Objective To investigate the relationship between intracranial pressure (ICP) monitoring of different parts in patients with craniocerebral injury and therapeutic effect. Methods A total of 82 patients with severe craniocerebral injury were divided into non-monitoring group and monitoring group according to the monitoring of ICP. The monitoring group was divided into intraventricular monitoring group and brain parenchymal monitoring group. The treatment effect of different groups was observed. Results The ICP was significantly lower in the intraventricular monitoring group (10.35±3.32）mmHg than that in the parenchymal group（ 13.46±4.71）mmHg（ t=4.017，P=0.028）. The time and dose of mannitol in the intraventricular monitoring group were significantly lower than those in the parenchymal monitoring group［ （6.05±2.03）d vs （8.46±2.41）d；（ 748.92±126.65）g vs（ 961.72±128.82）g， t=3.011，5.254；P＜ 0.05］. The GCS score of the intraventricular monitoring group on the 7th and the 14th day after surgery was（ 9.11±2.73），（ 12.06±2.37）， which was significantly higher than that of the parenchymal group［ （8.82±2.67），（ 10.09±2.23），t=3.403， 4.562；P ＜ 0.05］. After 6 months follow-up， the total effective rate of the monitoring group was significantly lower than that of the non-monitoring group（ 75.86%，68.97% vs 41.67%；χ2=5.432，6.401；P ＜ 0.05）. The incidence of hydrocephalus in the intraventricular monitoring group was significantly lower than that in the parenchymal group and non-monitoring group（ 10.34% vs 3.45%， 4.17%；χ2=3.709，4.421；P ＜ 0.05）. Conclusions Intracerebral ICP monitoring can significantly reduce the use of mannitol in patients with severe craniocerebral injury， reduce the intracranial hypertension， and the prognosis is good with less complications， which is worthy of clinical application. In addition， brain parenchymal monitoring is second to intraventricular monitoring.