Validity of radiographic evaluations of bone formation in a rat calvaria osteotomy defect model

Objective: The objective of this study was to evaluate the validity of radiographic evaluations of bone formation in a critical-size rat calvaria osteotomy defect model. Methods: Bilateral, critical-size (∅ 6 mm) calvaria osteotomy defects in 30 adult Sprague-Dawley rats treated with a rat platelet-rich plasma preparation or control treatments were evaluated by radiographic and histometric measures following a 4- or 8-week healing interval. Standardized radiographic images of the rat calvaria gross specimens were used to assess bone formation within the defect sites by visual evaluation of the grey scale by three masked examiners. The most central portion of each defect site was subject to histometric analysis using a PC-based image analysis system. Kappa statistics and percentage agreement between the radiographic and histometric analysis were estimated. Results: Radiographic evaluations of bone formation are associated with significant weaknesses poorly representing actual healing events; κ statistics (0.17) denoting slight agreement beyond chance. Perfect agreement between the histologic and radiographic analysis for defect sites showing complete and partial histologic bone fill was achieved 63% and 50% of the time, respectively. Agreement reached only 20% for sites with no/limited bone fill. When no/limited and partial bone fill occurred, the radiographic analysis tended to overestimate bone fill and underestimate bone fill when complete closure of the defect sites was observed in the histologic analysis. Conclusion: Low accuracy was observed when radiographic evaluations were employed in identifying and characterizing bone fill in the rat calvaria osteotomy defects. Assessment of bone healing in animal models aiming at treatment recommendations for clinical application must not solely be based on radiographic analysis, but should be confirmed using histologic observations.