← Back

CLOTBUST

Ultrasound-Enhanced Systemic Thrombolysis for Acute Ischemic Stroke

Year of Publication: 2004

Authors: Andrei V. Alexandrov; Carlos A. Molina; James C. Grotta; Zsolt Garami; Shiela R. Ford; Jose Alvarez-Sabin; Joan Montaner; Maher Saqqur; Andrew M. Demchuk; Lemuel A. Moyé; Michael D. Hill; Anne W. Wojner; for the CLOTBUST Investigators

Journal: The New England Journal of Medicine

Citation: N Engl J Med. 2004;351(21):2170-2178.

Link: https://www.nejm.org/doi/full/10.1056/NEJMoa041175

PDF: https://www.nejm.org/doi/pdf/10.1056/NEJMoa041175

Clinical Question

In acute ischemic stroke due to middle cerebral artery occlusion treated with intravenous t-PA within 3 hours, does continuous 2-MHz transcranial Doppler monitoring enhance early arterial recanalization and improve recovery compared with sham monitoring?

Bottom Line

Continuous 2-MHz transcranial Doppler during IV t-PA significantly increased early complete recanalization and the composite of early recanalization or dramatic clinical recovery without increasing symptomatic intracerebral hemorrhage; trends toward better 3-month outcomes were not statistically significant.

        Major Points
        CLOTBUST was the first randomized trial of ultrasound-enhanced thrombolysis (sonothrombolysis) — demonstrating that continuous 2-MHz TCD monitoring during IV tPA more than DOUBLED complete recanalization rates (46% vs 18%, p<0.001).
126 patients with MCA occlusion confirmed by TCD (TIBI grades 0–3) treated with IV tPA within 3 hours, randomized to continuous 2-MHz TCD monitoring vs sham monitoring for 2 hours.
Primary combined endpoint (complete recanalization by TIBI criteria OR dramatic clinical recovery within 2 hours) favored active TCD: 49% vs 30% (RR 1.6, 95% CI 1.03–2.6, p=0.03).
The mechanism: ultrasound energy at 2 MHz enhances tPA-mediated fibrinolysis by mechanically agitating the clot surface, increasing tPA penetration into the thrombus, and promoting acoustic streaming that delivers more drug to the clot interface.
Safety signal was reassuring: sICH rates identical at 4.8% in both groups — proving that ultrasound augmentation did NOT increase hemorrhagic risk despite dramatically increasing recanalization.
TIBI (Thrombolysis in Brain Ischemia) flow-grading system — previously validated (Demchuk 2001; Burgin 2000) — was used to diagnose MCA occlusion and monitor recanalization: grade 0 (absent), 1 (minimal), 2 (blunted), 3 (dampened), 4 (stenotic), 5 (normal). Sonographers had to pass a computerized tutorial (≥90% correct) on TIBI wave-form interpretation before enrolling patients at each site.
3-month outcomes trended favorably but were not significant (mRS 0–1: 42% vs 29%, p=0.20) — the trial was powered for the surrogate recanalization endpoint, not clinical outcomes, as a phase 2 proof-of-concept study.
Led to CLOTBUST-ER (2017): operator-independent hands-free headframe device tested in a phase 2 trial. However, the CLOTBUST-ER trial found no significant benefit, possibly due to device limitations and the evolving role of thrombectomy.
Innovative blinding: sonographers were unblinded (had to operate TCD), but treating physicians and outcome assessors were blinded. Device audio was muted and displays turned away from treating team in both groups.
Historical importance: CLOTBUST established the proof of concept for sonothrombolysis, but the field was ultimately overtaken by mechanical thrombectomy (MR CLEAN, ESCAPE, etc.) which achieves near-complete recanalization regardless of adjunctive therapies.

      

Design

Study Type: Phase 2, multicenter, randomized, placebo-controlled trial with blinded treating physicians

Randomization: 1

Blinding: Treating physicians and outcome assessors blinded to group assignment; sonographers unblinded to deliver monitoring; devices' audio muted and displays turned away from treating team

Follow-up Duration: 3 months

Centers: Memorial Hermann Hospital, Houston (62); St. Luke's Episcopal Hospital, Houston (8); Vall d'Hebron Hospital, Barcelona (51); University of Alberta Hospital, Edmonton (3); Calgary Foothills Hospital, Calgary (2)

Countries: United States, Spain, Canada

Sample Size: 126

Analysis: Comparisons by Fisher's exact test and chi-square; logistic regression adjusted for site, age, sex, baseline NIHSS, time to t-PA bolus, baseline TIBI grade, and glucose; two-sided α=0.05; prespecified primary combined endpoint; last-observation-carried-forward imputation for missing early TIBI and internal regression for missing glucose

Inclusion Criteria

Acute ischemic stroke due to MCA occlusion with residual/abnormal flow on TCD (TIBI 0–3) before t-PA bolus
IV t-PA eligibility per published criteria
Treatment initiated within 3 hours of symptom onset

Exclusion Criteria

Absence of transcranial Doppler evidence of obstructive residual flow in the middle cerebral artery (i.e., inability to demonstrate the required inclusion finding on baseline TCD).
Baseline TIBI grade 4 or 5 on TCD (grade 5 = complete recanalization; grade 4 = stenotic flow); only patients with grade 0, 1, 2, or 3 were eligible.
Standard IV t-PA exclusion criteria per published guidelines (e.g., recent surgery, bleeding diathesis, elevated INR, uncontrolled hypertension).
Isolated posterior circulation occlusion — the trial enrolled only MCA occlusions.
Inability to obtain informed consent within the 3-hour t-PA treatment window.
Participation in other investigational stroke trials was prohibited.

Baseline Characteristics

Characteristic	Control	Active
Age, mean ± SD (y)	70 ± 13	67 ± 12
Systolic blood pressure, mean ± SD (mm Hg)	158 ± 22	160 ± 22
Serum glucose, mean ± SD (mg/dL)	146 ± 69	138 ± 61
NIHSS, median	17	16
MCA occlusion - Proximal (%)	70	79
MCA occlusion - Distal (%)	30	21
Tandem 70–100% stenosis of proximal ICA (%)	27	38
Median depth of residual flow signals (mm)	50	50
TIBI grade, mean ± SD	1.6 ± 1.1	1.6 ± 0.9
Time from onset to t-PA bolus, median (min)	130	150
Treatment within 2 hours of onset (%)	43	27

Arms

Field	IV t-PA + continuous TCD monitoring	Control
Intervention	IV t-PA 0.9 mg/kg (10% bolus, 90% infusion) within 3 hours plus continuous 2-MHz TCD monitoring for 2 hours using FDA-approved diagnostic probes	IV t-PA 0.9 mg/kg within 3 hours plus sham monitoring (probe connected to inactive channel; audio muted and displays hidden)
Duration	2 hours of monitoring concurrent with t-PA administration	2 hours of sham monitoring concurrent with t-PA administration

Outcomes

Outcome	Type	Control	Intervention	HR / OR / RR	P-value
Composite of complete recanalization by TIBI criteria or early/dramatic clinical recovery (NIHSS ≤3 or decrease ≥10 points) within 2 hours after t-PA bolus	Primary	30%	49%	19.00%	0.03
Complete recanalization within 2 hours	Secondary	18%	46%		<0.001
Sustained complete recanalization at 2 hours	Secondary	13%	38%		0.002
Clinical recovery within 2 hours (NIHSS ≤3 or decrease ≥10)	Secondary	21%	29%		0.4
Recovery at 24 hours (NIHSS ≤3 or decrease ≥10)	Secondary	40% (21/52 eligible)	44% (24/54 eligible)		0.7
Favorable outcome at 3 months (mRS 0–1)	Secondary	29% (14/49 eligible)	42% (22/53 eligible)		0.20
Mortality at 3 months	Secondary	18%	15%		0.4
Symptomatic intracerebral hemorrhage ≤72 h	Adverse	4.8% (3/63)	4.8% (3/63)

Criticisms

Phase 2 proof-of-concept with only 126 patients — adequately powered for the surrogate recanalization endpoint but NOT for clinical outcomes (mRS at 90 days). The authors estimated 274 patients per group (~548 total) would be needed for a pivotal efficacy trial powered on 90-day mRS 0–1.
Sonographers necessarily unblinded — they had to operate the TCD device. Although treating physicians were blinded, early recanalization assessment by TCD inherently required the unblinded sonographer's involvement.
Operator-dependent technique requiring skilled TCD sonographers available 24/7 — severely limits real-world implementation. Many stroke centers lack TCD-trained personnel, especially off-hours.
Baseline imbalance: more control patients treated within 2 hours (43% vs 27%) — earlier treatment favors recanalization, potentially biasing against the treatment group. However, TCD effect remained significant after adjustment.
49% of patients were treated and randomly assigned at a single hospital (Memorial Hermann, Houston); the two Houston sites (Memorial Hermann 62 + St. Luke's Episcopal 8) together enrolled 70/126 (56%) — raising external validity concerns about generalizability to other clinical settings.
2-MHz diagnostic frequency used — this is standard diagnostic TCD frequency, but subsequent research explored lower frequencies (microsphere-enhanced, 2-MHz with microbubbles in NOR-SASS) and showed mixed results, leaving optimal ultrasound parameters undefined.
Did not compare to mechanical thrombectomy — in the modern endovascular era (post-2015), the clinical niche for sonothrombolysis is unclear since thrombectomy achieves ~90% recanalization rates.
CLOTBUST-ER (2017) follow-up with hands-free operator-independent device was negative — raising questions about whether the original CLOTBUST benefit was partly due to operator expertise or continuous optimization of insonation angle.
Short 2-hour monitoring window — some re-occlusions occurred after monitoring ceased, potentially underestimating the benefit of sustained ultrasound exposure.

Funding

Supported by NINDS (1K23NS02229-01, 1P50NS044227); Canadian funding agencies (CIHR, Alberta Heritage Foundation); investigator-sponsored (Genentech protocol A2207s); diagnostic ultrasound equipment provided at no charge by Spencer Technologies, DWL, Multigon, and Nicolet.

Based on: CLOTBUST (The New England Journal of Medicine, 2004)