TL;DR: ⚪ Inconclusive — thrombectomy for large infarct selected by noncontrast CT alone showed no significant functional benefit, but the wide credible interval doesn't rule out benefit either; more hemorrhage occurred with thrombectomy.
1. Publication
- Title: Thrombectomy for Stroke With Large Infarct on Noncontrast CT: The TESLA Randomized Clinical Trial
- Acronym: TESLA
- Year & Journal: JAMA, published September 23, 2024 (332(16):1355-1366)
- Citation: The Writing Committee for the TESLA Investigators. JAMA. 2024;332(16):1355-1366. doi:10.1001/jama.2024.13933
2. Context & Rationale
Background: Recent large-infarct thrombectomy trials used varied imaging modalities and time windows for selection; noncontrast CT is the most common stroke imaging method globally, but its adequacy alone (without perfusion imaging) for selecting large-infarct patients for thrombectomy was untested.
Research Question/Hypothesis: In patients with large infarcts identified by noncontrast CT alone within 24 hours, does thrombectomy improve 90-day functional outcome compared with medical management alone?
3. Design & Methods
- Study Type: Randomized clinical trial, internet-based stratified randomization
- Setting & Centers: 47 stroke centers
- Population: 300 patients with anterior-circulation LVO, ASPECTS 2-5, within 24h of onset
- Intervention: Endovascular thrombectomy plus medical care (n=152)
- Comparator: Medical care alone (n=148)
- Statistical Power & Follow-Up: Primary: 90-day utility-weighted mRS (UW-mRS). Secondary: mortality, symptomatic ICH, radiographic hemorrhage.
4. Key Results
300 patients randomized.
Outcome | Thrombectomy | Medical Care | Effect Size | 95% CI | Notes |
90-day UW-mRS (primary) | Mean 2.93 | Mean 2.27 | Adjusted difference 0.63 | Crosses null | Not statistically significant |
90-day mortality | 35.3% (53/150) | 33.3% (49/147) | RR 1.06 | 0.77–1.45 | No significant difference |
Intracranial hemorrhage (any) | Higher | Lower | — | — | More ICH with thrombectomy, including symptomatic |
Decompressive hemicraniectomy | 21.9% | 14.8% | RR 1.48 | 0.91–2.42 | Numerically more with thrombectomy |
Important nuance: The credible interval around the effect estimate encompassed both no effect and a clinically meaningful benefit — the trial did not definitively rule out benefit, per independent commentary.
5. Internal Validity Assessment
Moderate-sized (300-patient) RCT with internet-based stratified randomization (by age, ASPECTS, NIHSS, time window). Overall: Moderate — clear null on the primary endpoint, but the wide credible interval (encompassing both null and meaningful benefit) means this should be read as "not proven beneficial" rather than "proven ineffective," per independent commentary; increased ICH is a genuine safety signal.
6. External Validity Assessment
47-center population selected using noncontrast CT alone (not perfusion imaging) — directly relevant to the many centers worldwide without advanced perfusion-imaging capability for large-infarct patient selection.
7. Strengths & Limitations
Strengths: Directly tests noncontrast-CT-alone selection (the most globally accessible imaging modality), addressing a genuine implementation-relevant question distinct from perfusion-imaging-based trials (e.g., SELECT2, TENSION).
Limitations: Modest sample size (300) with a wide, inconclusive credible interval; increased hemorrhage with thrombectomy; single imaging-modality approach may under- or over-select compared with perfusion-based methods.
8. Interpretation & Practice Impact
Does not demonstrate a clear functional benefit from thrombectomy in noncontrast-CT-selected large-infarct stroke, though the result does not definitively exclude benefit either — an important, imaging-modality-specific nuance in the broader large-infarct thrombectomy evidence landscape.
9. Controversies & Subsequent Evidence
An accompanying commentary ("CT-Guided Thrombectomy for Large Core Stroke Up to 24 Hours — Another Piece in a Complex Puzzle") and a subsequent review ("Endovascular Treatment of Large Core Infarcts: no Limits?") situate TESLA alongside SELECT2, TENSION, and the Costalat trial (this handbook) as part of a genuinely complex, evolving large-infarct thrombectomy literature where imaging-selection modality appears to matter.
10. Summary & Executive Takeaway
Summary: TESLA randomized 300 patients with large infarcts selected by noncontrast CT alone (within 24h) to thrombectomy plus medical care or medical care alone. The primary outcome (90-day UW-mRS) showed no significant difference (adjusted difference 0.63, credible interval crossing null), with more intracranial hemorrhage in the thrombectomy group.
Overall Takeaway: Thrombectomy did not demonstrate a clear functional benefit when large-infarct stroke patients are selected using noncontrast CT alone, though the wide, inconclusive credible interval means this should be read as "not proven," not "disproven" — an important nuance for centers relying on noncontrast CT (rather than perfusion imaging) for large-infarct patient selection.
11. Bibliography
- Costalat V, Jovin TG, Albucher JF, et al. Trial of thrombectomy for stroke with a large infarct of unrestricted size. N Engl J Med. 2024;390(18):1677-1689 (this handbook).
- Sarraj A, et al. SELECT2 1-year outcomes. Lancet. 2024;403(10428):731-740.