Sonolysis during carotid endarterectomy: randomised controlled trial

Dear Editor

The Sonolysis in Prevention of Brain Infarctions During Internal Carotid Endarterectomy (SONOBIRDIE) trial is a well-conducted, multicentre, randomised controlled study evaluating the efficacy and safety of sonolysis, a low-intensity pulsed wave ultrasound technique in carotid endarterectomy (CEA). The trial investigates a neurovascular protection strategy aimed at reducing perioperative cerebral events. The primary endpoint was a composite of ischaemic stroke, transient ischaemic attack (TIA), and death within 30 days, while secondary outcomes included the appearance of new ischaemic lesions on brain magnetic resonance imaging (MRI), and the key safety endpoint was the incidence of intracranial bleeding.

The trial is methodologically strong, with a sham-controlled design, appropriate blinding (except for the sonographer), adequate statistical power, and clinically meaningful imaging and composite outcomes. The observed absolute risk reduction of 5.5% in the composite outcome (P < 0.001) is notable and suggests that sonolysis may offer significant benefit in improving perioperative outcomes following CEA.

However, several important limitations should be acknowledged. First, although blinding was maintained across most of the trial team, the sonographers were unblinded, which, while practically unavoidable, introduces the potential for bias. Second, the trial employed a variety of commercially available transcranial Doppler ultrasound machines without a unified protocol for sonolysis delivery across centres. This lack of standardisation may have introduced variability in insonation parameters, operator technique, and therapeutic dose delivery, potentially affecting reproducibility and limiting the translation of these findings into routine practice. Standardised, purpose-designed sonolysis devices would be beneficial in future studies.

In addition, given the study’s five-year duration, the natural evolution of surgical and ultrasound technique over time, along with increasing operator experience, may have contributed to improved outcomes. A year-by-year analysis of recruitment and complication rates by centre would be helpful to identify potential learning curve effects. The study population also included a clinically heterogeneous mix of symptomatic (45%) and asymptomatic (55%) carotid stenosis patients. These cohorts differ significantly in stroke risk, treatment indications, and guideline recommendations. Stratified reporting or subgroup analysis would clarify whether sonolysis confers equal benefit across these groups—particularly important given the ongoing debate over intervention in asymptomatic carotid disease.

Generalisation of the findings may also be limited by the study’s setting. All participating centres were based in Central Europe, and while the intervention demonstrated efficacy within this context, broader applicability to more diverse health systems remains uncertain. For instance, despite the observed relative risk reduction, both the composite outcome and the 30-day rate of stroke or TIA were higher than those reported in the 2024 UK National Vascular Registry1. Moreover, the finding that female sex independently increased the risk of adverse outcomes (Odds Ratio 2.0) is an important observation that aligns with previous reports but remains underexplored in this study. Further research should investigate sex-based differences in procedural outcomes more thoroughly.

The early termination of the trial following interim analysis for efficacy, while justified, raises concerns about potential overestimation of treatment effect and reduced statistical power for evaluating secondary and safety outcomes. Furthermore, the cost implications of implementing sonolysis were not addressed. A formal health economic analysis, especially within resource-constrained systems such as the NHS, would be essential to evaluate the feasibility and sustainability for wider adoption of this procedure.

In conclusion, the SONOBIRDIE trial marks an important step toward integrating therapeutic ultrasound into vascular practice. The results are promising, particularly for improving perioperative outcomes in patients with asymptomatic carotid stenosis. However, several key questions remain regarding long-term effectiveness, centre-level variability, cost-effectiveness, and generalisability. We encourage the authors to address these issues in future analyses or correspondence to support informed clinical implementation.

References
1. VASGBI. National Vascular Registry Report 2024: Summary for Anaesthetists. VASGBI, https://vasgbi.com/research-audit/nvr-summary-2024/ (Accessed 24 March 2025).