The “collateral story” may be more complicated than previously thought, highlighting the need for more research into their evolution and role.
Leptomeningeal collateral status is not predictive of functional independence or mortality among patients who undergo endovascular therapy for acute stroke beyond 6 hours, according to a prespecified analysis of the DEFUSE 3 trial.
The findings were presented recently at the International Stroke Conference in Honolulu, HI, and published online ahead of print in Stroke.
“Given that collaterals have been so robustly associated with outcome previously, we were convinced going into this study that we would see the same association, and we were surprised to not find it,” lead author Adam de Havenon, MD (University of Utah, Salt Lake City), told Neurovascular Exchange.
“The difference between our study and most of the prior research is the time window that patients’ collaterals were measured in,” lead author Adam de Havenon, MD (University of Utah, Salt Lake City), told Neurovascular Exchange. “The majority of the prior research has been in time windows closer to the onset of stroke, so the 4.5-hour window for tPA or 6-hour window for standard thrombectomy.”
For this prespecified analysis of DEFUSE 3, de Havenon and colleagues included the 130 patients with CT angiography as their baseline imaging, rating collateral status using the validated scales described by both Tan and Maas. The primary outcome was functional independence at 90 days (mRS ≤ 2).
Overall, collaterals were rated poor in 33 patients (25%) and good in 97 (75%). The presence of good collaterals was associated with significantly smaller ischemic core volume and less ischemic core growth.
Typically, the proportion of patients with good/poor collaterals is closer to a 50/50 split, explained de Havenon. Although the higher rate of good collaterals could be at least partially explained by the fact that this cohort of patients had to meet DEFUSE 3 inclusion criteria, the core lab nevertheless reevaluated collateral status using Maas criteria, which yielded a more typical split of 47% with good and 53% with poor collaterals. The relationship between collateral status and outcomes remained the same regardless of the grading scale used, however.
No Link With Functional Independence or Mortality
There was no difference between patients with good and poor collaterals with respect to functional independence at 90 days (30% vs 39%; P = 0.3), even after adjustment for treatment arm (OR 0.61; 95% CI 0.26-1.45). Similarly, there was no difference between good and poor collateral patients with respect to stroke-related death (19% vs 24%; P = 0.5) or the treatment effect of endovascular thrombectomy (P = 0.8).
“What [also] surprised me was that . . . the baseline collaterals in this cohort were not associated with many of the traditional things that have been considered predictors of collateral status [such as] older age [and] history of hypertension,” de Havenon said, adding, “That suggests that maybe this was a somewhat unique cohort or that collaterals don’t behave the same way later on than they do in the earlier time windows.”
But one limitation common to “all collateral research is we are just looking at one point in time and collaterals are dynamic,” de Havenon observed. “We don’t know enough about their natural history, and we don’t have practical ways of measuring them repeatedly. [So] the collateral story may be more complicated than we think.” He pointed out that all of the imaging tools currently available provide good information about large leptomeningeal collaterals but not end arterioles and small vessels that form the bulk of retrograde blood flow into collaterals.
Source:
de Havenon A, Mlynash M, Kim-Tenser MA, et al. Results from DEFUSE 3: good collaterals are associated with reduced ischemic core growth but not neurologic outcome. Stroke. 2019;Epub ahead of print.
Disclosures:
de Havenon reports no relevant conflicts of interest.
