Infective endocarditis post-transcatheter aortic valve implantation (TAVI), microbiological profile and clinical outcomes: A systematic review

Background The data on infective endocarditis after transcatheter aortic valve implantation (TAVI) is scarce and limited to case reports and case series in the literature. It is the need of the hour to analyze the available data on post-TAVI infective endocarditis from the available literature. The objectives of this systematic review were to evaluate the incidence of infective endocarditis after transcatheter aortic valve implantation, its microbiological profile and clinical outcomes. It will help us to improve the antibiotic prophylaxis strategies and treatment options for infective endocarditis in the context of TAVI. Methods EMBASE, Medline and the CENTRAL trials registry of the Cochrane Collaboration were searched for articles on infective endocarditis in post-TAVI patients till October 2018. Eleven articles were included in the systematic review. The outcomes assessed werethe incidence of infective endocarditis, its microbiological profile andclinical outcomes including major adverse cardiac event (MACE), net adverse clinical event (NACE), surgical intervention and valve-in-valve procedure. Results The incidence of infective endocarditis varied from 0%-14.3% in the included studies, the mean was3.25%. The average duration of follow-up was 474 days (1.3 years). Enterococci were the most common causative organism isolated from 25.9% of cases followed by Staphylococcus aureus (16.1%) and coagulase-negative Staphylococcus species (14.7%). The mean in-hospital mortality and mortality at follow-up was 29.5% and 29.9%, respectively. The cumulative incidence of heart failure, stroke and major bleeding were 37.1%, 5.3% and 11.3%,respectively. Only a single study by Martinez-Selles et al. reported arrhythmias in 20% cases. The septic shock occurred in 10% and 27.7% post-TAVI infective endocarditis patients according to 2 studies. The surgical intervention and valve-in-valve procedure were reported in 11.4% and 6.4% cases, respectively. Conclusion The incidence of post-TAVI infective endocarditis is low being 3.25% but it is associated with high mortality and complications. The most common complication is heart failure with a cumulative incidence of 37.1%. Enterococciare the most common causative organism isolated from 25.9% of cases followed by Staphylococcus aureus in 16.1% of cases. Appropriate measures should be taken to prevent infective endocarditis in post-TAVI patients including adequate antibiotics prophylaxis directed specifically against these organisms. Study registration PROSPERO registration number CRD42018115943.


Introduction
Infective endocarditis (IE) is an uncommon infectious disease but with significant mortality and morbidity [1]. The mortality rate of infective endocarditis is 25% [1]. In most population surveys, its incidence ranges from 3-7 per 100,000 people per year [2]. Infective endocarditis ranks fourth among the life-threatening infections. In 2010, it was estimated that IE causes 1.58 million disability-adjusted life-years worldwide [3]. The causative organism isolated is Viridans streptococci in 35-45% of the patients [4]. According to a survey in France, the microbiological profile of infective endocarditis has changed in recent years [5]. Staphylococcus aureus causes the majority of the cases of infective endocarditis in the industrialized world. There is a higher incidence of infective endocarditis in older age, those with prosthetic heart valves and cardiac devices while at the same time, there is a decreased proportion of IE in rheumatic heart disease [6].
Infective endocarditis presents with high-grade fever, valvulitis, peripheral emboli, immunological phenomenon and sustained bacteremia or fungemia. However, the typical history and clinical manifestations are not present in most of the patients. So, the diagnosis of infective endocarditis relies on a highly sensitive and specific diagnostic strategy. In 1994, a diagnostic scheme was formulated in Duke University Medical Center. According to this scheme, patients with suspected infective endocarditis were allocated into three classes: definite, possible and rejected cases (Table 1) [7]. Antibiotic prophylaxis is given for the prevention of infective endocarditis in high-risk persons before dental treatment [8]. The high-risk persons for infective endocarditis include those withtheprevious episode of infective endocarditis, congenital heart defects and prosthetic heart valves. Individuals with the history of rheumatic fever, heart murmur and native valve disease are at moderate risk of infective endocarditis [4].
Aortic stenosis is the most prevalent valvular abnormality in adults with a higherincidencein advanced age. The frequency of aortic stenosis has increased due to increasing life expectancies [9]. The patients with aortic stenosis are treated with conventional cardiac surgery, surgical aortic valve replacement (SAVR). Until recently due to significant mortality associated with SAVR, elderly patients with co-morbidities were not treated [10].
Transcatheter aortic valve implantation (TAVI) was introduced in 2006 as a revolutionary intervention for severe aortic stenosis. It is a less invasive procedure for high-risk patients or patients who are inoperable by SAVR [11,12].Sternotomy is not required in TAVI. A bioprosthetic valve is implanted over the native valve using a catheter. The mortality rate after TAVI is reported to be 14%-31% after 1 year [13]. The commonly used TAVI systems are the Edwards SAPIEN valve and the CoreValve1. Both of them are effective and safe [14]. Transcatheter aortic valve implantation can be performed through various approaches such as transfemoral, transapical, subclavian or direct aortic approach. The transfemoral approach is done in 80-90% of the cases and is appropriate for both types of the valve [15]. The success rate after TAVI is more than 90% and 30-day procedural mortality rates less than 10% [16]. The majority of the complications after TAVI are technical and device-related including conduction disturbances and periprosthetic paravalvular leaks [17].After the success of TAVI in aortic stenosis, several strategies have been established to treat aortic regurgitation with TAVI [18].
Infective endocarditis (IE) is a fatal complication of TAVI [19]. The incidence of infective endocarditis after TAVI is low, although it is a major cause of heart failure. As TAVI is becoming more popular with time, the magnitude of post-TAVI infective endocarditis will rise [20]. The patients with post-TAVI infective endocarditis have an atypical clinical presentation causing a delay its diagnosis and treatment [21].Surgical aortic valve replacement is often required to explant the valve with endocarditis and most of these patients are inoperable or at high risk for SAVR [12]. The majority of the patients who underwent TAVI are elderly and infective endocarditis could have the worst prognosis in them [22].The leaflets of transcatheter valve prostheses contain a greater quantity of metal in the stent frame in contrast to the surgical valves. This factor may change the outcome and management of IE [23].
There is limited data on infective endocarditis after TAVI, its microbiological profile, clinical outcomes and treatment modalities. Most of the data is limited to case reports and small series, which can lead to publication bias [19]. In the prospective randomized Placement of Aortic Transcatheter Valves (PARTNER) trial, the incidence of infective endocarditis after TAVI is reported from 0.1% to 3.03%, with no difference between TAVI and SAVR [24,25]. According to other studies, infective endocarditis occurs in 0.5% to 3.1% of patients after TAVI [26,27]. Similar incidence rates of infective endocarditis are reported after surgical valve replacement [28]. The clinical picture of post-TAVI IE varies from nonspecific symptoms to acute infection or sepsis with fever, heart failure or embolic stroke. As most of these patients are elderly, they frequently present with atypical symptoms and signs. Greater than 50% of the patients present with heart failure and 20% have non-specific symptoms e.g. malaise, weakness or weight loss. High-grade fever and heart murmur are relatively less common in post-TAVI IE as compared to native valve IE. The diagnosis of post-TAVI IE by echocardiography is more difficult than native valve IE due to differences in the technique of valve implantation. The detection of small vegetation in post-TAVI IE is limited as the valve contains large amounts of metal which create reflectance and a shadowing effect [30]. According to guidelines, early surgery is indicated in complicated cases of post-TAVI IE. Because these patients are high-risk, surgery is contraindicated in most of the patients with post-TAVI IE [31].  Infective endocarditis post-TAVI, microbiological profile and clinical outcomes: A systematic review

Rationale
Infective endocarditis is a rare but fatal complication of TAVI with its incidence ranging from 0.1%-3.03%. It has a high mortality rate with many complications [1,23].Transcatheter aortic valve implantation is gaining popularity day by day because it is feasible, less invasive technique and the only treatment option in patients at high surgical risk for severe symptomatic aortic stenosis [11]. Due to this, the incidence of post-TAVI infective endocarditis is expected to rise in the coming years. The data on infective endocarditis after TAVI is scarce and limited to case reports and case series in the literature [19,20]. It is the need of the hour to analyze the available data on post-TAVI infective endocarditis from the available literature. This study wasdesigned to conduct a systematic review of the incidence of infective endocarditis after TAVI, the causative pathogens isolated from these patients and the clinical outcomes. It will help us to improve the antibiotic prophylaxis strategies and treatment options for infective endocarditis in the context of TAVI.

Objectives
The objectives of this systematic review were to evaluate the incidence of infective endocarditis after TAVI, its microbiological profile and clinical outcomes.

Material and methods
The study was done according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines and is registered with PROSPERO International Prospective Register of Systematic Reviews (PROSPERO registration number CRD42018115943).

Eligibility criteria
The following criteria were used for the selection of studies:

Study designs
The retrospective, prospective and observational studies were eligible for the systematic review. The case reports and case series were excluded.

Participants
The studies included adult humans of either gender with age �70 years who underwent TAVI. The minimum follow-up time was 6 months.

Interventions
The intervention was TAVI. Healthcare provision and follow-up of the patients after TAVI to look for the incidence of infective endocarditis & its clinical outcomes. Investigating the causative organisms of infective endocarditis.
"Transcatheter heart valve endocarditis is defined following Duke's modified criteria or evidence of abscess/paravalvular leak/pus/vegetation confirmed as secondary to infection by histological or bacteriological studies or evidence of abnormal tracer uptake around the site of the prosthetic valve by F-fluoro-deoxyglucose positron emission tomography" [29].

Comparators
There was no control group for comparison.

Outcomes
The primary outcome was: ➢ Incidence of post-TAVI infective endocarditis The secondary outcomes were: ➢ Microbiological profile of infective endocarditis ➢ MACE including in-hospital mortality & mortality at follow-up, heart failure, stroke, major bleeding and arrhythmias Sepsis is defined as an infection that triggers a particular Systemic Inflammatory Response Syndrome (SIRS). This is characterized by body temperature outside 36˚C-38˚C, HR >90 beats/min, respiratory rate >20/min, WBC count >12,000/mm3 or <4,000/mm3. Patients with infections plus two or more elements of the SIRS meet the criteria for sepsis. Those who have end organ failure are considered as having severe sepsis; and those who have refractory hypotension along with the above said criteria are considered to be in septic shock [32]."

Timing
All the included studies had a follow-up time of at least 6 months after TAVI.

Setting
The study conducted in any type of setting was included.

Language
Studies reported in the English language only.

Information sources
All the relevant articles in English with text words related to infective endocarditis and transcatheter aortic valve implantation (TAVI) were searched in MEDLINE (PubMed), EMBASE (OVID interface) and the Cochrane Central Register of Controlled Trials (Wiley interface) till October 2018. The literature search was limited to human subjects. The case reports and case series were excluded.

Data management, selection process and data collection
The first and second authorswere the principal investigators. Each author individually read all the relevant articles. The articles meeting the eligibility criteria were included for the systematic review. The search results from each database were saved in EndNote X9 and duplicates were removed.

Data items
Data was recorded and tabulated including author name, year of publication, sample size, study type, follow-up time, mean age & gender of patients, primary and secondary outcomes. It ensured uniformity between the authors and integration of findings.

Outcomes and prioritizations
The outcomes were the incidence of infective endocarditis after TAVI, its microbiological profile and clinical outcomes of infective endocarditis.

Data synthesis
The search strategy shortlisted 137 articles, out of which 44 articles were relevant. Forty four articles on infective endocarditis in post-TAVI patients were assessed for full text. Out of these, 11 studies were included in the systematic review. All the studies were retrospective or observational with the follow-up duration of at least 6 months. Thirty three articles did not meet the inclusion criteria as these were case reports, case series and systematic reviews. The meta-analysis could not be done as there was no control group for comparison.The PRISMA flow diagram for the study protocol is shown in Fig 3.

Data analysis
The data entry and analysis was done using the Statistical Package for the Social Sciences (SPSS) version 25. The follow-up duration and incidence of IE were evaluated as mean. The microbiological profile and clinical outcomes of post-TAVI IE were expressed as frequency and percentage.
Risk of bias. The Newcastle-Ottawa-Scale was used to calculate the risk of bias in the included studies as shown in Table 2.

Results
A total of 44 articles on infective endocarditis in TAVI patients were assessed for full-text. Thirty three articles were excluded as these articles were case reports, case series and systematic reviews. Eleven articles were included in this systematic review with the average duration of follow-up 474 days (1.3 years). Table 3 shows the important parameters of the included studies.
Out of 11 included articles,the incidence of infective endocarditis varied from 0%-14.3%, the mean is 3.25%. This may be attributed to hospital, cardiologist and patient-related factors or the sample size. The incidence of post-TAVI infective endocarditis is shown in Fig 4. The clinical outcomes of post-TAVI infective endocarditis are summarized in Table 5.
Latib et al. conducted a multicenter study in which he stated that all the patients were given prophylactic antibiotics depending on institutional practices. However, the details of these antibiotics are not mentioned in the article [32]. In a study by Olsen et al., patients received prophylactic antibiotics (amoxicillin or roxithromycin) only before undergoing invasive dental procedure [27].
Studies  Infective endocarditis post-TAVI, microbiological profile and clinical outcomes: A systematic review cefuroxime or ceftriaxone in combination with fusidic acid/ciprofloxacin/rifampicin in 5 patients [27].
Puls et al. mentioned the following antibiotics used for treating 5 patients with post-TAVI IE: vancomycin & rifampicin alone or in combination with gentamicin in 2 patients, ampicillin with ciprofloxacin/gentamicin in 1 patient and ceftriaxone in 1 patient. One patient received parenteral antibiotics, the names of which are not mentioned in the study [34].
In a study by Martinez-Selles et al., out of 10 post-TAVI IE patients, 9 patients received β lactam antibiotics. Two out of these 9 patients were also given aminoglycosides. One patient had fungal endocarditis and was treated with fluconozole&caspofungin [22]. The antibiotics used for the treatment of post-TAVI IE in 5 studies are tabulated in Table 6.

Discussion
The results of our study showed that the mean incidence of post-TAVI IE is 3.25%. The most common causative organism is Enterococci (25.9%) followed by Staphylococcus aureus (16.1%) and coagulase-negative Staphylococcus species (14.7%). Amat-Santos et al. also reported Enterococci (34.4%) as the most common organism causing IE after TAVI but the second common organism was coagulase-negative Staphylococcus species (18.7%) [26]. In contrast, in a systematic review by Eisen et al., coagulase-negative Staphylococcus species was the most common (30%) cause of post-TAVI IE followed by Enterococci (20%) [19].
In our study, the mean incidence of in-hospital mortality is 29.5% and mortality at followup is 29.9%. Similarly, the high mortality rate was reported in other systematic reviews. Post-TAVI IE was responsible for 34.4% and 40% mortality in studies by Amat-Santos et al. and Eisen et al.,respectively [26,19]. According to our study, surgical intervention and valve-invalve procedure for the treatment of post-TAVI infective endocarditis were performed in

Amat-Santos[23]
Low Low Low Low Low Low Low Low

Puls[34]
Low Low Some concern Some concern Some concern Low Some concern Low

Latib[33]
Low Low Some concern Some concern Some concern Low Some concern Low

Olsen[27]
Low Low Low Low Low Low Low Low

Martinez-Selles[22]
Low Low Low Low Low Low Low Low

Regueiro [28]
Low Low Some concern Some concern Some concern Low Some concern Low

Seiffert[18]
Low Low Some concern Some concern Some concern Low Some concern Low

Spartera [29]
Low Low Some concern Some concern Some concern Low Some concern Low Kosek[36] Low Low Some concern Some concern Some concern Low Some concern Low
Another systematic review was conducted by Eisen et al. including 10 cases, 8 were from case reports and 2 cases were presented in congresses. The microorganisms isolated from blood cultures were coagulase-negative Staphylococcus species in 3(30%), Enterococci in 2 (20%), Candida albicans,Moraxella, Corynebacterium, other Streptococcus species and Histoplasma (1 in each) patients. Three (30%) patients underwent surgery and 4(40%) patients died Infective endocarditis post-TAVI, microbiological profile and clinical outcomes: A systematic review      Infective endocarditis post-TAVI, microbiological profile and clinical outcomes: A systematic review after post-TAVI infective endocarditis [19]. The major limitation of these systematic reviews was that most of the included articles were case reports and case series which may have precluded the real evaluation of the post-TAVI infective endocarditis. Lastly, certain cases of post-TAVI infective endocarditis might not have been published leading to potential publication bias. A retrospective multicenter study was conducted in 21 centers in America and Europe. A total of 7944 patients underwent TAVI out of which 53(0.67%) patients developed infective endocarditis. The mean follow-up time of patients was 1.1±1.2 years. Antibioticprophylaxis was given in all centers during the TAVI procedure. Cephalosporins, vancomycin and piperacillin/tazobactam were used in 14(67%), 6(28%) and 1(5%) centers respectively. In most of the centers, a single antibiotic dose was administered except 2 centers in which 2-3 doses were also given after TAVI. Coagulase-negative Staphylococcus specieswas the most causative organism isolated from 13(24.5%) cases.Staphylococcusaureus, Enterococci and Viridans streptococci were isolated from 11(20.8%), 11(20.8%) and 3(5.7%) patients, respectively. Infective   Table 4. https://doi.org/10.1371/journal.pone.0225077.g006 Infective endocarditis post-TAVI, microbiological profile and clinical outcomes: A systematic review endocarditis was caused by atypical microorganisms such as Escherichia coli, Serratia, Acinetobacter, Candida lusitaniaein 13(24.5%) patients. In 2(3.8%) patients, cultures were negative. The most common complication of IE was heart failure in 36(67.9%) patients. The other complications were acute kidney injury, septic shock, systemic embolism, stroke and persistent infection. The patients were managed with surgical explantation (7.5%) and valve-in-valve procedure (3.8%). In-hospital mortality was reported in 25(47.2%) and mortality at follow-up in 13(24.5%) patients [23].
In Grenoble Alpes University Hospital France, a retrospective study was done on 326 patients who underwent TAVI with the average follow-up period of 460 days. Out of 326 patients, 6 (1.8%) patients were diagnosed with infective endocarditis. Blood cultures were positive in 5 and negative in 1 patient. The pathogens responsible in 5 patients were Staphylococcus aureus, coagulase-negative Staphylococcus species, Enterococci, Streptococcus speciesand Escherichia coli. Two patients died after infective endocarditis, one 17 days and another 40 days after IE [35].
Kosek et al. retrospectively analyzed 7 patients with bicuspid aortic valve who underwent TAVI at the Institute of Cardiology, Poland. A total of 104 patients had TAVI at the institution and 7 of them had the bicuspid aortic valve. These patients were followed up to 12 months after TAVI. One patient developed infective endocarditis and died after 30 days of the procedure [36].
A study was carried out on 100 patients who underwent TAVI in Germany. The average follow-up duration was 3.8±2 years. The outcomes assessed were 30-day mortality, 5-year mortality, infective endocarditis, embolization, reintervention, stroke and valve thrombosis. None of the patients developed endocarditis [37].
Another study analyzed the outcomes of TAVI in patients with severe aortic regurgitation. Thirty one patients were included from 9 German centers. The follow-up time was 235 days. During follow-up, one of the patients had infective endocarditis and required surgical intervention (SAVR) [18].
Another retrospective study was performed in Italy in which 621 post-TAVI patients were followed for 402 days. Eight (6%) patients developed infective endocarditis. The microorganisms isolated were Enterococci in 2, Viridans streptococci in 2, coagulase-negativeStaphylococcusspecies in 2, Staphylococcus aureus in 1 and Streptococcus speciesin 1 patient. The mortality rate was 75% in patients [29].

Conclusions
The incidence of post-TAVI infective endocarditis is low being 3.25% but it is associated with high mortality and complications. The most common complication is heart failure with a cumulative incidence of 37.1%. Enterococciare the most common causative organism isolated from 25.9% of cases followed by Staphylococcus aureusin 16.1% of cases. Appropriate measures should be taken to prevent infective endocarditis in post-TAVI patients including adequate antibiotics prophylaxis directed specifically against these organisms.

Limitations of the study
• The meta-analysis could not be done as there was no control/comparison group in the included studies.
• The outcomes of post-TAVI IE have not been integrated with the impact of different treatment options as none of the included studies correlated the outcomes with the treatment modality.
• The review does not improve the discussion on IE prophylaxis, although the rationale sets this topic as a central objective. This is because the data on IE prophylaxis is not detailed in all the included studies.

Recommendations of the study
• Meticulous aseptic measures should be reinforced with special focus on sterilization and disinfection in the catheterization laboratory. The provision of laminar air flow has an added advantage.
• Antibiotic prophylaxis should be given in TAVI patients. As Enterococci are the most common organism causing post-TAVI infective endocarditis, the prophylactic antibiotics should be given directed against this organism. The traditional use of cephalosporins should be reconsidered as Enterococci are intrinsically resistant to all cephalosporins.