 |
 |
 |
When is Surgical Valvotomy a Better Choice than Balloon Valvotomy for Neonates with Isolated Critical Aortic Stenosis? A Decision Analysis Model
Anish Katta1, Samuel M. Hoenig1, Kunaal S. Sarnaik1, Justin Robinson2, Karl Welke3, Brian W. McCrindle4, *Tara Karamlou5
1School of Medicine, Case Western Reserve University, Cleveland, Ohio, United States, 2Heart Vascular and Thoracic Institute, Cleveland Clinic, Cleveland, Ohio, United States, 3Division of Pediatric Cardiothoracic Surgery, Atrium Health, Charlotte, North Carolina, United States, 4Labatt Family Heart Centre, Sick Kids Foundation, Toronto, Ontario, Canada, 5Department of Pediatric Cardiac Surgery, Akron Children's Hospital, Akron, Ohio, United States
Background: While Balloon Aortic Valvuloplasty (BAV) is preferentially used over Surgical Aortic Valvotomy (SAV) for neonates with isolated critical aortic stenosis (AS), there is limited evidence supporting this paradigm. We aimed to objectively compare these two strategies, accounting for short and long-term effects. We present a decision analysis utilizing a cost-effectiveness model to compare these strategies. Through this process, we highlight the challenges in modeling and the importance of critically assessing model reliability and clinical relevance. Methods: An economic evaluation was designed employing a cost-effectiveness analysis through a Markov decision tree with Monte Carlo microsimulations that simulated survival, aortic valve (AV) reinterventions, aortic valve replacement, and mortality for neonates with isolated critical AS (Figure 1A). Two hypothetical cohorts of 10,000 neonates underwent either index SAV or BAV. The model was run in 1-year cycles at the 20-year horizon. Cost, defined as the healthcare expenditure associated with hospitalization of each AV intervention, was derived and normalized to January 2024. Effectiveness was defined as the years a patient remained alive with their native AV, emphasizing sustained valve preservation as a key outcome. This definition contrasts with traditional measures like Quality Adjusted Life Years (QALYs) and highlights the importance of evaluating the durability of these procedures in neonates. The willingness-to-pay (WTP) threshold represented the theoretical amount society would pay to delay AVR by 1 year. Net Monetary Benefit (NMB) and the incremental cost-effectiveness ratio (ICER) were derived to assess the cost-effectiveness of each intervention. Tornado threshold analysis assessed how NMB changed with variance of the baseline values. Results: Mean mortality over 20 years was 14.95% (SD: 35.65%) in BAV and 6.53% (24.7%) in SAV, with highest mortality between 0-30 days. BAV had more reinterventions than SAV (2.37 [1.06] vs. 1.95 [0.78]) and lower cumulative costs ($111,243.32 [49,094.87] vs. $175,415.00 [39,692.41] for SAV). Mean age at AVR was higher in BAV: 12.69 years (7.67), compared to SAV: 12.64 (7.65). At baseline, the ICER score was $1,305,793.80, meaning that BAV dominates SAV in terms of cost-effectiveness due to BAV's similar effectiveness and lower costs than SAV (Figure 1B). SAV only offers a NMB over BAV if society is willing to exceed $1,305,793.80 per year AVR is delayed. At a WTP threshold of $100,000, sensitivity analysis indicated that high index SAV cost, high index SAV mortality and low index BAV mortality were factors that substantially impacted the NMB in BAV's favor. The results highlight the risks of relying on decision analysis models without robust data validation. Variation in input parameters such as mortality and reintervention rates, resulted in implausible estimates under certain conditions. For instance, the tornado diagram (Figure 1C) demonstrated that scenarios with a low BAV mortality and high SAV cost resulted in implausibly high NMB values for BAV, exaggerating its apparent cost-effectiveness. These patterns emphasize the need for external validation using datasets to improve prediction reliability and accuracy. Conclusion: This preliminary cost-effectiveness analysis establishes a theoretical framework to determine if BAV is more cost-effective than SAV for neonates with isolated critical AS and highlights the challenges of decision analysis modeling. While preliminary findings indicate cost-effectiveness hinges on mortality and reintervention rates of BAV and SAV, the model's sensitivity to changes in inputs emphasizes the need for external data validation. Future efforts will incorporate dataset-based costs and outcomes to strengthen model accuracy and guide clinical decision-making.
Back to 2025 Abstracts
