Introduction: When it is difficult to determine whether a muscle is grade 0 or 1, manual muscle test (MMT) accuracy can be further improved by using needle electromyography (EMG) as a supplementary and confirmatory examination tool. Objective: To evaluate concordance between needle EMG and MMT findings for key muscles with motor grades 0 and 1 on the International Standards of Neurological Classification of Spinal Cord Injury (ISNCSCI) examination, and to potentially improve the prognosis for grade 0 muscles with proven muscle activity based on needle EMG findings. Design: A retrospective analysis. Setting: Inpatient tertiary rehabilitation facility. Interventions: Not applicable. Patients: One hundred seven patients with spinal cord injury (SCI) admitted for rehabilitation (n = 1218 key muscles, grades 0 or 1). Main Outcome Measures: Inter-rater reliability between MMTs and needle EMG was analyzed using Cohen's kappa coefficient (.). A Mantel Haenszel linear-by-linear association chi-square test was used to determine whether the presence of motor unit action potentials (MUAPs) in muscles graded 0 on the initial MMT at admission was associated with MMT grades at discharge and readmission. Results: Moderate-to-substantial agreement between needle EMG and MMT findings was observed (. = 0.671, p <.01). Concerning key upper and lower extremity muscles, moderate and substantial agreement was identified, respectively. The lowest agreement was noted for C6 muscles. During follow up, 68.8% of muscles with proven MUAPs showed improved motor grades. Conclusions: At initial assessment, distinguishing between motor grades 0 and 1 is imperative because motor grade 1 muscles are more likely to have a better prognosis for improvement. Moderate-to-substantial agreement was observed between MMT and needle EMG findings. The MMT is a reliable method of muscle grading, yet needle EMG may be of value in certain clinical situations to evaluate for the presence of MUAPs when evaluating motor function.