Management of Central Post-Stroke Pain: Systematic Review and Meta-Analysis

Abstract

Central post stroke pain (CPSP) is a neuropathic pain condition prevalent in 8% to 35% of stroke patients. This systematic review and meta-analysis aimed to provide insight in the effectiveness of available pharmacological, physical, psychological, and neuromodulation intervention in reducing pain in CPSP patients. Secondary outcomes included mood, sleep, global impression of change, and physical responses. Data extraction included participant demographics, stroke aetiology, pain characteristics, pain reduction scores, and secondary outcome metrics. Forty two original studies were included with a total of 1451 participants. Twelve studies met requirements for a random-effects meta-analysis which found: neuromodulation to be the most effective with a moderate effect on pain scores (SMD = -0.60, 95% confidence interval [-0.97, -0.23]), followed by physical interventions with moderate effect (SMD = -0.55, [-1.28, 0.18), and pharmacological interventions with a small effect on pain (SMD = -0.36, [-0.68, -0.03]). Fourteen studies were included in proportional meta-analysis with pharmacological studies having a moderate effect (58.34% mean reduction, [36.50, 80.18]), and neuromodulation studies a small effect (31.70% mean reduction, [21.44, 41.96]). Sixteen studies were included in the narrative review. While the overall medium risk of bias limits generalisation of findings, fluvoxamine and repetitive transcranial magnetic stimulation was found to have consistently good pain alleviation and relatively low risk of side effects. Anticonvulsants were found to have a significant effect on pain reduction, but were found to have the most side effects. Virtual reality and acupuncture show promising results, but lack rigorous methodological investigation to understand their full effect.