| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Psychophysiology of Pain and Pathophysiology of Integrative Autonomic Systems Laboratories, University Centre for Adaptive Disorders and Headache (UCADH) (C.T., G.S., A.P.C., G.S.), IRCCS "C. Mondino" Foundation; Department of Neurological Sciences, University of Pavia, Pavia, Italy; the Department of Gynaecology (R.E.N.), IRCCS S. Matteo, University of Pavia, Pavia, Italy; and the University of Piemonte Orientale "A. Avogadro" (E.M.), Novara, Italy.
Address reprint requests to: G. Sandrini, Department of Neurological Sciences, "C. Mondino" Foundation, University of Pavia, Via Palestro 3, 27100 Pavia, Italy. Email: gsandrin{at}unipv.it
| ABSTRACT |
|---|
|
|
|---|
METHOD: We studied in 14 healthy women during the follicular and luteal phase the nociceptive flexion reflex (RIII reflex), an objective neurophysiological method that allows exploring possible abnormal functioning of the pain-control system. The basal body temperature (BBT) was used to evaluate the different phases of the ovarian cycle. The menstrual distress questionnaire (MDQ) was also applied for monitoring somatic and psychological symptoms during the cycle.
RESULTS: During the luteal phase, the threshold of the RIII reflex (Tr) and the psychophysical threshold for pain (Tp) were both significantly reduced compared with the follicular phase. Moreover, the reflex threshold in the luteal phase was negatively correlated to the total MDQ score of the recording day.
CONCLUSIONS: A higher sensitivity to pain stimuli was observed during the luteal phase of the menstrual cycle, which probably results from a reduction in the inhibitory descending control on spinal nociceptive flexion reflex. Complex neuromodulatory interactions of ovarian steroids with other systems of neurotransmission (especially serotonergic) may account for these observations.
Key Words: gender pain steroid hormones menstrual cycle nociceptive flexion reflex threshold
Abbreviations: NFR = nociceptive flexion reflex;; BBT = basal body temperature;; MDQ = menstrual distress questionnaire.
| INTRODUCTION |
|---|
|
|
|---|
Reproductive life events and sex hormone alterations have also been related to fibromyalgia, rheumatoid arthritis, and irritable-bowel syndrome (79). Findings from animal research suggest that pain sensitivity varies across the estrous cycle (1012). Numerous pain studies have been conducted in women, examining possible variations related to the menstrual phase, in response to experimentally induced noxious stimulation (13), but no conclusive findings have been obtained. As far as pressure stimulation, cold pressor pain, and ischemic muscle pain are concerned, a general pattern emerges, with higher thresholds emerging in the follicular phase than in ovulatory, luteal, and premenstrual phases, although some studies detected random pain threshold variations throughout the menstrual cycle.
Only two studies have investigated the variation of pain perception across the menstrual cycle by means of electrical stimulation, and these have produced contrasting results. Veith et al. (14) found no difference in pain thresholds across the menstrual cycle, while Giamberardino et al. (15) found higher thresholds in the luteal phase than in the periovulatory and perimenstrual phases.
Most methods for studying pain in humans are based on a psychophysical analysis of the responsehighly subjective and poorly quantifiableto various kinds of painful stimuli (1421). Nociceptive flexion reflexes, instead, are sensory-motor responses elicited by electrical noxious stimuli, which involve activation of spinal and supraspinal neuronal circuits, providing an objective and quantitative assessment of the function of the pain-control system. The study of withdrawal reflexes in humans in response to electrical stimuli offers two important advantages: 1) the reflex thresholds can be clearly and consistently quantified, and 2) the assessment of pain thresholds is objective, as suggested by the close correlation between the threshold of the reflex and the subjective pain threshold (22).
In the present study, the nociceptive flexion reflex (NFR) was evaluated at lower limb level in healthy women during the follicular and luteal phases of their menstrual cycle to assess and quantify changes in pain control across the menstrual cycle.
| METHODS |
|---|
|
|
|---|
Procedures
All the women were instructed to record their BBT, taken every day at the same of the day, on menstrual calendars to establish both cycle regularity and the occurrence of ovulation. BBT monitoring was performed for 3 months, the first month serving as a run-in. Throughout the entire study period, the subjects were also instructed to fill in, daily, the MDQ for monitoring the premenstrual symptomatology (23, 24).
Each participant was tested once in each of the following phases of her menstrual cycle: follicular (days 810 from the first day of menstrual bleeding) and luteal (days 68 from ovulation). After completion of the test sessions, the subjects were required to report the date of onset of their subsequent menstruation to allow retrospective confirmation of phase assessment. BBT changes and the dates of the most recent menses/cycles were used to verify the phase during which the experimental session actually occurred.
Before starting formal measurements for the study, the subjects were given an initial training session to familiarize them with the procedure of judging pain thresholds. During this session, the participants were clearly informed about the procedure and its purpose.
NFR was studied at the biceps femoris, according to the method described by Willer (22), at the same time of day (between 9:00 and 11:00 AM) for all subjects; the test was conducted in a manner identical to the training session. The subjects were seated in a comfortable armchair in a quiet environment where the temperature was kept at a constant level (around 23 ± 2°C). Their lower limbs were positioned to ensure complete muscle relaxation, with the knee flexed at 130° and the ankle at 90°. The stimulating surface electrodes were placed at the retromalleolar site or slightly further down along the course of the sural nerve, 2 cm apart, with the cathode placed proximally. The recording surface electrodes were placed on the tendon (anode) and over the belly of the biceps femoris capitis brevis muscle (cathode). Stimuli were delivered by means of a constant current isolation unit fitted with an amperometer. A unit for random delivery of stimuli was added to the system to avoid facilitation phenomena related to anticipation of the stimulus. The sural nerve was stimulated percutaneously by a volley of ten 1-millisecond rectangular pulses delivered over 20 milliseconds at 300 Hz of internal frequency. Muscular response (nociceptive flexion reflex; RIII reflex) was recorded from the biceps femoris muscle capitis brevis using an electromyographic technique. The staircase method devised by Willer (22) was used. The initial stimulus intensity corresponded to 5 mA, and subsequent stimuli were delivered with increments of 0.5 mA until stable values were reached. Intensity range varied from 5 to 25 mA. Each session lasted 6070 minutes and consisted of three tasks separated by 7- to 10-minute intervals. This was done to confirm reproducibility of the values during the session. Reproducibility and consistency of the responses in different sessions had been previously demonstrated (22, 44).
The RIII threshold (Tr) was defined as the intensity eliciting a flexor response with a probability of 80% to 90%. The subjective intensity of the painful sensation elicited by the sural stimulation was estimated by subjects on an 11-point numerical scale graded from 0 = no pain to 10 = unbearable pain (25). The pain threshold (Tp) corresponded to the intensity of the stimulus that evoked a painful sensation with an intensity equal to level 3 on the 11-point numerical scale.
The Tp/Tr ratio, which represents the ratio between stimulus intensities (mA) needed to obtain respectively Tp and Tr, was also measured (25, 26). To minimize order effects, the sequence of the two test phases (follicular/luteal) was not the same in all the women but rather was randomly distributed between them.
Statistical Analysis
Means and standard deviations of the pain thresholds and MDQ scores were calculated for each session. Total MDQ score was computed daily throughout the menstrual cycle by calculating the mean of the scores on the various items that make the different MDQ subscales (pain, concentration, behavioral changes, autonomic reactions, water retention, negative mood, arousal, and control). The items were rated 0 = no symptom, 1 = symptom with light intensity, 2 = symptom with moderate intensity, and 3 = symptom with severe intensity. Pain MDQ score was also calculated daily by computing the mean score on the pain subscale items. For statistical purposes, we evaluated total and pain MDQ scores recorded during the 24 hours before the follicular and luteal sessions, respectively.
Analysis of variance was performed to evaluate the presence of any order effect resulting from the randomization of the first test phase. A paired t test analysis was applied to examine a possible trend of threshold variations during the cycle. The Spearman test was used to analyze the relationship between total and subtotal MDQ scores and reflex thresholds during the days preceding the follicular and luteal neurophysiological evaluations.
| RESULTS |
|---|
|
|
|---|
Slightly, but significantly lower Tr and Tp values were observed during the luteal phase (Figure 1,a and b), with the Tp/Tr ratio remaining in a range between 0.9 and 1.1. Lower Tr and Tp values in the luteal phase were observed in the majority but not in all subjects. Tr and Tp remained stable across the cycle in two subjects, while their values were slightly lower in the follicular phase in other two women.
|
|
|
| DISCUSSION |
|---|
|
|
|---|
The mechanisms underlying the fluctuations of pain thresholds across the menstrual cycle and their relation to painful symptoms/syndromes are presently unknown. The influence of psychological variables has been investigated, but no phase-dependent changes in emotional symptomatology have been found to correlate with changes in pain perception (14, 16, 29). The recent demonstrations that female sex hormones can modulate neuronal excitability via effects on ion channels (30, 31) and that progesterone fluctuations are related to cortical excitability (32, 33) open up an entirely new perspective in which menstrual-related fluctuations of pain perception and pain symptoms are the results of spinal and supraspinal changes in neurotransmissional and hormonal assets.
Gender differences in the perception and modulation of pain have been extensively described (4), and there is no doubt that estrogen and progesterone play a pivotal role in pain thresholds during pregnancy and parturition by acting on the opioid system (3437). The lack of hormonal data represents a limitation of this study; however, other authors previously failed to demonstrate a correlation between gonadal hormonal changes and pain sensitivity (29). Fillingim et al. demonstrated that higher estrogen levels are associated with increased thermal pain sensitivity, which did not change across the menstrual cycle, while no hormonal correlation was found with ischemic pain sensitivity, which was lower in the follicular phase (16). Thus, it seems that hormonal influence might play a minor direct role in the individuals sensitivity and response to pain. This does not preclude the involvement of the endogenous opioid system since a clear effect of the menstrual cycle phase on opioid control of reproductive function has been observed (38). Animal studies indicate that the induction of a luteinizing hormone (LH) surge leads to a diminished analgesic response to morphine because of desensitization of brain opiate receptors (39, 40). Thus, one can speculate that hormonally induced (ovulation) opiate receptor desensitization could enhance luteal phase pain sensitivity among women.
The existence in humans of a descending opiate control system, which has an inhibitory effect on NFRs, is undisputed (26, 41, 42), and the opioid system is possibly responsible for the fluctuations in pain thresholds across the menstrual cycle (43). However, several human studies have suggested that the opiate control system does not play a relevant role in the modulation of NFRs. Sandrini et al. (44) showed that no correlation exists between the circadian fluctuation in the RIII reflex thresholds and the circadian rhythms of beta-endorphin and beta-lipotropin. In addition, naloxone pretreatment failed to change RIII reflex thresholds in healthy volunteers (45).
Alternatively, it has been demonstrated that serotonin modulates NFRs (46, 47), and ovarian steroids may exert a complex modulatory action on the serotonergic system, affecting metabolites, activity, and receptors (48). This effect is in keeping with the findings of Schoenen et al. (49) on the exteroceptive suppression (ES2) of the temporalis muscle assessed in women with tension-type headache. They report a marked menstrual reduction for the ES2 compared with the periovulatory phase, positively related to the plasma estradiol/progesterone ratios. As inhibitory interneurons responsible of the ES reflex are under the influence of aminergic pathways descending from limbic areas, partially common to the antinociceptive system, these data clearly show a modulatory effect of estrogen on serotonergic pathways that control excitability of brain stem interneurons. Actually, animal studies suggest changes in serotonergic levels during the estrous cycle and in ovariectomized estradiol-treated rats (50). In humans, serotonin concentration is lowest premenstrually (51), and there is evidence that serotonin levels in whole-blood and plasma and that platelet uptake and content are lower premenstrually in women with premenstrual syndrome (52). An inhibitory serotonergic control over the RIII reflex has been suggested in animal studies (41, 53) and has been recently confirmed in humans (46). Additionally, differences in response to electrically evoked pain might be the result of menstrual-related changes in the tone of the sympathetic system (5456). It ensues that many variables may be involved in the RIII threshold fluctuations observed in this study. These variables may be only partially associated with hormonal fluctuations. For instance, serotonin blockade does not influence LH pulsatile secretion, but it markedly inhibits pulsatile FSH release (57). We speculate that pain sensitivity across the menstrual cycle is under the influence of multiple oscillators regulated by hypothalamic centers, the rhythms of which may not be synchronous to that of ovarian hormones. Additional research is needed to test this hypothesis.
The present findings are in partial disagreement with the two previous studies that evaluated pain perception across the menstrual cycle in response to electrical stimulation (14, 15). At variance with our findings, these studies reported either a higher threshold or no change in the luteal phase in normal women. Methodological differences might explain the contrasting findings. First, the timing of sampling was not always comparable. For instance, Giamberardino (15) did not evaluate the follicular phase, while she investigated menstrual, periovulatory, and premenstrual phases, none of which was investigated in the present study. Veith et al. (14) used electric shock stimulation, which consisted of the application of concentric circular electrodes to the inside of the participants nondominant wrist. Giamberardino et al. used 18-millisecond trains of 0.5-millisecond monophasic square-wave pulses, repeated automatically every 2 seconds by means of surface electrodes for the skin and needle electrodes for muscle and subcutis. In neither study were quantitative or semiquantitative scales used to express the exact intensity of the stimulus needed to evoke a painful sensation; hence, in both cases, the pain response constituted a totally subjective element. In our study, pain perception was coupled with an objective marker of nociceptive pathways (RIII reflex). Interestingly, our findings of higher pain threshold during the follicular phase are in agreement with most of the studies that examined pain threshold in response to stimuli other than electrical. In the past, several hypotheses have been proposed to explain the different findings obtained when comparing pain response to electrical stimulation with responses to other stimulus modalities. These hypotheses have ranged from the proposal of a different impact of pain sensitivity for different stimulations to the speculation that electrical stimuli have different perceptual-emotional dimensions of pain (13).
In conclusion, though based on a small number of subjects, the present study suggests that women tend to have a higher susceptibility to the perception of painful symptomatology during the luteal phase. This is likely to be a result of complex central/peripheral interactions between specific neurotransmitters (ie, serotonin and opiates) and ovarian steroids. Additional studies are required to confirm the present data on a larger population and to explore the possible central mechanisms that are involved.
Received for publication February 27, 2001.
Revision received August 30, 2001.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |