The likelihood of good clinical and safety outcomes after endovascular thrombectomy in patients with large core on CT ASPECTS (Alberta stroke programme early CT score) declined substantially based on CT perfusion (CTP) core volume and time. This conclusion was presented during a late-breaking session by Amrou Sarraj, associate professor at the Department of Neurology, UT McGovern Medical School, Houston, USA, at the International Stroke Conference (ISC; 6–8 February, Honolulu, USA).
Although Sarraj acknowledged that CT ASPECTS detects hypodense tissue, CTP identifies regions of very low blood flow or volume, meaning that large core definition may differ between CT and CTP. As the majority of upcoming large core randomised controlled trials (RCT’s) use only CT for patient inclusion, Sarraj and colleagues set out to evaluate the variability of endovascular thrombectomy outcomes in patients with large core on CT in relation to CPT core volume and time.
In order to identify anterior circulation occlusions (internal carotid artery, middle cerebral artery, M1/M2) with no baseline disability and treated up to 24 hours, Sarraj and colleagues combined the cohort of a prospective multicentre study of imaging selection (SELECT) with the prospective multicentre registry—TREVO. CTP core volume was determined by automate software, while independent core labs adjudicated ASPECTS.
According to Sarraj, the primary outcome was defined as a good 90-day modified Rankin Scale (mRS; of 0–2), while symptomatic intracerebral haemorrhage, neurological worsening, and mortality were secondary outcomes; all of which were stratified by perfusion volume (<50cc, 50–100cc, and >100cc) and time.
Of the 2,453 patient combined cohort, 221 had ASPECTS ≤5, while Sarraj reported that the baseline characteristics were similar between the two SELECT and TREVO cohorts. In the current study, good outcome rate was 35% in ASPECT ≤5. Of importance, Sarraj remarked that outcome rates decreased as CTP core volume increased (<50cc: 48%, 50-100cc: 20%, >100cc: 0%; p trend=0.03). Further, Sarraj and his colleagues observed that for every 10cc increase in CTP core volume, the probability of a good outcome decreased by 27%. Moreover, a substantial reduction in good outcome was witnessed for patients with longer times from onset.
In terms of safety outcomes, both neurological worsening (4%, 22%, 100%, p trend<0.001) and mortality (4%, 30%, 50%, p trend=0.006) significantly increased with increasing CTP core volume. Similarly, rates of symptomatic intracerebral haemorrhage also significantly increased from 4% to 75% for volumes ≤100 compared to those >100cc.
Despite the contribution of this data in relation to its implications for future trials, Sarraj alluded to particular caveats present in the current study. He acknowledged that unlike the initial SELECT trial, there were no medical management controls present in the combined cohort, meaning that no benefit conclusions can be made. Furthermore, imaging selection in the TREVO registry was carried out by an investigators discretion.
Regardless of the limitations, Sarraj maintained that in terms of patients with large core on CT—as the volume increased and time to reperfusion progressed—the rates of functional independence declined, while both mortality and symptomatic haemorrhage rates increased. Acknowledging the current study’s contribution to optimising the stroke patient pathway, Sarraj also highlighted the scope of this data in relation to helping inform the design of future large core RCT’s.