Atypical Central Neuropathic Pain Following Stroke Presenting With Distal Movement-Evoked Shoulder Pain: A Case Report Authors: Joon Sin Ser, Sangita Kuparasundram Published: 2026 ABSTRACT This case report describes a unique presentation of central neuropathic pain (CNP) in a Chinese male in his 40s, following a large right basal ganglia hemorrhage with extension into the right middle cerebral artery (MCA) territory, characterized by left shoulder pain reproducibly triggered by passive wrist and finger extension. The report outlines the sequential treatment approach implemented for pain management. Although spasticity and shoulder pain are common sequelae of stroke, this case highlights the importance of considering central neuropathic mechanisms in patients with atypical or refractory symptoms. It also underscores the heterogeneity of CNP following stroke and the need for heightened clinical awareness of this condition. Shoulder pain is a common complication after stroke, with a reported incidence of 29% and prevalence of 33%, and represents a significant source of morbidity, contributing to prolonged rehabilitation, poorer functional recovery, and reduced quality of life.1 The differential diagnosis of poststroke shoulder pain is broad and includes spasticity-related pain, musculoskeletal conditions such as rotator cuff tendinopathy and adhesive capsulitis, as well as complex regional pain syndrome (CRPS) and central neuropathic pain (CNP) such as central post stroke pain syndrome (CPSP).2 Given the overlap in clinical features among these conditions, accurate diagnosis requires a systematic and methodical approach. Common mechanisms contributing to poststroke shoulder pain include a combination of motor weakness and sensory deficits, which may result in glenohumeral subluxation and subsequent musculoskeletal injury, such as rotator cuff tendinopathy. Patients with hemispatial neglect are at increased risk because of impaired protective awareness and positioning of the affected limb. Spasticity is another important contributing factor, particularly involving the pectoralis major, subscapularis muscles, which may lead to abnormal shoulder biomechanics.3 Meanwhile, CNP has complex underlying mechanisms that are not fully understood. The pathophysiology is thought to involve neuronal hyperexcitability with central sensitization or disinhibition, maladaptive plasticity and disruption of spinothalamocortical tract pathways within the brain and spinal cord.4,5 In this report, we describe an unusual presentation of CNP with sharp shoulder pain reproducibly triggered by passive wrist and finger extension following a right basal ganglia hemorrhage with extension into the right middle cerebral artery (MCA) territory. This anatomically discordant, stimulus-dependent pattern mimicked a peripheral shoulder pathology and, to our knowledge, has not been previously reported. This case underscores the importance of considering CNP mechanisms particularly when pain characteristics are incongruent with peripheral anatomical or biomechanical explanations and refractory to treatment. The report adheres to the CARE (Case Report) guidelines, and written informed consent was obtained from the patient. Institutional Research Ethics Board approval was not required. Case presentation A Chinese male in his late 40s with a background of hypertension presented in December 2024 to an acute tertiary hospital with acute onset left-sided weakness associated with drowsiness. Prior to admission, he was independent in activities of daily living and community ambulant without assistive devices. Computed tomography (CT) imaging of the brain reported a large acute intracerebral hemorrhage centered within the right basal ganglia with extension into the adjacent cortical regions within the right MCA territory (fig 1). He was admitted to the neurosurgery department, and subsequently underwent a right craniotomy with evacuation of the haematoma under the neurosurgical service.Fig. 1Axial computed tomography (CT) image of the brain demonstrating an acute intraparenchymal haematoma centered within the right basal ganglia, measuring 6.3×3.3×3.4 cm, with extension into the right lateral ventricle and right middle cerebral artery (MCA) territory.Fig 1 dummy alt text Axial computed tomography (CT) image of the brain demonstrating an acute intraparenchymal haematoma centered within the right basal ganglia, measuring 6.3×3.3×3.4 cm, with extension into the right lateral ventricle and right middle cerebral artery (MCA) territory. Following a 29-day acute hospital stay, he was deemed medically stable and transferred to an inpatient rehabilitation unit. He exhibited severe left hemispatial neglect with cognitive impairment (Montreal Cognitive Assessment score 22 out of 30) but retained functional communication ability. He had left hemiplegia with Medical Research Council (MRC) grade of 0 out of 5 for strength on motor examination of the left upper and lower limbs. Sensory examination revealed absent light-touch sensation over the left side of the body. Spasticity was noted with a Modified Ashworth Scale (MAS) score of 2 in the left finger flexors and 1+ in the elbow flexors. There was no spasticity of the shoulder muscles. Oral baclofen was initiated at 5 mg twice daily for spasticity management. He received multidisciplinary therapy including physiotherapy, occupational therapy, speech therapy, and music therapy at the rehabilitation unit. At the patient’s request, he also received weekly acupuncture sessions. After approximately 2 weeks of inpatient rehabilitation (42 days postadmission), patient reported left shoulder pain (fig 2). The pain was described as a sharp, ‘poking’ sensation occurring with passive movement and absent at rest. Passive shoulder range of motion was largely preserved though he experienced pain with ranging of the left shoulder to 120° of abduction and 140°of forward flexion, rated 3 out of 10 in severity. Notably, passive extension of the wrist and fingers beyond the neutral position reproducibly elicited severe sharp pain localized to the left shoulder, despite stabilization of the elbow during the maneuver. The pain was rated 8 out of 10 in severity. He denied having any focal pain at the fingers or wrist with this maneuver. There was no neck pain or radicular symptoms.Fig. 2Timeline of investigations and treatment.Fig 2 dummy alt text Timeline of investigations and treatment. Oral paracetamol, tramadol, and topical ketoprofen were prescribed with minimal benefit. Baclofen was continued but later reduced to 5 mg at night because of daytime somnolence. Acupuncture sessions provided only transient relief, with rapid recurrence of pain after treatment. Progressive worsening of finger flexor spasticity (MAS 3) was observed and appeared to exacerbate shoulder pain during passive stretching. Botulinum toxin injections were thus administered to the flexor digitorum superficialis and flexor digitorum profundus. One week following the injections, spasticity in the finger flexors improved to MAS 1. This was accompanied by reduction of shoulder pain during finger extension to 4 out of 10. Given persistent symptoms, further investigations were undertaken. A left shoulder radiograph was performed which demonstrated osteoarthritic changes involving the glenohumeral and acromioclavicular joints. Left shoulder ultrasound examination revealed a subacromial-subdeltoid bursa measuring 37 mm with increased vascularity, consistent with bursitis. Mild peribicipital edema was also reported, while rotator cuff tendons remained intact. An ultrasound-guided subacromial-subdeltoid bursal injection of 40 mg of triamcinolone and 4 mL 1% lignocaine was administered. Pain persisted at 5 out of 10 following the procedure. Pregabalin was initiated at 75 mg nightly and gradually titrated to 300 mg nightly without significant improvement in pain. An ultrasound-guided suprascapular nerve block was performed using 40 mg triamcinolone and 5 mL 1% lignocaine and similarly failed to provide pain relief. A triphasic bone scan was performed to evaluate for CRPS and demonstrated asymmetric delayed radiotracer uptake without increased flow or blood pool activity which was indeterminate for CRPS. Clinically, the patient did not meet Budapest diagnostic criteria, with no evidence of allodynia, edema, temperature asymmetry, or trophic changes.6 Duloxetine 30 mg daily was initiated without significant improvement. Nortriptyline 25 mg nightly was added resulting in mild pain reduction to 3 out of 10 and was subsequently increased to 50 mg nightly. The patient was discharged from inpatient rehabilitation after 4 months. He continued to have left hemiplegia with sensory deficits but demonstrated improvement in hemispatial neglect (Bell’s test score improving from 3 to 20 out of 35). Functionally, he was wheelchair based and required minimal-to-moderate assistance for transfers and self-care. The shoulder pain was considered tolerable, with no functional impact and did not hinder care; therefore, further treatment options were not pursued. He was discharged with pregabalin 300 mg nightly, nortriptyline 50 mg nightly, duloxetine 30 mg daily, and baclofen 5 mg 3 times daily. At outpatient follow-up 3 months after discharge, his shoulder pain and spasticity remained stable, and the existing medication regimen was continued. Discussion The patient developed sharp, severe left shoulder pain consistently elicited by passive extension of the ipsilateral finger and wrist flexors. Although mild discomfort was provoked by direct shoulder movement, the qualitatively distinct and disproportionately intense pain triggered by distal limb movement was clinically incongruent with a primary shoulder pathology. Pain perceived at a site remote from the mechanical stimulus is atypical for musculoskeletal conditions and prompted suspicion for a centrally mediated mechanism. Sensory innervation of the shoulder joint arises primarily via the upper branches of the brachial plexus (C3-C4 nerve roots), including the suprascapular, axillary, lateral pectoral, and lower subscapular nerves.7 In contrast, the ventral forearm muscles, including the finger flexors (flexor digitorum superficialis and flexor digitorum profundus), are innervated by the median nerve (C5-T1 nerve roots) and ulnar nerve (C8-T1 nerve roots). The absence of shared peripheral innervation between these regions makes classical referred pain improbable and supports a centrally mediated explanation for pain perception. CNP is classically associated with thalamic lesions as in the Dejerine-Roussy syndrome; however, involvement of other structures such as the basal ganglia and medulla, has also been implicated because of their role in sensorimotor integration and modulation of pain processing pathways.8 In CPSP, the timing of onset of symptoms can be variable with some patients having immediate onset but most develop symptoms around 1 month poststroke.9 This patient developed symptoms 42 days postevent. However, a contributing factor may have been the patient’s gradual recovery of attention and sensory awareness of the affected limb with improvement in hemispatial neglect following rehabilitation. This may have ‘unmasked’ pain that was previously not perceived or reported. However, this presentation was not in keeping with typical features of CPSP as defined by the International Association for the Study of Pain which commonly includes the presence of neuropathic symptoms of allodynia or hyperalgesia. Most patients with CPSP experience medium to high intensity of pain, with common pain qualities of ‘burning’, ‘tingling’, and ‘tearing’.9 Diagnostic criteria for CPSP may however also include negative sensory signs such as raised thresholds to temperature or reduced sensation to pinprick or light touch.5 A comprehensive systematic evaluation was also undertaken to exclude common contributors of poststroke shoulder pain. Structural abnormalities including osteoarthritic changes and subacromial bursitis were addressed with targeted interventions such as corticosteroid injections and suprascapular nerve block. These measures failed to provide sustained relief, and the character of the pain remained unchanged. Although worsening finger flexor spasticity appeared to exacerbate pain during passive extension, reduction in tone mediated by botulinum toxin injections resulted only in partial improvement. The persistence of the same anatomically discordant pattern despite resolution of potential peripheral drivers argues against a primary nociceptive source. Collectively, this reduces the likelihood of a musculoskeletal or peripherally mediated explanation. Furthermore, the patient demonstrated partial symptomatic improvement following treatment with centrally acting neuropathic pain agents. This incomplete treatment response is consistent with existing literature regarding CNP. Management of CNP remains challenging. First-line pharmacological treatment includes gabapentinoids (pregabalin, gabapentin) and antidepressants (duloxetine, venlafaxine, amitriptyline). Other medications that have been studied include anticonvulsants (lamotrigine, carbamazepine) and opioids. Nonpharmacological management include psychological therapy (cognitive behavioral therapy), acupuncture, electroencephalography neurofeedback and even virtual reality interventions.4 Neuromodulation with high frequency repetitive transcranial magnetic stimulation targeting the primary motor cortex (M1) has also had promising results.4 Overall evidence supports a multimodal and individualized treatment strategy tailored to patient response and tolerability. This report has limitations inherent to its design as a single-case study. A definitive diagnosis of CNP could not be established because of the absence of objective biomarkers. Advanced neuroimaging modalities such as functional magnetic resonance imaging, neurophysiological assessments, or quantitative sensory testing, which may have provided further insight into central sensory dysfunction, were not available.4 Pain assessment relied primarily on patient-reported outcomes, which are inherently subjective and may be influenced by cognitive and affective factors following stroke. Specific screening questionnaires such as Douleur Neuropathique en 4 Questions (DN4) were not performed.4,5 Finally, the sequential and overlapping implementation of multiple therapeutic interventions limits conclusions regarding the relative efficacy of individual treatments. Nevertheless, this is reflective of real-world clinical practice in managing complex poststroke pain syndromes. Conclusion This case report describes an atypical presentation of CNP manifesting as sharp shoulder pain reproducibly evoked by passive wrist and finger extension following a basal ganglia hemorrhage with extension into the right MCA territory. The pain pattern was anatomically discordant with peripheral neural pathways and persisted despite systematic exclusion of potential musculoskeletal and peripheral causes. Furthermore, the patient demonstrated partial responsiveness to centrally acting neuropathic pain medications. These findings highlight the heterogeneous clinical presentation of CNP after stroke and also underscore the complex interaction among spasticity, musculoskeletal factors, and central sensory dysfunction that can make clinical management of pain in a poststroke patient challenging. Clinicians should maintain an index of suspicion for CNP in patients with refractory poststroke shoulder pain, particularly when pain features are atypical, stimulus-dependent, or disproportionate to peripheral findings. There are no pathognomonic or uniform characteristics of CNP and it is a diagnosis of exclusion.8 A structured and systematic diagnostic approach is essential when evaluating poststroke shoulder pain, with stepwise exclusion of peripheral causes while maintaining awareness of centrally mediated pain syndromes. Early recognition of such presentations may facilitate timely initiation of appropriate neuropathic pain management. Further research is needed to better characterize atypical pain patterns in CNP and to elucidate the underlying neurophysiological mechanisms, including the potential role of advanced neuroimaging. Improved understanding may support the development of more precise diagnostic frameworks and evidence-based, individualized treatment strategies for this challenging and often under-recognized condition. Author contributions JSS contributed to the study design, analysis of results and writing of the manuscript, SK supervised the design of the case report, writing of the manuscript and coordination of the case report. Ethical declaration Institutional Research Ethics Board approval was not required for this case report. Data statements All data generated or analyzed during this study are included in this published article. Disclosure This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.