Abstract

Comparison of conventional radiographs (CR) of distal tibial growth plate fractures [Salter-Harris (SH) fracture types I-V/triplane fractures I-III] with computed tomography (CT) as the reference standard and assessment of diagnostic benefit of CT imaging in the affected patients. We retrospectively evaluated all growth plate injuries of the distal tibia with complete pre-therapeutic imaging material (CR and CT including MPR) obtained between August 2001 and December 2006. The imaging material was randomised and presented to two radiologists. Fracture of metaphysis, epiphysis and epiphyseal line were noted separately for distal tibia. In case of fracture, involvement of the articular surface, articular dehiscence and ridge formation, subluxation and number of tibial fragments were evaluated. All fractures were classified as SH type I-V or triplane fractures type I-III. Sensitivity, specificity, positive and negative predictive value and overall accuracy of CR were compared to CT. Thirty-three patients (mean age 14 ± 2 years) were evaluated. CR showed significantly less tibial fragments as compared to CT (1.39 ± 0.75 vs. 1.61 ± 1.25; p = 0.023). The overall accuracy of CR was <90% for fracture involving the metaphysis (82%), dehiscence of the articular surface (64%), ridge formation of the articular surface (61%) and subluxation (79%). The CR evaluation showed differing SH classification in CT in 10/33 cases (30%) with the highest misclassification rates in type-III SH fracture. For evaluation of triplane fractures, CR classification was incorrect in five cases (71%) out of seven. No misclassification occurred in types I and II SH fractures. The CR of distal tibial growth plate fractures showed a low overall accuracy for articular surface dehiscence, articular ridge formation and subluxation as compared to CT. CT revealed significantly more fragments. It is difficult to correctly classify type III/IV SH fractures and triplane fractures with CR thus emphasising the necessity of using CT in selected patients.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call