Economic analyses have determined that successful endovascular thrombectomy procedures are cost-effective in patients with moderate-to-severe pre-stroke disability, with more pronounced signals being observed in individuals with a modified Rankin scale (mRS) score of 3—as compared to 4. These findings were shared for the first time at the 2026 Society of NeuroInterventional Surgery (SNIS) annual meeting (20–24 July, Seattle, USA) by Na Young Evelyn Park (University of Tennessee College of Medicine, Memphis, USA).
“Previous studies have suggested that successful recanalisation is associated with better functional outcomes and lower mortality rates compared to unsuccessful recanalisation, which supports the clinical rationale for pursuing endovascular thrombectomy in patients with moderate-to-severe disability, or mRS 3–4,” Park noted, speaking at SNIS 2026. “And, while previous economic analyses have shown that endovascular thrombectomy is cost-effective for patients with mild pre-stroke disability—or mRS 1–2—patients with mRS 3–4 have not yet been evaluated. This is the question we’re trying to answer today.”
To this end, Park and colleagues built an analytic model comparing successful versus unsuccessful recanalisation in patients with pre-stroke mRS scores of 3 or 4, which were analysed separately via data from the Stroke Thrombectomy and Aneurysm Registry (STAR)—co-founded by Alejandro Spiotta (Medical University of South Carolina, Charleston, USA) and Ali Alawieh (Semmes-Murphey Clinic, Memphis, USA).
“Patients underwent endovascular thrombectomy, were assigned a 90-day utility-weighted mRS score, and then entered a long-run projection phase assuming an average life expectancy of 4.2 years,” Park explained. “Functional outcomes were measured using utility-weighted mRS, converted to QALYs [quality-adjusted life years], and ICERs [incremental cost-effectiveness ratios] were calculated so that we could determine cost-effectiveness and compare it against willingness-to-pay thresholds of US$50,000 and US$150,000.”
The results of the analysis showed successful recanalisation to be cost-effective across both mRS 3 and mRS 4. Regarding mRS 3 specifically, successful recanalisation carried an additional cost of roughly US$11,000 but led to gains of 0.52 QALYs, equating to an ICER of $US22,000 per QALY. And, for mRS 4, successful recanalisation carried an extra cost of approximately US$8,500 alongside 0.29 QALYs, resulting in an ICER of $US29,000 per QALY.
“Both ICERs were well below the US$50,000 threshold,” Park commented.
Touching on data from one-way sensitivity analyses performed by the researchers—whereby they introduced variability into the thrombectomy costs and time horizons—Park reported that their initial findings on mRS 3 were “robust”. However, for mRS 4, higher estimated procedure costs and pessimistic combined scenarios saw cost-effectiveness exceed the US$50,000 willingness-to-pay threshold.
“This indicates that, with changes in model assumptions, there was some uncertainty,” Park added.
Finally, the presenter shared the findings of probabilistic sensitivity analyses that saw the researchers run a total of 10,000 simulations in which they varied several key parameters, including thrombectomy procedure costs, annual care costs, and mRS scores and mortality rates at 90 days.
Regarding mRS 3, roughly 60% of those simulations demonstrated cost-effectiveness at the US$50,000 threshold, rising to 91% for the US$150,000 threshold—and the same figures on mRS 4 were approximately 43% and 64%, respectively, also indicating “greater uncertainty” in the latter group, according to Park.
“In conclusion,” she averred, “endovascular thrombectomy is cost-effective in both pre-stroke disability groups, but our findings were more robust for mRS 3. Prospective studies should include patient-level cost data from hospitals’ administrative records and long-term survival data to replace literature-based assumptions, and a societal perspective analysis incorporating caregiver burden would likely strengthen the economic case further—particularly for pre-stroke disabilities of mRS 4.”
