Electroacupuncture (EA) is a complementary therapy to improve morphine analgesia for postoperative pain, but underlying mechanism is not well-known. Herein, we investigated EA-induced analgesic effect in a plantar incision (PI) model in male Sprague-Dawley rats. PI was performed at the left hind paw. EA of 4 Hz and high intensity or sham needling was conducted at right ST36 prior to PI and repeated for another 2 days. Behavioral responses to mechanical and thermal stimuli, spinal phospho-ERK, and Fos expression were all analyzed. In additional groups, naloxone and morphine were administered to elucidate involvement of opioid receptors and for comparison with EA. EA pretreatment significantly reduced post-PI tactile allodynia for over 1 day; repeated treatments maintained analgesic effect. Intraperitoneal naloxone could reverse EA analgesia. Low-dose subcutaneous morphine (1 mg/kg) had stronger inhibitory effect on PI-induced allodynia than EA for 1 h. However, analgesic tolerance appeared after repeated morphine injections. Both EA and morphine could equally inhibit PI-induced p-ERK and Fos inductions. We conclude that though EA and morphine attenuate postincision pain through opioid receptor activations, daily EA treatments result in analgesic accumulation whereas daily morphine injections develop analgesic tolerance. Discrepant pathways and mechanisms underlying two analgesic means may account for the results.
Surgery is a necessary evil which can be associated with complications such as unbearable pain. Poor postoperative pain control prolongs hospitalization days, increases perioperative morbidity, and causes chronic postoperative pain [
Numerous studies demonstrated with convincible evidence that EA acts as a complementary treatment technique for various surgeries and invasive procedures (dental extraction, colonoscopy, and bronchoscopy) to reduce pain and related symptoms such as nausea, vomiting, and dizziness with or without PCA [
To investigate mechanistic processes of postsurgical pain, Brennan et al. [
Both Fos expression and phosphorylation of extracellular signal-regulated kinase (ERK), a member of mitogen-activated protein kinase (MAPK), are induced in the spinal dorsal horns by peripheral nerve injury, inflammation, and paw incision and have been viewed as molecular evidence of neuronal excitation [
Male Sprague-Dawley rats (250–300 gm; BioLASCO, Taipei, Taiwan) were used. Animals were housed in groups of two to three per cage at constant
The electrical stimulation was modified by lab protocol [
The plantar surgery had been reported before [
To test tactile allodynia, a series of von Frey filaments with incremental stiffness (0.4, 1, 2.0, 4.0, 6.0, 10.0, 15.0, and 26.0 g) (Stoelting, Wood Dale, IL) was used. Animals were individually acclimated in chamber (10 × 10 × 20 cm) of plexiglas cage on an elevated iron mesh floor 20 min before testing. The filaments, starting from the 4.0 g filament, were perpendicularly applied from underneath the mesh openings to stimulate the plantar surface at medial aspect adjacent to the wound for 5-6 seconds for each filament. Stimulation was conducted in an up-down method [
To test thermal hyperalgesia, animals were put in a plastic box placed on a glass plate prewarmed to constant 30°C (Plantar Test Apparatus, IITC, CA). The left plantar surface was exposed to a beam of radiant heat underneath the glass floor. The heat was adjusted to produce baseline latencies of 10–12 sec and a cut-off limit of 25 sec to prevent potential heat injury. Every withdrawal latency was an average of three tests, separated by a 5 min interval [
Animals were overanesthetized with high-dose isoflurane and then transcardially perfused with saline at room temperature, followed by 4°C 4% paraformaldehyde in 0.1 M phosphate buffer (PB). The L4-5 spinal segments were carefully removed, postfixed overnight, and cryoprotected in 4°C 30% sucrose/PB for another 24–48 h. Before slicing, a hole was made at right spinal ventral horn by a fine needle as side marker. The tissues were cut by a cryostat and the free-floating sections (30
Images of p-ERK-immunoreactive (p-ERK-ir) and Fos-immunoreactive (Fos-ir) cells were captured at a magnification of 20x under Nikon E600 (Tokyo, Japan) microscope in all cases. The immunoreactive cells in the dorsal laminae (I-V) were counted at randomly chosen sections and averaged from at least 6 sections for each lumbar segment. At least 4 rats in each group were included. The investigator who measured the staining was blind to group allocation.
The study designs are plotted in Figures
Effect of repeated daily EA treatments on PI-induced pain. (a) Summary of the protocols used in this experiment. BL: baseline on one day before PI; D: postplantar incision day; EA: electroacupuncture; h: hour; PI: plantar incision, marked by a solid triangle; pre: before daily EA treatment. (b) Mechanical allodynia, (c) heat hyperalgesia. Naïve: group without PI surgery or EA treatment, EA: group with repeated EA treatments, EA+PI: group for PI with repeated EA treatments, and Sham+PI: group for PI with repeated sham needle insertions; # < 0.05, ## < 0.01, ### < 0.001 groups versus Naïve; ∗ < 0.05, ∗∗ < 0.01, ∗∗∗ < 0.001 EA+PI versus Sham+PI; one-way ANOVA with Tukey’s multiple comparison test;
Intraperitoneal injections of naloxone (Nal) reversed EA analgesic effect on postincisional pain. (a) Summary of the protocols used in this experiment. Naloxone was injected at doses of 2 mg/kg immediately before anesthesia and 1 mg/kg immediately after PI and repeated the administration mode on D1 and D2. BL: baseline on one day before PI; D: postplantar incision day; EA: electroacupuncture; h: hour; i.p.: intraperitoneal injection; Nal: naloxone, marked with an arrow line; PI: plantar incision, marked with a solid triangle; pre: before daily Nal i.p. and EA treatment. (b) Intraperitoneal injections of naloxone almost completely antagonized EA analgesia to basal post-PI pain levels. Nal: group with i.p. naloxone, Saline+EA+PI: group for PI with repeated saline injections and EA treatments, Nal+Sham+PI: group for PI with repeated naloxone injections and sham needle insertion, and Nal+EA+PI: group for PI with repeated naloxone injections and EA treatments; ### < 0.001 groups versus Nal; & < 0.05 Nal+EA+PI versus Nal+Sham+PI; ∗ < 0.05, ∗∗ < 0.01, ∗∗∗ < 0.001 Saline+EA+PI versus Nal+EA+PI; one-way ANOVA with Tukey’s multiple comparison test;
Comparison of analgesic patterns between EA and subcutaneous injection of morphine in incisional pain. (a) Summary of the protocols used in this experiment. BL: baseline on one day before PI; D: postplantar incision day; EA: electroacupuncture; h: hour; s.c.: subcutaneous injection; Mor: morphine, marked with a blank triangle; PI: plantar incision, marked with a solid triangle; pre: before daily Mor i.p. and EA treatment. (b) Mechanical allodynia, PI: group for PI surgery, EA+PI: group for PI with repeated EA treatments, and Mor+PI: group for PI with repeated morphine s.c. injections; # < 0.05, ## < 0.01, ### < 0.001 groups versus PI; ∗∗ < 0.01, ∗∗∗ < 0.001 Mor+PI versus EA+PI; one-way ANOVA with Tukey’s multiple comparison test.
Second, opioid receptor-mediated effect was investigated by intraperitoneal (i.p.) naloxone (Nal) injection. Naloxone was injected at doses of 2 mg/kg immediately before anesthesia and 1 mg/kg immediately after PI and repeated the administration mode on D1 and D2. The dose is based on previous studies [
Last, we compared analgesic patterns between EA and morphine injection. Rats in the Mor+PI group received subcutaneous (s.c.) injection of 1 mg/kg morphine (Mor) and sham needling at 30 min before PI and at post-PI D1 and D2. The EA+PI group received EA stimulation and s.c. saline injection of the identical volume. We injected morphine at subcutaneous tissue over right thigh because the rat was placed in a restrained tube for anesthesia and i.p. technique became difficult and unsecure.
Some rats in the EA treatment study, at least 4 in a group, were sacrificed for immunohistochemical analysis of phosphorylated ERK (p-ERK) and Fos. The inductions of p-ERK and Fos were analyzed in samples from L4-5 spinal dorsal horn at 30 min and 3 h, respectively, after PI.
All study protocols were standardized, and baselines of mechanical and thermal thresholds were measured from at least 2 days before surgery to eliminate the hyper- or hyposensitive rats. There were at least 6 rats at each group for behavioral tests. Because our anesthetic apparatus can anesthetize maximally three rats at a time, we always included one sham and one EA rat at each test to minimize background biases.
All the results were expressed as mean ± SEM (standard error of the mean). Two-way analysis of variance (ANOVA) was conducted to analyze the influence of factors of time and treatments. Data from behavioral tests and the mean numbers of immunoreactive cells were analyzed by one-way ANOVA with post hoc Tukey’s test (PASW Statistics for Windows, Version 18.0. Chicago: SPSS Inc).
In consistence with previous studies and our studies [
EA stimulation significantly prevented and attenuated postoperative aversive responses. After EA pretreatment, the EA+PI group showed higher mechanical thresholds than the Sham+PI group at the post-PI 1 h (EA+PI versus Sham+PI:
Injection of high-dose naloxone (total 3 mg/kg, i.p.) did not alter normal withdrawal thresholds in the naive rats (the Nal group, Figure
Based on the above findings, we further compared effects of EA and morphine in PI model (Figures
The naive rats without PI or EA presented very few ERK activation or Fos expression in the spinal dorsal horn (Figures
Effect of EA and morphine on ERK activation 30 minutes after PI in the lumbar (L4 or L5) spinal dorsal horns. (a)–(d) are section slices showing spinal dorsal horn at the PI-ipsilateral side in the Naïve group ((a) rats without any operation or treatment), Sham+PI group ((b) rats with PI and sham needling and saline injection), EA+PI group ((c) rats with PI and pretreated EA), and Mor+PI group ((d) rats with PI and preinjected morphine). (e) Numbers of pERK-immunoreactive cells by lamina in the spinal dorsal horn among groups. Note no differences between the EA+PI and Mor+PI groups. Laminas I-II: superficial dorsal horn, laminas III-IV: middle dorsal horn, and lamina V: deep dorsal horn. DH: dorsal horn; ∗∗∗ < 0.001 groups versus Sham+PI, one-way ANOVA with Tukey’s multiple comparison test;
Effect of EA and morphine on Fos expression 3 hr after PI in the lumbar (L4 or L5) spinal dorsal horns. (a)–(d) are section slices showing spinal dorsal horn at the PI-ipsilateral side in the Naïve group ((a) rats without any operation or treatment), Sham+PI group ((b) rats with PI and sham needling and saline injection), EA+PI group ((c) rats with PI and pretreated EA), and Mor+PI group ((d) rats with PI and preinjected morphine). (e) Numbers of Fos-immunoreactive cells by lamina in the spinal dorsal horn among groups. Note no differences between the EA+PI and Mor+PI groups. Laminas I-II: superficial dorsal horn, laminas III-IV: middle dorsal horn, and lamina V: deep dorsal horn. DH: dorsal horn; ∗ < 0.05, ∗∗∗ < 0.001 groups versus Sham+PI, one-way ANOVA with Tukey’s multiple comparison test;
In Figure
We present in the study that EA stimulation is efficacious in reducing incision-induced tactile allodynia and heat hyperalgesia, as well as suppressing nociception-activated ERK phosphorylation and Fos expression in the spinal cord. Daily EA can maintain the analgesic effect. Most importantly, repeated EA does not show analgesic tolerance, as observed in the morphine treatment group.
The first study of EA effect in rat PI model was done by Oliveira and Prado [
The EA intensity we used is relatively strong but not intolerable. It is higher than those in most awake animal EA studies [
High dose of i.p. naloxone antagonizes EA-produced long-lasting analgesia. EA effect in this study cannot be a purely homosegmental inhibition (such as Gate control) in that EA stimulated at the opposite side of paw incision. It is likely that intense EA at the right hind limb sends information ascendingly to trigger the supraspinal structures (e.g., brainstem, midbrain, and cortex [
The reduction of Fos and p-ERK expression in the spinal dorsal horn supports spinal inhibitory mechanism exerted by EA and morphine treatments. We examined two markers at two respective time points because spinal ERK phosphorylation occurs early after peripheral injury (within tens of min) and Fos protein is an end product of immediately early gene c-
Based on these deductions, it is rational that EA-triggered opioid action can gradually increase and persist by activating complex spinal and supraspinal mechanisms and accumulating adequate endogenous opioids, compared to the immediate and short effect of injected morphine which passes absorption, circulation to reach spinal cord and brain targets, and metabolism. We injected 1 mg/kg morphine because we had demonstrated that morphine at 1 mg/kg is equipotent to EA of 10× basal intensity [
In this study, the first morphine injection produced strong and lasting antiallodynic potency, the second injection exerted strong but short effect, and then the third injection had only low and short effect. It is undoubted that the morphine group developed analgesic tolerance to injections of the same morphine doses, but not analgesic accumulation as the EA group did. Though EA tolerance could happen after chronic daily EA stimulations [
Growing evidence supports EA efficacy is stronger in pathological conditions than in a normal control. We found EA produced prolonged analgesia, much longer than duration of tail flick inhibition in the naïve rats (about 90–120 min) using the same EA setting [
In conclusion, intense EA stimulation suppressed incision-induced pain in a rat surgical pain model via an opioid-dependent analgesic effect. Particularly, repeated EA did not show analgesic tolerance as daily morphine administrations. Inhibition of Fos expression and ERK activity in the spinal dorsal horn implicates an important role of spinal mechanisms in EA analgesia. This preclinical study opens an alternative view on EA mechanisms.
The authors declare that there is no conflict of interests regarding the publication of this paper.
Sheng-Feng Hsu and Yeong-Ray Wen contributed equally to this study as the corresponding authors.
This study was sponsored by research grants from the National Science Council in Taiwan (NSC97-2314-B-039-044-MY3, NSC99-262S-B-039-00S-MY2, and NSC101-2314-B-039-005-MY3), in part by Taiwan Ministry of Health and Welfare Clinical Trial and Research Center of Excellence (DOH102-TD-B-111-004), by China Medical University under the “Aim for Top University Plan of the Ministry of Education, Taiwan” to Yeong-Ray Wen, and by the Joint Research Grant of Shin-Kong Wu Ho-Su Memorial Hospital and Taipei Medical University (SKH-TMU-98-15) to Yeong-Ray Wen and Julia Yi-Ru Chen. The authors thank Miss Ya-Hsin Lou for her technical and official assistance.