Reader response: Prognostic value of "tissue-based" definitions of TIA and minor stroke: Population-based study.

Abstract

In their study on DWI positivity after TIA and minor stroke,1 Hurford et al. pay little attention to a major issue plaguing many such studies, differentiating patients with TIA from those with other transient neurologic symptoms. Hurford et al. excluded 725 patients from their analysis because they were given an alternative diagnosis by the senior author, although why that diagnosis was given remains unclear. Including those patients, or even some fraction of them, would likely have altered their results. A recent prospective study found that even patients who neurologists believed were unlikely to have had TIA had a nontrivial rate of DWI positivity on subsequent MRI.2 We sought to explore the problem of TIA diagnosis in our own work—namely through a systematic review3 and a retrospective analysis of neurologists' decision-making.4 Prospectively correlating patients' presenting symptoms (motor and speech vs other), demographics (especially patient sex) with neurologists' diagnosis (definite TIA, possible TIA, and not TIA), and MRI positivity could help to arrive at criteria to ensure that patients with true TIA are not being missed. Enhanced diagnostic reliability would improve risk stratification, use of investigations, and efficacy of stroke prevention treatments.