Percutaneous Transapical Left Ventricular Access to Treat Paravalvular Leak and Ventricular Septal Defect

Abstract: Background. Surgical transapical (TA) access is an established technique for structural heart (SH) procedures, but is associated with considerable morbidity. Percutaneous TA puncture provides direct access for SH procedures and may overcome the disadvantages of surgical access. This study sought to evaluate the safety of percutaneous TA left ventricular access for SH interventions. Methods. We performed a retrospective analysis at a university hospital. Thirteen percutaneous TA procedures were performed on consecutive patients between January 2013 and July 2017 to provide LV access for transcatheter therapies. All procedures were performed under general anesthesia with three-dimensional transesophageal echocardiography guidance. Results. All TA punctures were successful. Delivery sheath sizes ranged from 5 Fr to 7 Fr. Eleven of the 13 TA sites were closed with a device. Total median procedural and fluoroscopy times were 106 minutes (interquartile range, 39-117 minutes) and 26.5 minutes (interquartile range, 8.3-43.8 minutes), respectively. The planned procedure was completed successfully in all cases. One access-site complication occurred, involving embolism of a duct occluder into the pleural space and extravasation from the apical puncture site. Hemostasis of the apex site was achieved immediately with placement of three vascular plugs from a femoral approach. Two patients died prior to discharge and neither death was related to a procedural complication. There were no significant pericardial effusions. Conclusion. Percutaneous TA access can be achieved safely in most cases to provide access for transcatheter procedures with short procedure times. Device closure of the TA access site is reliable, with a low complication rate and no procedure-related mortality.

J INVASIVE CARDIOL 2019;31(9):247-252. Epub 2019 June 15.

Key words: paravalvular leak, vascular closure, ventricular septal defect

Transapical (TA) access has been used for both interventional and electrophysiology procedures for over 50 years.^1,2 Interventional TA access has traditionally been achieved with direct surgical exposure of the left ventricular (LV) apex via a mini-thoracotomy during transcatheter valve replacement. More recently, percutaneous access of the LV apex has been described for interventional procedures.^3-6 This TA access technique was derived from LV puncture-needle catheterization, which was used to measure LV pressures in the presence of mechanical aortic and mitral valves. Presently, percutaneous TA access is most commonly used for structural interventions involving the mitral valve or as an alternative to femoral access in patients with severe peripheral arterial disease.^3-8 Given the growth in structural heart (SH) interventions and the need for direct access, TA may offer the interventionalist additional portals for therapy.

There have been limited data on the safety and utility of using TA access for SH procedures.^3,5 We report our single-center experience with percutaneous TA access of the LV for various SH interventions.

Methods

This study is a retrospective registry analysis of patients who underwent attempted percutaneous TA access for SH interventions between April 2014 and July 2017. Patients were identified by inclusion in a predefined registry, which included all patients who underwent TA access for any indication. All patients had clinically significant heart failure or hemolysis as indications for their structural procedure. The hospital institutional review board approved the retrospective analysis.

The selection of access site and procedural approach was made at the discretion of the heart team and the primary operator in each case. Percutaneous TA access was selected when other percutaneous access sites had unfavorable anatomy or had previously failed.

All charts and interventional reports of the included patients were examined. Efficacy endpoints included procedural success, freedom from repeat procedure, procedure time, and fluoroscopy time. Safety endpoints included rate of access-site complication, 30-day survival, 6-month survival, survival to discharge, major bleeding, minor bleeding, pericardial tamponade, and pericardial effusion. An access-site complication was defined as extravasation from the TA site, major coronary artery compromise, new LV dysfunction, or embolization of a TA site closure device. The severity of bleeding was defined according to TIMI definitions.⁹ Additional procedure-specific endpoints for patients undergoing paravalvular leak (PVL) closure were also collected, including the number and types of devices used and the presence of residual defects. Statistical analysis was performed using Stata (StataCorp).

Institution-specific TA access procedure. The informed consent process included a detailed discussion of the procedure, including possible complications and the expected off-label use of closure devices. Contraindications to TA access included cardiac surgery within 6 weeks, LV thrombus, and LV apical aneurysm. All patients had two-dimensional transthoracic echocardiographic (TTE) evaluation before the procedure. All TA procedures were done with general anesthesia. Two-dimensional and real-time three-dimensional transesophageal echocardiography (TEE) were used throughout the procedure. Femoral venous access was achieved in all cases.

After the induction of anesthesia, the anesthesiologist placed a dual-lumen airway. The left lung was then deflated to allow optimal positioning of the LV apex, while right lung ventilation continued. Identification of the LV apex occurred initially with palpation of the left fifth intercostal space, in the mid-clavicular line. Once the location between the ribs had been palpated, TTE was used to confirm apical location and a marker was placed on the skin at the site. Coronary angiograms of the left anterior descending (LAD) artery (or the left internal mammary artery if the patient had previous coronary artery bypass graft surgery) were performed with the marker in place (Figure 1). The apical puncture was performed once the marker position was confirmed to avoid the course of the LAD. Apical puncture was performed with a 21 gauge, 7 cm micropuncture needle under TTE guidance with the left lung deflated. A standard micropuncture wire was advanced into the LV, and the needle exchanged for a micropuncture sheath. A standard 5 Fr sheath was placed into the LV cavity over a standard 0.035˝ wire and sutured in place to prevent movement. Therapeutic anticoagulation was then achieved with a bolus of intravenous heparin to achieve an activated clotting time >250-300 seconds. The defect was then crossed with a hydrophilic-coated guidewire, typically an 0.035˝ curved Glidewire Advantage guidewire (Terumo Interventional Systems). The short sheath was then exchanged over this guidewire for an Amplatzer TorqVue Delivery System (Abbott Vascular) catheter, which was advanced across the defect. The occluder device was then delivered and positioned via this delivery system.

In the majority of cases, the TA site was closed with a 6 x 4 mm Amplatzer Duct Occluder (Abbott Vascular) (Figure 2). In patients without a mechanical valve in the aortic position, a pigtail catheter was placed in the LV and limited ventriculograms were performed to aid in the placement of the occluder. If no closure device was used, manual pressure was held in the intercostal space over the apex for approximately 10 minutes. In all cases, anticoagulation was reversed with 1 mg of protamine for every 100 units of heparin administered. After the closure device was deployed or hemostasis was achieved with manual pressure, repeat coronary angiography was performed to confirm patency of the LAD. In addition, a left ventriculogram was performed to demonstrate adequate hemostasis of the apical site. When ventriculography was not possible or not preferred, such as in the presence of mechanical aortic prosthesis or chronic kidney disease, two-dimensional TEE with color Doppler was used to confirm adequate TA closure. Patients were then admitted to the hospital and monitored overnight in the cardiac intensive care unit.

Results

Thirteen patients (61% male; median age, 69 years; interquartile range [IQR], 57-81 years) underwent percutaneous TA punctures during the study period (Table 1). All TA punctures were successful. The planned procedure was completed successfully in all cases. Delivery sheath sizes ranged from 5-7 Fr. Eleven of the TA sites (85%) were closed with a 6 x 4 mm Amplatzer Duct Occluder (Table 2). Manual pressure was used to achieve hemostasis in 2 cases; both patients had significant LV hypertrophy that was felt to preclude device closure. Total median procedural and fluoroscopy times were 106 minutes (IQR, 39-117 minutes) and 26.5 minutes (IQR, 8.3-43.8 minutes), respectively. Suppressive transvenous pacing was not required in any of the remaining 11 patients (0%) at the time of apical closure.

The indication for the majority of cases (n = 10; 77%) was mitral-position prosthetic PVL (Table 3). Of the 10 PVL cases, 5 (50%) were bioprosthetic mitral valves and 5 (50%) were mechanical mitral valves. Four patients (40%) also had a prosthetic valve in the aortic position. Seven procedures (70%) were done on an outpatient basis. One PVL closure (10%) was done emergently. Amplatzer Muscular VSD Occluders (Abbott Vascular) of various sizes (4-9 mm) were used to close the PVL. Three patients (30%) required more than one occluder device implanted; a second Amplatzer Muscular VSD Occluder was used in 2 cases (one 4 mm and one 6 mm) and one 10 mm Amplatzer septal occluder was used in the third case. Three patients (30%) had residual PVL present at the end of the case; all residual leaks were +0-1 (mild) in severity. Only 1 patient (10%) required a repeat procedure. This repeat procedure was not included in the registry, because it was performed from transfemoral approach.

Two ventricular septal defect (VSD) closures were performed and both cases were successfully closed with an Amplatzer Muscular VSD Occluder. No significant residual leak was present after VSD closure (Table 3).

The indication for 1 case was aortic-left atrial fistula. This was successfully closed with a single VSD occluder device without significant residual leak (Table 3).

Access-site complication occurred in 1 patient (8%), involving embolism of an Amplatzer Duct Occluder into the pleural space and extravasation from the apical puncture site (Table 4). Immediate ventriculography identified a long, serpiginous tract extending from the LV apex to the pleural space. Hemostasis of the apex site was achieved with placement of three Amplatzer Vascular Plug 4 devices (two 4 mm devices and one 6 mm device; Abbott Vascular) in the tract from a femoral approach. This patient was an adherent of the Jehovah’s Witness faith, so blood loss was managed with auto-transfusion.

Two patients had pericardial effusions after the procedure that were deemed not to be hemodynamically significant, and no additional therapy was required. No patients suffered complications requiring surgical repair or correction (Table 4).

Two patients died prior to discharge. All patients who survived to discharge were alive at 6-month follow-up. Neither death was related to a procedural complication (Table 4). One patient suffered from hypoxic arrest due to aspiration at the time of TEE probe removal. The other died of refractory hypoxic respiratory failure related to right ventricular failure.

Discussion

The literature addressing the safety and utility of interventional percutaneous TA access is limited. Tamponade, pneumothorax, and hemothorax have all been reported. Pitta et al reported an overall complication rate of 62% – mostly hemothorax – associated with manual hemostasis of interventional TA access.⁵ Jelnin et al published outcomes of a 26-patient cohort that underwent percutaneous PVL closure via TA approach.³ In the Jelnin group, TA access was achieved with the assistance of a previously acquired cardiac computed tomography angiography (CTA) for intraprocedural guidance of apical anatomy and the “safe puncture window.” Delivery sheaths ranged from 5-12 Fr. Of the 22 patients with access sites >5 Fr, a total of 20 were closed with an Amplatzer Duct Occluder. The 5 Fr access sites were closed with manual hemostasis. Two patients (7.1%) in the overall cohort had procedure-related complications. Total fluoroscopy time for primary TA access resulted in a 35% decrease compared with conventional arterial or venous access at the same center (27.4 ± 15.6 minutes vs a total 42.6 ± 29.9 minutes).³ It appears from these prior studies that routine closure of TA puncture sites is safer than manual hemostasis. In our study, the rate of access-site complications after device closure was 9% (1 of 11 patients). Our study adds to the literature demonstrating low procedure-related complications when a 5-7 Fr TA access site is closed with a device.

Unlike Jelnin et al, the patients in our study did not undergo CTA guidance when obtaining TA access. The similar complication rates indicate that the routine use of preprocedure CTA, although helpful, is not necessary to achieve safe TA access for SH intervention. Our standard technique for TA access includes general anesthesia and routine coronary angiography to ensure the puncture site will not compromise the coronary vasculature. We also routinely ask the anesthesiologist for solitary right-lung ventilation in order to deflate the left lung. The anesthesiologists at our center are adept with this technique, and we feel that it reduces the risk of lung injury at the time of TA puncture.

We do not routinely use CTA mapping for preprocedural planning. At our center, the majority of preprocedure planning is done with high-quality echocardiography, often TEE. However, the use of CTA is beneficial in cases with unclear anatomy or in patients with poor echocardiography windows. In addition, the use of CTA-derived three-dimensional printed models for preprocedure planning has great promise.

The majority of cases in the registry were percutaneous closures of mitral prosthetic PVL. At our institution, percutaneous therapy for PVL is preferred for patients with adequate anatomy for transcatheter intervention. Our team typically prefers surgical correction in cases of complicated, extensive PVL, particularly when a large leak (extending >50% around the valve annulus) is present. Contemporary data show largely similar long-term survival with percutaneous and surgical management of PVL.¹⁰ Meta-analysis data report higher technical success with surgical closure of PVL, but similar 1-year mortality and symptom improvement compared with percutaneous closure.¹¹ An observational study reported improved long-term outcomes with surgical closure of PVL at the cost of worse perioperative morbidity and mortality rates.¹²

Two VSD cases were included in the registry. One case was a post-myocardial infarction VSD, in which the location of the defect was deemed too difficult to access from a traditional right-sided or retrograde, transaortic approach. The other case was a Gerbode defect. Although this defect can often be approached from the internal jugular vein, it was felt that the probability for valve interaction was high from that approach. Therefore, the TA approach was used.

The primary indications for TA access may be limited to interventions of the mitral valve, the mitral valve annulus, and select VSDs. Navigation of interventional equipment to the mitral position or across the ventricular septum is technically challenging via either the transseptal or retrograde-aortic approach. All of the SH procedures in our series were completed successfully via the TA approach with relatively low median fluoroscopy time (26.5 minutes; IQR, 8.3-43.8 minutes) and median procedure time (106 minutes; IQR, 39-117 minutes). The median fluoroscopy time for mitral PVL cases was 28.9 minutes (IQR, 10.2-47.6 minutes; mean, 38.8 ± 21.1 minutes). For comparison, a large cohort study of percutaneous PVL closure that used a variety of access sites (transseptal, retrograde-aortic, and a small number of TA cases) reported a mean fluoroscopy time of 42.6 ± 29.9 minutes for mitral PVL cases.¹³ Previous cohorts of interventional TA access have reported improved fluoroscopy times when compared with more conventional vascular access.³ The direct, in-line access to the mitral valve, mitral valve annulus, and ventricular septum from TA access is therefore attractive for both procedural success and procedure time considerations.

Interventional TA access is a growing field. There are a number of closure devices used for percutaneous TA closure under investigation, some with CE mark in Europe.^14-16 These devices utilize various approaches to secure TA closure, including sutured and sutureless techniques. There are currently no approved TA closure devices available in the United States. Therefore, we typically use the 6 x 4 mm Amplatzer Duct Occluder for closure of the 6 Fr TA site.

Study limitations. There are multiple limitations to our study. First, this is a single-center, retrospective registry study that is limited in size. The selection of vascular access was made at the discretion of the primary operator. Patients were included in the registry if the heart team and operator chose percutaneous TA access. Therefore, inclusion in this registry was dependent on both expert opinion and the availability of an operator who was comfortable with TA access. Finally, the patients included in the registry and their indications for TA access are heterogeneous. Larger registries of TA access are needed to investigate specific patient characteristics that may predict or limit TA site complications.

Conclusion

Percutaneous TA puncture is effective and safe in the hands of experienced operators. It is a potentially preferred technique for select interventional procedures due to its direct approach and decreased procedure times. Device closure of the apical access site is reliable and should likely be done in most cases.

References

1. Brock SIRR, Milstein BB, Ross DN. Percutaneous left ventricular puncture in the assessment of aortic stenosis. Thorax. 1956;163:163-171.

2. Vurgun VK, Altin AT, Kilickap M, Candemir B, Akyurek O. Percutaneous transapical approach and transcatheter closure for ventricular tachycardia ablation. Pacing Clin Electrophysiol. 2018;41:334-337.

3. Jelnin V, Dudiy Y, Einhorn BN, Kronzon I, Cohen HA, Ruiz CE. Clinical experience with percutaneous left ventricular transapical access for interventions in structural heart defects: a safe access and secure exit. JACC Cardiovasc Interv. 2011;4:868-874.

4. Lim DS, Ragosta M, Dent JM. Percutaneous transthoracic ventricular puncture for diagnostic and interventional catheterization. Catheter Cardiovasc Interv. 2008;918:915-918.

5. Pitta SR, Cabalka AK, Rihal CS. Complications associated with left ventricular puncture. Catheter Cardiovasc Interv. 2010;997:993-997.

6. Brown SC, Boshoff DE, Rega F, et al. Transapical left ventricular access for difficult to reach interventional targets in the left heart. Catheter Cardiovasc Interv. 2009;142:137-142.

7. Venturini JM, Rosenberg J, Lang RM, Shah AP. Transapical access for percutaneous mitral paravalvular leak repair. Struct Hear. 2017;1:121-128.

8. Sharma BS, Kliger C, Jelnin V, Ruiz CE. Paravalvular leak closure: the apical approach. Card Interv Today. 2014:48-52.

9. Mehran R, Rao S V, Bhatt DL, et al. Standardized bleeding definitions for cardiovascular clinical trials. A consensus report from the Bleeding Academic Research Consortium. Circulation. 2011;123:2736-2747.

10. Alkhouli M, Rihal CS, Zack CJ, et al. Transcatheter and surgical management of mitral paravalvular leak: long-term outcomes. JACC Cardiovasc Interv. 2018;10:1947-1956.

11. Busu T, Alqahtani F, Badhwar V, Cook CC, Rihal CS, Alkhouli M. Meta-analysis comparing transcatheter and surgical treatments of paravalvular leaks. Am J Cardiol. 2018;122:302-309.

12. Millan X, Bouhout I, Nozza A, et al. Surgery versus transcatheter interventions for significant paravalvular prosthetic leaks. JACC Cardiovasc Interv. 2017;10:1959-1969.

13. Ruiz CE, Jelnin V, Kronzon I, et al. Clinical outcomes in patients undergoing percutaneous closure of periprosthetic paravalvular leaks. J Am Coll Cardiol. 2011;58:2210-2217.

14. Ziegelmueller JA, Lange R, Bleiziffer S. Access and closure of the left ventricular apex: state of play. J Thorac Dis. 2015;7:1548-1555.

15. Blumenstein J, Kempfert J, Van Linden A, et al. Sutureless transapical access and closure to facilitate transapical transcatheter aortic valve implantation: first-in-human use. J Am Coll Cardiol. 2013;62:763.

16. Ferrari E, Demertzis S, Angelella J, et al. Apical closure device for full-percutaneous transapical valve implantation: stress-test in an animal model. Interact Cardiovasc Thorac Surg. 2017;24:721-726.

From the ¹Section of Cardiology, Department of Medicine, University of Chicago Medical Center, Chicago, Illinois; and the ²Section of Cardiology, Department of Medicine, Feinberg School of Medicine, Northwestern University, Chicago, Illinois.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Shah is a consultant and proctor for Abbott Vascular, which is the manufacturer of the devices used in the study; Abbott Vascular was not involved in the design, structure, analysis of the study, or drafting of the manuscript. The remaining authors report no conflicts of interest regarding the content herein.

The authors report that patient consent was provided for publication of the images used herein.

Manuscript submitted January 22, 2019, provisional acceptance given February 11, 2019, final version accepted February 20, 2019.

Address for correspondence: Atman P. Shah, MD, Section of Cardiology, Department of Medicine, University of Chicago Medicine, 5841 S. Maryland Ave, MC 6080, Chicago, IL 60637. Email: Ashah@bsd.uchicago.edu