Author + information
- Received January 22, 2018
- Revision received March 8, 2018
- Accepted March 13, 2018
- Published online June 18, 2018.
- Janarthanan Sathananthan, MBChB, MPHa,
- Dale J. Murdoch, MBBSa,
- Brian R. Lindman, MD, MScib,
- Alan Zajarias, MDc,
- Wael A. Jaber, MDd,
- Paul Cremer, MDd,
- David Wood, MDa,
- Robert Moss, MDa,
- Anson Cheung, MDa,
- Jian Ye, MDa,
- Rebecca T. Hahn, MDe,
- Aaron Crowley, MAf,
- Martin B. Leon, MDe,f,
- Michael J. Mack, MDg and
- John G. Webb, MDa,∗ ()
- aCentre for Heart Valve Innovation, St. Paul’s Hospital, Vancouver, University of British Columbia, Vancouver, British Columbia, Canada
- bVanderbilt University Medical Center, Nashville, Tennessee
- cWashington University, St. Louis, Missouri
- dCleveland Clinic, Cleveland, Ohio
- eColumbia University Medical Center, New York, New York
- fCardiovascular Research Foundation, New York, New York
- gBaylor Scott & White Health, Plano, Texas
- ↵∗Address for correspondence:
Dr. John Webb, St. Paul’s Hospital, 1081 Burrard Street, Vancouver, British Columbia V6Z 1Y6, Canada.
Objectives The aim of this study was to assess the implications of concomitant tricuspid regurgitation (TR) in patients undergoing valve-in-valve (VIV) transcatheter aortic valve replacement.
Background Patients undergoing VIV transcatheter aortic valve replacement with concomitant TR may have worse outcomes, and optimal management remains undetermined.
Methods The multicenter PARTNER 2 (Placement of Aortic Transcatheter Valves) VIV trial enrolled patients with symptomatic degenerated surgical aortic bioprostheses who were at high risk for reoperation. Outcomes were assessed between patients with mild or no TR versus moderate or severe TR.
Results A total of 237 patients underwent VIV procedures (mean age 78.7 ± 10.8 years, mean Society of Thoracic Surgeons score 9.1 ± 4.8%). In this cohort, 162 patients (68.4%) had mild or no TR, and 75 patients (31.6%) had moderate or severe TR. Although there was no difference in New York Heart Association functional class III or IV symptomatic status (89.3% vs. 91.4%; p = 0.62) or moderate or severe right ventricular dysfunction (9.4% vs. 16.9%; p = 0.11), patients with moderate or severe TR were more likely to be at high surgical risk, with a Society of Thoracic Surgeons score of >8 (62.7% vs 46.9%; p = 0.02). There was no difference in a composite endpoint of death and rehospitalization between moderate or severe TR and mild or no TR, either at 30 days (10.7% vs. 9.9%; p = 0.85) or at 1-year follow-up (24.1% vs. 23.2%; p = 0.80). There was a significant reduction in overall moderate or severe TR from baseline at 30 days (31.1% vs. 21.1%; p = 0.002), which was sustained at 1-year follow-up (38.0% vs. 22.8%; p = 0.004).
Conclusions Despite higher predicted surgical risk, the presence of TR was not a predictor of long-term outcomes. Importantly, there was significant reduction in TR severity at both short- and long-term follow-up. In selected patients undergoing VIV transcatheter aortic valve replacement, it may be appropriate to conservatively manage concomitant TR.
Selected patients with bioprosthetic aortic valve failure who are at high surgical risk may be candidates for aortic valve-in-valve (VIV) transcatheter aortic valve replacement (TAVR) with a lower risk for morbidity or mortality (1–3). The prevalence of coexisting significant tricuspid regurgitation (TR) in aortic VIV TAVR patients is unclear. Limited studies have shown that in patients undergoing native valve TAVR, the prevalence of coexisting moderate to severe TR is between 11% and 26% (4–6). When patients do have concomitant TR, a reduction in echocardiographic TR grade has been reported in 15% to 50% following native valve TAVR (5,7).
In high-risk patients undergoing native aortic valve TAVR, moderate or severe TR may be associated with increased mortality at long-term follow-up, but reports are conflicting (5,6,8). In patients undergoing surgical aortic valve replacement, there are also conflicting reports of the change in functional TR grade at follow-up and impact on outcomes (9–12). The impact of coexisting TR in patients undergoing aortic VIV TAVR for failed bioprosthetic valves has not been described. The role of concomitant tricuspid valve interventions to improve clinical benefit is also undetermined. We sought to assess the implications of concomitant TR in patients undergoing VIV TAVR.
The PARTNER 2 (Placement of Aortic Transcatheter Valves) VIV registry was a multicenter study that prospectively enrolled patients with symptomatic degenerated aortic bioprosthetic surgical valves. An initial nested registry included 96 patients who were at high risk for reoperation and an additional 269 patients were subsequently enrolled in a continued access registry.
Patients were assessed by a heart team and considered at high risk if the risk for surgical mortality or major morbidity for reoperation exceeded 50%. Surgical bioprosthetic valve degeneration was defined as severe aortic stenosis with an effective orifice area of <0.8 cm2 or an indexed effective orifice area of <0.5 cm2/m2 and a mean gradient of >40 mm Hg or peak velocity of >4 m/s, or severe aortic regurgitation. Mixed bioprosthetic failure was defined as at least moderate stenosis and regurgitation. Patients with concomitant mitral and TR, irrespective of severity, were included. All patients needed to be in New York Heart Association functional class II or greater. Key exclusion criteria included greater than mild paravalvular regurgitation, bioprosthetic valve labeled size <21 mm, left ventricular ejection fraction <20%, or expected life expectancy <2 years. Complete inclusion and exclusion criteria are outlined in Online Table 1.
Patients were required to have anatomy suitable for aortic VIV TAVR with either a 23- or 26-mm SAPIEN XT transcatheter heart valve (Edwards Lifesciences, Irvine, California). The 20- and 29-mm SAPIEN XT valves were not available. This transcatheter heart valve incorporated a balloon-expandable cobalt chromium frame with bovine pericardial leaflets and a polyethylene terephthalate sealing cuff that covered the lower inner portion of the frame. Post-procedure, 6 months of treatment with aspirin and clopidogrel was recommended.
Patients underwent serial echocardiography at baseline, within 24 h of discharge, at 30 days, and at 1-year follow-up. All echocardiograms were independently assessed by a core laboratory. TR severity was determined as recommended by the American Society of Echocardiography. This included assessment of tricuspid valve morphology, right atrial and right ventricular (RV) size, inferior vena cava size, jet area, vena contracta width, jet density and contour, and hepatic vein flow (13). TR severity was reported for each echocardiogram using a 5-tier reporting model (none, trivial, mild, moderate, or severe TR). There were 2 consensus readers for all TR and RV function assessments to avoid interobserver variability, with a third person to break any ties. The American Society of Echocardiography guidelines support qualitative reporting of RV size and function.
Definitions and endpoints
The primary endpoint was a composite of all-cause mortality and rehospitalization at both 30 days and 1-year follow-up. Secondary endpoints were individual components of the primary endpoint, cardiovascular death, and reduction in TR severity at 30-day and 1-year follow-up. Additional endpoints were the predictive value of baseline moderate or severe RV dysfunction or mechanism of bioprosthetic valve failure on TR improvement at 30 days.
Endpoints are defined as per the Valve Academic Research Consortium-2 criteria (14). Clinical events were adjudicated by a clinical event committee, and all adverse events were reviewed by a data and safety monitoring board. Pulmonary hypertension was collected only at baseline in the operability risk assessment from each participating center.
Categorical variables are reported as percentages and were compared using the chi-square or Fisher exact test, as appropriate. Continuous variables are reported as mean ± SD and were compared using the Student's t-test. Kaplan-Meier failure estimates reflect time-to-event outcomes and were compared using the log-rank test. Adjusted comparisons were conducted using a multivariate Cox proportional hazards regression model. The association between baseline TR and outcomes was adjusted for the following covariate set: sex, Society of Thoracic Surgeons (STS) score, pulmonary hypertension, atrial fibrillation, permanent pacemaker, and moderate to severe mitral regurgitation at baseline. A logistic regression model was used to assess the effect of mechanism of bioprosthetic valve failure and baseline RV dysfunction on TR improvement at 30-day follow-up. All tests were 2 sided, and p values <0.05 were considered to indicate statistical significance. Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, North Carolina).
A total of 365 patients (96 in the initial registry, 269 in the continued access registry) underwent aortic VIV procedures across 46 centers between July 12, 2012, and October 8, 2015. Baseline TR was assessed in 237 patients (64.9%) (90 in the initial registry, 147 in the continued access registry), of whom 162 (68.4%) had mild or no TR, and 75 (31.6%) had moderate or severe TR. The numbers of patients in this cohort with moderate and severe TR were 61 (25.7%) and 14 (5.9%), respectively. Differences in baseline characteristics between the analysis population and those excluded due to missing TR assessment are detailed in Online Table 2.
The cohort had a mean age of 78.7 ± 10.8 years and a mean STS score of 9.1 ± 4.8%. Patients with moderate or severe TR were more likely to be women (48.0% vs. 33.3%; p = 0.03) and to have severe pulmonary hypertension (28% vs. 8%; p < 0.0001), atrial fibrillation (66.7% vs. 38.3%; p < 0.0001), or a permanent pacemaker (40% vs. 17.9%; p = 0.0003) or to be at high surgical risk, with an STS score of >8 (62.7% vs. 46.9%; p = 0.02) and higher logistic European System for Cardiac Operative Risk Evaluation score (14.8 vs. 11.6; p = 0.04). They were less likely to have prior myocardial infarction (6.7% vs. 16.7%; p = 0.04), prior stroke (6.7% vs. 17.3%; p = 0.03), and significant renal impairment (4% vs. 12.3%; p = 0.04). There was no difference in New York Heart Association functional class III or IV symptomatic status (89.3% vs. 91.4%; p = 0.62) (Table 1).
Patients with moderate or severe TR were more likely to have moderate or severe mitral regurgitation (56.2% vs. 25.9%; p < 0.0001), larger left atrial volume (51.4 vs. 44.3 ml/m2; p = 0.002), larger right atrial area (23.9 vs. 19.3 cm2; p < 0.0001), and higher RV systolic pressure (48.6 vs. 38.8 mm Hg; p < 0.0001) (Table 2).
There was no difference in a composite endpoint of all-cause death and rehospitalization between moderate or severe TR and mild or no TR at baseline, at 30 days (10.7% vs. 9.9%; p = 0.85), or at 1-year follow-up (24.1% vs. 23.2%; p = 0.80). At 1-year follow-up, there was also no difference between moderate or severe TR and mild or no TR at baseline, for individual endpoints of all-cause death (12.0% vs. 13.2%; p = 0.85), cardiovascular death (9.4% vs. 10.1%; p = 0.91), or rehospitalization (13.7% vs. 13.1%; p = 0.83) (Figure 1).
Multivariate analysis adjusting for sex, STS score, hypertension, atrial fibrillation, moderate or severe mitral regurgitation, and permanent pacemaker still did not demonstrate any significant difference for a composite endpoint of all-cause mortality and rehospitalization between mild or no TR and moderate or severe TR at baseline, at 30 days (hazard ratio: 0.83; 95% confidence interval [CI]: 0.32 to 2.16; p = 0.70), or at 1-year follow-up (hazard ratio: 0.75; 95% CI: 0.40 to 1.40; p = 0.36). Neither RV dysfunction nor RV size was found to be associated with the composite of death and rehospitalization in univariate analyses and thus neither was included in the final model. There was no interaction of mitral regurgitation with TR for the primary endpoint of all-cause death and rehospitalization (p = 0.30). There was no interaction of atrial fibrillation with TR for the primary endpoint of all-cause death and rehospitalization (p = 0.26). There was no significant difference in follow-up between patients with or without significant TR (median follow-up days 716 [interquartile range: 434 to 856] vs. 679 [interquartile range: 364 to 771], respectively; p = 0.10).
There was a significant reduction in overall moderate or severe TR from baseline at 30 days (31.1% vs. 21.1%; p = 0.002), which was sustained at 1-year follow-up (38.0% vs. 22.8%; p = 0.004) (Figure 2).
There was no difference in baseline demographics between patients with TR reduction and persistent TR at 30 days. Apart from higher left ventricular end-diastolic diameter (5.2 vs. 4.8 cm; p = 0.04), there were no differences in baseline echocardiographic variables in patients with TR reduction compared with those with persistent TR at 30 days. At 1-year follow-up, patients with persistent TR at 30 days were more likely to have moderate or severe RV dilatation (36.4% vs. 4.8%; p = 0.01), with no difference seen in RV function (19% vs. 4.8%; p = 0.15).
The mechanism of bioprosthetic aortic valve failure was not predictive of TR improvement at 30 days for either regurgitation (odds ratio: 0.88; 95% CI: 0.23 to 3.34; p = 0.92) or mixed failure (odds ratio: 0.88; 95% CI: 0.25 to 3.11; p = 0.92) compared with stenosis alone. Baseline moderate to severe RV dysfunction was also not predictive of TR improvement at 30 days (odds ratio: 0.92; 95% CI: 0.21 to 4.11; p = 0.91).
This study has shown that in patients undergoing aortic VIV TAVR, the presence of significant concomitant TR was not associated with 1-year mortality or rehospitalization. There was also a significant reduction in TR severity at early follow-up, which was sustained at 1 year. This study addresses a knowledge gap and aids in management decisions in patients with concomitant TR undergoing aortic VIV TAVR.
Identification of patients who will have TR reduction following aortic VIV TAVR is integral to guide the need for additional intervention. The mechanism of TR may predict which patients will improve following aortic VIV TAVR. TR is commonly the result of secondary causes such as left-sided heart disease or pulmonary hypertension (15–17). Once aortic stenosis is corrected with TAVR, subsequent improvement in left-sided cardiac function, such as a reduction in mitral regurgitation or left ventricular dysfunction, might subsequently lead to reduction in TR. There is a complex interaction between mitral regurgitation and TR, and it is conceivable that in some cases, untreated mitral regurgitation would prevent subsequent TR reduction (18). In long-standing disease with significant tricuspid annular or RV dilatation, TR reduction is also unlikely. Additionally, primary TR resulting from pacing leads or leaflet degeneration may be unlikely to diminish as a result of correction of aortic stenosis (15,19). Because secondary or functional TR accounts for as much as 90% of all TR, an initial conservative strategy may be appropriate in selected patients, because the majority of patients have the potential to have possible reduction in TR severity (16).
The prognostic impact of persistent significant TR following left sided heart surgery is poorly understood. The presence of TR has been shown to lead to progressive RV dysfunction and possibly worse survival with longer follow-up following surgery (20). Similarly, in our study persistent moderate or severe TR was associated with worsening RV dilatation at 1 year following TAVR. The prognostic impact of TR may also be associated with RV size or function rather than TR alone (11,21). Because there is no clear mortality impact of TR in VIV TAVR patients, the decision to proceed to TR intervention should be driven by factors such as patient symptoms, heart failure, or possibly deterioration in RV size or function. Patients with concomitant TR should be monitored following VIV TAVR to assess TR reduction and clinical status.
In lower risk patients with failing aortic bioprosthetic valves and concomitant native valve disease, redo surgery should be considered. However, outcomes from tricuspid surgery are not ideal (22,23), and many patients with both aortic and tricuspid disease may be at elevated surgical risk and may be better suited to a percutaneous approach. Given the improvement in TR following VIV therapy and the lack of adverse midterm outcomes in this study, it may be reasonable to treat aortic bioprosthetic valve failure first. Where TR remains significant or worsens in the setting of symptoms of right heart failure, staged additional intervention can be considered. Percutaneous options for the tricuspid valve continue to be developed and, although promising, remain unproved (24–26).
Only patients with high surgical risk were included, and comorbidities may be the primary determinants of outcomes. Among patients with moderate or severe TR, only a small proportion of patients had severe TR, with the majority having moderate TR. The low rate of patients with severe TR may reflect clinical case selection. Patients with bioprosthetic aortic valve failure and concomitant severe TR may have been treated surgically or alternatively considered too high risk and managed medically. These factors potentially limit the applicability of this study to patients with severe TR. This was a registry study and hence subject to limitations compared with a randomized controlled trial. Future studies with larger patient numbers may also help identify any potential adverse associations of significant TR that may be limited in this study because of insufficient power.
Further study is needed to understand the prognostic impact of severe TR alone on outcomes. Additionally, follow-up beyond 1 year may be required to identify any adverse effects of moderate or severe TR. In our series, the mechanism of TR was not known, which poses challenges for identifying which patients will improve following aortic VIV TAVR. The assessment of TR severity was made on the basis of primarily qualitative and semiqualitative parameters. Quantitative grading of TR severity could permit a more accurate assessment of outcomes (27). Additional RV assessment with techniques such as echocardiographic strain assessment or magnetic resonance imaging might provide useful information about RV size and function (28).
Baseline significant TR was not a predictor of outcomes 1 year following aortic VIV TAVR. Importantly, there was significant reduction in TR severity at both short- and long-term follow-up. In selected patients undergoing VIV TAVR, it may be appropriate to conservatively manage concomitant TR.
WHAT IS KNOWN? VIV treatment with TAVR is a viable option for patients with failing aortic bioprosthetic valves. However, the significance and optimal management of concomitant significant TR in these patients is unknown.
WHAT IS NEW? Despite being at higher risk, the presence of TR in patients undergoing VIV TAVR was not a predictor of long-term outcomes. Importantly, there was significant reduction in TR severity at both short- and long-term follow-up.
WHAT IS NEXT? Follow-up beyond 1 year is needed to understand the long-term prognostic implications of significant TR. An understanding of the mechanism of TR and impact on subsequent improvement following VIV TAVR is also needed in future studies.
The PARTNER 2 trial was funded by Edwards Lifesciences. Dr. Sathananthan was supported by a grant from the National Heart Foundation of New Zealand. Dr. Lindman has received grant funding from Edwards Lifesciences and Roche Diagnostics; and consulting fees from Medtronic. Dr. Zajarias has received consulting fees from Edwards Lifesciences. Drs. Jaber and Hahn have echocardiographic core laboratory contracts with Edwards Lifesciences (no direct compensation). Dr. Wood is a consultant to Edwards Lifesciences. Drs. Leon, Mack, and Webb are members of the PARTNER Trial Executive Committee (no direct compensation). Dr. Webb is a consultant for Edwards Lifesciences. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- confidence interval
- right ventricular
- Society of Thoracic Surgeons
- transcatheter aortic valve replacement
- tricuspid regurgitation
- Received January 22, 2018.
- Revision received March 8, 2018.
- Accepted March 13, 2018.
- 2018 American College of Cardiology Foundation
- Webb J.G.,
- Mack M.J.,
- White J.M.,
- et al.
- Dvir D.,
- Webb J.,
- Brecker S.,
- et al.
- ↵Kornowski R, Tchetche D, Verhoye J-P, Chevalier B. TAVI for failed surgical aortic bioprostheses using a self-expanding device: early results from the prospective VIVA post-market study. Presented at: EuroPCR; Paris, France; 2017.
- Schwartz L.A.,
- Rozenbaum Z.,
- Ghantous E.,
- et al.
- Lindman B.R.,
- Maniar H.S.,
- Jaber W.A.,
- et al.
- Pavasini R.,
- Ruggerini S.,
- Grapsa J.,
- et al.
- Taramasso M.,
- Maisano F.,
- De Bonis M.,
- et al.
- Mascherbauer J.,
- Kammerlander A.A.,
- Marzluf B.A.,
- Graf A.,
- Kocher A.,
- Bonderman D.
- Zoghbi W.A.,
- Enriquez-Sarano M.,
- Foster E.,
- et al.
- Cheng R.,
- Azarbal A.,
- Currier J.,
- et al.
- Praz F.,
- Windecker S.
- Mangieri A.,
- Montalto C.,
- Pagnesi M.,
- et al.
- Ito S.,
- Pislaru S.V.,
- Soo W.M.,
- et al.
- Ejiofor J.I.,
- Neely R.C.,
- Yammine M.,
- et al.
- Nickenig G.,
- Kowalski M.,
- Hausleiter J.,
- et al.
- Taramasso M.,
- Pozzoli A.,
- Guidotti A.,
- et al.
- Hahn R.T.,
- Meduri C.U.,
- Davidson C.J.,
- et al.
- Hahn R.T.
- Cavalcante J.L.,
- Simon M.A.,
- Chan S.Y.