Incidence and outcomes of prosthetic valve endocarditis in adults with congenital heart disease.
Hsu AR, Karnakoti S, Abdelhalim AT, Miranda WR, Connolly HM, Dearani JA, DeSimone DC, Egbe AC. Am Heart J. 2025 Apr;282:125-133. doi: 10.1016/j.ahj.2025.01.005. Epub 2025 Jan 17.PMID: 39826702
Take-home Points
- Prosthetic valve endocarditis (PVE) occurred in 2.7% of adults with congenital heart disease (CHD) who had prosthetic valves.
- The incidence rate of PVE was 5.2 events per 1000 patient-years.
- Key risk factors included male sex, younger age, type 2 diabetes, multiple prosthetic valves, and use of the Melody transcatheter valve.
- PVE was associated with more than a twofold increase in all-cause mortality.
Commentary from Dr. Mathias Posner, section editor of ACHD Journal Watch:
Introduction
Adults with congenital heart disease have an increased risk of developing infective endocarditis, a risk that is further amplified following prosthetic valve implantation. While it is known that this subpopulation is particularly vulnerable, a clearer understanding of the clinical characteristics and outcomes of PVE in CHD patients is necessary. This study aimed to determine the incidence, risk factors and outcomes of PVE in adults with CHD.
Study Design
This was a retrospective cohort study encompassing adult CHD patients (aged ≥18) with prior prosthetic valve implantation, who received care at Mayo Clinic between 2003 and 2023. The cohort consisted of 9161 patients, of whom 3150 had prosthetic valves. Patients who developed PVE were designated as the PVE group (n=86), while the rest served as the reference group. Diagnosis of PVE was made according to the 2023 European Society of Cardiology modified diagnostic criteria of infective endocarditis. The study assessed incidence rates, risk factors, mortality outcomes, and PVE-related complications over a mean follow-up of approximately 11.6 years.
Key Findings
A prosthetic valve was present in 34% of the entire cohort. Most patients with a prosthetic valve had moderate or complex CHD (90%). Of all prosthetic valves implanted, 59% were bioprosthetic valves, while 41% were mechanical valves. 5% of the patients with a prosthetic valve received a transcatheter valve. The most common prosthesis positions were pulmonary (42%) and aortic (40%) positions, with 12% of all patients with a prosthetic valve having received more than one valve.
PVE occurred in 2.7% of CHD patients with prosthetic valves, corresponding to an incidence rate of 5.2 events per 1000 patient-years. The cumulative 10-year incidence of PVE was 5.3%. The 10-year cumulative incidence of PVE was higher in patients with left-sided compared to right-sided surgical prosthetic valves (6.2% versus 4.1%, P = 0.04) but was similar between patients with surgical bioprosthetic versus mechanical valves (5.4% versus 5.1%, P = 0.7). Within the group of patients with a bioprosthetic valve, patients with a Melody valve had a higher 5-year cumulative incidence of PVE (14.8% vs 2.6%, P < 0.001).
Mean age at the time of PVE diagnosis was 35 ± 9 years, and the average interval between prosthetic valve implantation and PVE was 91 ± 27 months.
Multivariable analysis revealed that younger age, male sex, type 2 diabetes, presence of multiple prosthetic valves, and Melody valve use were independent predictors of PVE. Patients with PVE had over twice the risk of all-cause mortality compared to those without PVE. Complications occurred in nearly half of the PVE patients, including septic emboli and perivalvular abscesses. The 30-day, 1-year, and 5-year mortality rates following PVE diagnosis were 1.6%, 12%, and 15%, respectively.
Strengths
This study represents one of the largest single-center cohorts focused on PVE in adults with CHD. It includes extensive follow-up data and detailed clinical, echocardiographic, and procedural information. The rigorous use of time-to-event analyses and multivariable models strengthens the validity of the findings. Moreover, the study provides insight into real-world outcomes and management strategies within a specialized tertiary care setting.
Limitations
The retrospective, single-center design introduces potential for selection and ascertainment bias. The high prevalence of moderate and complex CHD lesions in this cohort may not reflect the broader spectrum of CHD severity.
Discussion
The study highlights a substantial burden of PVE among CHD patients with prosthetic valves, with a 10-year incidence of PVE of 5.3%. It confirms previous findings regarding high-risk valve types, particularly the Melody valve, and adds evidence for newly identified risk factors such as younger age and multiple prostheses. Unlike previous studies, this analysis found no significant difference in PVE risk between surgical bioprosthetic and mechanical valves, a finding that warrants further investigation. The findings emphasize the importance of early recognition, prevention, and a multidisciplinary approach to management.
Conclusion
Prosthetic valve implantation in adults with CHD significantly increases the risk of endocarditis. Key risk factors for PVE include younger age, male sex, type 2 diabetes, multiple prosthetic valves, and use of Melody valves. PVE is an independent predictor of mortality. These findings underline the need for improved risk stratification tools and preventative strategies to mitigate the incidence and impact of PVE in this growing population.
