Author + information
- Received April 17, 2017
- Revision received May 11, 2017
- Accepted May 16, 2017
- Published online August 7, 2017.
- John Mooney, MDa,
- Stephanie L. Sellers, MSca,
- Phillip Blanke, MDa,
- Philippe Pibarot, DVM, PhDb,
- Rebecca T. Hahn, MDc,
- Danny Dvir, MDd,
- Pamela S. Douglas, MDe,
- Neil J. Weissman, MDf,
- Susheel K. Kodali, MDc,
- Vinod H. Thourani, MDg,
- Hasan Jilaihawi, MDh,
- Omar Khalique, MDc,
- Craig R. Smith, MDc,
- Shaw Hua Kueh, MDa,
- Mickael Ohana, MDa,
- Romi Grover, MDa,
- Christopher Naoum, MDa,
- Aaron Crowley, MSi,
- Wael A. Jaber, MDj,
- Maria C. Alu, MSc,
- Rupa Parvataneni, MSi,
- Michael Mack, MDk,
- John G. Webb, MDa,
- Martin B. Leon, MDc and
- Jonathon A. Leipsic, MDa,∗ ()
- aSt. Paul’s Hospital, Vancouver, British Columbia, Canada
- bUniversité Laval, Quebec City, Quebec, Canada
- cColumbia University Medical Center, New York, New York
- dUniversity of Washington, Seattle, Washington
- eDuke Clinical Research Institute, Duke University Medical Center, Durham, North Carolina
- fMedstar Health Research Institute, Georgetown University School of Medicine, Washington, DC
- gEmory University School of Medicine, Atlanta, Georgia
- hNYU Langone Medical Center, New York, New York
- iCardiovascular Research Foundation, New York, New York
- jCleveland Clinic, Cleveland, Ohio
- kBaylor Scott and White Health, Plano, Texas
- ↵∗Address for correspondence:
Dr. Jonathon A. Leipsic, Department of Radiology, UBC, Centre for Heart Valve Innovation–St Paul’s Hospital, 1081 Burrard Street, Vancouver V6Z 1Y6, Canada.
Objectives This study sought to determine if indexed effective orifice area (EOAi), using left ventricular outflow tract measured from computed tomography (EOAiCT), reclassified prosthesis–patient mismatch (PPM) compared with conventional echocardiogram-defined measurements (EOAiTTE).
Background PPM does not predict mortality following transcatheter aortic valve replacement (TAVR). However, it is unknown if the EOAiCT of the left ventricular outflow tract improves risk stratification.
Methods A total of 765 TAVR patients from the PARTNER II (Placement of Aortic Transcatheter Valves II) trial S3i cohort were evaluated. EOAi was calculated using the continuity equation, and the left ventricular outflow tract area was derived from baseline computed tomography. Traditional echocardiographic categories defined PPM: absent (>0.85 cm2/m2), moderate (≥0.65 and ≤0.85 cm2/m2), or severe (≤0.65 cm2/m2). Correlation of EOAiCT and EOAiTTE to 1-year outcomes was performed.
Results The incidence of PPM was 24% with EOACT compared with 45% with EOAiTTE. Only 6% of PPM was graded severe by EOAiCT compared with 9% by EOAiTTE. EOAiTTE, but not EOAiCT, defined PPM showed association with reduced left ventricular mass regression (p = 0.03 vs. p = 0.52). There was no association between PPM and death or rehospitalization at 1 year with either modality. EOACT was associated with minor stroke at 1 year (log-rank p = 0.04), and EOAiTTE with stroke/transient ischemic attack (log-rank p = 0.030). Furthermore, when subjects with mild or greater paravalvular regurgitation were excluded, the presence of PPM did not show association with any outcome.
Conclusions EOAiCT downgrades frequency and severity of PPM in patients after TAVR, and was not associated with mortality 1 year after TAVR. EOAiTTE, but not EOAiCT, was associated with less left ventricular mass regression.
Prosthesis-patient mismatch (PPM) can occur after transcatheter aortic valve replacement (TAVR) and is defined as a valve prosthesis too small for body size (1). Classically, PPM has been identified after valve implantation by transthoracic echocardiogram (TTE), and defined as an effective orifice area (EOA) of the prosthesis indexed to body surface area (BSA) of ≤0.85 cm2/m2 (2).
PPM is conventionally defined by TTE-based derivation of the indexed EOA (EOAiTTE) using the continuity equation and relying on the assumption that the cross-sectional area of the left ventricular outflow tract (LVOT) is circular (3). However, the LVOT is more elliptical in cross-sectional area than circular, and assuming a circular LVOT for sizing using TTE has been found to underestimate LVOT area by up to 17% when compared with cardiac computed tomography (CT) (4–7). CT has been used to reclassify low gradient aortic stenosis (8). The implications of this underestimation of LVOT size are amplified by its use in the continuity equation, which squares the derived LVOT area (9). Notably, the presence of a stent post-implant can also potentially limit echo definition of the LVOT.
Although CT can potentially improve the accuracy of EOAi calculations, it is unclear what impact this calculation has on the classification of PPM post-TAVR, and whether EOAiCT is predictive of clinical risk following TAVR. To answer this, we aimed to determine if EOAiCT would reclassify the incidence and severity of PPM, and whether this was able to discriminate risk of all-cause mortality, and secondary analysis of outcomes inclusive of cardiovascular death, rehospitalization, stroke, and left ventricular (LV) mass regression.
The cohort consisted of intermediate-risk patients treated with the SAPIEN 3 valve in a nested registry of the PARTNER II (Placement of Aortic Transcatheter Valves II) trial (10). The trial’s design and inclusion and exclusion criteria have been reported separately. For this study, we included only patients with baseline systolic phase contrast-enhanced CT and TTE as well as post-implant discharge or 30-day TTE imaging with 1-year follow up.
Patients were considered to have an intermediate risk of 30-day mortality with a Society of Thoracic Surgeons (STS) score of 4 to 8, with the score equaling the percentage predicted 30-day mortality. Patients with an STS score <4 could also be enrolled if coexisting comorbidities were present that were not reflected in the STS model.
Transcatheter heart valve procedure
Vascular access for each patient was determined on vessel size to accommodate vascular sheaths. If transfemoral access was contraindicated, the transthoracic route was used. All patients received the SAPIEN 3 (Edwards Lifesciences, Irvine, California) balloon expandable valve (11). Patients received aspirin and clopidogrel before the procedure and heparin during the procedure. After the procedure, aspirin was continued indefinitely and clopidogrel for 1 month.
Patients underwent TTE at baseline, discharge, 30 days, and 1 year as evaluated by PARTNER II S3 Core Lab Consortium at Université Laval (Quebec City, Canada), MedStar Health Research Institute (Hyattsville, Maryland), and Cardiovascular Research Foundation (New York, New York). The process of image reproducibility, analysis, and quality assurance has previously been described (12). Stroke volume was measured in the LVOT using diameter and velocity measured just underneath the prosthetic stent (Figure 1A). LV mass was measured using American Society of Echocardiography guidelines for chamber quantification (13). Central, paravalvular, and total aortic regurgitation (AR) was measured using a unified grading scheme (14). AR was classed as none, trace, mild, moderate, or severe.
Patients underwent pre-TAVR CT imaging. CT image analysis was performed using standard methodology. LVOT measurements were taken 2 to 5 mm below the annular plane (Figure 1B). All CT examinations were reviewed and interpreted in a central core laboratory (St Paul’s Hospital, Cardiac CT Laboratory, Vancouver, University of British Columbia, Canada).
Effective orifice area
The TTE-derived valve EOA (EOATTE) was calculated according to the continuity equation (3) using standard Doppler velocity tracings from TTE. This value was then indexed for BSA (EOAiTTE) using the DuBois formula (15), where VTI is velocity time interval:The CT-derived EOA was determined by use of the LVOT cross-sectional area in the formula, represented as A:This value was also indexed for BSA (EOAiCT) using the DuBois formula (15). LVOT dimensions were measured in mid-systole. The VTI in the LVOT and across the aortic prosthesis relied on TTE derived values. The indexed EOA dimension derived from the equation for both imaging modalities was used to define presence or absence of PPM, and the grade of severity.
The EOA was indexed to BSA and derived from the first echocardiogram obtained after implant (either at discharge or 30 days). PPM was defined by EOAi as none (>0.85 cm2/m2), moderate (>0.65 cm2/m2 and ≤0.85 cm2/m2), and severe (≤0.65 cm2/m2) (2).
The primary outcome of interest was death from any cause within 365 days of device implantation. Secondary outcomes included cardiovascular death, stroke/transient ischemic attack (TIA), and LV mass regression within 365 days.
Cardiovascular death was defined as death caused by a cardiovascular event. This included any death due to proximate cardiac disease cause; unwitnessed death and death of unknown cause; all cardiovascular procedure-related deaths including complications of the procedure; and death caused by cerebrovascular disease, pulmonary embolism, and death from peripheral vascular disease.
Stroke was defined as a rapid onset of focal or global neurological deficit with at least 1 of the following: change in level of consciousness, hemiplegia, hemiparesis, numbness or sensory loss affecting 1 side of the body, dysphasia/aphasia, hemianopia, amaurosis fugax, other new neurological symptom or sign consistent with stroke. Duration was ≥24 h or <24 h if therapeutic intervention was performed (thrombolysis/intracranial angioplasty), neurological imaging confirmed a new lesion, or the neurological deficit resulted in death. No other readily identifiable nonstroke cause was found for clinical presentation (e.g., brain tumor, trauma, pharmacological). Confirmation of diagnosis was made with at least 1 of the following: neurology/neurosurgical consultation, neuroimaging by CT, magnetic resonance imaging, or cerebral angiography.
Major stroke was defined as a disabling stroke with modified Rankin score (mRS) ≥2 at 90 days post-stroke (16). This ranged from slight disability and inability to carry out all previous activities to severe disability with full care needs. Minor stroke was defined as mRS <2 indicating no functional impairment and slight or no symptoms (16). TIA was defined as new focal neurological deficit with rapid symptom resolution within 24 h or neurological imaging with no evidence of tissue injury.
Change in LV mass was assessed as a continuous variable, whereby LV mass was determined on TTE at baseline, discharge, 30 days, and 365 days.
Continuous data were presented as mean ± SD, or as median (interquartile range) when distribution was skewed. Categorical data were presented as percentages and fraction of occurrence. Comparisons were based on the chi-square test or Fisher exact test, as appropriate, for categorical variables; Student t test for means for continuous variables; and Wilcoxon rank sum test for medians.
Kaplan-Meier estimates and log-rank test were used to compare incidence of outcomes over 1 year stratified between presence of PPM (no PPM/PPM) and severity of PPM (none, moderate, or severe) using indexed TTE or CT derived area. Subgroup analysis was performed by exclusion of those with at least mild AR, to avoid any potential confounding on outcomes of AR with PPM. Receiver-operating characteristic curves were used to demonstrate the C-statistic for prediction of outcomes. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Inc., Cary, North Carolina) and a p value <0.05 was considered statistically significant.
Baseline characteristics and comparison between patients with and without PPM
A total of 765 patients were included in our analysis. The median age was 83 ± 6.3 with more than one-half being male (63%). Most patients (70.7%) were New York Heart Association class III or IV for symptoms of heart failure, and the average STS score for the cohort was 5.28 ± 1.26. Most cases (87%) were performed by the transfemoral route. Total cohort demographics with relevant clinical and procedural variables are given in Table 1.
The presence or absence of PPM is shown in Table 1 and is based on standard TTE definition (EOAi ≤0.85 cm2/m2 indicating presence). Those with PPM were younger (p = 0.004), more likely to be female (p = 0.021), and have increased BSA and body mass index (both p < 0.001). These were also found significant when presence or absence of PPM was defined by EOAiCT. PPM was also more common among transfemoral patients (p = 0.004 and less common for the transapical approach (p = 0.01). There were no other significant differences between the 2 groups.
At 1-year follow-up there were 54 deaths (7.1%), of which 31 (4.1%) were attributed to cardiovascular causes. A total of 92 (12.3%) patients were rehospitalized, and 44 (5.9%) had a stroke or TIA. The overall composite of patients with all these events was 149 (19.6%).
Comparison of TTE-derived and CT-derived LVOT in grading PPM
After implantation, the median indexed EOA from TTE (EOAiTTE) for the whole cohort was 0.87 cm2/m2 (0.75 to 1.01 cm2/m2). In comparison, the overall median EOA measured by CT (EOAiCT) was 1.02 cm2 (0.86 to 1.23 cm2), which was found to be 15% larger than the TTE-derived area (p < 0.001).
On dichotomization into PPM and no PPM groups, the median values for indexed EOAiTTE were 0.72 cm2/m2 (0.64 to 0.80 cm2/m2) and 0.92 cm2/m2 (0.82 to 1.05 cm2/m2), respectively. By comparison, with EOAiCT, those with PPM had a median value of 0.75 cm2 (0.65 to 0.81cm2) and those with no PPM a median value of 1.10 cm2 (0.97 to 1.31 cm2); (p < 0.001).
The incidence of PPM with EOAiTTE was 45%, although when defined by EOAiCT was only 24% (p < 0.001). When PPM was graded using EOAiTTE, 36% had moderate PPM and 9% severe PPM. In comparison, EOAiCT decreased to 18% for moderate PPM and 6% for severe PPM (Figure 2). Overall, EOAiCT saw a downgrading in both the incidence and severity grading of PPM compared with EOAiTTE as illustrated in Figure 2, which provides a comparison of the severity grading as proportions of the 2 separate imaging modalities.
Impact of PPM on regression of LV hypertrophy at 1 year
LV mass regression was significantly reduced compared with baseline, and this was evident regardless of how the EOAi was defined. The EOAiTTE group with PPM showed significantly less regression of LV mass indexed (median 103.11 g/m2 vs. 98.28 g/m2; p = 0.03) compared with the EOAiTTE group with no PPM. However, when PPM was defined by EOAiCT, there was no difference between those with or without PPM at any time point (median, 98.90 g vs. 100.88 g; p = 0.52). Subgroup analysis was performed using percent change, LV hypertrophy (≥115 g/m2 in men and ≥95 g/m2 in women) and EOAi <1.00 cm2 for EOAiCT defined cases. The optimal cutpoint to measure the absolute change in indexed LV mass for EOAiTTE was 0.88 cm2/m2 and for EOAiCT was 1.02 cm2/m2, again reflecting a 15% size difference. The discriminatory ability for both modalities for LV mass regression was poor (C-statistic: EOAiTTE, 0.52 [95% confidence interval (CI): 0.47 to 0.57]; EOAiCT, 0.54 [95% CI: 0.49 to 0.58]). Figure 3 demonstrates the change in LV mass dichotomized by presence of PPM defined by both imaging modalities.
Impact of PPM on mortality, readmission, and stroke/TIA
EOAiTTE demonstrated no relationship with all-cause mortality at 1 year follow-up on univariate analysis (Table 2). Dichotomized into PPM versus no PPM, the incidence of death was 5.5% and 8.4%, respectively (log-rank p = 0.114). Categorized into no PPM, moderate PPM, and severe PPM the incidences were 8.4%, 5.1%, and 7.4%, respectively (overall log-rank p = 0.230) (Figure 4). With cardiovascular death, there was again no significant difference between absence or presence of PPM (log-rank p = 0.246), or categorization of PPM (overall log-rank p = 0.234). There was also no association between EOAiCT and death from any cause on univariate analysis (Table 2). This included analyses categorized by presence or absence of PPM (8.0% and 4.4%, respectively; log-rank p = 0.098) and graded by severity (8.0% none, 5.0% moderate, and 2.3% of patients with severe PPM deceased by 1 year; overall log-rank p = 0.209) (Figure 5). Cardiovascular-related death was also not associated with PPM either dichotomized (log-rank p = 0.508) or graded by severity (overall log-rank p = 0.738).
No association was found between presence or absence of PPM and readmission to hospital, irrespective of how the EOAi was derived (Table 2). Where EOAiTTE was used to discriminate presence or absence of PPM there was an incidence of 11.2% and 13.2%, respectively (log-rank p = 0.363). Comparable results were found with EOAiCT (12.5% vs. 11.6%; log-rank p = 0.672).
We assessed stroke as a composite outcome with TIA, as a discrete outcome, and subcategorized as either debilitating (mRS ≥2) or nondebilitating (mRS <2). These were compared both according to presence or absence of PPM and by grade of PPM severity.
EOAiTTE was associated with the composite outcome of stroke/TIA, with an incidence of 7.9% among those with PPM compared with 4.2% without (p = 0.030). However, when subcategorized, it was not associated with stroke (p = 0.111), major debilitating stroke (p = 0.726), or minor nondebilitating stroke (p = 0.067). Moreover, when grade of PPM severity was assessed, there was no overall association between grade and any stroke-based outcome.
With EOAiCT there was no association with the composite stroke/TIA outcome (p = 0.386). Subcategorized, there was no association with stroke (p = 0.455) or major stroke (p = 0.266). There was, however, an association with minor stroke (incidence of minor stroke, 4.3% PPM vs. 1.7% no PPM; p = 0.040). Again, when grade of PPM severity was assessed, there was no overall association between grade and any stroke-based outcome.
Predictive utility of CT-derived EOAi to outcomes
Receiver-operating characteristic curves analysis of EOAiCT to predict outcomes was also assessed. For all-cause mortality, the area under the curve (AUC) was 0.56 (95% CI: 0.48 to 0.64) (as shown in Figure 6). Similarly, the C-statistic for cardiovascular death (AUC: 0.52; 95% CI: 0.41 to 0.63) and readmission to hospital (AUC: 0.53; 95% CI: 0.47 to 0.59) also demonstrated poor prediction.
Exclusion of cases with regurgitation
A subanalysis of 388 subjects, excluding all subjects with at least mild AR, either valvular or paravalvular, at discharge or 30-day follow-up was performed. Of these, 174 were considered to have PPM using EOAiTTE, and 96 using EOAiCT. Within this group, no association was found between PPM and outcomes at 1 year, regardless of whether EOAiTTE or EOAiCT was used to define PPM. Where the optimal cutoff for CT was assessed in association to death, the univariate hazard ratio was 1.39 (95% CI: 0.45 to 4.31). As shown in Figure 7, the AUC demonstrated that CT-derived PPM had poor predictive utility at 365 days (C-statistic, 0.53; 95% CI: 0.42 to 0.63).
Our study investigated the use of EOAiCT versus EOAiTTE for evaluation of PPM and yielded several notable findings. First, the frequency and severity of PPM are reduced using EOAiCT compared with EOAiTTE when using standard echocardiographic criteria to define PPM. Second, EOAiCT was associated with minor stroke, but not all-cause mortality, LV mass regression, or rehospitalization at 1 year. EOAiTTE demonstrated association with LV mass regression and stroke, although not with all-cause mortality or rehospitalization. The absence of a meaningful connection between the hybrid calculation of PPM dispels previously held notions that the lack of risk associated with PPM in TAVR simply reflected overestimation of PPM based on the limitations of traditional echocardiographic-based calculation of EOA. Finally, patients with PPM defined both by EOAiCT and EOAiTTE but without residual prosthetic regurgitation were not at increased risk of adverse events at 1-year follow-up.
Overall, EOAiCT defined a median EOA 15% larger than EOAiTTE, which is consistent with the findings of De Vecchi et al. (5) whereby they documented larger aortic EOAi using CT compared with echo in native valvular assessment. As a result, there was a subsequent reduction in the overall incidence of PPM. With EOAiTTE, almost one-half the patients were found to have PPM as described in previous studies (17,18). However, when CT LVOT measures were used for the calculation of EOAi, the incidence was reduced to a quarter of the cohort. The severity of the PPM was also reduced with CT-defined EOAi, with half the cases considered moderate PPM by echo downgraded to no PPM, and a one-third of those considered severe downgraded to moderate. Interestingly, there was a higher prevalence of PPM in patients with larger BSA and body mass index that remained significant when measured by either EOAiCT or EOAiTTE. This was reported previously in the PARTNER I trial (19) and again may reflect an overindexation of PPM severity in obese patients (20).
Neither PPM defined by EOAiCT nor EOAiTTE were associated with increased mortality after TAVR. These negative findings held true whether categorized by severity or as a continuous variable. This is consistent with the high-risk PARTNER I cohort wherein PPM was not found to be associated with increased mortality at 2 years post-TAVR (17). Interestingly, when cases with post-implant AR were excluded from the current analysis, there was no association with worse outcome identified. This contrasts with the higher risk PARTNER I findings where AR was considered to confound the presence of PPM such that exclusion of these patients showed that PPM was associated with worse outcomes, similar to the surgical aortic valve replacement group (17). The lack of impact on mortality may reflect several factors, including that this is a lower risk cohort combined with improved experience in TAVR sizing and implantation.
Interestingly, the association of EOAiTTE and EOAiCT to other secondary outcomes varied. PPM identified by EOAiTTE but not EOAiCT resulted in significantly less LV mass regression using traditional thresholds for PPM. Interestingly, subgroup analysis demonstrated a 15% size difference in the optimal cutoff between EOAiTTE and EOAiCT; however, the discrimination of LV mass regression was poor for both the conventional and hybrid CT approach. Further studies in larger series of patients with longer-term follow-up are necessary to help establish the appropriate thresholds of EOAi that should be used for the CT hybrid approach but more importantly to determine if PPM by either metric meaningfully identifies those less likely to experience LV mass regression. In addition, when analyzed for stroke, EOAiTTE was associated with overall stroke/TIA yet EOAiCT was associated with minor stroke alone. Again, the nature of this discrepancy is unclear and may require longer-term follow up to elucidate.
The lack of incremental benefit of EOACT in the determination of the severity of native aortic valve stenosis has been shown by Clavel et al. (21). In this head-to-head comparison of multidimensional CT and Doppler echocardiography, EOACT was larger than EOATTE but did not improve the correlation with transvalvular gradient, the concordance gradient-aortic valve area, or mortality prediction compared with EOATTE. In fact, their analysis showed that a larger cutpoint value should be used for severe aortic stenosis if EOACT (<1.2 cm2) is measured versus EOATTE (<1.0 cm2). Interestingly, EOACT did not contribute to prediction of risk irrespective of the threshold used. Our data provide similar clarity in the area of PPM and risk prediction highlighting the lack of incremental risk discrimination when PPM is calculated using a hybrid approach integrating CT measures of the LVOT.
First, the study relies on pre-procedural CT alone and does not incorporate post-procedural CT for comparison. This may not allow for comparison of the prosthesis on the LVOT, which may assume a more circular shape post-implantation (22). Post-procedural CT could potentially identify the influence of the prosthesis on geometry and flow; however, this is not clinically practical.
In addition, there is a potential survival bias, because patients in this analysis were only included if post-implant TTE and CT were available. Those with periprocedural or post-procedural mortality would have therefore been excluded. The impact of PPM therefore may not account for in-hospital events, identified previously as significantly impacting on mortality after surgical aortic valve replacement (17,18). To address this bias, post-procedural TTE and CT would be required, which may not be clinically practical. In addition, follow-up was limited to 1 year. Within surgical aortic valve replacement cohorts, longer-term follow-up has demonstrated that severe PPM was associated with adverse prognosis (18,22–24). Comparison with surgical cohorts needs to account for the fact that PPM has not been defined with CT after surgery, and frequency, severity, and relationship to outcomes could also be changed in these groups. Finally, EOAiCT requires TTE-derived LVOT and transaortic velocities to determine EOAi. Unlike the LVOT, these velocities are less prone to interuser variability, although they can be influenced by turbulent flow in the LVOT, hence affecting accuracy (3).
The use of EOAiCT results in reclassification of the prevalence and severity of PPM measurements, but does not provide any incremental value for the prediction of hard cardiovascular events post-TAVR than the traditional TTE-based measurements of PPM.
WHAT IS KNOWN? PPM is identified frequently after TAVR, although is not associated with increased mortality. The continuity equation used to define PPM assumes the LVOT is a circular shape, which potentially leads to underestimation of the LVOT area and consequent overestimation of the incidence and severity of PPM.
WHAT IS NEW? The EOA derived from a hybrid approach is approximately 15% larger than exclusively TTE-derived EOA. Although the cutpoint for PPM and the frequency of severity are consequently shifted, there was no association with mortality at 1 year. PPM defined from TTE was associated with reduced LV mass regression at 1 year.
WHAT IS NEXT? Longer term follow-up of this cohort for outcome assessment may provide insight into the significance of PPM defined by imaging modality.
The PARTNER 2 Trial was funded by Edwards Lifesciences and the protocol was developed collaboratively by the Sponsor and Executive Steering Committee. Dr. Blanke is a consultant for Edwards Lifesciences; and provides CT Core Lab services for Edwards Lifesciences, Medtronic, Neovasc, Guided Delivery System, and Tendyne Holdings for which he receives no direct compensation. Dr. Pibarot has Core Lab contracts with Edwards Lifesciences and Medtronic for which he receives no direct compensation; and is a speaker for St. Jude Medical. Dr. Hahn has Core Lab contracts with Edwards Lifesciences for which she receives no direct compensation; and is a consultant for Philips Healthcare, St. Jude Medical, and Boston Scientific. Dr. Dvir is a consultant for Edwards Lifesciences. Dr. Douglas has Core Lab contracts with Edwards Lifesciences for which she receives no direct compensation. Dr. Weissman has Core Lab contracts with Edwards Lifesciences, St. Jude Medical, Boston Scientific, Medtronic, Biostable, Sorin, Abbott Vascular, Direct Flow, and Mitralign for which he receives no direct compensation. Dr. Kodali is a consultant for Claret Medical, Edwards Lifesciences, and Medtronic; and holds equity in Dura Biotech and Thubrikar Aortic Valve, Inc. Dr. Thourani is a consultant for Edwards Lifesciences, Abbott, Sorin Medical, St. Jude Medical, and DirectFlow. Dr. Jilaihawi is a consultant for Edwards Lifesciences; and has an institutional research relationship with Medtronic and St. Jude Medical. Dr. Khalique is a consultant for Edwards Lifesciences and Boston Scientific; and is a reader for a Core Lab, which has contracts with Edwards Lifesciences for which he receives no direct compensation. Dr. Smith is a member of the PARTNER Trial Executive Committee for which he receives no direct compensation. Dr. Jaber has Core Lab contracts with Edwards Lifesciences for which he receives no direct compensation. Dr. Alu is a consultant for Claret Medical. Ms. Parvataneni is an employee of Jazz Pharmaceuticals; Dr. Mack is a member of the PARTNER Trial Executive Committee for which he receives no direct compensation. Dr. Webb is a member of the PARTNER Trial Executive Committee for which he receives no direct compensation; and a consultant for Edwards Lifesciences. Dr. Leon is a member of the PARTNER Trial Executive Committee for which he receives no direct compensation. Dr. Leipsic is a consultant for Edwards Lifesciences; and provides CT Core Lab services for Edwards Lifesciences, Medtronic, Neovasc, GDS, and Tendyne Holdings for which he receives no direct compensation. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- aortic regurgitation
- area under the curve
- body surface area
- confidence interval
- computed tomography
- effective orifice area indexed
- left ventricular
- left ventricular outflow tract
- modified Rankin score
- prosthesis–patient mismatch
- Society of Thoracic Surgeons
- transcatheter aortic valve replacement
- transient ischemic attack
- transthoracic echocardiogram
- velocity time interval
- Received April 17, 2017.
- Revision received May 11, 2017.
- Accepted May 16, 2017.
- 2017 American College of Cardiology Foundation
- Rahimtoola S.H.
- Pibarot P.,
- Dumesnil J.G.
- O'Brien B.,
- Schoenhagen P.,
- Kapadia S.R.,
- et al.
- Saikrishnan N.,
- Kumar G.,
- Sawaya F.J.,
- Lerakis S.,
- Yoganathan A.P.
- Thourani V.H.,
- Kodali S.,
- Makkar R.R.,
- et al.
- Webb J.G.,
- Doshi D.,
- Mack M.J.,
- et al.
- Lang R.M.,
- Bierig M.,
- Devereux R.B.,
- et al.
- Pibarot P.,
- Hahn R.T.,
- Weissman N.J.,
- Monaghan M.J.
- Kappetein A.P.,
- Head S.J.,
- Genereux P.,
- et al.
- Pibarot P.,
- Weissman N.J.,
- Stewart W.J.,
- et al.
- Head S.J.,
- Mokhles M.M.,
- Osnabrugge R.L.,
- et al.
- Mohty D.,
- Dumesnil J.G.,
- Echahidi N.,
- et al.
- Clavel M.A.,
- Malouf J.,
- Messika-Zeitoun D.,
- Araoz P.A.,
- Michelena H.I.,
- Enriquez-Sarano M.
- Dayan V.,
- Soca G.,
- Stanham R.,
- Lorenzo A.,
- Ferreiro A.