Cost-effectiveness of catheter ablation versus medical therapy for the treatment of atrial fibrillation in the United Kingdom

Abstract

Introduction: Research evidence has shown that catheter ablation is a safe and superior treatment for atrial fibrillation (AF) compared to medical therapy, but real-world practice has been slow to adopt an early interventional approach. This study aims to determine the cost effectiveness of catheter ablation compared to medical therapy from the perspective of the United Kingdom.

Methods: A patient-level Markov health-state transition model was used to conduct a cost-utility analysis. The population included patients previously treated for AF with medical therapy, including those with heart failure (HF), simulated over a lifetime horizon. Data sources included published literature on utilization and cardiovascular event rates in real world patients, a systematic literature review and meta-analysis of randomized controlled trials for AF recurrence, and publicly available government data/reports on costs.

Results: Catheter ablation resulted in a favorable incremental cost-effectiveness ratio (ICER) of £8614 per additional quality adjusted life years (QALY) gained when compared to medical therapy. More patients in the medical therapy group failed rhythm control at any point compared to catheter ablation (72% vs. 24%) and at a faster rate (median time to treatment failure: 3.8 vs. 10 years). Additionally, catheter ablation was estimated to be more cost-effective in patients with AF and HF (ICER = £6438) and remained cost-effective over all tested time horizons (10, 15, and 20 years), with the ICER ranging from £9047-£15 737 per QALY gained.

Conclusion: Catheter ablation is a cost-effective treatment for atrial fibrillation, compared to medical therapy, from the perspective of the UK National Health Service.

Keywords: antiarrhythmic drugs; atrial fibrillation; catheter ablation; cost-effectiveness; economic evaluation.

Figure 1

Model Structure for the Treatment of Atrial Fibrillation & Treatment Protocol for AF Recurrence. The structure and flow of patients through the model are depicted in the diagrams above. Figure 1A shows the general flow of patients through the various health states in the model, while Figure 1B depicts the protocol for patients that experience recurrence in the model. 1) All patients will begin in the “Treatment” state, incurring expenses related to the both the pretreatment work‐up and treatment itself. 2) Patients who undergo ablation are assumed to have initial procedural success. 3) In the “Normal Sinus Rhythm” and “AF Recurrence” health states, patients incur costs related to follow‐up/maintenance care. 4) All patients in the ablation arm that restore normal sinus rhythm from AF recurrence do so because of 1) starting AADs or 2) receiving a repeat ablation. In the AAD arm, it is assumed that all patients do so because of changing AADs. 5) In the catheter ablation arm, a subset of patients with AF recurrence will receive a single repeat ablation procedure. 6) After attempting 4 treatments, as outlined in the Recurrence Treatment Protocol, patients will cease rhythm control efforts, going on rate control drugs for remainder of model. 7) Patients may experience CV events (ischemic stroke, major bleeding event, or cardiac arrest) during the model time horizon. After a CV event, patients can recover without disability and continue rhythm control efforts moving to either the normal sinus rhythm or AF recurrence states, become disabled and move into a post‐AE state, or die. AAD: Antiarrhythmic drug; AF: Atrial fibrillation; CA: Catheter ablation; MT: Medical therapy

Figure 2

Estimated Freedom from AF Recurrence from Meta‐Analysis (Model Inputs/Transition Probabilities). A systematic literature review and meta‐analysis was performed to estimate the probability of a patient experiencing recurrence at 12‐, 48‐, and 144‐months after beginning treatment. These estimates were used, assuming an exponential decay over time, to develop the model inputs for the chance of a patient having a recurrence over time. The figure above depicts the probabilities over time, interpolated from the estimates at the three time points.CA: Catheter ablation; MT: Medical therapy

Figure 3

Scatterplot of ICER values. To test uncertainty in the model input values, a Monte Carlo probabilistic sensitivity analysis was conducted – running the analysis 10,000 times with different sets of model inputs. The ICER (incremental cost per QALY gained) from each of the 10,000 simulations are shown in the scatter plot above. The willingness‐to‐pay (WTP) line shows the threshold for which catheter ablation is considered to be cost effective – with each data point that is below the line representing a simulation that was cost‐effective. The analysis found ablation to be cost‐effective in 99% of the simulations that were run. ICER: Incremental cost‐effectiveness ratio; MT: Medical therapy; QALY: Quality‐adjusted life year; WTP: Willingness to pay