Commentary on: "safety of carotid intervention following thrombolysis in acute ischaemic stroke".

Indications for carotid endarterectomy (CEA) have evolved as stroke services have improved. Hyperacute stroke units are referring more patients with a significant carotid stenosis, some within hours of thrombolysis. In their metaanalysis, Mandavia et al. pooled the evidence to date to attempt to assess 30-day safety of urgent CEA for these patients. However, care must be taken with the interpretation and contextualization of these results. The introduction and discussion of this meta-analysis quote from NICE guidelines, ECST and NASCET, which is misleading in this setting. NICE guidelines have no pathway for carotid surgery after thrombolysis. Patients who received thrombolysis were not included in ECST, NASCET, or any other randomised trial of carotid intervention to date, so results are not directly comparable with the 30-day stroke and death rate from this analysis. Although the headline estimate of a 5% 30-day stroke/ death rate may seem reasonable, this is not being considered in the context of outcomes after thrombolysis for stroke. The reality is that the fate of the matched thrombolysed patient treated with modern best medical therapy is unknown. There are no published subgroup analyses from the stroke thrombolysis trials with which to compare these results. Recurrent ischaemic stroke in the first 7 days after thrombolysis was actually rare (1%). The 30-day estimate of stroke or death after CEA in this analysis was between 2% and 9%. This large confidence interval reflects the low number of events (4 strokes and 4 deaths) from which it was derived. If the headline estimate were 9%, as it could be, would the result still be considered evidence of safety? This highlights the major flaw in these data, the low-quality assessment of the included studies highlights the other. Long-term outcomes after thrombolysis for stroke were relatively poor. Pooled 90-day mortality was 17%. Thirtyone to 90-day mortality was 4.4% for patients with favourable clinical outcomes. Therefore the long term benefits of stroke prevention after CEA in this group are