Abstract
Objective. Response to lumbar epidural steroid injection in lumbar radicular pain varies. The purpose of this study is to characterize the changes in quantitative sensory testing (QST) phenotypes of subjects and compare the QST characteristics in patients who do respond to treatment of radicular pain with a lumbar epidural steroid injection (ESI).
Design. Prospective, observational pilot study.
Setting. Outpatient pain center.
Methods. Twenty subjects with a lower extremity (LE) radicular pain who were scheduled to have an ESI were recruited. At the visit prior to and four weeks following an ESI, subjects underwent QST measurements of both the affected LE and the contralateral unaffected UE.
Results. Following an ESI, nine subjects reported a greater than 30% reduction in radicular pain and 11 reported a less than 30% reduction in radicular pain. Subjects who had less than 30% pain reduction response (nonresponders) to an ESI had increased pre-injection warm sensation threshold (37.30 °C, SD = 2.51 vs 40.39, SD = 3.36, P = 0.03) and heat pain threshold (47.22 °C, SD = 1.38, vs 48.83 °C, SD = 2.10, P = 0.04). Further, the nonresponders also showed increased pre-injection warm sensation threshold as measured in the difference of warm sensation detection threshold difference in the affected limb and the unaffected arm (2.68 °C, SD = 2.92 vs 5.67 °C, SD = 3.22, P = 0.045). Other QST parameters were not affected.
Conclusions. The results show that the nonresponders to ESIs have increased detection threshold to heat pain and warm sensation, suggesting that a preexisting dysfunction in the C fibers in this group of subjects who can be detected by QST. Such altered QST characteristics may prognosticate the response to ESIs.
Introduction
The development of classic lower extremity radicular pain and parasthesias is associated with myriad pathophysiology such as intervertebral disc herniation and spinal osteoarthritis and/or localized inflammation through the release of inflammatory cytokines and arachidonic acid-derived mediators [1]. The local application of inflammatory substances will affect neuronal conduction primarily in vulnerable unmyelinated C fibers [2]. The application of steroids to the epidural space is a therapeutic option for the treatment of radicular pain. Epidural steroid injections (ESIs) have been used widely to treat lumbar radicular pain. However, the efficacy of ESIs is inconsistent and the procedure carries the risk of severe adverse complications [3,4]. Identifying patients who are most likely to benefit from ESIs for selective intervention may enrich the success rate of ESIs while avoiding unnecessary injections as well as the associated potential complications.
The chronic mechanical compression of lumbar nerve roots results in histologic and physiologic changes of pseudo-unipolar dorsal root neurons. Histological analysis of spinal nerve root tissue from subjects with lumbar radicular pain symptoms and evidence of spinal nerve root compression demonstrates decreased large myelinated Aβ and, to a lesser extent, Aδ fibers with relative preservation of myelinated C fibers [5]. The progressive decrease in number and/or changes in function of nerve fibers manifests clinically as decreased vibratory sensation with Aβ fiber dysfunction, followed by loss of cold sensation with Aδ dysfunction and finally loss of heat sensation with C fiber dysfunction [6].
Quantitative sensory testing (QST) utilizes thermal, mechanical, and electrical stimulation to interrogate the function of discrete neural fibers in vivo. Radicular pain has been associated with the development C fiber dysfunction as evidenced by increased warm thresholds in the affected extremity compared with the contralateral extremity [7]. QST measurement parameters are generally well preserved across populations, making the cause of deviations intriguing for both diagnostic and research purposes. QST measurements also have indicated that increased warmth sensation threshold has been associated with unsatisfactory reductions in pain following disc decompression surgery [8]. Other work has demonstrated that both heat and cold pain thresholds are affected in subjects with radicular pain compared with healthy volunteers [6]. In multiple studies, thermal hypoesthesia has been specifically associated with cervical radiculopathy, whereas increased cold pain sensitivity was noted to be associated with both cervical radiculopathy and nonspecific arm pain [9,10]. Additionally, evidence also exists for alterations solely in cold sensation in subjects with radicular pain [11]. Finally, studies of QST measurements suggest that increased cold sensation threshold, indicating Aδ fiber dysfunction, is correlated with a decrease in analgesic response to epidural steroid injections [12].
The purpose of this prospective pilot study is to evaluate and compare the thermal QST phenotypes of subjects who respond with a greater than 30% reduction in lower extremity radicular pain in response to ESI to subjects who do not receive such pain relief from an ESI.
Methods
Study Design
The study was approved by the Massachusetts General Hospital’s institutional review board. The study was registered with www.clinicaltrials.gov on April 24, 2014, and assigned the registration number NCT02130258. All subjects had the study fully explained and signed a provided statement of informed consent prior to participation. The study was designed as a single-center, prospective, nonrandomized observational trial with subject’s nonaffected extremity serving as an internal control. All subjects were recruited from the Center for Pain Medicine Clinic, Boston, Massachusetts, USA. Consecutive subjects between the ages of 18 to 80 years with a diagnosis of unilateral lower extremity radicular pain who were scheduled for a lumbar ESI were eligible and offered to participate in the study. Subjects were excluded from recruitment for the following reasons: use of illicit drugs or medications as detected by urine toxicology screening, administration of an interventional pain procedure capable of altering QST measurements within eight weeks of recruitment, change in the type or dose of any neuropathic pain medication within two weeks of recruitment, pregnancy, involvement in litigation regarding the current lumbar pain complaint, and history of major psychiatric disorder, such as major depression, bipolar, schizophrenia, anxiety disorder, or psychosis requiring hospitalization within one month of enrollment.
Subject demographics including gender, age, height, weight, body mass index (BMI), alcohol, and tobacco and illicit substance use were collected. A pain history was also obtained, including duration of pain, presence of weakness, presence of numbness, and medications for pain treatment. Subjects who were not working as a result of their pain were excluded from recruitment due to poor potential for improvement in their pain. However, subjects who were not working due to retirement and not the result of pain were recruited. Pain medications were classified as either opioid therapy, including tramadol, neuropathic pain medications, including anti-epileptics and antidepressants, or nonsteroidal anti-inflammatories (NSAIDs), which included acetaminophen except when combined with opioids by a manufacturer such as Vicodin.
Twenty-three subjects were recruited for the study. Three subjects were excluded after expressing a desire to not proceed with thermal QST analysis. No subject reported an adverse event related to participation in the trial. The age of patients who enrolled ranged from 30 to 72 years.
Epidural Steroid Injection Procedure
Subjects who met the inclusion criteria were scheduled to undergo a lumbar intralaminar ESI. Blood pressure, heart rate, and oxygen saturation were continuously monitored throughout the procedure. Subjects were positioned in the prone position on a fluoroscopy table. The back was prepped and draped in a sterile fashion. A 22-gauge, 3.5-inch Tuohy needle was advanced under fluoroscopic guidance into the intralaminar epidural space using the loss-of-resistance technique. Coaxial anterior-posterior, lateral, and contralateral oblique imaging confirmed correct needle location. Injection of a small volume (0.1–0.5 mL) of nonionic contrast demonstrated outline of epidural space and nerve roots. A solution of 2 mL of triamcinolone (40 mg/mL) and 3 mL of lidocaine (10 mg/mL) was injected slowly. The Tuohy needle was then removed, and the subject was observed for 30 minutes until discharge. Subjects were asked to return to the clinic in four weeks for repeat QST testing. At four weeks following the ESI, the subjects returned to clinic. Subjects were categorized as to whether they had a greater than or less than 30% reduction of self-reported pain in response to the ESI.
Thermal Quantitative Sensory Testing
Subjects underwent thermal QST phenotype measurements immediately before receiving an ESI and four weeks after the ESI was performed. At both times, thermal QST measurements were recorded from both the painful area of the affected lower limb with radicular pain and the contralateral, unaffected upper extremity. Thermal QST testing was conducted in the same location at both the initial and follow-up visits. QST testing was conducted in the affected dermatome distal to the knee or an unaffected dermatome distal to the elbow in a nonpainful dermatome. Although differences in QST measurements have been noted in healthy volunteers between different regions of the body, the upper extremity was thought to provide a more consistent control site as it would be less likely to be subject to the confounding effects of epidural steroid and local anesthetic, to which the contralateral lower extremity would be exposed [13]. QST testing was carried out by research staff who were not involved directly in subject’s medical care.
QST responses to thermal stimulation were examined using Medoc Thermal Sensory (Medoc Advanced Medical Systems, Durham, NC, USA). The QST examiner was blinded to the subject’s reported change in pain intensity in response to the ESI. Each QST session was carried out in a quiet room maintained at 25°± 2 °C. A contact thermode (30 mm x 30 mm) was gently attached to the affected lower extremity distal to the knee in the affected dermatome and the contralateral upper extremity distal to the elbow in each subject. By pressing a computer mouse button, each subject was able to stop stimulation at any time during a session. Four categories of thermal QST parameters were examined in the following sequence: cold and warm sensation, cold and heat pain threshold, cold and heat pain tolerance, and temporal summation of pain to heat stimulation. Each test was repeated three times with a three-minute interval between tests. The average of the three tests is reported. Pain was also assessed by an 11-point numerical rating scale, with 0 indicating no pain and 10 indicating the worst pain possible (NRS-11).
For cold and warm sensation thresholds, the temperature at the thermode changed at 1 °C/s from a neutral temperature of 32 °C to a temperature that the subject indicated as the initial perception of cold or warm, respectively. Subjects were instructed to stop stimulation when they first perceived cold or warm sensation as temperature changed. For heat pain and cold pain thresholds, the temperature at the thermode changed at 1.5 °C/s from a neutral temperature of 32 °C to a maximal cutoff temperature of 53 °C (heat stimulation) or a minimal cutoff temperature of 0 °C (cold stimulation) to avoid tissue injury. Subjects were instructed to stop the stimulation and indicate when they initially perceived a painful or uncomfortable sensation; the temperature at which the subject stopped the stimulation was recorded as threshold temperature.
To examine pain tolerance, two protocols were used. 1) To detect the maximal tolerable temperature (°C) for cold or heat pain, subjects were asked to tolerate the stimulation beyond their cold or heat pain threshold until it reached the maximal tolerable level. The maximum temperature was preset at 53 °C and 0 °C for heat pain and cold pain, respectively, to avoid tissue injury. 2) To detect the duration (seconds) of tolerance to supra-threshold heat pain stimulation, subjects were asked to tolerate, as long as he or she could, heat stimulation preset at 47 °C for a maximum of 60 seconds.
To examine temporal pain summation, a train of four identical stimuli at 47 °C, separated by a 2.2-second interval between stimuli, was applied to the subject’s forearm and affected lower extremity. Each time the thermode reached a temperature of 47 °C, it was referred to as a “peak.” Subjects were asked to rate their pain by 100 mm VAS at each peak. The VAS at the first peak was then subtracted from the VAS at the fourth peak.
To assess conditioned pain modulation (CPM), heat stimulation was used as the “test stimulation,” whereas cold stimulation was used as the “conditioning stimulation.” Conditioning cold stimulation was delivered by immersing the nontested arm (ipsilateral to the affected lower extremity) in cold water for 30 seconds. Heat stimulation (47 °C, four seconds) was delivered via a contact thermode attached to the affected lower extremity dermatome or unaffected contralateral upper extremity dermatome in each subject, and the pain response was rated. This was recorded as “baseline pain score.” The subject was then asked to immerse the nontested hand in a water bath with temperature controlled at 12 °C. Following 15 seconds of immersion, while the nontested upper extremity was still in the cold water bath, the second test stimulation was delivered to the contralateral control upper extremity or the affected lower extremity and pain intensity was recorded again (Test 1). Fifteen seconds later, subjects were asked to remove their nontested upper extremity from the cold water bath (with the total time of upper extremity immersion in the cold water bath being 30 seconds). Two additional heat stimuli to the tested contralateral control upper extremity or affected lower extremity were conducted 15 and 30 seconds subsequent to the removal of the nontested upper extremity from the cold water bath, designated as Test 2 and Test 3, respectively. Following the delivery of each test stimulus, subjects were asked to report the NRS-11 pain score. The average of the three tests is reported. The protocol for CPM is graphically explained in Figure 1.
Figure 1.
Conditioned pain modulation protocol and diagram. 1) Thermode starts at 37 °C. 2) Baseline pulse lasts four seconds. 3) Twelve seconds following the end of the pulse (not including the temperature decrease) the hand is submerged in the water bath. 4) Pulse 1 occurs once the hand has been submerged for 15 seconds (temperature increase starts at 14 seconds). 5) Hand is removed from water bath for a total duration of 30 seconds. 6) Pulse 2 occurs 15 seconds after removal of hand from water bath (increase starts at 14 seconds). 7) Pulse 3 occurs 30 seconds after removal of hand from water bath (increase starts at 29 seconds), which gives 11 seconds between the two stimuli.
The results of thermal QST measurements from affected lower extremity dermatomes were compared between those who had a greater than 30% decrease in pain in response to an ESI and those who did not. Additionally, the differences in thermal QST measurements between the affected lower extremity dermatome and the unaffected upper extremity dermatomes were compared. The postepidural thermal QST measurements were similarly compared.
Statistical Analysis
All data were coded in an Excel file (Microsoft, Redmond, WA, USA). Categorical variables were expressed as proportions and compared using the chi-square test. Continuous variables were expressed using mean (SD) and compared using the t test (if the normality assumption was true) or the Kruskal-Wallis test (if normality assumption was not true). Normality was determined using a Shapiro-Wilk test. Logistic regression method was used to assess the odds of less than 30% reduction in pain association with gender, alcohol use, opioid use, BMI, and postepidural pain. These variables were selected based on the AIC criteria. The analyses were performed using R version 3.2.2 (R foundation for Statistical Computing, Vienna, Austria). Statistical significance was set at a P value of less than 0.05, and all tests were two-sided.
Results
The demographic information and pain history of the 20 subjects who completed the trial are given in Table 1. Nine subjects reported a decrease in pain of greater than 30% from baseline in response to an epidural steroid injection. The average change in NRS-11 pain score for this group was 4.06 (SD = 3.17). Eleven subjects reported a less than 30% reduction in pain from baseline in response to an epidural steroid injection. The average change in NRS-11 pain score in this group was 0.27 (SD = 0.90). Three subjects in the second group had a worsening of their pain during the four weeks between the ESI and the follow-up visit. Two subjects in the second group reported no change in their pain score. The rest reported minor changes in their pain score (<30%).
Table 1.
Demographic and analgesic profile of subjects
| Greater than 30% pain relief (N = 9) | Less than 30% pain relief (N = 11) | P | |
|---|---|---|---|
| Gender | 2 male, 7 female | 5 male, 6 female | |
| Age, y | 57.55 (14.49) | 54.44 (11.56) | 0.61 |
| Smoke cigarettes | 0/9 | 1/11 | |
| Drink alcohol | 8/9 | 3/11 | |
| Weakness present | 5/9 | 4/11 | |
| Numbness present | 6/9 | 5/11 | |
| Height, in | 66.56 (3.05) | 66.73 (5.21) | 0.56 |
| Weight, lbs | 151.25 (25.51) | 180.91 (23.53) | 0.02* |
| BMI, kg/m2 | 24.92 (3.18) | 28.66 (3.24) | 0.02* |
| Opioid use | 1/9 | 6/11 | |
| Neuropathic use | 1/11 | 2/11 | |
| NSAID use | 5/9 | 3/11 | |
| Duration of pain | 2.33 (3.76) | 4.93 (4.55) | 0.26 |
| Pre-epidural pain intensity | 5.39 (2.47) | 5.45 (1.81) | 0.95 |
| Postepidural pain intensity | 1.56 (1.33) | 5.18 (1.66) | <0.01* |
| Change in pain score | 4.06 (3.17) | 0.27 (0.90) | <0.01* |
Indicates a P value of less than 0.05. NSAID = nonsteroidal anti-inflammatory.
There were several differences in demographics and medication use between the two groups. The subjects who had a greater than 30% reduction in their pain in response to an epidural were more often female (7/9 vs 6/11), currently consumed alcohol (8/9 vs 3/11, chi-square P = 0.02), had a lower weight (151.25 pounds, SD = 25.51, vs 180.91 pounds, SD = 23.53, P = 0.02) and a lower BMI (24.92, SD = 3.18, vs 28.66, SD = 3.24, P = 0.02). Additionally, subjects who reported at least a 30% reduction in pain less frequently utilized opioids (1/9 vs 6/11, chi square P = 0.12) and more frequently utilized nonsteroidal anti-inflammatory (NSAID) medications (5/9 vs 3/11, chi square P = 0.40). The reasons for the different rates of medication use were not fully explored in this study. The groups were approximately equivalent in terms of the number that smoked cigarettes, the number with preexisting weakness or numbness, the subject’s height, duration of pain, and pre-epidural rating of their pain.
Warm/Cold Sensation Threshold
The results of all thermal QST measurements are provided in Table 2. A difference between the two groups was noted for the warm sensation threshold in the pre-epidural affected lower extremity value. Those who had less than 30% reduction in pain had an increased warm sensation threshold (37.30 °C, SD = 2.51, vs 40.39 °C, SD = 3.36, P = 0.03). A difference was also noted between the two groups for the warm sensation threshold when the pre-epidural upper extremity value was subtracted from the pre-epidural lower extremity value. This did not meet criteria for significance with the P values adjusted for multiple comparisons (adjusted P = 0.0125). Those who had less than 30% pain reduction in response to an ESI had a wider difference between the two extremities in warm sensation threshold (2.68 °C, SD = 2.68, vs 5.67 °C, SD = 3.22, P = 0.04). Other warm sensation average measurements were statistically similar. No differences in cold sensation threshold measurements were observed between those who received at least a 30% reduction in pain in response to an epidural and those who did not.
Table 2.
Quantitative sensory testing phenotypes of those who responded to epidural steroid injections
| Pre-epidural Leg | Greater than 30% pain relief | Less than 30% pain relief | P | Test |
|---|---|---|---|---|
| Cold sensation threshold | 27.25 (2.66) | 25.96 (6.39) | 0.55 | t |
| Warm sensation threshold | 37.30 (2.51) | 40.39 (3.36) | 0.03* | t |
| Cold pain threshold | 10.49 (10.28) | 2.35 (5.98) | 0.06 | KW |
| Heat pain threshold | 47.22 (1.38) | 48.83 (2.10) | 0.04* | t |
| Cold pain tolerance threshold | 0.28 (0.76) | 1.28 (3.71) | 0.40 | t |
| Heat pain tolerance threshold | 50.83 (1.34) | 50.92 (1.40) | 0.80 | t |
| Tolerance test | 33.33 (23.28) | 31.15 (21.25) | 0.83 | KW |
| Temporal pain summation | 0.67 (0.84) | 1.56 (1.76) | 0.17 | t |
| Conditioned pain modulation | 5.58 (5.37) | 2.07 (3.03) | 0.10 | t |
| Difference between pre-epidural leg and pre-epidural arm | ||||
| Cold sensation threshold | −0.80 (2.71) | −2.51 (3.74) | 0.27 | KW |
| Warm sensation threshold | 2.68 (2.92) | 5.67 (3.22) | 0.04* | KW |
| Cold pain threshold | −0.03 (10.02) | −3.23 (2.83) | 0.37 | KW |
| Heat pain threshold | 0.49 (3.95) | 3.20 (2.47) | 0.09 | KW |
| Cold pain tolerance threshold | −0.34 (5.63) | 0.76 (3.53) | 0.62 | KW |
| Heat pain tolerance threshold | 0.29 (1.12) | 1.15 (1.16) | 0.10 | t |
| Tolerance test | −3.26 (19.87) | −4.61 (26.80) | 0.90 | KW |
| Temporal pain summation | −0.54 (0.53) | −0.62 (1.68) | 0.89 | t |
| Conditioned pain modulation | −0.44 (5.01) | −1.93 (1.73) | 0.41 | KW |
| Postepidural leg | ||||
| Cold sensation threshold | 27.40 (5.40) | 26.69 (3.54) | 0.73 | t |
| Warm sensation threshold | 36.91 (4.26) | 38.30 (4.21) | 0.47 | t |
| Cold pain threshold | 15.62 (17.19) | 3.05 (7.33) | 0.06 | KW |
| Heat pain threshold | 48.00 (1.91) | 47.48 (3.33) | 0.66 | t |
| Cold pain tolerance threshold | 0.84 (4.36) | 0.10 (0.22) | 0.53 | KW |
| Heat pain tolerance threshold | 51.15 (2.08) | 50.78 (1.74) | 0.67 | t |
| Tolerance test | 25.92.33 (44.08) | 32.36 (27.11) | 0.70 | KW |
| Temporal pain summation | 0.92 (1.67) | 0.96 (1.35) | 0.90 | KW |
| Conditioned pain modulation | 8.27 (8.97) | 1.88 (2.40) | 0.07 | KW |
| Difference between postepidural leg and postepidural arm | ||||
| Cold sensation threshold | −0.86 (2.79) | −0.63 (5.39) | 0.90 | KW |
| Warm sensation threshold | 2.30 (2.21) | 3.93 (3.40) | 0.21 | KW |
| Cold pain threshold | −0.60 (4.10) | −7.35 (9.11) | 0.04* | KW |
| Heat pain threshold | 2.07 (2.58) | 3.56 (4.23) | 0.34 | KW |
| Cold pain tolerance threshold | −2.24 (4.88) | −0.32 (0.61) | 0.27 | KW |
| Heat pain tolerance threshold | 0.95 (1.52) | 1.60 (1.44) | 0.34 | KW |
| Tolerance test | −8.63 (16.56) | −9.12 (16.84) | 0.94 | KW |
| Temporal pain summation | −0.47 (0.43) | −0.66 (1.06) | 0.60 | t |
| Conditioned pain modulation | −0.53 (2.01) | −4.08 (4.67) | 0.04* | KW |
Corrected P value for “difference between pre epidural leg and pre epidural arm” and “difference between post epidural leg and post epidural arm” is 0.0125. KW = Kruskal-Wallace test.
Heat/Cold Pain Threshold
A difference between the cold pain threshold of the postepidural unaffected upper extremity subtracted from the affected lower extremity was noted (−0.60 °C, SD = 4.10, vs −7.35 °C, SD = 9.11, P = 0.04). No other differences were observed. This did not meet criteria for significance with the P values adjusted for multiple comparisons (adjusted P = 0.0125).
A difference between the two groups was noted for the heat pain threshold in the pre-epidural affected lower extremity value. Those who had less than 30% reduction in pain had an increased heat pain threshold (46.54 °C, SD = 1.44, vs 48.83 °C, SD = 2.10, P = 0.01). No other differences were noted.
Heat/Cold Pain Tolerance
No differences in lowest cold temperature tolerated or highest heat pain temperature were observed between those who received at least a 30% reduction in pain in response to an epidural and those who did not both pre-epidural and postepidural.
Temporal Summation
No differences in the measured temporal summation average measurements were observed between those who received at least a 30% reduction in pain in response to an epidural and those who did not both pre-epidural and postepidural.
Conditioned Pain Modulation
A difference between the condition pain modulation threshold of the postepidural unaffected upper extremity subtracted from the affected lower extremity was noted (−0.53 °C, SD = 2.01, vs −4.08 °C, SD = 4.67, P = 0.04). This did not meet criteria for significance with the P values adjusted for multiple comparisons (adjusted P = 0.0125). No other differences in the measured conditioned pain modulation average measurements were observed between those who received at least a 30% reduction in pain in response to an ESI and those who did not both pre-epidural and postepidural.
Logistic Regression Analysis
A logistic regression analysis was performed to analyze factors associated with a less than 30% reduction in radicular pain, which are presented in Table 3. As expected, a higher absolute postepidural pain score was associated with having a less than 30% reduction in pain (odds ratio = 18.46, 95% CI = 1.99–2,225, P = 0.06). Gender, alcohol use, opioid use, and BMI were not found to be statistically associated with a 30% decrease in radicular pain.
Table 3.
Logistic regression analysis for risk of less than 30% reduction in pain
| Variables | Logistic regression |
|
|---|---|---|
| OR (95% CI) | P | |
| Gender | ||
| Male, ref. female | 1.14 (0.007,415.6) | 0.95 |
| Alcohol use | ||
| Yes, ref. no | 0.0008 (0.0001,19.3) | 0.28 |
| Opioid use | ||
| Yes, ref. no | 0.004 (0.0001,16.2) | 0.38 |
| BMI | 0.59 (0.13,1.54) | 0.33 |
| Postepidural pain | 18.46* (1.99,2225) | 0.06 |
P < 0.1. CI = confidence interval; OR = odds ratio.
Discussion
The primary finding of this prospective observational study is that subjects who experience less than 30% reduction in pain in response to an ESI at four weeks will have alterations in their initially measured QST phenotype with regards to warm sensation and heat pain threshold. Specifically, a larger difference is noted between the subject’s warm sensation threshold in the affected lower extremity and the unaffected upper extremity in subjects who failed to respond to an ESI with at least a 30% reduction in their pain. Additionally, an increased heat pain threshold is found in the affected lower extremity of those who report a less than 30% reduction in pain in response to an ESI. These findings suggest that a dysfunction in C fibers as manifested by the increased threshold to detect warm sensation and heat pain may be related to the response to ESI. This could implicate preservation of C fibers as a requisite factor in order to achieve therapeutic benefit with an ESI. The data does not suggest that the function of nerve fibers that conduct cold sensation, such as Aδ fibers, are related to the degree of pain relief experienced by an ESI.
Thermal QST measurements have been used to explore the functional changes that occur with certain pain states, including chronic radicular pain. Radicular pain has a thermal QST phenotype distinct from other sources of pain, such as fibromyalgia [14]. Increases in heat, cold, and mechanical thresholds have been reported in the affected dermatomes of subjects with radicular pain, indicating dysfunction of C, Aδ and Aβ, fibers, respectively [6]. Other studies of radicular pain have noted changes in both cold and warm sensation, indicating both Aδ and C fiber dysfunction [11,12,15]. Congruent with the presented data is evidence from surgical subjects indicating that an increase in warm sensation threshold was associated with a decreased response to surgical decompression [8]. Cold sensation was not altered.
There are several observations in the presented data. First, the data primarily demonstrated alterations in warmth and heat pain sensations. Prior histologic evaluation has demonstrated that chronic mechanical compression with features of radicular pain results initially in loss of large Aβ fibers followed by Aδ fibers [16]. However, lumbar pain with radicular symptoms can be caused by both compression and inflammation of lumbar nerve roots [1]. Herniation of the nucleus pulposis from intervertebral discs results in localized inflammation decreased axonal signal transduction [17]. Application of exogenous inflammatory mediators such as interleukin 1 (IL-1), interferon gamma (IFN-γ), and tumor necrosis factor alpha (TNF-α) produces a similar decrease in neural conduction [18]. In contrast to mechanical compression, unmyelinated C fibers have been demonstrated to be the most susceptible to chemical and inflammatory insults [7,19]. Steroid application to nerve roots has been shown to decrease painful ectopic transmission in C fibers, while the effect on Aβ and Aδ fibers has not been fully characterized [20]. The encouraging response to anti-inflammatory steroids provides indirect suggestion of an underlying inflammatory pathology. The exact pathophysiology was not investigated in this study. In a larger future study, it would be useful to elucidate how QST may be used to delineate pain pathologies and prognosticate the therapeutic response to an ESI.
The current investigation has several limitations. The lack of a placebo group makes changes in both subjective NRS-11 pain scores and thermal QST measurements difficult to attribute to ESI therapy or natural resolution during the four-week interval between the ESI and follow-up visit. The QST measurements that were undertaken were limited to thermal testing. Other forms of QST, such as vibratory testing, were not evaluated in this study. The use of single-dimension subjective measures for accurate and reproducible measurement of pain is valid in some situations, but could potentially have confounding shortcomings, such as a lack of ratio qualities, reliance on individuals’ subjective pain experiences, and overreporting of pain and derived pain relief [21–23]. This study attempted to minimize this by utilizing multiple measures to quantify alterations in pain processing and sensation in discreet fibers.
In keeping with the recommendations of the International Association for the Study of Pain (IASP), subjects of both genders were included in this study [24]; only 65% of the study subjects were female, but this population made up 77.8% of those who experienced a clinical benefit from an epidural compared with 54.5% of those that did not. Differences in the perception of pain interpretation among the genders may partially contribute to the observed results. However, prior work has found that both genders respond equally to epidural steroid therapy [25,26]. A logistic regression analysis (Table 3) failed to demonstrate an association between gender and response to an ESI. Additionally, subjects who experienced a reduction in pain in response to epidural steroid therapy weighed less (155.25 pounds vs 180.91 pounds) and had lower BMIs (24.92 vs 28.66) but had similar heights. Data from other studies suggest that a subject’s BMI has also not been associated with altered outcomes following epidural steroid injections [27]. The utilization of opioids has been associated with sensitization to pain through opioid-induced hyperalgesia (OIH) [28]. Subjects in this study were not formally evaluated for evidence of OIH. However, pre-injection use of opioid therapy has also not been associated with altered outcomes following epidural steroid injections [25]. In keeping with previously published reports, the logistic regression model presented in Table 3 again failed to demonstrate a relationship between relief derived from an ESI and either BMI or chronic opioid use.
In conclusion, the presented data add to the existing body of literature regarding thermal QST measurement as a means of characterizing low back pain with radiculopathy and prognosticating the response to therapeutic treatment with epidural steroid injections. Further investigation may better define the role of QST in the prognostication for the effectiveness of lumbar epidural steroid injections to treat radicular pain.
References
- 1. Goupille P, Jayson MI, Valat JP, Freemont AJ.. The role of inflammation in disk herniation-associated radiculopathy. Semin Arthritis Rheum 1998;281:60–71. [DOI] [PubMed] [Google Scholar]
- 2. Cuellar JM, Montesano PX, Antognini JF, Carstens E.. Application of nucleus pulposus to L5 dorsal root ganglion in rats enhances nociceptive dorsal horn neuronal windup. J Neurophysiol 2005;941:35–48. [DOI] [PubMed] [Google Scholar]
- 3. Friedly JL, Comstock BA, Turner JA, et al. A randomized trial of epidural glucocorticoid injections for spinal stenosis. N Engl J Med 2014;3711:11–21. [DOI] [PubMed] [Google Scholar]
- 4. Cohen SP, Hayek S, Semenov Y, et al. Epidural steroid injections, conservative treatment, or combination treatment for cervical radicular pain: A multicenter, randomized, comparative-effectiveness study. Anesthesiology 2014;1215:1045–55. [DOI] [PubMed] [Google Scholar]
- 5. Yoshizawa H, Kobayashi S, Morita T.. Chronic nerve root compression. Pathophysiologic mechanism of nerve root dysfunction. Spine 1995;204:397–407. [DOI] [PubMed] [Google Scholar]
- 6. Nygaard OP, Mellgren SI.. The function of sensory nerve fibers in lumbar radiculopathy. Use of quantitative sensory testing in the exploration of different populations of nerve fibers and dermatomes. Spine 1998;233:348–52. [DOI] [PubMed] [Google Scholar]
- 7. Zwart JA, Sand T, Unsgaard G.. Warm and cold sensory thresholds in patients with unilateral sciatica: C fibers are more severely affected than A-delta fibers. Acta Neurol Scand 1998;971:41–5. [DOI] [PubMed] [Google Scholar]
- 8. Nygaard OP, Kloster R, Solberg T, Mellgren SI.. Recovery of function in adjacent nerve roots after surgery for lumbar disc herniation: Use of quantitative sensory testing in the exploration of different populations of nerve fibers. J Spinal Disord 2000;135:427–31. [DOI] [PubMed] [Google Scholar]
- 9. Moloney N, Hall T, Doody C.. Sensory hyperalgesia is characteristic of nonspecific arm pain: A comparison with cervical radiculopathy and pain-free controls. Clin J Pain 2013;2911:948–56. [DOI] [PubMed] [Google Scholar]
- 10. Tampin B, Slater H, Hall T, Lee G, Briffa NK.. Quantitative sensory testing somatosensory profiles in patients with cervical radiculopathy are distinct from those in patients with nonspecific neck-arm pain. Pain 2012;15312:2403–14. [DOI] [PubMed] [Google Scholar]
- 11. Mosek A, Yarnitsky D, Korczyn AD, Niv D.. The assessment of radiating low back pain by thermal sensory testing. Eur J Pain (London, England) 2001;54:347–51. [DOI] [PubMed] [Google Scholar]
- 12. Schiff E, Eisenberg E.. Can quantitative sensory testing predict the outcome of epidural steroid injections in sciatica? A preliminary study. Anesth Analg 2003;973:828–32. [DOI] [PubMed] [Google Scholar]
- 13. Hafner J, Lee G, Joester J, et al. Thermal quantitative sensory testing: A study of 101 control subjects. J Clin Neurosci 2015;223:588–91. [DOI] [PubMed] [Google Scholar]
- 14. Blumenstiel K, Gerhardt A, Rolke R, et al. Quantitative sensory testing profiles in chronic back pain are distinct from those in fibromyalgia. Clin J Pain 2011;278:682–90. [DOI] [PubMed] [Google Scholar]
- 15. Zwart JA, Sand T.. Repeatability of dermatomal warm and cold sensory thresholds in patients with sciatica. Eur Spine J 2002;115:441–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Rempel D, Dahlin L, Lundborg G.. Pathophysiology of nerve compression syndromes: Response of peripheral nerves to loading. J Bone Joint Surg 1999;8111:1600–10. [DOI] [PubMed] [Google Scholar]
- 17. Olmarker K, Rydevik B, Nordborg C.. Autologous nucleus pulposus induces neurophysiologic and histologic changes in porcine cauda equina nerve roots. Spine 1993;1811:1425–32. [PubMed] [Google Scholar]
- 18. Aoki Y, Rydevik B, Kikuchi S, Olmarker K.. Local application of disc-related cytokines on spinal nerve roots. Spine 2002;2715:1614–7. [DOI] [PubMed] [Google Scholar]
- 19. Franz DN, Perry RS.. Mechanisms for differential block among single myelinated and non-myelinated axons by procaine. J Physiol 1974;2361:193–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Johansson A, Hao J, Sjolund B.. Local corticosteroid application blocks transmission in normal nociceptive C-fibres. Acta Anaesthesiol Scand 1990;345:335–8. [DOI] [PubMed] [Google Scholar]
- 21. Feine JS, Lavigne GJ, Dao TT, Morin C, Lund JP.. Memories of chronic pain and perceptions of relief. Pain 1998;772:137–41. [DOI] [PubMed] [Google Scholar]
- 22. Dannecker EA, George SZ, Robinson ME.. Influence and stability of pain scale anchors for an investigation of cold pressor pain tolerance. J Pain 2007;86:476–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Price DD, Bush FM, Long S, Harkins SW.. A comparison of pain measurement characteristics of mechanical visual analogue and simple numerical rating scales. Pain 1994;562:217–26. [DOI] [PubMed] [Google Scholar]
- 24. Greenspan JD, Craft RM, LeResche L, et al. Studying sex and gender differences in pain and analgesia: A consensus report. Pain 2007;132(suppl 1):S26–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Sivaganesan A, Chotai S, Parker SL, et al. Predictors of the efficacy of epidural steroid injections for structural lumbar degenerative pathology. Spine J 2016;68:928–34. [DOI] [PubMed] [Google Scholar]
- 26. Inman SL, Faut-Callahan M, Swanson BA, Fillingim RB.. Sex differences in responses to epidural steroid injection for low back pain. J Pain 2004;58:450–7. [DOI] [PubMed] [Google Scholar]
- 27. McCormick Z, Plastaras C.. Lumbosacral transforaminal epidural steroid injections are equally effective for treatment of lumbosacral radicular pain in the obese compared to non-obese population: A pilot study. J Back Musculoskel Rehabil 2013;262:183–8. [DOI] [PubMed] [Google Scholar]
- 28. Mao J. Clinical diagnosis of opioid-induced hyperalgesia. Reg Anesth Pain Med 2015;40:663–4. [DOI] [PubMed] [Google Scholar]