ADVERTISEMENT
Outcome of Patients Undergoing Transcatheter Aortic Valve Implantation After Prior Balloon Aortic Valvuloplasty
Abstract: Objectives. Transcatheter aortic valve implantation (TAVI) is an effective treatment for severe aortic stenosis. Yet, balloon aortic valvuloplasty (BAV) is still occasionally offered to apparently inoperable patients as bridge or destination therapy, with some eventually receiving TAVI. We aimed to determine whether prior BAV would unfavorably impact TAVI. Methods. The RISPEVA dataset was analyzed to compare the outcomes of non-BAV subjects undergoing TAVI vs those with prior BAV. Outcomes of interest were procedural results, hospital stay, and in-hospital outcomes, including major adverse events. Unadjusted and propensity-matched analyses were performed. Results. A total of 1683 patients were included (1541 [91.6%] non-BAV and 142 [8.4%] prior BAV). Device success, procedural success, major adverse events and their individual components were all similar in the two groups, in both unadjusted and propensity-matched analyses. However, non-BAV patients had significantly shorter intensive/coronary care unit stay (8.1 ± 23.0 days vs 13.4 ± 35.2 days; P=.03) and total hospital stay (12.4 ± 22.1 days vs 17.2 ± 32.7 days; P=.02). Similar findings were obtained even in the propensity-matched groups (6.3 ± 11.9 days vs 13.6 ± 35.5 days [P=.05] and 10.5 ± 13.8 days vs 17.4 ± 33.1 days [P=.03], respectively). No significant interaction was found between device type and prior BAV status. Conclusion. BAV continues to offer a palliative solution to patients with unclear or poor prognosis and deemed at too high risk for emergent TAVI or surgery. Subjects undergoing BAV and surviving the post-BAV period can undergo TAVI without a significantly increased risk of in-hospital adverse events in comparison with non-BAV patients.
J INVASIVE CARDIOL 2018;30(10):380-385.
Key words: balloon aortic valvuloplasty, transcatheter aortic valve implantation
The management of degenerative aortic stenosis (AS) has seen a paradigm shift with the introduction of transcatheter aortic valve implantation (TAVI) more than a decade ago.1-3 While surgery remains the treatment of choice for low-risk patients, high-risk patients – and possibly intermediate-risk patients – substantially benefit from TAVI.2-6 Historically, balloon aortic valvuloplasty (BAV) was considered a reasonable treatment for severe AS in inoperable patients, but the risk of complications and restenosis was often perceived as too high to consider this option in routine practice.7
Refinements in technique and patient selection, with the expansion in referral programs, have led to a renaissance in BAV, which is today often considered as a bridge treatment in those awaiting TAVI or as destination therapy in those unsuitable for TAVI.8-11 There is uncertainty, however, on the soundness of performing BAV in patients who might eventually be candidates for TAVI, as this might impact unfavorably on their procedural and short-term outcomes.12
Selected reports with small to moderate samples or focusing on first-generation devices have suggested that TAVI after BAV is associated with similar procedural and clinical results as TAVI in non-BAV patients.13-15 However, it is unclear whether this also applies to new-generation devices. We aimed to compare the outcome of patients with prior BAV undergoing TAVI vs those without any prior intervention, exploiting a large, prospective, national TAVI registry.
Methods
Design. This work represents a subanalysis of a large ongoing Italian prospective registry on TAVI (Registro Italiano GISE sull’impianto di Valvola Aortica Percutanea [RISPEVA]). Details on this study have been reported elsewhere.16 Notably, the study was approved by all ethics committees of participating centers, and all patients provided written informed consent. The study is registered at www.clinicaltrials.gov (NCT02713932).
Patients, procedures, and outcomes. As previously stated,16 all patients in whom TAVI was attempted at participating centers and willing to provide consent were offered inclusion in the registry, without other exclusion criteria. Subject selection, preprocedural management, procedural technique, device choice, and subsequent management were all largely informed by contemporary best-practice recommendations. While several baseline variables were collected in a dedicated electronic case report form, only the dichotomous feature of whether or not BAV before the index admission had been performed was entered into the dataset. Clinical assessments were planned post procedure and at discharge.
Details on the following outcomes were systematically collected: death, stroke, myocardial infarction (MI), major vascular complication, major bleeding, pacemaker dependency, device success, and procedural success. In addition, we computed major adverse event (MAE) rate, defined as the composite of death, MI, stroke, major vascular complication, or major bleeding (discounting multiple events occurring in the same patient). All definitions originated by the current Valve Academic Research Consortium recommendations.17
Statistical analysis. Continuous variables are reported as mean ± standard deviation, and categorical variables as count (percentage). Unadjusted analysis was based on unpaired Student’s t-test for continuous variables and Fisher’s exact test for categorical variables. One-to-one propensity-matching was performed with a nearest-neighbor approach, deriving propensity scores from an extensive set of baseline, preprocedural, and procedural features (Supplemental Table S1).18,19 Subsequently, unpaired Student’s t-test and Fisher’s exact test were used for inference. Statistical significance was set at the two-tailed .05 level without multiplicity adjustment. Computations were performed with Stata 13 (StataCorp).
Additional statistical details. After missing data imputation using the “mi impute mvn” command, the following variables were entered in a logistic regression model with the psmatch2 used to obtain propensity scores: age, gender, height, weight, body surface area, body mass index, aortic regurgitation, degenerated prosthesis, diabetes mellitus, dyslipidemia, smoking, peripheral artery disease, hypertension, systolic blood pressure, diastolic blood pressure, prior stroke, prior cardiac surgery, prior percutaneous coronary intervention, estimated glomerular filtration rate, cirrhosis, chronic obstructive pulmonary disease, prohibitive surgical risk, Logistic EuroScore, EuroScore II, Society of Thoracic Surgeons (STS) score, Frailty score, New York Heart Association (NYHA) class, left ventricular ejection fraction (LVEF), peak aortic gradient, mean aortic gradient, device, general anesthesia, percutaneous approach, and femoral access.
C statistic before matching was 0.77. Matching was based on a 1:1 approach with a nearest neighbor algorithm. Notably, 3 patients in the BAV group were not matched given their extreme propensity scores. The eventual propensity scores were 0.19 ± 0.15 in the non-BAV group and 0.19 ± 0.16 in the prior BAV group.
Results
A total of 1683 patients treated with new-generation TAVI devices were included. Of these, 1541 (91.6%) were non-BAV patients and 142 (8.4%) had undergone prior BAV (after 11.3 ± 36.5 months).
Unadjusted analysis highlighted several significant baseline differences (Table 1). Specifically, prior BAV patients were taller, with a higher prevalence of dyslipidemia, poorer renal function, higher prevalence of prior percutaneous coronary intervention and permanent pacemaker implantation, higher logistic EuroScore, EuroScore II, and STS scores, and worse NYHA class (all P<.05). In addition, LVEF was lower in prior BAV patients, whereas peak and mean aortic valve gradients were higher in non-BAV subjects (all P<.05).
Significant differences in procedural features were also found (Table 2). In particular, patients with prior BAV were less likely to undergo femoral access and percutaneous approach, and were more likely to receive antiembolic protection and Edwards TAVI devices (all P<.05). Contrast volume was higher in subjects with prior BAV (P<.001), but procedural time, device success, and procedural success were similar in those with vs without prior BAV (all P>.05). Postprocedural echocardiographic details were similar in the two groups, except for a significantly lower LVEF in patients with prior BAV (P<.001).
Clinical outcome rates were similar in the two groups, including MAE rates (Table 3; Figure 1). However, intensive/coronary care unit stay and total hospital stay were significantly longer in patients with prior BAV, whereas LVEF was lower and the rate of aortic regurgitation >2+ was higher (all P<.05). No significant interaction in the MAE rate was found after stratifying by device type, suggesting that new-generation devices fared similarly in those with vs without prior BAV (Supplemental Table S2). The MAE rate was also similar between patients stratified according to time between BAV and TAVI (Supplemental Table S3).
Propensity-matched analysis confirmed the unadjusted analysis, showing similar clinical and echocardiographic outcomes (Table 4). The only feature that remained significantly different in non-BAV patients vs those with prior BAV was total hospital stay, which proved to be longer in the latter group (P=.03).
Discussion
Severe AS in patients fit for surgery is optimally treated with surgical aortic valve replacement with a mechanic or bioprosthetic valve.20,21 The management of patients at high surgical risk has historically been challenging. BAV was introduced in the 1980s as a palliative treatment for subjects unfit for surgery, but the periprocedural risk and rate of restenosis were quite high, limiting its widespread adoption.7,21 The advent of TAVI and its clearly favorable risk-benefit balance in intermediate-risk and high-risk patients has revolutionized the role of interventional cardiology in subjects with severe AS, creating renewed interest in BAV as well.2-6,22
In particular, BAV has seen important refinements in devices, procedural details, and patient selection, leading to important improvements in early and subsequent outcomes.7-8,10,12 More importantly, BAV has been proposed in several case series as a potential bridge to TAVI in patients with a temporary contraindication or as destination therapy in subjects who are unsuitable for various reasons to TAVI but still merit a proactive management.12 A crucial issue in the concept of BAV as bridge to TAVI is whether such valvuloplasty procedures may favorably or unfavorably impact on the subsequent TAVI procedure.7
The present study builds upon prior reports on patients undergoing BAV and followed over time,8-11,23 as well as on studies focusing on the outcome of TAVI in subjects with prior BAV, albeit limited by small sample size and inclusion of first-generation TAVI devices only.13-15 In particular, we exploited the extensive RISPEVA dataset, appraising the outcomes of patients undergoing TAVI with, mostly, leading new-generation devices, and comparing non-BAV patients to patients with prior BAV. We found indeed that the latter group was characterized by a higher surgical risk and several other baseline and procedural features with unfavorable prognostic impact. Despite this, device and procedural success rates were similar in the two groups, as were all clinically relevant outcomes at short-term follow-up. The only significant difference was in-hospital stay (mainly intensive/coronary care unit and total stay). These findings thus, while confirming the role of BAV as a suitable palliation for subjects with contraindications to emergent TAVI or surgery, suggests that patients undergoing BAV and surviving the post-BAV period can undergo TAVI without a significantly increased risk of in-hospital adverse events in comparison to non-BAV patients. Beyond this, the question that remains most open relates to the timing of TAVI after BAV. It could be argued that TAVI should be performed before restenosis occurs (typically within 3-9 months).7,10,12,15,23
Conversely, the limitations of BAV as destination therapy should remain apparent, in light of the remarkable results of TAVI in this study and, most importantly, in the landmark randomized PARTNER trial (cohort B).3 This study demonstrated the superiority of TAVI in comparison to medical therapy, which actually included BAV in as many as 83.8%, in terms of overall and event-free survival. Indeed, while balloon predilation has been recommended routinely for the implantation of first-generation devices, recent research challenges even its intraprocedural role with new-generation devices.24 The higher rate of aortic regurgitation after TAVI in patients with prior BAV is worth emphasizing. Whether this is due to chance or an actual traumatic effect of the prior BAV procedure is unclear, but it can be argued that patients with prior BAV might be better off with a TAVI device capable of minimizing regurgitation risk.
Study limitations. This work has several drawbacks, mainly stemming from its non-randomized design and selection bias. Indeed, we did not include all patients who underwent BAV, but only those who eventually underwent TAVI. Thus, survivorship bias remains a major validity threat. Accordingly, while this still remains unproven, BAV might still be detrimental acutely or subsequently for patients later considered for TAVI. In addition, several key baseline and procedural details on BAV were not collected, and thus we cannot appraise the impact of periprocedural and postprocedural management on BAV results, and the same applies to the many clinically relevant comorbidities (eg, frailty, malignancy, end-stage liver or renal disease), which represent typical indications to BAV. Moreover, several patients underwent TAVI with the CoreValve (Medtronic) first-generation TAVI device. Finally, given the relatively small sample size and limited details in the dataset, we could not appraise the impact of TAVI timing or operator/center experience on clinical outcomes.
Conclusion
BAV maintains its role as a palliative solution in subjects with severe AS and unclear/poor prognosis who are deemed to be too high risk for emergent TAVI or surgery. Subjects undergoing BAV and surviving the post-BAV period can undergo TAVI without a significantly increased risk of in-hospital adverse events in comparison with non-BAV patients.
References
1. Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description. Circulation. 2002;106:3006-3008.
2. Kodali SK, Williams MR, Smith CR, et al; PARTNER Trial Investigators. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012;366:1686-1695.
3. Makkar RR, Fontana GP, Jilaihawi H, et al; PARTNER Trial Investigators. Transcatheter aortic-valve replacement for inoperable severe aortic stenosis. N Engl J Med. 2012;366:1696-1704.
4. Adams DH, Popma JJ, Reardon MJ, et al; U.S. CoreValve Clinical Investigators. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med. 2014;370:1790-1798.
5. Leon MB, Smith CR, Mack MJ, et al; PARTNER 2 Investigators. Transcatheter or surgical aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2016;374:1609-1620.
6. Reardon MJ, Van Mieghem NM, Popma JJ, et al; SURTAVI Investigators. Surgical or transcatheter aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2017;376:1321-1331.
7. Pendyala LK, Ben-Dor I, Waksman R. Evolution of percutaneous balloon aortic valvuloplasty in the treatment of patients with aortic stenosis. Minerva Med. 2012;103:415-429.
8. Moretti C, Chandran S, Vervueren PL, et al. Outcomes of patients undergoing balloon aortic valvuloplasty in the TAVI era: a multicenter registry. J Invasive Cardiol. 2015;27:547-553.
9. Alkhouli M, Zack CJ, Sarraf M, et al. Morbidity and mortality associated with balloon aortic valvuloplasty: a national perspective. Circ Cardiovasc Interv. 2017;10:e004481.
10. Saia F, Moretti C, Dall’Ara G, et al. Balloon aortic valvuloplasty as a bridge-to-decision in high risk patients with aortic stenosis: a new paradigm for the heart team decision making. J Geriatr Cardiol. 2016;13:475-482.
11. Kamperidis V, Hadjimiltiades S, Mouratoglou SA, et al. Aortic balloon valvuloplasty before transcatheter valve replacement in high-risk patients with aortic stenosis. Cardiac catheterization and echocardiographic hemodynamic study. Herz. 2016;41:144-150.
12. Feldman T, Guerrero M. Balloon aortic valvuloplasty in the TAVR era: not so new but definitely improved. Catheter Cardiovasc Interv. 2017;90:311-312.
13. Bière L, Durfort A, Fouquet O, et al. Baseline characteristics and outcomes after transcatheter aortic-valve implantation in patients with or without previous balloon aortic valvuloplasty: insights from the FRANCE 2 registry. Arch Cardiovasc Dis. 2017;110:534-542. Epub 2017 May 3.
14. Singh K, Carson K, Akbar Ali O, et al. Efficacy and complications of transcatheter aortic valve implantation with and without balloon aortic valvuloplasty. J Heart Valve Dis. 2017;26:139-145.
15. Ussia GP, Capodanno D, Barbanti M, et al. Balloon aortic valvuloplasty for severe aortic stenosis as a bridge to high-risk transcatheter aortic valve implantation. J Invasive Cardiol. 2010;22:161-166.
16. Giordano A, Corcione N, Biondi-Zoccai G, et al. Patterns and trends of transcatheter aortic valve implantation in Italy: insights from RISPEVA. J Cardiovasc Med (Hagerstown). 2017;18:96-102.
17. Kappetein AP, Head SJ, Généreux P, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. Eur Heart J. 2012;33:2403-2418.
18. Biondi-Zoccai G, Romagnoli E, Agostoni P, et al. Are propensity scores really superior to standard multivariable analysis? Contemp Clin Trials. 2011;32:731-740.
19. Biondi-Zoccai G, Marullo AGM, Peruzzi M, et al. Last nail in the coffin for propensity scores in observational cardiovascular studies? J Am Coll Cardiol. 2017;69:2575-2576.
20. Nishimura RA, Otto CM, Bonow RO, et al. 2017 AHA/ACC focused update of the 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2017;70:252-289.
21. Erbel R, Aboyans V, Boileau C, et al; ESC Committee for Practice Guidelines. 2014 ESC guidelines on the diagnosis and treatment of aortic diseases: document covering acute and chronic aortic diseases of the thoracic and abdominal aorta of the adult. The Task Force for the Diagnosis and Treatment of Aortic Diseases of the European Society of Cardiology (ESC). Eur Heart J. 2014;35:2873-2926.
22. Biondi-Zoccai G, Peruzzi M, Abbate A, et al. Network meta-analysis on the comparative effectiveness and safety of transcatheter aortic valve implantation with CoreValve or Sapien devices versus surgical replacement. Heart Lung Vessel. 2014;6:232-243.
23. Safian RD, Berman AD, Diver DJ, et al. Balloon aortic valvuloplasty in 170 consecutive patients. N Engl J Med. 1988;319:125-130.
24. Hamm K, Reents W, Zacher M, et al. Omission of predilation in balloon-expandable transcatheter aortic valve implantation: retrospective analysis in a large-volume centre. EuroIntervention. 2017;13:e161-e167.
From the 1Unità Operativa di Interventistica Cardiovascolare, Presidio Ospedaliero Pineta Grande, Castel Volturno, and Unità Operativa di Emodinamica, Casa di Salute Santa Lucia, San Giuseppe Vesuviano, both in Italy (AG, NC, PF); 2Department of Cardiology, IRCCS Policlinico San Donato, San Donato Milanese, Milan, Italy (FB, LT); 3Cardiovascular, Respiratory, Nephrologic and Geriatric Sciences Department, Umberto I Hospital, Sapienza University of Rome, Rome, Italy (GS, MM); 4Division of Cardiology, European Hospital, and Division of Cardiology, Aurelia Hospital, both in Rome, Italy (FT, GDP); 5Division of Cardiology, Ospedale Le Scotte, Siena, Italy (AI); 6Department of Medico-Surgical Sciences and Biotechnologies, Sapienza University of Rome, Latina, and Department of AngioCardioNeurology, IRCCS Neuromed, Pozzili, both in Italy (GF, GBZ).
Funding: The RISPEVA study was supported by unrestricted grants from Edwards Lifesciences and Medtronic.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Professor Biondi-Zoccai reports consultant income from Abbott and Bayer. The remaining authors report no conflicts of interest regarding the content herein.
Manuscript submitted February 23, 2018, provisional acceptance given May 7, 2018, final version accepted July 3, 2018.
Address for correspondence: Prof. Giuseppe Biondi-Zoccai, Department of Medico-Surgical Sciences and Biotechnologies, Sapienza University of Rome, Corso della Repubblica 74, 04100 Latina, Italy. Email: gbiondizoccai@gmail.com
