A study of radiological features of healing in long bone fractures among infants less than a year.

PURPOSE: To create a timetable for dating long bone fractures in infants aged less than 1 year using previously defined radiographic signs of fracture healing.

MATERIALS AND METHODS: A retrospective cross-sectional time series of long bone fractures in infants aged less than 1 year was conducted from 2006 to 2013. After exclusion criteria were applied 59 digital image series were available for review from 40 infants. Utilizing published criteria for dating fractures, the presence or absence of four pre-defined features of healing was scored: periosteal reaction, callus, bridging, and remodeling. Three radiologists independently scored radiographs with a 3-point scale, marking each feature as present, absent, or equivocal. The times in days when features were first seen, peaked (feature agreed present in >40% of images), and last seen were noted. Statistical analysis using free marginal kappa was conducted.

RESULTS: The level of agreement among the three radiologists was high (0.64-0.85). The sequence in which the features were seen was: periosteal reaction range 7-130 (present in the majority of cases between 9 and 49 days); callus range 9-130 (present in the majority of cases between days 9-26); bridging range 15-130 (seen in the majority of cases between 15 and 67 days); remodeling range 51-247 days.

CONCLUSION: This study provides a timetable of radiological features of long bone healing among young infants for the first time. Dating of incomplete long bone fractures is challenging, beyond the presence of periosteal reaction, but a consistent sequence of changes is present in complete fractures.