Author + information
- Michael G. Dwyer, PhD∗ (, )
- Ronald M. Lazar, PhD and
- Robert Zivadinov, MD, PhD
- ↵∗Buffalo Neuroimaging Analysis, University of Buffalo, 100 High Street, D2, Buffalo, New York 14203
Thank you for the opportunity to respond to Dr. Abrams’s letter. Although we disagree on some points, we greatly appreciate his interest in our work and believe the discrepancies may be related to simple misunderstandings.
First, Dr. Abrams notes that we did not find a relationship between worsening cognitive deficits and new diffusion-weighted imaging lesions in the SENTINEL study (1). We need to point out, however, that we in fact found such a relationship in that study (reported in the paper and as Online Figure 1B ). In the paper, we did observe that prior transcatheter aortic valve replacement studies did not observe such a relationship, which may be the source of the confusion.
Although somewhat outside the scope of the paper, we agree with Dr. Abrams’s main point that the question of protected territories is important in a broader context. Let us examine his claims. He performs a calculation and concludes that 65% of lesions were outside protected territories. Unfortunately, this seemingly straightforward calculation is invalid for 2 reasons. First, and most important, the estimate must be performed on the control group, not the treatment group. To see why, one can imagine an ideal device that could prevent every single lesion in its protected zone; then, by definition, 100% of lesions would be in the “nonprotected” region, no matter how small that region was. Updated with this important correction, the result of the calculation is 42%. Second, subtracting and dividing medians as proposed is not permitted statistically. The more appropriate approach is to calculate the ratio across subjects and then take its median. When both of these corrections are taken into account, the result is considerably lower: 35%.
Finally, we appreciate Dr. Abrams’s door analogy. However, the problem with this assertion is that it implies that emboli will actively seek out the back door after being shut out of the front, rather than just continuing down the street.
Please note: Drs. Dwyer, Lazar, and Zivadinov have received institutional funding and consulting fees from Claret Medical. Dr. Lazar holds stock options in Claret Medical.
- 2018 American College of Cardiology Foundation