Background
As TAVR use expands, some patients require later cardiac surgery, including SAVR with TAVR valve explant in certain scenarios. The long-term risks of SAVR after TAVR compared with other cardiac surgeries remain unclear.
Methods
We studied adults in the TriNetX network who underwent TAVR followed by either SAVR or non-SAVR open-heart surgery (OHS) between 2010 and 2023. Propensity-score matching was done on 26 clinical and demographic factors. We compared outcomes at 3 and 5 years.
Results
After matching, 132 patients were included in each group. The groups were well-balanced at baseline. At 3 years, mortality was similar (18.9% SAVR vs 22.0% OHS; HR: 0.83, 95% CI: 0.46 to 1.51). At 5 years, mortality remained comparable (20.5% vs 24.2%; OR: 0.80, 95% CI 0.45 to 1.44). Rates of stroke, acute coronary syndrome, heart failure hospitalization, major bleeding, new atrial fibrillation, and renal failure were also similar.
Conclusions
SAVR after TAVR was associated with similar long-term outcomes compared with OHS. These findings suggest that risks may be more related to patient complexity than to the valve explant itself.
As transcatheter aortic valve replacement (TAVR) expands into broader patient populations, understanding the implications of subsequent cardiac surgeries is increasingly important. ,,, While valve-in-valve TAVR is often preferred, surgical aortic valve replacement (SAVR) is required in certain scenarios such as prosthetic valve endocarditis, severe paravalvular leak not amenable to transcatheter repair, structural valve degeneration unsuitable for valve-in-valve expansion due to annular size constraints or coronary obstruction risk, and device malposition. ,,
Prior registry-based studies suggest that patients undergoing SAVR after prior TAVR may experience higher operative mortality compared to those undergoing redo SAVR after prior surgical prosthesis. , The perceived complexity and risks of explanting TAVR valves have raised concerns, particularly regarding long-term outcomes. , However, it remains unclear to what extent this risk reflects the complexity of TAVR explant itself versus the underlying patient comorbidities and cardiac surgery. Given the projected increase in TAVR utilization and subsequent need for reintervention, clarifying these long-term outcomes is critical for patient counseling, surgical planning, and shared decision-making in the heart team setting. ,, This study compares outcomes between SAVR and non-SAVR open-heart surgeries (OHS)after TAVR. We hypothesized that, when accounting for comorbidities, SAVR after TAVR would not carry greater long-term risk than other forms of cardiac surgery in this population.
Methods
We conducted a retrospective cohort study using the TriNetX global research network, a federated network of deidentified electronic health records from more than 100 healthcare organizations. We identified adult patients (≥18 years) who underwent TAVR followed by either SAVR or OHS between January 1, 2010, and December 31, 2023. Procedures were defined using standardized codes from the Current Procedural Terminology (CPT), the International Classification of Diseases, 10th Revision, Procedure Coding System (ICD-10-PCS), and the Systematized Nomenclature of Medicine (SNOMED). The SAVR cohort included patients who underwent open surgical replacement of the aortic valve with a mechanical, bioprosthetic, stentless, or allograft valve. The OHS cohort comprised patients who underwent other cardiac operations, including coronary artery bypass grafting (CABG), mitral or tricuspid valve surgery, and aortic (excluding aortic root) or septal repairs, without concurrent SAVR. Procedures were identified using a predefined list of codes ( Supplemental Table 1 ). All patients were required to have undergone TAVR at least 1 day before cardiac surgery.
Propensity-score matching was performed in a 1:1 fashion using logistic regression with a caliper of 0.1 pooled standard deviations, without replacement. Matching variables included age, sex, race, body mass index (BMI), left ventricular ejection fraction (LVEF), coronary artery disease, chronic kidney disease, diabetes mellitus, hypertension, heart failure, chronic lung disease, prior myocardial infarction, prior percutaneous coronary intervention, prior stroke or transient ischemic attack, dialysis dependence, and baseline use of statins, aspirin, P2Y12 inhibitors, beta-blockers, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers (ACEi/ARBs), and loop diuretics. Balance between groups was confirmed by standardized mean differences <0.1 for all covariates ( Figure S1 , Table 1 ).
Table 1
Baseline characteristics of propensity-matched patients undergoing surgical aortic valve replacement (SAVR) versus non-SAVR cardiac surgery following prior TAVR
| Variable | SAVR after TAVR ( N = 132) | OHS after TAVR (N = 132) | p-value | Std diff. |
|---|---|---|---|---|
| Age at Index (years) | 72.0 ± 10.4 | 72.2 ± 10.6 | 0.855 | 0.022 |
| Female sex, n (%) | 52 (39.4) | 54 (40.9) | 0.802 | 0.031 |
| White, n (%) | 101 (76.5) | 97 (73.5) | 0.570 | 0.070 |
| Black, n (%) | 13 (9.8) | 14 (10.6) | 0.839 | 0.025 |
| Hispanic/Latino, n (%) | <10 (<7.6) | <10 (<7.6) | 1.000 | <0.001 |
| Diabetes mellitus, n (%) | 63 (47.7) | 63 (47.7) | 1.000 | <0.001 |
| Chronic kidney disease, n (%) | 62 (47.0) | 64 (48.5) | 0.805 | 0.030 |
| Heart failure, n (%) | 107 (81.1) | 111 (84.1) | 0.516 | 0.080 |
| Prior MI (any STEMI/NSTEMI), n (%) | 59 (44.7) | 65 (49.2) | 0.462 | 0.090 |
| Prior stroke/TIA, n (%) | 22 (16.7) | 21 (15.9) | 0.868 | 0.021 |
| Hypertension, n (%) | 99 (75.0) | 113 (85.6) | 0.030 | 0.269 |
| Hyperlipidemia, n (%) | 109 (82.6) | 98 (74.2) | 0.100 | 0.204 |
| COPD, n (%) | 28 (21.2) | 35 (26.5) | 0.312 | 0.125 |
| Prior PCI, n (%) | 15 (11.4) | 13 (9.8) | 0.689 | 0.049 |
| Dialysis, n (%) | 10 (7.6) | 11 (8.3) | 0.820 | 0.028 |
| LVEF (%) | 56.1 ± 13.6 | 53.9 ± 16.1 | 0.531 | 0.148 |
| BMI (kg/m²) | 30.4 ± 6.6 | 28.1 ± 6.5 | 0.008 | 0.351 |
| Statin use, n (%) | 109 (82.6) | 122 (92.4) | 0.016 | 0.301 |
| Aspirin use, n (%) | 124 (93.9) | 119 (90.2) | 0.255 | 0.140 |
| P2Y12 inhibitor (clopidogrel), n (%) | 90 (68.2) | 90 (68.2) | 1.000 | <0.001 |
| Beta-blocker use, n (%) | 119 (90.2) | 120 (90.9) | 0.834 | 0.026 |
| ACEi or ARB use, n (%) | 110 (83.3) | 97 (73.5) | 0.049 | 0.242 |
| Loop diuretic use, n (%) | 99 (75.0) | 99 (75.0) | 1.000 | <0.001 |
Abbreviations: SAVR = surgical aortic valve replacement; TAVR = transcatheter aortic valve replacement; OHS = open-heart surgery.
This table summarizes demographic, clinical, and procedural characteristics of patients who underwent SAVR ( n = 132) or non-SAVR cardiac surgery ( n = 132) after prior transcatheter aortic valve replacement (TAVR), following propensity-score matching. Values are presented as mean ± standard deviation or as counts with percentages. There were no statistically significant differences between groups across baseline variables, indicating successful matching.
Outcomes were assessed over 5 years, starting 1 day after the index surgery. The primary outcome was all-cause mortality. Secondary outcomes included acute coronary syndrome, stroke, heart failure hospitalization, major bleeding, new-onset renal failure, and new-onset atrial fibrillation. Rare outcomes, such as liver failure and pulmonary embolism, were also recorded. Event rates were compared using Kaplan-Meier survival analysis, hazard ratios (HR) from Cox models, odds ratios (OR), and log-rank tests. Analyses were performed within TriNetX. The study was exempt from institutional review board approval due to the use of deidentified data.
