Abstract
Inter- and intra-observer variation has been noted in the analysis of radiographic examinations with regard to experience of surgeons, and the monitors used for conducting the evaluations. The aim of this study is to evaluate inter/intra observer variation in the measurement of mechanical alignment from long-leg radiographs.
40 patients from the elective waiting list for TKA underwent long leg radiographs pre-operatively and 6 months post-operatively (total of 80 radiographs). The x-rays were analysed by 5 observers ranging in experience from medical student to head orthopaedic surgeon. Two observers re-analysed their results 6 months later to determine intraobserver correlation, and one observer re-measured the alignment on a different monitor. These measurements were all conducted blindly and none of the observers had access to the others’ results.
80 radiographs were analysed in total, 40 pre-op and 40 post-op. The mechanical alignment was analysed using Pearson's correlation (r = 0 no agreement, r = 1 perfect agreement) and revealed that experience as an orthopaedic surgeon has little effect on the measurement of mechanical alignment from long leg radiograph. The results for the different monitor analysis were also analysed using Pearson's correlation of long leg alignment. Monitor quality does seem to affect the correlation between alignment measurements when reviewing both intra and inter observer correlation on different computer monitors.
Surgical experience has little impact on the measurement of alignment on long leg radiographs. Of greater concern is that monitors of different resolution can affect measurement of mechanical alignment. As there might be a range of monitors in use in different institutions, and also in outpatient clinics to surgical theatres, close attention should be paid to the implications of these results.
