We thank Tattevin and colleagues for their comments and we take this opportunity to clarify a few important points.
First, the very low in-hospital (4.4%) and 1-year (8.2%) mortality observed in our study1 were not the rates of death of all the cohort of patients hospitalized in our institution from 2002 to 2006. As indicated in our article, the mortality rates were obtained in a selected sample of patients who had been treated by a multidisciplinary medical team according to a standardized protocol for an episode of IE caused by common (streptococci, enterococci, and staphylococci) or unknown pathogens. During this period, 112 patients were not included in the study because of the following reasons: uncommon pathogens (n = 39), the patient was referred to our department at the end of the appropriate antibiotic therapy (n = 43), and no sign of IE on valvular histopathologic examination after early surgery (n = 30). The exclusion of patients with no sign of IE on valvular histopathologic examination despite a definite diagnosis according to the Duke criteria2 could be largely criticized. However, when those patients were included in the analysis of IE episodes caused by common or unknown pathogens, the in-hospital mortality was 12.3% without the protocol during the period 1 (1991-2001) and 5.3% with the protocol during the period 2 (2002-2006) (P = .01). In addition, when we pooled all the patients hospitalized at La Timone Hospital in the cardiology department for a definite IE whatever the causative pathogen, the histopathologic results, and the type of management (n = 453), the rate of 1-year death was 21.7% during the period 1 and 14.8% during the period 2 (P = .047). Similarly, the total in-hospital mortality of patients hospitalized during these periods was 13.8% during period 1 and 7.7% during period 2 (P = .03) (Table). Thus, the 1-year and in-hospital mortality in our cardiology department remains lower than that observed in all other series and is related to our multidisciplinary approach and standardized protocol. In their letter, Tattevin and colleagues argue that the management of IE by a multidisciplinary team did not allow for a decrease in IE-related mortality at their institution. What we tried to demonstrate in our work is not only that a multidisciplinary medicosurgical team is important to manage IE but also that each step of the management of the disease should be standardized with a survey in compliance, with codified and standardized therapeutic indications. In our experience, the implementation of a multidisciplinary team in 1991 did not allow for a significant reduction in mortality in our department for 10 years, probably because of the lack of a standardized management. Now, the reduction of mortality was noticed in such a department at the same time as the implementation of the standardized protocol and increased compliance.