Circulation: Cardiovascular Interventions

ORIGINAL ARTICLE

Contemporary Costs Associated With

Transcatheter Versus Surgical Aortic Valve
Replacement in Medicare Beneficiaries
Suzanne J. Baron , MD, MSc; Michael P. Ryan, MS; Kimberly A. Moore, MS, PharmD; Seth J. Clancy, MPH;
Candace L. Gunnarsson , EdD

BACKGROUND: In patients with severe aortic stenosis, treatment with transcatheter aortic valve replacement (TAVR) has been
shown to be cost-effective in the high-risk surgical population and cost-saving in the intermediate-risk population when
compared with surgical aortic valve replacement (SAVR) in early pivotal clinical trials. Whether TAVR is associated with
comparable or lower costs when compared with SAVR in contemporary clinical practice is unknown.

METHODS: Using data from the Medicare Dataset Standard Analytic Files 5% Fee for Service database, patients receiving
either TAVR or SAVR between 2016 and 2018 were identified. Patients were categorized as low, intermediate, or high
mortality risk based on 2 validated indices—the Hospital Frailty Risk Score and the logEuroScore. Health care costs out to
1 year were compared between TAVR and SAVR among the low, intermediate, and high-risk groups, after adjustment for
patient demographics.

RESULTS: Nine thousand seven hundred forty-six patients were identified (4834 TAVR; 3760 SAVR) and included in the
analysis. Patients receiving TAVR were older and more likely to be female. Index hospitalization costs were significantly
lower with TAVR compared with SAVR across all risk strata (logEuroScore: low: $61 845 versus $68 986; intermediate:
Downloaded from http://ahajournals.org by on June 27, 2023

$64 658 versus $76 965; high: $65 594 versus $91 005; P<0.001 for all). Follow-up costs through 1 year were generally
lower with TAVR and this difference was more pronounced in the low risk groups (logEuroScore: $9763 versus $14 073;
Hospital Frailty Risk Score: $10 116 versus $12 880). Accordingly, cumulative 1-year costs were substantially lower with
TAVR compared with SAVR.

CONCLUSIONS: At 1 year, TAVR is associated with lower health care costs across all risk strata when compared with SAVR in
contemporary practice. If long-term data continue to demonstrate similar clinical outcomes and valve durability with TAVR
and SAVR, these findings suggest that TAVR may be the preferred treatment strategy for patients with aortic stenosis from
an economic standpoint.

GRAPHIC ABSTRACT: A graphic abstract is available for this article.

Key Words:  transcatheter aortic valve replacement ◼ cost analysis

See Editorial by Goldsweig and Thourani

W
hen transcatheter aortic valve replacement TAVR has emerged as the preferred treatment strategy
(TAVR) was first introduced, it was regarded as an for patients at intermediate and high-surgical risk, who
alternative for patients with severe aortic stenosis are anatomically suitable for the procedure, because of
(AS), who were too sick to undergo surgical aortic valve comparable outcomes in terms of survival and quality of
replacement (SAVR). However, over the last 10 years, life.1–7 Most recently, TAVR has also been shown to both

 
Correspondence to: Suzanne J. Baron, MD, MSc, Lahey Hospital and Medical Center, 41 Mall Rd, Burlington, MA 01805. Email suzanne.j.baron@lahey.org
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/CIRCINTERVENTIONS.121.011295.
For Sources of Funding and Disclosures, see page 277.
© 2022 American Heart Association, Inc.
Circulation: Cardiovascular Interventions is available at www.ahajournals.org/journal/circinterventions

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 268

 Baron et al Contemporary Costs Associated With TAVR and SAVR

of TAVR versus SAVR has shifted over time. As such,

WHAT IS KNOWN we sought to compare health care costs out to 1 year
• Transcatheter aortic valve replacement is a viable in patients receiving TAVR and SAVR in contemporary
treatment strategy for the treatment of symptomatic practice using Medicare claims.
severe aortic stenosis in patients of all levels of sur-
gical risk.
• Compared with surgical aortic valve replacement METHODS
prior trial-based economic studies have demon-
strated that transcatheter aortic valve replacement Data Source
is a cost-effective strategy for the treatment of high- This retrospective cohort study utilized data from the Medicare
surgical risk patients and a cost-saving strategy for Dataset Standard Analytic Files 5% Fee for Service (FFS)
patients at intermediate surgical risk. database, which encompasses a nationally representative 5%
sample of the Medicare FFS payer database. The Medicare
WHAT THE STUDY ADDS FFS payer database includes information on all health care
• Using the Medicare Dataset Standard Analytic services (including institutional and noninstitutional files),
Files 5% Fee for Service database, transcatheter which are covered for beneficiaries enrolled in Medicare Parts
aortic valve replacement was found to be associ- A and B. Access to the Medicare FFS payer database may
ated with lower health care costs during the index be requested from the Centers for Medicare and Medicaid
hospitalization, at 90 days, and at 1 year across Services Research Data Assistance Center at www.resdac.org.
all risk strata, when compared with surgical aortic All data used to perform this analysis were de-identified and
valve replacement. accessed in compliance with the Health Insurance Portability
• These findings provide a present-day update of and Accountability Act of 1996 and the Health Insurance
prior trial-based analyses and suggest that trans- Portability and Accountability Act Omnibus Rule of 2013. As a
catheter aortic valve replacement has evolved to retrospective, noninterventional, observational analysis of a de-
become an economically attractive treatment strat- identified database, the research was exempt from IRB review
egy for both payors and providers across a broad under 45 CFR 46.101(b).
patient population.
Study Population
All Medicare patients with a record of either a SAVR or trans-
femoral TAVR being performed between January 1, 2016
Nonstandard Abbreviations and Acronyms
Downloaded from http://ahajournals.org by on June 27, 2023

and December 31, 2018 were identified. Patients undergoing

SAVR, including those who had concomitant coronary artery
AS aortic stenosis bypass grafting procedures and other valvular procedures,
HFRS Hospital Frailty Risk Score were included in the primary analysis since this was felt to be
FFS Fee for Service most reflective of current practice. A secondary analysis was
LES logistic EuroScore conducted evaluating patients receiving only isolated SAVR.
To ensure an appropriate baseline period for defining comor-
SAVR surgical aortic valve replacement
bidities, patients without a minimum of 6 months of continu-
STS-PROM Society of Thoracic Surgeons Pre- ous Medicare enrollment before their SAVR or TAVR procedure
dicted Risk of Mortality were excluded.
TAVR transcatheter aortic valve replacement Each of the 2 cohorts (SAVR and TAVR) were stratified into
groups considered to be at low, intermediate, and high risk for
mortality categories by 2 validated indices—the Hospital Frailty
safe and effective for low-risk patients8,9 and has been Risk Score and the logistic European System for Cardiac
given a class I indication for the treatment of all patients Operative Risk Evaluation (EuroSCORE). Although the Society
with symptomatic, severe AS between the ages of 65 for Thoracic Surgeons Predicted Risk of Mortality (STS-PROM)
and 80 after a shared decision-making discussion.10 Score has been more commonly used in clinical trials of TAVR
With the use of TAVR expanding substantially, it is versus SAVR, this instrument requires several clinical covari-
important to understand the evolving economic implica- ates that cannot be determined from claims databases.
tions of this technology. Prior economic analyses have The Hospital Frailty Risk score (HFRS) was initially devel-
demonstrated that TAVR is cost-effective, although not oped and validated in a population of older (≥75 years) patients
cost-saving, compared with medical therapy in inoper- using electronic hospital records and International Statistical
Classification of Diseases and Related Health Problems, Tenth
able AS patients11 and compared with SAVR in patients
Revision (ICD-10) diagnostic codes for the purpose of generat-
with AS at high-surgical risk.12 In the intermediate-risk ing a widely available tool with which to identify patients at a
population though, TAVR has been shown to be cost- greater risk of adverse outcomes. This risk score has also been
saving.13 As TAVR device technology, operator and pro- validated in a US population of patients receiving percutane-
cedural proficiency, and care pathways have advanced ous valvular therapy (either TAVR or transcatheter mitral valve
over the last 10 years, it is unknown whether these prior repair).14 For each patient in the analytic cohort, the HFRS was
analyses still hold true or whether the economic balance calculated based on 109 ICD-10 diagnostic code categories

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 269

 Baron et al Contemporary Costs Associated With TAVR and SAVR

(the first 3 characters) from all claims occurring at least 6 region, and year of index procedure), and Elixhauser comor-
months before the date of admission for the index hospital- bidity index to account for patient characteristics not captured
ization. The full list of diagnoses codes used in this algorithm within a risk score. The cost analysis was conducted comparing
are shown in Supplemental Methods S1. Patients were then TAVR with all SAVR patients as well as comparing TAVR with
categorized into low risk (score<5), intermediate risk (score patients receiving isolated SAVR only.
5–15), and high risk (>15) groups based on previously pub- Tabulation of summary statistics was performed using the
lished cut-points.14 Instant Health Data platform from Panalgo. Multivariable mod-
The logistic EuroSCORE (LES) is an established risk els were run using SAS 9.4. A 2-sided P of <0.05 was consid-
score used to predict 30-day mortality in patients undergoing ered statistically significant.
cardiac surgery.15 The score is calculated using the patient’s
age and the presence or absence of ICD-10 diagnosis and
procedure codes on certain claims from the 6 months before RESULTS
each patient’s index SAVR or TAVR procedure (Supplemental
Methods S2). As previously established, patients were con- Patient Population
sidered to be at low risk if LES <10%, intermediate risk if A total of 9746 patients, who underwent SAVR or
LES 10% to 20%, and high risk if LES >20%.16 While the TAVR, were identified in the Medicare 5% FFS Payer
updated EuroSCORE II has been shown to have better pre-
database between 2016 and 2018. After excluding
dictive discrimination for operative mortality in patients under-
going cardiac surgery,17 similar to the STS-PROM score, the
patients without a minimum of 6 months of continu-
EuroSCORE II cannot be calculated from the variables avail- ous Medicare enrollment before their valve procedure,
able in an administrative database. the analytic cohort included 4834 patients with TAVR
and 3760 patients with SAVR. Of the patients with
Outcome Measures SAVR, 1500 patients underwent a combined cardiac
surgical procedure, while 2260 underwent isolated
The primary outcome of interest of this analysis was health
care costs for specific periods through 1 year. Index hospital- SAVR only (Figure 1).
ization length of stay (defined as day of admission to day of Baseline demographics of the analytic cohort com-
discharge) was also examined as a secondary outcome. Costs paring TAVR with SAVR were sub-divided by procedure
were analyzed separately at each of the following mutually and risk category are displayed in Table 1 (HFRS) and
exclusive time windows: index hospitalization, 1 to 30 days Table 2 (LES). In general, SAVR patients were younger
post-procedure, 31 to 90 days post-procedure, 91 to 180 days and more likely to be male across all risk categories. As
post-procedure, and 181 to 360 days post-procedure. Patients expected, the Elixhauser comorbidity index increased
Downloaded from http://ahajournals.org by on June 27, 2023

were required to have all or partial Medicare enrollment during with risk group within both the SAVR and TAVR cohorts
each time window. If an individual patient’s enrollment ended (P for trend <0.001). There was a higher proportion of
during a specific time window, the data for that time window
TAVR patients considered to be high risk compared with
would be estimated, but they were not considered for the next
SAVR, regardless of risk score utilized (HRFS: 24.1%
time window. For example, if a patient has 25 days of enroll-
ment post their index hospitalization, they would only be con- versus 9.8%; LES: 38.5% versus 9.9%; P<0.001 for
sidered in the index hospitalization and the 1- to 30-day post both). Similar baseline characteristics were seen when
time period (the last 5 days of which would be imputed based TAVR patients were compared with patients receiving
on the first 25 days), and they then would be dropped in the isolated SAVR (Tables S1 and S2).
31- to 90-day analysis. This piecewise approach was chosen
to ensure an accurate reflection of follow-up costs across the
total follow-up period, regardless of any differences in mortality Health Care Resource Utilization Through 1
or follow-up between the 2 groups. Year After SAVR and TAVR
Tables 3 and 4 and Figure 2 demonstrate health care
Statistical Analyses costs associated with TAVR and SAVR out to 1 year
Continuous variables are expressed as mean±SD values and after the index procedure in all risk strata as defined
were compared using t tests. Categorical variables are pre- by both the HFRS and the LES. Index hospitalization
sented as frequencies and percentages and were compared costs were significantly lower with TAVR compared
using χ2 test. Because of the inherent right-skewness of cost with SAVR across all risk groups, as defined by the
data, total costs were estimated for each time window (index, HFRS (low: $61  845 versus $68  968; intermedi-
1–30, 31–90, 91–180, and 181–360 days) using gamma log
ate: $64 658 versus $76 965; high: $65 594 versus
link regression models. Index length of stay was analyzed using
$91 005; P<0.001 for all) and the LES (low: $64 925
a general linear model with a log-negative binomial distribu-
tion. Costs were derived using the charges submitted by the versus $71  953; intermediate: $61  911 versus
facility and adjusted using each facility-specific cost-to-charge $76 562; high: $63 806 versus $90 764; P<0.001 for
ratio as per Medicare’s standardized methodology. For all cost all). This was likely primarily driven by a significantly
estimates, separate models were run for each time window for lower length of stay on the order of 5 to 8 days for
SAVR and TAVR by each risk category (low, intermediate, and patients treated with TAVR as compared with SAVR
high) with adjustments for patient demographics (age, sex, race, during the index hospitalization (Figure 3). In the

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 270

 Baron et al Contemporary Costs Associated With TAVR and SAVR

Figure 1. Flowchart demonstrating

derivation of analytic cohort.
SAVR indicates surgical aortic valve
replacement; and TAVR, transcatheter aortic
valve replacement.

low-risk groups, follow-up costs were lower at all time compared with SAVR in all risk strata, regardless of
intervals for TAVR patients, although the difference did the risk instrument used (Table 5).
not meet statistical significance in the 31- to 90-day
interval. In contrast, there were no significant differ-
Health Care Resource Utilization Through 1
ences in follow-up costs in patients treated with TAVR
or SAVR in the high-risk groups. For patients consid- Year After Isolated SAVR and TAVR
ered to be intermediate risk, follow-up costs through Since SAVR performed in conjunction with concomi-
Downloaded from http://ahajournals.org by on June 27, 2023

1 year were lower with TAVR (HRFS: $19 352 versus tant surgical procedures could be associated with
$20 089; LES: $17 012 versus $22 096), although higher costs due to the complexity of the procedure
this was primarily driven by significantly lower costs as well as the comorbidities of the patient, a second-
in the 1- to 30-day follow-up interval. Overall, cumula- ary analysis was performed comparing costs associated
tive 90-day and 1-year costs were lower with TAVR as with TAVR and isolated SAVR. Findings were generally

Table 1.  Baseline Characteristics by Risk Category per Hospital Frailty Risk Score

Low Intermediate High

TAVR SAVR TAVR SAVR TAVR SAVR

N=1600 N=1968 P value N=2067 N=1423 P value N=1167 N=369 P value

Age, y 79.8±7.7 71.8±7.1 <0.001 80.1±7.6 71.7±8.1 <0.001 80.6±8.1 70.6±10.2 <0.001

Male sex 54.8% 67.0% <0.001 53.4% 58.7% 0.002 49.4% 62.1% <0.001

White 93.6% 91.4% 0.016 93.1% 90.1% 0.002 91.9% 84.6% <0.001

Elective 82.7% 76.0% <0.001 83.6% 78.0% <0.001 78.1% 64.2% <0.001

Year
 2016 28.9% 42.0% 26.0% 36.2% 29.0% 35.2%
 2017 32.8% 29.5% <0.001 34.9% 33.5% <0.001 34.0% 31.2% 0.080
 2018 38.4% 28.5% 39.1% 30.3% 36.9% 33.6%
Diabetes 30.6% 30.9% 0.806 43.7% 41.9% 0.278 52.2% 54.2% 0.499
Hypertension 71.3% 78.5% <0.001 94.0% 90.0% <0.001 98.5% 98.1% 0.552
Peripheral vascular disorders 41.5% 37.2% 0.009 63.6% 44.2% <0.001 71.1% 57.5% <0.001

Chronic pulmonary disease 24.1% 19.4% <0.001 41.3% 30.6% <0.001 54.5% 47.2% 0.014
Renal failure 11.8% 8.8% 0.003 39.1% 26.9% <0.001 58.4% 52.6% 0.051
Elixhauser comorbidity index 4.7±3.1 4.4±2.3 <0.001 8.0±2.7 6.6±2.7 <0.001 10.8±2.9 10.1±3.3 <0.001

Continuous variables are expressed as mean±SD. SAVR indicates surgical aortic valve replacement; and TAVR, transcatheter aortic valve replacement.

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 271

 Baron et al Contemporary Costs Associated With TAVR and SAVR

Table 2.  Baseline Characteristics by Risk Category per Logistic EuroSCORE

Low Intermediate High

TAVR SAVR TAVR SAVR TAVR SAVR

N=1683 N=2849 P value N=1291 N=539 P value N=1860 N=372 P value

Age, y 76.9±7.5 70.6±7.6 <0.001 81.0±7.3 75.2±7.5 <0.001 82.5±7.2 74.4±8.5 <0.001

Male sex 55.9% 65.4% <0.001 47.8% 55.5% 0.002 53.7% 59.7% 0.033
White 93.1% 90.5% <0.001 93.3% 90.0% 0.016 92.6% 89.0% 0.018
Elective 86.0% 78.3% <0.001 82.3% 71.6% <0.001 78.1% 60.5% <0.001

Year
 2016 24.4% 39.5% 27.0% 37.5% 31.1% 39.0%
 2017 33.7% 30.6% <0.001 35.6% 34.1% <0.001 33.1% 31.7% 0.008
 2018 41.9% 30.0% 37.4% 28.4% 35.8% 29.3%
Diabetes 36.7% 35.8% 0.577 39.0% 38.4% 0.799 47.4% 47.6% 0.939
Hypertension 73.9% 81.9% <0.001 92.1% 92.2% 0.937 96.8% 96.2% 0.558
Peripheral vascular disorders 40.9% 37.6% 0.024 61.0% 52.1% <0.001 71.7% 59.9% <0.001

Chronic pulmonary disease 25.7% 21.4% <0.001 36.9% 32.7% 0.080 51.9% 55.1% 0.255
Renal failure 21.4% 15.1% <0.001 30.1% 26.9% 0.166 50.0% 47.3% 0.344
Elixhauser comorbidity index 5.5±3.6 5.1±273 <0.001 7.3±3.0 6.9±3.0 0.007 9.7±3.1 9.5±3.2 <0.001

Continuous variables are expressed as mean±SD. SAVR indicates surgical aortic valve replacement; and TAVR, transcatheter aortic valve replacement.

similar as noted above (Tables S3 through S5; Figure 4). costs in the intermediate- or high-risk groups. Cumula-
Index hospitalization costs were significantly lower with tive 90-day and 1-year costs were lower with TAVR as
TAVR compared with isolated SAVR across all risk compared with isolated SAVR in all risk strata, although
groups when stratified according to the HFRS (cost the magnitude of cost difference between TAVR and
difference—low: −$3025; intermediate: −$9781; high: SAVR was less pronounced in the low-risk groups sec-
−$23 045; P≤0.01 for all) or the LES (cost difference— ondary to the lower index hospitalization costs associ-
Downloaded from http://ahajournals.org by on June 27, 2023

low: −$3688; intermediate: −$13 746; high: −$23 299; ated with isolated SAVR versus all SAVR.
P≤0.001 for all), although the difference in the low-risk
groups was of a smaller magnitude when qualitatively
compared with the analysis evaluating all SAVR proce- DISCUSSION
dures. In the low-risk groups, follow-up costs were again In this retrospective analysis of costs associated with
lower at all time intervals for TAVR patients, while there TAVR and SAVR in a contemporary setting, we found
was generally no significant difference in follow-up that patients with severe AS who were treated with TAVR

Table 3.  Comparison of Costs by Time Interval Within Each Risk Category per Hospital Frailty Risk Score

Risk category Time period TAVR SAVR Cost difference (95% CI) P value
Low Index hospitalization $61 845 $68 968 −$7123 (−$9444 to $4878) <0.001

Discharge to 30 d $1637 $2922 −$1285 (−$1877 to −$793) <0.001

31–90 d $1955 $2185 −$230 (−$635 to $111) 0.199

91–180 d $2089 $2540 −$451 (−$963 to −$24) 0.037
181–360 d $2943 $5136 −$2194 (−3265 to −$1307) <0.001

Intermediate Index Hospitalization $64 658 $76 965 −$12 307 (−$15 110 to −$9603) <0.001

Discharge to 30 d $2551 $5041 −$2490 (−$3441 to −$1690) <0.001

31–90 d $5151 $4169 $982 ($226 to $1621) 0.013

91–180 d $4676 $5052 −$376 (−1377 to $459) 0.401
181–360 d $6974 $5827 $1147 (−$84 to $2164) 0.066
High Index Hospitalization $65 594 $91 005 −$25 411 (−$31 034 to −$20 115) <0.001

Discharge to 30 d $4939 $4577 $362 ($1412 to $1640) 0.649

31–90 d $5323 $4786 $537 (−$1068 to $1739) 0.471
91–180 d $6524 $7594 −$1070 (−$3952 to $1019) 0.355
181–360 d $10 279 $8413 $1866 (−$1650 to $4345) 0.261

SAVR indicates surgical aortic valve replacement; and TAVR, transcatheter aortic valve replacement.

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 272

 Baron et al Contemporary Costs Associated With TAVR and SAVR

Table 4.  Comparison of Costs by Time Interval Within Each Risk Category per Logistic EuroScore

Risk category Time period TAVR SAVR Cost difference (95% CI) P value
Low Index hospitalization $64 925 $71 953 −$7028 (−$9217 to −$4904) <0.001

Discharge to 30 d $2061 $3109 −$1048 (−$1572 to −$600) <0.001

31–90 d $2376 $2477 −$101 (−$485 to $231) 0.571

91–180 d $2807 $3538 −$731 (−$1325 to −$222) 0.004
181–360 d $4011 $5046 −$1036 (−$1992 to $232) 0.009
Intermediate Index hospitalization $61 911 $76 562 −$14 651 (−$18 229 to −$11 232) <0.001

Discharge to 30 d $2021 $6219 −$4198 (−$5963 to −$2824) <0.001

31–90 d $5207 $4898 $309 (−$1047 to $1370) 0.624

91–180 d $3625 $4121 −$497 (−$1690 to $428) 0.322
181–360 d $6160 $6857 −$697 (−$2750 to $883) 0.422
High Index hospitalization $63 806 $90 764 −$26 958 (−$31 960 to −$22 217) <0.001

Discharge to 30 d $4060 $5268 −$1208 (−$2963 to $109) 0.076

31–90 d $4761 $4380 $382 (−$942 to $1398) 0.535
91–180 d $5539 $6208 −$670 (−$2799 to $915) 0.448
181–360 d $7706 $8065 −$359 (−$3183 to $1732) 0.766

SAVR indicates surgical aortic valve replacement; and TAVR, transcatheter aortic valve replacement.

had significantly lower costs at discharge and through vascular complications, and disabling stroke, all of which
1-year of follow-up compared with those patients, who have been associated with increased resource utiliza-
were treated with SAVR. Cost savings in the TAVR group tion and costs.19 In addition, more centers are employing
was primarily driven by lower expenditures during the streamlined care pathways that are directed at decreas-
index hospitalization; however, patients at lower risk also ing resources used during the procedure and facilitating
demonstrated reduced costs at follow-up time periods as next-day discharge. These programs have not only been
well when compared with SAVR. These findings provide associated with good clinical outcomes but also with
Downloaded from http://ahajournals.org by on June 27, 2023

a present-day update of prior trial-based analyses and shorter lengths of stay and resulting lower index hospi-
suggest that TAVR has evolved to become an economi- talization costs.20,21 With these improvements in proce-
cally attractive treatment strategy for both payors and dural outcomes and postprocedural care over time, it is
providers across a broad patient population. not surprising that the length of stay for these high-risk
patients has decreased substantially over time (mean
of 8–10 days in the PARTNER 1A and CoreValve High
Comparison With Prior Economic Studies of
Risk trial versus median of 3 days in our study). That
TAVR and SAVR said, it is important to note that patients were deemed
The cost-effectiveness of TAVR as compared with “high-risk” in the PARTNER 1A and the CoreValve High
SAVR has been studied in multiple economic sub-stud- Risk trial by the STS risk score as opposed to the LES
ies embedded within the pivotal clinical TAVR trials. In and HRFS instruments, which were used in this analysis,
patients considered to be high risk for SAVR, TAVR was and thus it is not possible to perform a direct compari-
associated with higher index hospitalization costs in the son between the economic results derived from these
CoreValve High Risk trial by ≈$11 00018 (84% of whom pivotal trials and our analysis. Nevertheless, it is unlikely
were treated with transfemoral access) and with simi- that low-risk patients, as defined by a low STS risk score,
lar index hospitalization costs in the transfemoral cohort were substantially represented in our high-risk cohorts,
of PARTNER 1A trial when compared with SAVR.12 In given the advanced age, multiple comorbidities, and high
contrast, TAVR patients deemed high risk by either the Elixhauser comorbidity index scores of these patients.
LES or the HRFS had significantly lower index hospi- In prior trial-based economic analyses performed
talization costs in our analysis by ≈$25 000. There are in patients with AS at intermediate surgical risk, index
several possible explanations for this finding. Over the hospitalization costs have been shown to be lower for
last 10 years, there have been multiple advancements patients treated with new generation TAVR devices by
in TAVR device technology and postprocedural manage- up to ≈$4000 when compared with SAVR.13,22 Since
ment, both of which have likely led to a decrease in costs index procedural costs were shown to be higher with
associated with the TAVR procedure and hospitalization. TAVR in these analyses because of the cost of the
Enhancements in the TAVR device and delivery system TAVR implant itself, the cost savings was driven primar-
as well as improved operator proficiency has led to lower ily by reductions in total length of stay (≈4–6 days).13
rates of periprocedural complications, such as bleeding, In the low and intermediate-risk cohorts in our study

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 273

 Baron et al Contemporary Costs Associated With TAVR and SAVR
Downloaded from http://ahajournals.org by on June 27, 2023

Figure 2. Healthcare costs associated with transcatheter aortic valve replacement (TAVR) and all surgical aortic valve
replacement (SAVR) through 1 y in low, intermediate, and high risk categories.
A, Risk categories stratified using Hospital Frailty Risk Score. B, Risk categories stratified using Logistic EuroScore.

(who likely represent an overlapping patient popula- substantially over time, it is possible that the greater
tion with the aforementioned trial-based analyses since cost-savings seen in our study is reflective of greater
TAVR was not approved for low-surgical risk patients differences in procedural resource utilization. For exam-
until after the time period of our study [eg, 2019]), we ple, registry analyses have shown that TAVRs are being
found more substantial cost savings during the index performed more frequently in the catheterization labo-
hospitalization costs by ≈$7000 in the low-risk cohort ratory as opposed to the operating room.23 In addition,
and by ≈$12 000 to $14 000 in the intermediate-risk TAVRs are now often done under conscious sedation
group with TAVR. Interestingly, we observed these sav- as opposed to general anesthesia, and this strategy has
ings despite noting differences in median length of stay been associated with lower mortality, shorter length of
between patients treated with TAVR and SAVR that stay, and higher rates of discharge to home.24 Accord-
were similar to those observed in the pivotal random- ingly, it would follow that these sorts of improvements
ized trials of intermediate-risk patients.3,7 Given that in resource utilization would contribute to lower costs
the cost of the valve implants has also not changed associated with TAVR over time.

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 274

 Baron et al Contemporary Costs Associated With TAVR and SAVR

Figure 3. Index hospitalization

median length of stay in days after
transcatheter aortic valve replacement
(TAVR) and all surgical aortic
valve replacement (SAVR) in low,
intermediate, and high risk categories
per Hospital Frailty Risk Score and
Logistic EuroScore.

With regards to costs accrued after the index hospi- reflect the trends toward lower postprocedural health
talization, costs were noted to be significantly lower with care resource utilization for these patients with more
TAVR at almost all follow-up time periods through 1 year patients being discharged home, as opposed to short-
in the low-risk groups. These findings are consistent with term nursing facilities or inpatient rehabilitation centers,
a prior cost-effectiveness analysis comparing patients and lower rates of re-hospitalizations after TAVR over
with AS at intermediate surgical risk treated with TAVR time.23,25 In contrast, the overall 1-year follow-up costs
versus SAVR, which also demonstrated lower follow-up observed in both TAVR and SAVR patients in the high-
Downloaded from http://ahajournals.org by on June 27, 2023

costs at 1 and 2 years with TAVR.13 Additionally, 1-year risk strata were very similar to those reported in prior
follow-up costs for TAVR patients were markedly lower trial-based economic studies of this patient population
by up to ≈$19 000 in the low-risk strata and ≈$10 000 (≈$25 000 to $30 000)12,18 and Medicare-linked regis-
in the intermediate risk strata than those observed in try analyses.26 Given that an average adult over the age
the cost-effectiveness analysis of the PARTNER 2A of 70 with 1 reported limitation in an activity of daily
study,13 a population presumed to overlap substantially living is estimated to utilize ≈$22 000 health care dol-
with the low and intermediate risk cohorts of this analy- lars per year (in 2018 dollars),27 it follows that patients
sis for the reasons noted above. These findings may considered to be high risk for TAVR or SAVR by any
definition will likely continue to use substantial health
care resources because of other comorbidities and
Table 5.  Cumulative 90-Day and 1-Year Health Care Costs
Stratified by Risk Category
thus, improvements in care processes related to their
valve replacement will not affect their cumulative health
Cumulative 90-d costs* Cumulative 1-y costs† care costs as substantially as it would in the lower-risk
Risk
Score category TAVR SAVR TAVR SAVR populations.
HFRS Low $65 436 $74 075 $70 468 $81 751
Intermediate $72 360 $86 175 $84 009 $97 054
Future Economic Implications
High $75 856 $100 368 $92 659 $116 375
LES Low $69 361 $77 539 $76 179 $86 124
When TAVR was initially introduced as a treatment strat-
egy for severe AS, there was concern that the high cost
Intermediate $69 139 $87 679 $78 923 $98 657
of the TAVR device as compared with a surgical valve
High $72 628 $100 411 $85 872 $114 685
would make the TAVR procedure economically unviable.
HRFS indicates Hospital Frailty Risk Score; LES, Logistic EuroScore; Indeed, a Medicare claims studies looking at all aortic
SAVR, surgical aortic valve replacement; and TAVR, transcatheter aortic valve
replacement. valve replacements (TAVR and SAVR) in 2012 found that
*Cumulative 90-d costs derived from summation of costs derived from indi- the contribution margin (defined as the median of the dif-
vidual models for index hospitalization, 1–30, and 31–90 d time intervals. ferences between the payment and estimated hospital
†Cumulative 1-y costs derived from summation of costs derived from indi-
vidual models for index hospitalization, 1–30, 31–90, 91–180, and 181–360 cost), was negative for TAVR hospitalizations by $3200
d time intervals. and positive for SAVR hospitalizations by $2700.28 From

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 275

 Baron et al Contemporary Costs Associated With TAVR and SAVR
Downloaded from http://ahajournals.org by on June 27, 2023

Figure 4. Healthcare costs associated with transcatheter aortic valve replacement (TAVR) and isolated surgical aortic valve
replacement (SAVR) through 1 y in low, intermediate, and high risk categories.
A, Risk categories stratified using Hospital Frailty Risk Score. B, Risk categories stratified using Logistic EuroScore.

a payor perspective, Brescia et al29 also demonstrated in TAVR and SAVR index hospitalization costs by 2016.30
that early era TAVR was associated with increased pay- Our study demonstrates that this trend has continued,
ments not only for the index hospitalization episode of such that costs are now not only lower during the index
care but also out to 90 days, which could carry significant hospitalizations but that these cost savings extend out
implications if the TAVR procedure was ever considered to 90 days and even to 1 year in the contemporary era.
for a bundled payment model. However, as TAVR proce- Since the difference in costs in the low-risk groups of this
dure has evolved in efficiency and safety over time, there analysis were statistically significant both for the index
have been suggestions that this procedure could become hospitalization and in follow-up, it seems likely that these
more economically attractive. Indeed, an analysis of the findings will extend to the cohort of patients, defined
Nationwide Readmissions Database found that TAVR as low surgical risk by the STS risk score, who became
index hospitalization costs decreased between 2012 eligible for commercial TAVR in 2019. Taken together,
and 2016, such that there was no significant difference these data are suggestive that TAVR may emerge as

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 276

 Baron et al Contemporary Costs Associated With TAVR and SAVR

an economically dominant strategy for all risk groups, with patients who were lost to follow-up or died were
assuming that the clinical outcomes and durability of the accounted for.
TAVR device remain comparable to SAVR.
Conclusions
Limitations In conclusion, based on a Medicare claims-based analy-
This study should be interpreted in the context of several sis utilizing data from 2016 to 2018, TAVR is associated
limitations. First, this is a retrospective cohort study that with lower health care costs across all risk strata when
relies on coding and claims from the Medicare FFS 5% compared with SAVR. If long-term data continue to dem-
dataset. As such, many clinical parameters were ascer- onstrate similar clinical outcomes and valve durability
tained through sources of automated data, which could with TAVR and SAVR, these findings suggest that TAVR
lead to the over- or under-coding of diagnoses. Addi- may be the preferred treatment strategy for patients with
tionally, since the dataset only included patients covered AS from an economic standpoint.
by Medicare Parts A and B, these findings may not be
generalizable to self-pay patients or patients covered by
ARTICLE INFORMATION
other insurance plans (eg, Medicare Advantage or pri-
Received July 20, 2021; accepted December 16, 2021.
vate insurances). Likewise, since the analytic cohort was
drawn from a 5% sample of Medicare patients, it is pos- Affiliations
sible that sampling error could lead to a nonrepresenta- Lahey Hospital and Medical Center, Burlington, MA (S.J.B.). Baim Institute
tive population and this could affect the generalizability for Clinical Research, Boston, MA (S.J.B.). MPR Consulting, Cincinnati, OH
(M.P.R.). Edwards LifeSciences, Irvine, CA (K.A.M.). Gunnarsson Consulting,
of our findings as well, Furthermore, claims databases, Jupiter, FL (C.L.G.).
such as the Medicare FFS 5% dataset, only collect data
on certain comorbidities and also lack some granularity Sources of Funding
The data were obtained, and this analysis was funded by Edwards LifeSciences.
of data (eg, echocardiogram parameters, hemodynamics
measures, hospital procedural volumes), thereby leading Disclosures
to the possibility of unmeasured confounding variables. Dr Baron reports research funding from Boston Scientific Corporation and
In fact, since it would be impossible to completely control Abiomed and consulting/advisory board fees from MitraLabs, Boston Scientific
Corporation, and Abiomed. K.A. Moore and S.J. Clancy are employees of Edwards
for all factors that would contribute to a patient being LifeSciences. M.P. Ryan and C.L. Gunnarsson received consulting fees from Ed-
chosen for TAVR versus SAVR outside of a random-
Downloaded from http://ahajournals.org by on June 27, 2023

ward LifeSciences for the performance of the statistical analysis.

ized controlled trial (eg, heart team evaluation; anatomic
Supplemental Material
considerations), it is likely that there are some inherent Supplemental Methods
differences between the 2 cohorts despite risk stratifi- Tables S1–S5
cation. Nevertheless, given that patients receiving TAVR
were likely of higher risk (since TAVR was not approved
REFERENCES
for low-risk patients between 2016 and 2018) and thus
1. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG,
would likely have utilized more health care resources, it
Tuzcu EM, Webb JG, Fontana GP, Makkar RR, et al; PARTNER Trial
is possible that our findings may represent an underesti- Investigators. Transcatheter versus surgical aortic-valve replacement in
mation of the true cost-savings associated with TAVR in high-risk patients. N Engl J Med. 2011;364:2187–2198. doi: 10.1056/
NEJMoa1103510
the current era.
2. Reynolds MR, Magnuson EA, Wang K, Thourani VH, Williams M, Zajarias A,
It is also important to note that the definitions of low, Rihal CS, Brown DL, Smith CR, Leon MB, et al; PARTNER Trial Investiga-
intermediate, and high surgical risk in this study are dif- tors. Health-related quality of life after transcatheter or surgical aortic valve
replacement in high-risk patients with severe aortic stenosis: results from the
ferent than those used in the pivotal randomized TAVR
PARTNER (Placement of AoRTic TraNscathetER Valve) Trial (Cohort A). J Am
trials (eg, the STS-PROM score), thereby limiting direct Coll Cardiol. 2012;60:548–558. doi: 10.1016/j.jacc.2012.03.075
comparisons between the economic analyses of those 3. Leon MB, Smith CR, Mack MJ, Makkar RR, Svensson LG, Kodali SK,
Thourani VH, Tuzcu EM, Miller DC, Herrmann HC, et al; PARTNER 2
studies and the findings described herein. Furthermore,
Investigators. Transcatheter or surgical aortic-valve replacement in
the HFRS has not been directly correlated with the STS- intermediate-risk patients. N Engl J Med. 2016;374:1609–1620. doi:
PROM score nor has it been studied in patients receiving 10.1056/NEJMoa1514616
4. Baron SJ, Arnold SV, Wang K, Magnuson EA, Chinnakondepali K,
SAVR (although it has been validated in patients receiv-
Makkar R, Herrmann HC, Kodali S, Thourani VH, Kapadia S, et al; PART-
ing percutaneous valve treatments14). Nevertheless, NER 2 Investigators. Health status benefits of transcatheter vs surgical
the use of the HFRS and LES in this study does lend a aortic valve replacement in patients with severe aortic stenosis at interme-
diate surgical risk: results from the PARTNER 2 Randomized Clinical Trial.
foundation for the use of these risk instruments in future
JAMA Cardiol. 2017;2:837–845. doi: 10.1001/jamacardio.2017.2039
claims-based economic analyses. Last, our chosen meth- 5. Adams DH, Popma JJ, Reardon MJ. Transcatheter aortic-valve replacement
odology did not allow us to statistically compare cumula- with a self-expanding prosthesis. N Engl J Med. 2014;371:967–968. doi:
10.1056/NEJMc1408396
tive 1-year costs; however, we think that our methods are
6. Arnold SV, Reynolds MR, Wang K, Magnuson EA, Baron SJ, Chinnakondepalli
more representative of actual costs incurred during each KM, Reardon MJ, Tadros PN, Zorn GL, Maini B, et al; CoreValve US Piv-
specific follow-up time period since costs associated otal Trial Investigators. Health status after transcatheter or surgical aortic

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 277

 Baron et al Contemporary Costs Associated With TAVR and SAVR

valve replacement in patients with severe aortic stenosis at increased surgi- valve replacement with a self-expanding prosthesis versus surgical aor-
cal risk: results from the corevalve US pivotal trial. JACC Cardiovasc Interv. tic valve replacement. J Am Coll Cardiol. 2016;67:29–38. doi: 10.1016/j.
2015;8:1207–1217. doi: 10.1016/j.jcin.2015.04.018 jacc.2015.10.046
7. Reardon MJ, Van Mieghem NM, Popma JJ, Kleiman NS, Søndergaard L, 19. Arnold SV, Lei Y, Reynolds MR, Magnuson EA, Suri RM, Tuzcu EM,
Mumtaz M, Adams DH, Deeb GM, Maini B, Gada H, et al; SURTAVI Inves- Petersen JL, 2nd, Douglas PS, Svensson LG, Gada H, et al; PARTNER
tigators. Surgical or transcatheter aortic-valve replacement in interme- Investigators. Costs of periprocedural complications in patients treated with
diate-risk patients. N Engl J Med. 2017;376:1321–1331. doi: 10.1056/ transcatheter aortic valve replacement: results from the Placement of Aor-
NEJMoa1700456 tic Transcatheter Valve trial. Circ Cardiovasc Interv. 2014;7:829–836. doi:
8. Popma JJ, Deeb GM, Yakubov SJ, Mumtaz M, Gada H, O’Hair D, Bajwa T, 10.1161/CIRCINTERVENTIONS.114.001395
Heiser JC, Merhi W, Kleiman NS, et al; Evolut Low Risk Trial Investiga- 20. Babaliaros V, Devireddy C, Lerakis S, Leonardi R, Iturra SA, Mavromatis K,
tors. Transcatheter aortic-valve replacement with a self-expanding valve Leshnower BG, Guyton RA, Kanitkar M, Keegan P, et al. Comparison of
in low-risk patients. N Engl J Med. 2019;380:1706–1715. doi: 10.1056/ transfemoral transcatheter aortic valve replacement performed in the cath-
NEJMoa1816885 eterization laboratory (minimalist approach) versus hybrid operating room
9. Mack MJ, Leon MB, Thourani VH, Makkar R, Kodali SK, Russo M, (standard approach): outcomes and cost analysis. JACC Cardiovasc Interv.
Kapadia SR, Malaisrie SC, Cohen DJ, Pibarot P, et al; PARTNER 3 Inves- 2014;7:898–904. doi: 10.1016/j.jcin.2014.04.005
tigators. Transcatheter aortic-valve replacement with a balloon-expandable 21. Lauck SB, Wood DA, Baumbusch J, Kwon JY, Stub D, Achtem L,
valve in low-risk patients. N Engl J Med. 2019;380:1695–1705. doi: Blanke P, Boone RH, Cheung A, Dvir D, et al. Vancouver transcatheter aortic
10.1056/NEJMoa1814052 valve replacement clinical pathway: minimalist approach, standardized care,
10. Otto CM, Nishimura RA, Bonow RO, Carabello BA, Erwin JP, 3rd, Gentile F, and discharge criteria to reduce length of stay. Circ Cardiovasc Qual Out-
Jneid H, Krieger EV, Mack M, McLeod C, et al. 2020 ACC/AHA guideline comes. 2016;9:312–321. doi: 10.1161/CIRCOUTCOMES.115.002541
for the management of patients with valvular heart disease: a report of the 22. Reardon MJ, Geisler BP, Popma JJ, van Mieghem NM, Serruys P,
American College of Cardiology/American Heart Association Joint Com- Deeb GM, Windecker S, Yakubov SJ, Sondergaard L, Pietzsch JB. Cost-
mittee on Clinical Practice Guidelines. Circulation. 2021;143:e72–e227. doi: effectiveness of transcatheter aortic valve implantation in patients at
10.1161/CIR.0000000000000923 intermediate surgical risk: a model-based analysis based on the SUR-
11. Reynolds MR, Magnuson EA, Wang K, Lei Y, Vilain K, Walczak J, TAVI Trial using contemporary cost data. Oral presentation at: EuroPCR;
Kodali SK, Lasala JM, O’Neill WW, Davidson CJ, et al; PARTNER Inves- May, 2019; Paris, France.
tigators. Cost-effectiveness of transcatheter aortic valve replacement 23. Carroll JD, Mack MJ, Vemulapalli S, Herrmann HC, Gleason TG, Hanzel G,
compared with standard care among inoperable patients with severe Deeb GM, Thourani VH, Cohen DJ, Desai N, et al. STS-ACC TVT Registry of
aortic stenosis: results from the placement of aortic transcatheter valves transcatheter aortic valve replacement. J Am Coll Cardiol. 2020;76:2492–
(PARTNER) trial (Cohort B). Circulation. 2012;125:1102–1109. doi: 2516. doi: 10.1016/j.jacc.2020.09.595
10.1161/CIRCULATIONAHA.111.054072 24. Butala NM, Chung M, Secemsky EA, Manandhar P, Marquis-Gravel G,
12. Reynolds MR, Magnuson EA, Lei Y, Wang K, Vilain K, Li H, Walczak J, Kosinski AS, Vemulapalli S, Yeh RW, Cohen DJ. Conscious sedation versus
Pinto DS, Thourani VH, Svensson LG, et al; PARTNER Investigators. Cost- general anesthesia for transcatheter aortic valve replacement: variation in
effectiveness of transcatheter aortic valve replacement compared with practice and outcomes. JACC Cardiovasc Interv. 2020;13:1277–1287. doi:
surgical aortic valve replacement in high-risk patients with severe aor- 10.1016/j.jcin.2020.03.008
tic stenosis: results of the PARTNER (Placement of Aortic Transcath- 25. Kawsara A, Sulaiman S, Linderbaum J, Coffey SR, Alqahtani F,
eter Valves) trial (Cohort A). J Am Coll Cardiol. 2012;60:2683–2692. doi: Nkomo VT, Crestanello JA, Alkhouli M. Temporal trends in resource use,
10.1016/j.jacc.2012.09.018 cost, and outcomes of transcatheter aortic valve replacement in the
Downloaded from http://ahajournals.org by on June 27, 2023

13. Baron SJ, Wang K, House JA, Magnuson EA, Reynolds MR, Makkar R, United States. Mayo Clin Proc. 2020;95:2665–2673. doi: 10.1016/j.
Herrmann HC, Kodali S, Thourani VH, Kapadia S, et al. Cost-Effectiveness mayocp.2020.05.043
of transcatheter versus surgical aortic valve replacement in patients with 26. Vemulapalli S, Dai D, Hammill BG, Baron SJ, Cohen DJ, Mack MJ,
severe aortic stenosis at intermediate risk. Circulation. 2019;139:877–888. Holmes DR, Jr. Hospital resource utilization before and after transcathe-
doi: 10.1161/CIRCULATIONAHA.118.035236 ter aortic valve replacement: The STS/ACC TVT registry. J Am Coll Cardiol.
14. Kundi H, Popma JJ, Reynolds MR, Strom JB, Pinto DS, Valsdottir LR, 2019;73:1135–1146. doi: 10.1016/j.jacc.2018.12.049
Shen C, Choi E, Yeh RW. Frailty and related outcomes in patients under- 27. Lubitz J, Cai L, Kramarow E, Lentzner H. Health, life expectancy, and
going transcatheter valve therapies in a nationwide cohort. Eur Heart J. health care spending among the elderly. N Engl J Med. 2003;349:1048–
2019;40:2231–2239. doi: 10.1093/eurheartj/ehz187 1055. doi: 10.1056/NEJMsa020614
15. Roques F, Michel P, Goldstone AR, Nashef SA. The logistic EuroSCORE. Eur 28. McCarthy FH, Savino DC, Brown CR, Bavaria JE, Kini V, Spragan DD,
Heart J. 2003;24:881–882. doi: 10.1016/s0195-668x(02)00799-6 Dibble TR, Herrmann HC, Anwaruddin S, Giri J, et al. Cost and contribu-
16. Clark MA, Arnold SV, Duhay FG, Thompson AK, Keyes MJ, Svensson LG, tion margin of transcatheter versus surgical aortic valve replacement.
Bonow RO, Stockwell BT, Cohen DJ. Five-year clinical and economic out- J Thorac Cardiovasc Surg. 2017;154:1872–1880.e1. doi: 10.1016/j.
comes among patients with medically managed severe aortic stenosis: jtcvs.2017.06.020
results from a Medicare claims analysis. Circ Cardiovasc Qual Outcomes. 29. Brescia AA, Syrjamaki JD, Regenbogen SE, Paone G, Pruitt AL,
2012;5:697–704. doi: 10.1161/CIRCOUTCOMES.112.966002 Shannon FL, Boeve TJ, Patel HJ, Thompson MP, Theurer PF, et al. Trans-
17. Ad N, Holmes SD, Patel J, Pritchard G, Shuman DJ, Halpin L. Com- catheter versus surgical aortic valve replacement episode payments and
parison of EuroSCORE II, original EuroSCORE, and the society of tho- relationship to case volume. Ann Thorac Surg. 2018;106:1735–1741. doi:
racic surgeons risk score in cardiac surgery patients. Ann Thorac Surg. 10.1016/j.athoracsur.2018.07.017
2016;102:573–579. doi: 10.1016/j.athoracsur.2016.01.105 30. Goldsweig AM, Tak HJ, Chen LW, Aronow HD, Shah B, Kolte D, Desai NR,
18. Reynolds MR, Lei Y, Wang K, Chinnakondepalli K, Vilain KA, Magnuson EA, Szerlip M, Velagapudi P, Abbott JD. Relative costs of surgical and transcath-
Galper BZ, Meduri CU, Arnold SV, Baron SJ, et al; CoreValve US High eter aortic valve replacement and medical therapy. Circ Cardiovasc Interv.
Risk Pivotal Trial Investigators. Cost-Effectiveness of Transcatheter aortic 2020;13:e008681. doi: 10.1161/CIRCINTERVENTIONS.119.008681

Circ Cardiovasc Interv. 2022;15:e011295. DOI: 10.1161/CIRCINTERVENTIONS.121.011295 March 2022 278