Objective:

To investigate the prognostic accuracy of TMS-MEP for predicting upper limb motor function and capacity following stroke, while assessing the technical and methodological quality of relevant studies, including specific aspects such as sample size and timing of assessments.

Key Findings:

• Sixteen prospective TMS-MEP studies included, all showing low risk of bias.
• Muscle strength: sensitivity 0.76 (95% CI: 0.64–0.85), specificity 0.96 (95% CI: 0.55–1.00), PPV 0.96 (95% CI: 0.85–0.99), NPV 0.62 (95% CI: 0.43–0.78).
• Muscle synergies: sensitivity 0.70 (95% CI: 0.57–0.81), specificity 0.98 (95% CI: 0.65–1.00), PPV 0.98 (95% CI: 0.54–1.00), NPV 0.69 (95% CI: 0.42–0.87).
• UL-capacity: sensitivity 0.84 (95% CI: 0.76–0.89), specificity 0.91 (95% CI: 0.78–0.96), PPV 0.91 (95% CI: 0.79–0.97), NPV 0.81 (95% CI: 0.65–0.91).
• No significant differences in accuracy based on outcome measures or timing of TMS-MEP assessments.

Interpretation:

TMS-MEP is a reliable predictor of favorable upper limb recovery in terms of muscle strength, muscle synergy, and UL-capacity, but has limitations in accurately predicting non-recovery early post-stroke.

Limitations:

• Methodological flaws in many studies, including small sample sizes and arbitrary timing of assessments, which may impact the reliability of findings.
• Limited ability to predict individuals who do not regain UL motor function and capacity early post-stroke.

Conclusion:

TMS-MEP demonstrates strong predictive value for upper limb recovery post-stroke, though its early predictive accuracy for non-recovery is limited.