In the series by de Sousa et al.,9 the mean age of patients was 55.5 ± 12.1 years, with a large predominance of males. This population is younger than that observed in EURO-ENDO and this may be related to the high number of intravenous drug users and to the inclusion of some relatively old studies with possibly younger patients. In three one-year population-based surveys conducted in 1991, 1999, and 2008,5,6 the total IE incidence remained stable over time, but mean age progressively increased from 57.9 ± 16.6 years in 1991 to 61.6 ± 16.3 years in 2008. Similarly, the incidence of cardiac device-related IE (CDRIE) is increasing. Although observed in only 6.1% of cases in the Portuguese series, it is now much more frequent (9.9%) in EURO-ENDO,7 reflecting the increasing burden of IE in this older population. Conversely, the relatively high incidence of enterococci (11.7 to 15% of patients) reported by the current study corroborates the current increasing incidence of this microorganism (15.8%) reported in EURO-ENDO, again reflecting the increasing age of this population.

One of the most important changes in the diagnosis of IE is the implementation of non-echocardiographic imaging techniques, mainly cardiac computed tomography (CT) and 18F-fluorodeoxyglucose positron emission tomography / computed tomography (18F-FDG PET/CT).8,10 These techniques are more and more frequently used in IE, particularly in the field of suspected prosthetic valve IE. For instance, [18F]FDG PET/CT was performed in 16.6% of patients in EURO-ENDO (7), and multislice CT in 53.1% of patients. The low use of these techniques in Portuguese centers probably reflects the difficult access to nuclear imaging in some Portuguese cities. But more importantly, as reported by the authors, the main explanation is the inclusion of patients before the publication of the 2015 ESC guidelines,8 in which these new imaging tools were included for the first time in the diagnostic and prognostic stratification of IE. Nevertheless, the use of [18F]FDG PET/CT varied largely between countries across the world, ranging from 0.5 to 33.9% of IE cases in EURO-ENDO.7

At Portuguese centers, the reported surgical rate ranged from 3.1 to 52%, with a trend toward a higher surgical rate in the most recent series, with three main indications for surgery: heart failure (50 and 74%), uncontrolled infection (27.3 and 46.8%) and embolization (13.6 and 15.6%). These results support current ESC guidelines,8 in which early surgery is recommended as soon as one of these three complication occurs. In EURO-ENDO,7 we focused on the implementation of the ESC guidelines in clinical practice and found that, although theoretical indication for cardiac surgery was present in 69.3% patients, surgery was eventually performed during hospitalization in only 51.2% patients. Failure to perform surgery when indicated was associated with the worst prognosis in EURO-ENDO, emphasizing the benefit of an aggressive surgical strategy in these patients. The relatively high in-hospital and long-term mortality reported at Portuguese centers probably reflects the low use of surgery (29.8%) in some relatively old reported series.

The report from de Sousa et al. (9), although limited by its retrospective nature and other issues reported by the authors, has the merit of emphasizing the main messages of EURO-ENDO, including the higher frequency of IE among the elderly, in prosthetic IE and in CDRIE, the value of new imaging techniques, still underused in several countries, the need for early surgical management, and the still poor prognosis of IE, particularly when surgery is indicated and cannot be performed. Endocarditis teams8 are now in existence at several hospitals across Europe. They will help implementing the ESC guidelines in the future, both in terms of diagnostic and therapeutic strategies, and improving the prognosis of this devastating disease.

The authors have no conflicts of interest to declare.