Correlation of Short Knee Radiographs and Full-length Radiographs in Patients Undergoing Total Knee Arthroplasty.

INTRODUCTION: The clinical success and longevity of a primary total knee arthroplasty (TKA) in large part depend on our ability to control coronal alignment. However, controversy exists regarding which radiographs to use for the most accurate interpretation. The study assesses the accuracy of coronal alignment measurements using a single short knee radiograph (SKR) in comparison with full-length radiographs (FLRs). METHODS: Using our institutional database, we retrieved radiographs of all patients who have had pre- and postoperative FLRs for their primary TKA in 2014. The following measurements were obtained on both short and long radiographs: femoral-tibial angle (FTA), anatomic lateral distal femoral angle, medial proximal tibial angle, condylar-plateau angle, and condylar-plateau distance. A reliability analysis was conducted between the pre- and postoperative SKRs and FLRs using the intraclass correlation coefficient (ICC). RESULTS: Radiographs of 236 limbs were included in the analysis. The FTA showed an ICC of 0.84 and 0.69 on the pre- and postoperative radiographs, respectively. Good ICC was seen in the lateral distal femoral angle in both the pre- and postoperative radiographs; these were 0.70 and 0.67, respectively. Also, the medial proximal tibial angle showed good to excellent correlation, with an ICC of 0.83 on the preoperative and 0.66 on the postoperative radiographs. CONCLUSION: This study illustrates that SKRs could be an appropriate substitute for FLRs for the evaluation of primary TKA coronal alignment, especially in the postoperative assessment of these patients. LEVEL OF EVIDENCE: Level III.