1 INTRODUCTION

Transcatheter aortic valve implantation (TAVI) is nowadays the first line therapy for the treatment of aortic valve stenosis in patients at high, intermediate, and low surgical risk.^1-3 Moderate paravalvular leak (PVL) after TAVI is, since early generation devices, a well-known complication occurring in up to 22% of the cases and has been associated with worse clinical outcomes and increased long-term mortality.^4-6 Balloon post-dilation (BPD) is a widely adopted strategy to reduce the degree of PVL after TAVI in case of prosthesis under-expansion.⁷ Furthermore, BPD may play a major role also in optimizing valve expansion and therefore may lead to decreased risk of prosthesis-patient mismatch.⁸ However, it might be associated with potential damage of the prosthetic leaflets leading to more rapid deterioration and structural failure of the transcatheter valve.⁹ In addition, initial single center observations with limited populations warned that BPD might be associated with an increased risk of iatrogenic cerebrovascular events.^{7, 10} However, latter larger multicentric observational studies did not confirm these results and showed that BPD after TAVI was safe and not associated with increased risk of stroke, aortic root injury, and mortality.¹¹ Nevertheless, follow-up duration of the aforementioned studies is limited (12 months), and the impact of BPD on long-term valve hemodynamics after TAVI remains unclear. Therefore, the aim of this large multicenter observational study was to evaluate the impact of BPD on valve durability and long-term clinical outcomes after TAVI with a self-expanding prosthesis.

2 METHODS

3 RESULTS

3.3 Long-term clinical outcomes

At 6-year follow-up, no differences among patients undergoing BPD and those not undergoing BPD were found for any clinical outcome (Table 3 and Figure 1).

Table 3. Clinical outcomes up to 6-years.

Outcome	Overall (N = 1835)	BPD (N = 417)	No-BPD (N = 1418)	Hazard ratio (95% CI)	p Value
All-cause mortality	937 (51.1)	204 (48.9)	733 (51.7)	1.05 (0.9−1.22)	0.557
Cardiovascular death	263 (14.3)	67 (16.1)	196 (13.8)	1.25 (0.94−1.64)	0.120
MI	49 (2.7)	11 (2.6)	38 (2.7)	1.06 (0.54−2.07)	0.874
Stroke/TIA	115 (6.3)	18 (4.3)	97 (6.8)	0.67 (0.40−1.11)	0.117
Aortic surgery	11 (0.6)	3 (0.7)	8 (0.6)	1.32 (0.35−4.96)	0.686
New PM implantation	568 (31.3)	131 (31.7)	437 (31.1)	1.08 (0.89−1.31)	0.442
Valve thrombosis	4 (0.2)	2 (0.5)	2 (0.1)	NA	NA
Endocarditis	8 (0.4)	2 (0.5)	6 (0.4)	NA	NA

• Abbreviations: BPD, balloon post-dilation; CI, confidence interval; MI, myocardial infarction; PM, pacemaker; TIA, transitory ischemic attack.

At multivariate analysis, age at procedure (HR: 1.02; 95% CI: 1.0−1.04, p = 0.019), creatinine >2 (mg/dL) (HR: 1.88; 95% CI: 1.46−2.42, p < 0.001), diabetes (HR: 1.36; 95% CI: 1.12−1.64, p = 0.002), prior stroke/TIA (HR: 1.36; 95% CI: 1.07−1.71, p = 0.011), severe chronic obstructive pulmonary disease (HR: 1.43; 95% CI: 1.17−1.74, p < 0.001), and left ventricular ejection fraction <50% (HR: 1.27; 95% CI: 1.06−1.52, p = 0.011), were associated with an increased risk of 6-year all-cause death. Instead, body mass index was associated with a decreased risk of 6-year all-cause death (HR: 0.97; 95% CI: 0.95−0.99, p = 0.014) (Supporting Information: Table 2).

3.4 Long-term SVD and echocardiographic outcomes

At 6-year follow-up echocardiographic data was available in 174 patients (40.2%). Six patients undergoing BPD and 30 patients not undergoing BPD experienced SVD within 6 years from implant with no statically significant differences between groups (1.4% vs. 2.1%, p = 0.381) (Table 4). Assuming death as a competing risk that can prevent SVD to happen, actual analysis resulted in a 6-year CIF for SVD of 1.04% (0.29%−2.84%) in patients undergoing BPD and 0.73% (0.34%−1.38%) in patients not undergoing BPD (Supporting Information: Figure 1).

Table 4. Echocardiographic follow-up to 6-years.

	Overall (N = 1835)	BPD (N = 417)	No-BPD (N = 1418)	p Value
SVD	36 (2)	6 (1.4)	30 (2.1)	0.381
Intravalvular leak				0.906
0	109/127 (85.8)	31/34 (91.2)	78/93 (83.9)
1	15/127 (11.8)	3/34 (8.8)	12/93 (12.9)
2	2/127 (1.6)	0/34 (0)	2/93 (2.2)
3	1/127 (0.8)	0/34 (0)	1/93 (1.1)
Mean aortic gradient (mmHg)	8.4 ± 5.7	8.4 ± 5.7	8.4 ± 5.7	0.441
Moderate or severe MR	62/164 (37.8)	16/36(44.4)	46/128 (35.9)	0.352
Moderate or severe PVL	18/169 (10.7)	4/39 (10.3)	14/130 (10.8)	1
LVEF (%)	54.9 ± 8.4	53.8 ± 8.9	55.2 ± 8.2	0.683

• Note: Values are expressed as n/N of patients (%) or mean ± standard deviation.
• Abbreviations: BPD, balloon post-dilation; LVEF, left ventricular ejection fraction; MR, mitral regurgitation; PVL, paravalvular leak; SVD, structural valve deterioration.

On echocardiography, no differences between groups were found in terms of mean aortic gradient, intravalvular, and PVL at 6 years follow-up (Table 4). Changes in mean aortic gradient by group over time are depicted in Figure 2.

4 DISCUSSION

PVL following TAVR results from incomplete prosthesis apposition to the native annulus due to the extent of calcification or annular eccentricity, undersizing of the device, and/or malpositioning of the valve.^14-16 Numerous studies have shown significant rates of moderate-to-severe PVL after TAVI, which has been associated with an increased 1-year mortality.^{5, 17, 18} However, the impact of mild PVL on mortality remains controversial. Some studies have suggested that mild PVL after TAVI leads to benign clinical outcomes,^19-21 while other investigations have demonstrated that mild PVL also results in significantly higher long-term mortality.^{22, 23} One potential explanation for this difference is variability in the assessment of PVL severity, resulting in different patient populations across studies. ²⁴

BPD is the first-line strategy to reduce the degree of PVL after TAVI. BPD has been shown to obtain optimal prosthesis frame expansion, reduce the severity of PVL, improve effective orifice area, and reduce patient-prosthesis mismatch.^{25, 26} However, BPD is not exempt from risks. Aortic annulus rupture, cerebrovascular events, conduction disorders, valve embolization, and prosthesis damage have been proposed as potential complications following BPD after TAVI. Although early studies had suggested a higher risk of cerebrovascular events after BPD, more recent analysis have shown that BPD was safe and not associated with increased risk of stroke, aortic root injury and mortality.^{11, 27} However, follow-up duration of the aforementioned studies is limited (12 months) and no prior study had evaluated the impact of BPD on long-term clinical outcomes and valve durability. The present investigation provides the first evidence about long-term safety of BPD. In our study, BPD was not associated with an increased risk of all-cause mortality, cardiovascular death, cerebrovascular events, myocardial infarction, and new pacemaker implantation at 6-year follow-up.

Another major concern after BPD is long-term SVD. SVD is an acquired intrinsic prosthesis valve abnormality due to deterioration of the leaflets or supporting structures resulting in valve hemodynamic dysfunction, manifested as stenosis or regurgitation.^{28, 29} Tissue disruption over time because of mechanical stress in conjunction with abnormal flow shear stress at the surface of valve leaflets, collagen fiber disruption, and tissue calcification have been suggested as mechanisms of SVD.^{28, 29} Many conditions may impact on the natural history of SVD such as remaining valve calcification, incomplete valve expansion and BPD.²⁸ BPD has been postulated to damage the pericardial leaflets originating leaflet tears or flails and leading to local prothrombotic conditions such as exposure of tissue factor (secondary to dilation-related endothelial disruption).³⁰ These local conditions may lead to thrombosis around the prosthesis and potentially cause valve thrombosis.³⁰ However, as shown in our investigation, no statistically significant differences were found at 6-year follow-up between patients undergoing BPD and those not undergoing BPD in terms of valve thrombosis. Of note, valve thrombosis diagnosis was based on clinical suspicion and no systematic evaluation with computed tomography was performed to detect subclinical leaflet thrombosis. Notwithstanding, the impact of subclinical leaflet thrombosis on subsequent thromboembolic complications and valve function has reported conflicting results.³¹ Similarly, no statistically significant differences were observed between groups in other adverse events which could have been potentially originated at long-term follow-up after BPD such as endocarditis or intravalvular leak. In addition, changes in mean aortic gradient over time were similar between groups. All these findings are concordant with the lack of statistically significant differences in SVD among patients undergoing BPD and those not undergoing BPD at 6-year follow-up. These findings remained consistent when assuming death as a competing risk that can prevent SVD to happen (Figure 3).

BPD is a key procedural aspect in the therapeutic armamentarium of patients undergoing valve-in-valve. In this setting, BPD at very high pressures to modify or rupture the structure of a previously implanted surgical valve is performed. This strategy may expose the recently implanted valve leaflets to the high pressures of the ballooning, which could lead to tears and potentially impair durability. Despite our investigation was not specifically performed in patients undergoing valve-in-valve, 10.3% of the patients in the BPD group underwent more than one valve implantation as a bailout procedure for acute implant failure. Though our results might shed light in this setting, the impact of BPD on long-term valve durability in patients undergoing bioprosthetic valve fracture yet needs to be determined.

In summary, the findings of the present investigation are of utmost clinical relevance, supporting the role and safety of BPD after TAVI at long-term follow-up.

5 LIMITATIONS

The present study should be interpreted considering some limitations. First, the indications and technique of BPD, as well as balloon choices, were left to the discretion of the physician of each center. Therefore, detailed reasons to perform BPD were not available. Second, clinical events were reported under the responsibility of all the participating centers without a validation from a central independent committee and no CoreLab has been involved in the evaluation of the echocardiographic data neither for the pre-BPD procedure nor for the post-BPD evaluation. Third, it cannot be excluded that the lack of statistically significant differences between groups for some endpoints (i.e., new pacemaker implantation) might be due to small sample size. Fourth, the higher than expected number of patients lost at follow-up might have contributed to low event rates (i.e., endocarditis, valve thrombosis). Fifth, the number of patients surviving and with an echocardiographic follow-up available at 6 years was relatively small but, to the best of our knowledge, this study represents the only one aimed to evaluate the long-term impact of BPD reported so far. Sixth, long-term SVD follow-up was performed in a cohort of surviving patients, with death possibly exerting a competing risk that may have biased our results. To limit this effect, we reported both actuarial and actual estimates, as recommended by the EAPCI/ESC/EACTS consensus statement. Seventh, whether these results can be extrapolated to lower risk patients and different prosthesis yet needs to be determined.

6 CONCLUSIONS

BPD following the implantation of the self-expanding CoreValve and Evolut R prosthesis does not seem to be associated with an increased risk of adverse clinical outcomes or SVD at 6-year follow-up.

CONFLICTS OF INTEREST STATEMENT

Dr. Jorge Sanz has received speaking fees from Cordis, Terumo, Boston, and Medtronic. Dr. Marco Barbanti is consultant for Medtronic, Boston Scientific, and Edwards LifeSciences. Dr. Matteo Montorfano has recieved consultant fees from Abbott, Boston Scientific, Edwards, Kardia, and Medtronic. Dr. Corrado Tamburino is consultant for Medtronic. Dr. Bernhard Reimers has received speaking honoraria from Boston Scientific. The remaining authors declare no conflict of interest.