Validation of a French version of the upper limb Erasmus Modified Nottingham Sensory Assessment with stereognosis (EmNSA-S) component in patients with stroke

Abstract

Sensory impairment of the upper limb (UL) is a common deficit post-stroke, associated with loss of functional ability. Currently, there is no standardized tool evaluating sensation of the UL in patients with stroke in the French language. The objective was to develop a French version of the Erasmus modified Nottingham Sensory Assessment with stereognosis component (EmNSA-S) for evaluation of the UL among adults with stroke. It was a monocentric prospective observational cross-sectional design study. The study consisted of two phases including (i) instrument translation, and (ii) validation phase with establishment of the concurrent criterion-related validity, internal consistency, intra and inter-rater reliability and minimal detectable change with 95% confidence interval (MDC95). French version of the EmNSA-S for UL was developed through multidisciplinary forwards-backwards translation. Outcome measures were the EmNSA-S, Fugl-Meyer Assessment sensory subscale (FMA-S) and the Fugl-Meyer Assessment motor subscale of UE (FMA-M UE). Fifty patients with hemiparesis due to a single cerebrovascular accident were recruited to establish psychometric properties (mean ± SD age, 55 ± 15.6 years; median [range] time since stroke, 3.1 [1.1 to 319.1] months; 70% of male; 52% right paresis; median [range] FMA-M UE scale, 33.5/66 [4 to 66]). The median [range] score of EmNSA-S was 51.5/64 [1 to 63], a score of 64 indicating no sensory impairment. Spearman rank correlation coefficient was very strong between EmNSA-S and FMA-S total scores ( r = 0.826, P < 0.05). EmNSA-S internal consistency is adequate across subscales with Cronbach α coefficient ranging between 0.821 and 0.966. For the total score, intra-rater reliability was excellent (ICC = 0.91) with a MDC95 equal to 11.5, and the inter-rater reliability was good (ICC = 0.78) with a MDC95 equal to 13.7. Analysis of the hierarchical aspect of the tactile sensation subscale has found a relatively high error rate (up to 24%) suggesting the need for more systematic assessment. The standardized French version of the EmNSA-S provides a valid and reliable scale. The study supports the use of the French version of this scale for comprehensive and accurate assessment of sensory modalities in adults post-stroke.