Mortality and morbidity benefits of surgical versus transcatheter aortic valve replacement in patients ≤60 years: Findings from a national cohort Authors: Jessan A Jishu, Yara Menassa, Patrick Kearns, Mahmoud Omar, Cameron Galic, Soubhi Tahhan, Shriya Chunduri, Aabha Divya, Joseph M Obeid, Jamil F Borgi Published: 2026 ABSTRACT AbstractObjectiveTo compare outcomes of surgical aortic valve replacement (SAVR) with transcatheter aortic valve replacement (TAVR) in patients aged ≤60 years throughout the United States.MethodsPatients aged ≤60 years with aortic stenosis undergoing either first-time SAVR or TAVR between January 2006 and December 2025 were identified from TriNetX, an administrative real-world database. The primary outcome was all-cause mortality over 5 years. Cardiac-related outcomes were also assessed. Results were analyzed after propensity matching for age, sex, presence of bicuspid aortic valve, lower respiratory disease, liver fibrosis/cirrhosis, liver failure, kidney disease, dialysis dependence, and previous cardiac surgery.ResultsAmong 9841 patients, 7896 underwent SAVR, whereas 1945 underwent TAVR. Matching resulted in 1666 patients per cohort. Before matching, patients who underwent SAVR were younger, more likely to have bicuspid aortic valves, and had lower rates of lower respiratory disease, liver or kidney disease, dialysis use, and previous cardiac surgery. After matching, SAVR was associated with comparable mortality risk at 1 month but lower mortality risks at 1 year (odds ratio, 0.69; 95% CI, 0.53-0.92) and 5 years (hazard ratio [HR], 0.58; 95% CI, 0.48-0.69). By the end of follow-up, SAVR was associated with decreased risks of acute myocardial infarction (HR, 0.65; 95% CI, 0.49-0.86), endocarditis (HR, 0.55; 95% CI, 0.37-0.82), complete atrioventricular block (HR, 0.45; 95% CI, 0.30-0.66), pacemaker implantation (HR, 0.61; 95% CI, 0.45-0.83), paravalvular leak (HR, 0.39; 95% CI, 0.26-0.87), and reintervention (HR, 0.54; 95%, CI 0.38-0.78).ConclusionsAmong patients aged ≤60 years, SAVR is associated with lower mortality and morbidity rates over TAVR.Key Words: TriNetX, propensity, valve repair, structural, TAVR, SAVR Survival benefits for patients aged ≤60 years undergoing SAVR over TAVR. Central Message SAVR is associated with better survival and clinical outcomes in patients ≤60 years of age compared with TAVR. Perspective Current guidelines recommend SAVR for patients ≤65 years of age with aortic stenosis, yet there is an increasing number of patients receiving TAVR. This study incorporated a nationwide analysis and found that SAVR is associated with similar short-term but improved long-term outcomes in age ≤60 years, highlighting the need for prospective studies to continue assessing age-specific outcomes. Transcatheter aortic valve replacement (TAVR) has transformed the management of aortic stenosis (AS) by making valve replacement accessible to a broader population of patients.1 Indeed, the use of TAVR has increased over the years, from 2.6% in 2012 to 12.5% in 2016 and eventually surpassing that of surgical aortic valve replacement (SAVR) in 2019.2,3 The expansion of TAVR into lower-risk populations was supported by the findings of the PARTNER 3 and Medtronic Evolut Low Risk trials in 2019, which demonstrated favorable short- and intermediate-term outcomes in low-risk patients.4, 5, 6 However, this has prompted scrutiny of how age relates to procedural risks and long-term benefits. Although older patients may clearly benefit from the minimally invasive nature of TAVR, younger patients present with distinct considerations, including valve durability, long-term survival, and the need for future reinterventions over a longer life expectancy.7 Current guidelines recommend SAVR for patients younger than 65 years, yet an increasing number of younger patients are undergoing TAVR with uncertain long-term results.8,9 Given this lack of clarity and growing real-world adoption of TAVR among younger individuals, this study uses a large-scale database representative of practices in the United States to confirm the findings from these previous studies by assessing short- and long-term outcomes of SAVR versus TAVR in patients younger than 60 years, a population in whom long-term durability and reintervention risk are of particular concern. Methods Data Source This retrospective cohort study used data from the TriNetX United States Collaborative Network, an administrative real-world database containing electronic health records from 69 participating health care organizations. The platform provides longitudinal clinical data, including diagnostic codes using International Classification of Diseases, 10th Revision, Clinical Modification, procedure codes using Current Procedural Terminology, and laboratory values. All data are collected from predominantly insured patients who receive medical care from various health care organizations, including community hospitals and academic centers.10 Because patient information within TriNetX is deidentified, approval from the institutional review board was not required. Study Population Adult patients (age ≥18 years) diagnosed with nonrheumatic AS who underwent either first-time isolated SAVR or first-time TAVR when they were younger than the age of 60 years were identified between January 1, 2006, and December 31, 2025. The procedure served as the index event; hence, all outcomes captured were relative to the date of the procedure. Within the SAVR cohort, patients who underwent previous TAVR or concomitant surgical procedures on the other heart valves or grafting for coronary artery bypass were excluded to ensure isolated aortic valve procedures. Within the TAVR cohort, patients who underwent previous SAVR were excluded. Furthermore, patients who had incomplete outcome data or outcomes before the index date were excluded. Diagnostic and procedural codes describing inclusion and exclusion criteria can be found in Table E1. Study Outcomes The primary outcome was all-cause mortality. The secondary outcomes described morbidity including acute myocardial infarction, stroke, endocarditis, complete atrioventricular block, pacemaker implantation, paravalvular leak, and reintervention with either SAVR or TAVR. All patients were followed for up to 5 years. If the outcomes did not occur during the follow-up period, patients were censored at their last recorded visits. Detailed definitions and codes for the study outcomes can be found in Table E2. Statistical Analysis All statistical analyses were performed using TriNetX and validated using R, version 4.5.2 (R Foundation for Statistical Computing). Categorical variables were compared using 2-sided χ2 testing whereas continuous variables were assessed with Wilcoxon rank-sum tests. Short-term outcomes at 1 month and 1 year were analyzed using logistic regression with results reported as odds ratios and 95% CIs. Long-term outcomes at 5 years were evaluated using Cox proportional hazards models, with effect estimates expressed as hazard ratios (HRs) and corresponding 95% CIs. Survival curves were generated using Kaplan-Meier methods and differences between cohorts were assessed using the log-rank test. Trend analysis of annual procedures performed were assessed using the Cochran-Armitage test. All study outcomes were assessed before and after 1:1 propensity matching, with a nearest-neighbor approach. Covariates included age, sex, and comorbidities including presence of bicuspid aortic valve, lower respiratory disease (including bronchitis, bronchiectasis, emphysema, asthma, or other chronic obstructive pulmonary disease), liver fibrosis/cirrhosis, liver failure, chronic kidney disease, dialysis dependence, and prior cardiac surgery. Diagnostic and procedural codes describing variables used for propensity matching are shown in Table E3. Covariate balance after matching was evaluated using standardized mean differences, with values less than 0.1 indicating appropriate balance. Results Baseline Characteristics Before matching, a total of 9841 patients were identified, including 7896 (80.2%) who underwent SAVR and 1945 (19.8) who underwent TAVR. After matching, there were 1666 patients per cohort (Table 1). Table 1Baseline characteristics of the study populationCharacteristicsBefore matchingAfter matchingSAVR (n = 7896)TAVR (n = 1945)P valueSMDSAVR (n = 1666)TAVR (n = 1666)P valueSMDAge at index, y ± SD48.7 ± 12.454.6 ± 6.9<.010.5954.2 ± 7.054.3 ± 7.3.960.01Female2439 (30.9)725 (37.3)<.010.13639 (38.4)623 (37.4).570.02Black or African American549 (7.0)216 (11.1)<.010.14173 (10.4)154 (9.2).270.04White6347 (80.4)1525 (78.4).040.051310 (78.6)1327 (79.7).470.03Not Hispanic or Latino6403 (81.1)1547 (79.5).090.041378 (82.7)1321 (79.3).010.09Hispanic or Latino492 (6.2)128 (6.6).580.0194 (5.6)115 (6.9).130.05Unknown ethnicity990 (12.5)270 (13.9).120.04194 (11.6)230 (13.8).060.07Body mass index, kg/m2 ±SD30.7 ± 7.532.3 ± 9.4<.010.1931.7 ± 7.832.4 ± 9.5.060.08Bicuspid aortic valve2208 (28.0)251 (12.9)<.010.37232 (13.9)238 (14.3).770.01Lower respiratory disease1625 (20.6)747 (38.4)<.010.39644 (38.7)594 (35.7).070.06Liver fibrosis/cirrhosis128 (1.6)219 (11.3)<.010.39109 (6.5)118 (7.1).540.02Liver failure116 (1.5)124 (6.4)<.010.2568 (4.1)70 (4.2).860.01Chronic kidney disease748 (9.5)657 (33.8)<.010.59434 (26.1)429 (25.8).840.01Dialysis dependence129 (1.6)218 (11.2)<.010.3995 (5.7)99 (5.9).770.01Previous surgical history on the heart or pericardium204 (2.6)241 (12.4)<.010.37120 (7.2)139 (8.3).220.04Data are reported as count (percentage). Two-sided χ2 or Student t test was used. Bold P values indicate statistical significance. SAVR, Surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; SMD, standardized mean difference; SD, standard deviation. Baseline characteristics of the study population Data are reported as count (percentage). Two-sided χ2 or Student t test was used. Bold P values indicate statistical significance. SAVR, Surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; SMD, standardized mean difference; SD, standard deviation. In the unmatched analysis, patients undergoing SAVR were significantly younger and less likely to be female and Black or African American compared with those undergoing TAVR. The patients in the SAVR cohort were more likely to have bicuspid aortic valves (28.0% vs 12.9%, P < .01) and demonstrated substantially lower burden of comorbidities, including lower respiratory disease (20.6% vs 38.4%), liver fibrosis/cirrhosis (1.6% vs 11.3%), liver failure (1.5% vs 6.4%), chronic kidney disease (9.5% vs 33.8%), dialysis dependence (1.6% vs 11.2%), and previous cardiac surgery (2.6% vs 12.4%) (all P < .01). In addition, patients undergoing SAVR had a lower mean body mass index (30.7 ± 7.5 vs 32.3 ± 9.4 kg/m2, P < .01). After matching, baseline characteristics were balanced between cohorts. Although there was a very minor difference in ethnicity, there were no significant differences in demographics or major comorbidities. Temporal Trends Before 2013, aortic valve replacement in the present study consisted exclusively of SAVR. In this early period, annual case counts of SAVR gradually increased and ranged from 47 procedures in 2005 to 255 procedures in 2012. TAVRs were first performed in 2013 and was followed by rapidly increasing use over subsequent years. Both approaches experienced a decrease in use in 2020 but gradually increased until the end date of the study. After 2020, TAVR accounted for between 25% and 32% of annual aortic valve replacement procedures (Figure 1). The Cochran-Armitage test for trend demonstrated a significant increase in TAVR use relative to that of SAVR through the duration of follow-up (P < .01). Annual trends in the occurrences of SAVR and TAVR. SAVR, Surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement. Follow-up Time Before matching, the median follow-up times for SAVR and TAVR cohorts were 1332 days (interquartile range [IQR], 1404 days) and 929 days (IQR, 1367 days), respectively. After matching, the median follow-up times for SAVR and TAVR cohorts were 1166 days (IQR, 1439 days) and 1000 days (IQR, 1415 days), respectively. One-Month Outcomes After matching, mortality rates at 1 month were comparable between SAVR and TAVR cohorts (2.0% vs 1.5%; OR, 1.37; 95% CI, 0.81-2.30). No significant differences were observed for other secondary outcomes (Table 2). Table 2Comparative 1-month outcomes in patients undergoing SAVR versus TAVROutcomeBefore matchingAfter matchingSAVR (n = 7896)TAVR (n = 1945)OR95% CISAVR (n = 1666)TAVR (n = 1666)OR95% CIMortality71 (0.9)34 (1.7)0.480.32-0.7334 (2.0)25 (1.5)1.370.81-2.30Acute MI112 (1.4)16 (0.8)1.360.81-2.3121 (1.3)13 (0.8)1.480.74-2.96Stroke44 (0.6)13 (0.7)0.740.40-1.38∗∗∗∗Endocarditis∗∗∗∗∗∗∗∗Complete AV block63 (0.8)27 (1.4)0.520.33-0.8216 (1.0)26 (1.6)0.600.32-1.11Pacemaker implantation246 (3.1)70 (3.6)0.800.61-1.0554 (3.2)70 (4.2)0.760.53-1.09Paravalvular leak22 (0.3)20 (1.0)0.240.13-0.45∗∗∗∗Reintervention59 (0.7)34 (1.7)0.400.26-0.6117 (1.0)29 (1.7)0.580.32-1.06Data are reported as percentage. Two-sided χ2 or Student t test was used. Bold values indicate statistical significance. SAVR, Surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; OR, odds ratio; MI, myocardial infarction; AV, atrioventricular.∗Count too small (<10) to calculate. Comparative 1-month outcomes in patients undergoing SAVR versus TAVR Data are reported as percentage. Two-sided χ2 or Student t test was used. Bold values indicate statistical significance. SAVR, Surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; OR, odds ratio; MI, myocardial infarction; AV, atrioventricular. Count too small (<10) to calculate. One-Year Outcomes After matching, mortality rates at 1 year were lower among patients undergoing SAVR when compared with those undergoing TAVR (5.5% vs 7.7%; OR, 0.69; 95% CI, 0.53-0.92). SAVR was also associated with lower rates of complete atrioventricular block (OR, 0.56; 95% CI, 0.34-0.92), pacemaker implantation (OR, 0.65; 95% CI, 0.46-0.91), paravalvular leak (OR, 0.29; 95% CI, 0.16-0.50), and reintervention (OR, 0.55; 95% CI, 0.33-0.91), whereas other secondary outcomes were similar between cohorts (Table 3). Table 3Comparative 1-year outcomes in patients undergoing SAVR versus TAVROutcomeBefore matchingAfter matchingSAVR (n = 7896)TAVR (n = 1945)OR95% CISAVR (n = 1666)TAVR (n = 1666)OR95% CIMortality189 (2.4)166 (8.5)0.250.20-0.3191 (5.5)128 (7.7)0.690.53-0.92Acute MI183 (2.3)56 (2.9)0.630.46-0.8545 (2.7)52 (3.10)0.790.53-1.18Stroke126 (1.6)57 (2.9)0.480.35-0.6635 (2.1)48 (2.9)0.720.46-1.12Endocarditis92 (1.2)42 (2.2)0.530.36-0.7620 (1.2)32 (1.9)0.660.38-1.17Complete AV block109 (1.4)47 (2.4)0.510.36-0.7225 (1.5)43 (2.6)0.560.34-0.92Pacemaker implantation273 (3.5)85 (4.4)0.730.57-0.9358 (3.5)87 (5.2)0.650.46-0.91Paravalvular leak75 (0.9)46 (2.4)0.360.25-0.5216 (1.0)52 (3.1)0.290.16-0.50Reintervention82 (1.0)48 (2.5)0.390.27-0.5624 (1.4)43 (2.6)0.550.33-0.91Data are reported as percentage. Two-sided χ2 or Student t test was used. Bold values indicate statistical significance. SAVR, Surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; OR, odds ratio; MI, myocardial infarction; AV, atrioventricular. Comparative 1-year outcomes in patients undergoing SAVR versus TAVR Data are reported as percentage. Two-sided χ2 or Student t test was used. Bold values indicate statistical significance. SAVR, Surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; OR, odds ratio; MI, myocardial infarction; AV, atrioventricular. Five-Year Outcomes After matching, mortality rates at 5 years remained lower in the SAVR cohort when compared with the TAVR cohort (11.3% vs 18.3%; HR, 0.58; 95% CI, 0.48-0.69) (Figure 2). SAVR was also associated with lower risks of acute myocardial infarction (HR, 0.65; 95% CI, 0.49-0.86), endocarditis (HR, 0.55; 95% CI, 0.37-0.82), complete atrioventricular block (HR, 0.45; 95% CI, 0.30-0.66), pacemaker implantation (HR, 0.61; 95% CI, 0.45-0.83), paravalvular leak (HR, 0.39; 95% CI, 0.26-0.57), and reintervention (HR, 0.54; 95% CI, 0.38-0.78) (Table 4). Kaplan-Meier survival curves between patients undergoing SAVR versus TAVR (A) before matching and (B) after matching. Shaded areas represent 95% CIs. SAVR, Surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement. Table 4Comparative 5-year outcomes in patients undergoing SAVR versus TAVROutcomeBefore matchingAfter matchingSAVR (n = 7896)TAVR (n = 1945)HR95% CISAVR (n = 1666)TAVR (n = 1666)HR95% CIMortality442 (5.6)415 (21.3)0.230.20-0.26189 (11.3)305 (18.3)0.580.48-0.69Acute MI329 (4.2)137 (7.0)0.450.37-0.5588 (5.3)116 (7.0)0.650.49-0.86Stroke331 (4.2)119 (6.1)0.570.46-0.7195 (5.7)101 (6.1)0.880.66-1.16Endocarditis187 (2.4)82 (4.2)0.530.41-0.6838 (2.3)69 (4.1)0.550.37-0.82Complete AV block168 (2.1)93 (4.8)0.390.30-0.5037 (2.2)76 (4.6)0.450.30-0.66Pacemaker implantation322 (4.1)116 (6.0)0.640.52-0.8067 (4.0)105 (6.3)0.610.45-0.83Paravalvular leak143 (1.8)102 (5.2)0.300.23-0.3937 (2.2)87 (5.2)0.390.26-0.57Reintervention166 (2.1)92 (4.7)0.390.30-0.5045 (2.7)77 (4.6)0.540.38-0.78Data are reported as percentage. Two-sided χ2 or Student t test was used. Bold values indicate statistical significance. SAVR, Surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; HR, hazard ratio; MI, myocardial infarction; AV, atrioventricular. Comparative 5-year outcomes in patients undergoing SAVR versus TAVR Data are reported as percentage. Two-sided χ2 or Student t test was used. Bold values indicate statistical significance. SAVR, Surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; HR, hazard ratio; MI, myocardial infarction; AV, atrioventricular. Discussion This study shows SAVR is associated with similar perioperative mortality and lower morbidity and mortality at 1 and 5 years compared with TAVR in patients younger than 60 years. Although current guidelines recommend SAVR in patients younger than 65 years, even younger patients are increasingly receiving TAVR as the result of it being less invasive.11 Current studies have identified age younger than 60 years to be a significant risk factor for accelerated valve deterioration, with supporting literature finding reintervention rates to be greatest for them.12,13 This highlights the importance of valve type selection in younger patients: mechanical valves are durable but associated with an increased risk of bleeding because of the lifelong need for anticoagulation, whereas bioprosthetic valves avoid this requirement but carry a greater risk of valve deterioration. Accordingly, mechanical aortic valve replacement has been the preferable choice for younger individuals, although valve type could not be evaluated in the present study.14 Moreover, Marsh and colleagues15 found that patients younger than 60 years are aware of the greater risks associated with TAVR in their age cohort yet opt for it anyway. In another study, Gad and colleagues16 compared 30-day outcomes after SAVR versus TAVR using this age threshold and found similar mortality rates but fewer complications after TAVR. As such, comparing SAVR and TAVR in patients younger than 60 years can expand upon these findings because this population faces a longer remaining life expectancy and greater cumulative exposure to valve-related issues. In this study, there were no significant differences in perioperative morbidity and mortality between SAVR and TAVR, likely reflecting the low baseline comorbidity burden of this young population as well as improvements in perioperative care through implementation of standardized Enhanced Recovery After Surgery protocols.17,18 Ando and colleagues19 performed a nationwide analysis to compare 1-month outcomes between SAVR and TAVR in patients younger than 65 years and found similar risk of mortality between procedures, although they noted greater risk of pacemaker implantation after TAVR. Although this was not observed in the present study as a short-term outcome, high-degree atrioventricular block is an established conduction abnormality after TAVR and can often present after discharge.20 With their analysis being performed before 2015, it is likely that the common practice was to hold a conservative approach with such arrythmias and proceed promptly with pacemaker implantation, whereas this analysis may reflect liberal practices in waiting first.21 This study also showed that SAVR is associated with improved survival and lower morbidity compared with TAVR at the 1- and 5-year marks, consistent with previous studies.22,23 Beyond survival, multiple studies have highlighted differences in complication profiles between SAVR and TAVR that are particularly relevant to younger patients. For example, permanent pacemaker implantation and conduction disturbances have been shown to adversely affect long-term functional status and survival in lower-risk cohorts.24 The present study additionally found lower rates of reintervention associated with SAVR than with TAVR, although it was unable to distinguish redo procedures from cross-procedure interventions. Nevertheless, the favorable reintervention rate associated with SAVR echoes findings from previous studies.25,26 This carries an important survival implication, given rising national trends in post-TAVR operations, particularly with SAVR after initial TAVR or redo-TAVR being both associated with greater risks of mortality when compared to redo-SAVR.27,28 This suggests that patients with appropriate structural cardiac anatomy may benefit from receiving SAVR as the initial treatment for AS. The better outcomes observed with SAVR likely reflect a combination of procedural and patient-specific factors. For instance, surgical valve replacement allows for complete decalcification, which can contribute to lower risk of paravalvular leaks and ultimately reintervention. Patients with coronary artery disease may also benefit from surgical intervention with the option of concomitant coronary revascularization, although the present study analyzed isolated SAVR and did not evaluate that particularly interesting cohort.29 Furthermore, transcatheter valves exert pressure on nearby conduction tissue and left ventricular outflow tract, contributing to greater rates of atrioventricular block and permanent pacing.30 From an economic perspective, previous studies have associated TAVR with greater hospital costs even after considering shorter length of stay, although this may be attributed to the greater comorbidity burden of older patients who more commonly undergo TAVR.31 Overall, SAVR may bring both patient-centered health advantages and system-wide economic benefits in these young patients. This study also revealed substantial reductions in SAVRs and TAVRs performed in 2020, which can be attributed to COVID-19. Reductions in these surgeries as the result of COVID-19 have been previously reported worldwide.32, 33, 34 Gradual increases have followed, likely from elective cases accumulating since this pandemic. This analysis is among the first studies to leverage nationwide data to compare outcomes of SAVR versus TAVR in younger patients. There is one other study in the literature that used TriNetX to compare outcomes of these approaches, with that one focusing instead on the influence of bicuspid valves.35 Most previous investigations focus on populations defined by the aforementioned guidelines, including age younger than 65 years. This large-scale study captured patterns throughout the United States, increasing generalizability beyond single-institution experiences. However, these findings should be cautiously interpreted as statistically significant differences that persisted after matching may not always translate into clinically meaningful differences. This study has limitations inherent to the use of an administrative database. These include miscoding and absence of procedural variables such as access site, valve type, and surgical technique. Some covariates, such as functional status or mobility, were unavailable and could not be accounted for, which may contribute to residual confounding. Regarding outcomes, there was no information on the extent of paravalvular leak or percutaneous closure of the leak which has been shown to be an alternative to surgical reintervention in high-risk patients.36 In addition, cause-specific mortality and echocardiographic measures were unavailable, which can limit interpretation of valve performance. Surgeon and institutional volume, variables known to influence patient outcomes, were also not captured.37 Conclusions In this analysis of patients younger than 60 years, SAVR was associated with similar perioperative outcomes, lower long-term mortality, and fewer cardiac-related complications compared with TAVR, with these differences persisting after matching. This supports SAVR as the preferred initial aortic valve replacement strategy in this age group, pending confirmation from prospective studies. Webcast You can watch a Webcast of this AATS meeting presentation by going to: https://www.aats.org/resources/mortality-and-morbidity-benefi-12917. Conflict of Interest Statement The authors reported no conflicts of interest. The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest. Appendix E1 Table E1Diagnostic and procedural codes used for study inclusion and exclusionCategoryCode(s)DefinitionInclusionICD-10-CM I35.0, I35.2Nonrheumatic aortic stenosisCPT 1006147, 1006151, 1029693, 1035167Surgical aortic valve replacementCPT 1021150Transcatheter aortic valve replacementExclusionCPT 1006159Surgical procedures on the mitral valveCPT 1006174Surgical procedures on the pulmonary valveCPT 1006167Surgical procedures on the tricuspid valveCPT 1006199, 1006207, 1006216Coronary artery bypass graft surgeryICD-10-CM, International Classification of Diseases, 10th Revision, Clinical Modification; CPT, Current Procedural Terminology. Diagnostic and procedural codes used for study inclusion and exclusion ICD-10-CM, International Classification of Diseases, 10th Revision, Clinical Modification; CPT, Current Procedural Terminology. Table E2Definitions of study outcomesCode(s)Definitionn/aAll-cause mortalityICD-10-CM I21Acute myocardial infarctionICD-10-CM I63StrokeICD-10-CM I33EndocarditisICD-10-CM I44.2Complete atrioventricular blockCPT 1006079Pacemaker implantationICD-10-CM T82.03Paravalvular leakCPT 1006147, 1006151, 1029693, 1035167, 1021150Reintervention with another surgical/transcatheter aortic valve replacementn/a, Not available; ICD-10-CM, International Classification of Diseases, 10th Revision, Clinical Modification; CPT, Current Procedural Terminology. Definitions of study outcomes n/a, Not available; ICD-10-CM, International Classification of Diseases, 10th Revision, Clinical Modification; CPT, Current Procedural Terminology. Table E3Variables used for propensity-score matchingCriteriaCode(s)DefinitionDemographicsn/aAge at indexn/aSexMedical historyICD-10-CM Q23.81Presence of bicuspid aortic valveICD-10-CM J40-J4ALower respiratory disease (bronchitis, bronchiectasis, emphysema, asthma, or other chronic obstructive pulmonary disease)ICD-10-CM K72Liver failureICD-10-CM N18Chronic kidney diseaseICD-10-CM 1012740Dependence on dialysisCPT 1006057Previous cardiac surgeryn/a, Not available; ICD-10-CM, International Classification of Diseases, 10th Revision, Clinical Modification; CPT, Current Procedural Terminology. Variables used for propensity-score matching n/a, Not available; ICD-10-CM, International Classification of Diseases, 10th Revision, Clinical Modification; CPT, Current Procedural Terminology. References