Prosthetic Valve Endocarditis After Transcatheter Valve ...(B) Location of infective endocarditis in patients with previous transcatheter pulmonary valve replacement (information available
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5
ª 2 0 1 5 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N I S S N 1 9 3 6 - 8 7 9 8 / $ 3 6 . 0 0
P U B L I S H E D B Y E L S E V I E R I N C . h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j c i n . 2 0 1 4 . 0 9 . 0 1 3
T he use of transcatheter valves for the treat-ment of valve dysfunction has experienceda very rapid expansion since the initial
experiences in the first years of the past decade(Online Refs. 1–4). Although the high proceduralsuccess rate and beneficial effects associated with
m the *Department of Cardiology, Quebec Heart & Lung Institute, Quebe
spital Clínico Universitario de Valladolid, Valladolid, Spain; and the zDepaebec City, Quebec, Canada. Dr. Amat-Santos received support from the In
nico Universitario de Valladolid, Spain (Rio Hortega Contract). Dr. Rod�e
Jude Medical. Dr. Dumont is consultant for Edwards Lifesciences. A
ationships relevant to the contents of this paper to disclose.
nuscript received June 27, 2014; revised manuscript received September
transcatheter valve replacement (TVR) are widelyrecognized, some of the well-known risks associatedwith standard surgical treatment for valve diseasealso exist in TVR (Online Ref. 5), although the compli-cations probably have modified features that mayboth make their diagnosis and management difficult
c City, Quebec, Canada; yDepartment of Cardiology,
rtment of Pediatric Cardiology, Centre Mère-Enfant,
stituto de Salud Carlos III, Madrid, and the Hospital
s-Cabau is consultant for Edwards Lifesciences and
J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5 Amat-Santos et al.F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6 Infective Endocarditis and Transcatheter Valves
335
and change their clinical impact and prognosis. Thisis the case with prosthetic valve endocarditis (PVE),a rare (3 to 9 cases per 100,000 people) but complexand life-threatening disease.
Although a few TVR series have reported the inci-dence of early infective endocarditis IE) (OnlineRefs. 1–9) (Figures 1 and 2), data on PVE in the fieldof transcatheter valves are limited to case reports orsmall series (1–28, Online Ref. 10), which precludedany appropriate evaluation of the clinical character-istics of patients with this syndrome as well as of itsmanagement and prognosis. The objective of thissystematic review was therefore to provide furtherinsight into the baseline characteristics, incidence,disease features, management, and clinical outcomesof patients with IE as a complication of TVR (trans-catheter aortic valve replacement [TAVR] and trans-catheter pulmonary valve replacement [TPVR]).
METHODS
All relevant articles in English about TAVR/TPVR andPVE published between December 2000 and June2013 were systematically searched in BioMedCentral,Google Scholar, and PubMed. The following queryterms were used: transcatheter/percutaneous pul-monary/pulmonic valve replacement/implantation,transcatheter/percutaneous aortic valve replacement/implantation, transcatheter heart valve, infectiveendocarditis, prosthetic valve endocarditis, valveinfection, congenital heart disease treatment andmodified Duke’s criteria. Further studies were soughtby means of a manual search of secondary sources,including references from primary articles (backwardsnowballing) and contacts with international experts.We also searched for these topics as case reports inmajor cardiology meetings between 2004 and 2012.
Citations were first screened at the title/abstractlevel by 2 independent reviewers (I.J.A.S., H.B.R.).Potential divergences were resolved after consensusto gather all pertinent case reports and case seriesconcerning PVE in TAVR and TPVR. Only cases withdefinite endocarditis according to modified Dukecriteria were included (Online Refs. 11,12). Someadditional cases of “probable endocarditis” were alsoincluded due to high suspicion of actual PVE andlimited sensitivity of the diagnostic criteria in theirparticular context. Early PVE was defined, accordingto the guidelines, as that occurring within the first 12months after the valve replacement (Online Ref. 11).
Gathered data included baseline clinical, echocar-diographic, and TVR procedural characteristics. Dataon PVE clinical presentation, invasive procedures(any potential source of infection), etiology, in-
hospital or 30-day complications, and mor-tality at any time point were also gathered.Main baseline characteristics of the TAVR-PVE patients were compared with those ofpatients included in previous TAVR registries(Online Refs. 13–20), the PARTNER (Place-ment of AoRTic TraNscathetER Valve) trial(Online Refs. 6,7), and largest surgical seriesincluding the main types of aortic bio-prosthesis (Online Refs. 21–25). A similarcomparison was also performed for TPVR-PVE patients, including the largest TPVR
series (Online Refs. 26–30) and the largest surgicalseries that reported the incidence of PVE in congen-ital heart disease (Online Ref. 31), pulmonary valvereplacement (Online Ref. 32), and Ross intervention(Online Ref. 33).
Categorical variables were reported as n (%), andcontinuous variables as mean � SD or median (25thto 75th interquartile range) depending on variabledistribution.
pulm
RESULTS
A total of 60 patients who underwent TVR, including32 TAVR (1–16) and 28 TPVR (17–28), and who had PVEwere included in the study. All cases were publishedbetween 2006 and 2013 (Online Table 1 for biblio-graphic sources and type of articles). The main base-line characteristics of TAVR and TPVR populationsare summarized in Tables 1 and 2, respectively. Themean age of TAVR patients was 80 � 7 years, 66% ofthem were men, and the mean logistic EuroSCOREwas 30.4 � 14.0%. A comparison of these data withthe data on the patients included large TAVR regis-tries, the PARTNER trial (Online Refs. 6–9,19–22), andin surgical studies (Online Refs. 22–25) is shown inFigure 3.
TPVR patients were a much younger population(mean age, 19 � 6 years), and only 10% of them werewomen. Tetralogy of Fallot was the most commonunderlying disease, and most patients (53%) had ahomograft as a right ventricular–pulmonary arteryconduit. The mean time between surgery and theTPVR procedure was 10 � 3 years. Stenosis of thevalve conduit was the most common reason for TPVR(60%). Baseline characteristics of the patients withPVE compared with those included in large TPVRseries are shown in Figure 4.
PROCEDURAL AND IN-HOSPITAL COURSE OF TVR
PROCEDURES. The main characteristics of TAVRand TPVR procedures are shown in Tables 1 and 2,respectively. Of the TAVR patients, 58% had received
FIGURE 1 Incidence of Early Infective Endocarditis After TAVR and Surgical Aortic
Valve Replacement
Incidence of early (1 year) infective endocarditis after TAVR (A) and surgical aortic valve
Amat-Santos et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5
Infective Endocarditis and Transcatheter Valves F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6
336
a balloon-expandable Edwards SAPIEN/SAPIEN XTvalve (Edwards Lifesciences, Irvine, California) and42% received a self-expandable CoreValve system(Medtronic, Minneapolis, Minnesota). The trans-catheter valve was replaced via a transfemoral andtransapical approach in 66% and 34% of patients,respectively. Post-procedural echocardiographic datawere available for 21 patients, and the presence ofmild or greater residual aortic regurgitation (AR) wasobserved in 16 of them (76%), with as many as 5patients (24%) with moderate AR. This incidence ofresidual AR was higher than that observed in previousTAVR registries (mild or greater, 45.3%; moderate tosevere, 11.8%) and in both cohorts of the PARTNERtrial (mild or greater, 54%; moderate to severe, 12.3%)(Online Refs. 6–9,19–22). All patients who had TPVRhad received a balloon-expandable Melody valve(Medtronic) implanted via the transfemoral venousapproach; no data on hemodynamic results of theprocedure were provided in any of the TPVR studies.
The location and environmental conditions wherethe interventions took place were reported in 2 TAVRstudies only: a 4-case series that described a hybridroom as the usual place for TAVR procedures (OnlineRef. 34) and the catheterization laboratory in a casereport (Online Ref. 23). In the TPVR population,no information concerning the place where theinterventions were performed was provided.
Data on antibiotic prophylaxis were not detailedin most cases. A wide variety of intravenous antibi-otic regimens were administered including ampi-cillin, vancomycin � ciprofloxacin, cefazolin �gentamicin, or teicoplanin. No details were providedon whether single or multiple antibiotic doses wereadministered.
All procedures were considered successful in theTAVR and TPVR groups. In-hospital complicationswere reported exclusively in TAVR cases and includedcomplete atrioventricular block leading to permanentpacemaker implantation in 2 cases, pneumonia/tuberculosis reactivation in 2 cases, and acute kidneyinjury requiring dialysis in 2 cases. The median lengthof stay after the TAVR procedure for those patientswho did not present in-hospital PVE (all but 2 cases)was 6 days (interquartile range: 2 to 7 days).
CLINICAL FEATURES OF PVE POST-TVR. The mainindividual characteristics regarding the timing, clin-ical presentation, etiology, and location of PVE in theTAVR and TPVR groups are shown in Tables 3 and 4,respectively. The median time from the interventionto the diagnosis of PVE was 5.0 months (interquartilerange: 3 to 9 months). The timeline of initial PVEsymptoms is shown depicted in Figure 5.
The suspected sources of PVE were as follows:respiratory infections (TAVR, 4 cases; TPVR, 2cases), dental interventions (TAVR, 3 cases; TPVR,2 cases), skin infections (TAVR, 3 cases; TPVR, 2cases), and urological (TAVR, 2 cases; TPVR, 0 cases)or gastrointestinal interventions (TAVR, 1 case).The 3 main sources (dental and respiratory/skininfections) were the same for both types of TVR, buta health care–related origin was more common inTAVR patients (42.1%) than in TPVR patients (18.2%)(p ¼ 0.246). The source of infection remained un-determined in as many as 50% of the patients. Verylittle information was provided on the managementof PVE prophylaxis after the TVR procedure, andfailure to comply with recommendations was re-ported in 2 cases (18,22). A history of PVE existed in2 pulmonary cases.
In 2 cases, only criteria for “possible” PVE wereachieved, but given the high suspicion, they werefinally included in the present review (7,8).
FIGURE 2 Incidence of Early Infective Endocarditis After TPVR and Surgical
Pulmonary Valve Replacement
Incidence of early (1 year) infective endocarditis after TPVR (A) and surgical pulmonary
J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5 Amat-Santos et al.F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6 Infective Endocarditis and Transcatheter Valves
337
Fever (80%) and heart failure (22%) were the mostcommon initial symptoms of PVE in the TVR popu-lation. Specific symptoms derived from septic embolisuch as neurological symptoms (exclusively in theTAVR group) or pulmonary abscess in the TPVR groupalso occurred in 12.5% and 3.7% of the patients,respectively. Less frequent symptoms includedcutaneous stigmata, local chills, loss of appetite,macrophage activation syndrome, and limb ischemiadue to septic emboli. One patient remained asymp-tomatic, and PVE was suspected by new valveregurgitation and later confirmed by microbiologicalfindings (Online Ref. 35).
The location of PVE as determined by trans-esophageal echocardiography is schematically depic-ted in Figure 6. The presence of vegetation wasdetected in 58.3% of the patients (50.0% and 68.4% inthe TAVR and TPVR groups, respectively). In theTAVR group, the vegetation was located on thetranscatheter valve leaflets in 7 patients (21.9%), onthe valve stent frame in 2 patients (6.2%), andaffected both structures in 3 patients (9.2%). Also,tricuspid and mitral valves presented vegetation in 1(3.1%) and 3 patients (9.4%), respectively. The infec-tion was restricted to the prosthetic valve in all TPVRcases (Table 5). No valve involvement could beconfirmed in 4 patients (3 patients in the TAVR groupand 1 in the TPVR group), with initial criteria fordefinite PVE and good response to antibiotic therapy.
The main microbiological findings differed in theTAVR and TPVR groups (Table 5). Although Entero-coccus was the most frequent microorganism re-sponsible for PVE in TAVR patients (34.4%),Staphylococcus aureus was predominant among TPVRpatients (29.4%). Globally, positive cultures for“typical” microorganisms (Online Refs. 11,12) wereobtained in about one-half of the patients with pre-vious TVR (53.1% of TAVR patients and 47.1% of TPVRpatients). Less common causal agents included Gram-negative bacilli in 3 cases, Corynebacterium (2 cases),fungal infections (Candida albicans and Aspergillusfumigatus), Histoplasma capsulatum, Bartonella hen-selae, and Granulicatella adjacens (1 case each). In 2patients, all cultures including those performed aftervalve explantation, remained negative, but previousantibiotic therapy had been administered in bothcases before blood samples were obtained.
MANAGEMENT AND OUTCOMES OF TVR-RELATED PVE.
All patients received antibiotic therapy accordingto the etiology of PVE. Cardiac surgery and valveexplantation were performed in 13 patients (41%)in the TAVR group and in 21 patients (75%) in theTPVR group. In the TAVR group, the rate of PVE-
related surgery was twice as high in patientswho had received a balloon-expandable valve (57%)compared with those who had a self-expandablevalve (23%), despite a similar baseline risk profile inboth groups as evaluated by the logistic EuroSCORE(30.5 � 15.6% vs. 27.8 � 13.5% in balloon- and self-expandable groups, respectively).
The most important surgical findings includedincomplete valve endothelialization several monthsafter TAVR in 1 case and sinotubular junctionpenetration by the struts in another case (OnlineRefs. 14,34).
The in-hospital mortality rate was 34.4% in theTAVR group, 30.8% in those patients who had surgeryand valve explantation and 36.8% in those managedmedically. In the TPVR group, the in-hospital mor-tality rate associated with the episode of PVE was7.1%, 9.5% in patients who required surgery and nonein those managed medically. Follow-up data were
TABLE 1 Baseline Data of Patients With Previous Transcatheter Aortic Valve With Infective Endocarditis
Patient #(Reference No.)
Age,yrs Sex
LogisticEuroSCORE Valve-in-Valve
Antibiotic ProphylaxisRegimen Approach Valve
Valve Size,mm Post-TAVR AR
1 (1) 85 M 25 No — TF Edwards SAPIEN XT* 26 —
2 (2) 83 F — No — TA Edwards SAPIEN 23 —
3 (3) 76 M 61.1 Yes — TA Edwards SAPIEN 23 —
4 (4) 80 F 14.2 No Cefazolin TA Edwards SAPIEN XT 23 —
5 (5) 80 M 30 No Nasal mupirocin TF CoreValve† 29 Mild
6 (5) 81 F 48 No — TA Edwards SAPIEN 23 Moderate
7 (5) 80 F 41 No — TA Edwards SAPIEN 23 Mild
8 (5) 85 M 23 No — TF Edwards SAPIEN 23 Moderate
9 (5) 91 F 25 No — TA Edwards SAPIEN 23 Trace
10 (6) 73 M 6.6 No — TF Edwards SAPIEN 26 Mild
11 (7) 64 F — No — TF CoreValve — —
12 (7) 81 M 40 No — TF CoreValve 29 Mild
13 (7) 83 F — No — TA Edwards SAPIEN 23 Trace
14 (7) 78 M — No — TF Edwards SAPIEN — —
15 (7) 66 M — No — TF CoreValve 29 Mild
16 (7) 88 M 35 No — TF Edwards SAPIEN 26 Trace
17 (7) 91 M — No — TA Edwards SAPIEN 23 Mild
18 (7) 84 M 24 No — TF CoreValve 29 Mild
19 (7) 85 M — No — TF CoreValve 29 —
20 (7) 80 M — No — TF CoreValve — —
21 (8) 84 M 23.5 No Vancomycin þciprofloxacin
TF CoreValve 29 Mild
22 (9) 86 M — No — — — — —
23 (10) 81 M 29 No Ampicillin TF CoreValve 29 Moderate
24 (11) 72 F 39.7 Yes — TF Edwards SAPIEN 23 None
25 (12) 71 M 24.3 No — — Edwards SAPIEN 23 Moderate
26 (13) 75 F — No — — Edwards SAPIEN XT 26
27 (14) 84 M — No — TA Edwards SAPIEN XT
28 (15) 70 M 33.1 No — TF CoreValve Mild
29 (16) 72 M 28.1 No Cefazolin þ gentamicin TA Edwards SAPIEN 23 Mild
30 (16) 91 F 15.2 No Teicoplanin TF CoreValve 29 Trivial
31 (16) 88 F 55.3 No Cefazolin þ gentamicin TF CoreValve 29 Mild
AR ¼ aortic regurgitation; TA ¼ transapical; TF ¼ transfemoral.
Amat-Santos et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5
Infective Endocarditis and Transcatheter Valves F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6
338
available in 15 patients in the TAVR group (47% ofstudy population, 71% of the patients alive at hospitaldischarge), with a mean length of follow-up of 11 �9 months. Two additional cases of PVE-related deathwere reported in this period, at 1 and 3 months afterhospital discharge. No follow-up information wasavailable for patients treated with TPVR, which pre-cludes long-term prognostic conclusions.
The rates of surgical management and mortality inTAVR and TPVR groups are summarized in Table 5.
DISCUSSION
PVE POST-TVR: INCIDENCE AND PREDISPOSING
FACTORS. Although the incidence of early PVEpost-TVR has been more than 2% in some smallseries, larger studies have usually reported an
incidence #1%, similar to that of surgical valve series(Online Refs. 21–25). In the PARTNER trial, the inci-dence of early PVE was 0.72%, which was comparableto the 1% rate observed in the surgical cohort (OnlineRef. 6). However, one may wonder whether a lowerrate of early PVE should be expected among TAVRpatients, considering the less invasive nature of theprocedure.
Some concerns have been raised about the ade-quacy of the sterile conditions in which the trans-catheter valves are prepared and finally implanted.Although no details were provided about the locationof TVR procedures in the reported cases of PVE, it iswell known that most procedures are performed inthe catheterization laboratory (Online Ref. 34), usu-ally not achieving the same level of sterile conditionsas an operating or hybrid room. Although the rate of
TABLE 2 Baseline Data of Patients With Previous Transcatheter Pulmonary Valve With Infective Endocarditis
20 (26) — — Congenital aortic stenosiswith Ross procedure
Homograft — — — Melody
21 (26) — — — — — — — Melody
22 (26) — — — — — — — Melody
23 (26) — — — — — — — Melody
24 (27) 8 M Tetralogy of Fallot Stentless porcine valve — — Stenosis/regurgitation Melody
25 (27) 14 M Tetralogy of Fallot Stented porcine valve — — Stenosis Melody
26 (27) 20 M Tetralogy of Fallot Bovine valve conduit — — Stenosis/regurgitation Melody
27 (27) 11 M Tetralogy of Fallot Stented porcine valve — — Stenosis/regurgitation Melody
28 (28) 14 M Tetralogy of Fallot Carpentier-Edwardsconduit
6 — Stenosis/regurgitation Melody
*Manufacturer: Medtronic (Minneapolis, Minnesota). †Median age for the series: 21.5 years (16.2 to 30.1 years). ‡Median (interquartile range).
LV ¼ left ventricular; TPVR ¼ transcatheter pulmonary valve implantation; other abbreviations as in Table 1.
J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5 Amat-Santos et al.F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6 Infective Endocarditis and Transcatheter Valves
339
bacterial infection in coronary cases is low (0.64%),positive blood cultures were obtained in 18% ofthe patients after the procedure (Online Ref. 35).Also, infective complications in the postoperativeperiod affect as many as 15% of the patients (OnlineRef. 36) and are a potential source of infectionfor PVE. Finally, the compression of the leaflets dur-ing transcatheter valve preparation and loadingcan be associated with some leaflet damage(Online Ref. 37), which indeed can favor the occur-rence of PVE.
The present study showed that the vast majority ofpatients with PVE post-TVR were men, with as manyas 66% and 90% of the patients in the TAVR and TPVR
groups, respectively. This is in accordance with pre-vious PVE studies (Online Ref. 5), in which about two-thirds of the patients with endocarditis were men.The protective asset of the female sex could bepartially explained by the hypothetical endothelialprotection by estrogen (Online Ref. 38).
The present study suggests that patients withendocarditis after TAVR are among those with thehighest risk profile, with a mean logistic EuroSCOREclose to 30% and frequent comorbidities such asdiabetes, immunosuppression (i.e., steroids, myelo-dysplastic syndromes), and renal failure that havebeen recognized as predisposing factors for PVE(Online Ref. 5).
FIGURE 3 Clinical Characteristics of TAVR Infective
Endocarditis Patients Compared With Previous
Transcatheter and Surgical Aortic Valve Series
Age* (A), sex* (B), and logistic EuroSCORE* (C) of TAVR infective
endocarditis patients (n ¼ 32) compared with patients included in
TAVR registries, the PARTNER trial, and surgical series of aortic
valve replacement with bioprosthesis. *From cases reporting
data (Online References). Abbreviations as in Figure 1.
Amat-Santos et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5
Infective Endocarditis and Transcatheter Valves F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6
340
Finally, the presence of residual AR is well recog-nized as one of the most important limitations ofTAVR. In the present study, most patients (75%) withechocardiographic data available had mild or greaterresidual AR, a much higher rate compared with pre-vious TAVR series (Online Refs. 6–9,19–22). Thepresence of residual AR as a source of endothelialdamage that may act as anchoring for the germsduring episodes of transient bacteremia and its role asa predisposing factor for PVE needs to be furtherevaluated.
In TPVR patients, underlying congenital heartdisease or the type of conduit used to repair the right-side anomalies were not related to a higher rate ofPVE, albeit the presence of homografts in patientswith PVE was w20% less frequent than in previousTPVR studies (Figure 2). On the other hand, stenoticmalfunction of the percutaneous pulmonary valves(isolated or combined with regurgitation) seemed tobe associated with a higher risk of PVE. Stenoticconduits may be more deteriorated and calcified andwith higher shear stress forces that may predispose toPVE. However, previous studies have not clearlydemonstrated a higher incidence of PVE in stenoticor regurgitant prosthetic or native valves (OnlineRefs. 26–30).
Antibiotic prophylaxis before TVR and beforedental and other invasive procedures after TVR iscurrently decided on a case-by-case basis or accord-ing to institutional protocols, with the inherent limi-tations and variability of such an individualizedstrategy. This may also play a role in the occurrenceof early PVE post-TVR. Of note, gastrointestinal pro-cedures such as colonoscopies no longer requireantibiotic prophylaxis as of 2009 European Society ofCardiology guidelines (Online Ref. 11). Even if publi-cation bias may overestimate this problem, severalpatients included in this review presented with PVEafter such procedures; it is therefore necessary todetermine whether a step back in the recommenda-tions may be necessary in TVR patients.
PVE POST-TVR: ETIOLOGY AND DIAGNOSTIC
FEATURES. Staphylococci, fungi, and gram-negativebacilli have been found to be the main causes ofearly prosthetic valve endocarditis (Online Ref. 11).These “contaminant” germs were also a frequentetiology of early PVE after TAVR (36.7%), withenterococci as the predominant causative agents ofearly PVE in this group (Table 5). Previous studiespredicted the increasing role of this pathogen as lifeexpectancy increases and more aggressive therapiesare administered to aged patients (Online Refs. 5,35).Enterococci are highly tolerant to antibiotic-induced
FIGURE 4 Clinical Characteristics of TPVR Infective Endocarditis Patients Compared With Previous TPVR Series
Age (A), sex (B), underlying disease (C), type of conduit (D), and main reasons (stenosis vs. regurgitation) for TPVR (E) in TPVR infective endocarditis patients compared
with the largest series of TPVR. *See the Online References. Abbreviations as in Figures 1 and 2.
J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5 Amat-Santos et al.F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6 Infective Endocarditis and Transcatheter Valves
341
killing, and eradication requires prolonged adminis-tration (as long as 6 weeks) of synergistic bactericidalcombinations. Moreover, these microorganisms canbe resistant to multiple drugs, including amino-glycosides, beta-lactams, and vancomycin (OnlineRef. 35). This high rate of failure of antibiotics hasmajor implications, as isolated medical managementremains the most frequent strategy for the treatmentof PVE after TAVR.
With regard to echocardiographic findings, valveprosthesis vegetation was present in about one-halfof the patients diagnosed with PVE post-TVR. InTAVR patients, complications such as abscesses(47%), fistulae (9%), or the involvement of othervalves (22%) were relatively common and much morefrequent than that observed in previous seriesincluding native and surgical prosthetic valves(Online Refs. 21–25). On the other hand, in TPVR pa-tients, the infection was limited to the valve pros-thesis (either the leaflets, stent frame, or both) in allcases.
The microbiological, structural, and clinical par-ticularities of patients treated with TVR may reduce
the sensitivity of the Duke criteria (Online Refs. 11,12).Hence, some specific recommendations in this fieldmay improve accuracy in the diagnosis of PVE (OnlineRef. 39).
PVE POST-TVR: MANAGEMENT AND OUTCOMES.
The rate of valve explantation in endocarditis studieshas been as high as 75% for native valves and 50% forsurgical bioprostheses (Online Ref. 40). In contrast,the rate of valve explantation in cases of PVE post-TAVR was 41%. Interestingly, this rate was muchhigher in balloon-expandable cases (57%) than in self-expandable ones (23%). Although many factors mayhave played a role in the differences between valvetypes, the much longer stent frame extending towardthe ascending aorta of the CoreValve (Medtronic)system may increase technical difficulties duringsurgical valve explantation and may have beenresponsible for a lower rate of valve explantation inthese cases. The high-risk profile of the patients un-dergoing TAVR may explain patients’ refusal of valveexplantation in some of these cases even if more thanone-third of them had a local extension of the
TABLE 3 Main Symptoms, Imaging/Pathological and Microbiological Findings, and Outcomes of TAVR Patients With Infective Endocarditis
ECMO ¼ extracorporeal membrane oxygenation; PS ¼ pulmonary stenosis; other abbreviations as in Tables 2 and 3.
J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5 Amat-Santos et al.F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6 Infective Endocarditis and Transcatheter Valves
343
infection and a significant number (30%) had com-plications such as heart failure and embolism that arefrequent reasons to decide on surgery (OnlineRefs. 11,12). Indeed, about two-thirds of the patientswith heart failure at admission were not operated on
despite the proven survival benefits of surgery in thatscenario (Online Refs. 11,41).
The rate of valve explantation observed in PVEpost-TPVR was as high as 75%. Apart from the factthat the TPVR population is much younger than the
FIGURE 5 Timing of Infective Endocarditis After Transcatheter Aortic Valve
Replacement and Transcatheter Pulmonary Valve Replacement
Time from transcatheter valve replacement to infective endocarditis onset of symptoms in
transcatheter aortic valve replacement and transcatheter pulmonary valve replacement
patients.
FIGURE 6 Location
Schematic location o
transcatheter aortic v
had vegetation in mu
for 19 of the 28 pati
Amat-Santos et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5
Infective Endocarditis and Transcatheter Valves F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6
344
usual IE patients, the right-side location (OnlineRefs. 41,42) and etiology of the endocarditis in suchcases (S aureus as the most frequent agent) mayexplain the high valve explantation rate (OnlineRefs. 41,43).
of Infective Endocarditis After Transcatheter Valve Replacement
f infective endocarditis according to echocardiographic and/or pathological fin
alve replacement (information available in all 32 patients included in the revie
ltiple locations. (B) Location of infective endocarditis in patients with previo
ents included in the review).
Interestingly, the surgical explantation of many ofthese infected transcatheter valves and previousautopsy series (Online Refs. 43–45) have contributedto a better understanding of the predisposinganatomic factors, including extensive inflammatoryreactions (Online Ref. 45), infection of the skirt andleaflets with extension and perforation of adjacentstructures (Online Ref. 44), and (more controversial)the lack of valve endothelialization or thromboticcomplications (Online Refs. 46–49). Also, Loeseret al. (Online Ref. 43) reported that signs suggestiveof PVE after TAVR may be more frequent thancommonly thought.
PVE has been associated with a high mortality rate(20% to 40%), with no major improvements in thesurvival rate of this life-threatening disease in thepast 30 years (Online Ref. 11). The mortality rateobserved in TAVR patients who had PVE (34%) was inaccordance with such data, even though we cannotexclude a potential underestimation of the realmortality rate due to a publication bias (authors maytend to publish the cases that end well). The 7%mortality rate in TPVR patients that may seem lowappears to be too high if we take into considerationthe young population involved. Overall, this high-lights the importance of maximizing the measures ofasepsis and appropriate antibiotic prophylaxis andprompts us to think about the most appropriatestrategy for the treatment of PVE post-TVR. Thepossibility of earlier and more frequent valve
dings. (A) Location of infective endocarditis in patients with previous
w). Thirteen patients had vegetation in 1 location, whereas 3 patients
us transcatheter pulmonary valve replacement (information available
TABLE 5 Comparative Information of TAVR and TPVR
Patients With IE
VariableTAVR
(n ¼ 32)TPVR
(n ¼ 28)
Initial symptoms
Fever 25 (78.1) 24 (85.7)
Heart failure 8 (25) 6 (21.4)
Neurological 4 (12.5) 0 (0)
Early IE 24 (95.2) 18 (69.6)
Echocardiographic characteristics
Presence of vegetation 12 (37.5) 13 (46.4)
Located in the transcatheter valve 23 (71.9) 28 (100)
Microbiological characteristics
Typical microorganisms* 17 (53.1) 8 (47.1)
Contaminant agents in early IE† 11 (36.7) 9 (56.2)
Staphylococcus aureus 2 (6.2) 5 (29.4)
Staphylococcus epidermidis 4 (12.5) 1 (5.9)
Enterococci 11 (34.4) 0 (0)
Management and outcomes
Surgical treatment 13 (40.6) 21 (75.0)
IE-related death 11 (34.4) 2 (7.1)
Values are n (%). Percentage provided according to the total number of patientswho provided the issued information. *Typical microorganisms according tomodified Duke criteria (16): Streptococcus viridans, Streptococcus bovis, HACEKgroup, Staphylococcus aureus, or enterococci. †Contaminant microorganismsinclude those more frequently leading to early prosthetic valve endocarditis(Staphylococcus epidermidis, Staphylococcus aureus, Gram-aerobic agents, andfungi).
Abbreviations as in Tables 1 to 3.
J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5 Amat-Santos et al.F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6 Infective Endocarditis and Transcatheter Valves
345
explantation in such cases to improve prognosis(Online Ref. 50) may be considered. Although majordifferences in the clinical profile between the TAVRand TPVR groups probably explain the differences inmortality rate, the higher rate of surgical treatmentamong TPVR patients may have contributed to thebetter outcomes compared with TAVR candidates.Further studies are needed to evaluate whether sur-gery improves outcomes despite the complex inter-vention and the risk of recurrent infection estimatedat 15% after PVE (Online Refs. 40,51).
STUDY LIMITATIONS. The present study has thelimitations inherent to a systematic review that col-lects only the information described in the publica-tions. Therefore, there might be relevant informationomitted in the publications that could shed some lighton this limitation. This also includes incompleteechocardiographic data, particularly regarding thecharacteristics (size, mobility) of the vegetation. Inaddition, all of the published papers found wereeither case reports or very small series, precludingcomparison with the entire TVR population at risk.Additionally, the patients reported might have ten-ded to have a better outcome than those who werenot reported (selection bias).
CONCLUSIONS
PVE is an uncommon but life-threatening complica-tion after TVR. Although the conditions of asepsisare frequently less strict than in surgical in-terventions, the incidence of early PVE post-TVR re-mains low (usually #1%) and similar to that ofsurgical series. In the TAVR population, this compli-cation seems to be more frequent in male patientsand in those with higher risk profile. In TPVR pa-tients, PVE seems to occur more frequently in malepatients with a stenotic (vs. regurgitation) conduit/valve as the main underlying disease. Although earlyPVE is considered to be acquired during the peri-procedural time, the low rate of classic contaminantagents in favor of others such as enterococci, espe-cially among TAVR patients, may suggest alternativesource of infection. This is of major clinical impor-tance because alternative antibiotic prophylaxisprotocols may reduce the incidence of the disease.About two-thirds of the TAVR-PVE patients weremanaged medically, despite the fact that more thanone-half of the patients had complications such aslocal extension, embolism, and heart failure. Themortality rate of PVE in these patients was high(more than 30%) and similar to that described inprevious PVE studies. Most TPVR-PVE patients weremanaged surgically and underwent surgical explan-tation of the infected conduit valve. However, themortality rate remained at 7% despite the very youngage of this population.
The syndromic characteristics of PVE vary accord-ing to the underlying disease and the microorganisminvolved. PVE in transcatheter valve carriers repre-sents a paradigm shift in PVE profile, involving veryold (and high risk) or very young patients, both with ahigh rate of health-care procedures. This systematicreview represents a first step toward a better under-standing of the profile conditions and predisposingfactors, etiology, management, and prognosis of PVEin patients with transcatheter valves. Future studieswill have to determine the potential usefulnessof improving asepsis/antibiotic prophylaxis in thischallenging group of patients and finding a betterstrategy for an earlier diagnosis and better manage-ment to decrease the incidence of PVE and increasethe survival associated with this life-threateningcomplication.
Amat-Santos et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 8 , N O . 2 , 2 0 1 5
Infective Endocarditis and Transcatheter Valves F E B R U A R Y 2 0 1 5 : 3 3 4 – 4 6
346
RE F E RENCE S
1. Seok Koh Y, Hyoung Moon M, Hyun Jo K, WookKim H. Infective endocarditis in transcatheteraortic valve implantation. Eur J Cardiothorac Surg2014;45:582.
2. Santarpino G, Fischlein T, Pfeiffer S. Prostheticvalve endocarditis 6 months after transcatheteraortic valve implantation. G Ital Cardiol 2013;14:138–40.
3. Wilbring M, Tugtekin SM, Matschke K,Kappert U. Surgery for fulminant prosthetic valveendocarditis after transapical transcatheter aorticvalve-in-valve implantation. Thorac CardiovascSurg 2014;62:80–2.
5. Puls M, Eiffert H, Hünlich M, et al. Prostheticvalve endocarditis after transcatheter aortic valveimplantation: the incidence in a single-centre cohortand reflections on clinical, echocardiographic andprognostic features. EuroIntervention 2013;8:1407–18.
6. Castiglioni A, Pozzoli A, Maisano F, Alfieri O.Endocarditis after transfemoral aortic valveimplantation in a patient with Osler-Weber-Rendusyndrome. Interact Cardiovasc Thorac Surg 2012;15:553–4.
7. Eisen A, Shapira Y, Sagie A, Kornowski R. Infec-tive endocarditis in the transcatheter aortic valvereplacement era: comprehensive review of a rarecomplication. Clin Cardiol 2012;35:E1–5.
8. Quiroga B, Arroyo D, Verde E, Eworo A, Luño J.Infective endocarditis on a percutaneous pros-thetic aortic valve with associated glomerulopathydue to Granulicatella adjacens. Braz J Infect Dis2012;16:601–2.
9. García-Pardo H, Revilla A, Sevilla T, López J,Ortiz C, San Román JA. Staphylococcus aureusendocarditis on transcatheter aortic valves. RevEsp Cardiol 2012;65:771–3.
10. Citro R, Mirra M, Baldi C, Prota C, et al.Concomitant dynamic obstruction and endocardi-tis after “valve in valve” TAVI implantation. Int JCardiol 2013;167:e27–9.
11. Bozda�g Turan I, Kische S, D Ancona G,Nienaber CA, Ince H. Suspected endocarditis afterCoreValve� implantation: a word of caution.Anadolu Kardiyol Derg 2013;13:395–6.
13. Hidiroglu M, Kucuker A, Erdogan KE. Infectiveendocarditis after transcatheter aortic valveimplantation. Interact Cardiovasc Thorac Surg2013;S2:S115–6.
14. Zytowski M, Erb M, Albes JM, Hartrumpf M.Infective endocarditis 4 months after trans-apical aortic valve implantation with EdwardsSAPIEN� XT. Eur J Cardiothorac Surg 2013;44:769.
15. Orban M, Sinnecker D, Mair H, et al. Trans-catheter aortic-valve endocarditis confirmed bytransesophageal echocardiography. Circulation2013;127:e265–6.
16. Aung T, Poon K, Horvath R, Coulter C,Walters DL. A case series of medically managedinfective endocarditis after transcatheter aortic valvereplacement. Scand J Infect Dis 2013;45:489–93.
17. Atamanyuk I, Raja SG, Kostolny M. Bartonellahensealae endocarditis of a percutaneouslyimplanted pulmonary valve. J Heart Valve Dis2011;20:94–7.
18. Bhat DP, Forbes TJ, Aggarwal S. A case of life-threatening staphylococcus aureus endocarditisinvolving percutaneous transcatheter prostheticpulmonary valve. Congenit Heart Dis 2012;8:E161–4.
19. Buber J, Bergersen L, Lock JE, et al. Blood-stream infections occurring in patients withpercutaneously implanted bioprosthetic pulmo-nary valve: a single-center experience. CircCardiovasc Interv 2013;6:301–10.
20. Butera G, Milanesi O, Spadoni I, et al. MelodyTranscatheter pulmonary valve implantation.Results from the Registry of the Italian Society ofPediatric Cardiology. Catheter Cardiovasc Interv2013;81:310–6.
21. Eicken A, Ewert P, Hager A, et al. Percutaneouspulmonary valve implantation: two-centre expe-rience with more than 100 patients. Eur Heart J2011;32:1260–5.
22. Yedidya I, Stein GY, Vaturi M. Positron emis-sion tomography/computed tomography for thediagnosis of endocarditis in patients with pulmonicstented valve/pulmonic stent. Ann Thorac Surg2011;91:287–9.
23. Patel M, Iserin L, Bonnet D, Boudjemline Y. Atyp-ical malignant late infective endocarditis of Melodyvalve. J Thorac Cardiovasc Surg 2012;143:e32–5.
24. Gillespie MJ, Rome JJ, Levi DS, et al. Melodyvalve implant within failed bioprosthetic valves inthe pulmonary position: a multicenter experience.Circ Cardiovasc Interv 2012;5:862–70.
25. Alsoufi B, Al-Joufan M, Al-Omrani A, Bulbul Z.Obstruction of a percutaneous pulmonary valve byan Aspergillus mycotic thrombus mimickingmassive pulmonary embolus. Ann Thorac Surg2012;94:e5–6.
26. Lurz P, Coats L, Khambadkone S, et al. Percu-taneous pulmonary valve implantation: impact ofevolving technology and learning curve on clinicaloutcome. Circulation 2008;117:1964–72.
27. Villafane J, Baker GH, Austin EH 3rd, et al.Melody pulmonary valve bacterial endocarditis:experience in four pediatric patients and review ofthe literature. Catheter Cardiovasc Interv 2014;84:212–8.
28. Johnson JN, Miller SG, Lodge AJ. Corynebac-terium endocarditis of a percutaneously placedtranscatheter pulmonary valve. Cardiol Young2014;5:932–4.
KEY WORDS infective endocarditis,transcatheter aortic valve implantation,transcatheter pulmonary valve implantation
APPENDIX For a supplemental table andreferences, please see the online version of thisarticle.