Do somatosensory deficits predict efficacy of neurorehabilitation using neuromuscular electrical stimulation for moderate to severe motor paralysis of the upper limb in chronic stroke?

Keita Tsuzuki; Michiyuki Kawakami; Takuya Nakamura; Osamu Oshima; Nanako Hijikata; Mabu Suda; Yuka Yamada; Kohei Okuyama; Tetsuya Tsuji

doi:10.1177/17562864211039335

Do somatosensory deficits predict efficacy of neurorehabilitation using neuromuscular electrical stimulation for moderate to severe motor paralysis of the upper limb in chronic stroke?

Ther Adv Neurol Disord. 2021 Aug 25:14:17562864211039335. doi: 10.1177/17562864211039335. eCollection 2021.

Authors

Keita Tsuzuki¹, Michiyuki Kawakami², Takuya Nakamura¹, Osamu Oshima¹, Nanako Hijikata¹, Mabu Suda¹, Yuka Yamada¹, Kohei Okuyama¹, Tetsuya Tsuji¹

Affiliations

¹ Department of Rehabilitation Medicine, School of Medicine, Keio University, Tokyo, Japan.
² Department of Rehabilitation Medicine, School of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan.

Abstract

Background: Various neurorehabilitation programs have been developed to promote recovery from motor impairment of upper extremities. However, the response of patients with chronic-phase stroke varies greatly. Prediction of the treatment response is important to provide appropriate and efficient rehabilitation. This study aimed to clarify whether clinical assessments, such as motor impairments and somatosensory deficits, before treatment could predict the treatment response in neurorehabilitation.

Methods: The data from patients who underwent neurorehabilitation using closed-loop electromyography (EMG)-controlled neuromuscular electrical stimulation were retrospectively analyzed. A total of 66 patients with chronic-phase stroke with moderate to severe paralysis were included. The changes from baseline in the Fugl-Meyer Assessment-Upper Extremity (FMA-UE) and the Motor Activity Log-14 (MAL-14) of amount of use (AOU) and quality of movement (QOM) were used to assess treatment response, and multivariate logistic regression analysis was performed using the extracted candidate predictors, such as baseline clinical assessments, to identify predictors of FMA-UE and MAL-14 improvement.

Results: FMA-UE and MAL-14 scores improved significantly after the intervention (FMA-UE p < 0.01, AOU p < 0.01, QOM p < 0.01). On multivariate logistic regression analysis, tactile sensory (p = 0.043) and hand function (p = 0.030) were both identified as significant predictors of FMA-UE improvement, tactile sensory (p = 0.047) was a significant predictor of AOU improvement, and hand function (p = 0.026) was a significant predictor of QOM improvement. The regression equations explained 71.2% of the variance in the improvement of FMA-UE, 69.7% of AOU, and 69.7% of QOM.

Conclusion: Both motor and tactile sensory impairments predict improvement in motor function, tactile sensory impairment predicts improvement in the amount of paralytic hand use, and motor impairment predicts improvement in the quality of paralytic hand use following neurorehabilitation treatment in patients with moderate to severe paralysis in chronic-phase stroke. These findings may help select the appropriate treatment for patients with more severe paralysis and to maximize the treatment effect.

Keywords: chronic stroke; motor function; motor recovery; neurorehabilitation; prediction model; somatosensory function.