OBJECTIVE: Fracture dating significantly shapes decisions in child protection. With a dearth of primary evidence underpinning fracture dating in children, we examined the key radiologic features of fracture healing and their timelines. MATERIALS AND METHODS: Digital radiographs of children younger than 72 months old with accidental long bone fractures of known timing were reviewed independently by three pediatric radiologists blinded to the age of the fractures. Six radiologic features of fracture healing were evaluated: soft-tissue swelling, periosteal reaction, soft callus, hard callus, bridging, and remodeling. Interobserver agreement was assessed using kappa analysis. RESULTS: Two hundred twenty-eight films of 82 fractures in 63 children (mean age, 4.8 years) were assessed. Soft-tissue swelling was identified by two or more radiologists in 59% of the radiographs at days 1-2 after fractures, and prevalence sharply declined thereafter. Periosteal reaction was first seen at day 5 and was present in 62% of the films obtained between 15 and 35 days after the fracture. Soft callus was first seen at day 12 and was prevalent in 41% between 22 and 35 days. Hard callus and bridging began at day 19, increasing to 60% prevalence from 36 days onward. Remodeling was observed only in fractures 45 days old or more. Kappa scores were between 0.55 and 0.80 overall, with greater agreement when there was no plaster cast. CONCLUSION: The results of this study show that fractures in young children may be dated as acute (< 1 week), recent (8-35 days), or old (≥ 36 days) on the basis of the presence of six key radiologic features in combination. Furthermore, good interobserver agreement suggests these results are reproducible.
OBJECTIVE:Fracture dating significantly shapes decisions in child protection. With a dearth of primary evidence underpinning fracture dating in children, we examined the key radiologic features of fracture healing and their timelines. MATERIALS AND METHODS: Digital radiographs of children younger than 72 months old with accidental long bone fractures of known timing were reviewed independently by three pediatric radiologists blinded to the age of the fractures. Six radiologic features of fracture healing were evaluated: soft-tissue swelling, periosteal reaction, soft callus, hard callus, bridging, and remodeling. Interobserver agreement was assessed using kappa analysis. RESULTS: Two hundred twenty-eight films of 82 fractures in 63 children (mean age, 4.8 years) were assessed. Soft-tissue swelling was identified by two or more radiologists in 59% of the radiographs at days 1-2 after fractures, and prevalence sharply declined thereafter. Periosteal reaction was first seen at day 5 and was present in 62% of the films obtained between 15 and 35 days after the fracture. Soft callus was first seen at day 12 and was prevalent in 41% between 22 and 35 days. Hard callus and bridging began at day 19, increasing to 60% prevalence from 36 days onward. Remodeling was observed only in fractures 45 days old or more. Kappa scores were between 0.55 and 0.80 overall, with greater agreement when there was no plaster cast. CONCLUSION: The results of this study show that fractures in young children may be dated as acute (< 1 week), recent (8-35 days), or old (≥ 36 days) on the basis of the presence of six key radiologic features in combination. Furthermore, good interobserver agreement suggests these results are reproducible.
Authors: Shiow-Wen Yang; Mary A Kernic; Beth A Mueller; Gregory E Simon; Kwun Chuen Gary Chan; Ann Vander Stoep Journal: JAMA Pediatr Date: 2020-08-03 Impact factor: 16.193