Assessment of sympathetic skin response for the diagnosis of post-stroke complex regional pain syndrome

Abstract

Introduction: To investigate whether absolute amplitude differences in sympathetic skin responses (SSR) and ratio of SSR amplitudes between hemiplegic and normal sides can help with the diagnosis of type-I complex regional pain syndrome (CRPS) after stroke. Methods: We included 51 hemiplegic patients (13 with CRPS, 38 without CRPS) undergoing stroke rehabilitation, and 27 healthy age- and sex-matched controls. CRPS was diagnosed according to the Budapest Clinical Diagnostic Criteria proposed by the International Association for the Study of Pain. SSR were measured in both hands by a standardised method of stimulating the median nerve. We compared the ratio of SSR amplitudes between hemiplegic and normal sides for the hemiplegic patients, and the maximum ratio of SSR amplitudes between left and right hands for the healthy controls, using non-parametric tests. Results: Baseline characteristics including age, sex, side of hemiplegia, and types of strokes were similar among groups. The mean SSR amplitudes of hemiplegic patients with CRPS, hemiplegic patients without CRPS, and healthy controls were not significantly different. However, the ratio of SSR amplitudes of hemiplegic to normal sides in hemiplegic patients with CRPS was significantly higher than that in hemiplegic patients without CRPS (P<0.001), and the maximum ratio of SSR amplitude in healthy controls (P<0.001). Receiver operating characteristic analysis revealed an optimal cutoff ratio of 1.8, with associated sensitivity of 92% and specificity of 74%. Conclusion: The ratio of SSR amplitude between the hemiplegic and normal sides, rather than the absolute amplitude differences in SSR, may help with the diagnosis of post-stroke type-I CRPS.