To the Editor:— The occurrence of severe brachial plexopathy after an ultrasound-guided single-injection nerve block for total shoulder arthroplasty in a patient with multiple sclerosis (MS) presented by Koff et al. 1raised several issues regarding the cause of this complication. Intraneural injection, the most feared complication when performing regional block, can in this case be definitely excluded. The possibility of having transfixed the upper or median cord during the procedure seems, although possible, unlikely. Moreover, it has been shown that even injection of local anesthetics beyond the epineurium does not invariably result in nerve damage.2The existence of a preexisting subclinical polyneuropathy has been shown to increase the toxic potential of local anesthetics in certain circumstances.3In the current case, MS has been highlighted as a risk factor. MS is a chronic disease characterized by multiple areas of central nervous system white matter inflammation, demyelination, and glial scaring or sclerosis.4Despite reports of peripheral nerve alterations, peripheral nervous system involvement remains rare and, if present, subclinical in most cases, due to subtle nerve lesions without any frank demyelination. This is supported by the work by Boerio et al. 5: In MS patients with no nerve conduction abnormalities, assessment of the absolute and relative refractory periods showed significant increase in refractoriness compared with a control group. However, these minor changes could not be considered as significant alteration of the nerve myelin sheath. A recent study described the occurrence of a new inflammatory demyelinating disease unlike MS or chronic inflammatory demyelinating polyradiculopathy occurring in MS patients with a relapsing–remitting course in which the central nervous system involvement preceded peripheral nerve system involvement.6The current case does not fulfill the criteria for this diagnosis. The authors have suspected an acute “inflammatory” neuritis, but unfortunately this was not further investigated by either sural nerve biopsy or cerebrospinal fluid analysis for elevation of protein content reflecting nerve root inflammation.7The presence of a preexisting polyneuropathy could have been disclosed if conduction studies had been performed on postoperative day 3. The recordings would have shown signs of demyelination because pathologic features found on peripheral nerves in patients with MS are either segmental demyelination or reduction in myelin thickness.8This was not the case in this patient, and unfortunately electroneuromyography studies of the contralateral arm have not been performed. The latter recording would have given an objective state of the peripheral nerve system. These elements make the likelihood of a previous polyneuropathy very unlikely. This assumption is also supported by normal electromyography performed on the patient’s unaffected limbs 3 months later. How, then, can this event be explained? First, the occurrence of burning pain—neuropathic character—despite a dense motor block 5–6 h after a successful block performed with 30 ml ropivacaine, 0.5%, is unusual because the duration of the sensory block is approximately 12–15 h. This suggests an “acute trauma” of the brachial plexus. Second, the long duration of surgery (3 h 45 min) let us think that the procedure was complicated, meaning that the placement of the prosthesis had probably required a large amount of traction—physically induced stress—on the brachial plexus. Studies have shown that abduction challenges the brachial plexus.9Arm extension, wrist extension, and head rotation to the contralateral side add further stress on the nerves.9,10Ikeda et al. 11have demonstrated in experimental studies that an elongated nerve is much more vulnerable to compression injury (surgical retractors). This constellation favors an acute “physically induced trauma” of the brachial plexus to explain the development of this complication. This is supported by the electromyography recordings on day 11, consistent with axonal loss. On the other hand, the toxic effect of local anesthetic placed outside the epineurium, as shown by ultrasound in the current case, would have more likely shown signs of demyelination. Last, testing the anterior part of the shoulder with cold ice gives information regarding blockade of the medial branch of the supraclavicular nerve, not the axillary nerve. Positioning and surgically induced stress are certainly greatly underestimated by anesthesiologists as causes of brachial plexus damage after shoulder surgery.
*Balgrist University Hospital, Zurich, Switzerland. aborgeat@balgrist.ch
