GSR Tandem Lesions
(2021)Objective
To assess the safety and efficacy of different endovascular strategies—including acute carotid treatment and sequence of intervention—in patients with anterior circulation stroke and tandem occlusions.
Study Summary
• Intracranial-first approach resulted in faster reperfusion and numerically better outcomes compared to extracranial-first.
• Use of IV thrombolysis (IVT) significantly improved the chance of successful reperfusion.
Intervention
Prospective, multicenter registry analysis (German Stroke Registry) including 874 patients with tandem lesions. Compared outcomes between those receiving acute extracranial ICA treatment vs. no ICA treatment, and between intracranial-first vs. extracranial-first procedural strategies.
Inclusion Criteria
Patients with anterior circulation stroke undergoing thrombectomy, diagnosed with tandem lesions (extracranial ICA stenosis >70% or occlusion + intracranial LVO), from 25 German centers between 2015 and 2019.
Study Design
Arms: Acute ICA Treatment (Stenting/Angioplasty) vs. No ICA Treatment; Intracranial-First vs. Extracranial-First
Patients per Arm: Acute ICA Treatment: 607; No ICA Treatment: 188; Intracranial-First: 227; Extracranial-First: 267
Outcome
• Mortality: 17.1% vs. 27.1%; P<0.001
• Successful reperfusion (TICI 2b–3): 88.3% vs. 62.8%; P<0.001
• Intracranial-first approach led to shorter time to flow restoration (53.5 vs. 72.0 min; P<0.001)
• IV thrombolysis improved odds of reperfusion (OR 10.6; 95% CI 10.0–20.4; P=0.033)
Bottom Line
In anterior circulation stroke patients with tandem lesions, acute treatment of the extracranial ICA lesion is associated with better clinical outcomes and lower mortality. The intracranial-first approach might provide advantages with a shorter time to flow restoration and a higher (non-significant) probability of good outcome, without increased periprocedural complications.
Major Points
- GSR-ET is the largest prospective registry analysis of tandem lesions in acute stroke — 874 patients with tandem pathology out of 6,635 LVO strokes (13.2%) from 25 German centers.
- Key finding: acute treatment of the extracranial ICA (stenting ± angioplasty) was associated with DRAMATICALLY better reperfusion (OR 40.63, 95% CI 30-70), good outcome (39.5% vs 29.3%, p<0.001), and lower mortality (17.1% vs 27.1%, p<0.001) vs no acute ICA treatment.
- The 'intracranial-first' approach (clear the intracranial clot first, then treat the extracranial ICA) showed shorter time to flow restoration (53.5 vs 72.0 min, p<0.001) and a trend toward better outcomes (45.8% vs 33.0%, p=0.24) compared to 'extracranial-first.'
- No significant difference in periprocedural complications between approaches (18.1% no ICA treatment vs 23.9% acute ICA treatment, p=0.107) — acute stenting did NOT significantly increase procedural risk.
- Addresses a major unresolved question in thrombectomy practice: should you treat the cervical ICA lesion acutely or defer? GSR-ET supports acute treatment, but selection bias limits causal conclusions.
- 69.5% of tandem lesion patients (607/874) underwent acute ICA treatment — reflecting contemporary German practice favoring immediate intervention.
- Most common tandem pattern: extracranial ICA + MCA (67%), followed by extracranial ICA + carotid-T (32%). The underlying pathology (atherosclerosis vs dissection) was NOT differentiated.
- IV thrombolysis was a significant predictor of successful reperfusion (OR 10.58) — suggesting tPA may help soften the intracranial clot before mechanical retrieval.
- Acute ICA treatment most commonly involved stenting + PTA (67.9%), followed by PTA alone (9.9%) and stenting only (9.6%). Antiplatelet adjuncts varied widely.
- Sets the stage for randomized trials (TITAN, EASI-TOC) that will definitively determine optimal tandem lesion management strategy.
Study Design
- Study Type
- Academic, independent, prospective, multicenter, observational registry study
- Randomization
- No
- Sample Size
- 6635
- Follow-up
- 3 months
- Centers
- 25
- Countries
- Germany
Primary Outcome
Definition: Good outcome, defined as modified Rankin Scale score of 0 to 2 at 3 months.
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 29.3% | 39.5% | - | <0.001 |
Limitations & Criticisms
- Observational registry data — cannot establish causality. Patients who received acute ICA treatment may have been systematically different (better collaterals, younger, easier anatomy) from those who did not.
- Major confounding by indication: the decision to treat the ICA acutely was at the operator's discretion, introducing selection bias. Sicker patients or those with difficult anatomy may have been deferred.
- 18.1% lost to follow-up at 3 months — this is substantial and could bias results if losses were differential (e.g., if more patients who did poorly after no ICA treatment were lost).
- No differentiation between atherosclerotic stenosis and dissection as the underlying ICA pathology — these have very different treatment implications (dissection may self-resolve, atherosclerosis is fixed).
- No radiological characterization of ICH events — cannot distinguish symptomatic from asymptomatic hemorrhage, which is critical for safety assessment.
- No long-term follow-up on in-stent thrombosis or restenosis — acute stenting with DAPT in a patient with a fresh brain infarct raises bleeding and re-occlusion concerns.
- Antiplatelet regimen was not standardized — varied widely between centers (tirofiban, aspirin, clopidogrel, ticagrelor, heparin). This heterogeneity confounds safety comparisons.
- Single-country registry (Germany) — practice patterns, device preferences, and patient demographics may differ from other countries.
- The OR of 40.63 for reperfusion with ICA treatment is implausibly large and likely reflects confounding — it is tautological that treating an ICA occlusion leads to better reperfusion of the downstream territory.
Citation
Stroke. 2021 52:1265-1275. DOI: 10.1161/STROKEAHA.120.031797