J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8

ª 2 0 1 8 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O UN DA T I O N

P U B L I S H E D B Y E L S E V I E R

FOCUS ON TRANSCATHETER AORTIC VALVE REPLACEMENT

The Learning Curve and AnnualProcedure Volume Standards forOptimum Outcomes of TranscatheterAortic Valve Replacement

Findings From an International Registry

Anthony W.A. Wassef, MD,a Josep Rodes-Cabau, MD,b Yaqing Liu, MSC,a John G. Webb, MD,c Marco Barbanti, MD,d

Antonio J. Muñoz-García, MD, PHD,e Corrado Tamburino, MD, PHD,d Antonio E. Dager, MD,f Vicenç Serra, MD,g

Ignacio J. Amat-Santos, MD, PHD,h Juan H. Alonso Briales, MD,e Alberto San Roman, MD,h Marina Urena, MD, PHD,i

Dominique Himbert, MD,i Lius Nombela-Franco, MD, PHD,j Alexandre Abizaid, MD, PHD,k Fabio S. de Brito, JR, MD,l

Henrique B. Ribeiro, MD, PHD,m Marc Ruel, MD,n Valter C. Lima, MD,o Fabian Nietlispach, MD,p

Asim N. Cheema, MD, PHDa

JACC: CARDIOVASCULAR INTERVENTIONS CME/MOC

This article has been selected as this issue’s CME/MOC activity, available

online at http://www.acc.org/jacc-journals-cme by selecting the JACC

Journals CME/MOC tab.

Accreditation and Designation Statement

The American College of Cardiology Foundation (ACCF) is accredited by

the Accreditation Council for Continuing Medical Education (ACCME) to

provide continuing medical education for physicians.

The ACCF designates this Journal-based CME/MOC activity for a

maximum of 1 AMA PRA Category 1 Credit(s). Physicians should only

claim credit commensurate with the extent of their participation in

the activity.

Method of Participation and Receipt of CME/MOC Certificate

To obtain credit for this CME/MOC activity, you must:

1. Be an ACC member or JACC: Cardiovascular Interventions

subscriber.

2. Carefully read the CME/MOC-designated article available online and

in this issue of the journal.

3. Answer the post-test questions. At least 2 out of the 3 questions

provided must be answered correctly to obtain CME/MOC credit.

4. Complete a brief evaluation.

5. Claim your CME/MOC credit and receive your certificate electronically

by following the instructions given at the conclusion of the activity.

ISSN 1936-8798/$36.00

From the aDivision of Cardiology, Department of Medicine, St. Michael’s

Institute, Laval University, Quebec City, Canada; cDivision of Cardiology, De

of British Columbia, Vancouver, Canada; dDivision of Cardiology, FerraroeDepartment of Cardiology, Hospital Universitario Virgen de la Victoria, Un

Cardiology, Clínica de Occidente de Cali, Cali, Colombia; gDepartment of I

d’Hebron,UniversitatAutònomadeBarcelona, Barcelona, Spain; hCIBERCV,H

Spain; iDepartment of Cardiology, Bichat Hôpital, AP-HP, University Paris Did

Universitario Clínico San Carlos, Madrid, Spain; kInstituto Dante Pazzanes

Cardiology Department, Hospital Israelita Albert Einstein, São Paulo, Brazil; m

CME/MOC Objective for This Article: At the end of the activity the reader

should be able to: 1) compare the differences in clinical outcomes

between high- and low- volume centers after transcatheter aortic valve

replacement (TAVR); 2) recognize the level of procedural experience

needed to achieve optimal clinical outcomes after TAVR; and 3) appre-

ciate the significance of TAVR learning curve leading to incremental

improvement in clinical outcomes.

CME/MOC Editor Disclosure: JACC: Cardiovascular Interventions CME/MOC

Editor Bill D. Gogas, MD, PhD, has reported that he has no disclosures.

Author Disclosures: Dr. Rodes-Cabau has received research grants from

Edwards Lifesciences and Medtronic. Dr. Webb has received research

grants from and served as a consultant for Edwards Lifesciences and

Abbott. Dr. Barbanti has served as a consultant for Edwards Lifesciences.

Dr. Ruel has received research grants from Medtronic and Edwards Life-

sciences. Dr. Himbert has served as a proctor for Edwards Lifesciences and

Medtronic. Dr. Nombela-Franco has served as a proctor for Abbott. Dr.

Abizaid has served as a proctor for Edwards Lifesciences. Drs. de Brito and

Nietlispach has served as a consultant for Edwards Lifesciences, Abbott,

and Medtronic. All other authors have reported that they have no re-

lationships relevant to the contents of this paper to disclose.

Medium of Participation: Print (article only); online (article and quiz).

CME/MOC Term of Approval

Issue Date: September 10, 2018

Expiration Date: September 9, 2019

https://doi.org/10.1016/j.jcin.2018.06.044

Hospital, Toronto, Canada; bQuebec Heart & Lung

partment of Medicine, St. Paul’s Hospital, University

tto Hospital, University of Catania, Catania, Italy;

iversidad de Málaga, Málaga, Spain; fDepartment of

nterventional Cardiology, Hospital Universitari Vall

ospital ClínicoUniversitario deValladolid, Valladolid,

erot, Paris, France; jInstituto Cardiovascular, Hospital

e de Cardiologia, São Paulo, Brazil; lInterventional

Heart Institute (InCor), São Paulo, Brazil; nDivision of

Wassef et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8

TAVR Learning Curve and Annual Volume Standards S E P T E M B E R 1 0 , 2 0 1 8 : 1 6 6 9 – 7 9

1670

The Learning Curve and An

nual ProcedureVolume Standards for Optimum Outcomes ofTranscatheter Aortic Valve Replacement

Findings From an International Registry

Anthony W.A. Wassef, MD,a Josep Rodes-Cabau, MD,b Yaqing Liu, MSC,a John G. Webb, MD,c Marco Barbanti, MD,d

Antonio J. Muñoz-García, MD, PHD,e Corrado Tamburino, MD, PHD,d Antonio E. Dager, MD,f Vicenç Serra, MD,g

Ignacio J. Amat-Santos, MD, PHD,h Juan H. Alonso Briales, MD,e Alberto San Roman, MD,h Marina Urena, MD, PHD,i

Dominique Himbert, MD,i Lius Nombela-Franco, MD, PHD,j Alexandre Abizaid, MD, PHD,k Fabio S. de Brito, JR, MD,l

Henrique B. Ribeiro, MD, PHD,m Marc Ruel, MD,n Valter C. Lima, MD,o Fabian Nietlispach, MD,p

Asim N. Cheema, MD, PHDa

ABSTRACT

Ca

Po

Ed

sci

Me

Fra

lis

no

Ma

OBJECTIVES The authors aimed to determine the procedural learning curve and minimum annual institutional volumes

associated with optimum clinical outcomes for transcatheter aortic valve replacement (TAVR).

BACKGROUND Transcatheter aortic valve replacement (TAVR) is a complex procedure requiring significant training and

experience for successful outcome. Despite increasing use of TAVR across institutions, limited information is available

for its learning curve characteristics and minimum annual volumes required to optimize clinical outcomes.

METHODS The study collected data for patients at 16 centers participating in the international TAVR registry since

initiation of the respective TAVR program. All cases were chronologically ordered into initial (1 to 75), early (76 to 150),

intermediate (151 to 225), high (226 to 300), and very high (>300) experience operators for TAVR learning curve charac-

terization. In addition, participating institutions were stratified by annual TAVR case volume into low-volume (<50),

moderate-volume (50 to 100), and high-volume (>100) groups for comparative analysis. Procedural and 30-day clinical

outcomes were collected and multivariate regression analysis performed for 30-day mortality and the early safety endpoint.

RESULTS A total of 3,403 patients comprised the study population. On multivariate analysis, all-cause mortality was

significantly higher for initial (odds ratio [OR]: 3.83; 95% confidence interval [CI]: 1.93 to 7.60), early (OR: 2.41; 95%

CI: 1.51 to 5.03), and intermediate (OR: 2.53; 95% CI: 1.19 to 5.40) experience groups compared with the very high

experience operators. In addition, the early safety endpoint was significantly worse for all experience groups compared

with the very high experience operators. Low annual volume (<50) TAVR institutions had significantly higher all-cause

30-day mortality (OR: 2.70; 95% CI: 1.44 to 5.07) and worse early safety endpoint (OR: 1.60; 95% CI: 1.17 to 2.17)

compared with the moderate- and high-volume groups. There was no difference in patient outcomes between

intermediate and high annual volume groups.

CONCLUSIONS TAVR procedures display important learning curve characteristics with both greater procedural safety

and a lower mortality when performed by experienced operators. In addition, TAVR performed at low annual

volume (<50 procedures) institutions is associated with decreased procedural safety and higher patient mortality.

These findings have important implications for operator training and patient care at centers performing TAVR.

(J Am Coll Cardiol Intv 2018;11:1669–79) © 2018 by the American College of Cardiology Foundation.

rdiac Surgery, Ottawa Heart Institute, Ottawa, Canada; oHospital São Francisco-Santa Casa de Misericórdia de Porto Alegre,

rto Alegre, Brazil; and the pUniversity Hospital Zürich, Zürich, Switzerland. Dr. Rodes-Cabau has received research grants from

wards Lifesciences and Medtronic. Dr. Webb has received research grants from and served as a consultant for Edwards Life-

ences and Abbott. Dr. Barbanti has served as a consultant for Edwards Lifesciences. Dr. Ruel has received research grants from

dtronic and Edwards Lifesciences. Dr. Himbert has served as a proctor for Edwards Lifesciences and Medtronic. Dr. Nombela-

nco has served as a proctor for Abbott. Dr. Abizaid has served as a proctor for Edwards Lifesciences. Drs. de Brito and Niet-

pach has served as a consultant for Edwards Lifesciences, Abbott, and Medtronic. All other authors have reported that they have

relationships relevant to the contents of this paper to disclose.

nuscript received September 26, 2017; revised manuscript received May 28, 2018, accepted June 26, 2018.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8 Wassef et al.S E P T E M B E R 1 0 , 2 0 1 8 : 1 6 6 9 – 7 9 TAVR Learning Curve and Annual Volume Standards

1671

AB BR E V I A T I O N S

AND ACRONYM S

CAD = coronary artery disease

CI = confidence interval

NYHA = New York Heart

Association

OR = odds ratio

TAVR = transcatheter aortic

valve replacement

T ranscatheter aortic valve replacement(TAVR) has revolutionized the treatment ofsevere symptomatic aortic stenosis, with

several randomized trials demonstrating equivalenceor superiority to conventional surgical aortic valvereplacement for inoperable (1), high-risk (2,3), andintermediate-risk patients (4,5). Concomitantly, therehas been a large increase in the number of proceduresbeing performed and the number of centers perform-ing the procedure in North America as well as interna-tionally (6,7). This trend is likely to continue, as thereis expected to be a large increase in the number ofelderly patients with aortic valve disease (8), as wellas an increase in the utilization of this technology inlower-risk patients (9).

SEE PAGE 1680

A learning curve phenomenon, defined as animprovement in outcome with increasing experiencehas been demonstrated for multiple cardiac (10,11)as well as noncardiac (12) procedures. In addition,maintaining minimal annual procedural volumes areassociated with improved clinical outcomes (13).Although improved procedural results with greaterTAVR experience have been described (14–17),adequate understanding of the TAVR learning curveand the minimal annual procedural volume requiredto achieve competence is lacking. In the presentstudy, we characterize the TAVR procedure learningcurve and investigate the relationship betweenannual institutional volume and clinical outcomesfrom the international multicenter TAVR registry.

METHODS

All consecutive patients who underwent TAVR at16 large, urban international academic teaching sitesin North and South America and Europe since theinitiation of the respective center’s TAVR programwere included in this study. All centers employeda heart-team model of multidisciplinary decisionmaking for patient selection, procedural planningand performance (6). Choice of TAVR device, valvesize, approach, post-procedure care, and antith-rombotic management were at the discretion oftreating center’s heart team, and participation inclinical trials as well as investigational devices wereincluded in this analysis. A total of 3,468 patientsunderwent TAVR during the study period with com-plete data were available for 3,403 (98.1%) patientscomprising the study population. Patients withincomplete data, totaling <2% of the population wereexcluded. Baseline, procedural, echocardiographic,

and outcome data were collected in a pro-spective manner at each center. Each centerwas responsible for collection of baselinedemographic and procedural details, as wellas relevant 30-day outcomes. Events wereadjudicated at each site, with no centraladjudication. The first patient included hadthe procedure performed January 6, 2005,and the last patient included had the pro-cedure January 29, 2016.

Patients from each center were chronologicallyordered from the initiation of the TAVR programfor the learning curve analysis. To determine theprocedural learning curve characteristics, TAVR casesfrom each center were chronologically grouped intoinitial (1 to 75), early (76 to 150), intermediate (151to 225), high (226 to 300), and very high (>300)experience groups, to determine the effect ofincreasing procedural experience on procedural andclinical outcomes. To determine the minimum annualinstitutional TAVR volume for optimum clinicaloutcomes, each individual center’s TAVR volumeper calendar year (January 1 to December 31) wasdetermined, and grouped into low (1 to 49), inter-mediate (50 to 100) and high (>100) TAVR volumegroups. Centers may contribute to different volumegroups depending only on their annual volume percalendar year. For this analysis, all TAVR performedin the first and last calendar year from each centerwere excluded from analysis as the completenessof annual volume data could not be assured. Com-plete data available for 2,205 (64.8%) patients from 16institutions.

All procedural and clinical outcomes were definedper Valve Academic Research Consortium-2 criteria(18) and determined at 30 days. Briefly, device successwas defined as successful access, delivery, deploy-ment in the correct position and anatomical locationwith aortic valve area >1.2 cm2 and mean aortic valvegradient <20 mm Hg or peak velocity <3 m/s, withoutmoderate or severe aortic insufficiency. Early safetyendpoint was defined as the composite of death,stroke, major bleeding, vascular complications, sur-gical conversion, and renal failure.

STATISTICAL ANALYSIS. Categorical variables werereported as frequency and percentages and contin-uous variables as mean � SD. For the learning curveanalysis, the very high-experience group was usedas a reference and baseline characteristics, proce-dural variables, and clinical outcomes for eachexperience group (initial, early, intermediate, andexperienced) were compared with the very high-experience group using paired wise comparisons

Wassef et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8

TAVR Learning Curve and Annual Volume Standards S E P T E M B E R 1 0 , 2 0 1 8 : 1 6 6 9 – 7 9

1672

with a Student’s t-test, Mann-Whitney U test, or achi-square test as appropriate, with Dunnett’s test tocontrol for multiple comparisons. A logistic regres-sion was performed to determine association be-tween experience and annual volume groupallocation and all-cause mortality and compositeearly safety endpoint. Covariate adjustment withlogistic regression has been demonstrated tocompare well to propensity adjustment methodsfor clinical trials (19). Variables chosen for the modelwere included if there were significant baselinedifferences between the groups, and if they hadpreviously been demonstrated to predict worseoutcomes in TAVR (20). The variables included inthe logistic regression for the learning curve analysisincluded baseline characteristics age, body surfacearea, sex, New York Heart Association (NYHA)functional class, prior coronary artery bypass graft-ing, prior coronary artery disease (CAD), chronickidney disease, and continent of the participatinginstitution (Europe, South America, North America),procedural variables (transfemoral, transapical, orother access) and prosthesis generation for theSAPIEN platform (SAPIEN, SAPIEN XT, SAPIEN 3) toaccount for the geographical differences in clinicalpractice and technological changes which mayimpact vascular access complications and para-valvular leak rates. The logistic regression analysiswas performed first using procedural volume as acontinuous variable and then repeated using theprocedural volume categories as defined previously.Results of the multivariate analysis were expressedas odds ratio (OR) with 95% confidence interval (CI).All analyses were performed with SAS softwareversion 9.4 (SAS Institute, Cary, North Carolina) anda p value < 0.05 defined statistical significance.

RESULTS

TAVR LEARNING CURVE AND CLINICAL OUTCOMES.

Among the 3,403 patients enrolled at 16 interna-tional sites, 1,141 (33.5%) cases were performed byinitial operators, 780 (22.9%) cases performedby early operators, 549 (16.1%) cases performed byintermediate-experience operators, 354 (10.4%)cases performed by high-experience operators, and579 (17.0%) cases were performed by very high-experience operators. 16 centers contributed datato the initial cohort, 13 centers to early, 8 centers tointermediate, 6 centers to the high, and 4 centersto the very high cohort. Most participating centersshowed gradually increasing TAVR volumes overthe period of the study without a rapid ramp-up

phenomenon from one year to the next (Figure 1,Online Figure). Baseline unadjusted clinical andprocedural characteristics of the study populationare shown in Table 1. The mean age of the studypopulation was 82 � 8 years, with 77% of patientsundergoing transfemoral TAVR, and a balloon-expandable valve as used in 59% of patients. Therewere statistically significant intergroup differencesin age (p < 0.001), mean aortic valve gradient(p < 0.001), body mass index (p < 0.001), Societyof Thoracic Surgeons Predicted Risk of Mortality(p < 0.001), NYHA functional class IV (p < 0.001),CAD (p < 0.001), and the type of valve implanted(p < 0.001).

The unadjusted clinical and procedural outcomesfor the different TAVR experience groups are shownin Table 2. There was a consistent decrease in allcause death with increasing TAVR experienceand was significantly higher for the initial experi-ence (9.6%; p < 0.001) and early experience (7.9%;p ¼ 0.002) groups compared with the very high-experience (3.3%) operators (Figure 1A). Similarly,the composite early safety endpoint decreased withincreasing procedural experience from 27.5% for theinitial experience to 14.9% for very high-experienceoperators (Figure 1B), largely driven by reductionsin major bleeding (Figure 1C) and major vascularcomplications (Figure 1D). The procedural time andlength of stay post-procedure were only high in theinitial experience with no significant differencebetween early, intermediate, experienced, and veryhigh-experience groups (Table 2). The contrastvolume administered was also high in the initialand early groups compared with very high-experience operators (p < 0.001). The rate of pro-cedural success did not change with increasingexperience, and the rates of stroke, myocardialinfarction, new dialysis, severe aortic regurgitation,and surgical conversion were similar acrossdifferent operator experience groups. Although,there were significant intergroup differences forprocedure time and permanent pacemaker inser-tion, these did not demonstrate a stepwise learningcurve phenomenon.

The results of multivariate logistic regressionanalysis showed that TAVR procedural volumewhen used as a continuous variable was indepen-dently associated with 30-day mortality (p <

0.0001) (Figure 2A), major bleeding (p < 0.0001),and major adverse cardiac events (p < 0.0001). Inaddition, the multivariate logistic regression anal-ysis identified TAVR performed by the initial (OR:3.83; 95% CI: 1.93 to 7.60; p < 0.001), early (OR:

FIGURE 1 Unadjusted Clinical and Procedural Outcomes According to TAVR Procedural Experience

Measures of clinical success including (A) 30-day mortality, (B) early safety endpoint as well as procedural complications including (C) major bleeding and (D) major

vascular complications decreased with increasing procedural experience. TAVR ¼ transcatheter aortic valve replacement.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8 Wassef et al.S E P T E M B E R 1 0 , 2 0 1 8 : 1 6 6 9 – 7 9 TAVR Learning Curve and Annual Volume Standards

1673

2.41; 95% CI: 1.51 to 5.03; p ¼ 0.020), and inter-mediate (OR: 2.53; 95% CI: 1.19 to 5.40; p ¼ 0.016)experience groups were independently associatedwith higher mortality compared with the very high-experience operators (Figure 2B). In addition, theearly safety endpoint was significantly higher forinitial (OR: 2.02; 95% CI: 1.41 to 2.89; p < 0.001),early (OR: 1.74; 95% CI: 1.19 to 2.56; p ¼ 0.005),intermediate (OR: 2.07; 95% CI: 1.40 to 3.07;p < 0.01), and high (OR1.70; 95% CI: 1.14 to 2.59;p ¼ 0.009) experience groups compared with the veryhigh-experience operators (Figure 2C). Higher Societyof Thoracic Surgeons score (OR: 1.84; 95% CI: 1.20 to2.82; p ¼ 0.005) and transapical access (OR: 1.84; 95%CI: 1.20 to 2.82; p ¼ 0.005) were also independentpredictors of increased mortality, whereas left ven-tricular ejection fraction, sex, NYHA functional class,

prior coronary artery bypass grafting, chronic kidneydisease, chronic obstructive pulmonary disease sta-tus, and balloon-expandable versus self-expandingvalve did not demonstrate a significant associationwith increased mortality or early safety endpoint (datanot shown).

ANNUAL VOLUME AND CLINICAL OUTCOMES. Atotal of 2,205 of 3,403 (64.8%) patients were includedin this analysis, with 569 (25.8%) cases performed atlow, 1,121 (50.8%) cases performed at intermediate,and 515 (23.3%) cases performed at high annualvolume centers (Table 3). There were significant dif-ferences in the proportion of women (p ¼ 0.018),mean body mass index (p ¼ 0.001), left ventricularejection fraction (p ¼ 0.001), Society of ThoracicSurgeons Predicted Risk of Mortality (p ¼ 0.001), CAD

TABLE 2 Unadjusted Clinical and Procedural Outcomes for Learning Curve Analysis

InitialExperience(n ¼ 1,141)

pValue

EarlyExperience(n ¼ 780)

pValue

ModerateExperience(n ¼ 549)

pValue

HighExperience(n ¼ 354)

pValue

Very-HighExperience(n ¼ 579)

Procedural outcomes

Device success 873 (76.5) 0.22 633 (81.2) 0.985 473 (86.2) 0.032 296 (83.6) 0.523 465 (80.3)

Procedure time, min 135.0 � 80.0 0.001 78.0 � 77.0 0.805 47.0 � 45.0 <0.001 79.0 � 53.0 0.940 83.0 � 38.2

Contrast volume, ml 143 � 95 0.001 139 � 124 <0.001 77 � 44 1.000 56 � 32 0.101 78 � 51

Surgical conversion 21 (1.8) 0.860 9 (1.2) 0.255 5 (0.9) 0.192 7 (2) 0.982 14 (2.4)

Clinical outcomes

Death 109 (9.6) <0.001 62 (7.9) 0.002 32 (5.8) 0.115 17 (4.8) 0.525 19 (3.3)

Early safety endpoint 314 (27.5) <0.001 222 (28.5) <0.001 143 (26.0) 0.032 80 (22.6) 0.098 86 (14.9)

Bleeding 129 (11.3) <0.001 63 (8.1) 0.111 35 (6.4) 0.767 26 (7.3) 0.429 30 (5.2)

Vascular complication 121 (10.6) 0.001 87 (11.2) 0.001 70 (12.8) <0.001 34 (9.6) 0.043 31 (5.4)

Stroke 35 (3.1) 0.532 27 (3.5) 0.336 19 (3.5) 0.387 9 (2.5) 0.963 12 (2.1)

Myocardial infarction 21 (1.8) 1.00 5 (0.6) 0.145 8 (1.5) 0.953 2 (0.6) 0.342 11 (1.9)

New dialysis 20 (1.8) 0.429 11 (1.4) 0.780 1 (0.2) 0.430 1 (0.3) 0.708 5 (0.9)

New PPM 172 (15.1) 0.007 106 (13.6) 0.085 64 (11.7) 0.612 72 (20.3) <0.001 56 (9.7)

Moderate-severe AI 59 (5.8) 0.914 20 (3.1) 0.974 5 (1.1) 0.891 15 (4.2) 1.000 42 (8.1)

LOS post-TAVR, days 15.4 � 37.4 0.001 12.2 � 13.8 0.117 11.1 � 22.2 0.452 9.0 � 8.0 0.988 9.0 � 18.0

Values are n (%) or mean � SD. Very high experience (>300 procedures) group used as a reference group for pairwise comparison.

AI ¼ aortic insufficiency; LOS ¼ length of stay; PPM ¼ permanent pacemaker; TAVR ¼ transcatheter aortic valve replacement.

TABLE 1 Baseline Unadjusted Clinical and Procedural Characteristics for Learning Curve Analysis

All(N ¼ 3,403)

Initial Experience(n ¼ 1,141)

Early Experience(n ¼ 780)

Moderate Experience(n ¼ 549)

High Experience(n ¼ 354)

Very High Experience(n ¼ 579)

Age, yrs 81.5 � 7.7 81.8 � 7.2 81.6 � 7.2 81.1 � 7.6 82.5 � 7.6 80.5 � 9.1

Female 1,652 (48.6) 576 (50.6) 384 (49.2) 271 (49.4) 170 (48.0) 251 (43.7)

Gradient, mm Hg 46.3 � 16.9 49.0 � 16.8 47.1 � 16.9 45.7 � 15.9 44.4 � 16.0 41.3 � 17.0

AV area, cm2 0.6 � 0.2 0.6 � 0.2 0.6 � 0.2 0.6 � 0.2 0.6 � 0.2 0.6 � 0.3

BMI, kg/m2 26.8 � 5.2 26.3 � 4.9 26.9 � 5.1 26.7 � 5.2 27.1 � 5.0 27.4 � 6.0

LVEF, % 54.5 � 14.1 55.2 � 14.1 55.3 � 13.7 53.3 � 14.7 53.9 � 13.9 53.7 � 14.1

STS PROM 8.2 � 6.4 9.1 � 7.4 7.7 � 5.9 8.7 � 7.2 7.4 � 5.0 7.0 � 4.5

Hypertension 2,676 (78.6) 855 (74.9) 598 (76.7) 446 (81.2) 294 (83.1) 483 (83.4)

DM 1,017 (29.9) 328 (28.7) 222 (28.5) 151 (27.5) 113 (31.9) 203 (35.1)

NYHA functional class IV 504 (14.8) 183 (16.0) 115 (14.7) 92 (16.8) 63 (17.8) 51 (8.8)

AF 1,033 (30.4) 299 (26.2) 244 (31.3) 184 (33.5) 113 (31.9) 193 (33.3)

CAD 1,900 (55.8) 622 (54.5) 390 (50.0) 306 (55.7) 209 (59.0) 373 (64.4)

Prior CABG 769 (22.6) 249 (21.9) 158 (20.3) 114 (20.8) 86 (24.3) 162 (28.1)

COPD 907 (26.7) 297 (26.0) 199 (25.5) 154 (28.1) 101 (28.5) 156 (26.9)

CKD 1,707 (50.2) 597 (52.3) 355 (45.5) 255 (46.4) 198 (55.9) 302 (52.2)

Prior MI 491 (14.4) 146 (12.8) 121 (15.5) 56 (10.2) 49 (13.8) 119 (20.6)

TF approach 2,622 (77.0) 935 (81.9) 611 (78.3) 426 (77.6) 255 (72.0) 395 (68.2)

SE THV 1,397 (41.1) 593 (52.0) 332 (42.6) 214 (39) 159 (44.9) 99 (17.1)

Values are mean � SD or n (%). Very high experience (>300 procedures) group used as a reference group for pair wise comparison.

AF ¼ atrial fibrillation; AV ¼ aortic valve; BMI ¼ body mass index; CABG ¼ coronary artery bypass grafting; CAD ¼ coronary artery disease; COPD ¼ chronic obstructivepulmonary disease; DM ¼ diabetes mellitus; LVEF ¼ left ventricular ejection fraction; NYHA ¼ New York Heart Association; MI ¼myocardial infarction; SE THV ¼ self-expandingtranscatheter heart valve; STS PROM ¼ Society of Thoracic Surgeons Predicted Risk of Mortality; TF ¼ transfemoral.

Wassef et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8

TAVR Learning Curve and Annual Volume Standards S E P T E M B E R 1 0 , 2 0 1 8 : 1 6 6 9 – 7 9

1674

FIGURE 2 Multivariate Logistic Regression Analysis of Mortality Using Procedural

Volume as a Continuous Variable, Categorical Variable, and Early Safety Endpoint

According to TAVR Procedural Experience

(A) The odds of mortality were independently associated with procedural experience and

significantly decreased with increasing experience. In addition, (B) a procedural experi-

ence of #225 procedures was independently associated with higher 30-day mortality

compared with procedural experience of >300 cases, although (C) the early safety

endpoint continued to improve beyond 225 TAVR case volume. TAVR ¼ transcatheter

aortic valve replacement.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8 Wassef et al.S E P T E M B E R 1 0 , 2 0 1 8 : 1 6 6 9 – 7 9 TAVR Learning Curve and Annual Volume Standards

1675

(p ¼ 0.001), and chronic kidney disease (p ¼ 0.001)among the 3 groups. The use of transfemoralapproach (p < 0.001) and self-expanding valve(p < 0.001) was highest at low compared with highannual volume centers. The unadjusted clinical andprocedural outcomes for the 3 annual TAVR volumegroups are shown in Table 4. All-cause mortalitywas significantly higher in the low-volume centerscompared with the high-volume centers (8.8% vs.3.9%; p ¼ 0.003), but was similar between interme-diate- and high-volume centers (Figure 3A). The earlysafety endpoint (34.3% vs. 26.4%; p ¼ 0.01), andmajor bleeding (11.1%, vs. 4.9%; p ¼ 0.037) werealso significantly higher for low-volume centerscompared with high-volume centers, with no differ-ence between intermediate- and high-volume centers(Figures 3B to 3D). The procedure time (184.3 � 101.6min vs. 84.8 � 36.3 min; p < 0.001), and contrastvolume (219.8 � 92.4 ml vs. 74.8 � 45.5 ml; p < 0.001)were higher in low-volume centers compared withhigh-volume centers. The device success rates andrisk of post-procedural myocardial infarction, stroke,surgical conversion, new pacemaker, or aortic regur-gitation were similar for the 3 volume groups.The results of logistic multivariate analysis showedthat low annual volume group was independentlyassociated with increased mortality (OR: 2.70; 95% CI:1.44 to 5.07; p ¼ 0.002) and high rate of compositeearly safety endpoint (OR: 1.60, 95% CI: 1.17 to 2.17;p ¼ 0.003) (Figure 4).

DISCUSSION

The main findings of the present study are: 1) animportant learning curve exists with incrementalimprovement in clinical outcomes for increasingTAVR experience; 2) a TAVR experience of at least225 procedures is associated with reduced TAVRmortality while the early safety endpoint continuesto improve beyond the 225-case volume; and 3) theinstitutional annual procedural volume affectsTAVR clinical outcomes and low volume centersdefined as an annual case volume <50 TAVR pro-cedures are independently associated with highermortality.

The number of TAVR procedures being performedhas increased markedly as the indications haveexpanded from inoperable patients (1) to high-risk(2,3) and intermediate-risk patients (4). There islikely to be a further increase in TAVR utilizationin view of demographic changes (21) and potentialconfirmation of safety and efficacy in lower riskpatients (7). However, TAVR remains a technicallychallenging procedure requiring a high level of

TABLE 4 Unadjusted

Analysis

Patients

Death

Device success

Early safety endpoint

Stroke

Major bleeding

Major vascular complica

Surgical conversion

New PPM

Aortic insufficiency (>2

LOS post-TAVR, days

Values are n (%) or meanpairwise comparison.

Abbreviations as in Tabl

TABLE 3 Baseline Unadjusted Clinical and Procedural Characteristics for Annual

Volume-Outcome Analysis

Low Volume(1–49)

IntermediateVolume

(50–100)

HighVolume(>100)

pValue

Patients 569 (25.8) 1,121 (50.84) 515 (23.36)

Age, yrs 81.5 � 7.2 81.6 � 7.2 80.7 � 9.7 0.09

Female 263 (46.3) 571 (50.9) 224 (43.8) 0.01

AV gradient, mm Hg 48.6 � 16.2 45.7 � 16.4 41.3 � 17.5 <0.001

AV area, cm2 0.7 � 0.2 0.6 � 0.2 0.6 � 0.3 0.05

BMI, kg/m2 26.4 � 4.9 27.4 � 5.3 26.5 � 5.7 <0.001

LVEF, % 54.8 � 14.1 55.1 � 13.8 52.2 � 14.7 <0.001

STS PROM 8.7 � 7.6 6.9 � 5.6 8.1 � 5.6 <0.001

CAD 314 (55.2) 579 (51.7) 324 (62.9) <0.001

CKD 300 (52.7) 504 (45) 280 (54.4) <0.001

COPD 144 (25.3) 303 (27) 141 (27.4) 0.68

DM 168 (29.5) 345 (30.8) 170 (33) 0.45

Hypertension 448 (78.7) 890 (79.4) 409 (79.4) 0.94

AF/atrial flutter 137 (24.1) 326 (29.1) 184 (35.7) <0.001

NYHA functional class IV 83 (14.6) 155 (13.8) 59 (11.5) 0.28

CABG 121 (21.3) 235 (21) 152 (29.6) <0.001

TF approach 470 (82.6) 905 (80.7) 342 (66.4) <0.001

Medtronic CoreValve 945 (42.9) 312 (54.8) 568 (50.7) <0.001

Edwards SAPIEN 430 (19.5) 90 (15.8) 277 (24.7)

Edwards SAPIEN XT/3 830 (37.6) 178 (29.4) 283 (24.6)

Values are n (%) or mean � SD. The high-volume (>100 procedures/year) group used as a reference group forpairwise comparison.

Abbreviations as in Table 1.

Wassef et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8

TAVR Learning Curve and Annual Volume Standards S E P T E M B E R 1 0 , 2 0 1 8 : 1 6 6 9 – 7 9

1676

skill and prior experience with percutaneous andendovascular procedures. In addition to a compre-hensive pre-procedural patient assessment by adedicated heart team, the TAVR requires a unique

Clinical and Procedural Outcomes for Annual Volume-Outcome

LowVolume(1-49)

pValue

IntermediateVolume(50–100)

pValue

HighVolume(>100)

569 (26.0) 1,121 (51.0) 515 (23.0)

50 (8.8) 0.003 64 (5.7) 0.22 20 (3.9)

444 (78.0) 0.88 933 (83.2) 0.07 407 (79.0)

195 (34.3) 0.01 313 (27.9) 0.73 136 (26.4)

21 (3.7) 0.30 35 (3.1) 0.54 12 (2.3)

63 (11.1) 0.03 82 (7.3) 0.24 25 (4.9)

tion 49 (8.6) 0.87 102 (9.1) 0.98 48 (9.3)

16 (1.4) 0.86 16 (1.4) 0.65 10 (1.9)

75 (13.2) 0.57 166 (14.8) 0.11 59 (11.5)

þ) 22 (4.2) 0.92 33 (3.4) 0.29 34 (7.4)

15.0 � 21.0 <0.001 11.0 � 23.1 0.1 8.9 � 7.7

� SD. High-volume (>100 procedures/year) group used as a reference group for

e 2.

skillset and expertise for implantation and expedi-tiously manage unexpected complications. There-fore, adequate training and experience play acritical role in improving procedural safety andclinical success. In addition, similar to other pro-cedural skills, it is likely that a minimum annualprocedural volume is required to provide optimumresults.

TAVR LEARNING CURVE. The 2 important compo-nents of any procedural learning curve include mea-sures of clinical outcome including patient survivalor procedural success and process efficiencyincluding procedural time and resource utilization.For cardiovascular procedures, prior learning curvestudies have demonstrated important relationshipfor increasing experience with both outcome mea-sures as well as process efficiency (12,22). Four years’experience post-training has shown to decreasemortality for coronary artery bypass surgery (23) and>150 cases required to optimize outcomes of percu-taneous balloon mitral valvuloplasty (10). Similarly,minimum procedural volume criteria for optimizingprocedural efficiency has been described for radialPCI (11,24).

In the present study, we examined 30-day mor-tality and composite early safety endpoint as mea-sures of clinical outcome and procedural time andcontrast media usage as measures of proceduralefficiency. In addition, the very high-experiencegroup was defined for operators with >300 casevolume as the reference group to assess thelearning curve of the all operators and effect onoutcome measures with increasing procedural vol-umes. The patients undergoing TAVR by the veryhigh-experience operators in the present studyshowed a mortality rate of 3.3%, which is lowerthan that reported for the STS/ACC TVT (Society ofThoracic Surgeons/American College of CardiologyTranscatheter Valve Therapy Registry (6) andsimilar to sites with the most TAVR experience inthe United States (25). TAVR learning curve analysisfrom the PARTNER I (Placement of Aortic Trans-catheter Valves) trial (14,15) showed an excessmortality and procedural failure only for initialTAVR experience but the findings are limited by asmall sample size. These results of the presentstudy are consistent with recently published datafrom the TVT registry, showing lower major adverseevents with increasing TAVR experience (25). Carrollet al. (25) observed a reduction in vascular andbleeding complications most noted in the first 100cases, and a reduction in mortality that becamestatistically insignificant after adjustment. When

FIGURE 3 Unadjusted Clinical and Procedural Outcomes According to Annual Institutional TAVR Volume

The risk of (A) 30-day death, (B) early safety endpoint, and (C) major bleeding were lower in centers with an annual TAVR volume >100

whereas the risk of (D) major vascular complications was similar among low- and high-volume centers. TAVR ¼ transcatheter aortic valve

replacement.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8 Wassef et al.S E P T E M B E R 1 0 , 2 0 1 8 : 1 6 6 9 – 7 9 TAVR Learning Curve and Annual Volume Standards

1677

used as a continuous variable, the present studyidentified procedural volume to be independentlyassociated with 30-day mortality (Figure 2A), majorbleeding and major adverse cardiac events andthe odds of mortality continued to decrease withincreasing procedural experience. On comparingprocedural experience as a categorical variable inmultivariate regression analysis, the 30-day mor-tality showed a consistent decrease for up to a 225-case volume (Figure 2B). For the early safetyendpoint up, we observed improvement beyond the225-case volume (Figure 2C), a finding consistentwith earlier reports (25). Similarly, the rates ofstroke, paravalvular aortic regurgitation, and dial-ysis requirement did not change with increasingexperience. The measures of procedural efficiencysuch as procedure time, contrast volume alsoimproved with increasing experience but no

significant difference was observed after initial andearly experience when compared with the veryhigh-experience operators.

ANNUAL TAVR VOLUMES AND CLINICAL OUTCOMES.

The present study showed that low-volume (<50cases/year) centers experienced a 30-day mortality of8.8%, which is significantly higher than 5.7% forintermediate-volume (51 to 100 cases/year) and 3.9%for high-volume (51 to 100 cases/year) centers(Figure 3). The mortality rates for high-volumecenters in the present study are lower than those re-ported for contemporary TAVR experience (6,25,26)confirming that high annual volume is an importantcontributor to improved clinical outcomes. Similarly,the early safety endpoint was also significantlyworse among the low-volume group and low-volumegroup was an independent predictor of worse 30-day

FIGURE 4 Multivariate Analysis of Mortality and Early Safety Endpoint According to

Annual Institutional TAVR Volume

Annual institutional transcatheter aortic valve replacement (TAVR) volume #50

procedures was independently associated with worse (A) 30-day mortality and

(B) early safety endpoint compared with institutional annual TAVR volume >100 cases.

Wassef et al. J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8

TAVR Learning Curve and Annual Volume Standards S E P T E M B E R 1 0 , 2 0 1 8 : 1 6 6 9 – 7 9

1678

mortality and the early safety endpoint (Figure 4).There is limited data for the relationship betweenannual institutional TAVR volume and clinicaloutcomes. A recent study by Khera et al. (27)showed that low-volume centers identified asperforming <50 TAVR cases per year have a signifi-cantly higher 30-day readmission rates comparedwith high-volume centers performing >100 casesper year.

Despite the large increase in the number of TAVRprocedures being performed at greater number ofinstitutions, many hospitals perform a relativelysmall number of procedures. There are many lowvolume centers performing <50 procedures a yearboth in the United States and internationally. Thistrend is likely to continue with potential utilization ofthis technology in lower risk patients (9). The findingsfrom the present study suggest a minimum annualvolume threshold to provide the best clinical out-comes for patients undergoing TAVR proceduresand can serve as a guide for optimal distribution ofresources and technology.

STUDY LIMITATIONS. First, there were significantdifferences in baseline demographic and procedural

characteristics between the chronological and vol-ume groups in the learning curve and annual volumeanalysis that may have confounded the outcomes ofinterest. Although, multivariate regression analysiswas performed to adjust for these differences; otherunmeasured confounders may still remain. Second,this study was only able to assess the center learningcurve and center annual volume, not of individualoperators. However, the current guidelines recom-mend a heart team approach to decision making andmost centers require a 2-member TAVR team forprocedures making individual operator data lessrelevant. Prior data from coronary interventionshave shown center volume to be a more robustpredictor of outcome compared with individualoperator volume (13,28). However, the interactionbetween low-volume TAVR operators and highannual volume centers and vice versa could not beassessed in the present analysis as all operatorsperformed TAVR procedures at a single institution.Our analysis included data from 16 centers onlywhich may have influenced the findings due to asmall sample size. Finally, many of the centers thatwere included in this study began their TAVRexperience early; whether the learning curve may beabbreviated in the current environment with morecenters performing the procedure, with newer tech-nology and more educational opportunities isnot known.

CONCLUSIONS

In this study examining the learning curve andannual procedural volumes of center’s performingTAVR using a large, international database, wefound that at least 225 procedures are required tooptimize mortality rates for TAVR and proceduralcomplications continue to decrease beyond the 225case volume. Annual institutional volume of <50procedures per year was associated with worseclinical and procedural outcomes. These findingshave important implications for operator trainingand patient care at centers performing TAVR.Further research is required to determine whethernewer TAVR technology, focused training, andproctoring can abbreviate the TAVR learningcurve.

ADDRESS FOR CORRESPONDENCE: Dr. Asim N.Cheema, Division of Cardiology, St. Michael’s Hos-pital, 30 Bond Street, Toronto, Ontario M5B 1W8,Canada. E-mail: cheemaa@smh.ca.

PERSPECTIVES

WHAT IS KNOWN? The present study analyzed data

from a large international multicenter registry and found

an important TAVR learning curve with 30-day mortality

that decreased with increasing procedural experience up

to 225 cases.

WHAT IS NEW? The early safety endpoint, however,

continued to improve beyond the 225-case volume.

In addition institutions with an annual TAVR volume

<50 cases were associated with worse 30-day mortality

and early safety endpoint compared with institutions

with an annual TAVR volume >100 cases.

WHAT IS NEXT? These findings have important impli-

cations for operator training and patient care

at institutions performing TAVR.

J A C C : C A R D I O V A S C U L A R I N T E R V E N T I O N S V O L . 1 1 , N O . 1 7 , 2 0 1 8 Wassef et al.S E P T E M B E R 1 0 , 2 0 1 8 : 1 6 6 9 – 7 9 TAVR Learning Curve and Annual Volume Standards

1679

RE F E RENCE S

1. Leon MB, Smith CR, Mack M, et al. Trans-catheter aortic-valve implantation for aortic ste-nosis in patients who cannot undergo surgery.N Engl J Med 2010;363:1597–607.

2. Adams DH, Popma JJ, Reardon MJ. Trans-catheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med 2014;371:967–8.

3. Smith CR, Leon MB, Mack MJ, et al. Trans-catheter versus surgical aortic-valve replacementin high-risk patients. N Engl J Med 2011;364:2187–98.

4. Leon MB, Smith CR, Mack MJ, et al., PARTNER 2Investigators. Transcatheter or surgical aortic-valve replacement in intermediate-risk patients.N Engl J Med 2016;374:1609–20.

5. Reardon MJ, Van Mieghem NM, Popma JJ,et al., SURTAVI Investigators. Surgical or trans-catheter aortic-valve replacement in intermediate-risk patients. N Engl J Med 2017;376:1321–31.

6. Grover FL, Vemulapalli S, Carroll JD, et al., STS/ACC TVT Registry. 2016 Annual Report of TheSociety of Thoracic Surgeons/American College ofCardiology Transcatheter Valve Therapy Registry.J Am Coll Cardiol 2017;69:1215–30.

7. Mylotte D, Osnabrugge RL, Windecker S, et al.Transcatheter aortic valve replacement in Europe:adoption trends and factors influencing deviceutilization. J Am Coll Cardiol 2013;62:210–9.

8. Nkomo VT, Gardin JM, Skelton TN,Gottdiener JS, Scott CG, Enriquez-Sarano M.Burden of valvular heart diseases: a population-based study. Lancet 2006;368:1005–11.

9. Thyregod HG, Steinbruchel DA, Ihlemann N,et al. Transcatheter versus surgical aortic valvereplacement in patients with severe aortic valvestenosis: 1-year results from the all-comersNOTION randomized clinical trial. J Am Coll Car-diol 2015;65:2184–94.

10. Sanchez PL, Harrell LC, Salas RE, Palacios IF.Learning curve of the Inoue technique of percu-taneous mitral balloon valvuloplasty. Am J Cardiol2001;88:662–7.

11. Ball WT, Sharieff W, Jolly SS, et al. Character-ization of operator learning curve for transradial

coronary interventions. Circ Cardiovasc Interv2011;4:336–41.

12. Hopper AN, Jamison MH, Lewis WG. Learningcurves in surgical practice. Postgrad Med J 2007;83:777–9.

13. Hannan EL, Racz M, Ryan TJ, et al. Coronaryangioplasty volume-outcome relationships forhospitals and cardiologists. JAMA 1997;277:892–8.

14. Minha S, Waksman R, Satler LP, et al. Learningcurves for transfemoral transcatheter aortic valvereplacement in the PARTNER-I trial: success andsafety. Catheter Cardiovasc Interv 2016;87:165–75.

15. Suri RM, Minha S, Alli O, et al. Learning curvesfor transapical transcatheter aortic valve replace-ment in the PARTNER-I trial: technical perfor-mance, success, and safety. J Thorac CardiovascSurg 2016;152:773–80.e14.

16. Gurvitch R, Tay EL, Wijesinghe N, et al.Transcatheter aortic valve implantation: lessonsfrom the learning curve of the first 270 high-riskpatients. Catheter Cardiovasc Interv 2011;78:977–84.

17. Alli OO, Booker JD, Lennon RJ, Greason KL,Rihal CS, Holmes DR Jr. Transcatheter aortic valveimplantation: assessing the learning curve. J AmColl Cardiol Intv 2011;5:72–9.

18. Kappetein AP, Head SJ, Genereux P, et al.Updated standardized endpoint definitions fortranscatheter aortic valve implantation: the ValveAcademic Research Consortium-2 consensusdocument. J Am Coll Cardiol 2012;60:1438–54.

19. Elze MC, Gregson J, Baber U, et al. Comparisonof propensity score methods and covariateadjustment: evaluation in 4 cardiovascular studies.J Am Coll Cardiol 2017;69:345–57.

20. Holmes DR Jr., Brennan JM, Rumsfeld JS,et al. Clinical outcomes at 1 year following trans-catheter aortic valve replacement. JAMA 2015;313:1019–28.

21. Holmes DR Jr., Nishimura RA, Grover FL, et al.,STS/ACC TVT Registry. Annual outcomes withtranscatheter valve therapy: from the STS/ACCTVT registry. J Am Coll Cardiol 2015;66:2813–23.

22. Le Morvan P, Stock B. Medical learning curvesand the Kantian ideal. J Med Ethics 2005;31:513–8.

23. Bridgewater B, Grayson AD, Au J, et al.Improving mortality of coronary surgery over firstfour years of independent practice: retrospectiveexamination of prospectively collected data from15 surgeons. BMJ 2004;329:421.

24. Hess CN, Peterson ED, Neely ML, et al.The learning curve for transradial percutaneouscoronary intervention among operators in theUnited States: a study from the National Cardio-vascular Data Registry. Circulation 2014;129:2277–86.

25. Carroll JD, Vemulapalli S, Dai D, et al. Proce-dural experience for transcatheter aortic valvereplacement and relation to outcomes: the STS/ACC TVT registry. J Am Coll Cardiol 2017;70:29–41.

26. Auffret V, Lefevre T, Van Belle E, et al.,FRANCE TAVI Investigators. Temporal trends intranscatheter aortic valve replacement in France:FRANCE 2 to FRANCE TAVI. J Am Coll Cardiol2017;70:42–55.

27. Khera S, Kolte D, Gupta T, et al. Associationbetween hospital volume and 30-day read-missions following transcatheter aortic valvereplacement. JAMA Cardiol 2017;2:732–41.

28. McGrath PD, Wennberg DE, Dickens JD Jr.,et al. Relation between operator and hospitalvolume and outcomes following percutaneouscoronary interventions in the era of the coronarystent. JAMA 2000;284:3139–44.

KEY WORDS learning curve, transcatheteraortic valve replacement, volume-outcomerelationship

APPENDIX For a supplemental figure, pleasesee the online version of this paper.

Go to http://www.acc.org/jacc-journals-cme totake the CME/MOC quizfor this article.