Background: The finding of pseudomeningocele from cervical myelogram is widely accepted as a pathognomonic sign for diagnosing root avulsion in brachial plexus injury. In general, motor power in this setting should be absent. However, in clinical practice, we observed that motor power in some patients was still preserved. The objective of this study is to evaluate the accuracy of pseudomeningocele from cervical myelogram for predicting root avulsion in brachial plexus injury. Methods: We retrospectively reviewed 201 patients with brachial plexus injury from 2007–2011. Four patients were excluded due to open wound injury. Motor power of the key muscle of each nerve root was evaluated by skilled hand surgeons. All cervical myelogram was interpreted by radiologists. Sensitivities, specificities, positive predictive values and negative predictive values were calculated with 95% confidence interval. Results: Thirty and 29% of pseudomeningocele occurred at C7 and C8 level, respectively. The sensitivity of pseudomeningocele of each root from C5 to T1 was low (range, 10–62%). The specificity was high only at C5 (91%) and T1 (96.2%). Over 20% of patients with pseudomeningocele at C6–8 levels had some motor function. The initial muscle power of these patients was M1 or M2 and 70% of them recovered to at least M3 at the final follow-up. Conclusions: Pseudomeningocele is not an absolute sign for diagnosing of root avulsion in brachial plexus injury due to its high false positive rate when compared with preoperative motor function of each root. Careful examination of the key muscle of each root is extremely crucial to prevent unnecessary operation on that cervical nerve root.
Correlation between motor function recovery and daily living activity outcomes after brachial plexus surgery