Neuroendoscopic Surgery versus Craniotomy for Supratentorial Hypertensive Intracerebral Hemorrhage: A Systematic Review and Meta-Analysis

BACKGROUND: No consensus has been achieved on the superiority between neuroendoscopy (NE) and craniotomy (CT) for the treatment of supratentorial hypertensive intracerebral hemorrhage (HICH). The purpose of this study is to analyze the efficacy and safety of NE versus CT for supratentorial HICH. METHODS: A systematic search of English databases (PubMed, Embase, the Cochrane Library, and Web of Science) was performed to identify related studies reported from September 1994 to June 2019. The Newcastle-Ottawa Scale and the Cochrane Reviewer's Handbook 5.0.0 were separately used to evaluate the quality of the included observational studies and randomized controlled trials. RevMan 5.3 software was adopted to conduct the meta-ana lysis. The outcome measures included the primary and secondary outcomes. Subgroup analysis was performed to explore the impact of year of publication, initial Glasgow Coma Scale (GCS) score, age, time to surgery, hematoma volume, and surgical methods on the outcome measures. RESULTS: Fifteen studies (3 randomized controlled trials and 12 observational studies), comprising 1859 patients with supratentorial HICH, were included in this meta-analysis. The pooled results showed that NE could increase the good functional outcome (GFO) (P < 0.0003) and hematoma evacuation rate (P = 0.0007) and reduce the mortality (P < 0.00001), blood loss (P = 0.004), operation time (P < 0.00001), hospital stays (P = 0.006), and intensive care unit stays (P < 0.0001) compared with CT. In addition, NE could also have a positive effect on preventing postoperative infection (P< 0.00001) and total complications (P < 0.00001). However, in postoperative rebleeding incidence (P = 0.12), no obvious difference was found between the 2 groups. Publication bias was low regarding GFO, mortality, and hematoma evacuation rate. Subgroup analysis suggested that year of publication, initial GCS score, age, hematoma volume, and surgical methods did not affect the hematoma evacuation rate significantly. The difference in mortality was not statistically significant in the subgroup of hematoma volume <50 mL (P = 0.44) and initial GCS score >8 ( P = 0.09). In addition, the data suggested that time to surgery and surgical methods might be the important factors affecting GFO and mortality. CONCLUSIONS: NE might be a safer and more effective surgical method than CT in the treatment of patients with supratentorial HICH. However, because of the existence of some limitations, the safety and validity of NE were weakened. More high-quality trials should be included to verify our conclusion.