ORIGINAL ARTICLE
Relationships Between Psychological Factors, Pain, and Disability in Complex Regional Pain Syndrome and Low Back Pain Debbie J. Bean, MSc,*w Malcolm H. Johnson, MA,w and Robert R. Kydd, PhDw
Objective: Cognitive and emotional factors are known to influence peoples’ pain experiences in many conditions, including low back pain. However, in complex regional pain syndrome (CRPS), their role is unclear. This study aimed to assess the relationships between psychological factors, pain, and disability in CRPS, compared with low back pain. This could help to identify target variables for psychological treatment. Materials and Methods: A total of 88 CRPS patients and 88 low back pain patients completed measures of pain, disability, depression, anxiety, and fear of movement and reinjury (kinesiophobia). Mean scores between the 2 groups were compared, and correlations between psychological factors, pain, and disability were compared between the 2 groups. Predictors of pain and disability were assessed using multiple regression analyses. Results: The 2 groups had remarkably similar scores on measures of pain, disability, depression, anxiety, and kinesiophobia. In both groups, those who were more depressed, anxious, and kinesiophobic were more disabled. For the CRPS group (but not the low back pain group), pain intensity significantly correlated with distress. Multivariate analyses showed that the unique predictors of disability for the 2 groups were pain and depression, and that depression had a stronger relationship with disability for the CRPS group. For both groups, pain intensity was predicted by kinesiophobia, and anxiety was a unique predictor in the CRPS group only. Discussion: In CRPS, disability and pain severity were more strongly associated with psychological factors than they were in low back pain. Cause and effect relationships could not be established by this cross-sectional study. Key Words: complex regional pain syndrome, low back pain, psychological factors, disability
(Clin J Pain 2014;30:647–653)
A
large body of literature has demonstrated that psychological and social factors influence people’s experience of pain in a variety of contexts, including experimental, acute, procedural, and chronic pain.1 The range of psychological factors that influence pain includes mood, cognitions and expectations, behavioral patterns, attention states, and learning history. Although such Received for publication February 20, 2013; revised May 20, 2014 accepted August 2, 2013. From *The Auckland Regional Pain Service (TARPS), Auckland District Health Board; and wDepartment of Psychological Medicine, The University of Auckland, Auckland, New Zealand. The authors declare no conflict of interest. Reprints: Debbie J. Bean, MSc, Department of Psychological Medicine, The University of Auckland, Private Bag 92019, Auckland, 1142, New Zealand (e-mail: [email protected]). Copyright r 2013 by Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 8, August 2014
psychological factors have been demonstrated to influence pain and disability in a range of chronic pain conditions, there have been relatively few studies demonstrating a significant influence of psychological factors in complex regional pain syndrome (CRPS). In fact, a recent systematic review identified no relationship between a number of psychological factors and CRPS type 1, except for a possible association of CRPS type 1 with more life events.2 Another recent literature review found that there was no evidence that personality or psychopathology play a causal role in CRPS; however, it did find that CRPS leads to a sequelae of negative psychological and social consequences.3 Prospective studies that have measured psychological factors among a cohort of fracture, surgery, or stroke patients then assessed patients at a later date for CRPS have reported few psychological risk factors for CRPS development.4–7 There are no published randomized controlled trials of psychological treatments for CRPS, although smaller trials have provided some evidence of the efficacy of exposure-based therapies aimed at reducing patients’ fear of movement or reinjury.8,9 Despite this lack of research, patients with longstanding CRPS are often treated in interdisciplinary pain clinics where psychological therapies are recommended and provided. The 2012 Royal College of Physicians’ guidelines recommends that psychological intervention is one of the 4 main arms of treatment for CRPS.10 Psychologists providing this treatment likely generalize from the research on other chronic pain conditions to identify appropriate target variables for treatment and set treatment goals and activities. One such condition is low back pain (LBP), where a large body of literature has established the importance of psychological factors in predicting pain intensity and heightened disability.11 Systematic reviews have shown that psychological factors such as depression, fear of movement and reinjury (kinesiophobia), catastrophic thinking, and poor expectations predict the progression of LBP from an acute to a chronic timeframe.12–14 Randomized controlled trials have demonstrated the efficacy of behavioral treatments for LBP, at least in the short term.15 There are several possible reasons why previous research has not demonstrated a relationship between psychological factors and CRPS. It might be that no such relationship exists. In contrast, it might be that previous research has suffered from methodological shortcomings, or has not assessed the right relationships. This may in fact be the case, as much of the previous research on psychological factors in CRPS has been based on a dualistic model of health and has sought to find psychological causes for CRPS (as mentioned above), or has focussed on identifying psychopathology or personality disorders among these patients over and above what would be expected of www.clinicalpain.com |
647
Clin J Pain
Bean et al
a sample of people in pain.16,17 This may be because traditionally CRPS was stigmatized as “psychogenic.”18–20 Research from other fields of pain has adopted a biopsychosocial model and suggests that psychological factors are not usually a singular cause of pain, but that unhelpful psychological and physical factors interact to produce heightened pain and disability. In line with this, the small number of studies that have aimed to identify more subtle influences of psychological factors on pain and disability in CRPS have shown some interactions between mood and pain levels,21 anger-expressiveness and pain severity,22 and fear of movement and disability.23,24 A prospective study reported that early increases in depression after knee replacement were associated with CRPS severity at 12 months, suggesting that psychological factors may interact with pain over time, leading to poorer outcome.25 Bruehl et al26 compared CRPS patients with non-CRPS limb pain and LBP patients and found that while there were few differences between the groups on measures of distress, there were stronger relationships between distress and pain severity for the CRPS and limb pain groups than the LBP group. The aim for the present study was to further assess associations between distress and pain in CRPS, and to compare the strength of any associations with a sample of patients with LBP. We aimed to further extend the findings of Bruehl et al26 by also looking at the role of fear of movement as a predictor variable and to look at disability and pain severity as outcome variables. LBP was chosen because in this group previous research demonstrates the importance of psychological factors, and therefore we hypothesized that if there were similarities between CRPS and LBP groups these would be meaningful.
MATERIALS AND METHODS Participants The participants were 88 consecutive CRPS patients and 88 LBP patients seen at The Auckland Regional Pain Service (Interdisciplinary Pain Center) between January 2009 and December 2011. The samples were matched for age, sex, and pain duration and thus the LBP patients were not a consecutive sample. Patients experiencing both CRPS and LBP concurrently were excluded. The interdisciplinary pain center typically sees patients with chronic pain who are referred after multiple treatments and who have seen specialists from other disciplines, and thus both samples are likely to be those severely affected by pain who have been difficult to treat. Both CRPS type 1 and 2 were included in the CRPS sample. Patients in the LBP sample were included if their primary complaint was pain in the low back and/or leg and this included patients with nonspecific back pain and other causes of LBP.
Procedure The study was a retrospective review and analysis of routinely administered psychological scales and medical records. The observational study was approved by the New Zealand Ministry of Health Northern X Ethics Committee. Patients were identified using previous clinic lists that code patients’ pain problems according to pain location. The first author searched the initial assessment reports from all patients coded with upper or lower limb pain to determine whether the patient was diagnosed with CRPS during their medical assessment at the interdisciplinary pain center. CRPS is only diagnosed at the pain center after a
648 | www.clinicalpain.com
Volume 30, Number 8, August 2014
comprehensive assessment and examination by pain specialist doctors, and patient records were checked to ensure that patients met the 1994 IASP criteria for CRPS.27 The first author also searched the records of all patients coded with LBP to determine whether this diagnosis was confirmed during the patient’s comprehensive assessment at the pain center. The following data were extracted from electronic patient records: age, sex, ethnicity, pain duration, condition (CRPS or LBP), CRPS limb, and employment status (working or not working). The following scores were also extracted from questionnaires that patients are required to fill in before receiving an appointment, these questionnaires are scanned and stored electronically on the hospital computer system, and scores were extracted by the first author: 1. Pain Numerical Rating Scale: Patients were asked to rate their average pain intensity on a scale of 0 (no pain) to 10 (worst pain you can imagine). 2. Pain Disability Index (PDI)28: This 7-item scale asks participants to rate their level of disability on a range of life domains. Scores range from 0 (no disability) to 70 (maximum disability). 3. Hospital Anxiety and Depression Scale (HADS)29: This 14-item scale measures symptoms of depression and anxiety, producing 2 separate scores for the separate constructs. The scale is designed to limit the bias of physical health symptoms on depression and anxiety scores. The scale ranges from 0 (very low levels of depression or anxiety) to 21 (very high levels of depression or anxiety). In the original paper, cut-off scores of 8 and above were suggested to identify borderline or possible cases of depression or anxiety, and scores of 11 and above were suggested to identify probable clinical cases of depression and anxiety. 4. Tampa Scale for Kinesiophobia (TSK)30: This is a 17item scale designed to measure kinesiophobia (fear of movement and/or (re)injury). Scores range from 17 (low kinesiophobia) to 68 (maximum kinesiophobia). Given that the data were collected as part of routine clinical care, there was some missing data that is reflected in the varying degrees of freedom reported for each of the statistical tests. There were more LBP patients with full data sets (80/88) compared with CRPS patients (65/88) because when selecting matched controls for the CRPS patients, efforts were made to select those without missing data.
Data Analysis A power analysis was conducted using G-Power.31 This was based on a 2-tailed, independent samples t test, and demonstrated that the study had 90% power to detect a moderate effect size (d = 0.5, a = 0.05), with 86 participants required in each group. Data were entered into SPSS for Windows. Not all variables displayed normal distributions. Thus, nonparametric tests were used, but these produced the same results as parametric tests, so parametric test results are presented here for ease of comprehension. Independent samples t tests and w2 tests were used to explore differences between the CRPS patients and LBP patients on demographic details and psychometric test scores. Pearson correlations were used to assess relationships between pain intensity, disability scores, and scores on psychological variables for each patient group. The strength of the correlations between psychological variables and disability scores were compared between the 2 r
2013 Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 8, August 2014
Psychological Factors in CRPS and Low Back Pain
diagnostic groups by computing Fisher z scores, with these compared statistically using z scores. This was done utilizing the calculator from Preacher.32 Multiple regression analyses were used to determine unique predictors of the 2 dependent variables (pain and disability), and to assess whether the psychological predictors were equivalent for the 2 diagnostic groups. The psychological variables were entered into the regression analyses, along with diagnostic group (dummy coded), and interaction terms for each psychological variable ((psychological variable mean of psychological variable) diagnostic group (dummy coded)), thus the interaction terms were mean-centered. For the multiple regression with pain as the dependent variable, the residuals were not normally distributed (negatively skewed), so a square root transformation was applied to pain scores. This was successful, resulting in normally distributed residuals, so the results from the transformed data are presented.
Demographics and Description of the Sample The demographic characteristics of the 2 samples are described in Table 1. Because the samples were matched, there were no differences between the groups on age, sex, or pain duration. Both samples had a fairly long pain duration of approximately 3½ years. The CRPS patients were significantly less likely to be working, w2(1, N = 172) = 5.49, P = 0.019 compared with the LBP patients. Of the CRPS sample, 66 (75%) had the upper limb affected and 22 (25%) had the lower limb affected. A total of 40 (45%) of the CRPS patients had the left side affected, 42 (48%) had the right side affected, and 6 (7%) had CRPS affecting both sides of the body. Of the CRPS patients, 75 had CRPS type 1 (85%), and 13 had CRPS type 2 (15%). We classified the LBP patients into 3 diagnostic groups according to the guidelines of the American College of Physicians and American Pain Society.33 A total of 72 TABLE 1. Demographic Details and Psychological Test Scores for the CRPS and Low Back Pain Patient Samples
CRPS (N = 88)
LBP (N = 88)
59 45 (12.44) 43.15 (53.73) 20 7.59 (1.77) 44.92 (16.09) 9.99 (4.04) 22.53
59 45 (12.60) 43.42 (53.14) 36* 7.69 (1.58) 43.89 (12.43) 9.65 (4.00) 40.70
35.20 10.34 (3.72) 33.80
38.37 11.06 (4.47) 27.91
49.30 43.42 (10.17)
55.81 45.08 (9.11)
Values displayed as mean (SD). *Statistically significant difference at the P < 0.05 level. CRPS indicates complex regional pain syndrome; HADS, Hospital Anxiety and Depression Scale; LBP, Low Back Pain; NRS, Numerical Rating Scale; PDI, Pain Disability Index; TSK, Tampa Scale for Kinesiophobia.
r
2013 Lippincott Williams & Wilkins
Differences Between Clinical Groups There were no significant differences between the CRPS and LBP samples on any of the psychometric test scores, as shown by t tests for pain: t170 = 0.40, P = 0.69; disability: t159 = 0.46, P = 0.65; depression: t155 = 0.52, P = 0.60; anxiety: t155 = 1.08, P = 0.28; and kinesiophobia: t162 = 1.10, P = 0.27. In fact, the groups scored very similarly. The results are presented in Table 1.
Correlations Between Psychological Variables and Disability Scores
RESULTS
Sex (% female) Age (y) Duration of pain (mo) Work status (% working) Pain (Numerical Rating Scale) Disability (PDI) Depression (HADS-D) % Scoring 8-10 (borderline cases) % Scoring Z11 (clinical cases) Anxiety (HADS-A) % Scoring 8-10 (borderline cases) % Scoring Z11 (clinical cases) Kinesiophobia (TSK)
(82%) had nonspecific LBP, 13 (15%) had radiculopathy or spinal stenosis, and 3 (3%) had other causes of LBP (eg, cauda equina syndrome or vertebral compression fracture). The percentage of patients in each group who scored above the “cut-offs” for the HADS are presented in Table 1. Over one third of each group scored over the cut-off for clinical cases of depression, and around half of each group scored above the threshold for clinical cases of anxiety disorders.
Correlations between pain, disability, depression, anxiety, and kinesiophobia for the 2 groups are presented in Table 2. Of particular interest, we wanted to see whether disability was associated with psychological distress. In both the CRPS and LBP groups, all of the psychological variables significantly correlated with disability (at the P < 0.05 level), suggesting that those who were more depressed, anxious, and kinesiophobic were more disabled. When we used a Bonferroni correction for multiple comparisons (ie, setting the critical P-value at 0.005 for 10 correlations), all correlations remained significant except for the correlation between kinesiophobia and disability in the LBP group. The correlation between depression and disability was significantly greater in the CRPS group than the LBP group (z = 1.78, P = 0.038). There was no significant difference between the CRPS and LBP groups in terms of the relative strength of the correlation coefficients between disability and pain, disability and anxiety, and disability and kinesiophobia.
Multiple Regression Analysis: Predictors of Disability A multiple regression analysis demonstrated that the linear combination of pain, the psychological variables, diagnostic group, and interaction terms was a significant predictor of disability, accounting for about 38% of the variance in disability scores (R2 = 0.38, F9,135 = 9.28, P < 0.001). The results are presented in Table 3. Note that none of the demographic variables were significantly associated with disability, nor was pain duration, so these were not included in the analysis. The significant predictors that emerged were pain, depression, and the interaction between depression and diagnostic group. Those with greater pain intensity and who were more depressed were more disabled. Figure 1 demonstrates the interaction effect: the positive association between depression and disability was stronger in the CRPS group compared with the low back pain group.
Correlations Between Psychological Variables and Pain Scores As shown in Table 2, in the CRPS group there were significant correlations indicating that pain intensity was associated with greater disability, greater depression, www.clinicalpain.com |
649
Clin J Pain
Bean et al
Volume 30, Number 8, August 2014
TABLE 2. Correlations Among Measures of Pain Intensity, Disability, and Psychological Variables CRPS group Pain Disability Depression Anxiety LBP group Pain Disability Depression Anxiety
Disability
Depression
Anxiety
Kinesiophobia
0.552**
0.375** 0.561**
0.384** 0.384** 0.611**
0.357** 0.426** 0.507** 0.391**
0.403**
0.172 0.306**
0.086 0.334** 0.595**
0.231* 0.289** 0.377** 0.417**
*P < 0.05. **P < 0.01.
greater anxiety, and greater kinesiophobia. In the LBP group, the only psychometric variables that significantly correlated with pain intensity (at the P < 0.05 level) were disability and kinesiophobia. After using a Bonferroni correction for multiple comparisons (ie, setting the critical P-value to 0.005 for 10 correlations), the correlations between psychological variables and pain remained significant in the CRPS group but in the LBP group the only correlation that remained significant was between disability and pain. The correlation coefficient was significantly greater for the CRPS group than the LBP group for the relationship between pain and anxiety (z = 1.94, P = 0.03). There was no statistically significant difference in the strength of the correlations for: pain and disability, pain and depression, and pain and kinesiophobia.
Multiple Regression Analysis: Predictors of Pain A multiple regression analysis demonstrated that the linear combination of psychological variables, diagnostic group, and the interaction term was a significant predictor of pain intensity, accounting for about 16% of the variance in pain scores (R2 = 0.16, F7,142 = 3.80, P = 0.001). The results are presented in Table 3. The significant predictors
that emerged were kinesiophobia and the interaction between diagnostic group and anxiety. Those who were more kinesiophobic had higher pain scores. Figure 2 illustrates the interaction effect and shows that a positive relationship between anxiety and pain existed for the CRPS group but not for the LBP group.
DISCUSSION The present study showed that matched patients with CRPS and LBP scored remarkably similarly on measures of pain, disability, and psychological distress. Although some significant associations between psychological factors, pain, and disability existed in the LBP group, these tended to be stronger in the CRPS group. The only significant differences between the groups were that CRPS patients were less likely to be working. These results support several previous studies that found there were few differences between CRPS and LBP patients on measures of distress, and the few differences that have been reported are not consistent between studies.17,26,34,35 The present study provides further support that CRPS is not associated with a distinct psychological profile or greater level of distress compared with other people experiencing chronic pain. Instead, it seems more likely that persistent pain leads to a pattern of
TABLE 3. Regression Tables to Predict Disability and Pain
B Predictors of disability Diagnostic group Pain Depression Anxiety Kinesiophobia Pain diagnostic group Depressiondiagnostic group Anxiety diagnostic group Kinesiophobia diagnostic group Predictors of pain Diagnostic group Depression Anxiety Kinesiophobia Depressiondiagnostic group Anxiety diagnostic group Kinesiophobia diagnostic group
SE B
t
b
0.49 3.33 0.91 0.23 0.06 0.62 0.73 0.49 0.05
1.97 0.617 0.32 0.32 0.12 0.62 0.32 0.32 0.12
0.02 0.40 0.26 0.07 0.04 0.07 0.21 0.14 0.03
0.25 5.41 2.81 0.70 0.53 100 2.25 1.50 0.412
0.03 0.02 0.02 0.01 0.01 0.03 0.01
0.07 0.01 0.01 < 0.001 0.01 0.01 < 0.001
0.03 0.14 0.14 0.19 0.06 0.24 0.01
0.35 1.36 1.29 2.18 0.53 2.27 0.15
P 0.805 < 0.001** 0.006** 0.483 0.598 0.318 0.026* 0.135 0.681 0.730 0.177 0.198 0.031* 0.596 0.025* 0.879
*Statistically significant difference at P < 0.05 level. **Statistically significant difference at P < 0.01 level.
650 | www.clinicalpain.com
r
2013 Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 8, August 2014
B
Diagnostic Group: CRPS
Diagnostic Group: LBP
70
70
60
60
50
50
Disability (PDI)
Disability (PDI)
A
Psychological Factors in CRPS and Low Back Pain
40 30 20
40 30 20 10
10
R2 Linear = 0.0942
R2 Linear = 0.314 0
0 0
5
10 15 Depression (HADS)
20
0
5
10 15 Depression (HADS)
20
FIGURE 1. Scatterplots showing the relationship between disability and depression for CRPS (A) and LBP patients (B). CRPS indicates complex regional pain syndrome; HADS, Hospital Anxiety and Depression Scale; PDI, Pain Disability Index.
psychological functioning that might be consistent between pain conditions, and this is likely a normal part of the human experience of chronic pain. We note that both groups’ psychological test scores were similar to those reported from patient groups in other interdisciplinary pain centers and published in normative data sets.36,37 The present study also found that greater disability was associated with higher levels of anxiety, depression, and kinesiophobia in both groups. This was further tested in a multiple regression analysis to predict disability scores, which showed that there were main effects for pain and depression, and an interaction effect of depression and diagnostic group. This means that disability was associated with depression after controlling for pain intensity, and that the relationship between depression and disability was stronger in the CRPS group than in the LBP group. There has been very little previous research looking at factors that influence disability in CRPS. De Jong et al24 found that perceived harmfulness of activities predicted functional
B
Diagnostic Group: CRPS
Diagnostic Group: LBP
10
10
8
8 Pain (NRS)
Pain (NRS)
A
limitations above and beyond pain severity in those with chronic CRPS (although not in acute CRPS). Our univariate results provide further support for the role of kinesiophobia in CRPS. Overall it seems that the fearavoidance model is promising in terms of understanding and treating disability in CRPS patients. We also assessed the relationships between psychological factors and pain intensity. For the CRPS group, higher pain intensity was associated with greater depression, anxiety, and kinesiophobia. For the LBP group, the only psychological variable that was associated with pain intensity was kinesiophobia, although this relationship did not remain significant after controlling for multiple comparisons. The multivariate analyses showed that there was a main effect for kinesiophobia and an interaction between anxiety and diagnostic group. This indicates that kinesiophobia is associated with pain for both groups, and that anxiety is associated with pain intensity for CRPS patients, but not for LBP patients. We found that the psychological
6
4
6
4 R2 Linear = 0.007
R2 Linear = 0.199 2
2 0
5
10 Anxiety (HADS)
15
20
0
5
10 Anxiety (HADS)
15
20
FIGURE 2. Scatterplots showing the relationship between pain and anxiety for CRPS (A) and LBP patients (B). CRPS indicates complex regional pain syndrome; HADS, Hospital Anxiety and Depression Scale; NRS, Numerical Rating Scale. r
2013 Lippincott Williams & Wilkins
www.clinicalpain.com |
651
Bean et al
variables accounted for 16% of the variance in pain scores, which, if interpreted using Cohen effect sizes, would indicate a moderate effect size. The psychological variables, together with pain intensity, accounted for more than double the variance in disability scores (38%), which would equate to a strong effect size according to Cohen. Our findings were also somewhat consistent with previous research which has sought to determine factors that influence pain severity in CRPS. Bruehl et al26 reported that depression and anxiety were correlated with pain severity in CRPS patients, and that these correlations were stronger than those that existed in LBP patients. The similarity of these results suggests this may be a fairly robust finding, and the present study goes further by also assessing the role of kinesiophobia. Feldman et al21 used a daily diary study to show that although pain leads to an increase in depression, anxiety, and anger, depression and social support also influenced pain in CRPS. Moseley23 found that catastrophic thinking about pain and kinesiophobia led to greater pain intensity and swelling in response to imagined movements in CRPS. There are several mechanisms by which psychological factors might be more strongly associated with disability and pain in CRPS than in LBP. It is possible that sympathetic nervous system activity could mediate this relationship. Research has shown that sympathetic activity influences CRPS symptoms in at least a subset of patients,38 and therefore any psychological distress that causes a sympathetic nervous system response could influence CRPS symptoms. Another possibility is that psychological factors might have a closer relationship with CRPS compared with LBP due to the range of symptoms experienced by CRPS patients, including swelling, color and temperature changes, and sensory symptoms. These symptoms may be distressing for patients, and could reinforce fears of serious pathology, leading to poorer mood and a closer pairing of pain and distress. Rommel et al39 reported that CRPS patients with allodynia were more distressed than those without allodynia, and it might be that the sensory abnormalities associated with CRPS also lead to a cycle of distress and avoidance, which could in turn reinforce pain and disability. We note that our results are inconsistent with the previous prospective studies that have been conducted in CRPS, that have measured psychological factors either before surgery or at the time of a fracture, and found few psychological differences between those who subsequently develop CRPS and those who do not.4,6,7 This suggests that psychological factors might not be particularly important in the initial development of CRPS, and it might be that physiological factors such as neurogenic inflammation40 or sympathetic nervous system function41 play a more important role at the early stage. Interestingly, a prospective study found that although preoperative psychological factors did not predict the development of CRPS, early postoperative increases in distress levels predicted later CRPS severity.25 More research is needed to assess the importance of psychological factors at different time points in the course of CRPS. There are several clinical implications of the present study. At present, psychologists treating CRPS patients have a limited body of research available on which to base treatment decisions about which psychological variables to target. The present study suggests that psychological treatments aimed at reducing depression, anxiety, and kinesiophobia are worth trialling in CRPS. Our data
652 | www.clinicalpain.com
Clin J Pain
Volume 30, Number 8, August 2014
suggest that it would be worth investigating whether psychological treatments that target depression lead to improvements in disability, and whether treatments that target anxiety lead to improvements in pain. There is already some evidence from multiple single-case design studies showing the efficacy of treatments aimed at kinesiophobia,8,9 and these data suggest that targeting depression and anxiety in addition may be worthwhile. The main limitation of the study is that the data were collected at a single time and the direction of relationships is unclear. It is not only possible but quite likely that while psychological factors might influence pain and disability, that higher pain intensity and greater disability lead to poorer psychological wellbeing. It is also possible that a third (unknown) variable accounts for the relationship. Prospective studies are needed to better understand these relationships. The data are also limited by self-report bias. Another limitation is that there was some missing data, more so in the CRPS than in the LBP group. Systematically missing data or selective attrition are a potential source of bias for this study, as it is possible that the difference between groups was due to the difference between those who complete questionnaires compared with those who do not. However, to investigate this, we reran the analyses utilizing only the 60 CRPS and low back pain matched pairs with complete data sets, and our results were virtually identical to those presented here. One strength of the study was that the samples were matched for age, sex, and pain duration, and this removes some possible sources of variation in measures of pain, disability, and psychological wellbeing. In conclusion, we found that there were remarkable similarities between CRPS patients and age-matched, sexmatched, and pain duration-matched LBP patients on measures of pain, disability, depression, anxiety, and kinesiophobia. In addition, we found that both pain and depression were associated with disability, and the association between depression and disability was stronger for the CRPS group than the LBP group. We found that kinesiophobia was associated with greater pain intensity, and that anxiety was associated with pain intensity in the CRPS group but not in the LBP group. Although previous research has demonstrated a limited role for psychological factors in causing CRPS, our results support the hypothesis that in chronic CRPS, psychological factors might play a role in maintaining pain and disability in the same way they influence other types of acute, experimental, and chronic pain. ACKNOWLEDGMENTS The authors thank Catherine Pollard, BSc (Hons), Physiotherapy Diploma HE Orthopaedic Medicine, Trevor Coe, MA, MB BChir, FRCA, FFPMANZCA, Department of Anaesthetist and Pain Medicine, The Auckland Regional Pain Service, Auckland District Health Board, Auckland, New Zealand and the rest of The Auckland Regional Pain Service team for assistance with data collection, and the Oakley Mental Health Research Foundation for supporting our CRPS Research. REFERENCES 1. Morley S. Psychology of pain. Br J Anaesth. 2008;101:25–31. 2. Beerthuizen A, van t Spijker A, Huygen FJ, et al. Is there an association between psychological factors and the Complex Regional Pain Syndrome type 1 (CRPS1) in adults? A systematic review. Pain. 2009;145:52–59. r
2013 Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 8, August 2014
3. Lohnberg JA, Altmaier EM. A review of psychosocial factors in complex regional pain syndrome. J Clin Psychol Med Settings. 2013;20:247–254. 4. Beerthuizen A, Stronks DL, FJ Huygen, et al. The association between psychological factors and the development of complex regional pain syndrome type 1 (CRPS1)—A prospective multicenter study. Eur J Pain. 2011;15:971–975. 5. Dilek B, Yemez B, Kizil R, et al. Anxious personality is a risk factor for developing complex regional pain syndrome type I. Rheumatol Int. 2012;32:915–920. 6. Harden RN, Bruehl S, Stanos S, et al. Prospective examination of pain-related and psychological predictors of CRPS-like phenomena following total knee arthroplasty: a preliminary study. Pain. 2003;106:393–400. 7. Puchalski P, Zyluk A. Complex regional pain syndrome type 1 after fractures of the distal radius: a prospective study of the role of psychological factors. J Hand Surg [Br]. 2005;30:574–580. 8. de Jong JR, Vlaeyen JWS, Onghena P, et al. Reduction of painrelated fear in complex regional pain syndrome type I: the application of graded exposure in vivo. Pain. 2005;116:264–275. 9. van de Meent H, Oerlemans M, Bruggeman A, et al. Safety of “pain exposure” physical therapy in patients with complex regional pain syndrome type 1. Pain. 2011;152:1431–1438. 10. Goebel A, Barker CH, Turner-Stokes L, et al. Complex Regional Pain Syndrome in Adults: UK Guidelines for Diagnosis, Referral and Management in Primary and Secondary Care. London: Royal College of Physicians; 2012. 11. Nicholas MK, Linton SJ, Watson PJ, et al. Early identification and management of psychological risk factors (“yellow flags”) in patients with low back pain: a reappraisal. Phys Ther. 2011;91:737–753. 12. Iles RA, Davidson M, Taylor NF. Psychosocial predictors of failure to return to work in non-chronic non-specific low back pain: a systematic review. Occup Environ Med. 2008;65: 507–517. 13. Heitz CAM, Hilfiker R, Bachmann LM, et al. Comparison of risk factors predicting return to work between patients with subacute and chronic non-specific low back pain: systematic review. Eur Spine J. 2009;18:1829–1835. 14. Chou R, Shekelle P. Will this patient develop persistent disabling low back pain? JAMA. 2010;303:1295–1302. 15. Henschke N, Ostelo RW, van Tulder MW, et al. Behavioural treatment for chronic low-back pain. Cochrane Database Syst Rev. 2010;CD002014. 16. Nelson DV, Novy DM. Psychological characteristics of reflex sympathetic dystrophy versus myofascial pain syndromes. Region Anaesth. 1996;21:202–208. 17. Monti DA, Herring CL, Schwartzman RJ, et al. Personality assessment of patients with complex regional pain syndrome type I. Clin J Pain. 1998;14:295–302. 18. Katz MM, Hungerford DS, Krackow KA, et al. Reflex sympathetic dystrophy as a cause of poor results after total knee arthroplasty. J Arthroplasty. 1986;1:117–124. 19. Zucchini M, Alberti G, Moretti MP. Algodystrophy and related psychological features. Funct Neurol. 1989;4:153–156. 20. Feliu MH, Edwards CL. Psychologic factors in the development of complex regional pain syndrome: history, myth, and evidence. Clin J Pain. 2010;26:258–263. 21. Feldman SI, Downey G, Schaffer-Neitz R. Pain negative mood, and perceived support in chronic pain patients: a daily diary study of people with reflex sympathetic dystrophy syndrome. J Consult Clin Psych. 1999;67:776–785.
r
2013 Lippincott Williams & Wilkins
Psychological Factors in CRPS and Low Back Pain
22. Bruehl S, Chung OY, Burns JW. Differential effects of expressive anger regulation on chronic pain intensity in CRPS and non-CRPS limb pain patients. Pain. 2003;104:647–654. 23. Moseley GL. Imagined movements cause pain and swelling in a patient with complex regional pain syndrome. Neurology. 2004;62:1644. 24. de Jong JR, Vlaeyen JWS, de Gelder JM, et al. Pain-related fear, perceived harmfulness of activities, and functional limitations in complex regional pain syndrome type I. J Pain. 2011;12:1209–1218. 25. Harden RN, Bruehl S, RSGM Perez, et al. Development of a severity score for CRPS. Pain. 2010;151:870–876. 26. Bruehl S, Husfeldt B, Lubenow TR, et al. Psychological differences between reflex sympathetic dystrophy and nonRSD chronic pain patients. Pain. 1996;67:107–114. 27. Merskey H, Bogduk N. Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms. 2nd ed. Seattle: IASP Press; 1994. 28. Tait RC, Chibnall JT, Krause S. The Pain Disability Index: psychometric properties. Pain. 1990;40:171–182. 29. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiat Scand. 1983;67:361–370. 30. Miller RP, Kori SH, Todd DD. The Tampa Scale for Kinesiophobia (Unpublished Report). Tampa, FL; 1991. 31. Erdfelder E, Faul F, Buchner A. GPOWER: a general power analysis program. Behav Res Meth Instrum Comput. 1996;28: 1–11. 32. Preacher KJ. Calculation for the test of the difference between two independent correlation coefficients. May 2002. Available at: http://quantpsy.org. Accessed September 20, 2012. 33. Chou R, Qaseem A, Snow V, et al. Diagnosis and treatment of low back pain: a joint clinical practice guideline from the American College of Physicians and the American Pain Society. Ann Intern Med. 2007;147:478–491. 34. DeGood DE, Cundiff GW, Adams LE, et al. A psychosocial and behavioral comparison of reflex sympathetic dystrophy, low back pain, and headache patients. Pain. 1993;54:317–322. 35. Ciccone DS, Bandilla EB, Wu W. Psychological dysfunction in patients with reflex sympathetic dystrophy. Pain. 1997;71: 323–333. 36. Nicholas MK, Asghari A, Blyth FM. What do the numbers mean? Normative data in chronic pain measures. Pain. 2008;134:158–173. 37. Chibnall JT, Tait RC. The Pain Disability Index: factor structure and normative data. Arch Phys Med Rehabil. 1994;75:1082–1086. 38. Drummond PD. Sensory disturbances in complex regional pain syndrome: clinical observations, autonomic interactions, and possible mechanisms. Pain Med. 2010;11:1257–1266. 39. Rommel O, Willweber-Strumpf A, Wagner P, et al. Psychische veranderungen bei patienten mit komplexem regionalem schmerzsyndrom (CRPS) [Psychological abnormalities in patients with complex regional pain syndrome (CRPS)]. Schmerz. 2005;19:272–284. 40. Birklein F, Schmelz M. Neuropeptides, neurogenic inflammation and complex regional pain syndrome (CRPS). Neurosci Lett. 2008;437:199–202. 41. Schurmann M, Gradl G, Zaspel J, et al. Peripheral sympathetic function as a predictor of complex regional pain syndrome type I (CRPS I) in patients with radial fracture. Auton Neurosci. 2000;86:127–134.
www.clinicalpain.com |
653
Relationships Between Psychological Factors, Pain, and Disability in Complex Regional Pain Syndrome and Low Back Pain Debbie J. Bean, MSc,*w Malcolm H. Johnson, MA,w and Robert R. Kydd, PhDw
Objective: Cognitive and emotional factors are known to influence peoples’ pain experiences in many conditions, including low back pain. However, in complex regional pain syndrome (CRPS), their role is unclear. This study aimed to assess the relationships between psychological factors, pain, and disability in CRPS, compared with low back pain. This could help to identify target variables for psychological treatment. Materials and Methods: A total of 88 CRPS patients and 88 low back pain patients completed measures of pain, disability, depression, anxiety, and fear of movement and reinjury (kinesiophobia). Mean scores between the 2 groups were compared, and correlations between psychological factors, pain, and disability were compared between the 2 groups. Predictors of pain and disability were assessed using multiple regression analyses. Results: The 2 groups had remarkably similar scores on measures of pain, disability, depression, anxiety, and kinesiophobia. In both groups, those who were more depressed, anxious, and kinesiophobic were more disabled. For the CRPS group (but not the low back pain group), pain intensity significantly correlated with distress. Multivariate analyses showed that the unique predictors of disability for the 2 groups were pain and depression, and that depression had a stronger relationship with disability for the CRPS group. For both groups, pain intensity was predicted by kinesiophobia, and anxiety was a unique predictor in the CRPS group only. Discussion: In CRPS, disability and pain severity were more strongly associated with psychological factors than they were in low back pain. Cause and effect relationships could not be established by this cross-sectional study. Key Words: complex regional pain syndrome, low back pain, psychological factors, disability
(Clin J Pain 2014;30:647–653)
A
large body of literature has demonstrated that psychological and social factors influence people’s experience of pain in a variety of contexts, including experimental, acute, procedural, and chronic pain.1 The range of psychological factors that influence pain includes mood, cognitions and expectations, behavioral patterns, attention states, and learning history. Although such Received for publication February 20, 2013; revised May 20, 2014 accepted August 2, 2013. From *The Auckland Regional Pain Service (TARPS), Auckland District Health Board; and wDepartment of Psychological Medicine, The University of Auckland, Auckland, New Zealand. The authors declare no conflict of interest. Reprints: Debbie J. Bean, MSc, Department of Psychological Medicine, The University of Auckland, Private Bag 92019, Auckland, 1142, New Zealand (e-mail: [email protected]). Copyright r 2013 by Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 8, August 2014
psychological factors have been demonstrated to influence pain and disability in a range of chronic pain conditions, there have been relatively few studies demonstrating a significant influence of psychological factors in complex regional pain syndrome (CRPS). In fact, a recent systematic review identified no relationship between a number of psychological factors and CRPS type 1, except for a possible association of CRPS type 1 with more life events.2 Another recent literature review found that there was no evidence that personality or psychopathology play a causal role in CRPS; however, it did find that CRPS leads to a sequelae of negative psychological and social consequences.3 Prospective studies that have measured psychological factors among a cohort of fracture, surgery, or stroke patients then assessed patients at a later date for CRPS have reported few psychological risk factors for CRPS development.4–7 There are no published randomized controlled trials of psychological treatments for CRPS, although smaller trials have provided some evidence of the efficacy of exposure-based therapies aimed at reducing patients’ fear of movement or reinjury.8,9 Despite this lack of research, patients with longstanding CRPS are often treated in interdisciplinary pain clinics where psychological therapies are recommended and provided. The 2012 Royal College of Physicians’ guidelines recommends that psychological intervention is one of the 4 main arms of treatment for CRPS.10 Psychologists providing this treatment likely generalize from the research on other chronic pain conditions to identify appropriate target variables for treatment and set treatment goals and activities. One such condition is low back pain (LBP), where a large body of literature has established the importance of psychological factors in predicting pain intensity and heightened disability.11 Systematic reviews have shown that psychological factors such as depression, fear of movement and reinjury (kinesiophobia), catastrophic thinking, and poor expectations predict the progression of LBP from an acute to a chronic timeframe.12–14 Randomized controlled trials have demonstrated the efficacy of behavioral treatments for LBP, at least in the short term.15 There are several possible reasons why previous research has not demonstrated a relationship between psychological factors and CRPS. It might be that no such relationship exists. In contrast, it might be that previous research has suffered from methodological shortcomings, or has not assessed the right relationships. This may in fact be the case, as much of the previous research on psychological factors in CRPS has been based on a dualistic model of health and has sought to find psychological causes for CRPS (as mentioned above), or has focussed on identifying psychopathology or personality disorders among these patients over and above what would be expected of www.clinicalpain.com |
647
Clin J Pain
Bean et al
a sample of people in pain.16,17 This may be because traditionally CRPS was stigmatized as “psychogenic.”18–20 Research from other fields of pain has adopted a biopsychosocial model and suggests that psychological factors are not usually a singular cause of pain, but that unhelpful psychological and physical factors interact to produce heightened pain and disability. In line with this, the small number of studies that have aimed to identify more subtle influences of psychological factors on pain and disability in CRPS have shown some interactions between mood and pain levels,21 anger-expressiveness and pain severity,22 and fear of movement and disability.23,24 A prospective study reported that early increases in depression after knee replacement were associated with CRPS severity at 12 months, suggesting that psychological factors may interact with pain over time, leading to poorer outcome.25 Bruehl et al26 compared CRPS patients with non-CRPS limb pain and LBP patients and found that while there were few differences between the groups on measures of distress, there were stronger relationships between distress and pain severity for the CRPS and limb pain groups than the LBP group. The aim for the present study was to further assess associations between distress and pain in CRPS, and to compare the strength of any associations with a sample of patients with LBP. We aimed to further extend the findings of Bruehl et al26 by also looking at the role of fear of movement as a predictor variable and to look at disability and pain severity as outcome variables. LBP was chosen because in this group previous research demonstrates the importance of psychological factors, and therefore we hypothesized that if there were similarities between CRPS and LBP groups these would be meaningful.
MATERIALS AND METHODS Participants The participants were 88 consecutive CRPS patients and 88 LBP patients seen at The Auckland Regional Pain Service (Interdisciplinary Pain Center) between January 2009 and December 2011. The samples were matched for age, sex, and pain duration and thus the LBP patients were not a consecutive sample. Patients experiencing both CRPS and LBP concurrently were excluded. The interdisciplinary pain center typically sees patients with chronic pain who are referred after multiple treatments and who have seen specialists from other disciplines, and thus both samples are likely to be those severely affected by pain who have been difficult to treat. Both CRPS type 1 and 2 were included in the CRPS sample. Patients in the LBP sample were included if their primary complaint was pain in the low back and/or leg and this included patients with nonspecific back pain and other causes of LBP.
Procedure The study was a retrospective review and analysis of routinely administered psychological scales and medical records. The observational study was approved by the New Zealand Ministry of Health Northern X Ethics Committee. Patients were identified using previous clinic lists that code patients’ pain problems according to pain location. The first author searched the initial assessment reports from all patients coded with upper or lower limb pain to determine whether the patient was diagnosed with CRPS during their medical assessment at the interdisciplinary pain center. CRPS is only diagnosed at the pain center after a
648 | www.clinicalpain.com
Volume 30, Number 8, August 2014
comprehensive assessment and examination by pain specialist doctors, and patient records were checked to ensure that patients met the 1994 IASP criteria for CRPS.27 The first author also searched the records of all patients coded with LBP to determine whether this diagnosis was confirmed during the patient’s comprehensive assessment at the pain center. The following data were extracted from electronic patient records: age, sex, ethnicity, pain duration, condition (CRPS or LBP), CRPS limb, and employment status (working or not working). The following scores were also extracted from questionnaires that patients are required to fill in before receiving an appointment, these questionnaires are scanned and stored electronically on the hospital computer system, and scores were extracted by the first author: 1. Pain Numerical Rating Scale: Patients were asked to rate their average pain intensity on a scale of 0 (no pain) to 10 (worst pain you can imagine). 2. Pain Disability Index (PDI)28: This 7-item scale asks participants to rate their level of disability on a range of life domains. Scores range from 0 (no disability) to 70 (maximum disability). 3. Hospital Anxiety and Depression Scale (HADS)29: This 14-item scale measures symptoms of depression and anxiety, producing 2 separate scores for the separate constructs. The scale is designed to limit the bias of physical health symptoms on depression and anxiety scores. The scale ranges from 0 (very low levels of depression or anxiety) to 21 (very high levels of depression or anxiety). In the original paper, cut-off scores of 8 and above were suggested to identify borderline or possible cases of depression or anxiety, and scores of 11 and above were suggested to identify probable clinical cases of depression and anxiety. 4. Tampa Scale for Kinesiophobia (TSK)30: This is a 17item scale designed to measure kinesiophobia (fear of movement and/or (re)injury). Scores range from 17 (low kinesiophobia) to 68 (maximum kinesiophobia). Given that the data were collected as part of routine clinical care, there was some missing data that is reflected in the varying degrees of freedom reported for each of the statistical tests. There were more LBP patients with full data sets (80/88) compared with CRPS patients (65/88) because when selecting matched controls for the CRPS patients, efforts were made to select those without missing data.
Data Analysis A power analysis was conducted using G-Power.31 This was based on a 2-tailed, independent samples t test, and demonstrated that the study had 90% power to detect a moderate effect size (d = 0.5, a = 0.05), with 86 participants required in each group. Data were entered into SPSS for Windows. Not all variables displayed normal distributions. Thus, nonparametric tests were used, but these produced the same results as parametric tests, so parametric test results are presented here for ease of comprehension. Independent samples t tests and w2 tests were used to explore differences between the CRPS patients and LBP patients on demographic details and psychometric test scores. Pearson correlations were used to assess relationships between pain intensity, disability scores, and scores on psychological variables for each patient group. The strength of the correlations between psychological variables and disability scores were compared between the 2 r
2013 Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 8, August 2014
Psychological Factors in CRPS and Low Back Pain
diagnostic groups by computing Fisher z scores, with these compared statistically using z scores. This was done utilizing the calculator from Preacher.32 Multiple regression analyses were used to determine unique predictors of the 2 dependent variables (pain and disability), and to assess whether the psychological predictors were equivalent for the 2 diagnostic groups. The psychological variables were entered into the regression analyses, along with diagnostic group (dummy coded), and interaction terms for each psychological variable ((psychological variable mean of psychological variable) diagnostic group (dummy coded)), thus the interaction terms were mean-centered. For the multiple regression with pain as the dependent variable, the residuals were not normally distributed (negatively skewed), so a square root transformation was applied to pain scores. This was successful, resulting in normally distributed residuals, so the results from the transformed data are presented.
Demographics and Description of the Sample The demographic characteristics of the 2 samples are described in Table 1. Because the samples were matched, there were no differences between the groups on age, sex, or pain duration. Both samples had a fairly long pain duration of approximately 3½ years. The CRPS patients were significantly less likely to be working, w2(1, N = 172) = 5.49, P = 0.019 compared with the LBP patients. Of the CRPS sample, 66 (75%) had the upper limb affected and 22 (25%) had the lower limb affected. A total of 40 (45%) of the CRPS patients had the left side affected, 42 (48%) had the right side affected, and 6 (7%) had CRPS affecting both sides of the body. Of the CRPS patients, 75 had CRPS type 1 (85%), and 13 had CRPS type 2 (15%). We classified the LBP patients into 3 diagnostic groups according to the guidelines of the American College of Physicians and American Pain Society.33 A total of 72 TABLE 1. Demographic Details and Psychological Test Scores for the CRPS and Low Back Pain Patient Samples
CRPS (N = 88)
LBP (N = 88)
59 45 (12.44) 43.15 (53.73) 20 7.59 (1.77) 44.92 (16.09) 9.99 (4.04) 22.53
59 45 (12.60) 43.42 (53.14) 36* 7.69 (1.58) 43.89 (12.43) 9.65 (4.00) 40.70
35.20 10.34 (3.72) 33.80
38.37 11.06 (4.47) 27.91
49.30 43.42 (10.17)
55.81 45.08 (9.11)
Values displayed as mean (SD). *Statistically significant difference at the P < 0.05 level. CRPS indicates complex regional pain syndrome; HADS, Hospital Anxiety and Depression Scale; LBP, Low Back Pain; NRS, Numerical Rating Scale; PDI, Pain Disability Index; TSK, Tampa Scale for Kinesiophobia.
r
2013 Lippincott Williams & Wilkins
Differences Between Clinical Groups There were no significant differences between the CRPS and LBP samples on any of the psychometric test scores, as shown by t tests for pain: t170 = 0.40, P = 0.69; disability: t159 = 0.46, P = 0.65; depression: t155 = 0.52, P = 0.60; anxiety: t155 = 1.08, P = 0.28; and kinesiophobia: t162 = 1.10, P = 0.27. In fact, the groups scored very similarly. The results are presented in Table 1.
Correlations Between Psychological Variables and Disability Scores
RESULTS
Sex (% female) Age (y) Duration of pain (mo) Work status (% working) Pain (Numerical Rating Scale) Disability (PDI) Depression (HADS-D) % Scoring 8-10 (borderline cases) % Scoring Z11 (clinical cases) Anxiety (HADS-A) % Scoring 8-10 (borderline cases) % Scoring Z11 (clinical cases) Kinesiophobia (TSK)
(82%) had nonspecific LBP, 13 (15%) had radiculopathy or spinal stenosis, and 3 (3%) had other causes of LBP (eg, cauda equina syndrome or vertebral compression fracture). The percentage of patients in each group who scored above the “cut-offs” for the HADS are presented in Table 1. Over one third of each group scored over the cut-off for clinical cases of depression, and around half of each group scored above the threshold for clinical cases of anxiety disorders.
Correlations between pain, disability, depression, anxiety, and kinesiophobia for the 2 groups are presented in Table 2. Of particular interest, we wanted to see whether disability was associated with psychological distress. In both the CRPS and LBP groups, all of the psychological variables significantly correlated with disability (at the P < 0.05 level), suggesting that those who were more depressed, anxious, and kinesiophobic were more disabled. When we used a Bonferroni correction for multiple comparisons (ie, setting the critical P-value at 0.005 for 10 correlations), all correlations remained significant except for the correlation between kinesiophobia and disability in the LBP group. The correlation between depression and disability was significantly greater in the CRPS group than the LBP group (z = 1.78, P = 0.038). There was no significant difference between the CRPS and LBP groups in terms of the relative strength of the correlation coefficients between disability and pain, disability and anxiety, and disability and kinesiophobia.
Multiple Regression Analysis: Predictors of Disability A multiple regression analysis demonstrated that the linear combination of pain, the psychological variables, diagnostic group, and interaction terms was a significant predictor of disability, accounting for about 38% of the variance in disability scores (R2 = 0.38, F9,135 = 9.28, P < 0.001). The results are presented in Table 3. Note that none of the demographic variables were significantly associated with disability, nor was pain duration, so these were not included in the analysis. The significant predictors that emerged were pain, depression, and the interaction between depression and diagnostic group. Those with greater pain intensity and who were more depressed were more disabled. Figure 1 demonstrates the interaction effect: the positive association between depression and disability was stronger in the CRPS group compared with the low back pain group.
Correlations Between Psychological Variables and Pain Scores As shown in Table 2, in the CRPS group there were significant correlations indicating that pain intensity was associated with greater disability, greater depression, www.clinicalpain.com |
649
Clin J Pain
Bean et al
Volume 30, Number 8, August 2014
TABLE 2. Correlations Among Measures of Pain Intensity, Disability, and Psychological Variables CRPS group Pain Disability Depression Anxiety LBP group Pain Disability Depression Anxiety
Disability
Depression
Anxiety
Kinesiophobia
0.552**
0.375** 0.561**
0.384** 0.384** 0.611**
0.357** 0.426** 0.507** 0.391**
0.403**
0.172 0.306**
0.086 0.334** 0.595**
0.231* 0.289** 0.377** 0.417**
*P < 0.05. **P < 0.01.
greater anxiety, and greater kinesiophobia. In the LBP group, the only psychometric variables that significantly correlated with pain intensity (at the P < 0.05 level) were disability and kinesiophobia. After using a Bonferroni correction for multiple comparisons (ie, setting the critical P-value to 0.005 for 10 correlations), the correlations between psychological variables and pain remained significant in the CRPS group but in the LBP group the only correlation that remained significant was between disability and pain. The correlation coefficient was significantly greater for the CRPS group than the LBP group for the relationship between pain and anxiety (z = 1.94, P = 0.03). There was no statistically significant difference in the strength of the correlations for: pain and disability, pain and depression, and pain and kinesiophobia.
Multiple Regression Analysis: Predictors of Pain A multiple regression analysis demonstrated that the linear combination of psychological variables, diagnostic group, and the interaction term was a significant predictor of pain intensity, accounting for about 16% of the variance in pain scores (R2 = 0.16, F7,142 = 3.80, P = 0.001). The results are presented in Table 3. The significant predictors
that emerged were kinesiophobia and the interaction between diagnostic group and anxiety. Those who were more kinesiophobic had higher pain scores. Figure 2 illustrates the interaction effect and shows that a positive relationship between anxiety and pain existed for the CRPS group but not for the LBP group.
DISCUSSION The present study showed that matched patients with CRPS and LBP scored remarkably similarly on measures of pain, disability, and psychological distress. Although some significant associations between psychological factors, pain, and disability existed in the LBP group, these tended to be stronger in the CRPS group. The only significant differences between the groups were that CRPS patients were less likely to be working. These results support several previous studies that found there were few differences between CRPS and LBP patients on measures of distress, and the few differences that have been reported are not consistent between studies.17,26,34,35 The present study provides further support that CRPS is not associated with a distinct psychological profile or greater level of distress compared with other people experiencing chronic pain. Instead, it seems more likely that persistent pain leads to a pattern of
TABLE 3. Regression Tables to Predict Disability and Pain
B Predictors of disability Diagnostic group Pain Depression Anxiety Kinesiophobia Pain diagnostic group Depressiondiagnostic group Anxiety diagnostic group Kinesiophobia diagnostic group Predictors of pain Diagnostic group Depression Anxiety Kinesiophobia Depressiondiagnostic group Anxiety diagnostic group Kinesiophobia diagnostic group
SE B
t
b
0.49 3.33 0.91 0.23 0.06 0.62 0.73 0.49 0.05
1.97 0.617 0.32 0.32 0.12 0.62 0.32 0.32 0.12
0.02 0.40 0.26 0.07 0.04 0.07 0.21 0.14 0.03
0.25 5.41 2.81 0.70 0.53 100 2.25 1.50 0.412
0.03 0.02 0.02 0.01 0.01 0.03 0.01
0.07 0.01 0.01 < 0.001 0.01 0.01 < 0.001
0.03 0.14 0.14 0.19 0.06 0.24 0.01
0.35 1.36 1.29 2.18 0.53 2.27 0.15
P 0.805 < 0.001** 0.006** 0.483 0.598 0.318 0.026* 0.135 0.681 0.730 0.177 0.198 0.031* 0.596 0.025* 0.879
*Statistically significant difference at P < 0.05 level. **Statistically significant difference at P < 0.01 level.
650 | www.clinicalpain.com
r
2013 Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 8, August 2014
B
Diagnostic Group: CRPS
Diagnostic Group: LBP
70
70
60
60
50
50
Disability (PDI)
Disability (PDI)
A
Psychological Factors in CRPS and Low Back Pain
40 30 20
40 30 20 10
10
R2 Linear = 0.0942
R2 Linear = 0.314 0
0 0
5
10 15 Depression (HADS)
20
0
5
10 15 Depression (HADS)
20
FIGURE 1. Scatterplots showing the relationship between disability and depression for CRPS (A) and LBP patients (B). CRPS indicates complex regional pain syndrome; HADS, Hospital Anxiety and Depression Scale; PDI, Pain Disability Index.
psychological functioning that might be consistent between pain conditions, and this is likely a normal part of the human experience of chronic pain. We note that both groups’ psychological test scores were similar to those reported from patient groups in other interdisciplinary pain centers and published in normative data sets.36,37 The present study also found that greater disability was associated with higher levels of anxiety, depression, and kinesiophobia in both groups. This was further tested in a multiple regression analysis to predict disability scores, which showed that there were main effects for pain and depression, and an interaction effect of depression and diagnostic group. This means that disability was associated with depression after controlling for pain intensity, and that the relationship between depression and disability was stronger in the CRPS group than in the LBP group. There has been very little previous research looking at factors that influence disability in CRPS. De Jong et al24 found that perceived harmfulness of activities predicted functional
B
Diagnostic Group: CRPS
Diagnostic Group: LBP
10
10
8
8 Pain (NRS)
Pain (NRS)
A
limitations above and beyond pain severity in those with chronic CRPS (although not in acute CRPS). Our univariate results provide further support for the role of kinesiophobia in CRPS. Overall it seems that the fearavoidance model is promising in terms of understanding and treating disability in CRPS patients. We also assessed the relationships between psychological factors and pain intensity. For the CRPS group, higher pain intensity was associated with greater depression, anxiety, and kinesiophobia. For the LBP group, the only psychological variable that was associated with pain intensity was kinesiophobia, although this relationship did not remain significant after controlling for multiple comparisons. The multivariate analyses showed that there was a main effect for kinesiophobia and an interaction between anxiety and diagnostic group. This indicates that kinesiophobia is associated with pain for both groups, and that anxiety is associated with pain intensity for CRPS patients, but not for LBP patients. We found that the psychological
6
4
6
4 R2 Linear = 0.007
R2 Linear = 0.199 2
2 0
5
10 Anxiety (HADS)
15
20
0
5
10 Anxiety (HADS)
15
20
FIGURE 2. Scatterplots showing the relationship between pain and anxiety for CRPS (A) and LBP patients (B). CRPS indicates complex regional pain syndrome; HADS, Hospital Anxiety and Depression Scale; NRS, Numerical Rating Scale. r
2013 Lippincott Williams & Wilkins
www.clinicalpain.com |
651
Bean et al
variables accounted for 16% of the variance in pain scores, which, if interpreted using Cohen effect sizes, would indicate a moderate effect size. The psychological variables, together with pain intensity, accounted for more than double the variance in disability scores (38%), which would equate to a strong effect size according to Cohen. Our findings were also somewhat consistent with previous research which has sought to determine factors that influence pain severity in CRPS. Bruehl et al26 reported that depression and anxiety were correlated with pain severity in CRPS patients, and that these correlations were stronger than those that existed in LBP patients. The similarity of these results suggests this may be a fairly robust finding, and the present study goes further by also assessing the role of kinesiophobia. Feldman et al21 used a daily diary study to show that although pain leads to an increase in depression, anxiety, and anger, depression and social support also influenced pain in CRPS. Moseley23 found that catastrophic thinking about pain and kinesiophobia led to greater pain intensity and swelling in response to imagined movements in CRPS. There are several mechanisms by which psychological factors might be more strongly associated with disability and pain in CRPS than in LBP. It is possible that sympathetic nervous system activity could mediate this relationship. Research has shown that sympathetic activity influences CRPS symptoms in at least a subset of patients,38 and therefore any psychological distress that causes a sympathetic nervous system response could influence CRPS symptoms. Another possibility is that psychological factors might have a closer relationship with CRPS compared with LBP due to the range of symptoms experienced by CRPS patients, including swelling, color and temperature changes, and sensory symptoms. These symptoms may be distressing for patients, and could reinforce fears of serious pathology, leading to poorer mood and a closer pairing of pain and distress. Rommel et al39 reported that CRPS patients with allodynia were more distressed than those without allodynia, and it might be that the sensory abnormalities associated with CRPS also lead to a cycle of distress and avoidance, which could in turn reinforce pain and disability. We note that our results are inconsistent with the previous prospective studies that have been conducted in CRPS, that have measured psychological factors either before surgery or at the time of a fracture, and found few psychological differences between those who subsequently develop CRPS and those who do not.4,6,7 This suggests that psychological factors might not be particularly important in the initial development of CRPS, and it might be that physiological factors such as neurogenic inflammation40 or sympathetic nervous system function41 play a more important role at the early stage. Interestingly, a prospective study found that although preoperative psychological factors did not predict the development of CRPS, early postoperative increases in distress levels predicted later CRPS severity.25 More research is needed to assess the importance of psychological factors at different time points in the course of CRPS. There are several clinical implications of the present study. At present, psychologists treating CRPS patients have a limited body of research available on which to base treatment decisions about which psychological variables to target. The present study suggests that psychological treatments aimed at reducing depression, anxiety, and kinesiophobia are worth trialling in CRPS. Our data
652 | www.clinicalpain.com
Clin J Pain
Volume 30, Number 8, August 2014
suggest that it would be worth investigating whether psychological treatments that target depression lead to improvements in disability, and whether treatments that target anxiety lead to improvements in pain. There is already some evidence from multiple single-case design studies showing the efficacy of treatments aimed at kinesiophobia,8,9 and these data suggest that targeting depression and anxiety in addition may be worthwhile. The main limitation of the study is that the data were collected at a single time and the direction of relationships is unclear. It is not only possible but quite likely that while psychological factors might influence pain and disability, that higher pain intensity and greater disability lead to poorer psychological wellbeing. It is also possible that a third (unknown) variable accounts for the relationship. Prospective studies are needed to better understand these relationships. The data are also limited by self-report bias. Another limitation is that there was some missing data, more so in the CRPS than in the LBP group. Systematically missing data or selective attrition are a potential source of bias for this study, as it is possible that the difference between groups was due to the difference between those who complete questionnaires compared with those who do not. However, to investigate this, we reran the analyses utilizing only the 60 CRPS and low back pain matched pairs with complete data sets, and our results were virtually identical to those presented here. One strength of the study was that the samples were matched for age, sex, and pain duration, and this removes some possible sources of variation in measures of pain, disability, and psychological wellbeing. In conclusion, we found that there were remarkable similarities between CRPS patients and age-matched, sexmatched, and pain duration-matched LBP patients on measures of pain, disability, depression, anxiety, and kinesiophobia. In addition, we found that both pain and depression were associated with disability, and the association between depression and disability was stronger for the CRPS group than the LBP group. We found that kinesiophobia was associated with greater pain intensity, and that anxiety was associated with pain intensity in the CRPS group but not in the LBP group. Although previous research has demonstrated a limited role for psychological factors in causing CRPS, our results support the hypothesis that in chronic CRPS, psychological factors might play a role in maintaining pain and disability in the same way they influence other types of acute, experimental, and chronic pain. ACKNOWLEDGMENTS The authors thank Catherine Pollard, BSc (Hons), Physiotherapy Diploma HE Orthopaedic Medicine, Trevor Coe, MA, MB BChir, FRCA, FFPMANZCA, Department of Anaesthetist and Pain Medicine, The Auckland Regional Pain Service, Auckland District Health Board, Auckland, New Zealand and the rest of The Auckland Regional Pain Service team for assistance with data collection, and the Oakley Mental Health Research Foundation for supporting our CRPS Research. REFERENCES 1. Morley S. Psychology of pain. Br J Anaesth. 2008;101:25–31. 2. Beerthuizen A, van t Spijker A, Huygen FJ, et al. Is there an association between psychological factors and the Complex Regional Pain Syndrome type 1 (CRPS1) in adults? A systematic review. Pain. 2009;145:52–59. r
2013 Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 8, August 2014
3. Lohnberg JA, Altmaier EM. A review of psychosocial factors in complex regional pain syndrome. J Clin Psychol Med Settings. 2013;20:247–254. 4. Beerthuizen A, Stronks DL, FJ Huygen, et al. The association between psychological factors and the development of complex regional pain syndrome type 1 (CRPS1)—A prospective multicenter study. Eur J Pain. 2011;15:971–975. 5. Dilek B, Yemez B, Kizil R, et al. Anxious personality is a risk factor for developing complex regional pain syndrome type I. Rheumatol Int. 2012;32:915–920. 6. Harden RN, Bruehl S, Stanos S, et al. Prospective examination of pain-related and psychological predictors of CRPS-like phenomena following total knee arthroplasty: a preliminary study. Pain. 2003;106:393–400. 7. Puchalski P, Zyluk A. Complex regional pain syndrome type 1 after fractures of the distal radius: a prospective study of the role of psychological factors. J Hand Surg [Br]. 2005;30:574–580. 8. de Jong JR, Vlaeyen JWS, Onghena P, et al. Reduction of painrelated fear in complex regional pain syndrome type I: the application of graded exposure in vivo. Pain. 2005;116:264–275. 9. van de Meent H, Oerlemans M, Bruggeman A, et al. Safety of “pain exposure” physical therapy in patients with complex regional pain syndrome type 1. Pain. 2011;152:1431–1438. 10. Goebel A, Barker CH, Turner-Stokes L, et al. Complex Regional Pain Syndrome in Adults: UK Guidelines for Diagnosis, Referral and Management in Primary and Secondary Care. London: Royal College of Physicians; 2012. 11. Nicholas MK, Linton SJ, Watson PJ, et al. Early identification and management of psychological risk factors (“yellow flags”) in patients with low back pain: a reappraisal. Phys Ther. 2011;91:737–753. 12. Iles RA, Davidson M, Taylor NF. Psychosocial predictors of failure to return to work in non-chronic non-specific low back pain: a systematic review. Occup Environ Med. 2008;65: 507–517. 13. Heitz CAM, Hilfiker R, Bachmann LM, et al. Comparison of risk factors predicting return to work between patients with subacute and chronic non-specific low back pain: systematic review. Eur Spine J. 2009;18:1829–1835. 14. Chou R, Shekelle P. Will this patient develop persistent disabling low back pain? JAMA. 2010;303:1295–1302. 15. Henschke N, Ostelo RW, van Tulder MW, et al. Behavioural treatment for chronic low-back pain. Cochrane Database Syst Rev. 2010;CD002014. 16. Nelson DV, Novy DM. Psychological characteristics of reflex sympathetic dystrophy versus myofascial pain syndromes. Region Anaesth. 1996;21:202–208. 17. Monti DA, Herring CL, Schwartzman RJ, et al. Personality assessment of patients with complex regional pain syndrome type I. Clin J Pain. 1998;14:295–302. 18. Katz MM, Hungerford DS, Krackow KA, et al. Reflex sympathetic dystrophy as a cause of poor results after total knee arthroplasty. J Arthroplasty. 1986;1:117–124. 19. Zucchini M, Alberti G, Moretti MP. Algodystrophy and related psychological features. Funct Neurol. 1989;4:153–156. 20. Feliu MH, Edwards CL. Psychologic factors in the development of complex regional pain syndrome: history, myth, and evidence. Clin J Pain. 2010;26:258–263. 21. Feldman SI, Downey G, Schaffer-Neitz R. Pain negative mood, and perceived support in chronic pain patients: a daily diary study of people with reflex sympathetic dystrophy syndrome. J Consult Clin Psych. 1999;67:776–785.
r
2013 Lippincott Williams & Wilkins
Psychological Factors in CRPS and Low Back Pain
22. Bruehl S, Chung OY, Burns JW. Differential effects of expressive anger regulation on chronic pain intensity in CRPS and non-CRPS limb pain patients. Pain. 2003;104:647–654. 23. Moseley GL. Imagined movements cause pain and swelling in a patient with complex regional pain syndrome. Neurology. 2004;62:1644. 24. de Jong JR, Vlaeyen JWS, de Gelder JM, et al. Pain-related fear, perceived harmfulness of activities, and functional limitations in complex regional pain syndrome type I. J Pain. 2011;12:1209–1218. 25. Harden RN, Bruehl S, RSGM Perez, et al. Development of a severity score for CRPS. Pain. 2010;151:870–876. 26. Bruehl S, Husfeldt B, Lubenow TR, et al. Psychological differences between reflex sympathetic dystrophy and nonRSD chronic pain patients. Pain. 1996;67:107–114. 27. Merskey H, Bogduk N. Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms. 2nd ed. Seattle: IASP Press; 1994. 28. Tait RC, Chibnall JT, Krause S. The Pain Disability Index: psychometric properties. Pain. 1990;40:171–182. 29. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiat Scand. 1983;67:361–370. 30. Miller RP, Kori SH, Todd DD. The Tampa Scale for Kinesiophobia (Unpublished Report). Tampa, FL; 1991. 31. Erdfelder E, Faul F, Buchner A. GPOWER: a general power analysis program. Behav Res Meth Instrum Comput. 1996;28: 1–11. 32. Preacher KJ. Calculation for the test of the difference between two independent correlation coefficients. May 2002. Available at: http://quantpsy.org. Accessed September 20, 2012. 33. Chou R, Qaseem A, Snow V, et al. Diagnosis and treatment of low back pain: a joint clinical practice guideline from the American College of Physicians and the American Pain Society. Ann Intern Med. 2007;147:478–491. 34. DeGood DE, Cundiff GW, Adams LE, et al. A psychosocial and behavioral comparison of reflex sympathetic dystrophy, low back pain, and headache patients. Pain. 1993;54:317–322. 35. Ciccone DS, Bandilla EB, Wu W. Psychological dysfunction in patients with reflex sympathetic dystrophy. Pain. 1997;71: 323–333. 36. Nicholas MK, Asghari A, Blyth FM. What do the numbers mean? Normative data in chronic pain measures. Pain. 2008;134:158–173. 37. Chibnall JT, Tait RC. The Pain Disability Index: factor structure and normative data. Arch Phys Med Rehabil. 1994;75:1082–1086. 38. Drummond PD. Sensory disturbances in complex regional pain syndrome: clinical observations, autonomic interactions, and possible mechanisms. Pain Med. 2010;11:1257–1266. 39. Rommel O, Willweber-Strumpf A, Wagner P, et al. Psychische veranderungen bei patienten mit komplexem regionalem schmerzsyndrom (CRPS) [Psychological abnormalities in patients with complex regional pain syndrome (CRPS)]. Schmerz. 2005;19:272–284. 40. Birklein F, Schmelz M. Neuropeptides, neurogenic inflammation and complex regional pain syndrome (CRPS). Neurosci Lett. 2008;437:199–202. 41. Schurmann M, Gradl G, Zaspel J, et al. Peripheral sympathetic function as a predictor of complex regional pain syndrome type I (CRPS I) in patients with radial fracture. Auton Neurosci. 2000;86:127–134.
www.clinicalpain.com |
653
