Classify patients with Parkinson’s disease (PD) tremor based on the heterogeneity of clinical features and electrophysiological characteristics of tremor. Methods: Clinical data (including motor symptoms, non-motor symptoms, progression rate, disease stage) and electrophysiological characteristics of 220 patients with PD tremor were collected. Based on these clinical features and electrophysiological parameters, these patients were classified by K-means cluster analysis. We used the silhouette coefficients corresponding to K ranging from 2 to 10 to determine the optimal number of clusters. Results: Based on the above methods, PD patients with tremor were clustered into two subtypes. (1) Subtype 1 (85 cases): rapid progression of tremor, good response to levodopa, and electromyographic analysis showing predominantly 4～6 Hz resting tremor; (2) Subtype 2 (135 cases): slow progression of tremor, poor response to levodopa, and electromyographic analysis revealing irregular jerky movements. Significant differences were observed between the two subtypes in disease duration, motor symptom scores, rigidity scores, tremor scores, axial symptom scores, levodopa-equivalent doses, rigidity/tremor score ratios, progression rates of tremor, responsiveness of motor symptoms and tremor to levodopa, frequency distribution of upper limb resting, postural, intentional, and postural action tremors, as well as patterns of muscle contraction (all P<0.05). Conclusion: There is heterogeneity in both clinical features and electrophysiological characteristics of PD patients with tremor. Based on this heterogeneity, PD patients with tremor can be classified into two subtypes.