Reproducibility of the knee signature system.

To assess trial-to-trial, installation-to-installation, and day-to-day effects of the Knee Signature System, we studied anterior/posterior translation measurements in six normal subjects. The protocol consisted of testing both knees on 6 days with three installations per day and three trials per installation. We recorded anterior, posterior, and total anterior/posterior translation at +/- 20 and +/- 40 pounds of force at 30 degrees of flexion. Additionally, after measuring anterior/posterior translation during passive knee extension, active extension, and active extension with 15 pounds of weight at the ankle, we calculated the difference in anterior translation at 30 degrees of flexion between each of these three conditions. We analyzed the size of the effects and computed 90% confidence limits using analysis of variance. Means and 90% confidence limits for total anterior/posterior translation measured at +/- 20 pounds of force were: right knee, 6.8 +/- 2.3; left knee, 8.2 +/- 2.0; and right/left difference, 1.5 +/- 2.7 mm. Left knee translation means were significantly greater than right knee means in all tests. Day-to-day effects, while statistically significant for right and left knee measurements, showed no significant effects for right/left differences. Trial-to-trial and installation-to-installation effects were not significant for any parameter. Right/left differences during knee extension testing were smaller than for the stress test, measuring 0.1 mm or less, but confidence limits were larger, being +/- 3.7 mm for active extension-passive knee extension, +/- 1.5 mm for 15 pounds of weight at the ankle-active extension, and +/- 4.6 mm for 15 pounds of weight at the ankle-passive knee extension. Active extension produced 6.3 mm more anterior translation at 30 degrees of flexion than passive extension.(ABSTRACT TRUNCATED AT 250 WORDS)