISA: interleukin-2 receptor α-chain (sCD25), scavenger receptor cysteine-rich-type-1 (sCD163), tumor-necrosis factor receptor superfamily-member-6 (sFas), hemeoxigenase-1 decycling (HO-1), interleukin-6 (IL-6), interleukin-18 (IL-18) and intercellular adhesion-molecule-1 (sICAM-1). RESULTS: After 2DE, 20 significantly differently expressed spots were identified by mass spectrometry analysis, revealing deregulation of six different proteins upon sepsis and 50% were validated by ELISA: Antithrombin-III (AT-III), Clusterin (CLUS) and Serum amyloid A-1 (SAA-1). Serum concentration of AT-III and CLUS was significantly lower in patients' serum, whereas SAA-1 showed higher values in septic patients. Serum concentration of the seven inflammatory proteins was significantly increased in septic patients. Functional analysis of the ten deregulated proteins revealed an enrichment of proteins related mainly to the activation of the immune response. CONCLUSION: We have identified a panel of ten potential sepsis marker proteins biologically connected and validated in a large number of patients, whose analysis could be considered as a complementary tool for the diagnosis of sepsis.