Abstract
Background/Objectives: The impact of stroke on upper extremity function in the older adult population underscores the need for accurate recovery prediction. Motor evoked potential (MEP) has been explored as a predictor of upper extremity function recovery in patients with stroke. However, research specifically targeting the geriatric population remains limited. Therefore, this study focused specifically on patients aged 65 years and older to investigate correlations between MEP parameters and upper extremity function. This study investigates correlations between MEP parameters (amplitude and latency) and upper extremity function-related measures, including Medical Research Council (MRC) scale, the Korean version of the Modified Barthel Index (K-MBI), and the Hand Function Test (HFT), including grip strength, pinch strength, the Box and Block Test, and the 9-Hole Peg Test, in older adults with stroke. Methods: A multiple linear regression model predicts upper extremity outcomes using initial MEP parameters, time, and function. The dataset includes 90 patients with stroke categorized by timing of the first MEP assessment: ≤3 months (n = 42) or >3 months (n = 48). Results: MEP amplitude and latency were significantly correlated with upper extremity function in both groups. No significant correlations were found between MEP amplitude or latency and outcome measures. Regression analysis showed that initial MEP amplitude had a limited association with outcomes, whereas latency was significantly associated with grip strength (β: -10.205, 95% CI: -19.374~-1.036) and the Box and Block Test (β: -10.204, 95% CI: -20.254~-0.154). Initial upper extremity parameters were significantly associated with K-MBI and HFT follow-up results (p < 0.05). Conclusions: Larger MEP amplitude and faster initial MEP latency were associated with improved upper extremity function in patients with stroke. In older patients, MEP latency, rather than amplitude, demonstrated greater predictive value for upper extremity function recovery, possibly due to age-related muscle atrophy, a factor not fully addressed in existing prognostic frameworks such as PREP2. These findings support the integration of MEP latency assessment into geriatric stroke prognostication, complementing existing frameworks such as PREP2, and may guide personalized rehabilitation planning to optimize functional recovery and independence.