To evaluate the variability of radiographic measurement of knee alignment by different observers, as well as repeated measurements by the same observer, standing anteroposterior radiographs of both knees of 36 patients presenting with knee pain were analyzed. Four physicians independently measured the anatomic tibiofemoral angle of both knees for a total of 72 measurements for each observer. These measurements were then repeated 1 month later in a random and blinded fashion. The same handheld goniometer was used for all measurements. The second measurement was within 3.1 degrees of the first measurement 95% of the time, and within 4 degrees 98% of the time. The maximum difference was 6 degrees. Among all four observers, measurements were within 3.7 degrees of each other 95% of the time with a maximum difference of 6 degrees. In addition to radiographic measurement, one physician also performed clinical measurement of knee alignment on these 36 patients using the same handheld goniometer. The clinical measurement was within 5 degrees of the same clinician's radiographic measurement 95% of the time with a maximum difference of 7 degrees. This variability in measuring radiographic alignment should be considered when making decisions on the need for surgical intervention or when evaluating results of procedures that relate to coronal plane alignment of the knee.
|Original language||English (US)|
|Number of pages||5|
|Journal||The American journal of knee surgery|
|State||Published - Jan 1 2001|
ASJC Scopus subject areas