P45-S Optimisation of the recording parameters of tibial somatosensory evoked potentials (SEPs)

Abstract

Background Tibial SEPs assess the function of the dorsal columns and supraspinal medial lemniscal somatosensory pathways. Lumbar N22 responses (T12/L1) and cervical P30 (Cv7) allow topographic localisation but are challenging to obtain in routine clinical practice. Here, we studied technical parameters that affect spinal and cortical (P40) responses. Materials and methods In a sample of consecutively referred patients, spinal response amplitudes and latencies were studied simultaneously with cup or subdermal needle and different reference electrode montages and correlated with BMI. P30 recordings were trialled and P40 amplitude optimisation was studied with three different montages (Cz’-Fz, Cz’-Cc, Ci-Cc). Results Responses with subdermal needles were of equal or smaller amplitude in 22/37 compared to cup electrodes. Contralateral iliac crest reference (0.74 ± 0.62 μV) gave larger amplitudes than ipsilateral (0.59 ± 0.61 μV; P = 0.017). Spinal N22 responses (cup electrode, contralateral reference) were present in 33/37patients with BMI 25(0.58 ± 0.54 μV) and were not strongly correlated with BMI (Spearman = −0.32). 30/136 recordings had absent spinal responses with recordable P40. A P30 response was not recordable in 115/140 when P40 was present. Amplitudes of Cz’-Cc (3.09 ± 2.01 μV) were significantly greater than Cz’-Fz (2.42 ± 1.83 μV) in 117/150 of recordings (P = 0.001) and Ci-Cc (2.48 ± 1.55 μV) in 97/150 (P = 0.008). There were no significant differences in latencies. Conclusions Subdermal needle and cup electrodes produce comparable N20 recordings and contralateral iliac crest reference provides largest amplitudes. Interestingly, BMI does not significantly influence N22 amplitude. The Cv7 response was often absent when N22 and P40 were normal. P40 amplitude is maximal at Cz’-Cc derivation.