Skull fractures can be difficult to confidently and accurately diagnose in infants due to the thin nature and variability of the skull and the sutures. We sought to determine whether 3D reconstructions improve the diagnostic yield and speed of diagnosis of skull pathology in infants receiving head CT for trauma
We did a retrospective review of head CT’s obtained in children less than 2 years of age over 12 month period. There were 292 studies, 180 of these had 3D reconstructions, and 24 of these had skull fractures. 10 CT studies (5 positive, 5 negative) where 3D reconstructions were performed were randomly selected and reviewed by radiologists of various levels of training (junior resident, senior resident, general pediatric fellow, pediatric radiologist with less than 2 years’ experience, pediatric radiologist with greater than 2 years’ experience, pediatric radiologist with > 10 years’ experience, and neuroradiologist with CAQ) who were blinded to the diagnosis. Radiologists reviewed each case on two separate occasions,3 weeks apart on one occasion they were shown the cases …show more content…
They were correctly identified 100% of the time with 3D reconstructions. Fractures were correctly identified 85% of the time without 3D reconstructions and 95% of the time with 3D reconstructions. Errors in interpretation were most commonly seen among junior radiology residents for both the 2D and 3D interpretation. Interpretations with 3D took 99±36 seconds, compared to 169±51 seconds without 3D imaging (p<0.0001). Interpretations with 3D were made with a greater degree of confidence. The improved confidence between 2D and 3D readers was inversely correlated to their level of training.
3D CT images increase speed, accuracy, and confidence of diagnosis of pediatric skull fractures. The degree of improvement in speed and accuracy was greater among those of lower levels of
We did a retrospective review of head CT’s obtained in children less than 2 years of age over 12 month period. There were 292 studies, 180 of these had 3D reconstructions, and 24 of these had skull fractures. 10 CT studies (5 positive, 5 negative) where 3D reconstructions were performed were randomly selected and reviewed by radiologists of various levels of training (junior resident, senior resident, general pediatric fellow, pediatric radiologist with less than 2 years’ experience, pediatric radiologist with greater than 2 years’ experience, pediatric radiologist with > 10 years’ experience, and neuroradiologist with CAQ) who were blinded to the diagnosis. Radiologists reviewed each case on two separate occasions,3 weeks apart on one occasion they were shown the cases …show more content…
They were correctly identified 100% of the time with 3D reconstructions. Fractures were correctly identified 85% of the time without 3D reconstructions and 95% of the time with 3D reconstructions. Errors in interpretation were most commonly seen among junior radiology residents for both the 2D and 3D interpretation. Interpretations with 3D took 99±36 seconds, compared to 169±51 seconds without 3D imaging (p<0.0001). Interpretations with 3D were made with a greater degree of confidence. The improved confidence between 2D and 3D readers was inversely correlated to their level of training.
3D CT images increase speed, accuracy, and confidence of diagnosis of pediatric skull fractures. The degree of improvement in speed and accuracy was greater among those of lower levels of