Summary
Background
Vascular complications (VCs) after transcatheter aortic valve implantation (TAVI) are frequent and their management is challenging.
Aim
To report the incidence, predictors and management of VCs following percutaneous transfemoral TAVI (TF-TAVI) at a single centre.
Methods
We analyzed 102 consecutive patients who underwent percutaneous TF-TAVI between August 2008 and December 2013. All endpoints were evaluated at 30 days and 6 months according to Valve Academic Research Consortium-2 criteria. VC percutaneous treatment success was defined as residual stenosis < 30%, absence of blood extravasation and absence of surgical or repeat endovascular intervention at 30 days.
Results
Twenty-two patients (22%) experienced VCs, including five patients (5%) with major VCs. Mortality at 30 days was significantly higher in patients with major VCs than in patients without major VCs (60% vs 3%; P = 0.001). Patients with VCs had more life-threatening or major bleeding (23% vs 5%; P = 0.02), but no difference in terms of need for blood transfusion was observed. Endovascular treatment was used in 13 of 22 patients with VCs (59%) and was successful in 11 of these 13 patients (85%). Primary surgical repair was necessary in only 1/22 (5%) patients, for a common femoral artery pseudoaneurysm 2 weeks after the TAVI procedure.
Conclusions
VCs following TF-TAVI are frequent. Major but not minor VCs are associated with increased mortality. Percutaneous management of VCs is feasible and safe, and surgery is rarely needed.
Résumé
Contexte
Les complications vasculaires (CV) liées au remplacement valvulaire aortique percutané (RVAP) sont fréquentes et leur prise en charge reste complexe.
Objectif
Rapporter l’incidence, les facteurs prédictifs et la prise en charge des CV liées au RVAP transfémoral (TF) dans notre institution.
Méthodes
Analyse basée sur 102 patients consécutifs ayant bénéficié d’un RVAP TF entre août 2008 et décembre 2013. Tous les critères d’évaluation étaient établis conformément au Valve Academic Research Consortium-2 et évalués à 30 jours et 6 mois. Le succès du traitement était défini par l’absence d’une sténose résiduelle > 30 %, l’absence d’un saignement résiduel, et l’absence d’une reprise endovasculaire ou chirurgicale à 30 jours.
Résultats
Vingt-deux patients (22 %) ont développé des CV, dont 5 % de CV majeures. La mortalité à 30 jours était significativement supérieure chez les patients avec des CV majeures que chez les patients sans CV majeure (60 % vs 3 %; p = 0,001). Les taux de saignements critiques ou majeurs étaient plus élevés parmi les patients avec une CV (23 % vs 5 %; p = 0,02), sans qu’aucune différence en termes de besoin de transfusion sanguine n’ait été mise en évidence. Cinquante-neuf virgule un pour cent des patients avec une CV (13/22) ont bénéficié d’un traitement endovasculaire, avec un taux de succès de 85 % (11/13). Une réparation chirurgicale en première intention n’a été nécessaire que chez 1/22 patients (5 %) pour la prise en charge d’un pseudo-anévrisme de l’artère fémorale commune 2 semaines après le RVAP.
Conclusions
Les CV liées au RVAP sont fréquentes. Les CV majeures, contrairement aux mineures, sont associées à une mortalité augmentée. Un traitement endovasculaire des CV est dans la majorité des cas envisageable.
Background
As the population is aging, physicians are increasingly confronted by patients with multiple morbidities and very elderly patients with symptomatic aortic valve stenosis (AS) who are not eligible or are at high risk for surgical aortic valve replacement. According to data from randomized trials, transcatheter aortic valve implantation (TAVI) has become the new standard of care for inoperable patients and is a valid (if not superior) option for high-risk patients . During the last decade, the reduction in diameter of the delivery systems, in conjunction with better preoperative vascular screening and increasing operator experience have led to a reduction in vascular complications (VCs). However, VCs remain the most frequent adverse events associated with transfemoral TAVI (TF-TAVI) procedures . In the early days of TAVI use, such complications were mainly managed surgically under general anaesthesia. With growing TAVI experience, endovascular treatment of VC has been increasingly adopted. In this report, we describe a single-centre experience of the incidence, predictors, management and outcomes of VCs at 30 days and 6 months in patients undergoing percutaneous TF-TAVI.
Methods
Patient population
Between August 2008 (the beginning of the TAVI programme at our institution) and December 2013, 102 consecutive patients underwent purely percutaneous TF-TAVI for symptomatic, severe AS, severe aortic regurgitation (one patient) or degenerated bioprosthesis (one patient). All procedures were performed by two operators (S. N., M. R.) using the self-expanding CoreValve ® device (Medtronic Inc., Minneapolis, MN, USA), which requires an 18-French introducer sheath for the four different available sizes (23, 26, 29 and 31). TF-TAVI accounted for 92% of the TAVI volume during this period of time. As four patients underwent planned, surgical, femoral cut-down because of significant vascular calcification, purely percutaneous TF-TAVI corresponded to 89% of our TAVI experience. Patients were selected for TF-TAVI by a local heart team, involving cardiac surgeons, interventional cardiologists, cardiovascular anaesthesiologists and intensive care specialists. Risk scores, such as the logistic EuroSCORE and the Society of Thoracic Surgeons’ (STS) score, were calculated as part of the evaluation. Initially, frailty was assessed using the “eye ball test”, but from 2012 onwards a standardized assessment was performed using the gait-speed test, grip strength assessment, serum albumin concentration and evaluation of recent weight loss, falls and mobility impairment (due to severe musculoskeletal or neurological disorders). All patients gave written informed consent to the TAVI procedure and the use of related data for research and publication.
Preoperative evaluation
All patients underwent a complete assessment of medical history, a physical examination, transthoracic echocardiography and baseline biological screening to assess blood count, coagulation, electrolyte and brain natriuretic peptide concentrations, as well as kidney and liver function. Aortic annulus and root dimensions were systematically assessed using multislice computed tomography (MSCT) or three-dimensional transoesophageal echocardiography versus magnetic resonance imaging for patients with a creatinine clearance < 40 μmol/L. An abdominal aortography (including iliofemoral vessels) and selective femoral angiographies were routinely performed at the time of the coronary angiogram and right heart catheterization, to assess the iliofemoral axis. The minimal vessel diameter (at least 6 mm), tortuosity and calcification were specifically evaluated. An additional vascular assessment with MSCT (extension of the cardiac MSCT down to the femoral bifurcation) was performed for the first time in December 2010 and routinely since January 2012.
Procedure
At the beginning of the TAVI intervention, the femoral angiography performed during screening was routinely reviewed to visualize the relation between the femoral head, the inferior epigastric artery and the femoral bifurcation. Once vascular access was obtained, femoral angiography was performed to document correct positioning. The self-expandable Medtronic CoreValve System was introduced through an 18-French introducer sheath (in the majority of cases Cook Medical, Bloomington, Indiana, USA) or a 19-French balloon expandable SoloPath ® sheath (Onset Medical, a subdivision of Terumo Medical Corporation, Irvine, CA, USA). The cutoff limit to use the introducer sheath was a 6-mm iliofemoral diameter. The procedure steps have been described in detail . Preclosure was performed in all patients using the Prostar ® XL 10 suture-based vascular closure device (Abbott Vascular, Reedwood City, CA, USA). At the end of the procedure, final crossover angiography of the main vascular access was systematically performed from the contralateral femoral puncture site to verify arteriotomy closure success as well as iliofemoral vessel integrity. The antithrombotic regimen during the procedure consisted of aspirin and a standard dose of unfractionated heparin (5000 U), with activated clotting time checks when the procedure was longer than expected. Initially a loading dose of clopidogrel was administered the night before the procedure, but since 2012 clopidogrel has been given immediately after the procedure and for 3 months — when no vitamin K antagonist is required — followed by aspirin or clopidogrel monotherapy indefinitely.
Management of vascular complications
Dissections detected during the final angiography were treated by balloon angioplasty and, in cases of a suboptimal angioplasty result, self-expanding uncovered nitinol stents were implanted (S.M.A.R.T. ® [Cordis Corporation, Hialeah, FL USA] and Misago ® [Terumo Corporation, Tokyo, Japan]). Incomplete arteriotomy closure or iliofemoral leakages were managed first, depending on the severity of the findings, with prolonged manual compression, balloon occlusion of the iliac axis using the contralateral access, reversal of unfractionated heparin with protamine administration or a combination of these manoeuvres. Covered stents were implanted in cases of persistent vascular leakage (other than trace) despite prolonged manual compression. Femoral pseudoaneurysm resistant to manual or echocardiography guided compression were treated, depending on their location, with thrombin injection, covered stent implantation or surgery. The covered stents used were either self-expanding elgiloy stents covered with a layer of polyethylene terephthalate (WALLGRAFT ® ; Boston Scientific, Natick, MA, USA) or self-expanding nitinol stents covered with polytetrafluoroethylene (Fluency ® ; Bard Peripheral Vascular, Phoenix, AZ, USA). VC percutaneous treatment success was defined as residual stenosis < 30%, absence of blood extravasation at final angiography and absence of surgical or repeat endovascular intervention at 30 days. Endovascular treatment of VCs was performed by two interventional cardiologists (M. R., S. N.), one of whom (M. R.) is trained in peripheral vascular treatment and performs more than 100 peripheral vascular interventions per year.
Study endpoints
All endpoints were evaluated at 30 days and 6 months following TAVI according to the Valve Academic Research Consortium-2 (VARC-2) criteria . VCs were stratified as major whenever a vascular injury led to death, life-threatening or major bleeding, visceral ischaemia or neurological impairment. Minor VCs were defined as vascular injury not leading to adverse events defining a major VC. A bleeding event was considered as life-threatening if it occurred in vital organs (i.e. intracranial, intraspinal, intraocular or pericardial), provoked haemorrhagic shock or needed at least 4 units of packed red blood cells, or whenever a > 5 g/dL drop in haemoglobin was reported. Bleeding was considered as major when overt bleeding was either associated with a > 3 g/dL drop in haemoglobin or required at least 2 units of packed red blood cells. The study primary outcome measure was the occurrence of any VC. The secondary outcome measure included 30-day incidence of life-threatening or major bleeding and 30-day as well as 6-month overall and cardiovascular mortality. An expert interventional cardiologist performed endpoint adjudication.
Data collection
Baseline characteristics, periprocedural data and adverse events during the index hospital stay, and at 30-day and 6-month clinical follow-up were routinely collected as part of our local prospective registry approved by the local ethics committee. Procedure angiographic images and iliofemoral MSCT, when available, were analyzed retrospectively with respect to iliofemoral calcification, atherosclerosis, tortuosity and femoral bifurcation location. A femoral artery calcification score using angiographic images just before contrast injection was adapted from the coronary artery calcification score first described by Yamanaka et al., allowing for a semiquantitative and reproducible assessment of iliofemoral vessel calcification . Briefly, it is defined as follows: 0 = no calcification; 1 = spotty wall calcification; 2 = unilateral linear calcification; 3 = unilateral linear calcification with spotty wall calcification on the opposite wall; 4 = bilateral linear calcification. The distance between the puncture site and the inferior epigastric artery as well as the femoral bifurcation was systematically measured on the final fluoroscopic acquisition using the most contrast-filled image. We used the slight Prostar-induced vessel stenosis to identify exact location of the puncture site. Finally, to assess the effect of the learning curve on incidence of VCs, the cohort was arbitrarily separated in two halves (51 patients in each group) before executing the statistical analysis.
Statistical analysis
The distribution of continuous variables was assessed using the Shapiro–Wilks test. Normally distributed variables are presented as means ± standard deviations and non-normally distributed variables as medians [interquartile ranges]. Differences were compared using Student’s t -test or the Wilcoxon rank-sum test, as appropriate. Categorical data are expressed as numbers and frequencies (%), and were compared with Fisher’s exact test or Pearson’s χ 2 test. Multivariable logistic regression was used to identify independent predictors of VCs. Variables with a P -value < 0.1 were entered in the regression model, and variables known to be associated with an increased risk of VCs (sex and vascular calcification) were included a priori into the analysis. All statistical analyses were performed using Stata SE (Stata Corp, College Station, TX, USA; 2011).
Results
Patient population
The baseline characteristics and echocardiographic variables of the 102 consecutive patients with percutaneous TF-TAVI interventions performed at our centre are presented in Table 1 . Table 2 compares the baseline characteristics of the patients with and without VCs; no significant differences between the two groups in terms of risk factors, medical history or preprocedure echocardiographic characteristics were found.
| Women | 64 (62.8) |
| Age (years) | 85.0 [76.0–94.0] |
| BMI (kg/m 2 ) | 25.6 ± 5.3 |
| Dyslipidaemia | 56 (54.9) |
| Diabetes mellitus | 30 (29.4) |
| Smoking | |
| Never or past | 96 (94.1) |
| Current | 6 (5.9) |
| Hypertension | 75 (73.5) |
| Chronic obstructive pulmonary disease | 25 (24.5) |
| Peripheral vascular disease | 20 (19.6) |
| Coronary artery disease | 54 (52.9) |
| Previous myocardial infarction | 20 (19.6) |
| Previous CABG | 13 (12.8) |
| Previous PCI | 42 (41.2) |
| Previous cerebral stroke | 10 (9.8) |
| NYHA | |
| 0–2 | 23 (22.6) |
| 3–4 | 79 (77.5) |
| GFR (mL/min/1.73 m 2 ) | 50.0 [36.0–64.0] |
| Aortic valve area (cm 2 ) | 0.68 ± 0.19 |
| Aortic peak velocity | 4.1 ± 0.7 |
| Mean aortic gradient (mmHg) | 39.2 [22.4–56.0] |
| Left ventricular ejection fraction (%) | 60 [47–73] |
| Logistic EuroSCORE (%) | 18.1 [3.6–32.5] |
| STS score (%) | 6.8 [2.1–11.5] |
| Overall | Patients with VCs | Patients without VCs | P | |
|---|---|---|---|---|
| ( n = 102) | ( n = 22) | ( n = 80) | ||
| Sex | 0.92 | |||
| Female | 64 (62.8) | 14 (63.6) | 50 (62.5) | |
| Male | 38 (37.3) | 8 (36.4) | 30 (37.5) | |
| Age (years) | 85.0 [76.0–94.0] | 85.0 [77.0–93.0] | 85.5 [76.5–94.5] | 0.59 |
| BMI (kg/m 2 ) | 25.6 ± 5.3 | 26.0 ± 5.1 | 25.4 ± 5.3 | 0.68 |
| Dyslipidaemia | 56 (54.9) | 10 (45.5) | 46 (57.5) | 0.32 |
| Diabetes mellitus | 30 (29.4) | 6 (27.3) | 24 (30.0) | 0.90 |
| Smoking | 0.34 | |||
| Never or past | 96 (94.1) | 22 (100.0) | 74 (92.5) | |
| Current | 6 (5.9) | 0 | 6 (7.5) | |
| Hypertension | 75 (73.5) | 14 (63.6) | 61 (76.3) | 0.24 |
| COPD | 25 (24.5) | 5 (22.7) | 20 (25.0) | 0.83 |
| Peripheral vascular disease | 20 (19.6) | 5 (22.7) | 15 (18.8) | 0.76 |
| Coronary artery disease | 54 (52.9) | 9 (40.9) | 45 (56.3) | 0.20 |
| Previous MI | 20 (19.6) | 2 (9.1) | 18 (22.5) | 0.22 |
| Previous CABG | 13 (12.8) | 4 (18.2) | 9 (11.3) | 0.47 |
| Previous PCI | 42 (41.2) | 7 (31.8) | 35 (43.8) | 0.31 |
| Previous cerebral stroke | 10 (9.8) | 4 (18.2) | 6 (7.5) | 0.22 |
| NYHA | 0.58 | |||
| 0–2 | 23 (22.6) | 4 (18.2) | 19 (23.8) | |
| 3–4 | 79 (77.5) | 18 (81.8) | 61 (76.3) | |
| GFR (mL/min/1.73 m 2 ) | 50.0 [36.0–64.0] | 49.8 [28.6–71.0] | 50.5 [37.0–63.0] | 0.60 |
| Aortic valve area (cm 2 ) | 0.68 ± 0.19 | 0.69 ± 0.17 | 0.68 ± 0.20 | 0.42 |
| Aortic peak velocity (m/s) | 4.1 ± 0.7 | 3.9 ± 0.1 | 4.1 ± 0.1 | 0.20 |
| Mean aortic gradient (mmHg) | 39.2 [22.4–56.0] | 35.5 [20.2–50.8] | 40.0 [21.7–58.3] | 0.26 |
| LVEF (%) | 60 [47–73] | 60 [57–63] | 60 [45–75] | 0.29 |
| Logistic EuroSCORE (%) | 18.1 [3.6–32.5] | 18.3 [7.3–29.2] | 18.1 [3.5–32.1] | 0.79 |
| STS score (%) | 6.8 [2.1–11.5] | 8.4 [3.7–13.1] | 6.5 [2.2–10.9] | 0.26 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
