Findings Associated With Nonaccidental Trauma in Children With Isolated Femoral Diaphyseal Fractures

Pediatric patients with isolated femoral diaphyseal fractures are difficult to assess for nonaccidental trauma (NAT). The purpose of this study was to determine (1) if there are any demographic features of isolated femoral diaphyseal fractures associated with suspected NAT and (2) if there are clinical signs associated with isolated femoral diaphyseal fractures associated with suspected NAT.

All patients with femoral diaphyseal fractures from January 2010 to June 2018 were reviewed. We included patients younger than 4 years old with isolated femoral diaphyseal fractures. We excluded patients 4 years old and older, polytraumas, motor vehicle collisions, and patients with altered bone biology. Diagnosis of suspected NAT was determined by review of a documented social work assessment. We recorded fracture characteristics including location along femur as well as fracture pattern and presence of associated findings on NAT workup including the presence of retinal hemorrhage, subdural hematoma, evidence of prior fracture, or cutaneous lesions. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of these associated findings were calculated.

Totally, 144 patients met the inclusion criteria. Social work was consulted on 50 patients (35%). Suspected NAT was diagnosed in 27 patients (19%). The average age of patients with suspected NAT was 0.82 and 2.25 years in patients without NAT ( P <0.01). The rate and type of skin lesions present on exam were not different between the 2 groups. Patients with suspected NAT had no findings of retinal hemorrhage or subdural hematoma, but 5 of 27 patients (19%) had evidence of prior fracture on skeletal survey. The sensitivities of retinal hemorrhage, subdural, and skeletal survey were 0%, 0%, and 19% and the specificities of all were 100%. The NPVs were 39%, 27%, and 63%, respectively. The PPV of skeletal survey was 100%. Since there were no patients in this study with positive findings of retinal hemorrhage or subdural hematoma, the PPV for these could not be assessed.

In the current study, signs of NAT such as skin lesions, retinal hemorrhage, subdural hematoma, and evidence of prior fracture on skeletal survey may not be helpful to diagnosis suspected NAT in patients with an isolated femoral diaphyseal fracture.

Level III-diagnostic study.