Introduction
Transcatheter device closure for the treatment of secundum atrial septal defect has become an established alternative to surgical repair in selected patients. However, eligibility for transcatheter closure depends on the defect size, adequacy of surrounding rims, and spatial relationship of the defect to adjacent cardiac structures. Careful patient selection is therefore essential, as suboptimal anatomical conditions may predispose patients to device-related complications. Reference Fraisse, Latchman and Sharma1,Reference Faccini and Butera2
Infective endocarditis is an uncommon complication following atrial septal defect device closure but may lead to severe systemic embolic events requiring urgent surgical management. Reference Amedro, Soulatges and Fraisse3 Although the risk of late infection is generally considered low after device implantation, cases occurring beyond the early post-procedural period continue to be reported.
Late-onset infective endocarditis after atrial septal defect device closure is rare, with most reported cases occurring within the early post-implantation period. However, data on very late device-related infective endocarditis and its underlying mechanisms remain limited. We report a case of device-related infective endocarditis presenting eight years after transcatheter atrial septal defect closure, focusing on mechanical irritation and adverse device–tissue interaction identified at surgery and their potential implications for long-term clinical management.
Case report
A 15-year-old boy presented to our department with a 10-day history of high-grade fever accompanied by chills. He had undergone transcatheter closure of a secundum atrial septal defect at another institution 8 years before admission. On presentation, his body temperature was 38.5°C, and laboratory findings were consistent with systemic infection. Computed tomography revealed splenic infarction, and physical examination demonstrated Janeway lesions on the toes, suggestive of peripheral septic embolisation.
Transthoracic echocardiography showed complete occlusion of the atrial septal defect with preserved biventricular function. A mobile vegetation-like mass measuring 10 mm × 9 mm was identified on the left atrial surface of the occluder device. Mild mitral and tricuspid regurgitation were present, without evidence of chamber enlargement or ventricular outflow tract obstruction.
Blood cultures obtained at another hospital were positive for Streptococcus pneumoniae, and intravenous antibiotic therapy was initiated according to the 2023 ESC guidelines for the management of infective endocarditis. The diagnosis of infective endocarditis was established based on modified Duke criteria, including positive blood cultures, echocardiographic evidence of vegetation, and systemic embolic manifestations. Given the diagnosis of device-related infective endocarditis complicated by multiple systemic embolic events, emergent surgical intervention was required. At the time of presentation, there was no documented history of recent pneumonia or other clinically apparent primary infectious focus. Due to the urgent need for surgical intervention, a comprehensive preoperative evaluation to identify a primary source of infection (including sinus, otologic, or ophthalmologic assessment) could not be fully performed.
The procedure was performed via a median sternotomy with cardiopulmonary bypass established using aortic and bicaval cannulation. Following a right atriotomy, the atrial septum was opened to access and remove the occluder device, permitting direct evaluation of the left atrial wall and adjacent structures. Intraoperatively, the occluder device showed relatively limited vegetation compared with the adjacent left atrial wall, where the predominant infectious involvement was observed (Figure 1). However, marked fibrotic thickening with associated vegetation was observed along the left atrial wall adjacent to the right lower pulmonary vein orifice. Although intraoperative photographs were not available, preoperative computed tomography demonstrated a mass-like lesion in this region, corresponding to the intraoperative findings (Figure 2). Extensive vegetation was identified along the affected right lower pulmonary vein orifice-adjacent left atrial wall, consistent with chronic mechanical contact between the inferior margin of the device and atrial endocardium, resulting in local flow disturbance and endocardial injury.
Explanted atrial septal defect occluder device. The device demonstrates incomplete endothelial coverage with exposed surface areas and focal adherent vegetation.
Preoperative imaging findings. Contrast-enhanced computed tomography demonstrates the atrial septal defect occluder positioned within the interatrial septum and a mass-like lesion along the left atrial wall adjacent to the right lower pulmonary vein orifice (arrow). Transthoracic echocardiography in the parasternal long-axis and apical four-chamber views shows a mobile mass on the left atrial side of the atrial wall (arrow).
The device was removed, and vegetations with surrounding fibrotic tissue were extensively debrided. Areas that could not be safely excised were treated with curettage. The atrial septal defect was definitively closed using a glutaraldehyde-treated autologous pericardial patch with continuous suturing. The patient was weaned from cardiopulmonary bypass without difficulty.
Intraoperative tissue samples were preferentially submitted for microbiological culture to maximise diagnostic yield; however, histopathological examination could not be performed due to limited specimen availability. Tissue cultures were negative.
The patient was extubated on postoperative day 1. Antibiotic therapy was continued for 6 weeks, with a change in regimen on postoperative day 17 due to a drug-induced fever and skin rash. Follow-up computed tomography at 5 weeks confirmed complete removal of the occluder device, preserved interatrial septal integrity, and resolution of the mass-like lesion adjacent to the right lower pulmonary vein (Figure 3). At the 4-month follow-up, the patient remained asymptomatic with no recurrence of fever or infection.
Postoperative imaging findings. Follow-up computed tomography obtained 5 weeks after surgery demonstrates complete removal of the occluder device, preserved interatrial septal integrity after patch closure, and resolution of the mass-like lesion along the left atrial wall adjacent to the right lower pulmonary vein orifice. Corresponding transthoracic echocardiographic views show no residual mass or abnormal lesion on the left atrial wall.
Figure 3. Long description
Panel A: A computed tomography scan in axial view of the chest. The scan shows the heart and surrounding structures. The interatrial septal integrity is preserved after patch closure, and there is no visible mass-like lesion along the left atrial wall adjacent to the right lower pulmonary vein orifice. Panel B: A transthoracic echocardiographic view of the heart. The echocardiogram shows the heart chambers and valves. There is no residual mass or abnormal lesion on the left atrial wall.
The identified S. pneumoniae was later confirmed to be the non-vaccine serotype 6C. Humoral and cellular immunity tests, including immunoglobulin G and lymphocyte subset, were performed, and no specific findings were observed.
Discussion
Late-onset infective endocarditis after transcatheter atrial septal defect device closure is uncommon, with the majority of reported cases occurring during the early post-implantation period. Endothelial coverage of the device surface is assumed to progress within several months after implantation, and the risk of device-related infection is therefore considered to decrease substantially over time. However, the present case demonstrates that infective endocarditis may occur many years after device implantation, indicating that the long-term infective risk cannot be regarded as negligible. Reference Białkowski, Pawlak and Banaszak4,Reference Tamura, Yonehara, Horibata, Uesugi, Sawamura and Sakamoto5
Device-related infective endocarditis is generally attributed to early peri-procedural microbial contamination or late haematogenous seeding. Current European guidelines, including the EACTS/ESC recommendations, do not support routine long-term antibiotic prophylaxis after atrial septal defect device implantation and restrict prophylaxis to selected high-risk patients undergoing specific procedures. However, in the setting of late-onset infective endocarditis, clear guideline-based recommendations regarding the use of antibiotic prophylaxis in clinical situations associated with transient bacteraemia are not clearly addressed. Reference Delgado, Ajmone Marsan and de Waha6 In the present case, Streptococcus pneumoniae was identified in blood cultures obtained at an outside institution. Although pneumococcal infective endocarditis is typically associated with an identifiable primary infectious focus, no definite source of infection was identified based on the available clinical history and evaluation. The patient also had a documented history of the Korean universal 13-valent pneumococcal vaccination. Although serotype 6C is classified as a non-vaccine serotype, preventive effects can be expected through the 13-valent pneumococcal vaccination. [PMID: 38553292].
In the present case, the intraoperative findings indicated that the infectious process was not confined to the device surface itself. Although echocardiography demonstrated vegetation on the occluder, the predominant infectious involvement was observed on the right lower pulmonary vein orifice adjacent to the left atrial wall. This area corresponded to the site of persistent contact with the inferior margin of a relatively large occluder device. Infective endocarditis is generally understood to be initiated by haematogenous seeding. The intraoperative findings suggest that device-related factors may have contributed to the localisation and extent of infection. Histopathological confirmation was not available, as tissue samples were prioritised for microbiological culture. However, the diagnosis of infective endocarditis was supported by positive blood cultures, echocardiographic evidence of vegetation, intraoperative findings, and systemic embolic manifestations.
This case emphasises the need for sustained awareness of late device-related issues following transcatheter atrial septal defect closure during childhood. In selected patients, mechanical factors related to device size and anatomical relationships may contribute to unfavourable long-term device–tissue interactions and predispose late infective endocarditis. Although current guidelines do not recommend routine long-term antibiotic prophylaxis after atrial septal defect device implantation, individualised consideration may be warranted in selected patients with persistent flow turbulence or potential device–tissue interaction undergoing procedures associated with transient bacteraemia. Long-term follow-up with echocardiographic assessment, particularly to identify persistent flow turbulence or abnormal device-tissue interaction, may identify patients requiring closer surveillance.
Conclusion
Despite the convenience of transcatheter atrial septal defect device closure, associated complications may be life-threatening, particularly when infective endocarditis is complicated by systemic embolic events. Although this risk is relevant across all age groups, greater caution is required in paediatric patients because device selection and implantation are limited by smaller cardiac structures and narrower anatomical margins. This case suggests that incomplete endothelialisation and chronic device-related mechanical interaction may create a vulnerable endocardial environment that facilitates bacterial adherence and subsequent infection. Therefore, careful patient selection, meticulous anatomical assessment, and prudent device sizing remain essential, particularly when considering device implantation in children.
Financial support
This study received no specific financial support or funding.
Competing interests
The authors declare that they have no conflicts of interest related to this study.
Ethical standard
Written informed consent for publication of clinical details and images was obtained from the patient.
Open access
