Santiago Medina MD, Presenter: Nothing to Disclose

Ilker Yaylali PhD, Abstract Co-Author: Nothing to Disclose

David Zurakowski PhD, Abstract Co-Author: Nothing to Disclose

Jennifer Ruiz BA, Abstract Co-Author: Nothing to Disclose

John Grossman MD, Abstract Co-Author: Nothing to Disclose

To determine imaging and non-imaging predictors of percent neuronal loss in perinatal brachial plexus injury.

Thirty-one children with perinatal brachial plexus injury who underwent surgical intervention were studied. Standard of reference was histopathologic percent neuronal loss. Standard of reference was correlated to MRI findings, pre and post operative evaluation, electromyography (EMG) and somatosensory evoke potentials (SSEP). Diagnostic performance characteristics including sensitivity, specificity, and ROC curves were determined and logistic regression was used to identify significant predictors of neuronal loss.

There were 70 biopsies evaluated (mean age 220 days, range 143-362). Median neuronal loss based on histopathology was 30% (range 0-100%). There was a moderate positive correlation between EMG (mAmp) and % neuronal loss (Pearson r = 0.42, P30% neuronal loss. ROC analysis using EMG indicated an area under the curve (AUC) of 0.69 for >30% neuronal and AUC = 0.74 for >50% neuronal loss. For >30% neuronal loss, MRI pseudomeningoceles had a sensitivity of 32% (95% CI: 21-43%) and specificity of 91% (95% CI: 84-98%).

Intraoperative scoring and EMG are independent predictors of neuronal loss in patients with perinatal brachial plexus injury. MRI pseudomeningoceles have a low sensitivity and a high specificity

Medina, S, Yaylali, I, Zurakowski, D, Ruiz, J, Grossman, J, Diagnostic Predictors of Perinatal Brachial Plexus Injury: A Prospective Study.  Radiological Society of North America 2005 Scientific Assembly and Annual Meeting, November 27 - December 2, 2005 ,Chicago IL. http://archive.rsna.org/2005/4406252.html