C. Gatlin, G. Vanbeek, J. Aguilera, J. Sattgast, T. LaPier

Eastern Washington University, Cheney, WA

Median sternotomy is frequently used during cardiac surgery. Patients are commonly prescribed Sternal Precautions (SP) during recovery to minimize post-surgical complications, but resumption of physical activity and arm movement are important for recovery. PURPOSE: To develop an objective methodology to distinguish between specific, purposeful arm movements and total, whole-body arm movements using 3 accelerometers. METHODS: ActiGraph GT9X Link, a validated 3-axis accelerometer (3AA), was used to measure movement. Subjects wore 3 activity monitors (3xAM), one on each wrist and at the waist for 4 consecutive days. The sum of acceleration measured in 3 axes was calculated and divided by 1000, resulting in a vector magnitude kilocount (VMC) expressed per day. Raw data from the wrist 3AA were filtered by a low-frequency extension filter. Data obtained from the waist monitor was used to determine periods of ambulation. Data was broken up into 15 second epochs during which arm movement was assumed to be arm swing and therefore non-purposeful when 5 or greater steps were taken. This allowed determination of both total arm movement and specific arm movement not associated with gait. Cross-sectional 3xAM data from adults > 65 years old (n = 17) were compared to data obtained with established activity measurements. Next, we analyzed the ability of the 3xAM technique to discriminate between whole body and arm specific activity. Statistical analyses included correlations, ANOVA, and t-tests with Bonferroni correction (P < 0.05). Currently, we have initiated 3xAM longitudinal data collection before, during, and after simulated sternal precautions. RESULTS: We found significant correlations (r = 0.4 – 0.7) between the 3xAM data and activity measures. When comparing the raw, filtered, and gait removed data, the raw data (15,000 VMC) was significantly greater than the modified data (6,000 VMC). Pilot data demonstrated a trend showing decreased activity during and immediately after SP that did not return to prior activity levels. CONCLUSIONS: The 3xAM technique described shows concurrent validity and successfully differentiates arm movement from whole body activity. Longitudinal pilot data suggests 3xAM VMC reflect arm movement patterns before, during, and after SP, indicating utility for measuring return to function.

This document is currently not available here.