The future of interventional cardiology

INTRODUCTION

Right ventricular outflow tract (RVOT) disease is a common finding in children and adults with congenital heart disease and often occurs as a sequel of previous surgery. Over the past 2 decades, percutaneous pulmonary valve implantation has become more widely used and is recommended by current clinical practice guidelines1 as the preferred option for patients with previous conduits or bioprostheses.

However, many patients have native or patched tracts (hereafter referred to as native RVOTs) with pulmonary regurgitation as the predominant lesion. In these patients, percutaneous valve placement is more complex due to the RVOT anatomy, its dynamic behavior, larger pulmonary annulus size, and lack of a proper landing zone for the valve. Because of the differences in underlying heart diseases, previous surgical repairs, and various pulmonary artery configurations, RVOT morphology varies widely but can be categorized into 5 subtypes2 (figure 1).

Figure 1. Five types of native RVOT anatomy: I - pyramidal; II - cylindrical or tubular; III - inverted pyramidal; IV - central enlargement; V - central narrowing. (Reproduced from Schievano et al.2 with permission from the author.)

 

Repaired tetralogy of Fallot serves as the paradigm, and in these cases, surgery remains the standard of care. However, the development of percutaneous procedures has enabled a larger number of patients with these substrates to be eligible for percutaneous treatment (figure 2).

Figure 2. Self-expanding pulmonary valve implantation in an aneurysmal right ventricular outflow tract (RVOT). A: severely dilated RVOT on magnetic resonance imaging, with interventricular septal flattening. B: balloon sizing test in the RVOT with simultaneous injection into the left main coronary artery. C: successful venus-P valve (MedTech, China) implantation in the RVOT.

 

Two different models of balloon-expandable valves have been authorized to treat dysfunctional bioprostheses and conduits: the Melody (Medtronic, United States) and the SAPIEN valves (XT model, Edwards Lifesciences, United States). Although they have not yet been authorized for implantation in native RVOTs, both (along with the SAPIEN S3) have been used off-label in this setting.

To address the specific characteristics of native RVOTs, several models of self-expanding valves have been developed, such as the Venus-P (Venus MedTech, China, with CE marking for use in Europe since 2022), PULSTA (TaeWoong Medical, South Korea), and Harmony valves (Medtronic, United States, with prior FDA approval). The Alterra valve (Edwards Lifesciences) has also been used. This valve serves as a self-expanding prestent onto which a SAPIEN valve is later implanted.

The characteristics of each of these devices have already been described in detail in a previous issue of REC: Interventional Cardiology.3

RESULTS OF PERCUTANEOUS VALVES IN THE NATIVE RIGHT VENTRICULAR OUTFLOW TRACT

More information has gradually become available on the favorable results and durability of percutaneous valves. The largest multicenter registry to date,4 with 2476 patients (82% implanted with the Melody valve and 18% with the SAPIEN device, including 16% with native RVOTs), reported an 8-year survival rate of 91.1% after implantation, and a reintervention rate of 25.1%, which is similar to the rates reported in some surgical series.5 Nonrandomized comparative studies6 and a recent meta-analysis7 also found similar reintervention rates. Some series report higher rates in patients implanted with the Melody compared with the SAPIEN valve,8,9 although the 2 groups were not directly comparable, with reintervention-free survival rates in patients with SAPIEN being similar to those reported in patients with surgical valves.8

The SAPIEN device can be implanted with or without prestenting, depending on the patient’s characteristics, with good outcomes. The largest trial published to date included 774 patients implanted with the XT and S3 models10, 51% of whom had native RVOTs (table 1).

 

Table 1. Summary of some of the main trials of patients with native right ventricular outflow tract

Author and year	Patients with native RVOT/Total patients	Valve type	Follow-up	Implant success	Other results	Complications
Malekzadeh-Milani et al.23, 2014	34/34	Melody	2.6 years	100%	Paravalvular leak in 2 patients during follow-up	3 acute complications (9%): 1 hemoptysis, 1 RVOT obstruction, 1 stent embolization
Meadows et al.18, 2014	31/31	Melody	15 months (1 month- 3.8 years)	100%	No mortality or valve regurgitation	Stent fractures (32%) associated with a higher rate of stenosis. 3 cases of endocarditis. Reintervention in 3 patients
Garay et al.13, 2017	10/10	Venus P	12 months (4 to 21)	100%	Normally functioning valve, no stent fractures, right ventricular remodeling, and NYHA functional class improvement	None
Martin et al.11, 2018	132/132	Melody	No follow-up		Complete cohort of 229 patients, but only 58% implanted. Good immediate hemodynamic outcomes	Complication rate of 4% (mostly due to stent migration)
Morgan et al.24, 2019	41/57	SAPIEN (S3, XT)	5.3 months (1 to 26)	100%	No prestenting. Normally functioning valve at follow-up. No mortality	1 aortic compression, 2 tricuspid valve injury, 1 valve regurgitation
Shahanavaz et al.10, 2020	397/774	SAPIEN S3 (78%) XT (22%)	12 months (n = 349)	97.4%	Normally functioning valve: 91.5%	Adverse events: 10%. Emergency surgery: 14 patients (1.8%). Tricuspid injury: 3%
Goldstein et al.19, 2020	143/530	Melody (88%) SAPIEN (22%)	1 year	98%	Normally functioning valve: 98%	1 death. Reintervention rate of 13.3% (mostly unrelated to the valve)
Lee et al.12, 2021	25/25	PULSTA	33 (± 14) months	100%	Zero cases of valve dysfunction	No significant adverse events
Gillespie et al.16, 2021	21/21	Harmony	5 years	100%	Implantation in all but 1 patient due to pulmonary hypertension. Normally functioning valve in nonreoperated patients	Valve explantation in 2 patients, 1 death 3 years after implantation, 2 reinterventions (valve-in-valve)
Morgan et al.25, 2021	38/38	Venus	27 months	97.4%	Normally functioning valve at follow-up	Migration: 2 cases (surgery in 1)
Houeijeh et al.9, 2023	99/214	SAPIEN XT/S3 (85%) Melody (15%)	2.8 years (3 months-11.4 years)	Only cases with successful implantation included	Reintervention-free survival at 5 to 10 years: 78.1% to 50.4% (Melody) and 94.3% to 82.2% (SAPIEN)	Severe complications: 2.3%, 1 valve-related death. Endocarditis 5.5/100 patient-years (Melody) and 0.2/100 patient-years (SAPIEN)
Álvarez et al.14, 2023	8/8	Venus	No follow-up	100%	Normally functioning valve in all	No significant adverse events
Lin et al.15, 2023	53/53	Venus (28%), PULSTA (72%)	27.5 months	98.1%	No valve regurgitation at 12 months	1 embolization, 1 endocarditis
NYHA, New York Heart Association; RVOT, right ventricular outflow tract. In studies that are not specific to native RVOT, the results and complications refer to the overall cohort, as the results of native, or non-native RVOTs are often not detailed independently. Some centers participated in > 1 study, which allowed the same patient to be included in multiple publications. The most recent series with larger numbers of patients have been prioritized.

 

In a study of patients with native RVOTs that included 229 candidates for the Melody valve, the device was finally implanted in 132 patients (58%).11 The most common reason for avoiding implantation was a prohibitively large RVOT, followed by coronary or aortic root compression. The immediate outcomes of patients with successful implantation were good. However, the low implantation rate demonstrates the limitation of treating native RVOTs with these valves.

Self-expanding valves fill this gap by allowing treatment of larger RVOTs, as they adapt to the anatomy of the RVOT and provide more stable attachment. The series published to date indicate a very high implantation success rate—close to 100%—with good short- and mid-term outcomes and few complications.12-16 (table 1 illustrates a selection of series representative of patients with native RVOTs).

Drawing comparisons between the results of surgical and percutaneous pulmonary valves is challenging because the types of patients and the anatomies treated are very different. Overall, patients undergoing percutaneous valve implantation are at higher risk and often have bioprostheses or small conduits that require smaller percutaneous valves, which is associated with a higher residual gradient, leading to a greater need for reintervention,4,8,9 and a higher rate of endocarditis. This aspect has traditionally led to less favorable results with the Melody series (with a maximum diameter of 22 mm), with durability being one of the issues initially identified, although more recent series have shown positive results. Only indirect comparisons can be drawn with valve procedures in the native RVOT.

PROCEDURAL PLANNING

Multimodal imaging modalities, especially magnetic resonance imaging and computed tomography, are crucial for patient selection and procedural planning. Key aspects include measurement of the RVOT and the pulmonary annulus, and assessment of both the RVOT and the coronary anatomy. Recently, computed tomography guidelines for RVOT assessment17 have been published with detailed information on the measurements that should be taken, and the cardiac cycle phase.

POTENTIAL PROBLEMS

Valve or stent migration or embolization is a potential complication in patients with large RVOTs due to the lack of an adequate landing zone, with incidence rates between 0% and 4.5%,11,10,18 and consequently proper valve sizing is of paramount importance.

Stent fractures can lead to loss of integrity and contribute to prosthetic valve dysfunction. Nonetheless, the implications of the occasional finding of an isolated strut fracture remain unclear.

Tricuspid valve injury has been reported in 3% to 6% of patients.10 However, this rate has dropped significantly with the use of DrySeal introducer sheaths (W.L. Gore & Associates, United States).10

Coronary compression is a rare complication nowadays, because coronary angiography is simultaneous and systematically performed during RVOT balloon inflation testing. However, it can be a reason for not performing percutaneous implantation in nearly 3% of patients.11

A higher incidence of infective endocarditis has been reported, especially after Melody valve implantation,6,8,9 with a higher risk when smaller valves are implanted and there is a greater residual gradient. The risk involved with other types of device such as the SAPIEN—whether because of its different composition (bovine jugular vein graft in the Melody compared with bovine pericardium in the SAPIEN) or because of its larger size—is much lower9 and seems comparable to that of surgical series.

Self-expanding valves are larger, and the proximal end often remains inside the RVOT, which could increase the risk of ventricular arrhythmias. The incidence of nonsustained ventricular tachycardia varies widely (between 0.6% and 40%11,19,20), although it is usually a transient phenomenon during the early postimplantation phase, and its long-term implications remain unclear. Of note, when comparing surgical with percutaneous pulmonary valve replacement, the early incidence of arrhythmias was lower in the latter.21 A potential caveat is that catheter access to the arrhythmic substrate can be limited after valve implantation.

The presence of the valve metal mesh with or without previous stents inside the RVOT can pose additional challenges for the surgeon if surgical valve replacement is subsequently required. This is a relative problem, because surgical pulmonary valve replacement also increases the risk of future reinterventions related to resternotomy.

BENEFITS OF PERCUTANEOUS VALVE IMPLANTATION

The possibility of performing percutaneous pulmonary valve implantation offers clear advantages: the procedure is much less invasive, length of stay is shorter,8 recovery is faster, the mortality rate is very low (from 0.2% to 0.8%),19 and the cost-effectiveness ratio is more favorable.22 In patients at high surgical risk, it might be the only available treatment option. The path followed by its “left-sided relatives”—transcatheter heart valves in the aortic position—illustrates that the threshold for the use of percutaneous techniques is decreasing as more experience is gained and technology becomes further refined. In cases of intermediate or low risk, percutaneous valve implantation may delay or avoid the need for surgery in patients who often require multiple interventions during their lifetime. This concept of avoiding sternotomies is relevant due to the added risk of further surgeries and for patients who are candidates for heart transplants. The decision on the best approach to valve implantation in each patient should be made by a multidisciplinary heart team, including health professionals experienced in these types of heart disease.

CONCLUSIONS

Percutaneous pulmonary valve implantation is particularly challenging in patients with native RVOTs. Nonetheless, it is a feasible option that is being used with a high success rate and few complications. However, appropriate candidate selection is essential. Several models of self-expanding valves have been specifically developed for this purpose, with good short- and mid-term results, allowing the treatment of patients with large RVOTs that were previously not amenable to balloon-expandable devices. The latest information suggests that the durability of percutaneous valves may be comparable to that of surgical bioprostheses, although long-term data are lacking, especially with the latest models. Although more studies and follow-up are necessary, percutaneous techniques are already an option for many patients and will likely become an alternative to surgical treatment in the near future.

FUNDING

None declared.

CONFLICTS OF INTEREST

None declared.

REFERENCES

INTRODUCTION

Transcatheter aortic valve implantation (TAVI) is a well-established procedure for severe symptomatic aortic stenosis. Previously, this procedure was used exclusively to treat inoperable and high-risk patients but is now a common approach for intermediate and low-risk populations. In parallel, there has been growing global experience in several off-label scenarios, as in bicuspid, valve-in-valve, and noncalcified aortic regurgitation (NCAR).1

AORTIC REGURGITATION

Due to abnormalities in the aortic leaflets or their supporting structures (ie, aortic root and annulus), or both, aortic regurgitation (AR) causes diastolic reflux of blood from the aorta into the left ventricle (LV). This leads to LV volume overload and dilatation, allowing the ejection of a larger stroke volume. However, over time, it results in a decline in systolic function causing symptoms conferring a poor prognosis.2 Current European and American guidelines recommend surgical intervention when significant AR is accompanied by symptoms, reduced LV ejection fraction, or severe LV dilatation.3-4 Although moderate or severe AR affects around 2.2% of the population aged 70 years or older, up to 30% of these individuals are deemed inoperable due to advanced age or comorbidities, with percutaneous options—including dedicated and nondedicated devices—gaining increased interest due to the absence of safe surgical alternatives.5

CHALLENGES OF TAVI IN NCAR

Specific challenges for transcatheter devices are due to the characteristics of patients with NCAR, including larger aortic annular dimensions, aortic root dilatation, insufficient annular calcification for anchoring, and larger stroke volume with backflow to the LV causing a “suction effect”. All of these factors create difficulties in selecting the appropriate device and positioning and deploying it correctly, and consequently increase the risk of embolization or malpositioning of the prosthesis, requiring a second valve implantation. Ultimately, these challenges are associated with increased mortality. In particular, the primary concern is valve embolization. To reduce the risk of this complication, prosthesis oversizing is routinely performed, but the exact degree of oversizing with each device has not been standardized. Furthermore, oversizing is associated with a high rate of conduction system disturbances and may carry a higher risk of annular rupture.

DEVICES AND EVIDENCE

There are 2 dedicated-devices for NCAR: The Trilogy system (JenaValve Technology Inc, California, United States) and the J-Valve (JC Medical Inc, California, United States). The most recent experience with these dedicated devices has been reported by Adam et al.6 and García et al.7 The Trilogy system was associated with a 30-day mortality rate of 1.7% (vs 3.7% with J-Valve) and both had 0% ≥ moderate residual AR. However, 7.4% required conversion to surgery with the J-Valve, leading to some technical changes in the technology. Of note, the pacemaker rate with both systems was above 10% (19.6% and 13%, respectively). These promising dedicated technologies still have certain shortcomings, the main one being the lack of sizes covering large annuli; indeed, for 10% oversizing, the proportion of AR patients whose aortic structures were above the recommended indications was up to ~50%.8

The first-in-human compassionate use of SAPIEN (Edwards Lifesciences, California, United States) was reported in 2012,9 followed by multiple case series and registries reporting the feasibility of TAVI in NCAR with new-generation nondedicated devices. The most recent summary of the experience was reported in a meta-analysis by Takagi et al.,10 and the pooled analysis at 30 days demonstrated 80.4% device success, 9.5% all-cause mortality, 7.4% ≥ moderate residual AR, and a pacemaker rate of 11.6%. Although these results were promising and significantly better than with early-generation devices in all aspects, there was still a substantial gap to achieve similar outcomes to TAVI in aortic stenosis and, indeed, the rate of valve embolization was high (above 9% with all the technologies, even the dedicated technologies).10

More recently, the new Myval balloon-expandable valve (Meril Lifesciences Ltd, Vapi, India) has demonstrated better outcomes than those reported by Takagi et al. likely due to the availability of extra-large sizes (30.5 and 32 mm), allowing a greater degree of oversizing and covering annuli up to 100.5 mm perimeter and 840 mm² area at its nominal volume, increasing the proportion of inoperable patients who can be treated percutaneously. Because of the high procedural success rate (94.7%), and the absence of severe anatomical complications (ie, annular rupture, aortic dissection, or coronary obstruction), the Myval balloon-expandable valve is a promising new approach to this condition until dedicated devices for these annular sizes become available. Nevertheless, prosthesis embolization occurs in 3.5% of patients, leaving room for improvement; according to this research, the interplay of the aortic annulus and the LV outflow tract could help to predict the risk of valve embolization, suggesting that, in borderline annular sizes, when 20% oversizing cannot be achieved and adverse (tapered) morphology is detected, the intervention should be avoided or might be carried out under mechanical circulatory support (figure 1).11

Figure 1. Anatomical predictor of valve embolization during transcatheter aortic valve implantation for noncalcified aortic regurgitation. LVOT, left ventricular outflow tract.

 

Poletti et al.12 recently compared latest-iteration nondedicated devices and suggested that Myval might provide the best outcomes compared with other devices; indeed, the subanalysis (not published yet) comparing the 2 balloon-expandable platforms suggested that, despite being used in patients with significantly smaller aortic annuli, SAPIEN-3 had a lower device success rate (72% vs 90%) and a higher rate of prosthesis migration/embolization (SAPIEN 17% vs MyVal 5%).12 A summary of the PURE-AR study11 with the Myval device is shown in figure 2. The current evidence on different devices is summarized in table 1.

Figure 2. Main outcomes reported in the Myval registry for treatment of noncalcified aortic annuli with large annuli. AR, aortic regurgitation; LVOT, left ventricular outflow tract; TAVR, transcatheter aortic valve regurgitation.

Table 1. Comparison of outcomes between the international registries

Registry	No. of patients	Type of device	Device success	All-cause mortality	≥ Moderate residual aortic regurgitation	Permanent pacemaker implantation rate
Yoon et al.13 (2017)	331 patients with NCAR and FSHV	Nondedicated devices	Overall: 74.3%	Overall: 10.9%	Overall: 9.6%	Overall: 18.2%
		Early-generation devices (CoreValve, SAPIEN XT)	Early-generation devices: 61.3%	Early-generation devices: 13.4%	Early-generation devices: 18.8%	Early-generation devices: 17.5%
		New-generation devices (Evolut R, JenaValve, Engager, Portico, ACURATE, Lotus, Direct Flow, SAPIEN 3)	New-generation devices: 81.1%	New-generation devices: 9.4%	New-generation devices: 4.2%	New-generation devices: 18.6%
De Backer et al.14 (2018)	254 patients with NCAR	Nondedicated devices Early-generation devices (CoreValve, SAPIEN XT)	Early-generation devices: 47%	Early-generation devices: 17%	Early-generation THV’s: 26%	Not reported
		New-generation devices (Evolut R, JenaValve, Engager, Portico, ACURATE, Lotus, Direct Flow, SAPIEN 3)	Newer-generation devices: 82%	Newer-generation devices: 8%	Newer-generation THV’s: 5%
Sawaya et al.15 (2018)	146 patients with NCAR and failing surgical heart valves (FSHV)	Early-generation devices (CoreValve SAPIEN XT)	NCAR: 72%	NCAR: 13%	NCAR: 13%	NCAR: 18%
		New-generation devices (Evolut R, JenaValve, Lotus, Direct Flow, SAPIEN 3)	FSHV: 71%	FSHV: 6%	FSHV: 6%	FSHV: 5%
			Early-generation devices: 54%	Early-generation devices: 22%	Early-generation devices: 27%
			New-generation devices: 85%	New-generation devices: 8%	New-generation devices: 3%
Sánchez-Luna et al.11 2023	113 patients with NCAR	New-generation nondedicated device: Myval	94.7%	9.7%	8.9%	22.2%
Poletti et al.12 2023	201 patients with NCAR	New-generation devices: SEV (Evolut R/Pro, ACURATE Neo/Neo2, Jena Valve, Navitor/Portico)	Overall: 76.1%	Overall: 5%	Overall: 9.5%	Overall: 22.3%
			SEV: 75.8%	SEV: 5.3%	SEV: 9.2%	SEV: 22.6%
		BEV (SAPIEN, Myval)	BEV: 76.8%	BEV: 4.4%	BEV: 10.1%	BEV: 21.8%
Adam et al.6 (2023)	58 patients with NCAR	Dedicated device: Trilogy system (JenaValve)	98%	1.7%	0%	19.6%
Garcia S et al.7 (2023)	27 patients with NCAR	Dedicated device: J-Valve	81%	3.7%	0%	13%
BEB, balloon-expandable valve; FSHV, failing surgical heart valve; NCAR, noncalcified aortic regurgitation; SEV, self-expandable valve; THV, transcatheter heart valve. Medtronic, United States: CoreValve, Engager, and Evolut R. Edwards Lifesciences, United States: SAPIEN XT, SAPIEN 3. Abbott Vascular, United States: Navitor/Portico. Meril Life, India: Myval. Boston Scientific, United States: ACURATE Neo/Neo2, Lotus. Trilogy, Germany: JenaValve. Direct Flow Medical, United States: Direct Flow. JC Medical, United States: J-Valve.

 

In conclusion, although the surgical approach remains the standard-of-care in NCAR patients, the outcomes of new-generation dedicated and nondedicated TAVI devices are rapidly improving. The main caveats include the lack of large-sized dedicated devices and the risk of valve embolization with nondedicated devices, although this risk can be minimized by extra-large sizes and morphological analysis of the LV outflow tract. The rate of conduction disturbances is still higher than for TAVI in aortic stenosis and its decrease will require specific technology and strategies, given its potentially negative impact in LV remodelling. Clinical trials comparing surgery and TAVI in this setting for high-risk patients are an urgent need.

FUNDING

None.

AUTHORS’ CONTRIBUTIONS

I.J. Amat-Santos and J.P. Sánchez-Luna equally contributed in the design, data gathering, analysis, and final approval of the manuscript.

CONFLICTS OF INTEREST

I.J. Amat-Santos is proctor for Boston Scientific, Medtronic, and Meril Life. There are no other potential conflicts of interest related to this work.

REFERENCES

Transcatheter aortic valve implantation (TAVI) is an established treatment for elderly patients with severe symptomatic aortic stenosis. Currently, most procedures are performed through transfemoral transcatheter aortic valve implantation (TF-TAVI) under conscious sedation via vascular access with safe and effective closure.1 Nonetheless, large-bore transcatheter arterial accesses are inherently associated with a higher risk of vascular and hemorrhagic complications.2

The use of vascular closure devices (VCDs) in endovascular procedures has grown exponentially. Their emergence has reduced time to hemostasis, facilitated early ambulation and discharge, and reduced hospitalization costs.2-4

However, failed VCDs are not an uncommon finding (1%-8%), and vascular and hemorrhagic complications—most commonly found at the access site—can sometimes determine procedural success or failure.5,6 Vascular complications significantly contribute to clinical outcomes in the short- and long-term, including length of stay, rehospitalizations, need for blood transfusions, and all-cause mortality.2,4 Therefore, successful arterial access and closure is critical in TAVI.2,7,8

CURRENT DEVICES

The most widely used VCDs for large-bore arteriotomy closure are the suture-mediated closure system, Perclose ProGlide system (Abbott Vascular, United States), and the most recent plug-based MANTA vascular closure device (Teleflex/Essential Medical, United States).

The failure mechanisms associated with the Perclose ProGlide device include suture-related malfunction, unsuccessful deployment, and incomplete vessel wall apposition. Potential errors associated with the MANTA device include failed deployment, intraluminal or subcutaneous deployment, collagen detachment, delivery handle-related arterial occlusion, and incomplete handle apposition.5,6,8

The first feasibility studies ever conducted on the MANTA device showed promising safety and efficacy. Afterwards, several registries have presented their findings that compare favorably with suture techniques.8-11

Two recent randomized clinical trials have compared the 2 techniques. The MASH trial included 210 patients and found no differences in its primary endpoint: puncture site-related major and minor complications (10% with the MANTA vs 4% with the ProGlide; P = .16).6 In contrast, in the CHOICE-CLOSURE trial of 516 patients, the MANTA device was associated with a higher rate of puncture site-related major and minor vascular complications (19.4% vs 12% with ProGlide; P = .029), and a similar incidence of puncture-site bleeding (11.6% vs 7.4%; P = .133).12

In an article published in REC: Interventional Cardiology, Martinho et al.13 present an interesting single-center observational study on the largest real-world experience with the MANTA device. The aim of the study was to evaluate the safety and efficacy profile of the device in a consecutive unselected cohort of patients referred for TF-TAVI. The registry included 245 patients from March 2020 through February 2022. The participants’ median age was 81 years, 52.7% were women, and the median EuroSCORE II was 3.15%. The 30-day TF-TAVI puncture site-related vascular complications were studied according to the Valve Academic Research Consortium (VARC) criteria. Regarding the primary outcome measure of the study—efficacy according to the VARC-3 criteria—successful puncture-site closure was achieved in 92.2% of participants. There were no major vascular or hemorrhagic complications, and only 8.6% experienced minor vascular complications according to the VARC-3 criteria (secondary safety outcome measure). The main predictors of device failure were a minimum femoral artery diameter and, consequently, a higher sheath-to-femoral artery diameter ratio, and the presence of more tortuous and calcified arterial accesses.13

Compared with the most recent clinical trials, the study by Martinho et al.13 found rates of device failure that were lower than those in the MASH trial (7.8% vs 20%), but higher than those reported by the CHOICE-CLOSURE trial (4.7%). According to the authors, some of these differences may be attributed to the varying and inconsistent definitions of device failure across the studies.6,12

According to researchers, the study limitations include its single-center retrospective design, the absence of comparisons with other closure devices, and the lack of postoperative ultrasound scans for some of the participants (which were mandatory in the CHOICE-CLOSURE trial, potentially contributing to a higher detection rate of access site-related complications).12 We should also mention that the registry included only patients considered eligible for closure with the MANTA device and excluded those with “unsuitable” anatomies.

To date, the available information, which includes numerous observational studies, 2 current randomized controlled trials, and more than 1 meta-analysis, suggests that vascular and hemorrhagic complications associated with collagen-based VCDs are quite similar to those of suture-based devices. No significant differences in the VCD failure rate have been reported, with identical 30-day all-cause mortality rates in the 2 groups.

CONCLUSIONS

The gradual reduction in the rate of vascular complications is the result of better preoperative assessment and technological support. Meticulous preoperative planning, with multidetector computed tomography for iliofemoral arterial axis study and ultrasound-guided access, should allow for an optimal approach to the access site and avoidance of incorrect cannulations.

Undoubtedly, cannulation of severely calcified arteries is associated with worse outcomes because of the difficulty involved in attaching the suture systems, and the higher risk of poor plug device apposition. Conversely, puncture sites without wall calcification should consistently yield better results, regardless of the type of VCD used.

Finally, regardless of the system used, angiographic or ultrasound confirmation after closure is always necessary to identify undetected VCD failures, vascular occlusions, or subcutaneous deployments, and expedite the implementation of corrective measures.5,13

In the search for the ideal VCD, several criteria have been proposed: successful deployment with immediate hemostasis in ≥ 98% of procedures, reduction of device-related major vascular complications to ≤ 1%, versatility to adapt to challenging anatomies, universal applicability with minimal exclusions, user-friendliness, easy access to the arterial site after closure, and the potential for early ambulation.5 Such a perfect closure device, however, is still far from being a reality, which is why VCD innovation must continue to minimize vascular complications and achieve such goals.

A less well studied yet undoubtedly pivotal variable to achieve favorable outcomes with any VCD is the operator’s experience, which is key to guaranteeing closure success and reducing VCD-related complications. However, measuring, quantifying, and analyzing this variable across different studies is a daunting task. Thus, although the perfect device may never become a reality, there will always be experienced operators who prefer certain types of VCDs.

With the current results, the decision to choose one VCD over another should be based on the patient’s anatomy and basically on the operator’s preferences and experience. An important consideration is that, to date, the MANTA device is more expensive than ProGlide, which, added to the substantially larger number of patients treated with TAVI, could pose a major obstacle to its selection in preference to suture-based systems.

In the meantime, we should consider innovations and technological advancements in the field of VCDs as welcome news for the interventional community. Soon, these advancements will likely position TF-TAVI as the technique of choice for most patients with symptomatic aortic stenosis.

FUNDING

None declared.

CONFLICTS OF INTEREST

None declared.

REFERENCES

Are the radiation doses we use in interventional procedures appropriate? Cardiologists should be able to answer this question, which is particularly important in pediatric patients. However, the answer matters not only to patients but also to the health professionals involved in these procedures. The occupational radiation doses received by health staff are associated with the doses received by patients, and “optimization” (keeping radiation doses to the minimum needed to achieve the clinical objective of the procedures involved) should be managed comprehensively for patients and professionals alike.1

The International Commission on Radiological Protection (ICRP) recommends using “diagnostic reference levels” (DRLs) to help in the optimization of imaging modalities with ionizing radiation (including interventional procedures).2

DRLs are indicative of “good clinical practice”. It is recommended that they be established for specific clinical indications and can be estimated for the local, national, or regional level by using the third quartile of the distribution of the median values of the dose indicators for patients from various centers representative of these clinical practices.2

The term “achievable dose value” has been proposed in the United States for the 50th percentile instead of the third quartile. Although the ICRP has stated that the median could be used as an additional step in optimization, the recommendation of using the third quartile recommendation to estimate DRLs still stands.2

For interventional procedures, the most widely used radiological measure is the kerma-area product (KAP), which is numerically equivalent to the dose-area product (DAP), and serves as one of the main indicators of the radiation doses received by patients. Secondary indicators that can also be used are the kerma at the patient entrance reference point (15 cm below the isocenter), fluoroscopy time, and the number of cine images acquired. These latter 2 indicators are becoming less relevant because doses depend on different image acquisition modes.

The ICRP recommends taking into consideration the complexity of interventional procedures, since it can increase DRLs significantly. Because complexity can vary widely for a single procedure, carried out for the same or similar clinical indications, it is important to assess its impact on the doses delivered to patients.3,4

The ICRP recommendations have been included in the European regulations (Directive 59/2013 EURATOM)5 and the corresponding practical guidelines of the European Union.6,7 For pediatric procedures, it is suggested that DRLs be estimated based on patient age and weight categories.2

The radiation doses received by pediatric patients vary widely depending on their size and weight. Although variations are inevitable, we should try to avoid those stemming from inappropriate use of imaging modalities (different fluoroscopy or cine modes) or protocols. DRLs help optimize radiation protection.

Different fluoroscopy and cine modes with varying dose rates (and image quality) can be used, substantially impacting the radiation doses received by patients. Factors that play a key role in delivered radiation doses are collimation, C-arm x-ray machine angles, fluoroscopy sequence recording to save cine sequences, and rotational acquisitions.

Ways to significantly reduce the radiation doses received by patients and health professionals are knowing the quality control results of x-ray machines (to understand dose differences between cine and fluoroscopy acquisitions) and fostering collaboration between hospital radiologists and cardiologists, along with continuous medical education programs on radiation safety.

All these variables associated with different operating modes can substantially change the doses delivered to patients and the quality of diagnostic information. Therefore, cardiologists’ knowledge and experience of their imaging equipment are crucial. In general, state-of-the-art machines reduce radiation doses while maintaining similar or improved diagnostic information. Quantifying all these factors and deciding whether corrective actions are needed involves comparing radiation doses for specific procedures with the DRLs.

The Royal Decree that transposes part of the European Directive to the Spanish legislation8 demonstrates the implementation and regular review of DRLs. If these DRLs are consistently and significantly exceeded, or if image quality deteriorates repeatedly, the corresponding local reviews should be undertaken and appropriate corrective measures should be implemented without delay.

Some automated dose management systems allow real-time reception and processing of the radiation doses received by patients and operators. These systems can create alerts for safer interventional practices.9,10

Several studies have been published on DRLs in interventional cardiology for adult patients (DOCCACI program) in Spain11,12 with collaboration from the Spanish Society of Cardiology.

No nationwide results have been published on the doses received by pediatric patients in interventional cardiology until now. However, Rueda Núñez et al.13 recently presented the results of the Radcong-21 Registry conducted by the Cardiac Catheterization Working Group of the Spanish Society of Pediatric Cardiology and Congenital Heart Disease (GTH-SECPCC) on the overall values from a sample of 1090 procedures across 10 different hospitals. This registry represents a significant initiative that could encourage other centers to compare their values with dose indicators obtained from a representative sample of multiple Spanish hospitals in patients with congenital heart disease treated with cardiac catheterization and categorized by type of procedure and estimated radiation risk (ERR).

The study authors used medians, although DRL values refer to the third quartile of the distribution of the median values in the different centers involved. Specific DAP/kg values are provided for certain specific procedures to treat prevalent conditions such as aortic coarctations, atrial septal defects, ductus arteriosus occlusions, aortic and pulmonary valvuloplasties and pulmonary valve implantations following the methodology proposed by Quinn et al.14 in the United States.

DAP/kg/fluoroscopy is a parameter that can be confusing when comparing radiation doses. This is because the total DAP includes contributions from fluoroscopic imaging—corresponding to different fluoroscopy modes with very different dose values—and cine acquisitions.

To facilitate comparisons and potential optimization efforts, the authors could provide dose indicator values (DAP/kg) tailored to a wider range of procedure types in future updates of the results. They could also use the third quartiles of the distribution of the median values for each center for the weight categories recommended by the ICRP and European guidelines.2,6,7

We could speculate whether it would be better to perform a global analysis across groups of different procedures or an analysis specifically designed for procedures with specific clinical indications. Quinn et al.14 choose the former, while managing the DAP/kg values as the primary dose parameter. However, a global approach does not allow analysis of specific procedures requiring corrective measures when the doses delivered to some patients may be very high. These doses exceeding the “good clinical practice” threshold can be used in certain procedures, but not in others, within the 3 REC (Radiation Exposure Category) groups proposed by Quinn et al.14 in their methodology. The advantage of using DRLs per weight group is that these DLRs are estimated for specific clinical indications, thus enabling easy comparisons with the dose indicators used in different hospitals for these procedures.

The global KAP/kg values for groups of different procedures may be balanced if there are procedures using higher radiation doses than necessary (due to excessive cine acquisitions, high-dose fluoroscopy modes, lack of collimation, etc) and other procedures requiring standard doses. This could indicate overall improvement (fewer doses in procedural groups), but does not necessarily indicate improvement in all types of procedures.

The effort made by the GTH-SECPCC to obtain and process PDA/kg values represents a significant advancement that could be further expanded in the future. This could involve establishing initial DRLs (KAP values) in Spain, based on weight and age groups, following the recommendations of the European guidelines and ICRP.

FUNDING

None.

CONFLICTS OF INTEREST

None reported.

REFERENCES