Introduction

Individuals with spinal cord injury (SCI) have several disabilities and sequelae such as loss of locomotor function and bladder, bowel, sexual and autonomic dysfunction. Pain and spasticity are other common and disabling consequences. Up to 80% of individuals with SCI complain of chronic pain, and they often experience several types of pain.1, 2, 3 Musculoskeletal pain is the most common pain type seen in about 80%, neuropathic pain is reported by 50–60% and visceral pain, which becomes increasingly prevalent over time, is present in about 30% with long-term SCI.1, 2, 4, 5 Within these broader categories of pain, different subtypes exist. Neuropathic pain felt at and below the level of injury differs in presentation and probably also mechanisms,1, 6, 7 and even within these subgroups, different phenotypes exist with specific combinations of pain descriptors and sensory abnormalities, which may represent different underlying mechanisms.8, 9 Chronic pain is associated with poorer psychological functioning and lower quality of life,8, 10, 11, 12, 13, 14 but few studies have looked into the role of specific pain types on quality of life.15, 16 Spasticity is another common complication present in about 70% of all SCI after the first year.17 Spasticity is a broad term that includes velocity-dependent increase in muscle tone, enhanced tendon reflexes, muscle spasms and clonus.18, 19 Although mild spasticity may be beneficial, severe spasticity may cause functional impairment,20 and spasticity is also suggested to decrease quality of life.21, 22, 23, 24, 25

Pain and spasticity overlap in individuals with SCI. Spasms may cause musculoskeletal pain, and both below-level central pain and spasticity developed over time can be evoked by non-painful stimuli and show spatial extension, suggesting common underlying mechanisms. In addition, both central neuropathic pain and spasms arise from central neuroplastic changes with neuronal hyperexcitability and loss of inhibition, causing decreased threshold, increased responses and expansion of receptive fields in spinal cord neurons.26 Still, very little is known about the correlations between different pain types and spasticity symptoms in individuals with SCI.

Quality of life27 is generally lower among the SCI than in the general population.11, 28, 29, 30, 31 Many factors, including older and lower age at injury, short time since injury, higher level of lesion, low functional ability, low social support, residence (cohabitation), low income, low mobility, marital status (single), neurogenic bowel and bladder dysfunction, spasticity, pain and pain interference, may be associated with lower quality of life in SCI.7, 11, 13, 24, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39

The purpose of this study was to estimate the prevalence, predictors and impact of self-reported pain and spasticity in a large nationwide study. In addition, we aimed to look at specific subtypes of pain and spasticity and their correlations. The study also evaluated the use of cannabis in the present population. These results are presented elsewhere (Andresen et al. unpublished).

Methods

Survey design

This was a cross-sectional anonymous nationwide postal survey in patients who had been hospitalized or been in contact with one of the two SCI outpatient clinics in Denmark from 1990 to 2012. Participants had to be ⩾18 years and have suffered a traumatic SCI. The responses were anonymous as the questionnaire also assessed cannabis use among SCI individuals. These results are reported separately (Andresen et al. unpublished).

Questionnaire

The questionnaire included 35 questions, combining general demographics, SCI characteristics, pain, spasticity, quality of life, sleep disturbance and information on cannabis use, effect and adverse events (Andresen et al. unpublished). As the results on cannabis use are reported elsewhere (Andresen et al. unpublished), this report focuses on pain and spasticity and the impact on quality of life.

Pain

Participants were asked about chronic pain, defined as continuous or daily recurring pain that had been present for >3 months. Participants with chronic pain rated average pain intensity and how much the pain interfered with their day-to-day activities in the last week on a numeric rating scale (NRS, 0–10), where 0 is no pain/no interference and 10 is pain as bad as you can imagine/extreme interference, using parts of the International Spinal Cord Injury Pain Basic Data Set.40 They also filled in the seven-item Douleur Neuropathique 4 questions (DN4 2005, Bouhassira D. All rights reserved).41, 42 The selection of at least three of the seven pain descriptors (burning, painful cold, electric shocks, tingling, pins and needles, numbness and itching) is suggestive of neuropathic pain with a sensitivity of 81.6% and a specificity of 85.7% in the original study41 and of 93% and 75% in SCI individuals.42 Pain treatment was established with a list of multiple treatment options and the possibility to add other treatments.

Spasticity

Participants were asked whether they had spasticity (muscle stiffness or spasms). If they had spasticity, they rated how much the spasticity interfered with their day-to-day activities in the last week on an NRS (0–10). They also rated the intensity of average muscle stiffness intensity in the last week, and spasm frequency was assessed using the Penn Spasm Frequency Scale.43, 44 Participants were given a treatment list for spasticity.

Quality of life

To estimate participants’ quality of life, we used the International Spinal Cord Injury Quality of Life Basic Data Set,45 which asks about satisfaction with life as a whole, with physical health and with psychological health, emotions and mood. Participants were asked to rate these three factors for the past week on an NRS (0–10), 0=completely dissatisfied to 10=completely satisfied. Sleep disturbance during the last week was reported using an NRS (0–10).

Statistical analysis

Data were checked for normality by using histograms and QQ-plots and then described as mean (s.d.) or median (range) or frequency and percentages. Categorical or numerical data were analyzed by using the Chi-square test or Fisher’s exact test and an unpaired t-test or the Mann–Whitney U-test, respectively. For comparisons between three groups (participants with no chronic pain, participants with chronic pain with descriptors not suggestive of neuropathic pain and participants with chronic pain and descriptors suggestive of neuropathic pain), we first used Bartlett’s test to show that there was no significant (P>0.05) difference in variance between the groups; then, we used one-way ANOVA with Bonferroni multiple comparison tests to analyze the differences in quality-of-life outcome scores between the three groups. The factors that showed association with quality-of-life questions in previous studies within this population were included in a multivariate regression model. Data missing in different variables were <2.4%. Statistical analysis was performed with STATA release 12 (StataCorp, College Station, TX, USA). P-values <0.05 were considered statistically significant.

Statement of ethics

Approvals to conduct the study were given by the Danish Health and Medicines Authority, Copenhagen, Denmark (no. 3-3013-621/1) and the Danish Data Protection Agency, Copenhagen, Denmark (no. 1-16-02-210-14). All applicable institutional and governmental regulations concerning the ethical use of human volunteers were followed during the course of this research.

Results

The questionnaire was sent out to 1101 individuals of which 544 responded (response rate 49.4%), and 537 responses were included in the analysis (7 were excluded because questionnaires were incomplete on main outcomes, further details in Andresen et al. unpublished).

Demographics and SCI characteristics

The average age was 54.6 (s.d. 14.6) years and 77% were men (Table 1). Sixty-five percent reported that they were unemployed, 19% had part-time work and 16% full-time work. Fourteen percent (74 of 530) of the participants reported that they drank more alcohol per week than recommended by the Danish Health and Medicines Authority (14 units for men and 7 units for women). Twenty-six percent of the participants smoked cigarettes daily or sometimes. Forty-nine percent were paraplegics, 64% had an incomplete SCI and mean time since injury was estimated to be 18.2 (s.d. 12.8) years. The leading reasons for their traumatic SCI were transportation (41%) and fall (30%). For detailed description, please see Andresen et al. unpublished.

Table 1 Demographics and clinical characteristics of the spinal cord injury (SCI) individuals included
Full size table

Pain

Chronic pain was reported by 73% (390/536). The average pain intensity was 5.6 (s.d. 2.3): 21% reported mild pain (NRS 1–4), 22% moderate pain (NRS 5–6), and 28% severe pain (NRS 7–10). Pain interference in day-to-day activities was 5.0 (s.d. 2.8). Pain treatment was reported by 67% (259 of 390), and the most common type was physiotherapy/training (Figure 1).

Figure 1
figure 1

Treatments reported for pain (n=390) and spasticity (n=378). Other pain treatment included methadone, acupuncture, transcutaneous electrical nerve stimulation, massage, cranio-sacral therapy, cannabis oil, mindfulness/meditation and hot water basin and other treatment for spasticity included physical therapy, botulinum toxin, cannabis and acupuncture.

Full size image

Two hundred and thirty-four participants had a DN4 score of ⩾3 corresponding to 60% of those with pain and 43.6% of the total sample. The average DN4 score among those with pain was 3.1 (s.d. 1.7). Participants with a DN4 score ⩾3 had a significantly higher average pain intensity and pain interference (6.0 (s.d. 2.1) and 5.5 (s.d. 2.7), respectively) than those with a DN4 score <3 (5.0 (s.d. 2.5) and 4.3 (s.d. 2.9), respectively) (P<0.001). The most common descriptors were pins and needles (75%), burning (60%) and tingling (59%) (Table 2). Significantly more tetraplegics than paraplegics reported painful cold (P=0.02), and more participants with an incomplete SCI had painful cold (P=0.01), numbness (P=0.003) and itching (P=0.03) than those with a complete SCI (Table 2).

Table 2 Pain descriptors in spinal cord-injured individuals with different lesion levels and injury severity
Full size table

Spasticity

Spasticity was reported by 71% (378 of 536) participants, of these 356 (94%) reported muscle stiffness and 361 (95.5%) muscle spasms. Among those with spasticity, the average muscle stiffness intensity was 3.9 (s.d. 2.4) and the average interference was 2.9 (s.d. 2.7). Thirty-three percent had mild (NRS 1–3), 23% moderate (NRS 4–6) and 12% severe (NRS 7–10) muscle stiffness. Among those with spasms, the average spasm score was 1.63 (s.d. 0.9); 83% (300/361) had mild or infrequent spasms and 17% (61/361) had more than one spasms per hour (Table 1). One hundred and seventy-two (46%) reported treatment for spasticity. Baclofen was the most frequently used medication for spasticity used by 40% (35% oral and 5% pump) of those with spasticity (Figure 1).

Quality of life

Seventy-seven percent (412/534) stated that their traumatic SCI had at least moderate influence on their daily life (Table 1). The mean quality of life scores was between 5.5 and 6.7, with the lowest score for physical health (Table 1). Participants with pain and a DN4 score ⩾3 had significantly lower quality-of-life scores than those without pain (P<0.001) and lower scores than participants with pain and a DN4 score <3 for physical and mental health (P<0.05) (Figure 2).

Figure 2
figure 2

Quality-of-life scores for participants without pain (n=146), pain and a DN4 score <3 (n=156) and pain and a DN4 score ⩾3 (n=234). *P<0.05; **P<0.001. Error bars are s.e.m.

Full size image

Variables associated with pain and spasticity

We compared demographics, SCI-related factors and quality of life among participants with and without pain, with and without pain with a DN4 score ⩾3 and with and without spasticity in univariate analyses (Table 3). Participants with pain were significantly older, more often female, less often employed and had a significantly higher impact of SCI on everyday life, sleep disturbance and muscle stiffness score and lower quality of life scores than those without pain (Table 3). Participants with pain and a DN4 score ⩾3 also had a higher spasm frequency score than those without. Participants with spasticity were more often tetraplegic and unemployed and had a higher impact of SCI on everyday life and sleep disturbance and lower satisfaction with physical health (Table 3).

Table 3 Characteristic of spinal cord-injured (SCI) individuals with and without pain, pain with specific pain descriptors and spasticity
Full size table

Variables associated with quality of life

Variables suggested to be associated with quality of life in SCI in previous studies7, 11, 13, 24, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39 were included in a multivariate regression model (Table 4). Female gender was associated with lower mental health scores (Table 4). High pain interference and shorter time since injury were associated with lower quality-of-life scores for all three parameters (Table 4). Tetraplegia was associated with lower physical health scores (Table 4). In total, quality-of-life items as life and life situation could be explained by 12.9%, physical health by 22.6% and mental health by 14.8% with the variables included. The quality-of-life scores increased with increasing time since injury up to 40–45 years after injury, after which they decreased again (Figure 3). There was a significant trend over time since injury for the three different parameters of quality of life (P⩽0.001; test for trend across ordered groups).

Table 4 Demographic and clinical characteristics assessed for association with quality of life in a multivariate a regression model
Full size table
Figure 3
figure 3

Mean score on all three quality-of-life items compared to time since injury (years). Error bars are s.e.m.

Full size image

Discussion

In this large nationwide questionnaire study of individuals who had sustained a traumatic SCI at least 3 years prior to the study, we found a prevalence of chronic pain of 73%, of chronic pain with typical neuropathic pain characteristics of 44% and of spasticity of 71%, supporting that these symptoms are frequent in this population.1, 5, 7, 12, 46, 47, 48, 49, 50, 51 Almost 50% of all participants had at least moderate pain and 28% had severe pain. Of those with chronic pain, 67% received treatment, most often physiotherapy and training followed by antiepileptics, paracetamol and opioids, whereas 46% of those with spasticity received treatment for this, most often baclofen. Pain was on average rated as more intense and had a higher interference on daily life than spasticity. In univariate analysis, individuals with spasticity had lower satisfaction with physical health than those without, whereas pain was associated with lower satisfaction in all three domains of life situation, physical health and mental health. Similarly, interference of pain but not spasticity was associated with lower quality of life. Both pain and spasticity were associated with unemployment and sleep disturbance.

Participants with pain and a DN4 score ⩾3, which suggests the presence of neuropathic pain, reported higher pain intensity and pain interference and had a lower satisfaction with their life situation and mental health than those with pain and a DN4 score <3. Pins and needles (75%), burning (60%) and tingling (59%) were the most common pain descriptors. Painful cold, numbness and itching were more often reported in those with an incomplete than a complete SCI, suggesting that, at least in the Danish version, numbness is not a measure of loss of sensation but rather reflects abnormal evoked sensations. Consistent with most previous studies,5 we found no clear relation between pain and injury-specific characteristics, whereas spasticity expectedly was more common in individuals with tetraplegia compared with paraplegia, which includes cauda equina. Individuals with pain were slightly older than those without pain, consistent with other studies,12, 52, 53 and were more often female. Individuals with pain more often had spasticity and higher muscle stiffness than those without pain and those with pain and a DN4 score ⩾3 also had more spasms. Although treatment may impact the presence and severity of pain and spasticity, this may reflect that spasticity can cause musculoskeletal pain and also supports previous suggestions that neuropathic pain and spasms share underlying pathophysiological mechanisms.54 We saw an increase in quality of life with increased time since injury up to 45 years after injury. A similar increase in quality of life has also been described in prospective studies.7, 13 A high pain interference was associated with a low quality of life for all three domains. We have previously shown that an increase in quality of life over time correlated with an increase in change in pain interference,7 suggesting that efforts to minimize pain’s impact are important. Being tetraplegic was only associated with lower satisfaction with physical health. Several other consequences of SCI that have been shown to impact quality of life, such as hand function, bowel, bladder and sexual dysfunction and cardiovascular autonomic dysfunction, were not addressed in our study.11, 13, 29, 34, 35, 37

Limitations

We had a relatively low response rate of 49%, which may be explained by the inclusion of questions related to cannabis use, which is illegal in Denmark. In addition, as the questionnaire for this reason was anonymous, we could not send out reminders or do an analysis of dropouts. However, our population was representative of the Danish SCI population with respect to gender, age and SCI characteristics (Andresen et al. unpublished). This limitation is most important for the frequency; the data on the correlation between symptoms and ‘consequence’ are not influenced as much. Another limitation is that the results rely on self-report. Although pain descriptors may be suggestive of neuropathic pain, we cannot determine the type of pain with certainty without a detailed pain history and clinical examination.

Conclusion

We confirm that pain and spasticity are common and disabling consequences of a traumatic SCI. As a new finding, we show that pain seemed to have more impact on quality of life than spasticity and that pain with descriptors suggestive of neuropathic pain was rated as more severe and with higher interference and associated with lower quality-of-life scores than pain without such descriptors. There was no relation between pain and injury-specific characteristics, but spasticity was more common in individuals with tetraplegia, and there was an association between the presence of pain and spasticity. We found that high pain interference, shorter time since injury, female gender and tetraplegia were associated with lower quality-of-life scores. The study emphasizes the need for attention to pain and spasticity in the SCI population and that further research on improving the treatment of pain and spasticity is important.

Data archiving

There were no data to deposit.