Development of the Minimalist Approach for Transcatheter Aortic Valve Replacement at a Veterans Affairs Medical Center

Abstract

Objectives. While a minimalist transcatheter aortic valve replacement (TAVR) approach has shown safety and efficacy at civilian hospitals, limited data exist regarding developing this approach at Veterans Affairs (VA) medical centers (VAMCs). We implemented TAVR with minimalist approach (MA) using conscious sedation (CS) with transthoracic echocardiography (TTE) and compared safety and outcomes with general anesthesia (GA) with transesophageal echocardiography (TEE) at a university-affiliated VAMC. Methods. A total of 258 patients underwent transfemoral TAVR at a VAMC between November 2013 and October 2019. Ninety-three patients underwent GA/TEE and 165 patients underwent CS/TTE with dexmedetomidine and remifentanil. Propensity-score matching with nearest-neighbor matching was used to account for baseline differences, yielding 227 participants (81 GA, 146 CS). Results. MA-TAVR had no effect on 30-day mortality or paravalvular leakage. No differences were found in permanent pacemaker implantation, major vascular complications, or postoperative hemodynamics. In this population, MA-TAVR did not reduce procedural time, hospital length of stay, or intensive care unit length of stay. Conclusions. Unlike civilian hospitals, MA with CS/TTE did not reduce overall length of stay in the veteran population; however, it was safe and effective for transfemoral TAVR without impacting clinical outcomes of mortality, major vascular complications, and paravalvular leakage.

J INVASIVE CARDIOL 2021;33(2):E108-E114. doi:10.25270/jic/20.00389

Key words: conscious sedation, general anesthesia, TAVR

Transcatheter aortic valve replacement (TAVR) has rapidly become an accepted alternative to surgery for the treatment of aortic stenosis in intermediate-, high-, and prohibitive-risk patients.^1,2 The results from the recently published PARTNER (Placement of AoRTic TraNscathetER Valve) 3 trial³ and Evolut Low-Risk trial⁴ demonstrated superiority of balloon-expandable TAVR and non-inferiority of self-expanding TAVR over surgical aortic valve replacement in low-risk patients, respectively, cementing the path for TAVR’s expansion to an even larger patient population.^1,5

As TAVR becomes more widely used, studies have examined the use of a minimalist approach (MA) with conscious sedation (CS) and transthoracic echocardiography (TTE) vs the use of general anesthesia (GA) and transesophageal echocardiography (TEE) for TAVR cases. These studies, often performed at high-volume TAVR centers, suggest the safety and efficacy of MA-TAVR.^6–8 The definition of MA varies, but typically encompasses minimizing procedural elements and implementing postprocedure protocolized pathways to achieve early discharge. CS in TAVR has been associated with shortened length of stay, decreased procedure time, and decreased costs, without impacting incidence of procedural mortality or major complications compared with GA.^9–11 In our MA, we opted to use CS and intraprocedural TTE as opposed to GA and intraprocedural TEE, and gradually minimized adjunctive procedures by selective use of central venous access, minimizing Foley catheter usage but maintaining admission to intensive care unit (ICU) post procedure. Outcomes of TAVR patients in the federal system have recently been reported,¹² demonstrating excellent short-term results. However, the MA has not been uniformly adopted at Veterans Affairs (VA) medical centers. Thus, outcomes for the safety and efficacy of MA-TAVR remain largely unknown, since VA data are not captured by the Society of Thoracic Surgeons/American College of Cardiology (STS/ACC) Transcatheter Valve Therapies (TVT) registry. VA outcomes are separately reported to national VA databases for benchmarking and quality-improvement purposes, but there is value in reporting these data publicly for transparency given the overall high comorbidity burden in this population, as well as for patient preference with regard to choice of healthcare facility.

Since the first VA-TAVR program was established in 2011, several centers have emerged across the country, including our VA medical center, which is 1 of the first 5 VA-approved TAVR sites.¹³ The perioperative outcomes at our VAMC have previously been described.^12,14 This study expands upon the previously reported results and investigates procedural outcomes of our TAVR program as the MA was adopted.

Methods

We retrospectively reviewed 289 patients at a VA medical center undergoing TAVR between November 2013 and October 2019. Patients undergoing TAVR with concomitant percutaneous coronary intervention (PCI) and cases of elective valve-in-valve procedures for surgical bioprosthetic valve degeneration were excluded, with 258 patients remaining. At our center, TAVR was initially performed using GA and TEE with a gradual transition to the use of CS and TTE while the program transitioned to nearly uniform use of MA. Our MA program involved using CS with dexmedetomidine and remifentanil, titrated such that the patient is only purposefully responsive to verbal or tactile stimulation while maintaining adequate spontaneous ventilation. It also used TTE rather than TEE, preferential use of condom instead of Foley catheters, and a selective insertion of internal jugular lines based on preoperative parameters. In contrast, the traditional approach used GA, TEE, central lines, and Foley catheters. All post-TAVR patients automatically were admitted to ICU, with subsequent transfer to a step-down unit, telemetry unit, or discharge depending upon clinical stability and bed availability.

The VA computerized patient record system (CPRS) records were used to investigate TAVR outcomes. Ninety-three patients underwent TAVR with GA, while 165 patients underwent MA with CS. Data were collected based on STS cardiac surgery guidelines.¹⁵ In the analysis of our data, patients were propensity-score matched to account for differences in baseline characteristics.^16,17 Propensity scores were estimated using a logistic regression model based on factors found to be significantly different at baseline, including STS mortality score, age, prior myocardial infarction (MI), prior coronary artery bypass grafting (CABG), prior PCI, chronic lung disease, and aortic valve area (Figure 1).¹⁸ Propensity-score matching using a nearest-neighbor algorithm with 1 neighbor yielded 227 patients whose data were supported by the algorithm, comprised of 81 patients who underwent TAVR with GA and 146 patients who underwent TAVR with CS. Statistical analyses were performed using Stata version 16 (StataCorp), with P<.05 considered statistically significant.

Results

Baseline patient characteristics. Between November 2013 and October 2019, a total of 258 TAVR procedures were performed at the San Francisco VA medical center. A total of 165/258 cases (64.0%) were performed under MA with CS/TTE and 93/258 cases (36.0%) were performed under GA/TEE. The procedure was converted from CS to GA in 4/258 cases (1.55%) for hypotension in 1 case, apnea in 1 case, and vascular complications in 2 cases. Preoperative demographics and comorbidities are summarized in Table 1. Eighty-nine of the 93 GA patients (95.7%) and 163/165 MA patients (98.8%) were male (P=.11). Mean age was 80.2 years for GA patients and 77.5 years for MA patients (P=.02). GA patients had higher surgical risk (STS predicted risk of mortality [PROM] score, 4.89% vs 3.03% for GA vs MA, respectively; P<.001). Preoperative ejection fraction (P=.81), smoking status (P=.06), diabetes (P=.34), and hypertension (P=.66) were not significantly different between GA and MA patients. GA patients were more likely to have a history of MI (P<.001), CABG (P<.01), PCI (P=.01), and chronic lung disease (P<.001). Prior stroke (P=.38) was not statistically significant between GA and MA patients. New York Heart Association class was not significantly different between GA and MA patients (P=.16). Differences in aortic valve area were statistically, although not clinically, significant (0.714 cm² for GA vs 0.776 cm² for MA; P<.01). Preoperative aortic valve mean pressure gradient was not significantly different (44.0 mm Hg for GA vs 42.2 mm Hg for MA; P=.21). After propensity-score matching was implemented, a total of 227 patients remained, including 81 patients who underwent TAVR with GA/TEE and 146 patients who underwent MA-TAVR with CS/TTE.

Procedural characteristics. Table 2 shows procedural characteristics in the GA and MA groups. Of the 93 GA patients, 17 were treated with Sapien, 28 with Sapien XT, and 19 with Sapien 3 valves (Edwards Lifesciences), while 17 were treated with CoreValve, 10 with CoreValve Evolut R, and 2 with CoreValve Evolut Pro (Medtronic). Of the 165 CS patients, 2 were treated with Sapien XT, 132 with Sapien 3, 19 with CoreValve Evolut R, and 12 with CoreValve Evolut Pro valves. Central lines were used in 97.8% of GA cases vs 83.0% of MA cases (P<.001). Foley catheters were used in 96.8% of GA cases vs 16.4% of MA cases (P<.001). Redo TAVR-in-TAVR procedures were later performed in 5 patients, with 4 under GA and 1 under CS, due to worsening paravalvular leak or device migration associated with patient-specific anatomy. The repeat procedures were performed under GA. Among the propensity-matched groups, no patients required cardiac surgical intervention for any complication during the index TAVR admission.

Procedural outcomes. Table 3 describes perioperative characteristics and outcomes. Our MA with CS was associated with shorter procedure time (93.1 minutes in the MA group vs 132.1 minutes in the GA group), although the difference was not statistically significant after propensity-score matching (P=.20). CS was associated with a decrease in contrast use (125.5 mL in the MA group vs 155.4 mL in the GA group; P<.01) and a decrease in fluoroscopy time (17.0 minutes in the MA group vs 26.1 minutes in the GA group; P<.01). Postprocedure hospital length of stay, defined as the number of days from TAVR procedure until hospital discharge (P=.46), and ICU stay (P=.54) were similar between the 2 groups. Thirty-day postoperative mortality rate was 1.37 in the MA group vs 1.23% in the GA group (P=.42). The difference in the need for blood transfusion was statistically significant (2.74% in the MA group vs 13.6% in the GA group; P=.02). Postoperative stroke was 2.05% in the MA group vs 3.70% in the GA group (P=.07). Rates of permanent pacemaker (P=.24), atrial fibrillation (P=.57), and cardiac arrest (P=.25) were not significantly different between the 2 groups. The rate of major vascular complications, as defined by the Valve Academic Research Consortium-2 criteria,¹⁹ was not significantly different between the 2 groups (P=.47). The need for a repeat valve procedure due to worsening paravalvular leak or device migration was not significantly different between the 2 groups (P=.57). Postoperative paravalvular leak was also not significantly different between the MA and GA groups (P=.07). Similarly, postoperative aortic valve mean pressure gradient was not significantly different between the MA and GA groups (P=.91). Mild paravalvular leak rates were 12.3% for the MA group vs 19.8% for the GA group, while moderate paravalvular leak rates were 0% for the MA group vs 4.94% for the GA group. No paravalvular leak exceeding moderate degree occurred in either group. Postoperative ejection fraction was not statistically significant between the 2 groups (P=.08).

Discussion

TAVR under MA in civilian populations has been shown to improve clinical outcomes and quality of life while lowering costs.⁹ However, the limited data reported regarding TAVR outcomes at federal institutions¹² did not examine the MA, which has not been uniformly adopted, and national TAVR registries do not capture VA outcomes. This retrospective review examined the outcomes of 258 patients at a single VA medical center as the TAVR program transitioned to nearly uniform use of MA with CS/TTE from GA/TEE. Propensity-score matching using a nearest-neighbor algorithm yielded analyses of 227 patients (88.0% of the original 258 patients). Our data showed that the adoption of MA did not impact 30-day mortality or paravalvular leakage rates. Differences in procedural time were not significantly different once propensity-score matching was employed. Our results demonstrate that the adoption of MA did not compromise outcomes in the VA patient population, while procedural benefits were conferred to patients.

In a smaller study of TAVR at a VA medical center, MA for TAVR was associated with decreased procedural times, fluoroscopy times, and contrast use.²⁰ Our much-larger study of 258 patients also showed decreased fluoroscopy time and contrast use, but not lower procedural times, and we accounted for baseline population differences with propensity-score matching. We also described MA development for our VA-TAVR program with avoiding Foley insertion and reducing central venous line placement in addition to conversion from GA/TEE to CS/TTE. We found no significant differences in postoperative mean pressure gradient or postoperative ejection fraction using MA vs GA/TEE. No significant differences were seen in 30-day mortality, stroke, new-onset atrial fibrillation, cardiac arrest, and major vascular complications between MA and GA/TEE. There was no significant difference in the rates of post-TAVR permanent pacemaker implantation in the MA and GA/TEE groups. Similarly, no significant difference was found for rates of repeat valve procedures between the MA and GA/TEE groups. Four of the 5 cases requiring repeat valve procedures took place under GA/TEE, suggesting that MA does not lead to increased rates of device malposition. The most likely explanation for these repeat procedures is the learning curve associated with the TAVR procedure itself. Overall, the results suggest that these outcomes are comparable between TAVR performed with MA and TAVR performed under GA.

Unlike the civilian population, our data found no significant difference in postprocedure hospital length of stay or in postprocedure ICU length of stay. Civilian population studies have shown an association between MA techniques and early discharge. The Vancouver TAVR clinical pathway showed significant differences in the use of local anesthesia/CS and avoidance of a urinary catheter when comparing early discharge at 1 day vs standard discharge at 3 days.^21,22 In contrast, our data showed no association, with 19.8% of GA patients discharged before 3 days compared with 19.2% of MA patients (P=.84). The literature suggests that lengths of stay >5 days are an independent predictor of 30-day readmission.²³ When examined by stays <5 days, our data showed that 60.4% of GA patients were discharged by day 5 compared with 75.3% of MA patients (P=.03) (Figure 2). These results suggest that our transition to MA was done safely.

The inability to reduce hospital length of stay is unique to our VA population. Most patients are referred from as far as Los Angeles to the south and Eureka to the north. As such, length of stay disposition in the hospital is frequently longer to arrange transportation. Many of our patients rely on VA shuttles, which are not available on the weekends, limiting our ability to offer early discharge at the end of the work week. Similarly, during our study, the ICU length of stay did not change because the telemetry floors were unable to accommodate patients who might be at risk of requiring a temporary pacemaker or at risk of vascular complications. As such, TAVR patients were kept in the ICU for 48 hours of monitoring. Interestingly, with the advent of COVID-19 and limitations in ICU beds after this study was completed, ICU bed utilization has decreased to <48 hours to accommodate sicker hospital patients. Our policy has evolved to include monitoring in the postanesthesia care unit for patients not requiring a pacemaker and without bleeding rather than sending these patients directly to the ICU, given our resource limitations in the setting of COVID-19.

Our structural heart team determined the use of TAVR vs surgical aortic valve replacement informed by concurrent guidelines, which changed over the course of the study period as high-, intermediate-, and low-risk trials extended the use of TAVR to more patients.³ For the TAVR cases, the decision to use GA/TEE vs MA with CS/TTE involved the collaboration of the anesthesia and structural heart teams. Contraindications to CS include patient preference for GA, which was rare, the inability to lie flat and/or still, the inability to cooperate under sedation, and difficult airway in a setting where conversion to GA was more likely from a cardiac standpoint. Since 2016, CS has become preferred over GA in the absence of these contraindications because of its theoretical benefits.

Our data showed that when compared with MA patients, GA patients had higher STS mortality scores, were older, were more likely to have had a prior MI, were more likely to have a prior CABG or prior PCI, and were more likely to have chronic lung disease. These demographic differences are consistent with findings at other institutions, which showed that patients receiving CS had lower STS mortality scores than patients receiving GA.²⁴ These differences are likely related to the necessity of using GA in patients with more severe cardiac and pulmonary disease, as well as the presence of other comorbidities. For instance, severe congestive heart failure affects a patient’s ability to lie flat, and dementia affects a patient’s ability to lie still. In these settings, a patient may not be a candidate for MA with CS/TTE. Propensity matching enabled us to account for some of these differences, with reductions in standardized percent bias in all significantly different categories (Table 1). Yet, given that patient characteristics were used in the determination of the anesthetic used, heart team judgment is necessary to appropriately administer MA with CS/TTE in patients undergoing TAVR.

The findings in this study are consistent with data from the literature comparing GA vs MA. In recent meta-analyses comparing TAVR under GA vs MA, no significant difference was found in rates of stroke, major vascular complication, or the need for a permanent pacemaker.^25,26 Interestingly, data from national registries and some meta-analyses showed that the use of CS was associated with lower 30-day mortality, suggesting that CS may be superior to GA.^11,25 This finding is encouraging, although 30-day mortality data remain mixed, with our data and the results from other studies showing no significant difference.²⁶ Nonetheless, given the non-inferiority of TAVR performed under MA and the potential advantages of cost saving, MA should be considered as the primary option for appropriately selected patients undergoing TAVR.

Cases under GA allow use of TEE, while our defined MA uses CS with intraprocedural TTE. Theoretical concerns have existed with the move from GA to MA, including concerns that complications such as annular rupture or cardiac tamponade would be diagnosed late. As we adopted MA, TTE immediately after valve deployment became our protocol and enabled us to quickly evaluate for tamponade or aortic dissection. We found no differences in postprocedural paravalvular leak, similar to literature comparing the use of TTE and TEE.²⁷ Similarly, no difference has been shown for postprocedural aortic regurgitation or tamponade when comparing the use of TEE and angiography with angiography alone.²⁸ Furthermore, our data showed no significant difference in major vascular complications for patients undergoing MA vs GA.

Study limitations. This study was limited by its retrospective and single-center design. The patient population studied is representative of the VA-TAVR program, which is predominantly male, with age affected by service years. Due to inherent differences in baseline characteristics of the GA/TEE vs MA cohorts, propensity-score matching was implemented. Another study limitation is that the learning curves for TAVR and development of MA occurred together, which makes it difficult to tease out the impact of learning curve for TAVR from the effects of MA. Nonetheless, our study suggests that our TAVR program was able to safely transition from GA/TEE to MA, which offers procedural benefits without compromising TAVR outcomes.

Conclusion

As one of the first 5 VA centers approved to perform TAVR, we showed that a federal facility was able to safely transition from GA/TEE to our MA utilizing CS/TTE for TAVR procedures. Furthermore, our results showed reduction in contrast use and fluoroscopy time as the MA program developed. These data are not captured by the STS/ACC registry and demonstrate transparency of TAVR outcomes at federal facilities. Our results suggest that the use of MA for TAVR can be extended to other federal facilities, with the possibility of reducing cost without compromising quality of care. These findings become increasingly important as TAVR becomes more widely used to treat aortic stenosis.

From the ¹Division of Cardiothoracic Surgery; ²Division of Cardiology; and ³Department of Anesthesia, University of California San Francisco and San Francisco Veterans Affairs (VA) Medical Center, San Francisco, California.

Presented at the Association of VA Surgeons 2018 Annual Meeting in Miami Beach, Florida.

Funding: Funded by NIH R01HL119857-01A1 and UCSF-CTSI Grant Number UL1 TR001872.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Ge reports grants from the National Institute of Health, and is the founder of ReValveMed. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript accepted June 26, 2020.

Address for correspondence: Elaine E. Tseng, MD, Professor of Surgery, University of California San Francisco Medical Center, Chief of Cardiothoracic Surgery, San Francisco VA Medical Center,

4150 Clement St. 112D, San Francisco, CA 94121. Email: Elaine.Tseng@ucsf.edu

1. Leon MB, Smith CR, Mack MJ, et al. Transcatheter or surgical aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2016;374:1609-1620.
2. Fröhlich GM, Lansky AJ, Webb J, et al. Local versus general anesthesia for transcatheter aortic valve implantation (TAVR) — systematic review and meta-analysis. BMC Med. 2014;12:41.
3. Mack MJ, Leon MB, Thourani VH, et al. Transcatheter aortic-valve replacement with a balloon-expandable valve in low-risk patients. N Engl J Med. 2019;380:1695-1705. Epub 2019 Mar 16.
4. Popma JJ, Deeb GM, Yakubov SJ, et al. Transcatheter aortic-valve replacement with a self-expanding valve in low-risk patients. N Engl J Med. 2019;380:1706-1715. Epub 2019 Mar 16.
5. Thourani VH, Suri RM, Gunter RL, et al. Contemporary real-world outcomes of surgical aortic valve replacement in 141,905 low-risk, intermediate-risk, and high-risk patients. Ann Thorac Surg. 2015;99:55-61.
6. Hosoba S, Yamamoto M, Shioda K, et al. Safety and efficacy of minimalist approach in transfemoral transcatheter aortic valve replacement: insights from the optimized transcatheter valvular intervention-transcatheter aortic valve implantation (OCEAN-TAVI) registry. Interact Cardiovasc Thorac Surg. 2018;26:420-424.
7. Jensen HA, Condado JF, Devireddy C, et al. Minimalist transcatheter aortic valve replacement: the new standard for surgeons and cardiologists using transfemoral access? Read at the 95th Annual Meeting of the American Association for Thoracic Surgery, Seattle, Washington, April 25-29, 2015. J Thorac Cardiovasc Surg. 2015;150:833-840.
8. Wood DA, Lauck SB, Cairns JA, et al. The Vancouver 3M (multidisciplinary, multimodality, but minimalist) clinical pathway facilitates safe next-day discharge home at low-, medium-, and high-volume transfemoral transcatheter aortic valve replacement centers: the 3M TAVR study. JACC Cardiovasc Interv. 2019;12:459-469.
9. Toppen W, Johansen D, Sareh S, et al. Improved costs and outcomes with conscious sedation vs general anesthesia in TAVR patients: time to wake up? PLoS One. 2017;12:e0173777.
10. Mayr NP, Michel J, Bleiziffer S, Tassani P, Martin K. Sedation or general anesthesia for transcatheter aortic valve implantation (TAVI). J Thorac Dis. 2015;7:1518-1526.
11. Hyman MC, Vemulapalli S, Szeto WY, et al. Conscious sedation versus general anesthesia for transcatheter aortic valve replacement. Circulation. 2017;136:2132-2140. Epub 2017 Sep 1.
12. Hall PS, O’Donnell CI, Mathew V, et al. Outcomes of veterans undergoing TAVR within Veterans Affairs medical centers: insights from the Veterans Affairs Clinical Assessment, Reporting, and Tracking Program. JACC Cardiovasc Interv. 2019;12:2186-2194. Epub 2019 Aug 28.
13. Omer S, Kar B, Cornwell LD, et al. Early experience of a transcatheter aortic valve program at a Veterans Affairs facility. JAMA Surg. 2013;148:1087.
14. Liang NE, Wisneski AD, Wang SX, et al. Veterans Affairs heart team experience with transcatheter aortic valve replacement and minimally invasive surgical aortic valve replacement. J Invasive Cardiol. 2019;31:217-222.
15. Adult Cardiac Surgery Database Data Collection. The Society of Thoracic Surgeons. Available at https://www.sts.org/registries-research-center/sts-national-database/adult-cardiac-surgery-database/data-collection. Accessed January 11, 2021.
16. Blackstone EH, Suri RM, Rajeswaran J, et al. Propensity-matched comparisons of clinical outcomes after transapical or transfemoral transcatheter aortic valve replacement. Circulation. 2015;131:1989-2000.
17. D’Errigo P, Ranucci M, Covello RD, et al. Outcome after general anesthesia versus monitored anesthesia care in transfemoral transcatheter aortic valve replacement. J Cardiothorac Vasc Anesth. 2016;30:1238-1243.
18. Balan P, Zhao Y, Johnson S, et al. The Society of Thoracic Surgery risk score as a predictor of 30-day mortality in transcatheter vs surgical aortic valve replacement: a single-center experience and its implications for the development of a TAVR risk-prediction model. J Invasive Cardiol. 2017;29:109-114.
19. Kappetein AP, Head SJ, Genereux P, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document (VARC-2). Eur J Cardio-Thoracic Surg. 2012;42:S45-S60.
20. Gurevich S, Oestreich B, Kelly RF, et al. Outcomes of transcatheter aortic valve replacement using a minimalist approach. Cardiovasc Revasc Med. 2018;19:192-195.
21. Lauck SB, Wood DA, Baumbusch J, et al. Vancouver transcatheter aortic valve replacement clinical pathway. Circ Cardiovasc Qual Outcomes. 2016;9:312-321.
22. Kotronias RA, Teitelbaum M, Webb JG, et al. Early versus standard discharge after transcatheter aortic valve replacement: a systematic review and meta-analysis. JACC Cardiovasc Interv. 2018;11:1759-1771.
23. Kolte D, Khera S, Sardar MR, et al. Thirty-day readmissions after transcatheter aortic valve replacement in the United States: insights from the Nationwide Readmissions database. Circ Cardiovasc Interv. 2017;10: e004472.
24. Marino M, Lilie CJ, Culp WC, Schepel SR, Tippett JC. The evolution of echocardiographic type and anesthetic technique for transcatheter aortic valve replacement at a high-volume transcatheter aortic valve replacement center. J Cardiothorac Vasc Anesth. 2019;33:29-35.
25. Villablanca PA, Mohananey D, Nikolic K, et al. Comparison of local versus general anesthesia in patients undergoing transcatheter aortic valve replacement: a meta-analysis. Catheter Cardiovasc Interv. 2018;91:330-342.
26. Panhwar MS, Li J, Zidar DA, et al. Outcomes among patients with heart failure with reduced ejection fraction undergoing transcatheter aortic valve replacement: minimally invasive strategy versus conventional strategy. J Invasive Cardiol. 2019;31:15-20.
27. Hayek SS, Corrigan FE, Condado JF, et al. Paravalvular regurgitation after transcatheter aortic valve replacement: comparing transthoracic versus transesophageal echocardiographic guidance. J Am Soc Echocardiogr. 2017;30:533-540.
28. Attizzani GF, Ohno Y, Latib A, et al. Transcatheter aortic valve implantation under angiographic guidance with and without adjunctive transesophageal echocardiography. Am J Cardiol. 2015;116:604-611.