Infective Endocarditis in the Transcatheter Aortic Valve Replacement Era: Comprehensive Review of a Rare Complication

Abstract

In recent years, transcatheter aortic valve replacement (TAVR) has emerged as a revolutionary alternative for surgical aortic valve replacement (SAVR) for the treatment of severe symptomatic aortic stenosis in patients at high risk for surgery. Prosthetic aortic valve endocarditis is a serious complication after SAVR with high morbidity and mortality. Although numerous TAVR procedures have been performed worldwide, infective endocarditis (IE) after TAVR was reported in the literature in few cases only and in 0% to 2.3% of patients enrolled in large TAVR cohorts. Our aim was to review the literature for IE following TAVR and to discuss the diagnostic and management strategies of this rare complication. Ten case reports of IE after TAVR were identified, 8 of which were published as case reports and 2 of which were presented in congresses. Infective endocarditis occurred in a mean time period of 186 days (median, 90 days) after TAVR. Most cases were characterized by fever and elevated inflammatory markers. Infective endocarditis after TAVR shared some common characteristics with IE after SAVR, yet it has some unique features. Echocardiographic findings included leaflet vegetations, severe mitral regurgitation with rupture of the anterior leaflet, and left ventricle outflow tract to left atrium fistula. Bacteriologic findings included several atypical bacteria or fungi. Cases of IE were managed either conservatively by antibiotics and/or using surgery, and the overall prognosis was poor. Infective endocarditis after TAVR deserves prompt diagnosis and treatment. Until further evidence is present, IE after TAVR should be managed according to SAVR guidelines with modifications as needed on a case‐by‐case basis.

The authors have no funding, financial relationships, or conflicts of interest to disclose.

Prosthetic aortic valve endocarditis is a serious complication after surgical aortic valve replacement (SAVR), occurring in about 10% to 15% of all cases lifelong with a high morbidity and mortality rate.1, 2 The diagnosis of prosthetic valve endocarditis is based on the combination of abnormal echocardiographic and bacteriologic findings in the appropriate setting and supported by additional clinical and laboratory markers, as suggested by the Duke criteria.1, 2 In the majority of cases, a repeat (ie, redo) SAVR is required.2 In recent years, transcatheter aortic valve replacement (TAVR) has emerged as a revolutionary alternative for SAVR for the treatment of severe symptomatic aortic stenosis in patients at high risk or in whom surgery is not an option. Numerous TAVR procedures have been performed worldwide, using either the CoreValve (Medtronic, Minneapolis, MN) or the Edwards SAPIEN (Edwards Lifesciences, Irvine, CA) prosthetic aortic valve systems. The most common complications after TAVR procedures are short‐term periprocedural and device‐related complications.3 Infective endocarditis (IE) after TAVR is a rare but serious complication reported up to date in only few case reports. The entity of IE after TAVR is an emerging one, and therefore there are no clear guidelines as to the approach toward diagnosis or treatment of this new entity. We therefore aimed to review the literature for all cases of IE reported following TAVR and to examine their clinical characteristics and echocardiographic findings, as well as treatment modalities used and their prognosis.

Methods

We searched for articles using MEDLINE‐EBSCO and PubMed databases as well as bibliographical cross‐referencing of all articles using the terms “endocarditis,” “TAVI (transcatheter aortic valve implantation),” “PAVI (percutaneous aortic valve implantation),” and “TAVR (transcatheter aortic valve replacement).” We have also searched for these topics as case reports in major congresses between 2008 and 2012, including congresses of the European Society of Cardiology and EuroPCR, the American College of Cardiology, Trans‐Catheter Cardiovascular Therapeutics, and the American Heart Association. Furthermore, we also searched few large TAVR cohorts of recent years for the occurrence of endocarditis.

Results

We identified 10 cases (8 published as case reports, 2 cases presented in congresses) of IE occurring after TAVR procedures (Table 1). Most TAVR procedures were uneventful, with no periprocedural complications. The majority of IE cases occurred few months after the procedure, with a mean time averaging 186 days after the initial valve implantation (range, 1–570 days; median, 90 days). The infected valve model was CoreValve in 6 cases and Edwards SAPIEN in 4 cases. Predisposing factors for IE included prior infections and procedures, as detailed in Table 1. Most cases were characterized by fever and elevated inflammatory markers. Echocardiography findings included valve vegetation (4 cases), severe mitral regurgitation with rupture of the anterior leaflet (2 cases), left ventricle outflow tract to left atrium fistula (2 cases), and an echo‐free space within the wall of the ascending aorta where the stents of the CoreValve were observed (1 case). In 1 case, the echocardiography study did not show any abnormality.4 Transesophageal echocardiography was performed in most cases of IE (7 cases). Different pathogens were identified, including Staphylococcus species (3 cases), Streptococcus species (2 cases), and atypical bacteria such as Enteroccocus faecium, Corynebacterium, and Moraxella nonliquefaciens, as well as fungi such as Histoplasma capsulatum and Candida albicans. Seven patients were treated only with antibiotics, whereas in 3 cases surgery was performed. Out of the 10 patients, 4 patients have died (40%), none of whom underwent surgery.

Table 1.

Case Reports of Infective Endocarditis Occurring in the Implanted Aortic Valve

No.	Reference	Age (y)	Sex	Major Comorbidities	Type of Aortic Valve	Time After TAVR (d)	Predisposing Factors for IE	Clinical Symptoms of IE	IE Signs on Physical Examination	Mean of Diagnosis of Valve Abnormality	Laboratory	Echocardiography	Pathogen	Treatment	Outcome
1	Rafiq et al21	64	F	Myasthenia gravis	CoreValve	60	NR	Fever, malaise	None	TTE + TEE	Elevated CRP and WBC, anemia	Echo‐free space in the wall of ascending aorta	Moraxella nonliquefaciens	6wk of antibiotics	Alive after 3 y
2	Gotzmann et al15	81	M	CAD, LV dysfunction, CRF	CoreValve 29 mm	570	PCI 1 mo prior to IE	Fever, rest dyspnea	NR	TEE	Elevated CRP, WBC	Paravalvular leak with LVOT‐LA fistula and valve vegetation (8×16 mm)	Staphylococcus lugdunensis	Antibiotics	Death after 13 d
3	Carnero‐Alcázar et al22	83	F	LV dysfunction, CRF	Edwards SAPIEN 23 mm	90	Recurrent UTIs in preceding 2 mo	CHF, fever	NR	Autopsy	NR	Valve vegetation (19 mm)	Enterococcus faecalis	Antibiotics	Death after 14 d
4	Head et al20	78	M	NR	Edwards SAPIEN	360	None	Fever	NR	TEE	Mild thrombocytopenia	Valve vegetation	Histoplasma capsulatum	SAVR, antifungal therapy	Alive after 9 mo
5	Comoglio et al17	66	M	Severe chronic lung disease, myelodysplasia	CoreValve	90	Mechanical ventilation 4 wk prior to IE	Fever, shivering, leukocytosis	NR	TEE	NR	Severe MR due to rupture of the anterior leaflet, pseudoaneurysm of the posterior aortic annulus	Corynebacterium	SMVR, aortic valve Carpentier‐Edwards prosthesis	Alive after 7 d
6	Wong et al18	88	M	CABG, CRF, prostate cancer, abdominal aneurysm	Edwards SAPIEN 26 mm	330	Dental visit 6 wk prior to IE	Fever	NR	TEE	NR	Severe MR due to rupture of the anterior leaflet	Streptococcus angiosus	SMVR, SAVR	Alive after 38 d
7	Santos et al14	91	M	CRF, pulmonary hypertension	Edwards SAPIEN 23 mm	1	NR	Fever	NR	Autopsy	Leukocytosis, elevated CRP and ESR	Transvalvular aortic gradient (20/10 mm Hg)	Candida albicans	Antifungal drug	Death after 54 d from TAVR
8	Chrissoheris et al4	84	M	CVA, chronic lung disease, chronic pancreatitis	CoreValve 29 mm	80	Repeated attempts at CoreValve implantation	Sepsis	NR	Not relevant	Elevated ESR and CRP	No vegetations	Staphylococcus epidermidis	Antibiotics	Alive after 12 mo
9	Loh et al19	85	M	CABG, LV dysfunction	CoreValve 29 mm	270	NR	Loss of appetite, apyrexia	None	TEE	Elevated CRP, anemia, acute renal failure	Suspected vegetation	Enteroccocus faecium	Antibiotics	Alive after 3 mo
10	Vidal‐Pérez et al16	80	M	Lung disease, CVA, CRF, CABG, AAA	CoreValve	14	Suspected pneumonia	Dyspnea, fever	NR	TEE	Leukocytosis	Mitroaortic union abscess (above the endoprosthesis) with fistulization from LVOT to left auricle	Staphylococcus lugdunensis	Antibiotics	Death after 1 mo

Abbreviations: AAA, abdominal aortic aneurysm; CABG, coronary artery bypass graft; CAD, coronary artery disease; CHF, congestive heart failure; CRF, chronic renal failure; CRP, C‐reactive protein; CVA, cerebrovascular accident; ESR, erythrocyte sedimentation rate; F, female; IE, infective endocarditis; LA, left atrium; LV, left ventricular; LVOT, left ventricle outflow tract; M, male; MR, mitral regurgitation; NR, not reported; PCI, percutaneous coronary intervention; SAVR, surgical aortic valve replacement; SMVR, surgical mitral valve repair; TAVR, transcatheter aortic valve replacement; TEE, transesophageal echocardiography; TTE, transthoracic echocardiography; UTI, urinary tract infection; WBC, white blood cell.

Cases of IE in large TAVR cohorts are depicted in Table 2. The prevalence of IE after TAVR ranged from 0% to 2.3% during a period of 1 to 3 years following implantation. There were no specific data of IE cases from these cohorts.5, 6, 7, 8, 9, 10, 11 One case of IE that occurred 20 months after TAVR was reported in an autopsy study of 26 TAVR patients.12 In this case, heavy valve calcification secondary to valve infection was noted.

Table 2.

Incidence of Infective Endocarditis in Large TAVR Cohorts

Name of Cohort	Incidence of Endocarditis, n (%)	Follow‐up, y
PARTNER A15	4 (1.5)	2
PARTNER B16	3 (2.3)	2
SOURCE17	10 (1.0)	1
REVIVAL18	0	1
Gurvitch et al19	1 (1.4)	3
Buellesfeld et al20	2 (1.4)	2
Ye et al21	0	3

Abbreviations: PARTNER, Placement of Aortic Transcatheter Valves trial; SOURCE, Edwards SAPIEN Aortic Bioprosthesis European Outcome registry; REVIVAL, Transcatheter Endovascular Implantation of Valves study; TAVR, transcatheter aortic valve replacement.

Discussion

Transcatheter aortic valve replacement is a novel procedure used for patients with severe symptomatic aortic stenosis with high surgical risk. Its short‐ and long‐term clinical outcomes are promising, but data are still emerging.3 The TAVR procedure is not free from complications, given that it is primarily reserved for elderly patients with multiple comorbidities.3 Infective endocarditis following TAVR is a rare but serious complication, and there are sparse data regarding it. The purpose of this article was to review all the reported cases of IE after TAVR up to date worldwide, to examine similar and different patterns and try to deduce some clinical implications.

From the data available so far, it seems that IE after TAVR is a rare entity. As mentioned, according to large study cohorts reported in recent years, the incidence of IE following TAVR ranges between 0% and 2.3%. This variability may depend upon the diagnostic criteria and the length and mode of follow‐up among treated patients.

According to our survey, IE after TAVR is extremely uncommon during the periprocedural and/or short‐term period after TAVR, and it is unclear whether IE soon after TAVR is the result of an intraoperative contamination or a postoperative event that may lead into chronic infectious complication. Fever during hospitalization shortly after the TAVR procedure is common and is usually due to procedure‐ and hospitalization‐related maneuvers. Transcatheter aortic valve replacement is an invasive procedure that involves insertion of several peripheral or central lines, catheters, and occasionally an intubation. Therefore, most fevers encountered during the hospitalization are due to a urinary tract infection, pneumonia, resolved hematomas, skin infections, and other causes.3 It should be recognized that soon after TAVR, systemic inflammatory response may occur in a substantial proportion of patients, manifested by elevated body temperature, increased heart and respiratory rate, and elevated leukocyte count and inflammatory biomarkers.13 This syndrome should be distinguished from IE based upon the short time interval elapsed from TAVR (ie, hours) to clinical onset, its transient nature, and lack of positive blood cultures. Nevertheless, IE shortly after a TAVR procedure should be highly suspected in a state of prolonged fever, new valve murmur and regurgitation, and persistent bacteremia. In these cases, a thorough evaluation for IE should be considered and repeat echocardiographic imaging, including a transesophageal echocardiography study, should be ordered as needed. Finally, in most TAVR procedures, a temporary pacemaker is implanted; in about 25% to 30% of CoreValve system procedures, this is followed by a permanent pacemaker implant. Although none of the cases reported an infection in a pacemaker pocket or electrode, theoretically this is an important port of infection and should be evaluated as needed.

As reviewed herein, IE occurring later after TAVR is a serious complication. Thus, it is essential to diagnose it early and provide prompt and definitive treatment. Which criteria should guide the diagnosis and treatment of IE following TAVR remains to be determined. Another question is, Do the same guidelines for IE after SAVR apply in TAVR cases? Due to a relatively small number of cases and follow‐up duration, we believe that meanwhile it would be reasonable to evaluate patients with suspected IE after TAVR in the same manner as in SAVR. The Duke criteria used for IE should be used also in a patient after TAVR who is evaluated for IE, yet a few characteristics may be different and should be kept in mind: First, the majority of TAVR patients are elderly, and, at least after SAVR, IE in the elderly has some unique features that pose a diagnostic challenge, including insidious symptoms, delayed diagnosis, negative blood cultures in up to 20% of cases, more virulent pathogens, and poorer prognosis.1 These characteristics also could be relevant among TAVR patients presenting with IE. Second, the major criteria according to the Duke classification include positive blood cultures for typical microorganisms for IE and evidence of cardiac involvement by echocardiography.1 The “typical” microorganisms after TAVR are not necessarily identical to those with SAVR. For example, none of the reported cases of IE after TAVR were due to Staphylococcus aureus, which is the most common pathogen in IE after SAVR. Third, echocardiographic findings of vegetations were not evident in all the cases described, but aortic regurgitation,14 fistulas and aneurysm,15, 16 and perforation of the mitral valve17, 18 were the only clues to IE in some cases. This variability may be due to the structural differences between surgical valves and transcatheter valves. In the former, for example, dehiscence of the valve can implicate prosthetic valve endocarditis, whereas this scenario is not feasible in the latter. Whether low implantation of the aortic valve, in particular the CoreValve system, is a risk factor for endocarditis and involvement of adjacent structures such as the mitral valve, as described in at least 2 cases reviewed,17, 18 is yet to be determined. In one case,19 no evidence of valve vegetations was observed and the diagnosis of IE was mainly based on positive blood cultures. This case emphasizes the necessity of determining the best means to distinguish between an isolated bacteremia after a TAVR procedure and that related to IE. Finally, although the Duke criteria are still the gold standard for the diagnosis of IE, in patients who underwent TAVR the exact criteria and definitions are yet to be established. It is noteworthy, however, that IE‐related findings by echocardiography could be obscured by the metal struts encircling the valve leaflets in both the CoreValve and the Edwards valve designs. Whether other diagnostic modalities, such as positron emission tomography–computed tomography, could be useful in this context is yet to be determined.

The treatment of choice for IE after TAVR could not be dictated by conventional guidelines, and until further data are available, a case‐by‐case decision based on clinical judgment or confirmed infectious diagnosis should be undertaken and managed individually. First, the antibiotic treatment should be specific for the pathogen identified in the blood culture. Second, as is the case after SAVR, several criteria should guide the medical staff to consider SAVR over medical treatment; for example, prolonged fever or clinical deterioration, such as congestive heart failure, despite pathogen‐specific antibiotic treatment. In particular in TAVR patients who are a priori at very high risk or for whom SAVR was not an option, this consideration should be reserved for complicated cases and life‐threatening clinical scenarios. A thorough multidisciplinary clinical evaluation by the heart team should be made to best estimate the risks associated prior to referring for surgery. In this case, the patient and family should be aware of the hazardous consequences and should be an integral part of the decision‐making process. Nevertheless, the 3 IE cases described above that were managed surgically17, 18, 20 were successful and the patients survived. This should remind us that although these patients are at very high risk for surgery, in some clinical scenarios of IE, SAVR will be the only viable life‐saving treatment of choice.

Endocarditis prophylaxis prior to TAVR is determined by institutional preference and is not yet evidence based. Yet, it is rational to use regimens that should be directed against skin pathogens, primarily S. aureus. A first‐generation cephalosporin or vancomycin in institutions with a high rate of methicillin‐resistant S. aureus is probably the treatment of choice. Whether endocarditis prophylaxis before dental and other procedures should be recommended is yet to be determined and should be evaluated by a case‐by‐case basis. Furthermore, whether paravalvular aortic regurgitation after the TAVR procedure confers a risk factor for developing IE remains unknown.

Finally, owing to the variability in the transcatheter valve technique, port of catheter entry, and clinical presentation, it is essential to further collect systematic and precise data on IE after TAVR to define the best diagnostic means and treatment options for this serious complication.

Study Limitations

This review is based on cases reported in the literature or in congresses, as well as on reports from TAVR cohorts. Certain cases of IE after TAVR, in particular typical cases with common microorganisms, might not have been reported or published, and therefore this could be a potential publication bias and underestimate the true prevalence of IE after TAVR.

Conclusion

Infective endocarditis after TAVR is a rare but serious complication that deserves prompt diagnosis and treatment. Until further evidence is present, IE after TAVR should be managed as IE after SAVR, with modifications as needed on a case‐by‐case basis.