We greatly appreciate the opportunity to reply to the very thoughtful and insightful letter by LeMaster, Mueller, and Sinacore (1). We believe that the comments raised are very important and deserve this sort of forum.
As to the first suggestion that waist-mounted accelerometers may not be robust enough to capture short shuffling steps taken by many of our less active patients with diabetes, our only response is that we agree. While the devices we used were highly sensitive and could be very finely calibrated, they do not approach the ability of ankle-worn devices to capture very fine movements of the lower extremity. Our decision to use waist-mounted devices was based on practicality. Waist-mounted pedometers and accelerometers (rather than ankle-worn devices) are highly acceptable to our patients and, we believe, led to a great deal of study adherence and acceptability. One outstanding study that we discuss in our manuscript, that of Maluf and Mueller (2), lasted several days (rather than several months). Many of our patients have a good deal of trouble applying their shoes and stockings in the morning secondary to their morphology and limited joint mobility. Applying an ankle-worn accelerometer (or activity monitors worn on multiple sites) over a prolonged period might not be all that attractive in this setting. That said, we may be entirely incorrect in this assumption. We have more recently begun using ankle-mounted devices with some success. The other fine study discussed by LeMaster et al. (3), used daily activity diaries, which, it might be argued, might be somewhat less accurate than a waist-mounted accelerometer in detecting subtle activity.
As to the second point made by the authors, we completely agree that assessment of total force per day or continuous measurement of pressure-time integral may be a critical factor in assessing risk for reulceration. We look forward to larger trials that can incorporate this metric.
The issue as to the potential for confounding factors affecting activity is a compelling one. It is entirely possible that this was the case in this study. However, the truly profound difference that was noted in variability of activity is a compelling finding and worthy of further investigation.
Finally, it (again) seems entirely reasonable, following the work from Mueller and Maluf (4) and, to a lesser extent, our unit, that modulating activity might have the potential to either increase or lower risk based on dose and duration. Perhaps a slowly applied (with continuous feedback to the patient and clinician) steady increase in activity could reduce risk by improving the quality of plantar tissue and its response to stress. Conversely, we believe, based on these aforementioned works, too little (and perhaps too much) activity might leave the patient outside of a zone of safety. Discovering the upper and lower limits of this zone will be a goal, we are sure, for many investigators in this most fascinating area of inquiry.