Endovascular treatment for ischemic stroke with the drip-and-ship model-Insights from the German Stroke Registry

Abstract

Background: Endovascular therapy (EVT) in acute ischemic stroke has been widely established. Globally, stroke patients are transferred either directly to a thrombectomy center (DC) or a peripheral stroke unit with a "drip-and-ship" (DS) model. We aimed to determine differences between the DS and DC paradigms after EVT of acute stroke patients with large-vessel-occlusion (LVO) in the database of the German Stroke Registry (GSR).

Methods: We performed a retrospective analysis of GSR patients between June 2015 and December 2019 in 23 German centers. Primary outcome was an ordinal shift analysis of modified Rankin Scale (mRS) 90 days after index event. Secondary endpoints included time from symptom onset to recanalization and complications. Tertiary endpoint was the association of imaging strategies in DS admissions with outcome.

Results: 2,813 patients were included in the DS and 3,819 in the DC group. After propensity score matching mRS after 90 days was higher in DS than DC admissions (OR 1.26; 95%-CI 1.13-1.40). Time from symptom-onset to flow-restoration was shorter in DC than DS (median 199.0 vs. 298.0 min; p < 0.001). DS patients undergoing magnetic resonance imaging (MRI; n=183) before EVT had a lower 90-day mRS than without (n = 944) (OR 0.63; 95%-CI 0.45-0.88). ASPECTS assessed on MRI correlated with 90-day mRS (ρ = -0.326; p < 0.001).

Conclusions: Clinical outcome was worse for EVT-eligible patients in the DS setting, even though patients were in a better state of health prior to stroke. A potentially mutable factor was the time delay of 99 min from symptom-onset to successful recanalization. Performing MRI before thrombectomy was associated with good outcome and MRI-ASPECTS was negatively correlated with mRS after 90 days.

Keywords: direct-to-center; drip-and-ship; endovascular treatment; ischemic stroke; mechanical thrombectomy.

Conflict of interest statement

Author FB reports receiving research grant from Stryker Neurovascular and Boehringer Ingelheim and speaker's honoraria from Alexion, Medtronic and Laerdal. Author WP reports speaker's honoraria from Laerdal. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2022 Schaefer, Kurka, Keil, Wagner, Steinmetz, Pfeilschifter, Bohmann and on behalf of the GSR-ET Study Group.

Figures

Figure 1

Forest plots showing multivariable ordinal regression analyses of the adjusted common odds ratio (OR) of modified Ranking Scales (mRS) 90 days after stroke for DC/DS admission status, sex, age, admission NIHSS, comorbidities and thrombolysis in all patients [(A), n = 5,107] and after propensity score matching [(B), n = 4,348]. The OR is presented logarithmically with 95%-confidence intervals.

Figure 2

The outcome on modified Rankin Scale (mRS) after 90 days for drip-and-ship (DS) and direct-to-center (DC) in all patients [(A); DS n = 2,266; DC, n = 3,172; OR 1.28; 95%-CI 1.15–1.41; p < 0.001] and after a propensity score matched analysis [(B); DS n = 2,114; DC n = 2,993; OR 1.28; 95%-CI 1.15–1.41; p < 0.001]. Favorable outcome (mRS 0–2) was also significantly more likely in DC than DS admissions after thrombolysis [(C); DS n = 1,273, DC n = 1,485; OR 0.73; 95%-CI 0.64–0.83; p < 0.001], but not if thrombolysis was not administered [(D); DS n = 977, DC n = 1,663; OR 0.87; 95%-CI 0.72–1.05; p = 0.149].

Figure 3

Correlation between the time from symptom onset to successful endovascular recanalization and the absolute difference of NIHSS scores at hospital admission and discharge. Values n = 2,366; ρ = 0.27; p < 0.001).

Figure 4

Modified Ranking scale (mRS) outcome data after 90 days in all drip-and-ship (DS) patients without secondary imaging (n = 1,171), with secondary imaging in interventional center (n = 1,045), only CT-imaging (n = 944) and MRI (n = 183). Secondary imaging in general did not significantly affect mRS compared to only initial imaging in peripheral centers (p = 0.206). After multivariable adjustment DS patients undergoing MRI before EVT had lower mRS scores than those without (OR 0.63; 95%-CI 0.46–0.88; p = 0.006).