Pathological features of post-stroke pain: a comprehensive analysis for subtypes

Post-stroke pain is heterogeneous and includes both nociceptive and neuropathic pain. These subtypes can be comprehensively assessed using several clinical tools, such as pain-related questionnaires, quantitative somatosensory tests and brain imaging. In the present study, we conducted a comprehensive assessment of patients with central post-stroke pain and non-central post-stroke pain and analysed their clinical features. We also performed a detailed analysis of the relationships between brain lesion areas or structural disconnection of the white matter and somatosensory dysfunctions. In this multicentre cross-sectional study, 70 patients were divided into 24 with central post-stroke pain, 26 with non-central post-stroke pain and 20 with no-pain groups. Multiple logistic regression analysis was used to summarize the relationships between each pathological feature (for the central post-stroke pain and non-central post-stroke pain groups) and pain-related factors or the results of quantitative somatosensory tests. Relationships between somatosensory dysfunctions and brain lesion areas were analysed using voxel-based lesion-symptom mapping and voxel-based disconnection-symptom mapping. All pathology feature models indicated that central post-stroke pain was associated with cold hypoesthesia at 8°C (β = 2.98, odds ratio = 19.6, 95% confidence interval = 2.7-141.8), cold hyperalgesia at 8°C (β = 2.61, odds ratio = 13.6, 95% confidence interval = 1.13-163.12) and higher Neuropathic Pain Symptom Inventory scores (for spontaneous and evoked pain items only; β = 0.17, odds ratio = 1.19, 95%, confidence interval = 1.07-1.32), whereas non-central post-stroke pain was associated with joint pain (β = 5.01, odds ratio = 149.854, 95%, confidence interval = 19.93-1126.52) and lower Neuropathic Pain Symptom Inventory scores (β = -0.17, odds ratio = 0.8, 95%, confidence interval = 0.75-0.94). In the voxel-based lesion-symptom mapping, the extracted lesion areas indicated mainly voxels significantly associated with cold hyperalgesia, allodynia at 8°C and 22°C and heat hypoesthesia at 45°C. These extracted areas were mainly in the putamen, insular cortex, hippocampus, Rolandic operculum, retrolenticular part of internal and external capsules and sagittal stratum. In voxel-based disconnection-symptom mapping, the extracted disconnection maps were significantly associated with cold hyperalgesia at 8°C, and heat hypoesthesia at 37°C and 45°C. These structural disconnection patterns were mainly in the cingulum frontal parahippocampal tract, the reticulospinal tract and the superior longitudinal fasciculus with a widespread interhemispheric disconnection of the corpus callosum. These findings serve as important indicators to facilitate decision-making and optimize precision treatments through data dimensionality reduction when diagnosing post-stroke pain using clinical assessments, such as bedside quantitative sensory testing, pain-related factors, pain questionnaires and brain imaging.