Introduction: Guideline-based eligibility criteria for endovascular thrombectomy (EVT) are commonly used as criteria to select patients for transfer from referring hospitals (RH) to thrombectomy capable stroke centers (TCSC), resulting in a large amount of futile transfers. We sought to determine the baseline clinical and imaging factors associated with greater likelihood of receiving EVT after transfer, and explored the impact of simulated vascular imaging availability at RH on reducing the number of futile transfers for EVT.
Methods: We identified stroke patients transferred from 30 RH between 2010 and 2016 for which (1) the RH computed tomography (CT) and (2) either CT/CT angiography or MRI at arrival were available for review. Initial ASPECT scores and baseline imaging characteristics were evaluated by 2 raters. The main outcome variable was pre-defined as receiving EVT at the TCSC after transfer. To analyze different options for improving triaging criteria for EVT at the RH, we generated various data-based situations accounting for the presumed vascular occlusion status based first on data available only at the RH and then by imputing vascular imaging information gathered at the TCSC presuming it could have been captured at the RH. The predictive ability of each situation, using varying NIHSS thresholds, was evaluated to find an optimal sensitivity (Se) - specificity (Sp) threshold.
Results: A total of 508 patients were included in the analysis (mean age 69.7 ± 14.8, 41.9% females). Application of current guidelines for EVT eligibility (NIHSS >6, ASPECTs≥6, Time<6h) yielded a sensitivity (Se) of 92% (95%CI, 0.84-0.96) and a specificity (Sp) of 53% (95%CI, 0.48-0.57) for receiving EVT. Repeated simulations identified an optimal selection threshold of NIHSS≥9 if accompanied by RH vascular imaging (Se=91%; 95% CI, 83%-95%and Sp=80%; 95% CI 75%-83%). Using these criteria, for every 5 patients transferred for EVT, 2.5 would receive the treatment and one futile transfer would be prevented.
Conclusion: Implementation of vascular imaging at RH would result in significantly fewer futile transfers for EVT. The optimal trade-off between time, cost and imaging will impact the number needed to transfer and will depend on local resources and capabilities.