Object: The purpose of this study was to investigate the usefulness of preoperative evaluation based on clinical testing and computed tomography (CT) myelography in differentiating root rupture (that is, graftable root) from root avulsion in total brachial plexus palsy.
Methods: Thirty-two patients with total brachial plexus palsy were clinically tested for the presence of phrenic nerve palsy, supraclavicular Tinel sign, shoulder protraction, Bernard-Horner syndrome, and hand pain. The patients underwent CT myelography and then underwent surgery. The combination of a positive Tinel sign and a positive shoulder protraction test accurately predicted the presence of a graftable root in 93.7% of the cases. A 96.8% rate of accuracy was attained if the results of the CT myelography were considered together with the clinical signs. The presence of Bernard-Horner syndrome and hand pain accurately indicated avulsion of the lower roots in 93.7% of the patients. Computed tomography myelography accurately predicted the condition of the lower roots in 100% of the cases. Total avulsion injury was observed in five cases (16%). The lower roots were avulsed in 94% of the cases. The C-5 and C-6 roots were grafted 40 times, and a suitable root stump for grafting lay in a retroscalenic position in 18 (45%) of the 40 cases.
Conclusions: Preoperative assessment based on clinical examination and CT myelography allowed correct surgical planning in more than 90% of the cases.