To investigate the therapeutic effect of recombinant human erythropoietin combined with predictive bilateral decompressive craniectomy in patients with severe craniocerebral injury. Methods: In this study, 98 patients with severe craniocerebral injury were randomly divided into the observation group and control group with 49 patients per group. On top of basic treatment, the control group was treated with predictive bilateral decompressive craniectomy and the observation group with recombinant human erythropoietin combined with predictive bilateral decompressive craniectomy. The expression level of related indexes of serum nerve injury [S-100β protein and neuron specific enolase (NSE)] and degree of neurological impairment (NIHSS score) and coma (GCS score) before and after treatment were compared between the two groups. After 6 months, the prognosis (GOS) of the two groups was compared. Results: NIHSS scores of the two groups after treatment were lower than those before treatment, and scores of the observation group were lower than those of the control group. After treatment, GCS scores of both groups were higher than those before treatment with the observation group showing higher scores than the control group (P＜0.05). After treatment, the serum levels of S-100 β protein and NSE of the two groups were lower than those before treatment, and levels in the observation group were lower than those in the control group (all P＜0.05). Six months after treatment, the number of patients in the observation group with a GOS of 5 or 4 was greater than that in the control group (P＜0.05). Conclusion: Recombinant human erythropoietin combined with predictive bilateral decompressive craniectomy in the treatment of patients with severe craniocerebral injury can significantly reduce the degree of neurological impairment and coma in patients and improve their prognosis.