| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Departments of Oral Medicine (J.J.S., L.L., K.H.H., J.C.S., S.F.D.), Dental Public Health Sciences (L.A.M.), and Psychiatry and Behavioral Sciences (S.F.D.), University of Washington, Seattle, Washington.
Address correspondence and reprint requests to Jeffrey J. Sherman, PhD, Box 356570, Department of Oral Medicine, University of Washington, Seattle, WA 98195-6370. E-mail: jeffreys{at}u.washington.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Fifty-six women with TMD and 59 pain-free controls underwent three experimental pain procedures, including palpation at fixed amounts of pressure, pressure pain thresholds, and an ischemic pain task. Levels of depression and somatization were assessed using the Research Diagnostic Criteria for TMD. Multiple regression analyses were performed to determine the extent to which depression and somatization were associated with experimental pain response.
RESULTS: After controlling for characteristic pain intensity and depression, somatization explained a significant proportion of variance in numbers of masticatory sites rated as painful (R2 change = 6.7%, p = .046) with the full model explaining 16.4% of the variance (p = .024). This did not meet an adjusted level of statistical significance (p = .008). After controlling for characteristic pain, only depression added significantly to the model predicting ischemic pain threshold and tolerance. The full models including characteristic pain and depression explained 49% and 20% of the variance in ischemic pain threshold and tolerance, respectively.
CONCLUSIONS: These findings suggest that depression and somatization are associated with different measures of experimental pain. Somatization may be related to more attentional and perceptual measures of clinically relevant pain while depression may be related to more behavioral measures of pain.
Key Words: pain, • depression, • somatization, • threshold, • tolerance.
Abbreviations: TMD = temporomandibular disorders;; TMJ = temporomandibular joint;; RDC = research diagnostic criteria;; MTPS = Manual Tender Point Survey;; CPI = characteristic pain intensity;; IPT = ischemic pain threshold;; IPTol = ischemic pain tolerance;; IPS = ischemic pain sensitivity;; PPT = pressure pain threshold;; Kpa/s = kilo-Pascals per second.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Patients with TMD have lower pain thresholds and greater sensitivity to experimental pain when compared with age- and gender-matched pain-free controls (8,9). Numerous explanations such as central sensitization (8), neuroendocrine (10–12), and autonomic dysregulation (13) have been offered to explain these differences, but more psychological explanations have received less attention. TMD shares common behavioral and psychological characteristics with other chronic pain conditions (14–16). Depression and somatization in particular have been shown to influence the expression of signs and symptoms of TMD (17) as well as treatment outcome (18,19). Furthermore, up to 50% of those with chronic TMD may have a diagnosable Somatoform Disorder and 33% may have an Affective Disorder (20). These same psychological characteristics may be associated with experimental pain response.
Somatization is defined as the tendency to experience numerous physical symptoms for which no apparent organic cause can be determined. Alternatively, if a medical cause is present, somatization is said to occur when complaints about the bodily disturbance and dysfunction are in excess of the pathology (21). Barsky (22) suggests that somatization is related to an amplification of bodily signals. Individuals high in somatization are believed to perceive exogenous and autonomic physical sensations as more intense. In support, Rief et al. (23) showed that those high in somatization interpret relatively minor autonomic sensations in a catastrophic manner and more recently, Rief et al. (24) suggested that people high in somatization demonstrate impaired heart rate deceleration and report feeling more tense in response to a cognitive/attentional stressor compared with healthy controls. Controls habituate to stress whereas individuals high in somatization have an impaired central nervous system ability to regulate the physiologic and psychological consequences of stress. In the case of TMD, we have found it useful to distinguish between this level of physiologic arousal, perception of the physical intensity of pain, and appraisal of its cognitive and emotional meaning (17,25).
Several studies have found a relationship between somatization and clinical pain measures in chronic pain populations. Specifically, patients with high levels of somatization report pain that is more diffuse and that does not follow anatomic distributions when compared with pain patients with lower levels of somatization (26,27). In TMD patients, somatization is related to more widely dispersed pain as indicated by self-report and as elicited by palpations during a clinical examination (17,18). Although the relationship between somatization and clinical pain reports has been established, the relationship between somatization and laboratory pain perception has not been thoroughly explored.
As with somatization, the relationship between depression and clinical pain suggests an association between depression and onset of some chronic pain conditions (28), numbers of reported pains (29), and treatment outcome (30). In contrast, the relationship between depression and experimental pain perception has resulted in discrepant findings. Some studies suggest that depressed subjects have reduced ischemic pain tolerances and greater ischemic pain intensity (31), whereas others using different experimental stimuli indicate that depressed subjects have higher pain thresholds and tolerances (32–37). Recently, Lautenbacher et al. (36) examined clinical pain reports and experimental pain sensitivity in patients with major depressive disorder and panic disorder and found an interesting divergence between clinical pain reports and laboratory pain sensitivity. Both patient groups experienced more frequent and intense clinical pain complaints when compared with healthy controls. However, despite greater clinical pain complaints, depressed patients had higher pressure pain threshold than healthy controls and those with panic disorder. Depressed patients also had higher cold-pressor pain thresholds than those with panic disorder. The discrepant findings in the experimental pain literature have been attributed to differences in methods of pain stimulation, inadequate reporting of comorbid physical illnesses in the subjects studied and in particular, whether or not depressed patients were experiencing clinical pain (33). The latter limitation is particularly salient because patients with chronic pain conditions typically have altered experimental pain perception and are more likely to be depressed.
The discrepant findings in experimental pain studies for those who are depressed, the lack of similar studies on those high in somatization and the lack of experimental research conducted on those with a chronic pain condition highlight the need to study these conditions in an experimental pain setting. The aims of the present study were to examine the relationships of somatization and depression with perception of and response to three different pain stimuli in TMD patients after controlling for levels of clinical pain. To our knowledge, the relationships of somatization and depression with experimental pain measures have not been studied simultaneously and have not been studied in the context of a chronic pain condition. We hypothesized that in chronic TMD pain patients, somatization would be associated with alterations in pain perception (eg, pain sensitivity) and that depression would be more closely related to behavioral avoidance measures of pain (eg, pain tolerance times).
|
METHOD |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Twenty-nine women with TMD and 28 female controls were taking low-dose combination oral contraceptives. The remainder of subjects used no exogenous hormones. Comparisons between those taking and not taking oral contraceptives revealed that those taking oral contraceptives were younger than the normally cycling women (25.1 vs. 30.9 years, t(53) = 3.33, p = .002), but there were no other differences on demographic, mood, or pain measures. Experimental sessions were equally distributed across the four phases of the menstrual cycle (menses, ovulatory, mid-luteal, late-luteal). These phases were determined based on self-report of onset of menstruation and, in women not taking oral contraceptives, luteinizing hormone surge confirmed using the ClearPlan Easy ovulation prediction kit. No statistically significant phase differences were found on mood, TMD pain intensity, or laboratory pain measures.
Subjects were contacted by telephone and the study was explained. Interested subjects were screened for inclusion and exclusion criteria and subjects meeting criteria were scheduled for a baseline appointment. At the baseline appointment, written informed consent was obtained and the subject was examined by a reliable, calibrated dental hygienist using the examination procedures of the RDC/TMD (2) to confirm that subjects met inclusion criteria for myofascial pain and arthralgia/arthritis. Subjects completed a baseline questionnaire including information on demographics, TMD pain and related symptoms, history of other pain problems, depression, anxiety, and somatization.
Experimental Procedures
During the experimental session participants underwent three consecutive pain-inducing procedures:
1. Palpation pain. Standardized digital pressure was applied to diagnostically relevant anatomic sites, according to RDC/TMD Axis I specifications and Manual Tender Point Survey (MTPS) for fibromyalgia specifications (38). A Somedic Type II Algometer (Solentuna, Sweden) was used to deliver pressure to sites on the head, face, neck, back, and arms. The pressure was applied using a probe diameter of 1 cm. Site location and amount of pressure delivered corresponded to the recommendations from the RDC/TMD and the MTPS. The sites and pressure delivered included bilateral posterior, middle and anterior temporalis, superior, middle and inferior masseter at two pounds of pressure, occiput, trapezius, supraspinatus and lateral epicondyle sites at 8.8 pounds of pressure. Although there is considerable overlap between TMD and fibromyalgia (39), we did not conduct a diagnostic examination for fibromyalgia. We used the MTPS sites (38) in our experiment because they are replicable sites, peripheral to the face, and tested with significantly greater pressure than that used in the RDC/TMD examination.
2. Pressure pain thresholds. Pain thresholds at RDC/TMD and MTPS sites were assessed using the Somedic algometer on 5 bilateral sites corresponding to the anterior temporalis, middle masseter, occiput, trapezius, and supraspinatus. Examiners delivered continuously increasing pressure at the rate of 40 kPa/s until the subject pushed a switch and said "stop" to indicate the point at which "the pressure turned to pain."
3. Ischemic pain. Subjects then rested for 10 minutes before beginning the submaximal effort tourniquet procedure. The nondominant arm was raised for 20s to promote venous drainage. The arm was then occluded with a cuff placed on the upper arm inflated to 230 mm Hg (Hokanson E20 Rapid Cuff Inflator with the AG101 Cuff Inflator Air Source; Hokanson, Bellevue, WA) and lowered. The subject then performed 20 hand grip exercises at 30% of her maximum force of grip. The duration of each squeeze was 2 seconds with an intersqueeze interval of 2 seconds. The tourniquet was maintained for a maximum of 20 minutes. Times to reach ischemic pain threshold and tolerance were recorded and subjects then rated their ischemic pain intensity.
Interrater Reliability
There were a total of 7 examiners in the study. The examiners were trained at the start of the study and recalibrated three times during the study. Overall, the examiners had fair to good agreement on palpation, pressure pain, and pressure threshold measures. The intraclass correlation coefficient ranged from 0.48 to 0.74 for sites painful to palpation, from 0.42 to 0.81 for VAS ratings of palpation pressure pain, and from 0.53 to 0.76 for pressure pain thresholds.
Baseline Measures
In addition to demographic and health history information, subjects completed the RDC/TMD Axis II self-report measures (2). These include the Graded Chronic Pain Scale (40) and measures of depression and somatization. The GCPS includes measures of characteristic pain intensity (CPI), pain-related interference and number of disability days. CPI is the average of three 0 to 10 numeric rating scales asking subjects to report on average, worst, and present facial pain levels. Pain interference is the average of three 0 to 10 numeric rating scales asking subjects to report on interference with daily, recreational/social/family and work-related activities.
Somatization and depression were assessed using SCL-90 items as described in the RDC/TMD (2). Subjects indicated the extent to which they had been distressed by specific symptoms in the past month, on a 0 to 4 scale. The depression and somatization subscales have been widely used and validated with TMD pain patients (2,17). The somatization subscale of the RDC/TMD Axis II includes 5 pain-related items (e.g., headache, pains in heart or chest, pain in lower back) and 7 nonpain items (e.g., dizziness, hot/cold spells, numbness/tingling in parts of the body). Subjects report their level of distress arising from these 12 physical symptoms and their scores are classified into three groups corresponding to Normal, Moderate and Severe categories according to RDC/TMD criteria (2). Cronbach alpha levels for this sample were 0.78 and 0.91 for the somatization and depression scales, respectively.
Normative data defining these cutoff scores are derived from population based studies with more than 1000 subjects (2). We have previously shown that these scales have high reliability, concurrent and construct validity, and clinical utility when compared with other measures of psychological status and psychosocial functioning (17). For example, we have shown that subjects with severe levels of somatization report more painful muscle and placebo palpation sites on clinical examination (17). Further, the RDC/TMD depression scale is closely associated with other measures of depression (eg, Beck Depression Inventory (41), Center for Epidemiologic Studies-Depression Scale (42)) often used in chronic pain settings (17).
Experimental Session Measures
Current TMD Pain and Psychological Distress
Facial pain before experimental testing was measured using a VAS scale of pain intensity anchored with the terms "no pain" and "worst possible pain." Current psychological distress was measured with VAS scales for anger, fear, sadness, emotional upset, and stress taken from the Emotion Assessment Scale (43). This scale has been shown to have adequate reliability and utility for the measurement of psychological distress in chronic pain patients (43–45).
Palpations
Examiners delivered a fixed amount of palpation pressure at a standardized rate. Subjects reported whether they experienced pain at that fixed amount of pressure and, if so, rated their pain using a 0 to 100 VAS.
Pressure/Pain Thresholds
Examiners delivered increasing pressure at a rate of 40 kPa/s. Subjects reported the point at which they "felt the pressure turn to pain" by pressing a patient controlled switch and saying "stop."
Ischemic Pain
The point at which subjects first began to feel pain in their hand or forearm (ischemic pain threshold, IPT), the point at which participants could no longer tolerate ischemic pain (ischemic pain tolerance, IPTol), and a 0 to 100 verbal rating of perceived pain sensitivity (IPS) were measured in response to the tourniquet task.
Data Reduction and Analyses
Independent samples t tests were used to compare differences between TMD and control groups for demographic characteristics, and baseline and day of testing psychological characteristics. Analysis of variance and 
2 procedures were used to compare somatization groups (Normal, Moderate, Severe) at baseline. Two composite measures were created for palpations: (1) masticatory muscle sites (bilateral anterior, middle, and posterior temporalis, superior, middle, and inferior masseter); and (2) fibromyalgia sites (bilateral occiput, supraspinatus, trapezius, lateral epicondyle). Regression analyses were run for the number of palpations rated as painful and a mean pain severity rating for each of these composite measures of palpation pain. Similarly, two composite measures were also created for mean pressure pain threshold (PPT) at: (1) masticatory muscle sites (bilateral masseter and temporalis) and (2) fibromyalgia sites (bilateral occiput, supraspinatus, trapezius). We examined the extent of association between depression and somatization and response to palpations at fixed amounts of pressure, pressure pain thresholds, and ischemic pain response (IPS, IPT, IPTol) after controlling for characteristic pain intensity by performing regression analyses. Characteristic pain intensity, depression, and somatization were entered sequentially into the regression models for subjects with pain. Regression models were also run for pain-free control subjects with only depression and somatization entered sequentially. Since we conducted nine regression analyses, we used the Holm multiple comparison procedure (46) to control for a Type I error. We examined scatterplots of standardized residuals against standardized predicted values to check for linearity assumptions and equality of variances and examined histograms to check for outliers and normality assumptions. The distributions for ischemic pain tolerance and threshold were slightly skewed; therefore, we used a square root transformation on these variables in all analyses. The results were nearly identical to the nontransformed results so we present the nontransformed results for ease of interpretation.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Table 1 shows the baseline demographic characteristics for subjects with TMD pain classified into Normal, Moderate, and Severe somatization subgroups based on RDC/TMD (2) criteria. There were no statistically significant differences across the groups for age, number of years of education, income, race, or Hispanic ethnicity. There were significant baseline differences for depression (F(2,53) = 6.9, p = .01) with the Severe somatization group having higher levels depression than either the Normal or Moderate groups (Tukey Studentized Range test, p = .05). Characteristic pain intensity differences approached significance (p = .057), but there were no differences in interference scores or pain duration across subgroups. Bivariate correlations between somatization and depression (r(54) = 0.61, p = .01), somatization and CPI (r(54) = 0.37, p = .01), and depression and CPI (r(54) = 0.37, p = .01), were all significant. Although these correlations are significant, Tabachnick and Fidell (47) suggest that multicollinearity is only a problem if the Pearson correlation between two variables is r > 0.70. Further, observations of scatterplots of standardized residuals against standardized predicted values demonstrated that the assumption of linearity was met. Table 2 shows pain and emotional distress levels on the day of testing. There were no statistically significant differences across the groups for current facial pain intensity or any of the psychological distress measures except for current anger levels (F(2,53) = 3.8, p = .03), with Severe somatizers reporting more anger than the Moderate group. When compared with pain-free controls, subjects with TMD reported greater fear (mean = 5.1 vs. 1.4, t(113) = 2.9, p = .004) and upset (16.2 vs. 6.3, t(113) = 2.7, p = .009) on the day of the session.
|
|
Experimental Pain Measures
Table 3 shows the results of the regression analyses for palpation pain measures in the TMD group. The full model predicting numbers of masticatory muscle sites rated as painful explained 16.4% of the variance (R2 = 0.164, F(3,52) = 3.4, p = .024) but was no longer statistically significant after the Holm adjustment for multiple comparisons. In the full model predicting numbers of masticatory sites rated as painful, characteristic pain intensity explained 6.4% of the variance (p = .06). Adding depression led to a model that explained 9.7% of the variance, a nonsignificant change in F (p = .17). Adding somatization led to a model that explained 16.4% of the variance in numbers of masticatory sites rated as painful (R2 change = 6.7, p = .046). In the model predicting severity of pain related to masticatory muscle palpation, characteristic pain intensity explained 12.3% of the variance (p = .008). Adding depression yielded a model that explained 18.9% of the variance, a significant R2 change (p = .043). Somatization added a marginally significant R2 change (p = .1) and resulted in a model that explained 23% of the variance. The full models for number of painful fibromyalgia sites and severity of pain related at fibromyalgia sites were nonsignificant.
|
Table 4 shows the results of the regression analyses on pressure pain thresholds and the ischemic pain task measures for TMD subjects. The full models predicting pressure pain thresholds on masticatory muscle and fibromyalgia sites were all nonsignificant. On the ischemic pain task, characteristic pain intensity explained 10.8% of the variance (p = .013) for ischemic pain sensitivity. Adding depression and somatization led to a model that explained 12.9% of the variance, a nonsignificant F change. For time to pain threshold, characteristic pain intensity explained 13.1% of the variance (p = .006). Adding depression resulted in a model that explained 23.6% of the variance, a significant R2 change (p = .009). Somatization did not add to the predictive model for pain threshold. Similarly, for ischemic pain tolerance, characteristic pain intensity explained 15.4% of the variance (p = .003). Adding depression led to a model that explained 20% of the variance, a marginally significant R2 change (p = .089). Somatization did not add to the predictive model for tolerance to ischemic pain.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Barsky and Wyshak (48) suggest that selective attention, amplification, and misinterpretation of bodily signals are the main processes in the development of somatization. We found that somatization was associated with pain sensitivity to palpation at a fixed amount of pressure in masticatory areas but not other areas of the body or ischemic pain. Attentional processes directing focus to the face but not other sites of the body may result in greater regional but not widespread laboratory pain sensitivity. The perceived threat associated with palpation to the face rather than palpation to other body sites is likely far greater in those with TMD and high somatization in contrast to those low in somatization.
In addition to attentional and appraisal processes, somatization is also associated with behavioral responses such as increased complaints and concerns about symptoms leading to increased utilization of medical care (21). Ischemic pain threshold and tolerance, the amount of time until a subject perceives pain from the ischemic task and the amount of time that a subject chooses to withstand the tourniquet task, can be viewed as behavioral response characteristics. Although we hypothesized that somatization would be associated with ischemic pain threshold and tolerance, no significant relationship was observed after controlling for depression. We found significant and negative correlations between characteristic pain intensity and threshold and tolerance times such that greater characteristic pain intensity resulted in lowered times to pain threshold and tolerance. However, after controlling for characteristic pain, we saw a positive and significant relationship between depression and ischemic pain threshold and tolerance. Those with higher levels of depression took longer to report onset of pain and withstood the ischemic pain task for a longer period of time. These data are consistent with the preponderance of studies finding higher pain thresholds for brief noxious thermal and electrical stimulation in depressed patients compared with controls (33,37).
In contrast to our findings of a positive association between depression and ischemic pain tolerance and thresholds times, Pinerua-Shuhaibar et al. (31) found lowered ischemic pain tolerance and no differences in ischemic pain threshold when comparing those with depressive disorders to normal control subjects. There were significant design differences between that study and our own that might explain the discrepancy. These authors compared a sample of dysthymic subjects diagnosed using DSM-III criteria that were predominantly female to a group of nondepressed controls who were predominantly male. Our own sample included only females, and we did not assess for depression using diagnostic criteria. Further, these authors defined ischemic pain tolerance by measuring the time elapsed from the first pain perception (ischemic pain threshold) until the subject could no longer tolerate the task or at 24 minutes. We defined ischemic pain tolerance times as do others (8,49,50) as the total time subjects tolerated the task following commencement of handgrip exercises or at 20 minutes.
Several psychological mechanisms have been suggested to explain the association between depression and laboratory pain perception, threshold, and tolerance times. Depressed patients may have a more stoic or affectively indifferent response to pain induction procedures (51) and may have slower reaction times or volitional impairments affecting reaction times (32). We suggest that the attentional, appraisal, and volitional processes underlying both depression and somatization may be influential contributors to experimental pain response and chronic pain presentation. There is considerable comorbidity between the clinical characteristics of affective and somatoform disorders with many of the criteria associated with depression involving bodily symptoms, behavioral avoidance, and appraisal of events as threatening. Our findings suggest that while many of the clinical characteristics of depression and somatization overlap, the effects of each condition on laboratory pain responses may differ. High somatizers who may be highly vigilant for internal bodily cues that can be appraised as threatening may attend more closely, appraise more negatively, and perceive more pain in response to fixed palpations pressure to an area of the body that is vulnerable. Depression may influence experimental pain response at a more behavioral and volitional level and at a less bodily specific level. These underlying processes may be influential in the presentation of the chronic pain patient and in designing treatment for chronic pain patients. Matching behavioral treatments to psychosocial profiles has been shown to be successful in TMD patients (52–54), and it is possible that treatment effects for a depressed patient could be enhanced by targeting passive behavioral responses to pain while treatment for a somatically focused patient could be enhanced by training patients in cognitive, behavioral, and self-regulation exercises to reduce dysfunctional attentional, appraisal, and perceptual tendencies.
Furthermore, these relationships may be most meaningful in the context of chronic pain. We tested these relationships in a pain-free control group, but all models testing the relationships between depression, somatization, and experimental pain were nonsignificant. The absence of a significant association in pain-free controls could be due to the more restricted range in depression and somatization in pain-free controls but could also suggest that the attentional, appraisal, and behavioral response characteristics influencing experimental pain are most salient in the presence of a chronic pain condition. Further research with subjects diagnosed with a somatoform disorder or a depressive disorder and without a primary chronic pain complaint could help in elucidating why these relationships were seen in chronic pain patients but not in pain-free controls.
Several researchers have shown that, when compared with pain-free controls, subjects with TMD have lowered pressure pain thresholds (8,55), lowered ischemic pain tolerance (8), lowered forearm heat pain thresholds (56), decreased responsiveness to electrical stimulus (57), and increased pain sensitivity to tonic saline infusion into the masseter (55). Physiological explanations for these differences include sensitization of peripheral nociceptors in the affected area (58), central sensitization (8), autonomic dysregulation (13), and alterations in opioidergic pain inhibitory processes (49). An alternative, although not mutually exclusive, explanation for differences in pain perception between those with TMD and their pain-free counterparts is that psychological differences such as depression and somatization result in altered pain perception and behavioral response to pain. Further, it is likely that there is considerable shared underlying pathophysiology between chronic pain conditions and psychiatric conditions such as depression and somatization.
This study has several limitations. First, while the RDC/TMD scales of depression and somatization have considerable data in support of reliability, validity, and clinical utility (17), they merely provide an assessment of clinical characteristics and are not diagnostic. Further studies using clinical samples diagnosed with Affective and Somatoform Disorders would elucidate the relationship between these clinical conditions and experimental pain. Our data are based on subjects with relatively low levels of depression, somatization, and clinical pain reports when compared with those seeking treatment for these conditions. Using samples with clinical diagnoses may produce more robust effects. Second, extrapolation of our laboratory pain findings to clinical pain presentation should be done cautiously. There are problems comparing studies using experimental pain stimuli with those involving clinical pain. The psychosocial context is different between experimental and clinical pain and could influence subjects’ response. Subjects understood the transitory and nonharmful nature of the laboratory pain stimuli in contrast with their experience of intransigent pain of perhaps unclear etiology. Third, our choice of TMD sample was restricted to women of childbearing age with RDC/TMD Axis I diagnoses of myofascial pain and arthralgia. Our choice of inclusion criteria was based on prevalence rates and the most common clinical presentation of TMD symptoms. TMD is 1.5 to 2 times more prevalent in women than in men in the community, and 80% of TMD treatment seekers are women (29,59). Prevalence rates for both somatization and depression are higher in women as well (21). Some evidence suggests that TMD patients with myofascial pain are more responsive to experimental pain (55) and that psychosocial profiles and treatment seeking in TMD differs by type of RDC/TMD physical diagnosis such that those with myofascial pain are more likely to seek treatment and have elevated levels of depression and somatization (60,61). It is possible that our findings are limited to a TMD population with myofascial pain as pain is a primary criterion for that TMD diagnosis and pain is a primary criterion for diagnosis of somatization. Some TMD diagnoses (e.g., Disc Displacement) do not include pain as a diagnostic criterion, and it is certainly possible that the observed relationships would be different in such TMD populations.
After reviewing the discrepant results in studies measuring the impact of depression on experimental pain perception, Dickens et al. (33) concluded that depressed subjects have higher pain perception thresholds in response to experimental pain stimuli. They noted that the extant research is limited by studies using too few pain modalities, not measuring pain tolerance, and not assessing other conditions of interest. This study assessed the relationships of somatization and depression with experimental pain perception, threshold, and tolerance using three different pain modalities and did so in a group of subjects with chronic TMD pain. We found independent and distinct relationships for depression and somatization with different measures of experimental pain. We suggest that future studies using experimental pain paradigms in chronic pain populations consider the effects of both depression and somatization.
Received for publication November 10, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
