300 Section - Stair performance to identify fallers with multiple sclerosis. A cross-sectional diagnostic study

The primary aim was to determine whether stair performance better identifies fallers with pwMS than BBS and TUG. The secondary aim was to identify the determinants of stair performance in pwMS.

Design:

A multicenter cross-sectional, diagnostic study in pwMS. Falls in the previous year were collected. Assessments included stair performance, BBS, TUG, knee extensor and flexor strength (dynamometry), lower limb strength (manual testing), lower limb spasticity, locomotor performance (10-meter walk test [10MWT] at comfortable and maximal speed, 6-minute walk test [6MWT]), fatigue, quality of life, anxiety and pain. Receiver Operating Characteristic (ROC) curves assessed fall discrimination with the area under the curve (AUC). Backward stepwise regressions identified determinants of stair ascent and descent.

Results:

We included 134 pwMS (median EDSS score = 3.5). 47.8% had fallen in the past year. Fallers were discriminated with stair ascent (AUC=0.69, 95%CI 0.60-0.78, Sensitivity 72%, Specificity 64%), stair descent (AUC=0.69, 95%CI 0.60-0.78, Sensitivity 67%, Specificity 71%) and BBS (AUC=0.66, 95%CI 0.57-0.75, Sensitivity 77%, Specificity 45%). Stair threshold for falls was between 1.42 and 1.50 steps/sec. TUG was not associated with falls. Determinants of stair ascent were maximal speed 10MWT, followed by 6MWT, TUG and quality of life, explaining 65.1% of the variance. Stair descent was determined by 6MWT then the TUG, ankle dorsiflexor strength and spasticity, explaining 64.2% of the variance.

Conclusion:

Stair performance is a quick and essential test to identify fallers among independently ambulant pwMS. It is more discriminant than the usual BBS and TUG. Locomotor performances determined stairs performance more than strength, balance or fatigue.