Upper limb pain and injury are highly prevalent in people with SCI and consequences are significant. The Consortium for Spinal Cord Medicine (2005) and Dyson-Hudson and Kirshblum (2004) reported through surveys and cross-sectional studies that shoulder pain in chronic spinal cord injured persons are common in both paraplegia and tetraplegia. The incidence of shoulder pain in acute tetraplegia (less than six months post injury) has been reported to range from 51%-78% (Salisbury et al., 2003). In the acute phase after SCI, shoulder pain is reported in approximately 75% of patients (Silfverskiold & Waters 1991; Waring & Maynard 1991), and 33%-63% of patients in the chronic phase (greater than 6 months) experience shoulder pain (Nepomuceno et al., 1979; Sie et al., 1992; Silfverskiold & Waters 1991). Curtis et al. (1995a), Nichols et al. (1979) and Pentland and Twomey (1991) reported in cross sectional studies that 60-100% of long-term wheelchair users experience shoulder pain. Subbardo et al. (1994) and Sie et al. (1992) reported that shoulder pain is still prevalent in individuals 15-20 years’ post injury. Pain experienced above the injury level during the first three to six months after injury is different than the pain experienced one year or more after injury (Apple 2001). Pain above the level of injury in chronically injured person assumes the character of overuse syndromes (Apple 2001). Injury involving the shoulder, elbow, wrist, and hand are seen at an earlier age in spinal cord injured individuals than in the general population because of the stresses of weight bearing and mobility that are added to the normal use of the upper limb (Pentland & Twomney 1994). Individuals who are older at the time of injury may experience functional changes earlier than people who are injured at a younger age (Thompson 1999), this results in difficulty determining whether shoulder pain is a function of duration of SCI or simply a part of the normal aging process (Neer & Welsh 1977).
The range in prevalence of shoulder pain is likely a reflection of the heterogeneity of participant populations between the studies with respect to duration of injury, age, neurologic level and severity of injury, as well as body mass index. Small sample size, selection bias, and variations in participant populations across the different studies make it difficult to assess the true prevalence of shoulder pain in individuals with shoulder pain (Pentland & Twomey 1991).
Nichols et al. (1979) was one of the first groups to report an association between chronic SCI and shoulder pain coining the term “wheelchair user’s shoulder”. Due to the prevalence of shoulder pain, Curtis et al. (1995b) developed a Wheelchair Users’ Shoulder Pain Index (WUSPI) that measures the severity of pain for 14 functional activities.
Table 10: Shoulder Injury Intervention
Outcome studies of surgical treatment in SCI also very limited. Two small studies report the outcome of rotator cuff repair – one showing relatively poor results (Goldstein et al., 1997) and another study showing relatively good outcomes (Robinson et al., 1993). Both studies recommend non-surgical approaches prior to surgical intervention. One RCT found that supervised exercise produced results similar to arthroscopic surgery for patients with impingement syndrome (Brox et al., 1993), however; this study did not include SCI patients.
Exercise has been shown to reduce pain in a RCT in which subtypes of pain were not reported (Hicks et al., 2003). Two studies found an association between restricted ROM and pain, reduced activity and/or injury (Ballinger et al., 2000; Waring & Maynard 1991). RCT’s incorporating stretching into an exercise program for individuals who use manual wheelchairs found stretching exercises were associated with decreased reported shoulder pain intensity (Curtis et al., 1999; Hick et al., 2003).
One RCT demonstrated that acupuncture was no more effective than Trager Treatment in the treatment of shoulder pain (Dyson-Hudson et al., 2001). There are several studies that address the use of complementary or alternative medicine (CAM) with the spinal cord population, which is used at similar rates to the general population. It was reported that the most common reason CAM was used, was for dissatisfaction with conventional medicine for treatment of chronic pain (Nayak et al., 2001). The only CAM technique evaluated in the SCI population is acupuncture although studies do not provide conclusive evidence of effectiveness (Dyson-Hudson et al., 2001; Nayak et al., 2001; Rapson et al., 2003).
The following are the many identified risk factors for the development of injury and pain in the shoulder:
- The shoulder is the most common joint above the level of injury where pain complaints are reported with persons with paralysis (tetraplegia or paraplegia) (Apple 2001).
- The shoulder is not well designed to handle the higher intra-articular pressures required for both weight bearing and mobility (Apple 2001).
- Partial innervation and impaired balance of shoulder, scapular and thoracolumbar muscles place individuals with tetraplegia at a higher risk for developing shoulder pain especially during weight-bearing upper limb activities such as wheelchair propulsion, transfers, and pressure reliefs.
- Due to differences in trunk postural control, differences may also occur between individuals with high paraplegia (T2-T7) and low paraplegia (T8-T12).
- Individuals with C1-C4 motor levels of injury are also at risk for shoulder pain.
- SCI severity also may be associated with shoulder pain (Dyson-Hudson & Kirshblum 2004).
- Lack of use of immobilization of the shoulder girdle muscles can limit their active joint movement and lead to muscle shortening and shoulder capsule tightness.
- The development of pain is associated with decreased shoulder ROM.
- Weakness and paralysis in these muscles can lead to increased reliance on the trapezius, which can result in overuse and pain in this muscle.
- Shoulder pain can occur from nerve root injury or radicular pain with dysesthesias or phantom sensations.
- People of certain age groups, those with higher cervical lesions, and those with shorter lengths of bed rest may be at a greater risk.
- Gender may be associated with shoulder pain in individuals with SCI (Pentland & Twomey 1991).
- Body mass index (BMI) also may play a role in shoulder injuries in manual wheelchair using individuals with SCI because it directly relates to the amount of physical strain experienced during ADLs in these individuals (Boninger et al., 2001; Jensen et al., 1996).
- Shoulder pain is more common in individuals with tetraplegia and complete injuries and in women and duration of injury, older age, and higher BMI all may be risk factors for developing shoulder pain and/or abnormalities in persons with SCI (Dyson-Hudson & Kirshblum 2004).
There is level 1b evidence (from two randomized controlled trials; Hicks et al., 2003; Curtis et al., 1999) that a shoulder exercise and stretching protocol reduces the intensity of shoulder pain post SCI.
There is level 1b evidence (from one randomized controlled trial; Dyson-Hudson et al. 2001) that general acupuncture is no more effective than Trager therapy in reducing post-SCI upper limb pain.
Shoulder exercise and stretching protocol reduces post SCI shoulder pain intensity.
Acupuncture and Trager therapy may reduce post-SCI upper limb pain.
Prevention of upper limb injury and subsequent pain is critical.