Post-transcatheter pulmonary valve (TPV) replacement (TPVR) risk of death and repeated intervention is comparable to that of surgical pulmonary valve replacement patients, according to an international TVPR procedures registry.
However, in an accompanying editorial, experts argued that this conclusion could “easily be challenged through the rate of endocarditis following TPVR,” stressing that endocarditis rates remain a “major concern” many years after initial device approvals.
The findings were published Monday online ahead of the Jan. 4 – 11 issue of the Journal of the American College of Cardiology, by lead author Doff B. McElhinney, MD, from Stanford University School of Medicine, together with colleagues.
TPVR has become the standard treatment for patients with a prosthetic conduit or valve who develop post-operative pulmonary outflow tract dysfunction, the researchers said. However, information over mid- and long-term outcomes has remained limited, prompting the new analysis.
A total of 2,475 patients undergoing TPVR who were implanted with either Medtronic’s Melody Valve (2,038 patients, 82%) or any type of Edwards Lifesciences’ Sapien valves (Sapien, Sapien XT or Sapien 3 – 438 patients, 18%) across 15 centers between July 2005 and March 2020 were included from the registry.
Of these, 95 died after TPVR, most commonly from heart failure (24 patients), followed by endocarditis (12 patients) and complications of the TPVR procedure (seven patients). The cumulative incidence of death was 8.9% (95% confidence interval [CI]: 6.9% to 11.5%) 8 years after TPVR.
The seven procedure-related deaths were from complications of coronary artery compression (three patients), complications of conduit rupture (two patients), hemorrhagic stroke (one patient), and one patient had a jailed branch pulmonary artery during TPVR.
At baseline, patients were aged between 10 months and 79 years at implant, though excluding hybrid implants, the youngest participant was 3.1 years. Stratified into age groups, the highest number of patients were aged 19 to 29 years (n = 705), followed by those aged 13 to 18 years (n = 611), then those of 0 to 12 months (n = 484), followed by age 30- to 39-year-olds (n = 338) and, finally, over 40 years (n = 330). Approximately 7% of the cohort were 50 years or older at implant, and 3% were 60 years or older.
The patients were followed for a total of 8,475 patient years, until the earliest of death, heart transplant, surgical pulmonary valve/conduit replacement, or most recent clinical contact or documentation. Median follow-up duration among the cohort was 2.8 years, although 713 patients were followed for 5 years or more and 236 were followed for 8 years or more, with significantly longer follow-up times for those receiving the Melody valve (median 3.3 years) versus Sapien (median 1.2 years).
Overall, the most common underlying cardiac diagnosis was tetralogy of Fallot, seen in 1,373 (55%) of the cohort, followed by truncus arteriosus/double-outlet right ventricle/transposition of the great arteries in 481 patients (19%), prior Ross procedure in 348 (14%) and valvar pulmonary stenosis or pulmonary atresia intact ventricular septum in 196 (8%).
The most common right ventricular outflow tract (RVOT) conduit or valve type was homograft conduit, in 1,104 of the patients (45%), followed by stented bioprosthetic valve in 553 (22%).
On multivariable analysis, age at TPVR was associated with greater death rates (hazard ratio [HR]: 1.04 per year; 95% CI: 1.03-1.06 per year; P < 0.001), with progressively and significantly shorter survival among patients who were older at TPVR. Meanwhile, for those 18 years or younger at implant, there were no significant differences in survival according to age.
The existence of a prosthetic valve in other positions (HR: 2.1; 95% CI: 1.2-3.7; P = 0.014), and an existing transvenous pacemaker/implantable cardioverter-defibrillator (HR: 2.1; 95% CI: 1.3-3.4; P = 0.004) were also associated with death.
A total of 258 patients underwent TPV reintervention, with a cumulative incidence of any TPV reintervention of 25.1% at 8 years (95% CI: 21.8% to 28.5%), and of surgical TPV reintervention of 14.4% (95% CI: 11.9% to 17.2%).
The initial reintervention was surgical pulmonary conduit/valve replacement in 136 of these patients (53%). A total of 122 underwent transcatheter reintervention – with valve-in-valve TPVR for 74 patients (29%) and balloon angioplasty of the TPV in 48 patients (19%) – 10 of whom subsequently had a surgical valve replacement.
Surgical reintervention risk factors included age (0.95 per year [95% CI: 0.93-0.97 per year]; P < 0.001), prior endocarditis (2.5 [95% CI: 1.4-4.3]; P = 0.001), TPVR into a stented bioprosthetic valve (1.7 [95% CI: 1.2-2.5]; P = 0.007) and postimplant gradient (1.4 per 10 mmHg [95% CI: 1.2-1.7 per 10 mmHg]; P < 0.001).
Ten patients died within 90 days of reintervention and seven died within 90 days of explant.
Comparability with surgical reports?
“These outcomes are comparable to what has been reported in published surgical reports,” the researchers concluded, citing for comparison a 288-patient cohort of patients who received large-diameter conduits (median age: 19 years), in which estimated survival was 87% 10 years after implant, versus a younger population of 216 patients in which 89% to 95% survival was reported at 5 to 10 years, depending on conduit type.
“Assessment of mortality risk in this cohort is probably more about the patient population than the therapy per se, aside from procedural and endocarditis-related deaths,” they added.
The findings should help inform decision-making, counseling and future studies in TPVR, the researchers said, calling for additional studies into the long-term impact of TPVR on survival, particularly in adults.
However, in an accompanying editorial, Alain Fraisse, MD, PhD, from the Royal Brompton Hospital and Harefield National Health Service Foundation Trust, London, and colleagues argued that the rate of endocarditis following TPVR could easily pose a challenge to the conclusion over similar TPVR and surgery outcomes.
They cited a multicenter registry in 845 patients treated across 42 centers, published in 2019, which concluded “infective endocarditis remains a significant concern causing significant morbidity and mortality.”
“More than 10 years after the approval of the Melody valve, the rate of endocarditis remains a major concern. Similarly, more than 15 years after the first Edwards Sapien implantation in the pulmonary position, we still do not have long-term follow-up on large cohorts (4), especially to clarify the rate of endocarditis,” said Fraisse and colleagues.
The editorialists added that a prospective post-approval study was launched in April 2021 to monitor device performance and long-term outcomes of the Sapien 3 device in the pulmonary position, but urged: “We need a more time-efficient answer with existing data. If we aim to compare in the long-term the differences between the Melody and Sapien, we may have no choice other than to work on a retrospective study.
“One limitation might be that both valves are applied to slightly different target populations, with different RVOT sizes or even different types of hemodynamic lesions.”
The editorialists also lamented the “short” median follow-up of 2.8 years as a limitation for the study.
“We believe that for a slowly progressive and chronic condition like dysfunction of the RVOT, we should target more than 8 years to reasonably assess freedom from reintervention and compare with surgical outcomes,” the editorialists said, lamenting that only 10% of the current study’s cohort falls within this scope.
“Clearly, the question of long-term transcatheter heart valve durability after TPVR and comparison with surgery remains unresolved.”
McElhinney DB, Zhang Y, Levi DS, et al. Reintervention and Survival After Transcatheter Pulmonary Valve Replacement. J Am Coll Cardiol 2022;79:18-32.
Fraisse A, Kempny A, Bautista-Rodriguez C. Outcome After Transcatheter Pulmonary Valve Replacement: Are We Focusing on the Long Term Now? J Am Coll Cardiol 2022;79:33-34.
Image Credit: sakurra – stock.adobe.com