Interobserver and intraobserver variation in the assessment of the healing of tibial fractures after intramedullary fixation

The reliability of the radiological assessment of the healing of tibial fractures remains undetermined. We examined the inter- and intraobserver agreement of the healing of such fractures among four orthopaedic trauma surgeons who, on two separate occasions eight weeks apart, independently assessed the radiographs of 30 patients with fractures of the tibial shaft which had been treated by intramedullary fixation. The radiographs were selected from a database to represent fractures at various stages of healing. For each radiograph, the surgeon scored the degree of union, quantified the number of cortices bridged by callus or with a visible fracture line, described the extent and quality of the callus, and provided an overall rating of healing. The interobserver chance-corrected agreement using a quadratically weighted kappa (K) statistic in which values of 0.61 to 0.80 represented substantial agreement were as follows: radiological union scale (kappa= 0.60); number of cortices bridged by callus (kappa= 0.75); number of cortices with a visible fracture line (kappa= 0.70); the extent of the callus (kappa= 0.57); and general impression of fracture healing (kappa= 0.67). The intraobserver agreement of the overall impression of healing (kappa= 0.89) and the number of cortices bridged by callus (kappa= 0.82) or with a visible fracture line (kappa= 0.83) was almost perfect.