A remarkable case of late-onset biatrial bacterial endocarditis of the Gore HELEX septal occluder

Abstract

The HELEX septal occluder (W.L. Gore & Associates, Newark, DE, USA) is a popular closure device for patent foramen ovale (PFO) and secundum atrial septal defect (ASD). Composed of a versatile polymer and a wire frame, it has been used for nearly a decade. A 50-year-old African American man with a PFO repaired 5 years previously with the HELEX was admitted to our hospital with methicillin-resistant Staphylococcus aureus bacteremia. A transesophageal echocardiogram demonstrated large mobile vegetations on the left and right atrial aspects of the occluder. Cardiothoracic surgery was consulted for removal of the vegetations and device, but hesitated owing to concurrent foot osteomyelitis and concern for re-infection of a potential replacement device. The patient was persistently bacteremic, his course complicated by septic pulmonary and cerebral emboli, but eventually cleared his infection on antibiotic therapy. Seventy days since admission, the patient was discharged to a rehabilitation facility. The case illustrates that infection is possible years after PFO/ASD closure despite a theoretical neoendothelialization within 6 months of implantation. Furthermore, it accentuates the success of timely and aggressive medical treatment, thereby averting high-risk open cardiac surgery.

<Learning objective: The Gore HELEX atrial septal occluder was designed to minimize post-implantation infection. However, our case illustrates that the device can be vulnerable to infection even years after placement. Additionally, treatment success with non-operative management underpins the value of swift and aggressive antimicrobial therapy.>

Introduction

Metaanalyses of trials comparing medical treatment versus surgical closure of patent foramen ovale (PFO) have shown that percutaneous closure is more effective for preventing recurrent thromboembolic events [1]. Common complications of any occluder device include device thrombus formation, transient ischemic attack, pericardial tamponade, device dislocation, arrhythmias, and endocarditis [1]. In this article, we take a closer look at endocarditis post-device implant. It is established that up to 6 months post-implant, there is a high likelihood of device infection because endothelialization is incomplete [2]. Late-onset (>6 months) endocarditis of any occluder device is exceptionally rare. There have been scattered reports of late endocarditis with the popular AMPLATZER (St. Jude Medical, St. Paul, MN, USA) device [2], [3], [4]. There has been one reported case of late endocarditis with the HELEX septal occluder (W.L. Gore & Associates, Newark, DE, USA) – where the patient was treated with open-heart surgery [5]. Here, we present a case of delayed-onset right- and left-sided bacterial endocarditis of the HELEX device – made even more remarkable by the resounding success of non-surgical management.

Case report

A 50-year-old African American man presented to our medical center with weakness, fever, lethargy, and chills. His medical history was significant for a 4 mm PFO that was repaired in 2010 (five years prior to current admission, repair reason unknown), paroxysmal atrial fibrillation, type 2 diabetes mellitus with resultant foot ulcer, hypertension, chronic hepatitis B, chronic kidney disease, and non-ischemic heart failure (ejection fraction of 30%, non-significant coronary disease in 2010 heart catheterization). He was noted to be in severe sepsis and admitted to the intensive care unit (ICU). Bacteremia with methicillin-resistant Staphylococcus aureus was discovered — suspected sources being a diabetic foot ulcer, calcaneal osteomyelitis, and/or a long-term intravenous line placed recently for osteomyelitis treatment. Vancomycin was initiated and endocarditis work-up was pursued.

Initial transesophageal echocardiogram (TEE) images were deceptively innocent, requiring advanced color adjustments and contrast to delineate the pathology. On a repeat TEE performed two weeks into the hospitalization, larger biatrial masses were noted attached to the septum (Fig. 1). The masses were noted to be echolucent, with a gelatinous consistency that initially raised the possibility of septal aneurysm or liquefied thrombus. Given the persistent bacteremia, infectious vegetation was presumed to be the diagnosis.

Fig. 1.

Transesophageal echocardiogram, day 14 of admission; Mid-esophageal bicaval views (A and B); Arrows pointing to masses later accepted as vegetations (left-sided vegetation 2.7 × 1.4 cm, right-sided vegetation 2.4 × 1.2 cm). Stealth arrowheads (A) point to outline of occluder device.

LA, left atrium; RA, right atrium; SVC, superior vena cava.

A cardiothoracic surgeon was invited to the case, but was hesitant to intervene due to concern for re-infection of a potential replacement device, given the patient’s diabetic foot ulcer and calcaneal osteomyelitis as existing sources of infection. Bacteremia persisted for 15 consecutive days before consistent clearance. Vancomycin was substituted with daptomycin based on susceptibilities early in the course, and the patient received a full 8-week course while being an inpatient. Ceftaroline was added for synergy and overlapped with daptomycin for approximately 3 weeks. The patient’s course was complicated by septic emboli to the lungs and watershed infarcts to the brain. Three TEEs were performed during his extensive stay — on hospital day 7, 14, and 54. The third TEE showed vegetations reduced in size compared to the prior findings (left-sided vegetation 2.0 × 1.2 cm, right-sided vegetation 1.0 × 0.7 cm). On a fourth TEE (performed 106 days after original admit date), there were minimal residual vegetations noted (Fig. 2). The patient remained an inpatient for 70 days, with 24 of these days in the ICU. He was briefly intubated due to poor mentation and concern for airway protection during TEE procedures. He required nearly no vasopressor support. On hospital day 70, the patient was discharged to a rehabilitation facility, breathing room air, alert, and fully oriented.

Fig. 2.

Transesophageal echocardiogram, 106 days after admission; Mid-esophageal aortic valve short-axis view (A), and bicaval view (B). Arrows point to minimal residual vegetation.

Ao, aorta; LA, left atrium; RA, right atrium.

Discussion

Percutaneous closure of PFO and atrial septal defect (ASD) has been in practice since the 1970s. The two most common devices in current use are AMPLATZER and HELEX, the latter in use since 2006 [6]. The AMPLATZER is a double-disk nitinol mesh that is self-centering and has firm contact with the septum, but may permit small spaces that are at risk for abscess formation and device infection [4], [6], [7]. In contrast, the HELEX is composed of a single piece of nickel–titanium wire with a patch of micro-porous expanded polytetrafluoroethylene (ePTFE) (Fig. 3) [8]. This ePTFE material is designed to conform to the irregularities of the septal surface and facilitate rapid cell penetration, thus promoting quicker tissue ingrowth, resulting in permanent defect closure and device stability [7]. In animal studies conducted by Zahn et al. in early stages of the HELEX, they discovered that implanted devices were nearly completely covered by fibrous connective tissue that deeply infiltrated the ePTFE interstices. Furthermore, an endothelial layer covered the luminal surfaces of the device. No thrombi were observed at any follow-up interval (1 month, 3 months, 6 months, or 12 months) [7].

Fig. 3.

HELEX septal occluder (W.L. Gore & Associates, Newark, DE, USA); Dashed arrow points to the expanded polytetrafluoroethylene patch. Solid arrows point to the nickel titanium wire. Curved arrow points to the end of the wire, which is curved and effectively “pins” the patch from unraveling.

LA, left atrium; RA, right atrium.

From studies performed on other devices, endocarditis is most feared in the first 3–6 months post device implantation [9]. This minimal duration is expected to ensure complete endothelialization by neoendothelium, thereby reducing risk of the foreign body being favored as a site for bacterial infection [2]. From human autopsy findings, incomplete endothelialization has been noted as late as 5 months post-implant [10]. Antiplatelet therapy and endocarditis prophylaxis is recommended for 6 months post-implant [2].

Late-onset endocarditis (>6 months) has been noted in the AMPLATZER device in at least 4 reports [2], [3], [4]. One case of late-onset endocarditis has been reported with the HELEX — 15 months post-implant, and treated surgically. The exact reasons for late device infection in our patient are unclear. We speculate that the protruding curved tip of the titanium wire on either side of the patch (curved arrow in Fig. 3 illustration) could have been the site of initial infection. Subsequent erosion of the endothelialized layers by infectious biofilm and exposure of the titanium wire could have propagated formation of large vegetations. Our case adds to the literature because infection took place years after implant and was treated non-surgically. Surgical specialists concurred that surgery is necessary for the patient’s survival, but timely and synergistic antimicrobial therapy precluded an open-heart procedure.

The HELEX remains a popular device for ASD and PFO closure, but our case demonstrates that it is not infallible, and late infectious complications are possible. By reporting this case, we aim to spur enhancement and innovation in percutaneous intervention for septal defects.

Conflicts of interest

The authors declare that there are no conflicts of interest.