Demonstration of autonomic dysfunction in traumatic brachial plexus injury using quantitative sudomotor axon reflex test: Preliminary results
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.217967
Source of Support: None, Conflict of Interest: None
Objective: To objectively document autonomic dysfunction in the affected arm with traumatic brachial plexus injury (TBPI) using quantitative sudomotor axon reflex test (QSART).
Keywords: Autonomic, axon reflex, brachial plexus, quantitative sudomotor axon reflex test, sweat
During the evaluation and treatment of numerous patients with traumatic brachial plexus injury (TBPI), we found that patients often complain of symptoms that are not directly attributable to the observed neurological deficits. These symptoms include non-dermatomal dysesthesia and altered sweating in the affected arm. Even during surgery, under general anesthesia, we found wide variations in heart rates at the time of dissection of the involved arm. These symptoms and changes in heart rates probably suggest involvement of the autonomic nervous system in patients with TBPI. The overt autonomic disturbances in the injured arm in the form of reflex sympathetic dystrophy have been well described.,,
To the best of our knowledge, till date there has been no objective documentation of autonomic dysfunction in patients with TBPI. In the present study, we have attempted to objectively document autonomic dysfunction in the affected arm with TBPI using quantitative sudomotor axon reflex test (QSART).
A prior institution ethics committee clearance was obtained to conduct the study. Patients presenting to the neurosurgical outpatient department from August 2013 to November 2014 with TBPI were included in the study. The QSART was administered to each patient with prior informed consent detailing the procedure. QSART was performed in our autonomic laboratory under the Department of Neurophysiology. It was done using the Q-sweat® machine (WR Medical Electronics, 123 North Second Street, Stillwater, MN - 55082).
A total of 20 patients with TBPI were included. The injuries were grouped as preganglionic or postganglionic injuries based on either of the following:
Clinical findings suggestive of preganglionic injury
Sensory nerve action potentials
Sensory nerve action potentials (SNAP) can sometimes differentiate a preganglionic from a postganglionic injury. Injury proximal to the dorsal root ganglion (preganglionic) produces a complete distal sensory loss but preserves distal sensory conduction. This is because the dorsal root ganglion, which is also avulsed, is still in contact with the peripheral nerve fibres. However, if the injury is postganglionic or both pre- and postganglionic, no SNAP will be obtained.
Magnetic resonance imaging findings suggestive of a preganglionic injury
A traumatic meningocele is a valuable sign of a preganglionic lesion although it is not a pathognomic sign. Absence of roots is also an important sign in detecting a preganglionic lesion. Abnormal enhancement of paraspinal muscles is an accurate indirect sign of root avulsion injury. Denervated muscles show enhancement as early as 24 hours after a nerve is injured. Abnormal enhancement of the multifidus muscle is the most accurate finding in the paraspinal muscles because the multifidus muscle is innervated by a single nerve root.
Participants were advised to abstain from coffee, tea, nicotine, and other such stimulants for 24 hours prior to the test. They were advised to not take any oral feed for at least 3 hours prior to the test. The test was carried out in a room where the temperature was maintained at 23°C. Participants were made to lie down on a comfortable bed and given adequate time to relax prior to the commencement of the test. The areas for placing the recording capsules were cleaned with 70% alcohol. The Q-Sweat instrument was then switched on to the pre-warm mode before starting the test.
The recording sites
On the right and left forearm, the sites selected were 75% of the distance from the ulnar epicondyle to the pisiform bone approximately over the area innervated by the ulnar nerve.
Recording the evoked sweat response
Acetylcholine was used as the reagent. Approximately 1 ml of 10% acetylcholine (1 g of acetylcholine was mixed with 10 ml of distilled water) was injected into each sweat compartment of the device.
The settings of the stimulator were as follows:
Patient information was entered into the computer and verified. Baseline sweat response was recorded until it stabilized (for 5 minutes). Then, the stimulation was started and the area on the computer graph was marked with a header “stimulation started.” After 5 minutes of electrical stimulation, it was stopped and the site on the graph was marked “stimulation stopped.” Recording was done for another 5 minutes and then stopped. At the end of the procedure, the software was used to analyze and generate a report from the data. The total volume of sweat, latency time, baseline sweat output, and ending offset sweat volume were recorded. The result for each patient was available as a graph, which was generated by the software along with the actual values.
The evoked response is measured, wherein a small amount of current applied to the sweat cell (a part of the Q-sweat device) causes iontophoresis of acetylcholine. Iontophoresis of acetylcholine helps it to enter the skin and stimulate the postganglionic sudomotor fibres. If the axons of the postganglionic sudomotor fibres are intact, a sweat response is elicited which is then recorded by the device. The iontophored acetylcholine stimulates the nicotinic receptors on the axons and elicits an axonal reflex. The sudomotor pathway evokes action potentials, which are conducted orthodromically as well as antidromically to axonal branching points. These are then conducted orthodromically to activate peripheral sweat glands in an axon reflex. Acetylcholine which is released at the neuroglandular junction binds to the muscarinic (M3) receptors on the eccrine sweat glands and leads to sweat secretions. This in turn is measured by the Q-Sweat machine sweat capsule.
The cholinergic local sweat response consists of two components; the first is the direct activation of sweat glands via muscarinic receptors, and the second is the axon reflex sweating in the vicinity of the stimulation site. Sweating is initiated by stimulation of nicotinic receptors on sudomotor axons and impulse distribution by peripheral nerve arborisations [Figure 1].
Demographics (n = 20) A total of 20 patients (19 males) with their age ranging from 15 to 50 years underwent the QSART procedure. The mean age was 30.65 years with a standard deviation of 8.86. Nineteen out of 20 participants were males. A total of 13 (65%) injuries were right sided and 7 (35%) were left sided. All patients had traumatic pan brachial plexus injuries, which were preganglionic; 13 (65%) were incomplete (preserved distal function).
Using paired t-tests, the total volume of sweat, latency period, and baseline sweat output in the affected and normal sides in patients (study group) was calculated. The mean total sweat volume was lower on the affected side [Table 1]. The mean latency period after starting stimulation was prolonged on the affected side [Table 1]. However, none of these values were significantly different between the affected and normal arm. The baseline sweat output was significantly lower on the affected side (P = 0.01) [Table 1] and [Figure 2].
Relation between sweat output and type of injury
The mean total volume, after starting stimulation, was lower in complete injuries (no function) [Table 2].
The latency period was longer for complete injuries and the baseline sweat volume was lower in complete injuries [Table 2]. However, none of the values were statistically significant [Table 2].
Pain and sweat output
The mean total sweat output and baseline sweat output on the affected arm in patients with pain was lower as compared to patients without pain [Table 3].
The mean latency period in the affected arm in patients with pain was longer than those without pain [Table 3]. However, none of the values were statistically significant in patients with and without pain [Table 3]. There was no statistically significant relationship between the presence and absence of pain and sweat output [Table 3].
Correlation between pain and sweat output
There was no statistically significant correlation between sweat output and pain [Table 4].
Correlation between duration of injury and sweat output
There was no statistically significant correlation between sweat output and duration of injury [Table 5].
Autonomic dysfunction following injuries to peripheral nerves and/or injury to the whole limb has been studied since the days of Sir Silas Weir Mitchell (1829–1914)., Many attempts have been made in the past to define the various symptoms and signs associated with such injuries, and presently the scientific community has agreed upon the use of the term 'complex regional pain syndrome' (CRPS) to describe such complex symptoms. In CRPS, abnormalities related to the sympathetic nervous system include changes in sweating and skin blood flow.,,,,,, Sympathetic denervation and mechanisms of denervation hypersensitivity cannot account for vasomotor and sudomotor abnormalities in CRPS-I because there is no visible nerve lesion. However, there is direct evidence of reorganization of central autonomic control in these syndromes. Resting sweat output, as well as thermoregulatory and axon reflex sweating are increased in patients with CRPS-I., Increased sweat production cannot be caused by a peripheral mechanism because, unlike blood vessels, sweat glands do not develop denervation supersensitivity.
In 1985. Horowitz et al. documented autonomic dysfunction in 4 patients with iatrogenic brachial plexus injury. This appears to be the only such documentation. To explain the unilateral sweating dysfunctions in patients with CRPS-I, Janig et al., proposed the involvement of spinal autonomic circuits. In a review article, Janig and Baron have studied the central sympathetic reflexes in cutaneous vasoconstrictor innervation induced by thermoregulatory (whole body warming and cooling) and respiratory stimuli. They measured the skin temperature and skin blood flow in the limbs. In normal conditions, these reflexes do not show differences between the two sides of the body. In patients with CRPS, they found three distinct vascular regulation patterns associated with the duration of the disorder: the warm regulation type (acute stage, <6 months), the intermediate type, and the cold regulation type (chronic stage). They have suggested that the secondary changes of neurovascular transmission that lead to the development of supersensitivity of the vascular smooth muscle, as a consequence of chronically decreased activity in the vasoconstrictor neurons, may account for the severe vasoconstriction and cold skin in chronic CRPS.
During our clinical examination of TBPIs, we have often found patients describing symptoms suggestive of autonomic dysfunction. We embarked on this project with the basic premise to document autonomic dysfunction in such injuries, if any. QSART has been utilized to evaluate the sudomotor dysfunction. Although the axon reflex, measured by QSART, is mediated almost entirely by the postganglionic sympathetic nerve fibres, we decided to use QSART as it is patient compliant, efficient, short, reproducible, economical, and the healthy arm can be used as a control.
Our study is an explorative study involving 20 patients with TBPI. We did not find any statistically significant difference between the affected and normal arm for total sweat volume and latency period. However, the average mean values for total sweat volume were lower in the affected arm as compared to the normal. Similarly, the average mean value for latency period after stimulation was increased in the affected arm as compared to normal arm [Table 1]. The baseline sweat output, however, was significantly lower in the affected arm as compared to the normal arm, as shown in the [Table 1] and [Figure 2].
It can be argued that there is no autonomic dysfunction as the main findings showed no statistically significant difference in the sweat output as assessed by QSART in the affected arm as compared to the normal. Baseline sweating may be an inconsequential marker compared to the main result of the test. However, the overall picture in the affected arm as compared to the normal arm following TBPI is that of hypohidrosis and resembles the chronic stage (cold regulation type CRPS) according to the phases described by Janig et al.
QSART is a test for evaluating the axon reflex, which is a function of the postganglionic sympathetic nerve fibres. A dysfunction in QSART demonstrates abnormalities in the small sympathetic C fibres which innervate the sweat glands. The injuries in our explorative study were all preganglionic. Our results have demonstrated hypohidrosis in the affected arm in general, and a statistically significant difference in the baseline sweat output. Thus, it can be argued that there is autonomic dysfunction, as measured by QSART. However, the dysfunction measured by the test is at the level of the postganglionic sympathetic C fibres while the injuries were all preganglionic. To explain this finding, we propose an antegrade (orthograde) axonal degeneration following preganglionic TBPI. This can explain the dysfunction of small C fibres in such patients.
The study was unable to demonstrate any statistically significant correlation between pain and the total sweat volume, latency period, and baseline sweat output in the affected and normal sides [Table 3] and [Table 4].
Similarly, there was no statistically significant difference in the sweat output in complete and incomplete injuries, in the affected and normal sides [Table 2]. There was no statistically significant correlation between duration of injury and sweat output [Table 5]. Further studies employing a wider range of tests for autonomic functions, along with a larger sample size is being undertaken to state with confidence the mechanism and type of autonomic dysfunction in TBPI.
This study has observed the presence of autonomic dysfunction in patients with TBPI. Though none of the patients complained of overt autonomic dysfunction, it was revealed by QSART. This shows that these patients might suffer from subclinical autonomic dysfunction.
Limitation of the present study is that the recording was done only in 20 patients. Further studies with a larger sample size are required to confirm the findings. The other limitation is the difference in duration after injury between the patients.
QSART has demonstrated reduced baseline sweat output in the affected arm in patients with TBPI. This was found to be statistically significant as compared to the baseline sweat output in the normal arm. This indicates the presence of autonomic dysfunction in the injured arm.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]