Summary
Background
Percutaneous pulmonary valve replacement (PPVR) has achieved standard of care for the management of dysfunctional right ventricular outflow tract.
Aim
Because of increasing reports of Melody ® valve (Medtronic, Inc., Minneapolis, MN, USA) infective endocarditis (IE), we aimed to evaluate its incidence and predictors of occurrence.
Methods
We collected data on all patients who received a Melody valve in the pulmonary position from 2009 to 2012.
Results
A total of 86 consecutive patients underwent PPVR from 2009 to 2012; of these, five developed IE (5.8%). Freedom from IE was 91% at 50 months. Baseline demographics, type of conduit, PPVR procedural success, residual gradients and in situ duration of Melody valve were similar in IE and non-IE patients. A significant number of IE patients had additional unprotected invasive procedures during follow-up and abruptly discontinued antiplatelet therapy ( P = 0.0139 and P = 0.002, respectively). The cumulative probability of survival without cardiovascular events for IE patients was 20% at 20 months, compared with 98.1% for non-IE patients ( P < 0.0001). Death was statistically associated with IE.
Conclusion
Early- and late-onset IE of the Melody valve is emerging as a catastrophic complication of PPVR. Abrupt aspirin discontinuation and additional unprotected invasive procedures during follow-up are significant predictors of Melody valve IE. Owing to its rapidly progressive nature, aggressive invasive management should not be delayed.
Résumé
Contexte
Le remplacement valvulaire percutané (RVP) est désormais considéré comme un traitement de première ligne des dysfonctions de la voie d’éjection droite.
Objectif
À cause de rapports croissants d’endocardites sur Melody ® (Medtronic, Inc., Minneapolis, MN, États-Unis), nous avons évalué son incidence et les facteurs de risque de survenue.
Méthodes
Nous avons collecté les données de tous les patients ayant eu la mise en place d’une valve Melody en position pulmonaire de 2009 à 2012.
Résultats
Quatre-vingt-six patients consécutifs ont reçu une valve Melody. Cinq ont eu une endocardite, soit une incidence de 5,8 %. La survie actuarielle sans endocardite était de 91 % à 50 mois. Les données démographiques, le type de conduits, le gradient résiduel, la durée de la valve étaient similaires dans les groupes avec et sans endocardites. Un nombre significatif de patients avec endocardites ont eu des actes invasifs non protégés au cours du suivi et ont interrompu leur traitement anti-plaquettaire subitement (respectivement p = 0,0139 et p = 0,002). Les probabilités cumulées de survie sans événement cardiovasculaire étaient de 20 % et 98,1 % à 20 mois, respectivement dans les groupes avec endocardites et sans endocardite ( p < 0,0001). Le décès était statistiquement associé à l’endocardite.
Conclusion
La survenue d’endocardites de la valve Melody émerge comme une complication catastrophique du remplacement valvulaire percutané. L’interruption brutale de l’aspirine et les actes invasifs non protégés au cours du suivi sont prédictifs d’endocardites sur Melody. En raison de leur caractère sub-aigu, il est nécessaire de les traiter de manière agressive et rapide.
Background
Percutaneous pulmonary valve replacement (PPVR) is now recognized as a treatment for dysfunctional right ventricular outflow tract (RVOT) . The device was initially developed with the aim of extending the lifespan of conduits surgically placed between the right ventricle (RV) and the pulmonary arteries (PAs) in patients with various congenital heart diseases (CHDs) . After more than a decade, its utility has evolved to include patients with dysfunctional native outflow tracts . The Melody ® valve (Medtronic, Inc., Minneapolis, MN, USA) is used most frequently for PPVR, and is made of a bovine jugular vein valve mounted on a platinum stent. PPVR using the Edwards SAPIEN valve (Edwards Life Sciences, Irvine, CA, USA) has been reported; however, long-term follow-up outcomes are awaited . PPVR can be safely performed with very low morbidity and mortality and excellent long-term haemodynamic results in the majority of patients ; it offers a good alternative to surgery for patients with haemodynamically significant RVOT diseases. Despite such promising results, there are increasing numbers of reports of Melody valve infective endocarditis (IE) . So far, however, there is little information on the incidence, predictors and potential risk markers associated with IE in patients with Melody valves .
This study was undertaken to evaluate the incidence of IE in a cohort of patients who had undergone Melody PPVR, and to identify potential risk markers in patients who developed IE in this prospective survey.
Methods
The primary endpoint was to evaluate the incidence and predictors of IE, and to extensively study the profile of patients with IE. IE was diagnosed based on modified Duke criteria with occurrence of bacteraemia with at least two positive blood cultures and vegetations seen on transthoracic or transoesophageal echocardiography or positive positron emission tomography (PET) scan.
The secondary endpoint was to identify associated risk markers, which were divided into three groups: preprocedural characteristics, perprocedural factors and postprocedural data.
Patient selection
All consecutive patients who underwent Melody PPVR for dysfunctional RVOT between April 2009 and June 2012 at a single institution were included in this prospective study. The study was conducted in accordance with local ethics committee guidelines; informed consent was obtained from patients and/or their parents after explanation of the procedure.
At the time of the procedure, no patient had ongoing infection. During the procedure, prophylactic antibiotics (cefamandol) were administered and continued for 24 hours (three doses in total). In case of allergy, vancomycin was given once during the procedure. Heparin was given during the procedure, followed by low-molecular-weight heparin for the next 24 hours (three doses in total). Patients were prescribed lifelong full-dose aspirin (100 mg/day) after PPVR or clopidogrel in case of allergy.
Preprocedural, perprocedural and postprocedural data were collected prospectively. Demographic characteristics, numbers of surgeries and cardiac catheterizations, type of conduit, history of severe infection and endocarditis, co-morbidities (defined as significant chronic medical conditions, including genetic abnormalities, undefined syndromes, chronic infection) and echocardiography were recorded for all patients. Procedural data (length of the procedure, access, haemodynamic data, associated procedures and complications) were recorded.
After PPVR, all patients were followed up at 1, 3, 6 and 12 months and yearly. The last gradient across the Melody valve on echocardiogram was reviewed. The evolution of the gradient over time was evaluated by subtraction of the haemodynamic gradient measured in the catheterization laboratory after Melody PPVR from the gradient measured on the last echocardiogram. The gradient over the RVOT at the time of IE was also compared with the gradient at the last echocardiogram before IE. Blood work and imaging modalities to diagnose IE, the evolution of the disease, the time to diagnosis and management, including surgery, were noted.
Statistical analysis
SAS software version 9.3 (SAS Institute, Inc., Cary, NC, USA) was used for statistical analysis. Nominal variables are expressed as numbers and percentages and were compared using Fisher’s exact test or the chi 2 test, as appropriate. Ordinal variables are presented as means ± standard deviations and were compared using Wilcoxon’s rank sum test. Continuous variables are expressed as means ± standard deviations and were compared using the independent variables t -test. All tests were two-sided and a P value < 0.05 was considered statistically significant. Survival curves were constructed using the Kaplan–Meier method. Groups were compared using log-rank statistics. The risk factors for IE were studied using univariate Cox proportional hazards survival analyses. The risk markers of IE were studied by univariate and multivariable Cox proportional hazards survival analyses. The variables used in the multivariable analysis were those with a P value < 0.05 in the univariate analysis.
Results
From 2009 to 2012 (a 4-year period), 86 consecutive patients underwent Melody PPVR. Five patients developed IE (5.8%, 95% CI 0.9–10.7%) during a mean follow-up period of 23.6 ± 15 months (range 2.6–28.3 months after Melody PPVR). The detailed profile of patients with IE is presented in Table 1 . Three patients had early (< 1 year) and two patients had late (> 1 year) Melody valve IE. At admission, all patients had a history of fever and malaise. Microorganisms were identified in all patients. Vegetations were seen on echocardiography (transthoracic or transoesophageal) in three patients, one patient had a positive PET scan and one patient presented with cardiogenic shock and severe obstruction of the RVOT, but no thrombus or vegetation was seen by standard imaging. Autopsy examination showed well-preserved stent struts with large loose obstructive vegetations made of fibrinous material filling up the cusps. Microorganisms were found in the histology and valve culture.
| Patient 1 | Patient 2 | Patient 3 | Patient 4 | Patient 5 | |
|---|---|---|---|---|---|
| Age at presentation (years) | 18 | 26 | 46 | 50 | 14 |
| Sex | Female | Male | Male | Male | Male |
| Underlying diagnosis | TOF-PA | TOF-PA | Rheumatic aortic stenosis; Ross procedure | TOF | Congenital aortic stenosis; Ross procedure |
| Co-morbidities | None | None | Heterozygote beta thalassaemia | Di George Syndrome; HCV | Portal cavernoma; oesophageal varices |
| Indication for PPVR | Obstruction | Mixed | Obstruction | Mixed; RV failure | Mixed |
| RVOT characteristics and size | Hancock 16 | Hancock 18 | Homograft | Homograft 26 | Carpentier-Edwards 20 |
| Surgery to PVR (years) | 14 | 15 | 15 | 11 | 4 |
| PPVR IE (months) | 28.3 | 2.6 | 4.5 | 26 | 11.9 |
| Previous IE | None | None | None | None | None |
| Previous infection | None | Cerebral abscess | None | None | None |
| Residual lesions: maximal velocity | 4.7 m/s | 3.5 m/s | 2.2 m/s | 2.2 m/s | 2.1 m/s |
| Presentation | Fever; RVOTO; RV failure; V max 5.5 m/s | Fever; RVOTO; RV failure; V max 5.9 m/s | Fever; V max 3 m/s | Fever; RVOTO; RV failure; V max 3.5 m/s | Fever; RVOTO; V max 3.3 then 4.6 m/s |
| Abrupt AAT discontinuation | Yes | Yes | No | Yes a | No b |
| Cardiac catheterization | Yes | No | No | Yes | No |
| Invasive procedure post-PPVR | Yes | No | No | Yes | Yes b |
| Potential source/portal of entry | Lingual piercing; abortion (3 months before IE) | Multiple tooth decay | None | Flutter ablation 3 months before IE | Hepatic biopsy b |
| Involvement of other valves | None | None | None | None | Aortic valve |
| Location of vegetations | PA side | PA side | PA and RV | PET scan + | PA side |
| Diagnosis with TTE | + | + | |||
| Diagnosis with TOE | + | ||||
| Other | Post mortem | PET scan | |||
| Pulmonary valvular gradient | Severe | Severe | Moderate | Severe | Severe |
| Causative microorganism from blood culture | Streptococcus sanguis | Staphylococcus epidermidis | Staphylococcus epidermidis ; methicillin-sensitive | Staphylococcus aureus ; methicillin-sensitive | Streptococcus sanguis |
| Initial treatment planned | Surgery; IV antibiotics (amoxicillin/gentamicin) | Surgery planned; IV antibiotics (gentamycin, tazobactam-piperacillin, vancomycin) | IV antibiotics (oxacillin/gentamycin) | IV antibiotics (oxacillin/pefloxacin); balloon dilatation of Melody | IV antibiotics (amoxicillin/gentamycin); bare-metal stent implantation |
| Outcome | Cardiogenic shock; death before surgery | Cardiogenic shock; death before surgery | Alive; no surgery; stable gradient | Severe RV failure; no surgery; death 6 months after IE | Alive and well after surgery |
a On warfarin; AAT stopped 3 months before IE.
b Hepatic biopsy complicated by sepsis 6 months before IE; transient discontinuation of AAT during sepsis due to thrombocytopoenia.
All patients with IE had an increase in their gradient across the RVOT when comparing gradients on the last echocardiogram and at hospital admission, confirming the involvement of the Melody valve.
Patients 1, 2 and 5 presented with rapidly progressive severe obstruction and heart failure. Patients 1 and 2 died (ventricular fibrillation) within 12 hours of hospital admission: one at arrival in the operating room; and one in the cardiac intensive care unit during the night before the surgery. Patient 5 needed cardiac catheterization to relieve obstruction of the RVOT (bare-metal stent insertion) and subsequent surgery to explant the Melody and treat concomitant aortic valve IE. Patient 4 with Staphylococcus aureus IE was stabilized with antibiotics, and balloon dilatation of RVOT obstruction (RVOTO) was performed before discharge. However, this patient developed severe heart failure without residual RVOTO and died 6 months after the onset of IE, despite sustained sterilization of the IE. Patient 3 was treated medically and is currently free of infection.
Preprocedural data
Patients with IE tended to be older than those without IE (mean age at implantation 29.8 ± 16.3 years vs 23.5 ± 10 years for patients without IE) ( Table 2 ). Most of the patients with IE were men (4 of 5). Co-morbidities were noted in 18 patients (20%) and were associated with a trend towards developing IE ( P = 0.06). The numbers of previous surgeries, cardiac catheterizations and stent implantations in the RVOT, the type of conduit, the maximal gradient across RVOT and the interval between surgery and PPVR were similar in both groups. In this cohort, five patients had a history of IE and six had a history of severe infections (all > 1 year before PPVR); these were not found to be a risk marker for IE ( P = 0.34 and 1, respectively). The presence of a stent in the RVOT was not associated with occurrence of IE.
