A multi-observer study examining the radiographic visibility of fishbone foreign bodies.

The use of plain radiographs to localize a suspected fishbone foreign body is the subject of controversy. Accordingly radiographs of 14 species of fishbone, impacted in a soft tissue phantom, were assessed by a series of observers from the ENT department (consultant surgeons, senior registrars and house officers). The agreement was assessed by graphical description of the data and tested by a Spearman's rank correlation test. The overall results showed that, for the clinician, radiography is very useful to detect the bones of: cod, haddock, lemon sole, cole fish, grey mullet and plaice; useful for red snapper, monk fish, gurnard and salmon; and unhelpful in detecting bones from herring, pike, mackerel and trout. The use of radiographs to locate these impacted fishbones can be rationalized in the light of these findings.