Pain threshold reflects psychological traits in patients with chronic pain: a cross-sectional study

The participants were selected from 81 patients with chronic pain who had been admitted to the department of Psychosomatic Medicine of Kansai Medical University. Based on previous studies [40‐42], the participants were diagnosed with nonmalignant chronic pain that had persisted for three or more months by attending physicians with clinical experience in treating chronic pain. The staff members of the Department of Psychosomatic Medicine are physicians, not psychiatrists, and the chief complaints of almost all patients are their physical symptoms. Patients with any of the following criteria were excluded: (1) an age of 18 years or less, (2) extensive peripheral neuropathy, (3) pain in the non-dominant hand, (4) opioid use, or (5) a diagnosis of major depression, schizophrenia, or dementia. After exclusions, the data of 57 patients with chronic pain was available for analysis. In accordance with the study protocol approved by the ethics committee of Kansai Medical University Hospital, written informed consent was obtained from the participants.

In this study, the QST was performed with a Neurometer CPT (Neurotron, Incorporated, Towson, MD, USA). With this device, an electrical stimulus of either 250 or 5 Hz was selectively applied to Aδ primary afferent fibers, which transmit sharp pain, and C primary afferent fibers, which transmit dull pain. The QST was performed in a quiet room at a comfortable temperature. The participants sat on a chair and a stimulating electrode was attached around the distal interphalangeal joint of the fourth finger of the non-dominant hand. The electrical stimulation current was increased from 0 to 9.99 mA at a set rate while the button on the device was pressed, and the current stopped when the button was released. After the patients were informed that stimulation with the highest electrical current would not cause tissue or other damage to the body, the participants then operated the device by themselves. They were instructed to release the button when the stimulus reached an unbearable level of pain, and this was defined as the PTT. The QST was performed once for each level of stimulation, and measurements were taken every 30 s. Previous studies have shown only a small level of variation among measurements [43‐45].

Subjective assessments of the pain were performed with the Short-Form McGill Pain Questionnaire (SF-MPQ). The participants evaluated the pain with 15 expressions that described the sensations and the emotions of pain, which were rated on a 4-point scale, while the severity of the pain was evaluated with both a visual analog scale (VAS) and the Present Pain Intensity 6-point scale. The reliability of the Japanese version of the SF-MPQ has been confirmed [46].

The MMPI questionnaires were distributed to the participants before the experiment and then collected at the time of the PTT measurement. The MMPI, which is a self-reported questionnaire on personality, is highly reliable for less invasively evaluating psychological traits from various perspectives. The MMPI consists of four validity scales (cannot say, lie, infrequency, and defensiveness) and ten clinical scales: Hypochondriasis (Hs), Depression (D), Hysteria (Hy), Psychopathic deviate, Masculinity/femininity, Paranoia, Psychasthenia, Schizophrenia, Hypomania, and Social introversion that are assessed with 550 questions that are answered on a 3 point scale (agree, disagree, and neither). The scores are calculated by assigning two points to agree and one point to neither. Higher scores indicate a greater tendency for that trait. The results are expressed as numerical values and profile forms [29, 30].

In Japan, the MMPI has been widely used in the clinical and academic fields [47‐49]. In the United States, the transition to the MMPI-2 has already been completed. The basic scales of the MMPI are compatible with those of the MMPI-2, and the items considered important in the clinical and academic fields are common between the two versions. Thus, the assessment results are considered similar for the MMPI and MMPI-2.

All values are presented as mean ± standard deviation (SD). The statistical analyses were performed on the data of the 57 participants (22 male and 35 female). According to the scatter diagrams of the PTTs to the 250-Hz and 5-Hz stimuli (Fig. 1), the PTTs were not distributed evenly among the patients with chronic pain. Therefore, a cluster analysis (Ward’s method) was performed with PTTs as a variable of interest to extract a characteristic group. Inter group comparisons were performed with a t-test, one-way analysis of variance (ANOVA), χ² test, or residual analysis. The SPSS software program (version 11.5, IBM Corporation, Armonk, NY, USA) was used for the analyses.

Of the participants, 61.4% were female. The mean age was 47.7 years, with a SD of 17.5 years. The mean pain duration was 59.3 months, with a SD of 75.0 months (Table 1). The main diagnosis was Chronic pain (40%), followed by Functional dyspepsia (10%), Fibromyalgia syndrome (9%), and Premenstrual syndrome (9%). The main sites of pain were the upper and low back (26%), the lower extremities (23%), the abdomen (16%), and the neck (14%) (Table 2). The SF-MPQ scores were 15.30 ± 7.78 for the sensory components and 6.00 ± 3.60 for the affective components, while current pain intensity was rated as 3.38 ± 1.21 on a six point scale ranging from 0 (no pain) to 5 (unbearable pain). The severity of pain at the injured site in the past week was rated as 6.71 ± 2.50 cm on a 10-cm VAS (Table 1).

The PTTs to 250-Hz stimulus ranged from 0.80 mA to 9.99 mA, with a mean ± SD of 4.59 ± 3.00 mA. The PTTs to 5-Hz stimulus ranged from 0.75 mA to 9.99 mA, with a mean ± SD of 4.23 ± 2.93 mA.

The mean ± SD PTTs to 250-Hz stimulus were 5.74 ± 3.19 mA for male and 3.87 ± 2.67 mA for female participants, while those to 5-Hz stimulus were 5.53 ± 3.19 mA and 3.41 ± 2.47 mA, respectively. The PTTs to both stimuli were significantly lower for female participants (250 Hz, p < 0.05; 5 Hz, p < 0.01).

The cluster analysis done to create the High-Sensitivity group selected 23 participants with decreased PTTs (n = 23) (Fig. 1). The PTTs ranged from a minimum of 0.75 mA to a maximum of 3.00 mA. In contrast, the PTTs of the remaining participants were distributed in a wide range, from a minimum of 1.60 mA to a maximum of 9.99 mA. Since we thought it was not appropriate to describe the rest as a group with identical characteristics, this group was named Others (n = 34).

Of the participants in the High-Sensitivity group 78.2% were female, as were 50.0% of the participants in the Others group. A χ² test showed that the male-to-female ratio did not differ between the groups. The mean ± SD ages were 56.30 ± 15.83 years in the High-Sensitivity group and 41.94 ± 16.35 years in the Others group. The mean ± SD pain durations were 40.60 ± 76.42 months in the High-Sensitivity group and 73.23 ± 81.90 months in the Others group. The t-tests showed that neither age nor pain duration differed between the groups. In addition, t-tests showed that the scores on the sensitive and affective components of the SF-MPQ, VAS scores, and current pain ratings did not differ between the groups (Table 3).

In the MMPI profiles, both groups showed high values on the Hs, D, and Hy, scales (t ≥ 70), which are typical profiles for patients with chronic pain (Fig. 2). Patients were classified by three patterns of these three scales: Conversion V pattern, in which the scores on the Hs and Hy scales were higher than the score on the D scale by 10 or more points; the Neurotic Triad pattern, in which the scores on the Hs and Hy scales were lower than the score on the D scale; and the others patterns. The High-Sensitivity group contained no participants with the Conversion V pattern, 11 with the Neurotic Triad pattern, and 12 with other patterns, whereas the Others group contained 17, 7, and 10 participants in each of these categories, respectively (Table 4). A χ² test (Yates’ correction) showed that the distribution of these patterns differed significantly between the groups. The residual analyses showed that the proportion of participants with the Conversion V pattern was significantly lower in the High-Sensitivity group, while the Neurotic Triad pattern was significantly higher (p < 0.01 for both). No significant difference in the proportions of the participants with the other patterns was observed.

Moreover, the scores on the MMPI clinical scales were compared with a one-way ANOVA. The scores on the Hs and Hy scales were significantly lower in the High-Sensitivity group than in the Others group (p < 0.05 for both) (Table 5).

In this study, patients with chronic pain were classified using cluster analysis of the PTT at non-injured sites and their psychological traits were evaluated. As a result, we identified a characteristic High-Sensitivity group. Several previous studies done to classify patients with chronic pain have suggested a relationship between pain and psychological traits. Murphy et al. showed that there were multiple detectable subgroups of patients with chronic pain by cluster analysis of clinical pain intensity and psychological variables [53]. In addition, Cruz-Almedia et al. classified patients with chronic pain by cluster analysis of psychological variables and clarified the association between psychological characteristics and both clinical pain intensities and pain thresholds [37].

Previous studies were classified according to psychological variables. In this study, we classified patients with chronic pain by cluster analysis with only reproducible PTT as a physical variable to assess the relationship between psychological factors and PTTs. The results strongly supported the study of Cruz-Almedia et al., which showed close relationship among psychological traits, somatosensory sensations, and central sensitization to chronic pain [37]. This shows that patients with chronic pain can be classified by PTT as a physical variable.

This study yielded results indicating an association between classifications based on PTTs and psychological traits. Previous studies that used MMPI for patients with chronic pain found that their participants could be classified into three to six types [54‐56], and the inclusion of the following three patterns is common to all of these classifications: the Conversion V pattern, the Neurotic Triad, and the normal patterns, which show scores within the normal range on all scales. The Conversion V and the Neurotic Triad patterns are known MMPI profiles of patients with chronic pain, and they were found for 35 of the 57 participants observed, representing over 60% of our sample. Furthermore, the High-Sensitivity group in this study included significantly more participants with the Neurotic Triad pattern than did the Others group, but it did not include any participants with the Conversion V pattern. Moreover, the scores on the Hs and Hy scales in the High-Sensitivity group fell on the border between moderate and high scores and were significantly lower than the ones in the Others group. This indicates that the participants in the High-Sensitivity group had different psychological traits from other participants with chronic pain. The lack of any significant group-to-group difference in subjective pain intensity in this study suggested that, even though no apparent differences are detected in the severity of pain reported by patients in clinical practice, their responses to QST might imply their psychological traits. Thus, assessment of pain sensitivity using the QST, which focuses on the close association between pain thresholds and psychological traits, may be particularly useful for prediction of the psychological traits of patients with chronic pain who are resistant to psychological intervention, such as psychological testing.

In this study, analysis of the MMPI showed that the High-Sensitivity group contained significantly more participants with the Neurotic Triad pattern, but none with the Conversion V pattern. Moreover, the scores on the Hs and Hy scales were significantly lower than those of the Others group. The Conversion V pattern relates to a characteristic profile of patients with chronic pain. It shows the tendency to replace psychological problems with physical complaints, and its socially incompatible personality is also known to cause difficulties in treatment. On the other hand, the Neurotic Triad pattern is characterized by depressive tendency and hypochondriac concerns [34]. While people with the Neurotic Triad pattern are introverted and nervous, previous studies have shown that they are less likely to engage in self-harming [29, 30] and that they respond well to multimodal treatment [57]. The Hs scale indicates a hypochondriac tendency, strong health concerns, and catastrophizing. Those who show high scores on the Hs scale tend to associate their normal physical sensations with somatic symptoms. The characteristics of the Hy scale are common to the Conversion V pattern described above. The above suggest that the High-Sensitivity group has quite different psychological aspects than the Others group.

In this study, data on income, social status, and working conditions of the participants were not collected, and further studies with such data will be necessary. As ethnic differences have been suggested to affect assessment of psychological traits when using the MMPI, further studies with multiple ethnic groups are needed. Since several reports have indicated that pain thresholds are affected by hormonal levels during the cycle phases [58, 59] the subjects also need to be classified by sex for analysis in future studies. Many patients with chronic pain have depression, which is known to affect pain sensation; for this reason, these patients were excluded from the study, resulting in a smaller final sample size. Furthermore, the involvement of central sensitization can be analyzed by measuring thresholds multiple times at multiple points in unaffected tissue. This study was a clinical study, thus the results might have been affected by the fact that the participants may not have been completely free from the effects of medication. However, no studies have sufficiently described the effects of drugs, such as nonsteroidal anti-inflammatory drugs and opioids, on QST [60, 61].