Shoulder Restraints as a Potential Cause for Stretch Neuropathies: Biomechanical Support for the Impact of Shoulder Girdle Depression and Arm Abduction on Nerve Strain

Stretch-induced neuropathies of the brachial plexus and median nerve are the second most prevalent perioperative neuropathies. In 2002, we demonstrated that movements that elongate the nerve bedding, such as shoulder girdle depression and wrist extension, significantly reduce the available range of elbow extension in healthy subjects in a brachial plexus tension test position.1Although an increase in strain in the brachial plexus and median nerve was the most plausible explanation, we could only speculate that this was the underlying mechanism for the reduced range of motion.

Because shoulder girdle depression and abduction of the arm greater than 90° have been associated with stretch-induced perioperative neuropathies, we measured strain in the median nerve in three embalmed undisturbed male cadavers in four different arm positions: arm by the side, without shoulder girdle depression (1, reference position) and with shoulder girdle depression (2), and in 90° arm abduction without depression (3) and with depression of the shoulder girdle (4). Because insertion of strain gauges in the brachial plexus requires excision of several structures that may alter nerve biomechanics, we decided to insert miniature linear displacement transducers (Differential variable reluctance transducers; Microstrain, Burlington, VT) into the median nerve at the level of the humerus and also just proximal to the carpal tunnel where the nerve runs relatively superficially. Mean values representing the change in strain relative to the strain in the reference position are reported. Because there is at least mild tension in a peripheral nerve in most positions, the strain gauges were inserted with the nerve under some tension. Therefore, it was impossible to calculate absolute strain values. Electrogoniometers (Biometrics, Blackwood, Gwent, United Kingdom) were used to guarantee that the position of the elbow was maintained at 170° extension throughout the experiment and the wrist in a neutral position (180°). The mean available depression, measured at the acromion, equaled 17 mm when the arm was by the side and 10 mm with the arm abducted. Excursion of the median nerve in relation to its surrounding structures was measured at both sites during arm abduction. A digital caliper was used. The Anatomy Ethics Committee of the School of Biomedical Sciences of The University of Queensland, Brisbane, Australia, approved the study.

Figure 1demonstrates the increase in strain in the different positions. As anticipated, strain in the proximal part of the median nerve increased with shoulder girdle depression, both with the arm by the side (+1.2%) and with the arm abducted (+1.0%). In addition, an increase in strain in the median nerve at the level of the wrist could be observed in one cadaver (+0.6%). In all cadavers, arm abduction resulted in a large increase in nerve strain at the level of the humerus (+3.1%) and a relatively large increase at the level of the wrist (+1.4%). This substantial increase was to our surprise as we allowed the physiologically coupled movement of shoulder girdle elevation to occur during arm abduction. It also demonstrated that strain is transmitted well beyond the site at which the nerve bedding is elongated. Abduction of the arm resulted in a proximal excursion of the median nerve of 11.8 mm at the humerus. The nerve movement at the wrist was too small to quantify.

It is well documented that strain may impair several processes, such as microcirculation, axonal transport, and nerve conduction. Animal models have demonstrated that 5–10% elongation results in impaired blood flow.2At 11% strain, axonal transport is inhibited,3and the amplitude of the action potential is reduced by 70% after a 1-h sustained elongation of 6%.4There is only indirect evidence, but it is to be expected that even lower strain values can affect these processes when blood pressures are reduced. Hypotension occurs frequently during anesthesia and surgery.

Although there is evidence that the magnitude of nerve strain obtained in cadaveric experiments is meaningful,5extrapolation of these values should be made with caution. Our findings may be an underestimation of the true strain increase because the available range of shoulder girdle depression was limited as a result of embalmment. In addition, the increase in tension in the brachial plexus with shoulder girdle depression is probably larger than the increase we recorded at the level of the humerus. There is ample evidence that the largest increase in strain occurs nearest the site of nerve bed elongation.6Finally, it is important to realize that the reported strain measures are increases in strain, not absolute values. Because the strain in the reference position to which values were normalized was not zero, absolute strain values are higher than the reported increases. Therefore, the cumulative strain recorded in this experiment may be of a magnitude that impairs microcirculation, axonal transport, and nerve conduction.

The findings of this study provide experimental support for the experientially based recommendations regarding positioning in anesthesia and surgery. It supports the guidelines to minimize the use of shoulder restraints and to monitor the position of the shoulder girdle rigorously if restraints must be used. However, because we observed a substantial increase in strain with 90° abduction, the recommendation to not exceed 90° abduction may even have to be adjusted to a more conservative guideline to further limit the occurrence of stretched-induced perioperative neuropathies. However, the substantial increase in strain associated with movements that are well within physiologic ranges and the fact that strain is transmitted well beyond the site of nerve bed elongation strengthen our previous statement that even when the positioning of all upper limb joints is carefully considered, complete prevention of perioperative neuropathy seems almost inconceivable.1 

School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia. m.coppieters@uq.edu.au