Cost-Effectiveness of Valve-in-Valve Transcatheter Aortic Valve Replacement Versus Redo-Surgical Aortic Valve Replacement for Degenerated Bioprosthetic Aortic Valve Authors: Elina E Pliakos, Andrew M Acker, Paul N Fiorilli, Allison L Tsao, Kendall M Lawrence, Nimesh D Desai, Toby Rogers, Jaffar M Khan, Jay S Giri, Ashwin S Nathan Published: 2026 ABSTRACT AbstractBackgroundDegenerated bioprosthetic aortic valves are associated with substantial morbidity, mortality, and health care resource use. Treatment options include redo-surgical aortic valve replacement (redo-SAVR) and valve-in-valve transcatheter aortic valve replacement (ViV-TAVR). ViV-TAVR has been associated with shorter length of stay, but its higher device costs have raised uncertainty about its overall economic value. We constructed a decision-analytic model to compare the cost-effectiveness of ViV-TAVR vs redo-SAVR for the management of degenerated bioprosthetic aortic valves.MethodsCost-effectiveness was evaluated by calculating deaths averted and incremental cost-effectiveness ratios (ICERs). The main outcome was ICER defined as US dollars/deaths averted. Uncertainty was addressed by plotting cost-effectiveness planes and acceptability curves for various willingness-to-pay thresholds.ResultsIn the probabilistic analysis, the mean total cost of ViV-TAVR was $88,829 (95% CI, $88,554-$89,104) with a 2-month survival probability of 0.97 (95% CI, 0.969-0.970). For redo-SAVR, the mean cost was $91,411 (95% CI, $91,163-$91,659) with a 2-month survival probability of 0.96 (95% CI, 0.959-0.961). ViV-TAVR was the dominant strategy, resulting in a mean ICER of −$259,323 per death averted (95% CI, −$304,599 to −$220,679). ViV-TAVR was cost effective in 57% to 58% of simulations for willingness-to-pay thresholds ranging from $0 to $150,000.ConclusionsViV-TAVR is an economically dominant strategy for the management of degenerated bioprosthetic aortic valves, yielding lower costs and marginally improved short-term survival relative to redo-SAVR. However, given a moderate degree of decisional uncertainty, trial-level comparative data and longer-term modeling are required to refine these estimates.Keywords: cost-effectiveness, decision-analytic model, degenerated bioprosthetic valve, incremental cost-effectiveness ratio, redo-surgical aortic valve replacement, structural heart disease, valve-in-valve transcatheter aortic valve replacement Introduction Bioprosthetic aortic valve degeneration occurs in approximately 30% of patients within 10 years of surgical implantation.1 Untreated valve degeneration results in adverse remodeling of the left ventricle, recurrent hospitalizations for heart failure, and ultimately death.2,3 Current management options include the surgical approach of redo-surgical aortic valve replacement (redo-SAVR) and the percutaneous option of valve-in-valve transcatheter aortic valve replacement (ViV-TAVR). Redo-SAVR has historically been the standard of care as it offers favorable outcomes in terms of hemodynamics, patient–prosthesis mismatch,2 and coronary access.3 However, redo sternotomy can be associated with high perioperative mortality and morbidity, especially for higher-risk patients.4 In this context, ViV-TAVR has emerged as a widely adopted alternative and is increasingly considered a preferred strategy in contemporary practice for selected patients. In recent observational studies and meta-analyses, ViV-TAVR is associated with lower short-term mortality, fewer major complications such as bleeding and stroke, and shorter hospital length of stay (LOS) compared with redo-SAVR.5, 6, 7, 8, 9 However, both strategies are resource-intensive procedures that incur substantial costs for patients and health care systems, although the cost contributions are different. Prior studies have demonstrated higher costs and lower contribution margins for TAVR vs SAVR in native aortic stenosis.10,11 Similarly, ViV-TAVR has substantial procedural and device-related expenses, primarily due to the higher price of the transcatheter valve itself, rendering its overall economic value uncertain. Furthermore, the emergence of novel coronary protection strategies, such as leaflet laceration, are introducing additional procedural complexity and costs.12 With the rising demand for cost-conscious, outcome-driven approaches to structural valve disease, the aim of this study was to perform a cost-effectiveness analysis comparing ViV-TAVR to redo-SAVR for the management of degenerated bioprosthetic aortic valves. Methods Model structure We constructed a decision-analytic model (Figure 1) assessing the cost-effectiveness of ViV-TAVR compared with redo-SAVR for the management of degenerated prosthetic aortic valves. The patient population of our analysis consisted of adult hospital inpatients undergoing ViV-TAVR or redo-SAVR for a degenerated bioprosthetic aortic valve. The main comparative strategies of our analysis were redo-SAVR vs ViV-TAVR. Of note, the root of the model is a decision node; therefore, no baseline percentage or probability was assigned to allocate patients to a specific strategy. However, given that ViV-TAVR is associated with a risk of coronary obstruction by displacing the diseased bioprosthetic valve leaflets,12 we also included the use of coronary protection strategies in our model. More specifically, we assumed that a small percentage of patients (3% for each group) would also receive coronary protection with either (1) the Bioprosthetic Aortic Scallop Intentional Laceration to prevent Iatrogenic Coronary Artery (BASILICA) procedure; (2) the chimney stenting procedure, also known as coronary ostial stenting with a snorkel stent13; or (3) the ShortCut technique, where the aortic leaflet is mechanically split.14 Decision tree model. The square indicates the decision to choose between the use of ViV-TAVR or redo-SAVR for the management of degenerated aortic valves. The circles indicate chance nodes, and the triangles, end points. SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; ViV, valve-in-valve. Costs and outcomes were calculated for a time horizon of 2 months, and the analysis was performed from a health care perspective. As the time horizon of the analysis was short, costs and clinical effects were not discounted. Our study included an impact inventory, as recommended by guidelines.15 The impact inventory is a checklist of health and nonhealth outcomes that were considered in this analysis, shown in Table 1. Cost data were obtained from sources that reported values in US dollars. Mortality was defined as 30-day mortality and readmission as 30-day readmission. The model was developed using the software TreeAge Pro 2019 (TreeAge). The analysis followed the recommendations made by the Consolidated Health Economic Evaluation Reporting Standards Statement16 and the guidelines reported in 2017 by the Second Panel on Cost-Effectiveness Analysis.15 Institutional review board approval was not required for this study. Table 1Impact inventorySectorType of impactIncluded in this analysis from the health care perspective?Formal health care sectorHealthHealth outcomes (effects) Mortality✔ Medical Costs Paid for by third-party payers✔ Paid for by patient’s out-of-pocket✔ Future related medical costs (payers and patients)✔ Future unrelated medical costs (payers and patients)✘Nonhealth care sectorProductivityLabor market earnings lost due to absence from work✘Uncompensated household production, patient✘ Impact inventory Model inputs: assigning probabilities To identify studies that provide data on the effectiveness of ViV-TAVR compared to redo-SAVR, we used the most relevant systematic reviews and meta-analyses.5, 6, 7, 8, 9 More specifically, we included studies that directly compared the 2 strategies and that included >100 patients in their study population.17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34 LOS was obtained through a weighted average of LOS.23, 24, 25 Probability estimates and CIs for mortality, readmission, stroke, bleeding, acute kidney injury, and pacemaker placement, were obtained by pooling with the use of random-effects meta-analysis (Der Simonian and Laird) (MedCalc V19.8).35 This method was chosen as it accounts for the considerable interstudy differences and heterogeneity among the included studies.35 Moreover, to define our primary effectiveness parameter, survival probabilities were first converted to mortality probabilities. These were then multiplied by 100 to yield the expected number of deaths per 100 treated patients for each strategy. Finally, deaths averted per 100 patients, was calculated by subtracting the expected deaths in the ViV-TAVR cohort from the expected deaths in the redo-SAVR cohort. Model inputs: assigning costs Costs were obtained from the literature were adjusted to August 2025 US dollars using the consumer price index inflation calculator provided by the Bureau of Labor Statistics.36 The procedural costs for redo-SAVR and ViV-TAVR, were estimated at $10,439 and $8082 respectively, accounting for personnel costs (anesthesiology, cardiology, cardiothoracic surgery) as well as miscellaneous costs such as technologists and supplies.37 The costs of valves themselves were estimated at $6285 for a surgical aortic valve and $34,621 for a TAVR valve.38 The costs of complications, including the cost of stroke, bleeding, renal insufficiency, and pacemaker placement, were obtained from a study by Arnold et al39 that calculated costs of periprocedural complications for patients treated with TAVR. The costs of readmission for each strategy were obtained from a study that used the nationwide readmissions database to compare readmissions costs of TAVR vs SAVR.40 The cost of hospitalization was estimated by multiplying the cost of hospitalization per day for the state of Pennsylvania, provided by the Kaiser Family Foundation,41 with the length of hospital stay for patients who received ViV-TAVR or redo-SAVR.23, 24, 25 Costs for Pennsylvania were used as they closely approximate the national average. The cost of the intensive care unit (ICU) per day was also obtained from the literature,42 and it was multiplied for the estimated ICU LOS for each strategy.43 Postprocedural follow-up costs including rehabilitation/skilled nursing facility stays, outpatient services, and physician fees were obtained from the PARTNER 3 trial comparing TAVR and SAVR costs in the 0 to 30 day follow-up period.38 For the BASILICA procedure, we accounted for the cost of a FineCross microcatheter ($150),44 a coronary transversal wire, ($106),45 an Amplatz gooseneck snare ($299),46 and for laboratory and physician costs for 1 additional hour of procedure time.47 For the chimney/snorkel stenting procedure, we accounted for the cost of a coronary stent ($2294)48 and for laboratory and physician costs of 30 additional minutes of procedure time.47 For the ShortCut technique, we assumed a total additional cost of $9,750.49 The laboratory and physician costs included physician fees, room/overhead costs, and procedure medication costs per minute.47 Furthermore, acknowledging that in the United States, ViV-TAVR was initially approved for patients with high or greater surgical risk (Society of Thoracic Surgeons Predicted Risk of Mortality >8%),50 which is higher than the mortality used in our model (4%), we performed an additional sensitivity analysis to look at the effect of varying redo-SAVR mortality on ICER. Outcome and data analysis In the base case analysis, our primary outcome was the incremental cost-effectiveness ratio (ICER), defined as the ratio of the incremental cost between 2 two strategies (ViV-TAVR or redo-SAVR) over their incremental difference in effectiveness.15 The incremental cost was defined as the excess cost of ViV-TAVR compared with the cost of redo-SAVR. In turn, the incremental effectiveness was defined in terms of deaths averted. The robustness of our model was evaluated with the use of deterministic (1-way sensitivity) and probabilistic sensitivity analysis (Monte Carlo). In the 1-way sensitivity analysis, each parameter was tested across a range of multiple point estimates, whereas in the probabilistic analysis, we varied all parameters of the model simultaneously. The base case estimates, ranges, and distributions for all parameters are presented in Table 217, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34,37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49 Table 2Model inputs and baseline estimates for probabilities, LOS, and costsBase case value (range and distribution)Probabilities (with references) Probability of mortality with ViV-TAVR17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 340.03 (Range: 0.02-0.04)Beta (0.03; SD: 0.01) Probability of mortality with redo-SAVR17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 340.04 (Range: 0.03-0.06)Beta (0.04; SD: 0.06) Probability of pacemaker placement with ViV-TAVR18,23, 24, 250.11 (Range: 0.06-0.15)Beta (0.11; SD: 0.02) Probability of pacemaker placement with redo-SAVR18,23, 24, 250.07 (Range: 0.06-0.09)Beta (0.07; SD: 0.01) Probability of bleeding with ViV-TAVR18,23, 24, 250.23 (Range: 0.08-0.42)Beta (0.23; SD: 0.06) Probability of bleeding with redo-SAVR18,23, 24, 250.32 (Range: 0.11-0.59)Beta (0.32; SD: 0.08) Probability of acute kidney injury with ViV-TAVR23,250.17 (Range: 0.11-0.25)Beta (0.17; SD: 0.2) Probability of acute kidney injury with redo-SAVR23,250.23 (Range: 0.21-0.24)Beta (0.23; SD: 0.01) Probability of stroke with ViV-TAVR18,23,24,340.03 (Range: 0.01-0.10)Beta (0.03; SD: 0.02)Probability of stroke with redo-SAVR18,23,24,340.04 (Range: 0.04-0.11)Beta (0.04; SD: 0.01) Probability of using coronary protection strategy model assumption0.03 (Range: 0.02-0.06)Beta (0.03; SD: 0.01) Probability of 30-d readmission with ViV-TAVR23,240.13 (Range: 0.08-0.19)Beta (0.13; SD: 0.02) Probability of 30-d readmission with redo-SAVR23,240.11 (Range: 0.10-0.12)Beta (0.11; SD: 0.01)Costs (US dollars) Procedural costs for redo-SAVR3710,439 (Range: 5220-20878)Gamma (10439; SD: 2610) Procedural costs for ViV-TAVR378082 (Range: 4041-16164)Gamma (8082; SD: 2021) Valve cost for ViV-TAVR3834,621 (Range: 17311-69242)Gamma (34621; SD: 8655) Valve cost for redo-SAVR386285 (Range: 3143-12570)Gamma (6285; SD: 1571) Cost of pacemaker placement3921,386 (Range: 10693-42772)Gamma (21386; SD: 5347) Cost of bleeding3959,411 (Range: 29706-118822)Gamma (59411; SD: 14853) Cost of stroke3942,503 (Range: 21252-85006)Gamma (42503; SD: 10626) Cost of renal insufficiency3929,381 (Range: 14691-58762)Gamma (29381; SD: 7345) Cost of hospitalization per day for the state of Pennsylvania413336 (Range: 1668-6672)Gamma (3336; SD: 834) Cost of ICU hospitalization per day426450 (Range: 3225-6450)Gamma (6450; SD: 538) Follow-up costs for ViV-TAVR38846 (Range: 423-1692)Gamma (846; SD: 212) Follow-up costs for redo-SAVR382770 (Range: 1385-5540)Gamma (2770; SD: 693) Cost of rehospitalization for redo-SAVR409020 (Range: 4510-18040)Gamma (9020; SD: 2255) Cost of rehospitalization for ViV-TAVR409686 (Range: 4843-19372)Gamma (9686; SD: 2422) Cost of FineCross wire44150 (Range: 75-300)Gamma (150; SD: 38) Cost of Amplatz gooseneck snare46299 (Range: 150-598)Gamma (299; SD: 75) Cost of transversal wire45106 (Range: 53-212)Gamma (106; SD: 27) Cost of a drug-eluting stent482294 (Range: 1147-4588)Gamma (2294; SD: 574) Cost of catheterization laboratory per minute4765 (Range: 33-130)Gamma (65; SD: 16) Cost of shortcut499750 (Range: 4875-19500)Gamma (9750; SD: 2438)LOS, d Total hospital LOS with ViV-TAVR23, 24, 255.6 (Range: 2.8-10)Gamma (5.6; SD:1.2) Total hospital LOS with redo-SAVR23, 24, 259.5 (Range: 4.8-19)Gamma (9.5; SD:2.4) ICU LOS with ViV-TAVR430.8 (Range: 0.4-1.6)Gamma (0.8; SD:1.2) ICU LOS with redo-SAVR433.3 (Range: 1.7-6.6)Gamma (3.3; SD: 0.6)ICU, intensive care unit; LOS, length of stay; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; ViV, valve-in-valve. Model inputs and baseline estimates for probabilities, LOS, and costs ICU, intensive care unit; LOS, length of stay; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; ViV, valve-in-valve. Probabilities were modeled as beta distributions, whereas costs and LOS were modeled as gamma distributions, as recommended by guidelines.15 When a range was not available for a variable, we approximated it by allowing the variable to vary between 50% and 200% of its base case value.51 If a standard deviation was not available, it was estimated by dividing the range by 6, as suggested for data that do not follow the normal distribution (approximation obtained with the use of Chebyshev inequality).52 In the Monte Carlo analysis, the model was run 10,000 times,51 and each time, a value from the predetermined distributions (Table 217, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34,37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49) was randomly selected for each variable. The results of each simulation were plotted on an incremental cost-effectiveness plane as points with coordinates (x, y), with x representing incremental effectiveness and y representing incremental cost. Points located within the southeast quadrant of the graph were considered to be cost effective and dominant.53 Finally, cost-effectiveness acceptability curves were used to evaluate the cost-effectiveness for various willingness-to-pay thresholds.15 Results Probabilistic analysis In the probabilistic analysis, the mean cost for ViV-TAVR was estimated to be $88,829 (95% CI, $88,554-$89,104), and the mean cost for redo-SAVR was estimated to be $91,411 (95% CI, $91,163-$91,659). The mean effectiveness (survival probability) was 0.97 for ViV-TAVR (95% CI, 0.969-0.970) and 0.96 (95% CI, 0.959-0.961) for redo-SAVR. Consequently, ViV-TAVR emerged as the dominant strategy, yielding a mean ICER of −$259,323 (95% CI, −$304,599 to −$220,679). In the cost-effectiveness acceptability curve (Figure 2), we present the probability that ViV-TAVR is cost effective compared with redo-SAVR across a range of willingness-to-pay thresholds ranging from $0 to $150,000. ViV-TAVR was cost effective in 57% to 58% of simulations for a willingness-to-pay ranging from $0 to $150,000. The cost-effectiveness acceptability plane, which shows the uncertainty around the cost-effectiveness outcomes, is depicted in Figure 3. Cost-effectiveness acceptability curve with a willingness to pay ranging from $0 to $150,000. The figure shows the proportion of probabilistic sensitivity analysis iterations in which ViV-TAVR (blue) and redo-SAVR (red) are cost effective across a range of willingness-to-pay thresholds. ViV-TAVR demonstrates a higher probability of being cost effective compared with redo-SAVR across all thresholds examined. SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; ViV, valve-in-valve. Incremental cost-effectiveness plane for ViV-TAVR compared to redo-SAVR. Each point represents the incremental cost and effectiveness of ViV-TAVR compared with redo-SAVR from a probabilistic sensitivity analysis iteration. The dashed diagonal line denotes the willingness-to-pay threshold of $50,000 per death averted. Points below the line indicate iterations in which ViV-TAVR is cost effective at this threshold. Green points represent iterations in which ViV-TAVR is less costly and/or more effective than redo-SAVR, whereas red points represent iterations in which ViV-TAVR is more costly and/or less effective. The ellipse indicates the 95% confidence region of the joint distribution of incremental costs and effectiveness. SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; ViV, valve-in-valve; WTP, willingness-to-pay. Base case analysis In the base case analysis the cost for the ViV-TAVR strategy was calculated to be $88,821. This included procedural,37 valve,38 complications,39 hospitalization,41,42 follow up,38 and rehospitalization.40The overall probability of survival with ViV-TAVR at 1 month was estimated to be 0.97. This probability of survival was estimated by pooling the available studies.17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34 Similarly, for the redo-SAVR strategy the cost was calculated to be $91,421, and the probability of survival at 1 month was 0.96. The included costs were procedural,37 valve,38 complications,39 hospitalization,41,42 follow up,38 and rehospitalization.40 The incremental difference in effectiveness between ViV-TAVR and redo-SAVR was 0.01 (0.97 vs 0.96). Overall, ViV-TAVR was associated with 1 fewer death per 100 patients treated compared with redo-SAVR. The incremental difference in cost between ViV-TAVR and redo-SAVR was $2,600 ($88,821 vs $91,421). ViV-TAVR resulted in savings of $259,939 per death averted (ICER, −$259,939/death averted) compared with redo-SAVR, suggesting that it is a cost-effective strategy. Sensitivity analysis The sensitivity analysis, which allowed us to test each model variable for thresholds by varying each base case value within the limits specified in Table 217, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34,37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49 suggested that in clinical practice, compared to ViV-TAVR, the redo-SAVR strategy would become the cost-effective strategy if the cost of the TAVR prosthesis is higher than $37,220, if the cost of the SAVR is less than $3,186, if the procedural costs of ViV-TAVR are more than $10,681, if the procedural costs of redo-SAVR are less than $7,840, if the cost of ICU per day is less than $5,410, or if the cost of bleeding is less than $30,529. Similarly, the findings are reversed if the LOS of redo-SAVR is shorter than 8.7 days, if the LOS of ViV-TAVR is longer than 6.4 days, or if the ICU LOS for redo-SAVR is shorter than 2.5 days. The reversal is also seen if the probability of death with ViV-TAVR is over> 0.4, if the probability of bleeding with redo-SAVR is <0.28, if the probability of bleeding with ViV-TAVR is <0.27, or if the probability of stroke with ViV-TAVR is >0.09. The findings of the sensitivity analysis are summarized in the tornado diagram (Figure 4), which is a graphical representation of how variations in each model variable affect the cost-effectiveness output. Furthermore, in an additional sensitivity analysis, where we varied the mortality of redo-SAVR from 0% to 20%, ViV-TAVR remained the more cost-effective strategy (Supplemental Figure S1). Additional graphical sensitivity analyses, for cost of hospitalization per day, ViV-TAVR valve cost, and ICU cost per day can be found in the Supplemental Materials. Tornado diagram. This graph is a summary of the one-way sensitivity analysis on the ICER of ViV-TAVR vs redo-SAVR. Each horizontal bar represents the range of ICER values obtained when a single model parameter was varied across its plausible lower and upper bounds, while holding all other parameters constant. Blue bars indicate the ICER associated with the lower bound of the parameter value, and red bars indicate the ICER associated with the upper bound. The vertical dashed line represents the willingness-to-pay (WTP) threshold of $50,000 per quality-adjusted life-year (QALY). Parameters are ordered from top to bottom according to their influence on the ICER, with those producing the greatest variation displayed at the top. AVR, aortic valve replacement; ICER, incremental cost-effectiveness ratio; ICU, intensive care unit; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; ViV, valve-in-valve. Discussion In this decision-analytic study comparing ViV-TAVR with redo-SAVR for the management of degenerated bioprosthetic aortic valves, we found that ViV-TAVR was the dominant strategy, yielding lower costs and improved short-term survival relative to redo-SAVR. More specifically, ViV-TAVR resulted in savings of $259,323 (95% CI, −$304,599 to −$220,679) per death averted (ICER, −$259,323/death averted) compared with redo-SAVR. It also remained cost effective across multiple sensitivity analyses and in 57% to 58% of simulations for a willingness-to-pay ranging from $0 to $150,000. The findings are summarized in the Central Illustration. Cost-Effectiveness ofViV-TAVRvsredo-SAVR.ViV-TAVR is the economically dominant strategy for degenerated bioprosthetic aortic valves. Compared with redo-SAVR, it yields lower costs and higher short-term survival, resulting in savings of $259,323 per death averted. ViV-TAVR remained cost effective in 57% to 58% of probabilistic simulations. ICER, incremental cost-effectiveness ratio; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement; ViV, valve-in-valve. This is the first modeling study examining the financial implications of ViV-TAVR vs redo-SAVR. A prior study that studied the cost the effectiveness of TAVR vs SAVR among intermediate risk patients and found that despite the higher procedural costs of TAVR (∼$20,000 higher for TAVR than SAVR), there was a small difference in the total index hospitalization costs (∼$2888 higher with TAVR) owing to the reductions in LOS associated with TAVR and due to reduced follow-up costs.54 Keuffel et al55 compared a staged approach of SAVR followed by ViV-TAVR to using a mechanical valve and estimated that it decreased 25-year expenditures from a payer perspective,55 suggesting a consistent financial benefit of ViV-TAVR across clinical scenarios, including for younger patients who may outlive their first surgical bioprosthetic valve replacement. Our base case analysis demonstrates that ViV-TAVR is the dominant strategy; however, our probabilistic sensitivity analysis reveals substantial decisional uncertainty due to the marginal absolute differences in both costs and clinical effects. The absolute difference in survival between the 2 strategies is small (0.97 vs 0.96), reflecting the high procedural success and low complication rates of both interventions. Notably, the 95% credible intervals for total costs notably do not cross. This indicates that the cost-saving nature of ViV-TAVR remains stable even when accounting for parameter uncertainty. For modeling purposes, given the lack of high-quality trial data, we used large ranges and generous distributions for uncertainty, which may have contributed to smaller detectable differences. In the sensitivity analysis, which is graphically represented in our tornado diagram (Figure 2), we found that length of hospital stay and procedural costs were influential variables on the findings of the analysis. LOS is often used in economic analyses as a proxy for resource use and efficiency and it commonly heavily impacts total hospitalization costs.56 In our study the extended LOS for redo-SAVR vs ViV-TAVR is consistent with many studies that have shown higher median LOS with SAVR vs TAVR as well as with higher ICU LOS.57,58 This is likely due to the fact that a surgical approach is associated with more major morbidity and intensive postoperative monitoring than a percutaneous approach. However, our analysis suggested that when the TAVR valve costs more than $37,220, or if the procedural costs of redo-SAVR are less than $7840, then redo-SAVR becomes the dominant strategy. This is a finding that reflects the sensitivity of the model to pricing structures and resource utilization patterns as they vary across different clinical settings and institutions. It also highlights the weight of costs associated with the index admission. Although our economic evaluation represents a snapshot in time, through the sensitivity analyses it offers a foundational framework to assess how future pricing trajectories will impact outcomes. Recognizing that coronary obstruction represents one of the major procedural concerns associated with ViV-TAVR, we included the use of various techniques as part of our analysis. If unaddressed prior to ViV-TAVR and obstruction occurs, it is a potentially catastrophic complication associated with high mortality (estimated around 40%-50%) that can result in the need for rescue percutaneous coronary intervention or emergency coronary artery bypass grafting.12,59 If addressed, it prevents coronary obstruction, but it is a technically complex procedure that requires additional cardiac catheterization laboratory time, personnel time, and supplies. Fortunately, a minority of ViV-TAVR cases carry significant risks of coronary obstruction. Taken together, these data support the use of ViV-TAVR for bioprosthetic aortic valve degeneration from a health care perspective, despite multiple prior studies demonstrating higher costs associated with TAVR procedure. Given the safety and efficacy of both procedures, these findings confirm that both redo-SAVR or ViV-TAVR are excellent strategies for the treatment of bioprosthetic aortic valve stenosis, and the use of one therapy or the other should be dictated by the clinical scenario. In younger, low-risk patients, redo-SAVR should be the preferred strategy and in older, higher-risk patients, ViV-TAVR is safe, effective, and cost effective. Of note, as ViV-TAVR has historically been approved for higher-risk patients, those undergoing the procedure may have been sicker at baseline than patients undergoing redo-SAVR, which could bias our findings toward underestimating the economic advantage of ViV-TAVR in comparably risk-matched populations. When interpreting the findings of our analysis, several limitations should be considered. First, cost-effectiveness analyses rely on mathematical decision models that simplify and condense complex clinical scenarios, and their results are, therefore, dependent on the assumptions and parameter estimates applied. Detailed baseline clinical characteristics of patients undergoing redo-SAVR vs ViV-TAVR such as comorbidity burden, Society of Thoracic Surgeons-predicted surgical risk, and urgency or emergent status, were not clearly available. Moreover, the effectiveness data for our model were derived from retrospective observational studies rather than randomized controlled trials. Although a substantial portion of the included studies used propensity score matching, residual confounding inevitably remains, introducing the possibility that unmeasured differences in patient selection or procedural complexity may have influenced the pooled event rates. In addition, as detailed granular cost data for the ViV-TAVR procedure and its complications are limited, we obtained cost data from multiple sources, which may limit their comparability. Although we itemized costs for the catheter-based procedures, some specific low-cost equipment may have been omitted. Given the lack of information, we also did not differentiate among specific valve types, surgical techniques, or anatomic subgroups, which can affect procedural risk, resource use, and costs. Similarly, the short time horizon that we had to use to maximize data reliability does not capture long-term considerations such as valve durability, gradients, patient–prosthesis mismatch, or the future need for reintervention. This narrower time horizon was chosen to avoid extensive long-term assumptions, recognizing that a substantial portion of the economic burden is driven by upfront, periprocedural costs. There is some evidence for more favorable long-term outcomes with redo-SAVR,60 which could have important economic implications beyond the timeframe considered in this analysis. We also did not capture heterogeneity in surgical complexity, including the need for aortic root enlargement or other adjunctive procedures in selected redo-SAVR cases. Furthermore, our model evaluated effectiveness based on short-term survival rather than quality-adjusted life years. Because ViV-TAVR is a less invasive procedure, these 2 strategies likely have significantly different profiles of short-term functional recovery and quality of life improvement, which our current analysis does not capture. Finally, because this analysis was conducted strictly from a health care sector perspective, it inherently does not account for indirect societal costs, such as lost productivity, out-of-pocket expenses, or caregiver burden, which may differentially impact patients recovering from these interventions. When additional information is available, a Markov model lifetime analysis incorporating realistic treatment sequences such as staged or combination strategies would be very helpful to understand the consequences of waxing valve durability over time and cumulative costs. Structural valve degeneration of bioprosthetic aortic valves is associated with significant mortality, morbidity and health care resource use. Our study, which integrated data from the literature on both clinical effectiveness and costs, showed that ViV-TAVR is a cost-effective strategy when compared with redo-SAVR. Through technological advancements, procedural streamlining, and increased operator and center experience, the cost of ViV-TAVR is likely to decrease over time. Our modeling also revealed decisional uncertainty, primarily driven by the parameter uncertainty in the current, predominantly retrospective literature. In the future, high quality, prospective trial-level comparative data are needed to reduce uncertainty and refine economic and clinical estimates. CRediT authorship contribution statement Elina E. Pliakos: Conceptualization, Data curation, Formal analysis, Methodology, Writing – original draft, Writing – review & editing, Investigation. Andrew M. Acker: Investigation, Methodology, Writing – review & editing, Writing – original draft. Paul N. Fiorilli: Investigation, Methodology, Writing – review & editing. Allison L. Tsao: Investigation, Methodology, Writing – review & editing. Kendall M. Lawrence: Investigation, Methodology, Writing – review & editing. Nimesh D. Desai: Investigation, Methodology, Writing – review & editing. Toby Rogers: Investigation, Methodology, Writing – review & editing. Jaffar M. Khan: Investigation, Methodology, Writing – review & editing. Jay S. Giri: Conceptualization, Investigation, Methodology, Project administration, Software, Supervision, Writing – review & editing. Ashwin S. Nathan: Conceptualization, Data curation, Investigation, Methodology, Project administration, Resources, Software, Supervision, Writing – original draft, Writing – review & editing. Declaration of competing interest Paul Fiorilli has received consulting honoraria from Edwards Lifesciences. Kendall Lawrence has received speaking fees from Edwards Lifesciences. Nimesh Desai is a consultant for Gore Medical and Terumo Aortic. Toby Rogers received consulting honoraria from Edwards Lifesciences, Medtronic, Boston Scientific, Abbott, Anteris, and Transmural Systems, serves on the advisory boards of Medtronic and Boston Scientific, has equity interest in Transmural Systems, and is a co-inventor on patents, assigned to National Institutes of Health, for transcatheter electrosurgery devices. Jaffar Khan is a co-inventor on patents assigned to the National Institutes of Health for purpose-built devices intended to lacerate valve leaflets. Jay Giri has received research funding and served on advisory boards for Abiomed, Edwards, Boston Scientific, and Inari Medical, and holds equity in Endovascular Engineering. Ashwin Nathan has received research funding from Abiomed, Biosense, Webseter, Edwards Lifesciences, and Boston Scientific. He has received speaking fees from Edwards Lifesciences. Elina Pliakos, Andrew Acker, and Allison Tsao have no conflicts of interest to declare. Acknowledgments Funding sources The authors did not receive any financial support for this study. Ethics statement and patient consent Institutional review board approval was not required for this study. Footnotes To access the supplemental material accompanying this article, visit the online version of the Journal of the Society for Cardiovascular Angiography & Interventions at 10.1016/j.jscai.2026.105400. Supplementary material Supplementary Figures S1-S4 References Associated Data Supplementary Materials Supplementary Figures S1-S4