Rhee et al. should be congratulated on their effort to describe the prevalence of spinal cord injury in their cohort of patients who were referred to their tertiary care center with brachial plexus injuries. The 12.2% prevalence rate in their study is significantly greater than the 2% to 5% prevalence rate previously reported. Their description of associated findings, such as supraclavicular vascular injury, cervical spine fracture, Horner syndrome, or phrenic nerve dysfunction, will help to define which patients have the greatest risk for having a combination of both injuries. They describe in detail how this may affect the outcomes of treatment in patients with combined injuries.
There remains a very important question on this topic, which this paper does not answer. This study cohort contains a very heterogeneous group of spinal cord injuries. The authors take great pains to rigorously classify the nature of the spinal cord injury as definitive, probable, or possible. Yet out of thirty-one patients with combined injuries, only eight had identifiable spinal cord injury patterns. As described, the remaining twenty-three subjects had spinal cord injuries of unclear clinical significance. The authors defined “major” clinical evidence of spinal cord injury as a neurologic deficit such as motor, sensory, or sphincter dysfunction on the basis of the history and/or the physical examination. No distinction was made between patients with transient neurological symptoms at the time of injury versus permanent or prolonged neurological deficit. Of twelve subjects with “major” clinical evidence, five had only neurological symptoms in the extremity affected by the brachial plexus injury, but with imaging findings consistent with spinal cord injury. The authors do not explain how they differentiated the neurological deficit that was related to the spinal cord injury from the neurological deficit that was related to the brachial plexus injury in this subgroup. Another eleven subjects had “minor” clinical findings only, such as upper motor neuron signs on clinical examination, and imaging consistent with spinal cord injury. Out of thirty-one patients with combined injuries, only twenty-two had advanced imaging available; that imaging showed increased T2 signal intensity in the spinal cord in nineteen patients, and epidural or subdural hematoma, or both, in three.
The authors discuss the significance of Brown-Séquard syndrome on the reconstruction outcomes of patients with brachial plexus injuries. What is not clear is whether the much more subtle spinal cord injuries described in this paper have any long-term impact on the outcome of treatment for the brachial plexus injury. The authors suggest that the presence of a transient neurological deficit at the time of injury or the presence of isolated upper motor neuron findings with increased signal intensity on magnetic resonance imaging (both scenarios that the authors would have defined as definite or probable spinal cord injuries) may result in worse outcomes at the time of reconstruction some months after injury. Without outcomes data, this supposition is unconvincing. Hopefully, the authors will continue to publish their outcomes data from their impressive clinical experience, which may allow them to answer this critical question.