Quantitative evaluation of shoulder proprioception 6 months following stroke

Stroke is a global problem with a burden that continues to increase due to population growth and aging [1]. Estimated prevalence of stroke in Egypt is 722/100,000 [2] with a higher prevalence among men and illiterate patients [3] which are by majority manual worker.

Improvement of quality of life (QoL) following stroke occurs up to 6 months following the event [4]. Shoulder pain is a major problem that occurs in nearly third of stroke patients with a majority ranging between moderate and severe degree. It appears around the fourth month [5] and persists in 21% of patients 16 months following stroke [6].

The sensorimotor system is defined as all of the sensory, motor, and central integration and processing components involved in maintaining joint stability [7]. Proprioceptive impairment as much as motor deficit is of outmost importance in planning of future management of stroke patients [8]. Chronic post-stroke shoulder pain (PSSP) has been attributed to altered kinematics of the joint leading to secondary soft tissue damage [9]. Rehabilitation of position sense showed promising results to avoid such problem [10].

The aim of this study is to evaluate shoulder proprioception in the ipsilateral paretic arm and contralateral normal side 6 months following the cerebrovascular event.

This is a case-control study including 30 adults with anterior circulation ischemic cerebrovascular stroke patients (group 1) and 30 control healthy subjects (group 2) matching for age, sex, weight, and height. Patients were recruited from outpatient clinic of neuromuscular disorders and its surgery, Faculty of Physical Therapy, Cairo University. Included patients have stroke for more than 6 months and less than 1 year. The patients should have sufficient cognitive abilities to enable them understanding the requirements of the study. We excluded patients with frozen shoulder or any other orthopedic problems affecting the shoulders, language or hearing deficits, and diabetes mellitus. All patients had an almost equal physiotherapy program prior to the recruitment as they were following the standard program of the physical therapy department of neuromuscular disorders and its surgery, Faculty of Physical Therapy, Cairo University.

Both patients’ shoulders were assessed. The power of the affected arm ranged from 3 to 4+ according to “Medical Research Council Scale (MRC)” and spasticity from 1 to 1+ according to “Modified Aschworth Scale (MAS)” [11] to assure enough control of the patient over his range of movement. The control group (G2) was examined in the dominant shoulder only. According to brain imaging, patient’s lesions were subdivided according to site of the lesion into right or left and cortical (n = 5) or subcortical (n = 18). Patients with mixed cortical and subcortical lesions were excluded (n = 7).

The statistical analysis was done using the Statistical Package of Social Science Software program, version 21 (SPSS, IBM Inc., Chicago, IL). The descriptive data were summarized using the arithmetic mean and the standard deviation for quantitative variable and the frequency percentage for qualitative variables. Comparison of the data within the same group was done using the paired t test and between groups using the unpaired t test.

Subgroups with left, right, or bilateral lesions showed no statistical difference in degree of angular error of the affected arm in both active and passive external or internal rotations (P > 0.05).

This study demonstrates that shoulder proprioception deficit is an underestimated problem that persists beyond 6 months in paretic arm of stroke patients despite adequate physical therapy. This deficit may be more common in patients presenting with cortical lesions.

The upper limit of age for patients and controls was 63 years. Anatomical and physiological changes of the central and peripheral nervous system affect proprioception with advancing of age [13, 14]. We chose patients with stroke onset at least 6 months prior to inclusion period as maximum recovery of sensory functions usually occurs within this period [15] although continuous response to rehabilitation programs was shown to extend beyond this limit of time [16].

Patient group (G1) with cerebrovascular stroke (> 6 months) had an increased absolute angular displacement error in all tested positions (active external rotation, passive external rotation, active internal rotation, and passive internal rotation) when compared to unaffected contralateral arm and same aged control group (G2) (P > 0.05). Niessen et al. showed similar results when he compared proprioception of the shoulder in the paretic arm compared to controls [12]. Yet he showed that the contralateral arm as well was affected to some extent unlike our results that shows no affection of the contralateral side. This difference can be attributed to the time of testing; he chose patients with sub-acute stroke, but our patients were assessed 6 months following stroke. This time lapse can be sufficient for contralateral recovery but not for the paretic side. Normal function of the shoulder was stated to depend on shoulder stability; static stabilization that depends on capsulolabral complex and dynamic stabilization that depends on scapulo-thoracic and glenohumeral muscles both coordinated by neuromuscular control [17]. Gooey et al. [18] showed that the muscle spindle is the outmost important structure in determining joint position. The shoulder seems to contain very few mechanoreceptor [19] that is why Neissen et al. [12] said that the kinesthetic deficit in stroke patients is primarily due to gamma motor neuron dysfunction of muscle spindles sensitivity so there is a delay in detection of muscle stretch, thus creating the cascade of shoulder instability, soft tissue injury, and chronic pain [9].

Angular error of passive external and internal rotation of the shoulder was statistically higher in patients with cortical lesions (n = 5) compared to patients with subcortical lesions (n = 18) (P = 0.009 and 0.000 respectively). In a previous study, patients with cortical lesions were shown to recover more rapidly than patients with subcortical lesions, yet recovery was used in terms of motor function only [20]. Proprioception deficit seems to be different as it is more linked to cortical sensory integrating regions. Right supramarginal gyrus was proved to play a major role in proprioception loss in stroke patients [21] thus may be explaining prolonged recovery of proprioception in patients with cortical lesions. On the other hand, relation between lesion site (right or left) and proprioceptive deficit was rejected by Dukelow et al. [8], which support our results where no difference was found between site or bilaterality of the lesion and proprioception deficit. Yet the small number used in both studies could not give conclusive results.

Chronic shoulder proprioceptive deficit persists beyond 6 months following cerebrovascular event but only in the affected paretic arm. This persistence may be linked to cortical lesions rather than subcortical ones. A new perspective in rehabilitation of stroke patients should include an extended program targeting deep sensory system as much as motor system through the first year following the event in order to achieve a better quality of life increasing independency and decreasing chronic debilitating pain.