Endoscopic sympathectomy impairs forearm cutaneous vascular conductance and sweating responses during passive whole-body heat stress

Abstract

Hyperhidrosis is characterized by excessive sweating, typically in the palmar and axillary regions. Individuals with severe cases of hyperhidrosis can be treated surgically via endoscopic thoracic sympathectomy to disrupt the neural connection between the spinal cord and the affected regions, thereby attenuating sweating in the hyperhidrotic regions. However, the effects of this procedure on skin blood flow and sweating responses to whole-body heat stress in regions not targeted by the sympathectomy, such as the forearm, are unknown. We tested the hypothesis that surgical sympathectomy attenuates the increase in forearm cutaneous vascular conductance and sweating during whole-body heat stress. Prior to, and 7±1 weeks following bilateral thoracic surgical sympathectomy, forearm skin blood flow (laser-Doppler imaging) and local sweat rate (capacitance hygrometry) were assessed during whole-body passive heat stress sufficient to elevate core temperature ~1.0°C (p = 0.74 between trials) in 10 subjects (27±3 yrs; 7 female). The mean increase in forearm cutaneous vascular conductance was attenuated following sympathectomy (Δ3.67±1.34 vs. Δ1.62±1.19 flux*mmHg⁻¹; p < 0.01). Similarly, mean sweat rate was attenuated in the palmar (Δ0.68±0.42 vs. Δ0.20±0.25 mg*min⁻¹*cm⁻²; p = 0.015) and forearm (Δ0.62±0.31 vs. Δ0.24±0.26 mg*min⁻¹*cm⁻²; p = 0.03) regions following sympathectomy. These data suggest that the increase in forearm cutaneous vascular conductance and sweat rate in response to whole-body heat stress is attenuated by ~50% after surgical sympathectomy. However, the implications of these responses on whole-body thermoregulation remain unknown.

Graphical Abstract

Introduction

Changes in skin blood flow can be neurally mediated by adrenergic and cholinergic branches of the sympathetic nervous system. In response to an increase in core temperature, there is an initial rise in skin blood flow mediated by withdrawal of sympathetic adrenergic vasoconstriction (1). As core temperature continues to rise, acetylcholine and other neurotransmitters are released from sympathetic cholinergic nerves, allowing further increases in skin blood flow and initiating sweating (1). This increase in skin blood flow and sweating is necessary to promote convective and evaporative heat loss to maintain heat balance. Individuals with attenuated increases in skin blood flow and sweating during heat stress have greater increases in core temperature and may be at increased risk for heat-related illnesses (2, 3).

Primary hyperhidrosis is a condition characterized by excessive sweating in the palmar, plantar, and/or axillary regions primarily in response to emotional distress and is typically due to over functioning of sympathetic cholinergic nerve fibers (4–6). Individuals with severe cases of hyperhidrosis can be treated surgically via bilateral endoscopic thoracic sympathectomy that consists of electrocautery or clipping of a selected ganglion between the second and fifth thoracic vertebrae. As such, sweat glands in the hyperhidrotic and surrounding regions become functionally denervated, resulting in varying degrees of hypohidrosis and/or anhidrosis in the affected regions in response to emotional distress (7). However, the impact of this procedure on efferent thermoregulatory responses remains unknown.

Previous studies have reported that endothelium-dependent cutaneous vasodilation to local skin heating is preserved in individuals with sympathectomized limbs (8, 9); however, the vasodilatory mechanisms in response to local heating are fundamentally distinct relative to the neural reflex-mediated mechanisms in response to increases in core temperature. The purpose of this study was to assess the effect of the sympathectomy procedure on efferent thermoregulatory responses driven by increases in core temperature. We hypothesized that the increase in forearm cutaneous vascular conductance (FVC), as well as forearm and palmar sweating, would be attenuated during whole-body heat stress following surgical sympathectomy.

Methods

Subjects

All experimental procedures were approved by the Institutional Review Boards at the University of Texas Southwestern Medical Center and Texas Health Presbyterian Hospital Dallas (#0700323). The study was performed in accordance with the principles outlined in the Declaration of Helsinki.

Ten adults (27±3 yrs; 169 ± 11 cm; 71 ± 17 kg; 7 female) exhibiting clinical manifestations of palmar and/or axillary primary hyperhidrosis were tested pre and post (7±1 weeks) bilateral endoscopic thoracic sympathectomy. The same surgeon performed the procedure for all subjects. The specific thoracic ganglion that was removed following monopolar cautery was reported in the surgical notes for 6 out of the 10 participants [the second (n = 1), third (n = 4), or fourth (n = 1) thoracic ganglion] (10). Though for the remaining 4 participants it is unknown which ganglion was removed, it is standard procedure to remove either the second, third, or fourth thoracic ganglion for hyperhidrosis (11). Sympathectomy was first performed on the left side, followed by sympathectomy on the right side for each subject, with confirmation that the removed tissue was neural in origin via histology.

Laboratory Measurements

Subjects wore a two-piece, tube-lined water perfusion suit that covered their entire body except the head, forearms, hands, and feet. An index of core temperature was measured using an ingestible telemetric pill (n = 5) or via a thermocouple placed in the sublingual sulcus (n = 5); for each participant, the same approach to measure core temperature was used for pre- and post-sympathectomy visits. For oral temperature measures, the thermocouple was kept in the sublingual sulcus for the entirety of the heating protocol. For both measures, pre-heating core temperatures were recorded once this variable had stabilized. Mean skin temperature was calculated as the weighted average of measurements from thermocouples placed on the calf (11%), thigh (14%), abdomen (14%), chest (22%), lower back (19%), and upper back (19%). Heart rate was obtained from an electrocardiogram (GE Healthcare, Milwaukee, WI, USA), and blood pressure was measured via auscultation of the brachial artery. Local sweat rate was measured using capacitance hygrometry on the palmar and forearm regions of either the left or right arm. Red cell flux was measured at the beginning and end of heat stress using laser-Doppler imaging (Moor LDI, Moor Instruments, UK) on the forearm and was normalized to mean arterial pressure to obtain FVC. It should be noted that the regions where FVC and sweating were assessed were not in contact with the water-perfusion suit. Thus, any changes in these variables (particularly skin blood flow/conductance) were due to reflex, not local heating, induced changes.

Experimental Protocol

Upon arrival at the laboratory, the participant was instrumented to measure ECG and arterial blood pressure (sphygmomanometry) from the brachial artery at both pre- and post-sympathectomy visits. The participant put on the water-perfusion suit, followed by resting supine for ~30 min with normothermic water (34 °C) perfusing the suit. During this resting period, sweat capsules (surface area of 2.83 cm²) were placed on the forearm and palmar areas. Anhidrotic nitrogen was perfused (300 ml/min) through these capsules such that any sweat on the skin’s surface under the capsule would fully evaporate into the nitrogen. Local sweat rate was calculated via capacitance hygrometry based on the absolute humidity of the effluent nitrogen gas and the flow rate through the capsule (Vaisala, Woburn, MA). Normothermic baseline (pre-heating) skin blood flow values were obtained via laser-Doppler imaging (Moor LDI, Moor Instruments, UK) from the dorsal aspect of both forearms, with the responses from both forearms averaged. The scanned areas were 3.4±0.8 cm x 7.9±4.3 cm, with each scan taking less than 2 min to complete. Using anatomical landmarks and/or marking the skin, care was taken to ensure that sweat rate and skin blood flow were assessed from the same areas within an experimental visit (for skin blood flow) and pre/post-sympathectomy visits (for both skin blood flow and sweat rate). The participants were then heat-stressed by perfusing 48-50 °C water through the tube-line suit, with a goal of increasing their core temperature by 1.0 °C. Participants were not permitted to drink water at any time during the heating protocol. Reported data depict the magnitude of the increase in assessed values from the normothermic baseline (pre-heating) to the end of heat stress in the pre- and post-sympathectomy trials.

Data and Statistical Analyses

Continuous measures (e.g., sweat rates, core and skin temperatures, ECG and heart rate, etc.) were collected using data acquisition software (Biopac MP150, Santa Barbara, CA, USA) at a minimum sampling frequency of 50 Hz. FVC was quantified as laser-Doppler flux units divided by mean arterial blood pressure that was obtained immediately before or after each scan. Post-hoc power analyses, as well as effect size calculations, were conducted based on the obtained means and standard deviations of the variables and the sample size of our study. This resulted in a statistical power ≥ 0.98 for both FVC and sweat rate. Paired samples t-tests were performed (GraphPad, Prism v. 10.5, GraphPad Software, San Diego, CA) to compare the magnitude of the increase in core temperature, skin temperature, local sweat rate (both forearm and palm), and FVC in response to whole-body heating pre- and post-sympathectomy. Statistical significance was accepted at p < 0.05.

Results

By design, the change in core temperature during whole-body heating was not different between pre- and post-sympathectomy (0.95 ± 0.26 vs. 0.97 ± 0.33 °C; p = 0.74). The time to increase core temperature by ~1°C was not different pre- and post-sympathectomy (51 ± 14 vs. 52 ± 12 min; p = 0.94). Likewise, there was no difference in the increase in mean skin temperature during heating pre- to post-sympathectomy (Δ3.39 ± 0.81 vs. Δ3.57 ± 0.74 °C; p = 0.47). However, local sweat rate from the forearm (Figure 1A; Δ0.62 ± 0.31 vs. Δ0.24 ± 0.26 mg*min⁻¹*cm⁻²; p = 0.030; Cohen’s d = 1.3) and palmar (Figure 1B; Δ0.68 ± 0.42 vs. Δ0.20 ± 0.25 mg*min⁻¹*cm⁻²; p = 0.015; Cohen’s d = 1.4) regions during heat stress were both attenuated post-sympathectomy. Similarly, the magnitude of the increase in FVC in response to whole-body heating was attenuated following sympathectomy (Figure 2; Δ3.67 ± 1.34 vs. Δ1.62 ± 1.19 flux*mmHg⁻¹; p = 0.008; Cohen’s d = 1.6). There was no difference in FVC during the normothermic baseline (pre-heating) period between pre- to post-sympathectomy trials (1.45 ± 0.45 vs. 1.44 ± 0.61 flux*mmHg⁻¹; p = 0.96).

Figure 1.

Change in local forearm (panel A) and palm (panel B) sweat rate (SR) responses to whole body heating pre (open circles) and post (blue circles) sympathectomy. The triangle represents the individual who received a T2 sympathectomy, the circles represent individuals who received a T3 sympathectomy, the square represents an individual who received a T4 sympathectomy, and the diamonds represent individuals whose level of sympathectomy was not reported. Data are shown as mean and standard deviation with individual data points. Each individual’s response is illustrated by connecting lines. Data were analyzed using a paired t-test.

Figure 2.

Change in forearm cutaneous vascular conductance (FVC) to whole body heating pre (open circles) and post (blue circles) sympathectomy. The triangle represents the individual who received a T2 sympathectomy, the circles represent individuals who received a T3 sympathectomy, the square represents an individual who received a T4 sympathectomy, and the diamonds represent individuals whose level of sympathectomy was not reported. Data are shown as mean and standard deviation with individual data points. Each individual’s response is illustrated by connecting lines. Data were analyzed using a paired t-test.

Discussion

We found that following endoscopic sympathectomy, local sweat rates from the forearm and palmar regions are attenuated to a whole-body heat stress sufficient to cause a ~1.0 °C increase in core temperature. Further, we show that the increase in FVC in response to whole-body heating is attenuated by ~50% following endoscopic sympathectomy.

Factors such as stress, anxiety, fear, and pain trigger the release of catecholamines from sympathetic fibers, activating apocrine and eccrine sweat glands and inducing emotional sweating. Individuals with primary hyperhidrosis experience excessive sweating in the facial, axillary, palmar, and plantar regions in response to such emotional stimuli. This condition is likely due to a hyperactive sympathetic nervous system, rather than dysfunction of the sweat glands (12). As such, a surgical treatment for primary hyperhidrosis is an endoscopic sympathectomy to inhibit/block sympathetic outflow to these regions, including the stressors outlined above. This procedure is highly successful in treating primary hyperhidrosis, with 80–90% of patients reporting immediate symptom relief and improved quality of life due to reduced emotional sweating (13, 14). Our findings in the present study demonstrate that endoscopic sympathectomy may also attenuate thermoregulatory sweating in the forearm and palm. Following a ~1°C increase in core temperature, local sweat rate in the palmar and forearm regions was attenuated by more than 50%. Thermoregulatory sweating is primarily mediated by sympathetic cholinergic nerves through the release of acetylcholine from postganglionic fibers acting on eccrine sweat glands. Thus, disrupting sympathetic outflow via sympathectomy likely inhibited activation of eccrine sweat glands, thereby attenuating sweat rate. Further, as the morphology of sweat glands is not altered or dysfunctional in individuals with primary hyperhidrosis (12), it is hypothesized that the observed attenuated sweat rate resulted from fewer sympathetically activated sweat glands rather than a reduced secretion rate per gland. However, it remains unknown whether sympathectomy alters sweat gland morphology.

Previous studies have shown that endothelium-dependent vasodilation in response to local heating or to exogenous acetylcholine administration is not altered following a sympathectomy (8, 9). Local warming of the skin results in cutaneous vasodilation, mediated first by local sensory nerves, followed by a longer, neurally independent phase predominantly mediated by endothelial nitric oxide (15). In contrast, reflex-induced cutaneous vasodilation in response to whole-body heating (i.e., increases in core temperature) requires the release of acetylcholine and co-neurotransmitters from sympathetic cholinergic nerves (1, 16–18). As such, the attenuated increase in FVC following sympathectomy in the present study was likely due to altered efferent neural outflow and inhibition of neurotransmitter release from sympathetic nerves, rather than impaired endothelial sensitivity. This inhibition of the release of neurotransmitters likely prevented, or attenuated, the synthesis of nitric oxide and therefore resulted in attenuated FVC.

Interestingly, the magnitude of FVC attenuation post-sympathectomy was highly variable: some participants showed essentially no change in skin FVC during whole-body heat stress, while others maintained a moderate, though attenuated, increase. This variance may be due to differences in the level of thoracic ganglia cauterized during sympathectomy. Endoscopic sympathectomy to treat primary hyperhidrosis is typically performed at either the second, third, or fourth thoracic ganglia. Cauterization at the second ganglia may result in greater disruption and denervation of sympathetic outflow than cauterization at lower ganglia levels (i.e., third or fourth ganglia), and therefore the former may result in a greater attenuation in the increase in FVC during heat stress. Consistent with this thought, one of the participants in the current study who received cauterization at the second ganglia had an increase in FVC during heating pre-sympathectomy (Δ 3.31 flux*mmHg⁻¹) but had no change in FVC in response to heat stress post-sympathectomy (Δ −0.08 flux*mmHg⁻¹). Although unconfirmed, it is possible that the level of cauterization contributes to the variability observed in the magnitude of FVC impairment during whole-body heat stress and thus warrants further investigation.

During ambient heat exposure, increases in sweat rate and skin blood flow are critical for promoting heat loss and maintaining safe core temperatures. Individuals with compromised thermoregulatory responses are at greater risk for heat-related illnesses, often due to complications associated with greater increases in core temperature (3). While the present study demonstrates attenuated responses in forearm sweat rate and FVC to heat stress following sympathectomy, it is important to note that these results do not directly inform us about any thermoregulatory risks that may occur in this population during extreme heat events. The present study utilized a water-perfused suit to increase core temperature pre- and post-sympathectomy. This approach allowed us to answer important mechanistic questions regarding thermoregulatory responses in sympathectomized individuals, but whether these responses affect changes in core temperature during ambient heat exposure requires further investigation. Secondly, we do not know how extensive the impairments in FVC and sweating post-sympathectomy are, given that we assessed only responses from the forearm (skin blood flow and sweating) and palmar (sweating only) regions. Plantar hyperhidrosis may also be treated by sympathectomy or a sympathetic block between the second and fourth ganglia. Patients report reductions in plantar sweating following these procedures (19, 20), but it remains unknown whether sweating or skin blood flow responses are also attenuated in the leg and plantar regions during thermal strain. Finally, a common side effect reported by patients following sympathectomy includes compensatory hyperhidrosis in other regions of the body, such as the back, chest, and thigh (21–23). However, whether this compensatory hyperhidrosis contributes to differences in whole-body sweat rate, and thus changes in skin and core temperatures during thermal stress, remains unknown.

Conclusion

To our knowledge, this is the first study to examine efferent thermoregulatory responses in individuals pre- and post-sympathectomy. We demonstrate that endoscopic sympathectomy attenuates the increase in forearm vascular conductance, as well as palmar and forearm sweating, during whole-body passive heat stress. These data show that the consequences of the sympathectomy procedure can extend beyond the regions (i.e., palmar, axillary, plantar) targeted by that procedure. Whether these impaired responses are extensive enough to compromise thermoregulatory function and heighten the risk of heat-related illnesses warrants further investigation.

New and Noteworthy.

Thoracic surgical sympathectomy is often performed to attenuate sweating in the palmar and axillary regions in individuals with hyperhidrosis. However, the effect of this procedure on efferent thermoregulatory responses to heat stress are unknown. We show that the increase in forearm cutaneous vascular conductance and local sweat rate is attenuated during whole-body passive heating following a surgical sympathectomy.

Data Availability

Deidentified participant data will be available from the corresponding author (Dr. Craig Crandall, craigcrandall@texashealth.org)