| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Departments of Behavioral Sciences (M.A., A.H.), Family Medicine (M.A.), and Physiology & Pharmacology (M.A., L.W.), University of Minnesota Medical School, Duluth, MN; and the Department of Psychology (C.F., J.F.), Ohio University, Athens, OH.
Address correspondence and reprint requests to Mustafa al'Absi, PhD, Department of Behavioral Sciences, University of Minnesota Medical School, Duluth, MN 55812. E-mail: malabsi{at}umn.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Methods: Ninety-nine participants completed two sessions during which a placebo or 50 mg naltrexone was administered using a double-blind, counterbalanced design. Participants rated their pain and completed the McGill Pain Questionnaire (MPQ) after three assessments of the nociceptive flexion reflex and after assessment of nociceptive pain threshold and tolerance. Saliva samples were obtained throughout the sessions.
Results: Salivary cortisol levels increased after pain assessment after the ingestion of naltrexone, but not after placebo, with the low-risk group exhibiting an earlier peak of cortisol response. Participants reported greater pain ratings and higher MPQ scores in the naltrexone versus placebo condition, and these effects were more pronounced in women. Pain threshold and tolerance were higher among high-risk men relative to low-risk men.
Conclusions: The results are consistent with the inhibitory effects of the endogenous opioids on cortisol response and suggest an altered response timeline among hypertension-prone individuals. The results demonstrate that hypoalgesia may be a marker of hypertension risk in men but not in women.
Key Words: cortisol • hypertension risk • pain • opioid blockade • naltrexone • gender
Abbreviations: ACTH = adrenocorticotropin; BP = blood pressure; CPT = cold pressor test; CRF = corticotrophin-releasing factor; HPA = hypothalamic-pituitary-adrenocortical axis; MPQ = McGill Pain Questionnaire; NFR = nociceptive flexion reflex; EMG = electromyographic activity.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Based on results of opioid blockade studies, McCubbin (13,14) hypothesized that both exaggerated cardiovascular responsivity and increased HPA axis activation may be observed among at-risk individuals as a result of an attenuation of normal opioidergic inhibitory feedback to hypothalamic areas involved in autonomic and HPA axis regulation. Specifically, McCubbin (13) observed that administration of an opioid antagonist (naloxone) resulted in greater blood pressure, ACTH, and plasma cortisol responsivity to arithmetic stress in individuals with low versus high resting blood pressure levels. These findings were interpreted as a relatively greater "unmasking" of the tonic inhibitory influence of endogenous opioids in low- versus high-risk individuals after opioid blockade.
As a further test of McCubbin's model of diminished opioid inhibition in individuals at risk for hypertension, the present study examined the effect of opiate blockade on cortisol responsivity to a painful stressor in individuals with and without a parental history of hypertension. As compared with the prior study, we chose to use a painful stressor to increase the likelihood of significant endogenous opioid activation. Specifically, using a double-blind, counterbalanced design, participants received either placebo or naltrexone before assessment of their nociceptive flexion reflex (NFR) threshold. The NFR paradigm has been used extensively in pain research, in part because the reflex threshold provides an objective physiological correlate of nociceptive responding. Salivary samples were obtained repeatedly during testing to provide for subsequent analysis of cortisol responsivity. In line with McCubbin's model, we hypothesized that offspring of normotensive parents would show greater cortisol responsivity to the painful stressor after endogenous opioid blockade. Based on existing evidence of decreased nociceptive responding in individuals at high versus low risk for hypertension (15), we further hypothesized that offspring of hypertensive parents would show higher NFR thresholds as compared with offspring of normotensive parents.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
A blood pressure history survey (18) was mailed to the biological parents of participants to confirm parental hypertension and antihypertensive medication history. Participants were classified as high risk (having a positive parental history of hypertension if at least one parent indicated that they had been diagnosed with hypertension by their physician; n = 39) or low risk for hypertension (neither parent diagnosed with hypertension; n = 60). The study was approved by the Institutional Review Board of the University of Minnesota and Ohio University. All participants signed a consent form and received a monetary incentive for participation.
Procedures
After the medical screening session, participants were scheduled for two sessions of approximately 3 hours each scheduled a minimum of 72 hours apart. To control for circadian rhythm effects, most testing sessions were conducted in the afternoon. As a result of scheduling difficulties 14% completed sessions in the morning hours with both sessions conducted within the same period of the day. Women were tested while in the follicular phase (i.e., within 3–11 days after the onset of menstruation). Before each session, participants received a reminder concerning the restrictions on the use of any alcohol or analgesic medication for 24 hours and narcotic medication for 3 days.
Figure 1 provides a graphic overview of the study protocol for each session. On arrival at the laboratory, participants were led to a quiet room and received information about the session. A 10-minute resting baseline measurement of blood pressure and heart rate was then completed using a Dinamap blood pressure monitor. Participants then ingested a capsule containing either 50 mg naltrexone (Trexan; DuPont, Wilmington, DE) or placebo. Participants were asked to sit quietly for 1 hour to allow for drug absorption. During this period, they completed a number of questionnaires, and electrodes were attached for NFR assessments. After the absorption period, a second 10-minute baseline of blood pressure and heart rate was obtained. Three NFR assessments were then conducted followed by a pain threshold and tolerance assessment. NFR assessments were conducted before, during, and after performance of a video game task.
|
To prepare for NFR assessment, sural nerve stimulation sites and biceps femoris electromyogram (EMG) recording sites were cleaned using Omni Prep electrode paste and an impedance of less than 10 K
was verified using a UFI Checktrode (model MKII). Del Sys EMG electrodes were then placed over the left biceps femoris muscle with a reference electrode attached over the lateral epicondyle of the femur. The EMG electrodes were connected to a DelSys, Bagnoli-2 EMG amplifier, and during NFR assessments the EMG was filtered (20–450 Hz), amplified (x10,000), and recorded at 2 KHz using a CED Micro1401 analog-to-digital converter and Spike2 software. After attachment of the EMG electrodes, a Nicolet bar electrode was attached to the left leg over the retromalleolar pathway of the sural nerve. The stimulating electrode was then connected to a Digitimer DS7A constant-current stimulator. After attachment of the recording and stimulating electrodes, participants were seated in a Hi-Seat rehabilitation chair with a leg rest adjusted to maintain knee flexion at approximately 60° from horizontal position.
Assessment of the NFR threshold followed procedures reported in prior studies (15–17) and included repeated administration of electrocutaneous stimulation applied over the sural nerve using an up-down staircase method. Specifically, NFR threshold (in milliamps) was computed by averaging the current levels required to elicit a reflex during an ascending, or "up" sequence, with the current levels that no longer elicited a reflex during a descending, or "down" sequence.
To assess the NFR threshold, electrocutaneous stimulation was applied over the sural nerve at a variable interval schedule of 20 seconds (range, 15–25 seconds) to reduce stimulus habituation and predictability. Each stimulation trial consisted of a volley of five 1-ms rectangular pulses with a 3-ms interpulse interval (total duration = 17 ms). Using an up-down staircase method (19), stimulation intensity was increased in 4-mA increments until a nociceptive flexion reflex was detected (or a maximum intensity of 40 mA was reached) and then was decreased in 2-mA increments until a reflex was no longer detected. Continuing from this intensity, the procedure was then repeated using 1-mA increments so that the reflex appeared and subsided two more times. Nociceptive flexion reflex occurrence was defined as a mean rectified EMG response in the 90- to 150-ms poststimulation interval that exceeded mean rectified EMG activity during the 60-ms prestimulation baseline (from –65 to –5 ms) interval by at least 1.5 standard deviations. The 90- to 150-ms interval was chosen because it avoids possible contamination by the low-threshold cutaneous flexor reflex, startle reactions, and voluntary movements (15). The NFR threshold (in milliamps) was defined as the average of the peaks during the last two ascending sequences (current intensity that elicited a reflex) and troughs during the last two descending sequences (current intensity that no longer elicited a reflex).
During the first and third NFR threshold assessments, participants rated the perceived magnitude of each stimulation using a verbal rating scale with anchors of one (sensory threshold), 25 (uncomfortable), 50 (painful), 75 (very painful), and 100 (maximum tolerable). During the second reflex threshold assessment, participants played a common video game (Nintendo Tetris) for the duration of the assessment period. This task was included as a distracting task to assess its effect on NFR threshold. The hypothesis was that because of enhanced activation of descending inhibitory pain modulation pathways, NFR threshold would increase during this task. Immediately after each of the three reflex assessments, participants completed the Short-Form MPQ (20) as well as a retrospective rating of overall pain during the reflex assessment using a zero (no pain) to 100 (maximum pain tolerable) rating scale of their overall pain experience.
In addition to the NFR assessments, an electrocutaneous pain threshold and pain tolerance procedure was conducted. Pain threshold was defined as the first stimulation intensity (in milliamps) that received a rating of 50 "painful" or greater. Pain tolerance was defined as the stimulation intensity (in milliamps) corresponding to a rating of 100 "maximum tolerable." A value of 40 mA was assigned if the participant reached the maximum current permitted. After the tolerance assessment, participants completed a retrospective rating of the pain tolerance procedure using the same verbal rating scale as described previously and the Short-Form MPQ to describe overall pain experience.
During each session, six saliva samples were collected at time points indicated in Figure 1. Samples were collected at the indicated times to allow for the examination of changes in cortisol concentrations after opioid blockade and exposure to the nociceptive stimuli consistent with previous protocols (21,22). Samples were collected using cotton dental rolls held in the mouth until saturated and collected into a plastic tube (Salivette tubes; Sarstedt, Rommelsdorf, Germany). Salivary cortisol assays were conducted using a time-resolved immunoassay with fluorometric end point detection. The assay has a minimum sensitivity of 0.5 nmol/L (23). Data collection for this study occurred between 2001 and 2003.
Data Analysis
The primary variables for this study were salivary cortisol and pain measures (intensity ratings, MPQ scores, and NFR threshold). Data analyses were conducted using analyses of variance methods as outlined for each variable subsequently. All repeated measure analyses used Wilks' lambda correction to test sampling time effect and to correct for repeated measures (24,25). Analyses were conducted using the SYSTAT software package (SYSTAT, Inc., Evanston, IL). Analyses were conducted on maximum numbers of subjects with available cortisol data from either session.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
.001), whereas women had a higher mean heart rate than men (F[1, 95] = 5.69, p < .05).
|
Cortisol Concentrations
Salivary cortisol concentration data (Table 2) were analyzed using a 2 (risk for hypertension: high, low) x 2 (gender) x 2 (medication: placebo, naltrexone) x 6 (samples) multivariate repeated-measure analysis of variance (MANOVA) with risk and gender as between-subject factors and medication and samples as within-subject variables. This analysis demonstrated a significant risk x medication x sample interaction (F [5, 80] = 2.77, p = .02). As depicted in Figure 2, this interaction reflects a significant increase in cortisol concentration after the pain assessment procedures in the naltrexone condition that was more pronounced in the low-risk group. To further test differences between low- and high-risk groups, levels obtained in the placebo conditions were subtracted from corresponding values obtained in the naltrexone condition and then premedication levels were subtracted from postmedication levels. Results of this analysis revealed a significant risk x time interaction (F [4, 81] = 3.12, p < .02), which reflected the fact that participants in the low-risk group exhibited earlier peak cortisol concentrations after naltrexone administration as compared with the high-risk group. These results were not altered by covarying for weight or BMI.
|
|
Pain Measures
Pain reports were analyzed using a 2 (risk: high, low) x 2 (gender) x 2 (medication: placebo, naltrexone) x 3 (trials) MANOVA with gender and risk as between-subject factors and medication and trials as within-subject variables. Participants reported greater pain during the naltrexone condition than placebo (F [1, 93] = 4.13, p < .05). A similar effect was found using the MPQ scores, with a trend indicating greater MPQ scores in the naltrexone condition (F [1, 95] = 4.04, p = .05). Furthermore, this effect was more pronounced in women, as indicated by the gender x medication interaction (F [1, 95] = 4.57, p = .04). In addition, the effects of trial as well as the risk x trial interaction were significant (F [2, 94] >3.84, p < .03), reflecting a decline in MPQ scores after the second and third assessment of NFR in the low- but not in the high-risk group (see Table 3).
|
For NFR thresholds, a trial main effect was observed (F [2, 64] = 12.50, p < .0001), reflecting declines in NFR threshold in the second assessment conducted during the video game. A gender main effect reflected higher NFR thresholds in men versus women (F [1, 65] = 6.71, p = .02). Finally, as seen in Figure 3, a gender x risk interaction (F [1, 65] = 5.56, p = .02) reflected higher NFR thresholds in high- versus low-risk men but the opposite pattern in women.
|
Electrocutaneous pain threshold was higher among participants at high risk for hypertension compared with those at low risk (F [1, 95] = 4.45, p < .05). However, a marginal gender x risk interaction (F [1, 95] = 3.82, p = .053) suggested higher electrocutaneous thresholds in men, but not women, at high versus low risk for hypertension. Electrocutaneous pain tolerance was higher among men compared with women (F [1, 95] = 12.17, p < .001). A marginal gender x risk interaction (F [1, 95] = 3.18, p = .08) also reflected greater tolerance in men at high risk for hypertension (p = .01), but no significant difference in women (F < 1) (see Table 3). These results were not altered by covarying for weight or BMI.
Correlation Analyses
In both drug conditions, resting systolic blood pressure level was positively associated with pain tolerance (r = 0.25, p = .01). Separate correlation analyses within the high- and the low-risk groups revealed consistent correlations of systolic and diastolic blood pressure with pain threshold and tolerance in individuals at high risk for hypertension (r = 0.34 to 0.48, p < .05), but not in the low-risk group (r < 0.07).
Correlation analyses were also conducted to assess the extent to which pain measures vary with salivary cortisol measures within each drug condition. During the placebo condition correlation among pain ratings, MPQ scores obtained after each of the three NFR assessments, and cortisol concentrations obtained after NFR assessment were consistently positive (r = 0.22 to 0.39, p < .05). Positive correlations were also found between salivary cortisol levels and NFR thresholds (r = 0.29 to 0.37, p < .01), suggesting that increased cortisol responses to the NFR assessment were associated with increased pain reports, and increased NFR threshold.
During the naltrexone session, less consistent correlations were found between pain measures and cortisol concentra-tions, with significant correlations found between cortisol levels obtained immediately after the NFR assessment and MPQ score (r > 0.23, p < .05). No significant correlations were found with NFR measures (r < 0.22, p > .07), or between cortisol concentrations and pain threshold or tolerance (r < 0.14, p > .19).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Differences between high- and low-risk groups in the cortisol response pattern may reflect differential sensitivity to the naltrexone dose required to produce effective endogenous opioid blockade. Because hypertensives and those at increased risk for the disorder exhibit greater endogenous opiate activity (26), it is possible that a 50-mg dose of naltrexone provides a more effective and earlier blockade of endogenous opiate activity in those at low versus high risk. Previous research has demonstrated that individuals with low resting systolic blood pressure exhibit relatively exaggerated epinephrine and cortisol responses to stress under opiate blockade (13), suggesting that the normal inhibitory effects of endogenous opioids on HPA responses may be disrupted in individuals at high risk for hypertension (27,28). Our findings are consistent with this notion.
In addition to the observed effects on cortisol responses, the present study showed that endogenous opioid blockade was associated with increased pain reports. Although these findings are physiologically consistent with the functions of the endogenous opioid system, they are inconsistent with previous work showing either no effect or an effect in the opposite direction after consumption of the same dose of naltrexone (22). It should be noted that in the earlier study, we used cold pressor and heat pain induction procedures, and it is possible that qualitative and physiological differences in pain induction procedures may have contributed to the different results. For example, if NFR assessment is associated with greater apprehension and aversion relative to the cold pressor, this may elicit greater endogenous opioid activity (29). Similar qualities have been associated with increased chances of demonstrating effects of opioid blockade on pain sensitivity in animal studies (30,31). Unfortunately, we did not collect measures of state distress and anxiety after each assessment, and future studies should carefully document mood changes and compare effects of opioid blockade across multiple pain assessment procedures.
Results from this study add to accumulating research demonstrating functional differences between men and women in endogenous opioid activity (32,33). Although we did not ad-dress underlying mechanisms to explain the different patterns of cortisol and pain results in men and women and in high and low hypertension risk groups, we speculate, based on available research, that gender differences may be in part the result of different effects of steroids on central opiate receptors (34–36) and to differences in the role of the opioid system in the stress response (37). For example, recent studies have demonstrated sex differences in the activation of mu-opioid receptors with men exhibiting enhanced activation in response to sustained pain (38), whereas women exhibit increased mu-opioid binding during resting conditions (36). These observations suggest that sensitivity, quantity, and ratio of the different classes of opioid receptors may differ between males and females.
It is important to note that our findings are limited by the use of a standard dose of naltrexone without adjusting for differences in body weight, especially between men and women. Furthermore, only cortisol was measured, leaving open the question about the extent to which the results were because of central activation at the hypothalamic-pituitary levels or to differential effect on the adrenal cortex. In future studies, a more detailed examination of the HPA and hemodynamic measures may provide a more inclusive profile of the effects of endogenous blockade in this group. Assessment of plasma levels of naltrexone should also be considered to determine differences in metabolic rate between men and women.
In summary, the results of this study confirm the inhibitory effects of the endogenous opiate system on cortisol production and pain sensitivity. They indicate differential time-related dynamics in how the endogenous opioid system regulates the hypothalamic-pituitary-adrenocortical functions in those at high and low risk for hypertension.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This study was supported in part by an NIH grant HL64794. During this study, the first author was also funded by NIH CA88272 and by an American Heart Association Grant-in-Aid support (Northland affiliate).
DOI:10.1097/01.psy.0000203240.64965.bd
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
