Índice SatO2/FiO2 versus PaO2/FiO2 para predecir mortalidad en pacientes con COVID-19 en un hospital de altura

• Autores: Marmanillo Mendoza, Gabriela, Zuñiga Manrique, Rosario Elsa Jesús, Cornejo Del Valle, Oscar Eduardo, Portilla Canqui, Luis Alberto
• Formato: artículo
• Fecha de Publicación: 2022
• Institución: Colegio Médico del Perú
• Repositorio: Acta Médica Peruana
• Lenguaje: español
• OAI Identifier: oai:amp.cmp.org.pe:article/2033
• Enlace del recurso: https://amp.cmp.org.pe/index.php/AMP/article/view/2033
• Nivel de acceso: acceso abierto
• Materia: Covid-19; Mortalidad Hospitalaria; Valor Predictivo de las Pruebas; Indice de Severidad de la Enfermedad; Consumo de Oxígeno; Hospital Mortality; Precctive Value of Tests; Severity of Illness Index; Oxygen Consumption

Objective: To compare the SatO2/FiO2 rate against PaO2/FiO2 rate for predicting mortality in COVID-19 patients in a high altitude hospital. Materials and methods: This is an observational, retrospective, and cross-sectional study, where sensitivity(S), specificity (Sp), positive predictive value (PPV), and negative predictive value (NPV) for both rates were compared. Results: The average age of COVID-19 patients was 62.2 ± 12.8 years. Three quarters (73.7 %) of all patients were male, most frequent comorbidities were diabetes mellitus (21.6 %), and high blood pressure (27.2 %). On the other hand, average SatO2/FiO2 rate was 234.9 ± 121.9, and PaO2/FiO2 was 169.7 ± 84, and 30.5 % of all patients died. A significant relationship between the aforementioned rates and mortality was evidenced. The cutoff point for SatO2/FiO2 was 114.4, with 43.1 % sensitivity, 91.2 % specificity, 68.3 % positive predictive value, and 78.5 % negative predictive value. The cutoff point for PaO2/FiO2 was 134.2, with 63.1 % sensitivity, 67.6 % specificity, 46.1 % positive predictive value, and 80.7 % negative predictive value. Conclusion: Both rates are predictors for mortality; however, SatO2/FiO2 showed better results with respect to the following parameters: sensitivity, specificity, positive predictive value, and negative predictive value.