RESUMO
Objective@#Community-acquired pneumonia (CAP) in older patients is a potentially life-threatening infection with a poorprognosis. Therefore, is important to predict the mortality rate of CAP for older patients. This study examined the effectsof predictive increases on CAP mortality by adding a biomarker to known CAP severity prediction tools. @*Methods@#A retrospective analysis of information was conducted on patients older than 65 years, who were treated withCAP in five emergency departments from October 2016 to February 2017. The primary outcome was the 28-day mortality.The following were calculated for each patient: qSOFA (quick Sequential Organ Failure Assessment), A-DROP (Age,Dehydration, Respiratory failure, Orientation, blood Pressure), CURB-65 (Confusion, Urea level, Respiratory rate, Bloodpressure, age≥65 years), SMART-COP (Systolic blood pressure, Multilobar infiltrates, Albumin, Respiratory rate,Tachycardia, Confusion, Oxygen and pH), NLR (neutrophil:lymphocyte ratio), PLR (platelet:lymphocyte ratio), and CAR(high-sensitivity C-reactive protein:albumin ratio). The prognostic value for the 28-day mortality was determined by multivariatelogistic regression analysis. @*Results@#The 28-day mortality was 12.0% of 693 CAP patients. Multivariate logistic regression analysis showed that lactate(odds ratio [OR], 1.589; P<0.001) and CAR (OR, 1.208; P=0.006) were correlated with the 28-day mortality. NLR(OR, 1.00; P=0.983) and PLR (OR, 1.00; P=0.784) were not correlated. The area under curve (AUC) was significant asCAR 0.649, lactate 0.737, and SMART-COP 0.735 (P<0.001), and the AUC of lactate+SMART-COP increased significantlyto 0.784 compared to SMART-COP (P=0.014). @*Conclusion@#A combination of lactate and SMART-COP can be used as a tool to assess the severity of older hospitalizedCAP patients who visited emergency departments.