Introduction: Tibia and femur fractures are common in orthopedics. Their standarad treatment is intramedullary nailing. This study investigated the best pre-surgery method for estimating the intramedullary nail length.Methods: First, we assessed intra-and inter-observer reliabilities by examining healthy volunteers. Then, we measured the presurgery intramedullary nail length of 82 patients with tibia and 25 patients with femur shaft fractures. This size was compared with the suitable nail length used at operation.Results: Almost all of the examined methods had "good" inter-and intra-observer reliability. In patients with femur fractures, the greater patella method had the best intraclass correlation coefficient (0.876) and the lowest standard error of measurement (0.777).In tibia fractures, medial malleolus-tubercle method had the best intraclass correlation coefficient (0.860) and the lowest standard error of measurement (0.602). Bland-Altman plots were created for the findings of the second part of the study. For femur fractures, 64% of the pre-surgery measurements with greater patella method were more than the actual nail used at operation. These percentages were 93% and 100% in olecranon fifth finger and greater-epicondyle methods, respectively. For tibia fractures, 40.9% of medial malleolus-tubercle, 37.3% of olecranon-metacarp and 65.06% of radiogram methods' measurements were more than the actual nail at operation.
Conclusion:The medial malleolus-tuberosity method seems to be more accurate for tibia fractures. The greater patella method seems to be more accurate for femur fractures. However, we found no completely accurate method in the examined methods.