RESUMEN
PURPOSE: The aim of this study was to analyse the characteristics of patients with IE in three groups of age and to assess the ability of age and the Charlson Comorbidity Index (CCI) to predict mortality. METHODS: Prospective cohort study of all patients with IE included in the GAMES Spanish database between 2008 and 2015.Patients were stratified into three age groups:<65â¯years,65 to 80â¯years,andâ¯≥â¯80â¯years.The area under the receiver-operating characteristic (AUROC) curve was calculated to quantify the diagnostic accuracy of the CCI to predict mortality risk. RESULTS: A total of 3120 patients with IE (1327â¯<â¯65â¯years;1291 65-80â¯years;502â¯≥â¯80â¯years) were enrolled.Fever and heart failure were the most common presentations of IE, with no differences among age groups.Patients ≥80â¯years who underwent surgery were significantly lower compared with other age groups (14.3%,65â¯years; 20.5%,65-79â¯years; 31.3%,≥80â¯years). In-hospital mortality was lower in the <65-year group (20.3%,<65â¯years;30.1%,65-79â¯years;34.7%,≥80â¯years;pâ¯<â¯0.001) as well as 1-year mortality (3.2%, <65â¯years; 5.5%, 65-80â¯years;7.6%,≥80â¯years; pâ¯=â¯0.003).Independent predictors of mortality were ageâ¯≥â¯80â¯years (hazard ratio [HR]:2.78;95% confidence interval [CI]:2.32-3.34), CCIâ¯≥â¯3 (HR:1.62; 95% CI:1.39-1.88),and non-performed surgery (HR:1.64;95% CI:11.16-1.58).When the three age groups were compared,the AUROC curve for CCI was significantly larger for patients aged <65â¯years(pâ¯<â¯0.001) for both in-hospital and 1-year mortality. CONCLUSION: There were no differences in the clinical presentation of IE between the groups. Ageâ¯≥â¯80â¯years, high comorbidity (measured by CCI),and non-performance of surgery were independent predictors of mortality in patients with IE.CCI could help to identify those patients with IE and surgical indication who present a lower risk of in-hospital and 1-year mortality after surgery, especially in the <65-year group.
Asunto(s)
Factores de Edad , Comorbilidad , Endocarditis/mortalidad , Adulto , Anciano , Anciano de 80 o más Años , Área Bajo la Curva , Bases de Datos Factuales , Endocarditis/etiología , Femenino , Insuficiencia Cardíaca/mortalidad , Mortalidad Hospitalaria , Humanos , Masculino , Persona de Mediana Edad , Modelos de Riesgos Proporcionales , Estudios Prospectivos , Curva ROC , Factores de Riesgo , España/epidemiología , Infecciones Estafilocócicas/mortalidadRESUMEN
BACKGROUND AND OBJECTIVE: It is well-known the relationship between Streptococcus bovis (S. bovis) bacteremia and colon cancer, liver cirrhosis and others neoplasms. However, a study protocol to rule out these underlying diseases has not been carried out yet. Our objective was to describe S. bovis bacteremia and associated diseases. PATIENTS AND METHOD: Multicenter, retrospective cohort study. S. bovis bacteremias episodes between 2001 and 2009 were included. Mean variables: colon neoplasm, non-colonic neoplasm or liver cirrhosis. Epidemiologist aspects, bacteremia related variables, personal and familiar history and clinical and analytical data were collected. RESULTS: Ninety three patients were included. One out of four individuals had a colon neoplasm. Fifty seven per cent were concomitant cases with bacteremia and six cases were diagnosed after bacteremia (time bacteremia-diagnosis of neoplasm [months], median [Q1-Q3], 2.6 [1-11]). Fourteen (15%) patients were diagnosed with any non-colonic neoplasm (mainly biliary and pancreatic [6 cases] or esophagus-gastric [3 cases]). There were three patients (21%) with concomitant bacteremia non-colonic neoplasm and two after it (1.2 and 10.4 months). Twenty-one (23%) patients suffered from liver cirrhosis. CONCLUSIONS: Patients with S. bovis bacteremia must undergo a study designed to rule out underlying diseases. We suggest that this study should include: a colonic evaluation, ideally by colonoscopy, a liver evaluation by serum chemistry, an abdominal ultrasound scan or a method of liver fibrosis assessment, a gastroscopy and an evaluation of biliary and pancreatic areas by magnetic resonance imaging.