Clinical scores unreliable in minor stroke and TIA

Research shows that imaging findings, rather than clinical scores, are the best means of predicting recurrent events in patients with minor stroke or transient ischaemic attack (TIA).

"The emphasis on early vascular imaging is generalizable and can be implemented widely in clinical practice", say Shadi Yaghi (Brown University, Providence, Rhode Island, USA) and co-researchers.

The various forms of the ABCD risk score did not in general predict recurrence in either of the two study cohorts. Although the ABCD³-I score was predictive in one cohort, this "result was driven by imaging alone", the team notes in JAMA Neurology.

All study participants had TIA or stroke with a baseline National Institutes of Health Stroke Scale score no higher than 3. The first cohort included 505 patients, 6.1% of whom had neurological deterioration or a recurrent event, and the validation cohort included 753 patients, 5.3% of whom had a recurrence.

The main independent predictor of recurrence was the presence of large-vessel disease on vascular imaging, which was associated with 6.69- and 8.13-fold increases in recurrence risk in the first and validation cohorts, respectively.

There was a tendency for the presence of an infarct on neuroimaging to be predictive in the validation cohort. In the other cohort, it could not be included in multivariate analysis, because only one patient with a recurrence had negative imaging findings. An infarct was present in 96.8% of patients with recurrence versus 49.4% without in the first cohort and 71.1% versus 46.6% in the validation cohort.

Of note, the recurrence rate in patients with neither of these predictors was no more than 2%, rising to about 10% for those with infarcts, 20% for those with large-vessel disease and 30% for those with both.

"Our study emphasizes the importance of urgent parenchymal and vascular imaging to risk stratify patients" with minor ischaemic stroke or TIA, say the researchers. They add that "whether rapid outpatient evaluations can be streamlined by our predictors remains to be studied in a clinical trial."

In a linked editorial, Deena Nasr and Robert Brown, both from the Mayo Clinic in Rochester, Minnesota, USA, suggest that further large prospective studies including the latest imaging techniques may result in a "moderate improvement" in predictive tools.

But they conclude that "given the prediction score challenges noted thus far, it is unlikely that any predictive scale will entirely replace the expertise and judgement of a well-trained stroke specialist in making a decision regarding the optimal setting, evaluation type, and level of urgency following presentation with TIA or minor ischemic stroke."

By Eleanor McDermid

Licensed from medwireNews with permission from Springer Healthcare Ltd. ©Springer Healthcare Ltd. All rights reserved. Neither of these parties endorse or recommend any commercial products, services, or equipment.