João Gabriel Rosa Ramos1, Rogério da Hora Passos2, Mauricio Brito Teixeira3, Andre Luiz Nunes Gobatto4, Rafael Viana Dos Santos Coutinho5, Juliana Ribeiro Caldas6, Suzete Farias da Guarda7, Michel Pordeus Ribeiro8, Paulo Benigno Pena Batista9. 1. Intensive Care Unit, Hospital Sao Rafael, Salvador, Brazil; Palliative Care Team, Hospital Sao Rafael, Salvador, Brazil. Electronic address: joao.ramos@hsr.com.br. 2. Intensive Care Unit, Hospital Sao Rafael, Salvador, Brazil; Internal Medicine, Hospital Sao Rafael, Salvador, Brazil; Intensive Care Unit, Hospital Portugues, Salvador, Brazil. 3. Internal Medicine, Hospital Sao Rafael, Salvador, Brazil; Nephrology Department, Hospital Ana Nery, Salvador, Brazil; Medical School, Escola Bahiana de Medicina, Salvador, Brazil; Medical School, Universidade do Estado da Bahia, Salvador, Brazil. 4. Intensive Care Unit, Hospital Sao Rafael, Salvador, Brazil; Internal Medicine, Hospital Sao Rafael, Salvador, Brazil; Medical School, Universidade Salvador, Salvador, Brazil; Intensive Care Unit, Hospital da Cidade, Salvador, Brazil. 5. Medical School, Universidade Federal da Bahia, Salvador, Brazil. 6. Intensive Care Unit, Hospital Sao Rafael, Salvador, Brazil; Medical School, Escola Bahiana de Medicina, Salvador, Brazil. 7. Intensive Care Unit, Hospital Sao Rafael, Salvador, Brazil; Internal Medicine, Hospital Sao Rafael, Salvador, Brazil; Medical School, Universidade Federal da Bahia, Salvador, Brazil; Medical School, Uniao Metropolitana para o Desenvolvimento da Educação e Cultura UNIME, Lauro de Freitas, Brazil. 8. Intensive Care Unit, Hospital Sao Rafael, Salvador, Brazil; Medical School, Uniao Metropolitana para o Desenvolvimento da Educação e Cultura UNIME, Lauro de Freitas, Brazil; Intensive Care Unit, Hospital Sagrada Familia, Salvador, Brazil. 9. Intensive Care Unit, Hospital Sao Rafael, Salvador, Brazil; Medical School, Escola Bahiana de Medicina, Salvador, Brazil; Medical School, Uniao Metropolitana para o Desenvolvimento da Educação e Cultura UNIME, Lauro de Freitas, Brazil. Electronic address: paulo.benigno@hsr.com.br.
Abstract
OBJECTIVES: Sepsis identification in older patients is challenging. We evaluated the performance of qSOFA across different age groups of patients with suspected infection outside the intensive care unit (ICU). METHODS: Retrospective cohort in a tertiary hospital in Brazil, from January 2016 to December 2016. Outcomes were hospital mortality, ICU admission and bacteremia. Performance of qSOFA was compared over three age groups: (1) reference: ≤65 years, (2) old: 65 to 79 years and (3) very old: ≥80 years. RESULTS: There were 420 patients in the study, of which 259 (61.7%) were ≤65 years, 80 (19%) were 65 to 79 years and 81 (19.3%) were ≥80 years. Old and very old patients had higher qSOFA scores and lower SIRS scores. Overall, qSOFA ≥2 was associated to hospital mortality [OR (95% CI) = 5.8 (3.3-10.4), p < 0.001], ICU admission [OR (95% CI) = 2.7 (1.6-4.6), p < 0.001] and bacteremia [OR (95% CI) = 3.1 (1.7-5.8), p < 0.001]. Those associations were stronger in old and very old patients. qSOFA and SIRS demonstrated overall AUROCs for hospital mortality of 0.72 and 0.50, respectively. CONCLUSION: qSOFA demonstrated good overall accuracy and was more strongly associated to outcomes in old and very old patients, when compared to younger patients.
OBJECTIVES: Sepsis identification in older patients is challenging. We evaluated the performance of qSOFA across different age groups of patients with suspected infection outside the intensive care unit (ICU). METHODS: Retrospective cohort in a tertiary hospital in Brazil, from January 2016 to December 2016. Outcomes were hospital mortality, ICU admission and bacteremia. Performance of qSOFA was compared over three age groups: (1) reference: ≤65 years, (2) old: 65 to 79 years and (3) very old: ≥80 years. RESULTS: There were 420 patients in the study, of which 259 (61.7%) were ≤65 years, 80 (19%) were 65 to 79 years and 81 (19.3%) were ≥80 years. Old and very old patients had higher qSOFA scores and lower SIRS scores. Overall, qSOFA ≥2 was associated to hospital mortality [OR (95% CI) = 5.8 (3.3-10.4), p < 0.001], ICU admission [OR (95% CI) = 2.7 (1.6-4.6), p < 0.001] and bacteremia [OR (95% CI) = 3.1 (1.7-5.8), p < 0.001]. Those associations were stronger in old and very old patients. qSOFA and SIRS demonstrated overall AUROCs for hospital mortality of 0.72 and 0.50, respectively. CONCLUSION: qSOFA demonstrated good overall accuracy and was more strongly associated to outcomes in old and very old patients, when compared to younger patients.