Endocarditis in adults with congenital heart disease: new answers–new questions

This editorial refers to ‘Incidence, risk factors, and predictors of infective endocarditis in adult congenital heart disease: focus on the use of prosthetic material’†, by J.M. Kuijpers et al., on page 2048. Patients with congenital heart disease (CHD) represent an evolving and growing population afflicted by numerous late complications. Unfortunately, despite the great technological advances of the last decades, these patients are—by and large—not cured and continue to present with late sequelae of disease.1 , 2 One of the main complications encountered in the late follow-up of adult patients with congenital heart disease (ACHD) remains infective endocarditis. The condition is often challenging to manage and associated with considerable morbidity and mortality. Over the last few decades, the mode of presentation has changed in various regards. First, infective endocarditis nowadays primarily affects patients with repaired congenital heart defects, especially those with prosthetic valves or other forms of implanted prosthetic material. This is in contrast to historical cohorts of patients with largely uncorrected CHD, cyanotic conditions, or rheumatic heart disease. Secondly, a shift in the underlying microbiological organisms causing endocarditis has been reported over the last decades. Presumably due to the increasing number of invasive procedures and hospitalizations required in this population, a trend towards more cases caused by Staphylococcus rather than Streptococcus species is seen.3 Although mortality has been reduced by improved antibiotic treatment options, it remains considerable even in the current era, ranging at ∼10–20%.4–6 The leading risk factors for mortality have been previously reported to be patient age, signs and symptoms of heart failure at presentation, cyanosis, a larger vegetation size (>20 mm), requirement for surgical therapy, and Staphylococcus aureus as a causative agent.7 Considering the high morbidity and mortality associated with infective endocarditis, prevention has traditionally been given a lot of attention and has motivated the historic practice of recommending antibiotic endocarditis prophylaxis to all patients with CHD. This prudent, proactive but largely empiric approach has been significantly revised during the last decade.8 Various factors beyond the mere lack of prospective data, including concerns about the efficacy of antibiotic prophylaxis, frequent low-grade bacteraemia during routine daily activities (mainly tooth brushing), concerns about increasing antibiotic resistance among microorganisms, and the common lack of an identifiable index procedure pre-dating endocarditis have motivated these changes in recommendations. Not surprisingly, however, the paradigm change from an extremely proactive to a largely restrictive use of antibiotic prophylaxis has created some confusion among patients and practitioners alike. As a consequence, further data are required both to investigate the impact of revised recommendations on the incidence of endocarditis as well as to delineate risk factors for endocarditis in contemporary ACHD patients. Based on data from the Dutch National Register for ACHD (Concor registry), Kuijpers and co-workers investigated the risk factors and predictors of infective endocarditis in contemporary ACHD patients.9 Overall, 14 224 ACHD patients were included in this retrospective analysis of prospectively collected data after 2001. The reported absolute incidence of endocarditis in this ACHD population, with a median age of 34 years, was low and is in broad agreement with previous studies. Infective endocarditis affected <1% of the ACHD patients, with an annual incidence of ∼1 per 1000 patient-years. The current data highlight, however, that endocarditis rates are likely to be higher in ACHD patients compared with children.10 In relative terms, of course, the risk of endocarditis is much higher in ACHD patients compared with the general population. Considering the anatomic location of the underlying heart defect, the study shows that ‘left heart’-sided disease is more common in ACHD patients with endocarditis compared with paediatric populations. The main findings of the investigation published in this issue of the journal are, however, (i) the fact that especially prosthetic valves (including valved conduits) carried a far greater risk of endocarditis compared with other types of prosthetic material and (ii) the results of the analysis of the time course of onset of endocarditis. Not only was the risk of developing endocarditis far greater in patients with prosthetic valves, but the study highlights that the increased risk of endocarditis associated with non-valve-related prosthetic material seems to be transient and limited to the first 6 months after the procedure. Based on their results, the authors also provide a risk score for endocarditis in ACHD based on clinical parameters. This score is likely to be of limited immediate clinical value but may be useful for further analyses including cost-effectiveness studies in this setting. Overall, the results of the study add credibility to current endocarditis guidelines, recommending antibiotic endocarditis prophylaxis to patients with prosthetic valves (indefinitely) and to those with other types of prosthetic material for the first 6 months post-operatively. Other risk factors identified in the current study are also already acknowledged by the guidelines and include complexity of disease and a history of previous endocarditis. Unfortunately—probably due to data limitations—the impact of cyanosis on the risk of endocarditis could not be clarified in the study by Kuijpers et al. In contrast to paediatric cohorts, the risk of endocarditis was higher in men, and male gender emerged as an independent predictor for developing endocarditis.3 , 10 The reasons for this finding remain speculative but may include—as discussed by the authors—general lifestyle factors and worse dental hygiene in males. The findings reported by Kuijpers et al. raise the obvious question of whether some types of valves/valved conduits are more prone to endocarditis compared with others. Unfortunately, the limited depth of the data precluded such analysis. From a clinical point of view, it will be especially important for further studies to clarify a possible difference in the risk of endocarditis between surgically and percutaneously implanted valves. The latter are increasingly used in this ageing population requiring a multitude of interventions during their life course. In addition, pulmonary atresia with ventricular septal defect and tetralogy of Fallot were two diagnoses identified as representing risk factors for endocarditis in the current study. Patients with these diagnoses frequently undergo percutaneous pulmonary valve implantation and may, therefore, be particularly prone to develop endocarditis. Currently, data on the risk of endocarditis associated with percutaneous pulmonary valve implantation is limited and the results are inconclusive. While some signals point towards a higher annual event rate in patients undergoing a Melody-type valve system implantation compared with surgical pulmonary valve replacement,11–13 robust data for other percutaneous valve types, especially the Edwards Sapien pulmonic system, are currently unavailable. Highlighting the various risk factors for endocarditis and incorporating clinical parameters in a formal model represents one of the main strengths of the study of Kuijpers et al. It emphasizes that comparing raw incidence rates is not appropriate. Rather, further studies need to compare endocarditis risk between surgical and transcatheter valves after accounting for haemodynamics, the underlying diagnosis, and patient demographics. Endocarditis risk in ACHD appears to be related to the type of repair procedure. Patients with prosthetic valves (including valved conduits) are at particular risk. Whether risk is even higher in percutaneous valves is under investigation. Panel A: schematic of Tetralogy of Fallot (ToF). Panel B: ToF with patch repair. Panel C: ToF repair with a valved conduit. Panel D: ToF repair with a percutanously implanted valve within a failed conduit. Patients with CHD are also prone to develop arrhythmias and sudden cardiac death. As a consequence, device therapy is gaining importance in this setting. With increasing use of implantable pacemakers and defibrillators comes increased complications, including lead infection and endocarditis.14 The problem is aggravated by inherent challenges and concerns regarding lead extraction in patients with CHD.14 , 15 Unfortunately, the study by Kuijpers et al. provides limited insight into the problem of device-related endocarditis. Taken together, the current study is largely confirmative, supporting current endocarditis prophylaxis guidelines and clinical practice. The authors have to be commended, however, as their study adds greatly to the literature by studying a large cohort of contemporary ACHD patients and for establishing a formal model for endocarditis risk. The limited depth of the data is compensated by the broad patient population. It remains for further focused studies to fill the gaps with regard to endocarditis related to transcatheter bioprosthetic valves, electrical devices, or prosthetic material in close proximity to residual haemodynamic defects. Conflict of interest: none declared.