CABG: Coronary Artery Bypass Graft; CMS: Centers of Medicare and Medicaid Services; FY: Fiscal Year; ICD-9-CM: International Classification of Diseases, 9th Edition, Clinical Modification; MB: Medicare Beneficiaries; MedPAR: Medicare Provider Analysis and Review; SAS: Statistical Analysis System; SAVR: Surgical Aortic Valve Replacement; TAVR: Transcatheter Aortic Valve Replacement

Symptomatic severe aortic stenosis necessitates mechanical intervention, as medically managed symptomatic aortic valve disease is associated with a high mortality [1-3]. In an aging United States population, the utilization of aortic valve procedures to manage aortic stenosis is expected to rise [1,4,5]. Historically, management of symptomatic aortic valve disease required surgical aortic valve replacement (SAVR). Advanced age, among other factors, contributes to a high-risk profile that impacts the referral and performance of SAVR in certain subgroups [4,6,7]. Transcatheter aortic valve replacement (TAVR) has been performed in patients deemed ineligible for surgery (2012), or as an alternative for patients with intermediate or greater surgical risk (2016), and lastly for low-risk patients (2019). First approved for commercial use in the United States in fiscal year (FY) - 2012, TAVR has demonstrated similar clinical outcomes to SAVR [6,8-13].

This report presents trends in isolated aortic valve procedures in Medicare beneficiaries (MBs) from FY-2016 to FY2020 and extends a previous report for the period FY-2009 to FY2015 [14]. Annual volumes for isolated aortic valve procedures including SAVR and TAVR, are detailed for FY-2016 through FY-2020. Overall mortality rates in-hospital, and 30 and 90- days post-discharge; along with nine selected adverse events in aggregate and by aortic valve procedure type are reported.

This retrospective study used the annual fiscal year (October 1 to September 30) version of the, Medicare Provider Analysis and Review (MedPAR) files, from 2016 to 2020. The MedPAR file, maintained by the Centers for Medicare and Medicaid Services (CMS), contains all inpatient claims submitted by hospitals without patient identifiers. For each hospitalization, the MedPAR record includes selected patient information, International Classification of Diseases, 10th Edition, Clinical Modification (ICD10-CM) diagnoses and procedure codes, discharge status, and days from admission to death. In addition, the MedPAR files contain up to 25 diagnosis and 25 procedure codes per admission.

The study population consisted of MBs undergoing an isolated aortic valve procedure in a U.S. hospital during the study period. For each fiscal year, the MedPAR file was searched for all hospital admissions with the following procedures: 1) aortic valve replacement with a tissue valve; 2) aortic valve replacement with other material or 3) TAVR. A total of 448,178 Medicare beneficiaries underwent one of the three aortic valve procedures during the study period. Of these, 95,796 were excluded due to performance of concomitant procedures including 78,425 coronary artery bypass graft operations; 10,280 non-aortic valve replacements; 7,081 non-aortic valve repairs, and 10 non-aortic valve revisions (Figure 1). The final study population included 352,382 MBs undergoing an isolated aortic valve procedure. Appendix A lists the procedure codes used for each of the inclusion and exclusion criteria.

To report trends, each MB’s isolated aortic valve procedure was classified according to type of valve utilized and approach. All aortic valves include all MBs in the study population, including those who received more than one type of aortic valve procedure during their study hospitalization. For each aortic valve procedure category, the annual trend for the number of MBs undergoing that procedure, the number of procedures per 100,000 MBs, and the procedure share of all aortic valve procedures during the associated FY were reported. Demographic variables of interest included age (<65, 65-69, 70-74, 75-79, ≥80), sex (male), and race (White, Black, Hispanic, and Other).

Outcome measures included three measures of mortality and nine in-hospital adverse events. Medicare beneficiaries who underwent both TAVR and SAVR were included in the allaortic MB category, but were not included in either TAVR or SAVR categories. The total number of hospitalizations for dual procedures was too small to report separately according to our data use agreement. Mortality was assessed at 3 timepoints: in-hospital, 30-days post-discharge and 90-days post discharge. The nine adverse events of interest were: transfusion; vascular complications (bleeding or surgical repairs); infection (postoperative infection or sepsis); post-operative stroke; pulmonary edema or heart failure; post-operative adult respiratory distress syndrome; acute renal failure; new onset hemodialysis; and pacemaker implantation during the hospitalization. The any adverse event rate included these nine events as well as inhospital mortality.

All tables report standard descriptive statistics using counts or proportions. Observed adverse event rates are reported as the proportion of hospitalizations during which a MB experienced an adverse event out of all study hospitalizations in that FY for the appropriate type of aortic valve procedure. The annual number of Medicare enrollees was obtained from the CMS, Medicare Enrollment for selected years and rounded to the nearest thousand beneficiaries [15]. All p-values were assessed with chi-square analysis comparing rates over the study period. All analyses were performed using Statistical Analysis System (SAS) 9.4 (SAS Institute, Cary, North Carolina).

The number of MBs undergoing isolated aortic valve procedures increased annually from 56,958 in FY-2016 to 79,972 in FY-2020 (Table 1). In addition, the number of aortic valve procedures performed per 100,000 MBs increased annually from 102.5 to 132.1 in Fy-2019 before declining to 129.9 in FY-2020. The number of MBs undergoing SAVR declined from 43.5 per 100,000 in FY-2016 to 19.2 per 100,000 MBs by FY2020. However, the number of MBs undergoing TAVR increased from 58.9 per 100,000 beneficiaries in FY-2016 to over 110 per 100,000 by FY-2020. During FY-2020, TAVR accounted for 85.1% of all isolated aortic valve replacement procedures in the Medicare program.

MBs undergoing SAVR were overwhelmingly white (over 85%) and more likely to be male (between 58.6% and 63.0%) (Table 2). Among the SAVR MBs, the proportion of MBs age 75 and older decreased from 42.9% in FY-2016 to 26.8% by FY2020. During FY-2020 over 60% of the SAVR MBs were age 65 to 74 up from slightly less than 50% in FY2016. Over 80% of all MBs undergoing TAVR were age 75 or older in FY2016 compared to approximately 72% in FY-2020. By FY-2020 over 25% of the TAVR MBs were between age 65 and 74. TAVR patients were sicker as denoted by a greater proportion of MBs having each comorbid condition more often than those MBs undergoing SAVR procedures.

During the study period, observed mortality rates for all MBs undergoing an isolated aortic valve procedure declined at each timepoint (in-hospital, 30-days and 90-days post discharge) (Table 3). MBs undergoing SAVR with tissue compared to nontissue valves experienced lower mortality rates each year; except for 30-day and 90-day post discharge mortality rates in FY-2019. In-hospital mortality rates for MBs undergoing TAVR decreased from 1.8% to 0.97% during the study period.

Observed in-hospital event rates for nine adverse events by type of aortic valve procedure are reported in Table 4. Among all MBs undergoing isolated aortic valve replacement, the rate of any adverse event decreased from 37.34% in FY-2016 to 23.77% in FY-2020. The annual proportion of TAVR MBs experiencing an observed adverse event fell to a low of 19.56% by FY-2020, compared to a rate of between 47.44% to 49.07% among SAVR MBs.

This study documents major volume trends among MBs undergoing isolated aortic valve procedures between 2016 and 2020 and demonstrated several key findings. First, isolated aortic valve procedures grew by 8.9%/year—continuing a trend that has been evident since the approval of TAVR in the US in late 2011[14]. Second, this increase was driven primarily by a 20.1% compounded annual growth rate in the number of MBs undergoing isolated TAVR between FY-2016 and FY-2020. During FY-2016, MBs undergoing TAVR accounted for 57.5% of all isolated aortic valve procedures in the Medicare population compared to 85.1% during FY-2020. Over this same time period, the number of MBs undergoing SAVR (with either a tissue or mechanical valve) decreased from 42.4 to 19.2 per 100,000 MB/ year—a decrease of 16.4%/year.

In addition to these marked changes in procedural volumes, our study highlights four trends in observed mortality rates. First, over the 5-year study period, observed in-hospital mortality rates for all procedures decreased from 2.0% to slightly less than 1.3%. In addition, post-discharge mortality rates decreased at both 30-days (from 1.6% to 1.3%) and 90-days (from 3.4% to 3.1%) for all procedures. Second, the observed in-hospital mortality rate among MBs undergoing TAVR decreased from 1.8% in FY-2016 to 0.97% in FY-2020, despite the rapid growth in this procedure. Notably, the cumulative mortality rate through 90-days post-discharge also decreased from 4.2% to 3.1%. Third, among patients undergoing SAVR, mortality rates at all 3 timepoints increased between 2016 and 2020, but there was substantial variation year to year. Further research is warranted to determine if changes in demographic and comorbid conditions among MBs undergoing TAVR impacted the observed mortality rates. Of note, in a study of all aortic valve procedures between 2012 and 2019, Mori and colleagues found similar trends in both unadjusted and risk-adjusted TAVR and SAVR 30-day mortality rates [15].

Finally, we found that rates of adverse events during the initial hospitalization for isolated aortic valve replacement decreased substantially over our study period. However, nearly all of this improvement in in-hospital outcomes was driven by patients undergoing TAVR (31% in FY-2016 to 20% in FY-2020) rather than SAVR (45% in FY-2016 to 47.6% in FY-2020). Among patients undergoing TAVR, the largest contributors to reduced complications were reductions in transfusions, acute renal failure, and permanent pacemaker implantation, although eight of 9 complications actually decreased with vascular complications having a slight increase. Overall, adverse event rates for TAVR in this study appear to be similar to adverse event rates reported in other studies [11,13].

Our study should be interpreted in light of several important limitations. First, all mortality rates and adverse event rates reported in this study are observed rates. Trends in observed mortality and adverse event rates should be interpreted with caution because these rates have not been adjusted for changes in severity of illness among MBs over time. However, the observed rates are of interest in that they report what happened to all MBs undergoing aortic valve procedures in a given year. A second limitation is that the MedPAR dataset lacks echocardiographic data or other relevant clinical data such as mean aortic valve gradients, left ventricular function, or the extent of coronary artery disease. A third limitation is identification of MBs undergoing isolated aortic valve procedures depends on ICD-10- CM procedure codes, which restricts the ability to evaluate the appropriateness of the procedure performed. A related limitation is the identification of study adverse events was dependent on ICD-10-CM coding. This limitation is mitigated by the fact that the MedPAR file contains 25 procedures and diagnosis codes along with present on admission codes to differentiate between diagnoses that existed on admission from those that occurred during the hospitalization.

Between FY 2016 and FY-2020, the number of isolated aortic valve replacement procedures in MBs has continued to increase—driven entirely by increased use of TAVR. Despite this continued growth, both morbidity and mortality after AVR has continued to decrease on a population level. These findings suggest that the introduction and expansion of TAVR has been an important advance in the care of patients with severe aortic stenosis.

Steven Culler, April Simon, Aaron Kugelmass and Phillip Brown have nothing to disclose.

David J Cohen reports a relationship with Edwards Lifesciences Corporation that includes: consulting or advisory and funding grants. David J Cohen reports a relationship with Medtronic Inc that includes: funding grants. David J Cohen reports a relationship with Boston Scientific Corp that includes: consulting or advisory and funding grants. David J Cohen reports a relationship with Abbott Laboratories that includes: consulting or advisory and funding grants.

Matthew Reynolds reports a relationship with Edwards Lifesciences Corporation that includes: consulting or advisory. Matthew R. Reynolds reports a relationship with Medtronic Inc that includes: consulting or advisory.

Marc Katz reports a relationship with Abbott Cardiovascular Structural Heart Division that includes: consulting or advisory. Marc Katz reports a relationship with Boston Scientific Corp that includes: consulting or advisory. Marc Katz reports a relationship with Edwards Lifesciences Corporation that includes: consulting or advisory. Marc Katz reports a relationship with Medtronic Inc that includes: consulting or advisory.

1. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin JP, Guyton RA, et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014: 129: e521-e643.
2. Freeman RV, Otto CM. Spectrum of calcific aortic valve disease. Circulation. 2005; 111: 3316-3326.
3. Carabello BA, Paulus WJ. Aortic stenosis. The Lancet. 2009; 373: 956- 966.
4. Barreto-Filho JA, Wang Y, Dodson JA, Desai MM, Sugeng L, Geirsson A, et al. Trends in aortic valve replacement for elderly patients in the United States, 1999-2011. JAMA. 2013; 310: 2078-2084.
5. Supino PG, Borer JS, Preibisz J, Bornstein A. The epidemiology of valvular heart disease: a growing public health problem. Heart Failure Clinics. 2006; 2: 379-393.
6. Webb JG, Pasupati S, Humphries K, Thompson C, Altwegg L, Moss R, et al. Percutaneous transarterial aortic valve replacement in selected highrisk patients with aortic stenosis. Circulation. 2007; 116: 755-763.
7. Brown JM, O’Brien SM, Wu C, Sikora JAH, Griffith BP, Gammie JS. Isolated aortic valve replacement in North America comprising 108,687 patients in 10 years: changes in risks, valve types, and outcomes in the Society of Thoracic Surgeons National Database. J Thorac Cardiovasc Surg. 2009; 137: 82-90.
8. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson L, et al. for the PARTNER Trial Investigators. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010; 363: 1597-1607.
9. Smith CR, Leon MB, Mack M, Miller DC, Moses JW, Svensson L, et al. for the PARTNER Trial Investigators Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011; 364: 2187-2198.
10. Adams DH, Popma JJ, Reardon MJ, Yakubov SJ, Coselli JS, Deeb G, et al. for the U. S. CoreValve Clinical Investigators. Transcatheter aorticvalve replacement with a self-expanding prosthesis. N Engl J Med. 2014; 370: 1790-1798.
11. Mack MJ, Leon MB, Smith CR, Miller DC, Moses JW, Tuzcu EM, et al. for the PARTNER 1 trial investigators. 5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial. The Lancet. 2015; 385: 2477-2484.
12. Webb JG, Wood DA. Current status of transcatheter aortic valve replacement. J Am Coll Cardiol. 2012; 60: 483-492.
13. Carroll JD, Mack MJ, Vemulapalli S, Herrmann HC, Gleason TG. for the STS/ACC TVT Registry of Transcatheter Aortic Valve Replacement. Ann Thorac Surg: 2021; 111: 701-722.
14. Culler SD, Cohen DJ, Brown PP, Kugelmass AD, Reynolds MR, Ambrose K, et al. Trends in Aortic Valve Replacement Procedures Between 2009 and 2015: Has Transcatheter Aortic Valve Replacement Made a Difference? Ann Thorac Surg. 2018; 105: 1137-1143.
15. Mori M, Gupta A, WangY, Vahl T, Nazif T, Kirtane AJ, et al. Trends in Transcatheter and Surgical Aortic Valve Replacement among Older Adults in the United States. J Am Coll Cardiol. 2021; 78: 2161-2172.
16. The Centers for Disease Control and Prevention (CDC), National Vital Statistics Reports (NVSR), Vol. 66, No. 1: Births: Final Data for 2015.