Vegetation size at diagnosis in infective endocarditis: Influencing factors and prognostic implications María Luaces a, ⁎ , Isidre Vilacosta b , Cristina Fernández b , Cristina Sarriá c , José Alberto San Román d , Catherine Graupner a , Iván J. Núñez-Gil b a Cardiología, Fuenlabrada University Hospital, Camino del Molino, 2. 28942 Fuenlabrada, Spain b Cardiology, Cardiovascular Institute, San Carlos University Hospital, Madrid, Spain c Infectious Diseases, La Princesa University Hospital, Madrid, Spain d Cardiology, ICICOR, Valladolid University Hospital, Valladolid, Spain Received 30 January 2008; Accepted 3 May 2008 Available online 9 August 2008 Abstract The role of vegetation as the key feature of infective endocarditis is universally recognized. Nowadays, the wide availability of transesophageal echocardiography has made of it the most employed technique to establish the diagnosis by visualizing vegetations. However, the factors which influence the size of vegetation when first detected are not clearly determined. Furthermore, there is considerable controversy regarding the prognostic implications of the size of vegetation. This is of paramount significance to early identify patients at high risk for complications, which might benefit from aggressive attitudes. We present a study based on TEE. Our results show that the size of vegetation at admission is mostly determined by anatomical and not microbiological factors, and the prognostic influence of vegetations on the risk of embolisms, need of surgery, persistent infection and septic shock. © 2008 Elsevier Ireland Ltd. All rights reserved. Keywords: Endocarditis; Transesophageal echocardiography; Prognosis ⁎ Corresponding author. Tel.: +34 91 600 6455; fax: +34 91 600 6186. E-mail address: mluaces.hflr@salud.madrid.org (M. Luaces). 1. Introduction Despite all advances in diagnostic tools and therapeutic armamentarium, infective endocarditis (IE) still remains a serious entity with high in-hospital mortality [1,2]. The vegetation is the cornerstone of the diagnosis and echocardiography is used to detect it. Notwithstanding, some controversial issues are pending about the determi- nants of vegetation size. Furthermore, the prognostic value of vegetation size is open to debate. Many efforts have been made to early identify a high-risk subgroup of patients which might benefit from aggressive attitudes [3,4]. Different studies yielded varied results, due to a scarce number of patients, use of transthoracic echocar- diography, or different diagnostic criteria [5–9]. Taking all these considerations in mind, we conducted a study based on the first transesophageal echocardiogram (TEE) with a double aim: 1) to know, if any, the determinant factors of vegetation size at admission; 2) to establish the prognostic influence of vegetation size at admission in adverse events and final outcomes. 2. Methods Clinical, multicenter, prospective study. All the patients admitted to 5 tertiary hospitals meeting criteria for definite IE with vegetations found and measured in the first TEE were prospectively included [10]. For each case, epidemiological, clinical, microbiological and echocardiographic data were collected, along with the clinical course, adverse events, need of cardiac surgery and mortality. Baseline characteristics were related to vegetation size by single univariate analysis. In order to find the factors determining vegetation size at admission, variables shown to have a statistical significance by univariate analysis and those clinically relevant were included to build a linear regression model. Prognostic implications of vegetation size were addressed by multivariate analysis. Events included need of surgery, embolisms, persistent infection, septic shock, and mortality (Table 1). 3. Results Our clinical cohort consisted of 330 episodes. Of those, 281 were left-sided IE, and 49 were right-sided IE. Most cases involved the mitral valve (41.8%). There were 93 prosthetic IE. Mean vegetation size at admission was 14.45 ± 7.66 mm (range, 0.5 to 47 mm). By univariate analysis, we found larger 76 Letters to the Editor