Effect of Spinal Cord Stimulation on Skin Temperature in Patients with Neuropathic Pain: An Observational Study Authors: Markus Gehling, Katrin Bertelmann-Barge, Christian Roth Published: 2025 ABSTRACT BackgroundSpinal cord stimulation (SCS) can successfully treat neuropathic pain when conventional medication fails. Current evaluation methods, however, are primarily subjective. This study explored the use of infrared thermography as an objective method to detect physiological effects of SCS.MethodsThis prospective observational study included 25 patients treated with SCS for neuropathic pain. Skin temperature was measured using infrared thermography before and during the first 60 minutes of stimulation. Pain intensity was assessed using an 11-point Numerical Rating Scale (NRS). The primary hypothesis was that SCS would lead to a significant skin temperature increase of at least 1°C.ResultsMean skin temperature in the affected area increased significantly from 29.1 ± 2.3°C to 30.8 ± 1.8°C (p = 0.0002). On the contralateral side, it increased from 29.3 ± 2.2°C to 30.0 ± 1.8°C (p = 0.0037). Temperature responses varied among individuals. A post hoc analysis showed that patients with a temperature increase of ≥1°C (responders) demonstrated significantly greater long-term improvement in pain chronification scores compared to non-responders (p = 0.002).ConclusionSCS produces measurable increases in skin temperature in patients with neuropathic pain. A ≥1°C temperature increase may indicate favorable long-term response. Infrared thermography may serve as a valuable objective measure for evaluating SCS efficacy. Introduction Spinal cord stimulation (SCS) has been shown to effectively relieve neuropathic pain even in cases where conventional pharmacological interventions have failed.1–4 Indications for SCS are determined based on the underlying condition. Randomized controlled trials support its use for failed back surgery syndrome with back or leg pain, complex regional pain syndrome (CRPS), and polyneuropathy. Decisions about permanent SCS implantation are based on both diagnosis and acute effects observed during a trial phase of 5–7 days.5 Psychosocial factors, including depression, anxiety, catastrophizing, low self-efficacy, personality disorders, and poor social support, can influence long-term outcomes, although standardized assessments of these contraindications remain lacking. Currently, SCS indications rely heavily on subjective patient reports during the trial phase. We aimed to support decision-making with objective parameters that are independent of patient cooperation and feasible in routine clinical settings. Increased skin temperature, as a marker of sympathetic inhibition (sympatholysis), appeared promising in this context. Sympatholysis has been shown to alleviate neuropathic pain in CRPS,6,7 diabetic neuropathy,8 and postherpetic neuralgia.9 While its effects are typically transient, SCS may offer a more sustainable alternative. Previous studies have demonstrated the sympatholytic effect of SCS, improving coronary and peripheral blood flow.10–13 These effects may explain improvements in wound healing observed in peripheral artery disease.14,15 Enhanced microcirculation has also been documented with SCS. Sympathetic dysfunction has been particularly studied in CRPS,16 where SCS has been proven effective.2,17 We hypothesized that part of the analgesic effect of SCS may be mediated by sympatholysis, detectable via changes in skin temperature. However, some studies question the clinical relevance of sympatholysis in SCS. For example, microcirculation did not improve in some patients with critical limb ischemia.18 In neuropathic pain, consistent changes in skin blood flow or skin temperature were not demonstrated.19,20 Furthermore, SCS-induced pain relief in CRPS was not always associated with microcirculatory changes,2 and SCS failed to reduce central sympathetic tone in patients with heart failure.21 Objective biomarkers reflecting the physiological effects of spinal cord stimulation (SCS) in neuropathic pain may help optimize patient selection for this therapy. The assessment of SCS-induced changes in skin temperature could serve as a simple and objective measure, independent of patient perception or reporting. We conducted this prospective observational study to investigate whether SCS leads to measurable increases in skin temperature in patients with neuropathic pain. Our hypothesis was that SCS would result in a regional skin temperature increase of at least 1°C. Methods After obtaining ethics approval (Ethics Committee of the State Medical Association of Hesse, 2020–1515-evBO) and registering the study design (DRKS-ID 00022524), the study was conducted at an outpatient pain center in 2020. Participants gave written informed consent. This study complies with the Declaration of Helsinki. Inclusion criteria were patients treated with SCS for neuropathic pain. Exclusion criteria were known peripheral arterial disease, inability to pause SCS for pain control, or lack of consent. Skin temperature was measured in the target area of stimulation over the first 60 minutes after SCS initiation. Pain intensity was assessed using an 11-point NRS (0 = no pain, 10 = worst imaginable pain). Patients received low-frequency stimulation (80–150 Hz; pulse width 80–350 µsec with a Medtronic™ System). Demographic and clinical data were extracted from patient records. Temperature data were collected with a commercial infrared camera (Seek Shot Pro™, accuracy ±0.1°C). Each data point represented the average of ~1000 pixel readings. Measurements were taken after a minimum 6-hour SCS pause, in a temperature-controlled room (20°C ± 1°C to 23°C ± 1°C) following a 30-minute acclimatization. Baseline temperature was recorded from the affected area with active SCS turned off. After switching on stimulation, temperature measurements were repeated at 5, 10, 15, 30, 45, and 60 minutes. Pain intensity was recorded at each time point. Sample size calculations (GraphPad) indicated that 25 measurements with a standard deviation of 0.29 had 80% power to detect a 0.3°C difference (p < 0.05, two-tailed). Data distribution was normal. One-way ANOVA was used for skin temperature and pain scores; other comparisons used non-parametric tests or Fisher’s exact test. In three individual skin temperature measurements, data completion was performed by using an interpolated value (n = 1) or by carrying forward the last valid measurement at the end of the recording sequence (n = 2). All other missing data were excluded from analysis. Results Of 35 eligible patients, 25 were included. One had died, three declined, and six could not be contacted. Table 1 summarizes demographic data. Most patients were long-term responders with reduced pain chronification, disability, and improved quality of life. Pain score also declined promptly after stimulation (Figure 1).Table 1Descriptive DataStudy Groupp-valuen25GenderF / M /On15 / 10 / 0AgeYearsMedian (range)59 (40–80)DiagnosisPeripheral neuropathyn5Radicular painn9Failed back surgeryn7Polyneuropathyn3Complex regional pain syndromen1Duration of SCS-use (months)Median (range)40 (23–59)Grading of chronic pain (Von Korff) prior to implantationMedian (range)3 (1–4)0.0223Grading of chronic pain (Von Korff) at studyMedian (range)2 (1–4)Pain disability prior to implantationMedian (range)41 (14–64)0.0130Pain disability at studyMedian (range)31 (5–59)Depression prior to implantationMedian (range)10 (1–23)Quality of life VR12 prior to SCS-implantationMedian (range)32.6 (12.1–50.5)0.0074Quality of life VR12 at studyMedian (range)43.8 (18.9–62.7) Figure 1Pain intensity in response to spinal cord stimulation (mean ± sem). Table 1Descriptive DataStudy Groupp-valuen25GenderF / M /On15 / 10 / 0AgeYearsMedian (range)59 (40–80)DiagnosisPeripheral neuropathyn5Radicular painn9Failed back surgeryn7Polyneuropathyn3Complex regional pain syndromen1Duration of SCS-use (months)Median (range)40 (23–59)Grading of chronic pain (Von Korff) prior to implantationMedian (range)3 (1–4)0.0223Grading of chronic pain (Von Korff) at studyMedian (range)2 (1–4)Pain disability prior to implantationMedian (range)41 (14–64)0.0130Pain disability at studyMedian (range)31 (5–59)Depression prior to implantationMedian (range)10 (1–23)Quality of life VR12 prior to SCS-implantationMedian (range)32.6 (12.1–50.5)0.0074Quality of life VR12 at studyMedian (range)43.8 (18.9–62.7) Descriptive Data Pain intensity in response to spinal cord stimulation (mean ± sem). After pausing stimulation pause ≥ 6 hours, reactivation led to a significant increase in skin temperature on the affected side from 29.1 ± 2.3°C to 30.8 ± 1.8°C (p = 0.0002, Figure 2). The contralateral side rose from 29.3 ± 2.2°C to 30.0 ± 1.8°C (p = 0.0037, Table 2). Differences between sides were not statistically significant at baseline or at 60 minutes.Table 2Skin Temperature After Start of Spinal Cord StimulationANOVATO10 Min20 Min30 Min40 Min50 Min60 Minp-valueNumber of values25252525252525Ipsilateral to stimulationMean29.1329.4129.6629.8329.9730.1430.110.0006Std. Deviation2.2692.1212.0022.0151.8641.8141.719Std. Error of Mean0.45380.42430.40030.40290.37290.36280.3438Contralateral to stimulationMean29.2629.4729.5429.8229.829.9629.960.0170Std. Deviation2.2252.1171.8951.8781.7921.7341.779Std. Error of Mean0.44490.42350.37910.37560.35830.34680.3558 Figure 2Mean skin temperature changes (mean + sem). Ipsilateral (left) and contralateral (right) skin temperature response to spinal cord stimulation. Table 2Skin Temperature After Start of Spinal Cord StimulationANOVATO10 Min20 Min30 Min40 Min50 Min60 Minp-valueNumber of values25252525252525Ipsilateral to stimulationMean29.1329.4129.6629.8329.9730.1430.110.0006Std. Deviation2.2692.1212.0022.0151.8641.8141.719Std. Error of Mean0.45380.42430.40030.40290.37290.36280.3438Contralateral to stimulationMean29.2629.4729.5429.8229.829.9629.960.0170Std. Deviation2.2252.1171.8951.8781.7921.7341.779Std. Error of Mean0.44490.42350.37910.37560.35830.34680.3558 Skin Temperature After Start of Spinal Cord Stimulation Mean skin temperature changes (mean + sem). Ipsilateral (left) and contralateral (right) skin temperature response to spinal cord stimulation. Figure 3 illustrates individual variability in temperature responses (Figure 3). A post hoc analysis identified responders (≥1°C increase), who showed greater improvement in Von Korff pain grading (1.6 ± 1.1 vs 0 ± 1.0; p = 0.002; Figure 4). Figure 3Individual patient temperature responses. Colored lines represent individual skin temperature responses to spinal cord stimulation. Figure 4Pain grading improvements by thermal response (median and range). Improvement of chronic pain grading in patients with temperature increase > 1°C (black dots) and with temperature increase < 1°C (black squares). Individual patient temperature responses. Colored lines represent individual skin temperature responses to spinal cord stimulation. Pain grading improvements by thermal response (median and range). Improvement of chronic pain grading in patients with temperature increase > 1°C (black dots) and with temperature increase < 1°C (black squares). Discussion SCS led to a continuous, significant increase in skin temperature in most patients. However, responses varied. Post hoc analysis indicated that only some patients experienced a clinically meaningful rise. These responders also showed better long-term outcomes. The differential response may be explained by variation in sympathetic nerve function and damage. Animal studies suggest that SCS induces peripheral vasodilation not only via regional sympathetic inhibition,22,23 but also through antidromic depolarization of sensory C- and A-delta fibers.24–26 These fibers release calcitonin gene-related peptide (CGRP), which stimulates nitric oxide production in endothelial cells, promoting vasodilation.26 While these mechanisms are well documented, direct links between sympatholysis and analgesia remain unproven. Pain relief via SCS is thought to involve spinal and supraspinal mechanisms, including GABA release27,28 and modulation of nociceptive signaling via microglia and serotonergic pathways.29 Clinical data on thermographic effects of SCS are sparse. Augustinsson et al (1985) reported increased skin temperature in vascular patients. Robaina et al (1989) observed similar effects in a mixed cohort including CRPS and Raynaud’s. Velasco et al (2009) reported temperature increases with motor cortex stimulation in neuropathic pain.30–32 However, Devulder et al (1996) found no temperature change in a thermography study involving post-laminectomy patients.20 Our methodology differed in terms of measurement site, duration, and pre-conditioning, which may explain these discrepancies. The study design and sample size do not allow for confirmatory statistical analysis. Although our sample was small, it was representative of our treatment population. Infrared thermography proved to be a reliable, non-invasive tool, independent of subjective factors, with potential value in SCS evaluation. Conclusion Our findings suggest that a skin temperature increase ≥1°C following SCS may indicate favorable long-term outcomes in patients with neuropathic pain. Infrared thermography may offer an objective clinical marker. Further studies are needed to determine sensitivity, specificity, and underlying mechanisms.