Pulmonary valve endocarditis in adults with congenital heart disease: the role of echocardiography in a case series

Abstract Background Pulmonary valve (PV) endocarditis is a frequent complication during follow-up in patients with repaired right ventricular outflow tract (RVOT) obstruction and poses relevant diagnostic and treatment challenges. We aimed to describe in details the possible different clinical presentations of this rare condition and to highlight the role of both transthoracic and transoesophageal echocardiography which, in experienced hands, may provide comprehensive useful information for the clinicians. Case summary We below describe the clinical presentation and the echo findings of three cases of pulmonary valve endocarditis complicating disease course after different repair modalities of congenital right ventricular outflow tract obstruction. Discussion The present case series outlines the diagnostic challenges of this increasingly frequent complication during follow-up of patients with congenital RVOT dysfunction after both surgical and percutaneous repair. Despite the diffusion of multimodality imaging, echocardiography with PV-dedicated views play a pivotal role in diagnosing such condition and guiding clinical management. Furthermore, this case series highlight that the suspicion of infective endocarditis should be raised whenever a sudden increase in transvalvular gradient is found during follow-up.


Introduction
In the general population, infective endocarditis (IE) involving pulmonary valve (PV) is an extremely rare event, accounting for 1.5-2% of all cases. 1 However, the risk of PV endocarditis significantly rises in the rapidly growing subset of adults with repaired congenital right ventricular outflow tract (RVOT) defects. 2,3 Here, we present three cases of PV endocarditis complicating three different repair modalities of congenital RVOT obstruction.

Timeline
Case presentation   relief. Three years later, the patient was referred to our Adult Congenital Heart Disease (ACHD) Unit for fever following a dental abscess and persisting despite appropriate antibiotic treatment. On physical examination at admission, the patient was febrile, the heart rate (HR) was 110 b.p.m., blood pressure was 100/60 mmHg, and the oxygen saturation was 99% on room air. Inflammatory markers were markedly raised and transthoracic echocardiogram (TTE) showed normal biventricular systolic function [left ventricular ejection fraction (LVEF) 60%, tricuspid annular plane systolic excursion (TAPSE) 20 mm] and a well-seated and normally functioning bioprosthetic PV with thin and mobile leaflets. However, from high parasternal shortaxis view, a mobile filamentous echogenic mass was visualized attached to the struts of the proximal segment of the stent in RVOT ( Figure 1A and B and Supplementary material online, Videos S1 and S2). No signs of pulmonary embolization were demonstrated at the positron emission tomography (PET)/computed tomography (CT) imaging. Serial blood cultures identified Streptococcus anginosus as the causative microorganism and, accordingly, a treatment with teicoplanine was commenced. Six weeks later, the patient was afebrile and in good clinical status and the inflammatory markers returned to normal values. However, TTE at 2-month follow-up showed persistent PV mass, albeit significantly reduced in size, hyperechogenic, and less mobile ( Figure 2 and Supplementary material online, Video S3). Considering that the patient was haemodynamically stable with no signs of active infection, and relying on the echocardiographic features suggestive of low risk of embolization, a conservative watchful  Pulmonary valve endocarditis in congenital heart disease management was followed. The 6-month follow-up TTE showed complete resolution of the vegetation.

Case 2
A 19-year-old man with complex congenital heart disease was referred to our division for persisting fever 1 month after a percutaneous procedure. The cardiac diagnosis was mesocardia, congenitally corrected transposition of the great arteries with non-restrictive ventricular septal defect (VSD), pulmonary atresia with hypoplastic pulmonary arteries supplied by numerous major aortopulmonary collateral arteries. During infancy, he underwent bilateral BT shunts followed by transcatheter perforation of the PV. As significant PS with clinical signs of severe pulmonary hypoperfusion persisted, 1 month before attending our ACHD Unit, a percutaneous PV balloon dilatation was successfully performed at other institution with subsequent improvement in effort tolerance. On admission at our unit, the patient reported severe shortness of breath. On physical examination, the patient was febrile, HR was 105 b.p.m., room air oxygen saturation was 90%, and a harsh 4/6 Levine systolic murmur was appreciable at the cardiac base. At TTE, a narrow RVOT with a  dysplastic PV overriding a large, unrestricted VSD were visualized from subcostal and modified apical views ( Figure 3A and C). Normal biventricular systolic function (LVEF 58%, TAPSE 18 mm) was demonstrated. Doppler analysis revealed severe PS with peak gradient measured at 93 mmHg with no significant pulmonary regurgitation (PR) detectable ( Figure 3B

Case 3
A 27-year-old man with tetralogy of Fallot (TOF) repaired during childhood and reoperated at the age of 14 with an RV to PA Contegra V R (Medtronic, MN, USA) conduit placement, presented to our ACHD Unit after a syncopal episode occurred during a mild effort. He reported, 1 month earlier, an episode of intermittent fever associated with gastrointestinal symptoms. On admission, he was afebrile and well perfused with normal vital signs. A harsh 4/6 Levine systolic and a 3/6 diastolic murmur were heard at the cardiac base. Routine blood tests demonstrated elevation of the inflammatory markers. TTE showed dilated and mildly hypokinetic RV [TAPSE 14 mm, fractional area change (FAC) 30%] with significantly raised RV systolic pressure. Although the conduit was non-visible on 2D echocardiography, Doppler signal at RVOT level revealed severe PS with peak gradient of 171 mmHg ( Figure 5C). LV systolic function was preserved (LVEF 60%). The transoesophageal echocardiogram (TOE) with upper oesophageal (UE) views allowed clear visualization of the severely stenotic conduit with degenerated and thickened PV leaflets determining moderate PR. Moreover, multiple small, mobile masses arising from the ventricular side of the PV were visualized ( Figure 5A and B and Supplementary material online, Videos S7 and S8). The CT imaging demonstrated severely reduced inner conduit lumen size with a minimum diameter of 3 mm, while no signs of embolization were found. Staphylococcus aureus was isolated from serial blood cultures. Based on the TOE results, the patient underwent a successful surgical PV replacement with a homograft.

Discussion
The findings from our series show that, in ACHD patients, PV IE may involve both native and bioprosthetic PV after surgical or transcatheter repair leading to relevant diagnostic and clinical difficulties during follow-up.
Nowadays, PV is the cardiac valve less frequently affected by IE, with the majority of cases confined to patients with congenital RVOT dysfunction. In the last decades, surgical advances allowed to a growing proportion of patients with successfully repaired congenital heart disease to reach adulthood with residual lesions and sequelae, often requiring further treatment during their lifetimes. Accordingly, the increase in prevalence of patients who had undergone either surgical or percutaneous PV procedures results in a parallel increase in the population at risk for IE. In fact, both surgical PV replacement 2 and percutaneous procedures 3 predispose to IE during follow-up. In particular, evidence from several reports in the literature suggested that bovine-derived PV prosthesis might be particularly susceptible to infections. 4 Even though echocardiography remains the mainstay imaging modality in the diagnostic flowchart of IE, difficulties in images acquisition resulting from the variable position of the conduits or shadowing artefacts from stented valves may prevent accurate PV leaflets visualization. Notwithstanding, our case series demonstrates that, in tertiary centres with expertise in imaging in ACHD, echocardiography provides fundamental and reliable morphological and functional information for a correct diagnosis, for guiding the management strategy and for monitoring treatment response. Multiple and off-axis unconventional echo views may be helpful to clearly visualize all the structures in the RVOT.
Transoesophageal echocardiogram is often underused to image RVOT because PV visualization may be limited by its anterior position in chest, thus appearing in the far field. However, TOE role in suspected PV endocarditis should not be neglected, especially when TTE is insufficient to accurately rule out the involvement of other cardiac structures, such as other valves and subvalvular apparatus, patches or intra-cardiac shunts. Furthermore, using PV-dedicated views, TOE may in some cases allows a comprehensive visualization of the anteriorly seated valved conduits as shown in Case 3.
Moreover, it should be noted that, unlike IE involving other cardiac valves, PV endocarditis more often presents with a stenotic lesion, as previous studies have demonstrated. 5 The high burden of infected material described in the surgical resection specimens, along with some degree of inflammatory thickening and thrombosis of the bioprosthetic leaflets have been hypothesized as possible causes. 6 However, non-infective subclinical thrombosis has been demonstrated in explanted prosthetic PV 7 and it could be speculated that thrombotic material deposition and consequent turbulent flow across the valve may predispose to infections. Even though further studies are necessary to clarify the physiopathologic substrate of PV endocarditis presenting with stenosis, this unique feature distinguishing PV endocarditis should raise the suspicion of valve infection whenever a sudden increase in transvalvular gradient is found during follow-up.

Conclusions
Pulmonary valve endocarditis is a frequent complication during follow-up of patients with congenital RVOT dysfunction after both surgical and percutaneous repair. Echo images may be challenging, nevertheless, using unconventional PV-dedicated views, echocardiography provides fundamental info not just for diagnosis but also for treatment guidance.

Lead author biography
Dr Giancarlo Scognamiglio, MD, PhD, is a Consultant Cardiologist and Lead of the Imaging Programme in Adult Congenital Heart Disease at the GUCH Unit of Monaldi Hospital in Naples, Italy. During his training, he specialized in transthoracic and transoesophageal echocardiography and congenital heart disease. His fields of interest include congenital heart disease, valvular heart disease, pulmonary hypertension, ICU, transthoracic and transoesophageal echocardiography, and 3D echo. He is author of chapters in 4 scientific books and more than 40 scientific papers on Peer-Reviewed