Studies have shown that a certain amount of exercise can effectively reduce pain (1–3). Such a phenomenon is widely recognized as exercise-induced hypoalgesia (EIH), which is closely associated with the efficiency of endogenous pain inhibitory pathways (1). Currently, EIH has been frequently observed in aerobic exercise (4,5), dynamic circuit resistance exercise (6), and isometric exercise (7–9), which is often characterized by an elevation in pain thresholds and a reduction in pain ratings. For instance, the hypoalgesia responses to isometric exercise could be demonstrated as an increase in pressure pain thresholds (PPT) (1), a decrease in pressure pain ratings (PPR) (10,11), and an increase in heat pain thresholds (HPT) (12,13).
In practice, isometric exercise is a specific form of exercise involving the static contraction of a muscle with less changes of the joint angle, and it is effective in relieving pain symptoms (1). Studies showed that a low-intensity isometric exercise (20% maximum voluntary contraction [MVC] of muscles sustaining for 5 min) could increase the PPT in both healthy controls and myalgia patients (2). In addition, a relatively high-intensity isometric exercise (40% MVC of muscles lasting for 3 min) is more likely to increase the PPT and HPT than low-intensity exercise in different types of pain patients and healthy populations (1). Therefore, it is inferred that the intensity of exercise may have a strong influence on the modulation of pain perception.
Up to the present, the underlying neural mechanisms responsible for EIH have not been entirely understood. Electrophysiological and functional imaging studies have shown that the stimulation of the motor cortex can activate supraspinal regions involved in pain modulation, including the thalamus, the anterior cingulate cortex, the anterior insula, and the periaqueductal gray (14–16). Evidence from animal studies also indicated that the activation of the motor cortex during exercise leads to a relatively long-lasting inhibition of nociceptive system. In particular, the activation of the motor cortex in the primate could inhibit the excitability of the spinothalamic neurons (17,18). Similarly, stepwise-changed electrical stimuli to the motor cortex can significantly inhibit the response of spinal dorsal horn neurons to nociceptive stimuli in rats (19,20), via the inhibition of thalamic nuclei and activation of the periaqueductal gray (20,21). Therefore, it could be hypothesized that exercise may induce hypoalgesia by activating a top-down descending pain modulation pathway. In addition, a recent study using laser-evoked potentials to explore the mechanisms of EIH proposed that the hypoalgesic effect induced by exercise could be possibly related to the modulation of nociceptive afferents (i.e., the signal transmission via Aδ and C fibers) (12). However, the role of exercise in pain modulation is not totally clear. A more thorough investigation of the mechanisms of exercise in pain control is needed.
In this study, we explored the effects of isometric exercise with different intensities on the modulation of pain perception. We further investigated the neural mechanisms of EIH through tracing the dynamic EEG changes to nociceptive heat stimuli induced by the isometric exercise.
MATERIALS AND METHODS
Participants
A total of 48 participants (24 females; mean ± SD age, 23.5 ± 2.3 yr) were recruited from Shanghai University of Sport through advertisements. Inclusion criteria were as follows: 1) no history of alcohol, smoking, and intense physical activities 24 h before the experiment; 2) no pain symptoms, neurological diseases, cardiovascular disease, and any other type of concomitant medical illness; and 3) ages ranging from 18 to 30 yr old. Written informed consent was obtained before testing. All procedures were approved by the local ethics committee and were carried out in accordance with the approved guidelines. Before the experiment, participant’s characteristics were collected, including age, height, weight, body mass index, and years of education. Their physical conditions and willingness to express pain experience were assessed using the self-report scale of physical activity and the Gender Role Expectation of Pain Questionnaire—Willingness (22,23).
Submaximal isometric exercise
Participants were seated upright in an adjustable chair, with their elbow joints of the exercised limb flexed to 90° and rested on a padded support parallel to the floor. When participants voluntarily activated their elbow flexor muscles, they were required to keep their forearm displaced midway between pronation and supination. The upward-directed force of elbow flexion was recorded by a built-in mobile software in a force transducer (K-Force, Kinvent Biomecanique SAS, France). Participants’ hands were relaxed, and the force was applied via the transducer interfacing with the forearm. Before the experiment, participants were instructed to match a target force of 20% MVC or 40% MVC displayed on a monitor. MVC for each participant was tested three times, and the average value was obtained as the recorded MVC. The exercised sides of the elbow flexor muscle were randomly assigned for all participants across groups. The intensity of the isometric contraction was structured to produce a circuit exercise training consisting of four sets, with 10 repetitions in each set, and lasted for 15–20 min. Each repetition was set with a work-to-rest ratio of 3:1 and lasted for 20 s. The interval between two adjacent sets was 30 s.
Pressure pain thresholds and pressure pain ratings
PPT was assessed on the dorsal surface of the hand and biceps brachii muscles of the exercised limb. We used a handheld algometer (Wagner Force 10 FDX-25; Wagner Instruments, Riverside, CT) to perpendicularly deliver pressures to the participants’ skin, and the pressure force was increased at a rate of approximately 1 kg·s−1. The pressure was stopped when they felt the pressure became painful. This procedure was repeated three times for each test site. Each site was marked with a water-soluble pen to ensure consistent and reliable location during the experimental procedure (24). To minimize the impacts of repeated stimuli on the experimental outcomes, a recovery period of 20–30 s was set between each measurement (25,26). All the participants were further required to report their subjective feelings after each recovery to confirm their pain sensation from the previous stimulus had dissipated. The PPT was recorded as the average of these three measures.
PPR was assessed over the midway part of the index finger on the unexercised limb. For each participant, their index finger was placed under the algometer and was supported on the surface of solid and flat support. To maintain the force and fix the algometer vertically, two objects with a total weight of 1.5 kg were attached to the handle of the algometer. The rubber-tipped probe of the algometer exerted a constant vertical force of 1.5 kg on the index finger for 1 min. Simultaneously, all participants were asked to rate their pain ratings every 10 s using an 11-point numeric rating scale ranging from 0 to 10 (0 indicated no pain, 10 indicated extreme pain). The sum of the pain scores was obtained as PPR.
Heat pain thresholds
HPT was assessed on the dorsal surface of the hand and biceps brachii muscles of the exercised limb. Heat stimuli were delivered through a surface thermode (30 × 30 mm2) controlled by the Medoc sensory evaluation system (ATS Model; Pathway Ltd., Ramat Yishai, Israel). The heat temperature was increased at a rate of 1°C·s−1 starting from the baseline of 32°C. All participants were instructed to say “stop” as soon as the heat stimuli became painful. The highest temperature was set within a safe range to prevent burns (≤49°C). This procedure was repeated three times for each tested site. The location of the test site and the recovery period for HPT measurement were set in the same experimental procedure as that in the PPT measurement. The average value of the three measurements was recorded as HPT.
Heat stimuli for heat-evoked potentials
A contact heat-evoked potential (HEP) stimulator of the Medoc sensory evaluation system (CHEPS Model, Pathway Ltd.) was applied to release heat pulses in the experiment. Heat stimuli were delivered through a round thermode with a contacting cutaneous area of 572.5 mm2 (27 mm in diameter). The temperature of the heat stimuli was increased by 70°C·s−1 from the baseline of 32°C and was decreased at a rate of 40°C·s−1 from 51°C (51°C as the target temperature). Each heat pulse lasted for 432 ± 2 ms (rise time, ~157 ms; fall time, ~275 ms).
Experimental design
The experiment consisted of two phases (i.e., phase 1 and phase 2), with three sessions (i.e., preintervention, intervention, and postintervention) in each phase (Fig. 1). A 48-h washout period was set between phase 1 and phase 2 to ensure possible exercise-induced alterations in pain perception vanished before commencing the next phase. All participants would first engage in a training program to familiarize with the experimental procedures. At the start of the experiment, participants were randomly assigned to one of the three different groups: 20% MVC group (n = 16), 40% MVC group (n = 16), and control group (n = 16), to receive different interventions as follows: 1) for MVC groups, participants were required to do the submaximal isometric exercise for 15–20 min; 2) for the control group, participants were merely required to have a rest for 25 min. The PPT and the PPR assessments were conducted in phase 1, whereas the HPT assessment and the EEG recordings were conducted in phase 2. All these assessments and EEG recordings were obtained in an equivalent duration in both pre- and postintervention sessions of each phase. During the EEG recordings, a total of 20 heat stimuli (10 stimuli for the ipsilateral side and contralateral side, respectively) were delivered to the participant’s forearm 2–5 cm below the cubital crease, with a fixed stimulus sequence in the pre- and postintervention sessions. The target temperature of each heat stimulus was 51°C, with an interstimulus interval varied randomly between 20 and 25 s. Participants were required to report the perceived heat pain ratings (HPR) to each stimulus on a 11-point scale (0 = no pain, 10 = extremely unbearable pain). All participants were asked to count the number of stimuli delivered to their forearm during the whole experiment to ensure that their attention was focused on the heat stimuli.
;)
FIGURE 1: Experimental design. The whole experiment consisted of two phases (i.e., phase 1 and phase 2), with three sessions in each phase: preintervention (~10 min in phase 1; ~20 min in phase 2), intervention (~25 min), and postintervention (~10 min in phase 1; ~20 min in phase 2). Forty-eight participants were randomly assigned to three different groups (20% MVC group, 40% MVC group, and control group; 16 participants per group) for intervention. PPT and PPR assessments were conducted in phase 1, whereas HPT assessment and HPR/EEG recordings were conducted in phase 2. All these assessments and EEG recordings were obtained in both pre- and postintervention sessions of each phase. During EEG recordings, 20 heat stimuli were delivered to participants’ forearm, either exercised or unexercised limb (10 stimuli per limb).
EEG data recording
EEG data were documented using 64 Ag-AgCl scalp electrodes (Brain Products, Gilching, Germany; pass band, 0.01–100 Hz; sampling rate, 1000 Hz) placed according to the 10–20 International System. A ground electrode was placed on the forehead, whereas a reference electrode was placed over the parietal. To monitor eyeblinks and ocular movements, electrooculographic signals were simultaneously recorded using another electrode placed ~10 mm below the right eye. All electrode impedances were kept lower than 10 kΩ.
EEG data preprocessing
EEG data were preprocessed using EEGLAB, an open-source toolbox running in the MATLAB environment. Continuous EEG signals were band-pass filtered from 1 to 30 Hz. Epochs were defined using a time window of 1500 ms, ranging from −500 ms (prestimulus) to 1000 ms (poststimulus). Baseline correlation was applied using the prestimulus interval. The independent component analysis (ICA) algorithm was used to correct trials contaminated by eye blinks and movements. After preprocessing, EEG data were referenced to the average of right and left mastoid electrodes.
Heat-evoked potentials
For each participant, epochs belonging to the same stimulated side (i.e., exercised and unexercised sides) and sessions (i.e., pre and postinterventions) were averaged, yielding four average waveforms time locked to the stimulus onset. It should be noted that N2 and P2 waves in HEP are sensitive to pain stimuli, and their amplitudes can accurately reflect the brain activity related to pain information processing (27,28). Moreover, N2 and P2 waves are stable indicators for pain assessment with fewer confounding factors (27). Therefore, the peak latencies and amplitudes of N2 and P2 waves, as well as the peak-to-peak amplitudes of the N2–P2 complex, were measured at the central electrode of Cz from the averaged waveforms for each participant. Single-participant average waveforms of each stimulated side and session were averaged across participants to obtain group-level waveforms. Group-level scalp topographies at peak latencies of both N2 and P2 waves were computed by spline interpolation.
Time–frequency distributions (TFD) were estimated to extract both phase-locked and non–phase-locked brain responses elicited by the heat stimuli (29). TFD of EEG trials were calculated using a windowed Fourier transform with a fixed 200-ms Hanning window. In each trial, windowed Fourier transform yielded a complex time–frequency estimate F(t,f) at each time–frequency point (t,f) extending from −500 ms to 1000 ms (in steps of 1 ms) in the time domain and from 1 to 30 Hz (in steps of 1 Hz) in the frequency domain. The resulting spectrogram, P(t,f) = |F(t,f) |2, represented the signal power as a joint function of time and frequency at each time–frequency point. The spectrogram was baseline corrected with a reference interval ranging from −400 to −100 ms at each frequency using the subtraction approach (30). Based on previous observation (29,30), two regions of interests (ROI) were defined to extract the magnitude of time–frequency responses within the baseline-corrected TFD for each stimulated side and session (event-related potential [ERP] = 350–600 ms, 1–10 Hz, central midline electrodes; event-related desynchronization [ERD], 700–1000 ms, 8–12 Hz, bilateral parietal–occipital electrodes). Magnitudes of these time–frequency responses for each ROI were measured by computing the mean value of all time–frequency points within the given ROI for each participant, stimulated side, and session. Group-level scalp topographies of the magnitude of each time–frequency response were computed by spline interpolation.
Statistical analyses
One-way analysis of variance (ANOVA) and chi-square test were used to compare the group differences of participants’ characteristics and baseline pretest scores of pain perception. The possible effect of isometric exercise on both the subjective pain reports and the brain responses elicited by heat stimuli was evaluated by calculating the difference of each measure between sessions (postintervention minus preintervention) (12,31). The resulting differences of PPT, PPR, HPT, and HPR between groups were examined using one-way ANOVA. Post hoc independent-sample t-tests were conducted when significant main effects were found. The resulting differences of the peak latencies and amplitudes of HEP in the time domain (i.e., N2 and P2 waves), as well as the magnitudes of ERP and ERD in the time–frequency domain between groups, were examined using the two-way mixed-design ANOVA, with “group” (three levels: 20% MVC group, 40% MVC group, and control group) as a between-subject factor and “limb” (two levels: exercised limb and unexercised limb) as a within-subject factor. When significant effects were found, post hoc paired sample t-tests were performed to compare the changes of brain responses elicited by heat stimuli delivered to the forearm on the exercised and the unexercised limb. Bonferroni corrections were used for multiple comparisons. In addition, we used the Shapiro–Wilk statistic to assess the normality of the data and performed a normal transformation on the data when necessary. Greenhouse–Geisser corrections were used to adjust the result if sphericity was violated. To quantify the relationship between subjective pain reports and heat-evoked brain responses, Spearman correlation analyses were performed across the whole sample, and all P values were corrected with false discovery rate (32). All data analyses were conducted using SPSS 23 (IBM Corp., Armonk, NY).
RESULTS
Participant characteristics
There were no significant group differences in age, height, weight, BMI, and years of education (all P > 0.050; Table 1). In addition, no significant group differences were found among the scores of self-reported physical activities, Gender Role Expectation of Pain Questionnaire—Willingness, and MVC (all P > 0.050; Table 1).
TABLE 1 -
Participants’ characteristics.
|
20% MVC Group (n = 16) |
40% MVC Group (n = 16) |
Control Group (n = 16) |
Statistics |
P
|
| Age |
24.0 ± 2.4 |
23.3 ± 2.3 |
23.3 ± 2.3 |
F = 0.459 |
0.635 |
| Height (cm) |
166.8 ± 9.3 |
168.6 ± 8.1 |
168.0 ± 8.7 |
F = 0.169 |
0.845 |
| Weight (kg) |
62.2 ± 10.6 |
62.7 ± 9.2 |
59.7 ± 11.5 |
F = 0.377 |
0.688 |
| BMI (kg·m−2) |
22.2 ± 1.8 |
22.0 ± 2.1 |
21.0 ± 2.2 |
F = 1.662 |
0.201 |
| Education years |
16.6 ± 1.5 |
16.6 ± 1.1 |
16.9 ± 1.4 |
χ
2 = 1.838 |
0.399 |
| Self-report activity |
|
| Exercise times per week |
3.7 ± 1.3 |
3.1 ± 1.5 |
4.4 ± 2.6 |
χ
2 = 2.951 |
0.299 |
| Exercise duration (min) |
72.6 ± 44.1 |
62.0 ± 31.7 |
58.7 ± 48.3 |
χ
2 = 1.214 |
0.545 |
| Fatigue condition (0–5) |
3.1 ± 0.5 |
3.0 ± 0.7 |
3.1 ± 0.8 |
χ
2 = 0.415 |
0.813 |
| GREP—Willingness |
|
| Self to typical man |
55.0 ± 25.5 |
47.3 ± 22.9 |
52.8 ± 17.8 |
F = 0.509 |
0.605 |
| Self to typical woman |
48.7 ± 26.2 |
48.5 ± 23.9 |
45.8 ± 18.8 |
F = 0.076 |
0.927 |
| Typical man to typical woman |
38.2 ± 20.6 |
39.1 ± 26.0 |
45.6 ± 20.6 |
F = 0.503 |
0.608 |
| Typical woman to typical man |
67.3 ± 25.7 |
58.1 ± 17.6 |
62.7 ± 17.8 |
F = 0.786 |
0.462 |
| MVC (kg) |
22.8 ± 8.0 |
19.6 ± 7.3 |
– |
T = 1.182 |
0.247 |
Data are presented as mean ± SD. Fatigue condition (0–5): 0 indicate no fatigue, 5 indicate extreme fatigue.
BMI, body mass index; GREP, gender role expectation of pain questionnaire; MVC, maximal voluntary contraction.
Effects of isometric exercise on subjective pain perception
For the pretest scores of PPT, no significant group differences were found on the dorsal surface of the hand (F = 0.459, P = 0.635) and the biceps brachii muscle (F = 1.002, P = 0.375). For the pretest scores of PPR, there was no significant group difference (F = 3.143, P = 0.055). For the pretest scores of HPT, there were no significant group differences on the dorsal surface of the hand (F = 0.979, P = 0.384) and the biceps brachii muscle (F = 0.875, P = 0.424). For the pretest scores of HPR, there were no significant group differences on the exercised limb (F = 1.394, P = 0.258) and the unexercised limb (F = 1.510, P = 0.232).
The isometric exercise induced consistent changes in the pressure pain (i.e., PPT and PPR) and heat pain (i.e., HPT and HPR; Fig. 2). For PPT, significant group differences induced by the isometric exercise were observed on both the dorsal surface of the hand (F(2,45) = 16.002, P < 0.001, η2 = 0.416; Fig. 2 top panel) and the biceps brachii muscle (F(2,45) = 20.038, P < 0.001, η2 = 0.471; Fig. 2 top panel). Post hoc pairwise comparisons revealed that 1) both the isometric exercise with 20% MVC (P = 0.020) and the isometric exercise with 40% MVC (P < 0.001) induced a larger increase of PPT on the dorsal surface of the hand compared with that in the control group (20% MVC group, 0.156 ± 0.223 kg·cm−2; 40% MVC group, 0.262 ± 0.160 kg·cm−2; control group, −0.026 ± 0.054 kg·cm−2), whereas the changes of PPT on the dorsal surface of the hand in the two MVC groups were similar (P = 0.134); 2) both the isometric exercise with 20% MVC (P = 0.005) and the isometric exercise with 40% MVC (P < 0.001) induced a larger increase of PPT on the biceps brachii muscle compared with that in the control group (20% MVC group, 0.136 ± 0.128 kg·cm−2; 40% MVC group, 0.350 ± 0.261 kg·cm−2; control group, −0.006 ± 0.091 kg·cm−2), and the increase of PPT on the biceps brachii muscle induced by the isometric exercise with 40% MVC was larger than that induced by the isometric exercise with 20% MVC (P = 0.015).
FIGURE 2: Effects of isometric exercise on the pressure pain and heat pain. The effects of isometric exercise on the pressure pain (i.e., PPT and PPR) and heat pain (i.e., HPT and HPR) are evaluated as the differences between preintervention and postintervention sessions. Both the two different intensities of isometric exercise (i.e., 20% MVC and 40% MVC) induced a larger increase of PPT and HPT, with a larger decrease of PPR and HPR compared with those in the control group. Data are presented as mean ± SD.
Significant group differences of PPR induced by the isometric exercise were observed (F(2,37) = 4.030, P = 0.026, η2 = 0.179; Fig. 2 middle left panel). Post hoc pairwise comparisons revealed that the isometric exercise with 40% MVC (P = 0.033) induced a larger decrease of PPR compared with that in the control, but no significant decrease in 20% MVC group (P = 0.105) compared with control group (20% MVC group, −2.429 ± 4.033; 40% MVC group, −2.857 ± 2.656; control group, 0.231 ± 2.488), whereas the changes of PPR in the two MVC groups were similar (P > 0.050).
Significant group differences of HPT induced by the isometric exercise were observed on the dorsal surface of the hand (F(2,43) = 13.015, P < 0.001, η2 = 0.377; Fig. 2 bottom panel) and the biceps brachii muscle (F(2,45) = 5.300, P = 0.009, η2 = 0.198; Fig. 2 middle right panel and bottom left panel). Post hoc pairwise comparisons revealed that 1) the increase of HPT on the dorsal surface of the hand (20% MVC group, 0.693°C ± 0.954°C; 40% MVC group, 1.953°C ± 1.449°C; control group, 0.011°C ± 0.737°C) induced by the isometric exercise with 40% MVC was larger than those in the 20% MVC group (P = 0.009) and the control group (P < 0.001), and the changes of HPT on the dorsal surface of the hand in the 20% MVC group were similar with the control group (P = 0.149); 2) both the isometric exercise with 20% MVC (P = 0.021) and the isometric exercise with 40% MVC (P = 0.019) induced a larger increase of HPT on the biceps brachii muscle compared with that in the control group (20% MVC group, 0.966°C ± 1.654°C; 40% MVC group, 1.036°C ± 0.880°C; control group, 0.046°C ± 0.676°C), whereas the changes of HPT on the biceps brachii muscle in the two MVC groups were similar (P > 0.050).
Significant group differences of HPR induced by the isometric exercise were observed on the exercised limbs (F(2,30) = 6.969, P = 0.003, η2 = 0.317; Fig. 2 bottom right panel), but not on the unexercised limbs (F(2,30) = 0.238, P = 0.790, η2 = 0.016). Post hoc pairwise comparisons revealed that the changes of HPR on the exercised limbs induced by the isometric exercise with 40% MVC were larger than those in the control group (P = 0.002), but the changes of HPR on the exercised limbs in the 20% MVC group were similar to the 40% MVC group (P > 0.050) and control group (P = 0.053) (20% MVC group, −0.503 ± 0.430; 40% MVC group, −0.574 ± 0.453; control group, −0.016 ± 0.594).
Effects of isometric exercise on heat-evoked brain responses
Group-level HEP waveforms and scalp topographies of N2 and P2 waves in the time domain for each group are shown in Figure 3. In accordance with previous studies, N2 and P2 waves were centrally distributed (29,33). A strong evidence for a main effect of “limb” was observed on the N2 amplitudes elicited by the heat stimuli on the exercised limb (F(2,30) = 6.635, P = 0.004, η2 = 0.307), but not on that elicited by the heat stimuli on the unexercised limb (F(2,30) = 0.330, P = 0.721, η2 = 0.022; Fig. 3). In addition, a weak evidence for an interaction between “group” and “limb” (F(2,30) = 3.466, P = 0.044, η2 = 0.188) was observed on the N2 amplitudes. Post hoc pairwise comparisons showed that for the N2 amplitudes elicited by the heat stimuli on the exercised limb, 40% MVC group induced a larger decrease than that in the control group (P = 0.003), 20% MVC group induced a small decrease but not significant than that in the control group (P = 0.091), whereas the changes of N2 amplitudes in the two MVC groups were similar (P > 0.050) (20% MVC group, 3.947 ± 5.207 μV; 40% MVC group, 4.794 ± 4.325 μV; control group, −1.107 ± 4.224 μV).
FIGURE 3: Effects of isometric exercise on the heat-evoked brain responses in the time domain. Left panels: group-level waveforms and scalp topographies of N2 and P2 waves (Cz) are displayed for each group. In each group, HEP elicited by the heat stimuli at forearm on the exercised and unexercised limbs in the preintervention and postintervention sessions are superimposed. Scalp topographies are plotted at the peak latency of the N2 and P2 waves. Right panels: effects of isometric exercise on heat-evoked brain responses are evaluated as the differences between preintervention and postintervention sessions (postintervention minus preintervention). The decrease of almost all heat-evoked brain responses (except for P2 amplitudes) on the exercised limb was significantly larger in the two MVC groups than that in the control group. Data are presented as mean ± SD.
No significant main effect of “group” was observed on neither the P2 amplitudes elicited by the heat stimuli on the exercised limb (F(2,30) = 2.477, P = 0.101, η2 = 0.142) nor the P2 amplitudes elicited by the heat stimuli on the unexercised limb (F(2,30) = 0.839, P = 0.442, η2 = 0.053; Fig. 3). In addition, no significant interaction between “group” and “limb” (F(2,30) = 0.340, P = 0.715, η2 = 0.022) was found on the P2 amplitudes.
A strong evidence for a main effect of “group” was observed on the N2–P2 peak-to-peak amplitudes elicited by the heat stimuli on the exercised limb (F(2,30) = 7.938, P = 0.002, η2 = 0.346), but not on that elicited by the heat stimuli on the unexercised limb (F(2,30) = 0.178, P = 0.837, η2 = 0.012; Fig. 3). Post hoc pairwise comparisons showed that both the two different intensities of isometric exercise (20% MVC group, P = 0.015; 40% MVC group, P = 0.004) induced a larger decrease of N2–P2 peak-to-peak amplitudes than that in the control group (20% MVC group, −8.595 ± 6.816 μV; 40% MVC group, −8.505 ± 5.607 μV; control group, −0.486 ± 5.963 μV), whereas the changes of N2–P2 peak-to-peak amplitudes in the two MVC groups were similar (P > 0.050). No significant interaction between “group” and “limb” (F(2,30) = 2.073, P = 0.144, η2 = 0.121) was found on the N2–P2 peak-to-peak amplitudes.
Group-level TFD and scalp topographies of ERP and ERD responses are shown in Figure 4. A large phase-locked response (ERP, 350–600 ms, 1–10 Hz; central midline electrodes) and a non–phase-locked response (ERD, 700–1000 ms, 8–12 Hz; bilateral parietal–occipital electrodes) elicited by heat stimuli were clearly identified. A strong evidence for a main effect of “group” was observed on the ERP magnitude elicited by the heat stimuli on the exercised limb (F(2,30) = 3.381, P = 0.047, η2 = 0.184), but not on that elicited by the heat stimuli on the unexercised limb (F(2,30) = 0.790, P = 0.463, η2 = 0.050; Fig. 4). Post hoc pairwise comparisons showed that the isometric exercise with 20% MVC induced a larger decrease of the ERP magnitude (P = 0.035) than that in the control group, whereas the changes of the ERP magnitudes in the two MVC groups were similar (P = 0.457) (20% MVC group, −0.835 ± 0.925 μV2·Hz−1; 40% MVC group, −0.630 ± 0.855 μV2·Hz−1; control group, −0.171 ± 0.968 μV2·Hz−1). No significant interaction between “group” and “limb” (F(2,30) = 0.210, P = 0.812, η2 = 0.014) was observed on the ERP magnitudes.
FIGURE 4: Effects of isometric exercise on the heat-evoked brain responses in the time–frequency domain. Left panels: group-level TFD and scalp topographies of ERP and ERD responses, averaged across groups and sides. The color scale represents the increase or decrease of the oscillatory magnitude, relative to a prestimulus interval (−400 to −100 ms). The displayed TFD contain both phase-locked (ERP: 350–600 ms, 1–10 Hz, top left) and non–phase-locked brain responses (ERD, 700–1000 ms, 8–12 Hz, bottom left), highlighted by the dashed lines. ERP and ERD magnitudes were measured at central and parietal–occipital electrodes, respectively. Electrodes showing the maximal response for each time–frequency feature are highlighted in white in the scalp topographies. Right panels: the effects of isometric exercise on the magnitudes of the ERP (top right panel) and ERD responses (bottom right panel) were expressed as differences between preintervention and postintervention sessions (postintervention minus preintervention).
No significant main effect of “group” was observed neither on the ERD magnitude elicited by the heat stimuli on the exercised limb (F(2,30) = 0.312, P = 0.734, η2 = 0.020), nor on that elicited by the heat stimuli on the unexercised limb (F(2,30) = 1.353, P = 0.274, η2 = 0.083; Fig. 4). There was no significant interaction between “group” and “limb” (F(2,30) = 0.634, P = 0.537, η2 = 0.041) on the ERD magnitude.
Correlations between subjective pain perception and heat-evoked brain responses
Significant correlations were observed between 1) the changes of PPR at the index finger and the ERP magnitudes elicited by the heat stimuli on the exercised limb (ρ = 0.504, P = 0.005; Table 2); 2) the HPR of the forearms and the N2–P2 peak-to-peak amplitudes elicited by the heat stimuli on the unexercised limb (ρ = 0.249, P = 0.028; Table 2); 3) the HPR of the forearms and ERP magnitudes elicited by the heat stimuli on the unexercised limb (ρ = 0.300, P = 0.018; Table 2); and 4) the HPR of the forearms and ERD magnitudes elicited by the heat stimuli on the exercised limb (ρ = 0.220, P = 0.048; Table 2) and on the unexercised limb (ρ = 0.256, P = 0.028; Table 2).
TABLE 2 -
The correlations between subjective pain perception and heat-evoked brain responses.
|
N2–P2 Peak-to-Peak Amplitude (Exercised) |
N2–P2 Peak-to-Peak Amplitude (Unexercised) |
ERP Magnitude (Exercised) |
ERP Magnitude (Unexercised) |
ERD Magnitude (Exercised) |
ERD Magnitude (Unexercised) |
| PPT-dorsal surface of the hand |
−0.323 |
−0.247 |
−0.213 |
−0.232 |
0.132 |
−0.105 |
| PPT-biceps brachii |
−0.365 |
−0.236 |
−0.109 |
−0.206 |
0.160 |
−0.026 |
| PPR-index finger |
0.267 |
−0.043 |
0.504* |
−0.077 |
0.098 |
0.397 |
| HPT-dorsal surface of the hand |
−0.271 |
−0.008 |
−0.263 |
0.017 |
−0.081 |
−0.074 |
| HPT-biceps brachii |
−0.220 |
0.058 |
−0.208 |
0.066 |
0.100 |
−0.077 |
| HPR |
0.130 |
0.249* |
0.068 |
0.300* |
0.220* |
0.256* |
For HPR, the correlation were performed at the same limb side.
All values set in boldface are significant.
*P < 0.050, two-tailed.
DISCUSSION
We used a randomized controlled design to investigate the hypoalgesic effects of isometric exercise and its underlying neural mechanisms with EEG techniques in the current study. Three main findings were obtained. First, EIH not only could happen in the moving part of the body (local EIH) but also can extend to the distal part of the body (global EIH). This is revealed by the increased PPT and HPT on the dorsal surface of the hand and the biceps brachii muscle of the exercised limb (closed to the contracting muscle) and the decreased PPR at the indexed finger of the unexercised limb (remote to the contracting muscle; Fig. 2). These findings are consistent with previous studies (2,11,12,34,35). However, such global EIH vanished when measured with the HPR (i.e., nonsignificant changes of HPR on the unexercised limb; Fig. 2), indicating that EIH could vary depending on the pain modalities. Second, both the local EIH and the global EIH were modulated by the intensities of the isometric exercise, which is supported by the fact that more reductions of pain sensitivity on both the biceps brachii muscle and the dorsal surface of the hand were induced by the high-intensity (i.e., 40% MVC) isometric exercise than the low-intensity (i.e., 20% MVC) isometric exercise (Fig. 2). Third, the recording of heat-evoked brain responses provided physiological support to the modulation of subjective pain perception: Both the high-intensity and the low-intensity isometric exercise induced the reduction of N2 amplitudes and N2–P2 peak-to-peak amplitudes, as well as the reduction of ERP magnitudes elicited by the heat stimuli on the exercised limb (Fig. 3). Notably, the heat-evoked brain responses were significantly correlated with the subjective pain perception (i.e., the changes of PPT on the exercised limb and the HPR on the unexercised limb; Table 2), implying that the hypoalgesic effects induced by isometric exercise could be possibly associated with the neural activities involving pain modulation.
Local EIH and global EIH
Increasing evidence suggests exercise as a potentially effective strategy for pain modulation (1,6,36). Consistently, we found a robust hypoalgesic effect induced by the isometric exercise, which is a type of resistance exercise. Previous studies indicated that resistance exercise could lead to decreased pain sensitivity close to the site of muscle contraction (i.e., local EIH) and at the remote site of the body, distant to the muscle contraction (i.e., global EIH) (2,11,34). In line with these previous findings, we detected both the local EIH response and the global EIH response elicited by the isometric exercise within a healthy, pain-free population. This could be of potential value in various clinical applications. For instance, patients with localized pain conditions might be able to obtain pain-relieving effects by exercising remote but nonpainful parts of the body, as such a strategy could still elicit global EIH. In addition, we observed that the global EIH could vary depending on the pain modalities (i.e., mechanical pain indexed by PPT and PPR; thermal pain indexed by HPT and HPR; Fig. 2), which is consistent with previous studies (1,9,12,13). Indeed, EIH is more profound and more consistent when using mechanical stimuli rather than thermal stimuli to evoke pain (1,9,13), but the exact reason for this phenomenon is still unknown. Considering the huge difference between experimental pain and clinical pain, as well as the complexity of clinical pain conditions, the effectiveness of global EIH needs to be further verified in different clinical populations.
Another critical point ought to be mentioned is the timing of the hypoalgesic effect induced by the exercise. Previous research indicates that hypoalgesic effect is usually dissipated about 15–30 min after exercise (1). For isometric exercise, moderate to large level of EIH was produced immediately after the muscle contraction (1). The hypoalgesic effect gradually decreased both on the exercised muscles and unexercised muscles over time but dissipated more quickly on the unexercised muscles (2). Although such a hypoalgesic effect is limited, isometric exercise is still an effective treatment for transient relief from pain symptoms in clinical conditions such as acute pain. However, the application of isometric exercise for chronic pain could be challenging due to this short timing of hypoalgesic effect (37,38). Extensive work on how to optimise the hypoalgesic effect of isometric exercise in clinical pain conditions are warranted in future studies.
Effect of exercise intensity on EIH
Consistent with previous findings, the local EIH response was greatly affected by the intensities of exercise (39,40). Notably, our study further demonstrated that the global EIH responses elicited by the isometric exercise could also be modulated by the intensities of the exercise itself. To some extent, the intensities of the exercise determine whether the exercise has to produce fatigue or not. In fact, more intense and longer exercise may produce more muscle fatigue that might change the corticomotor excitability affecting the detection of cutaneous stimuli (35,41,42). The muscle fatigue exerted by high-intensity exercise might also have a substantial benefit with regard to downregulating pain-related metabolites (43). Notably, our study indicated that the intensities of exercise are critical when dealing with pain control. However, when this issue comes to a clinical situation, it becomes even more critical when using an exercise therapy for patients with its double contrasting action on pain, namely, the hypoalgesic effects activated by a therapeutic movement and the hyperalgesia effect generated by a movement incompatible with the patients” clinical conditions (44). Such importance of the exercise intensity is evident in animal studies, where it demonstrated that exercise exerted to fatigue induces the development of exercise-induced pain by the activation of acid-sensing ion channels and that mice do not develop exercise-induced pain by blocking these channels (43,45). Clinical trials with a longitudinal design concerning different intensities are warranted to verify the intensity-dependent modulation of EIH and its long-term effect in patients.
Neural mechanisms of EIH
We observed significant correlations between the changes of subjective pain perception and the changes of heat-evoked brain responses in the present study, indicating that the EIH might be related to the neural activities involving the pain modulation. According to the gate control theory, the activation of large-diameter Aβ fibers could result in a segmental inhibition of the nociceptive information transmission through small-diameter Aδ and C fibers at the dorsal horn level in the spinal cord (46). Notably, previous findings indicated that the muscle vibration caused by the isometric exercise could stimulate the Aβ fibers (1,47), thus producing the EIH via a potential spinal gating mechanism. In line with this, we observed that both the high-intensity and the low-intensity isometric exercise produced clear hypoalgesic effects to the nociceptive stimuli on the exercised limb: the increase of PPT and HPT, the reduction of N2 amplitudes and N2–P2 peak-to-peak amplitudes, and the reduction of ERP magnitudes elicited by the heat stimuli on the exercised limb (Fig. 3). Importantly, strong global EIH was induced by both the high-intensity and the low-intensity isometric exercise, as evidenced by the decrease of PPR at the index finger of the unexercised limb (Fig. 2). Therefore, the hypoalgesic effects of isometric exercise in the present study cannot be fully explained by a homotopical spinal gating mechanism, and the concomitant contribution of a descending inhibition mechanism at the supraspinal level could remain a possibility. In partly support of this possibility, we observed that: 1) a significant reduction of N2–P2 amplitudes elicited by the heat stimuli on the exercised limb, with a little change for that on the unexercised limb (Figs. 3 and 4), and 2) the HPR was significantly correlated with the heat-evoked brain response (i.e., the N2–P2 amplitudes and the ERP magnitudes) on the unexercised limb (Table 2). It is worth mentioning that there was a strong similarity of cerebral dipoles activated by contact heat stimuli and laser stimuli, as evidenced by previous studies (48,49). Similar to that in laser-evoked potentials, the N2–P2 complex in HEP is mainly generated from insular and anterior cingulate cortices (50). Given that the N2–P2 complex is related to the activity of both insular and anterior cingulate cortices (50), the global EIH could be possibly consequent to a top-down inhibitory control of the nociceptive information. Such a top-down inhibitory control could be triggered by the function of the two core regions involved in the descending pain inhibitory pathway, which is likely modulated by the motor cortex activated during exercise (51,52). Indeed, evidence from a recent study showed that selective inhibition of the neural projection from the anterior cingulate cortex directly to the spinal cord could induce hypoalgesic effects in rats (53). In sum, the hypoalgesic effects of isometric exercise not only could be attributed to the modulation of nociceptive information transmission via a spinal gating mechanism but also could rely on a top-down descending pain inhibitory mechanism.
Limitations
There were still some limitations in our study. First, there is a lack of EEG recording during mechanical stimuli in the current study. The thermal and the mechanical stimuli could activate different nociceptors, thus possibly resulting in different stimulus-evoked brain responses (32). Considering this point, future studies are needed to clarify the neural mechanisms underlying EIH by comparing the brain responses elicited by different modes of pain stimuli (i.e., thermal, mechanical, and chemical). Second, we did not pay attention to the timing of the hypoalgesic. Such an acute exercise may generate limited and transient hypoalgesic effect (1), which hinders its application on pain relief in clinic. Future studies focusing on extending the hypoalgesic effect of exercise are warranted. As evidenced by previous studies, the exercise dose (i.e., exercise frequency, exercise intensity, exercise time and exercise type) could affect the efficiency of EIH (1). Thus, a deep insight into the relationship between the exercise dose and the timing of EIH could be a potential way to modify the EIH effect in clinical conditions.
CONCLUSIONS
The hypoalgesic effects induced by the isometric exercise were not only localized to the moving part of the body but also can be extended to the distal part of the body. The intensities of the isometric exercise play a vital role in modulating such effects. The EIH responses could be related to the modulation of nociceptive information transmission via a spinal gating mechanism and also rely on a top-down descending pain inhibition mechanism. Our study provides insights into the neural mechanisms of EIH and promotes the possible applications of isometric exercise to clinical pain management in the future.
This work was supported by the National Natural Science Foundation of China (32071061 and 31822025); Fok Ying-Tong Education Foundation of China (161092); the scientific and technological research program of the Shanghai Science and Technology Committee (fund number 19080503100); and the Shanghai Key Lab of Human Performance (Shanghai University of Sport) (11DZ2261100).
No conflict of interest was reported by the authors. The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The results of this study do not constitute endorsement by the American College of Sports Medicine.
REFERENCES
1. Naugle KM, Fillingim RB, Riley JL 3rd. A meta-analytic review of the hypoalgesic effects of exercise.
J Pain. 2012;13(12):1139–50.
2. Lannersten L, Kosek E. Dysfunction of endogenous pain inhibition during exercise with painful muscles in patients with shoulder myalgia and fibromyalgia.
Pain. 2010;151(1):77–86.
3. Luan X, Tian X, Zhang H, et al. Exercise as a prescription for patients with various diseases.
J Sport Health Sci. 2019;8(5):422–41.
4. Vaegter HB, Handberg G, Graven-Nielsen T. Similarities between exercise-induced hypoalgesia and conditioned pain modulation in humans.
Pain. 2014;155(1):158–67.
5. Meeus M, Roussel NA, Truijen S, Nijs J. Reduced pressure pain thresholds in response to exercise in chronic fatigue syndrome but not in chronic low back pain: an experimental study.
J Rehabil Med. 2010;42(9):884–90.
6. Baiamonte BA, Kraemer RR, Chabreck CN, et al. Exercise-induced hypoalgesia: Pain tolerance, preference and tolerance for exercise intensity, and physiological correlates following dynamic circuit resistance exercise.
J Sports Sci. 2017;35(18):1–7.
7. Umeda M, Newcomb LW, Koltyn KF. Influence of blood pressure elevations by isometric exercise on pain perception in women.
Int J Psychophysiol. 2009;74(1):45–52.
8. Koltyn KF, Umeda M. Contralateral attenuation of pain after short-duration submaximal isometric exercise.
J Pain. 2007;8(11):887–92.
9. Koltyn KF. Analgesia following exercise: a review.
Sports Med. 2000;29(2):85–98.
10. Alsouhibani A, Vaegter HB, Hoeger Bement M. Systemic exercise-induced hypoalgesia following isometric exercise reduces conditioned pain modulation.
Pain Med. 2019;20(1):180–90.
11. Koltyn KF, Brellenthin AG, Cook DB, Sehgal N, Hillard C. Mechanisms of exercise-induced hypoalgesia.
J Pain. 2014;15(12):1294–304.
12. Jones MD, Taylor JL, Booth J, Barry BK. Exploring the mechanisms of exercise-induced hypoalgesia using somatosensory and laser evoked potentials.
Front Physiol. 2016;7:581.
13. Vaegter HB, Hoeger Bement M, Madsen AB, Fridriksson J, Dasa M, Graven-Nielsen T. Exercise increases pressure pain tolerance but not pressure and heat pain thresholds in healthy young men.
Eur J Pain. 2017;21(1):73–81.
14. Garcia-Larrea L, Peyron R. Motor cortex stimulation for neuropathic pain: from phenomenology to mechanisms.
Neuroimage. 2007;37(1 Suppl):S71–9.
15. Peyron R, Faillenot I, Mertens P, Laurent B, Garcia-Larrea L. Motor cortex stimulation in neuropathic pain. Correlations between analgesic effect and hemodynamic changes in the brain. A PET study.
Neuroimage. 2007;34(1):310–21.
16. García-Larrea L, Peyron R, Mertens P, et al. Electrical stimulation of motor cortex for pain control: a combined PET-scan and electrophysiological study.
Pain. 1999;83(2):259–73.
17. Yezierski RP, Gerhart KD, Schrock BJ, Willis WD. A further examination of effects of cortical stimulation on primate spinothalamic tract cells.
J Neurophysiol. 1983;49(2):424–41.
18. Favorov OV, Pellicer-Morata V, DeJongh Curry AL, et al. A newly identified nociresponsive region in the transitional zone (TZ) in rat sensorimotor cortex.
Brain Res. 1717;2019:228–34.
19. Senapati AK, Huntington PJ, Peng YB. Spinal dorsal horn neuron response to mechanical stimuli is decreased by electrical stimulation of the primary motor cortex.
Brain Res. 2005;1036(1–2):173–9.
20. Lopes PSS, Campos ACP, Fonoff ET, Britto LRG, Pagano RL. Motor cortex and pain control: exploring the descending relay analgesic pathways and spinal nociceptive neurons in healthy conscious rats.
Behav Brain Funct. 2019;15(1):5.
21. Pagano RL, Fonoff ET, Dale CS, Ballester G, Teixeira MJ, Britto LRG. Motor cortex stimulation inhibits thalamic sensory neurons and enhances activity of PAG neurons: possible pathways for antinociception.
Pain. 2012;153(12):2359–69.
22. Black CD, Huber JK, Ellingson LD, et al. Exercise-induced hypoalgesia is not influenced by physical activity type and amount.
Med Sci Sports Exerc. 2017;49(5):975–82.
23. Wesolowicz DM, Clark JF, Boissoneault J, Robinson ME. The roles of gender and profession on gender role expectations of pain in health care professionals.
J Pain Res. 2018;11:1121–8.
24. Kiałka M, Milewicz T, Sztefko K, Rogatko I, Majewska R. Metformin increases pressure pain threshold in lean women with polycystic ovary syndrome.
Drug Des Devel Ther. 2016;10:2483–90.
25. Moraska AF, Stenerson L, Butryn N, Krutsch JP, Schmiege SJ, Mann JD. Myofascial trigger point-focused head and neck massage for recurrent tension-type headache: a randomized, placebo-controlled clinical trial.
Clin J Pain. 2015;31(2):159–68.
26. Celenay ST, Akbayrak T, Kaya DO. A comparison of the effects of stabilization exercises plus manual therapy to those of stabilization exercises alone in patients with nonspecific mechanical neck pain: a randomized clinical trial.
J Orthop Sports Phys Ther. 2016;46(2):44–55.
27. Meng J, Jackson T, Chen H, et al. Pain perception in the self and observation of others: an ERP investigation.
Neuroimage. 2013;72:164–73.
28. Rütgen M, Seidel EM, Riečanský I, Lamm C. Reduction of empathy for pain by placebo analgesia suggests functional equivalence of empathy and first-hand emotion experience.
J Neurosci. 2015;35(23):8938–47.
29. Peng WW, Tang ZY, Zhang FR, et al. Neurobiological mechanisms of TENS-induced analgesia.
Neuroimage. 2019;195:396–408.
30. Hu L, Zhang ZG, Mouraux A, Iannetti GD. Multiple linear regression to estimate time–frequency electrophysiological responses in single trials.
Neuroimage. 2015;111:442–53.
31. Colloca L, Corsi N, Fiorio M. The interplay of exercise, placebo and nocebo effects on experimental pain.
Sci Rep. 2018;8(1):14758.
32. Boca SM, Leek JT. A direct approach to estimating false discovery rates conditional on covariates.
PeerJ. 2018;6:e6035.
33. Peirs C, Seal RP. Neural circuits for pain: recent advances and current views.
Science. 2016;354(6312):578–84.
34. Vaegter HB, Madsen AB, Handberg G, Graven-Nielsen T. Kinesiophobia is associated with pain intensity but not pain sensitivity before and after exercise: an explorative analysis.
Physiotherapy. 2018;104(2):187–93.
35. Hoeger Bement MK, Weyer A, Hartley S, Yoon T, Hunter SK. Fatiguing exercise attenuates pain-induced corticomotor excitability.
Neurosci Lett. 2009;452(2):209–13.
36. Gajsar H, Titze C, Hasenbring MI, Vaegter HB. Isometric back exercise has different effect on pressure pain thresholds in healthy men and women.
Pain Med. 2017;18(5):917–23.
37. La Touche R, Fernández Pérez JJ, Martínez García S, Cuenca-Martínez F, López-de-Uralde-Villanueva I, Suso-Martí L. Hypoalgesic effects of aerobic and isometric motor imagery and action observation exercises on asymptomatic participants: a randomized controlled pilot trial.
Pain Med. 2020;21(10):2186–99.
38. Rice D, Nijs J, Kosek E, et al. Exercise-induced hypoalgesia in pain-free and chronic pain populations: state of the art and future directions.
J Pain. 2019;20(11):1249–66.
39. Koltyn KF. Exercise-induced hypoalgesia and intensity of exercise.
Sports Med. 2002;32(8):477–87.
40. Hoeger Bement MK, Dicapo J, Rasiarmos R, Hunter SK. Dose response of isometric contractions on pain perception in healthy adults.
Med Sci Sports Exerc. 2008;40(11):1880–9.
41. Sacco P, Thickbroom GW, Byrnes ML, Mastaglia FL. Changes in corticomotor excitability after fatiguing muscle contractions.
Muscle Nerve. 2000;23(12):1840–6.
42. Svoboda J, Sovka P, Stancák A. Effects of muscle contraction on somatosensory event-related EEG power and coherence changes.
Neurophysiol Clin. 2004;34(5):245–56.
43. Khataei T, Harding AMS, Janahmadi M, et al. ASICs are required for immediate exercise-induced muscle pain and are downregulated in sensory neurons by exercise training.
J Appl Physiol (1985). 2020;129(1):17–26.
44. Casale R, Chimento PL, Bartolo M, Taveggia G. Exercise and movement in musculoskeletal pain: a double-edged problem.
Curr Opin Support Palliat Care. 2018;12(3):388–92.
45. Gregory NS, Brito RG, Fusaro MCGO, Sluka KA. ASIC3 is required for development of fatigue-induced hyperalgesia.
Mol Neurobiol. 2016;53(2):1020–30.
46. Melzack R, Wall PD. Pain mechanisms: a new theory.
Science. 1965;150(3699):971–9.
47. Ackerley R, Sverrisdόttir YB, Birklein F, Elam M, Olausson H, Krämer HH. Cutaneous warmth, but not touch, increases muscle sympathetic nerve activity during a muscle fatigue hand-grip task.
Exp Brain Res. 2020;238(4):1035–42.
48. Iannetti GD, Zambreanu L, Tracey I. Similar nociceptive afferents mediate psychophysical and electrophysiological responses to heat stimulation of glabrous and hairy skin in humans.
J Physiol. 2006;577(Pt 1):235–48.
49. Valeriani M, Le Pera D, Niddam D, Chen AC, Arendt-Nielsen L. Dipolar modelling of the scalp evoked potentials to painful contact heat stimulation of the human skin.
Neurosci Lett. 2002;318(1):44–8.
50. Garcia-Larrea L, Frot M, Valeriani M. Brain generators of laser-evoked potentials: from dipoles to functional significance.
Neurophysiol Clin. 2003;33(6):279–92.
51. Porreca F, Ossipov MH, Gebhart GF. Chronic pain and medullary descending facilitation.
Trends Neurosci. 2002;25(6):319–25.
52. Da Silva JT, Zhang Y, Asgar J, Ro JY, Seminowicz DA. Diffuse noxious inhibitory controls and brain networks are modulated in a testosterone-dependent manner in Sprague Dawley rats.
Behav Brain Res. 2018;349:91–7.
53. Chen T, Taniguchi W, Chen QY, et al. Top-down descending facilitation of spinal sensory excitatory transmission from the anterior cingulate cortex.
Nat Commun. 2018;9(1):1886.
