Detection of significant abnormalities on lumbar spine radiographs.

The purpose of this study was to determine the proportion of significant abnormalities detected on anteroposterior (AP) and lateral radiographs of the lumbar spine when viewed separately, in a series of cases where the prevalence of abnormalities had been artificially increased. Five radiologists of varying experience were required to report separately on the AP and lateral films of 300 cases in which randomly included were 30 cases with metastatic disease, a disc infection or an inflammatory spondylitis. At a later date, unaware of their initial observations, the radiologists repeated the exercise reviewing all the films together. As might be expected the false positive rate was relatively high, particularly in the least experienced radiologists' responses. Nevertheless the overall results indicate that the majority of early inflammatory spondylitis cases will be missed on a solitary lateral film as will many of the metastases. Conversely, fractionally more of the disc infections were observed on the lateral film than on the AP. Further analyses in terms of sensitivity, specificity, positive and negative predictive values are presented. The authors conclude that it is preferable to reduce the overall number of lumbar spine examinations performed, by adherence to accepted guidelines, than to prejudice the detection of significant, albeit rare, abnormalities by restricting the routine series of radiographs.