We thank Drs Broderick and Tomsick for their insights regarding our editorial ‘The tribulations of stroke trials’.¹ As the title of the editorial implies, acute stroke trials are difficult to design and complete. Our editorial was not meant to discount the results of the Interventional Management of Stroke III (IMS III), or those of the Retrieval and Recanalization of Stroke Clots Using Embolectomy (MR-RESCUE) and SYNTHESIS Expansion trials, but rather to temper the interpretation of the results and to emphasize potentially confounding methodological nuances that should be examined prior to designing future stroke trials.

In the first paragraph of their letter, Broderick and Tomsick refer to IMS III as a ‘failed’ endovascular stroke trial. With regard to the pre-trial hypothesis that the combination of a partial dose of tissue plasminogen activator (tPA) and endovascular therapy would prove superior to IV tPA alone, the term ‘failed’ is technically accurate. But like other words and phrases, the term ‘failed’ regarding IMS III is already being misused by some groups to imply that endovascular therapy for stroke is ineffective. In fact, IMS III demonstrated that endovascular treatment strategies worked at least as well as IV tPA alone. Importantly, in certain subsets of patients in IMS III, there was a strong trend towards the superiority of endovascular techniques. Furthermore, IMS III demonstrated very favorable safety profiles for both treatment arms. Thus one objective of our editorial was to temper the interpretation, or misinterpretation, of the word ‘failed’ as it relates to the endovascular arm of IMS III.

The main methodological nuances regarding IMS III raised in our editorial centered on ‘index disease’ and technological advances during the study period. Broderick and Tomsick relate that the index disease for IMS III was not large vessel occlusion but rather major ischemic stroke, as defined by a National Institutes of Health Stroke Scale (NIHSS) score of >8. In point of fact, the NIHSS cut-off that was chosen as an entry criterion for the study was a surrogate for large vessel occlusion based on prior data. Non-invasive vascular imaging was not utilized as an entry criterion at the beginning of the study due to the immaturity of CT angiography (CTA) technology and the unavailability of rapid MR angiography at some study sites. Part way through the trial, the utility of CTA in documenting pretreatment large vessel occlusion was recognized by the study investigators and was added as an enrollment criterion. A major aim of our editorial was to emphasize the importance of pre-enrollment documentation of large vessel occlusion in the design of future endovascular stroke trials. Only through the selection of a valid ‘index disease’ will such trials accurately assess the comparative available therapies.

Similarly, the discussion in our editorial of endovascular device advances during the study period of IMS III was meant to point out the difficulty of conducting stroke trials during periods of rapid technological evolution. It is certainly unfortunate that IMS III and MR-RESCUE took so long to enroll patients within the context of such an unprecedented period of progress in both neuroimaging and neuroendovascular therapeutics. The reality is that the vast majority of patients in the endovascular arm of IMS III were imaged, triaged, and treated using technologies and approaches which are now obsolete. While the IMS-III investigators did ultimately attempt to incorporate some of these new technologies into the trial during its prolonged recruitment period, it is erroneous to imply that these data can, or should, be viewed as reflective of contemporary stroke treatment as most of the endovascular patients within the trial actually did not receive contemporary stroke treatment.

The IMS III trial provides a platform for the development of future investigations. Methodological flaws, as we delineated in our editorial, underscore the need for critical evaluation of the treatment methods, current status of endovascular technology, and the specific disease that we seek to analyze prior to forthcoming trial development. The future of our field hinges on the performance of such well conceived trials.