Introduction Central poststroke pain (CPSP) develops commonly after stroke, which impairs the quality of life, mood, and social functioning. Current pharmacological approaches for the treatment of CPSP are not satisfactory. Repetitive transcranial magnetic stimulation (rTMS) is a noninvasive technique which has been recommended for the treatment of chronic CPSP. However, few studies have evaluated the analgesic effects of rTMS in patients with acute neuropathic pain after stroke. Methods We evaluated the analgesic effects of rTMS applied over the upper extremity area of the motor cortex (M1) in patients with acute CPSP. Forty patients were randomized to receive either rTMS (10 Hz, 2000 stimuli) (n = 20) or a sham intervention (n = 20) for 3 weeks. The Numeric Rating Scale (NRS), Short-form McGill Pain Questionnaire-2 (SF-MPQ-2, Chinese version), Hamilton Anxiety Scale (HAM-A), Hamilton Depression Scale (HAM-D), brain-derived neurotrophic factor (BDNF) levels, and motor-evoked potentials (MEP) were analyzed at baseline, 3 days, 1 week, 2 weeks, and 3 weeks. Results Significant treatment-time interactions were found for pain intensity. Compared with the sham group, the NRS and SF-MPQ-2 scores were significantly lower on the seventh day of treatment in the rTMS group (P < 0.001, Cohen's d = 1.302) (P = 0.003, Cohen's d = 0.771), and this effect lasted until the third week (P = 0.001, Cohen's d = 0.860) (P = 0.027, Cohen's d = 0.550). The HAM-A and HAM-D scores did not change in the rTMS group when compared with the sham group (P = 0.341, Cohen's d = 0.224) (P = 0.356, Cohen's d = 0.217). The serum BDNF levels were significantly higher in the treated group (P = 0.048, Cohen's d = -0.487), and the resting motor threshold (RMT) decreased by 163.65%. Conclusion Our findings indicate that rTMS applied over the upper extremity area of the motor cortex can effectively alleviate acute CPSP, possibly by influencing cortical excitability and serum BDNF secretion.