Reliability and necessity of computerized tomography in distal tibial physeal injuries

Complex distal tibial physeal fractures can be difficult to characterize on plain radiographs. The role of computed tomography (CT) scans in the evaluation and treatment decision of these injuries is unclear. We aimed to determine whether or not the addition of CT would improve the reliability of fracture classification and treatment decision.

Five independent observers evaluated 50 distal tibial physeal fractures on 2 separate occasions for Salter Harris (SH) classification and treatment decision (surgical/nonsurgical) using plain radiographs (round 1) and combination of radiographs and CT (round 2). During round 1, observers were asked if they would order a CT, and during round 2, they were asked if the CT was useful. These rounds were repeated at 2 to 4 weeks to assess intraobserver reliability. Statistical analyses were performed to assess inter and intraobserver reliability using Kappa coefficient (κ).

Intraobserver reliability for SH classification showed substantial agreement, κ=0.76 and κ=0.80, respectively, during round 1 and 2. Interobserver agreement on the SH class was lower during round 1 and 2 (κ=0.67 and κ=0.57, respectively). There also was almost perfect intraobserver and interobserver agreement in the measurement of displacement at the fracture site during both rounds 1 and 2. Intraobserver reliability for treatment decision was substantial, κ=0.74 and κ=0.80, respectively, during round 1 and 2. However, interobserver agreement for treatment decision was moderate (κ=0.48) and fair (κ=0.36), respectively, during round 1 and 2. Surgeons indicated that they would like to order CT scans for 66% of the time in round 1, but the interobserver agreement as to who would best benefit from the CT was only fair (κ=-0.23). The main purpose of ordering the CT was to delineate fracture anatomy (55% of the time) and the observers felt CT would add to their treatment decision only 26% of the time. During round 2, 75% of time surgeons felt that CT scan was useful. CT was thought to be most useful in guiding screw placement (56% of the time) and not as useful (28% of time) for treatment decision making.

Addition of CT in complex distal tibial physeal fractures did not increase interobserver reliability to classify the fracture or the treatment decision. Surgeons reported that the CT was most useful to plan screw placement and changed their treatment decision in about a fifth of the cases.