Neurosurgical outcomes in a modern series of 400 craniotomies for treatment of parenchymal tumors

OBJECTIVE: The goals were to critically review all complications resulting within 30 days after craniotomies performed for excision of intra-axial brain tumors relative to factors likely to affect complication rates and to assess the value of these data in predicting the risk of surgical morbidity, particularly for surgery in eloquent brain regions. METHODS: Neurosurgical outcomes were studied for 327 patients who underwent 400 craniotomies for removal of intra-axial parenchymal brain neoplasms in a 21-month period. Tumors removed included gliomas (206 tumors) and metastases (194 tumors) located both supratentorially (358 tumors) and infratentorially (42 tumors). RESULTS: The major complication incidence was 13%, and the operative mortality rate was 1.7%. The overall morbidity rate was 32%, but more types of complications were considered than in previous studies. The major neurological morbidity rate was 8.5%. Based on pre-versus postoperative (at 4 wk) Karnofsky Performance Scale scores, 9% of patients deteriorated neurologically, 32% improved, and 58% showed no change. The median postoperative hospital stay was 5 days. Tumors were defined as Grade I, II, or III based on their location relative to brain function, and this tumor functional grade was the most important variable affecting the incidence of any neurological deficit. Patients with tumors in eloquent (Grade III) or near-eloquent (Grade II) brain areas incurred more neurological deficits than did patients with tumors in noneloquent areas (Grade I). Neither repeat surgery for recurrent disease nor extent of surgical resection affected outcome significantly. Although most tumors in this study, including those in eloquent regions, were removed by gross total resection, this did not lead to more major neurological deficits. Regional complications (at the surgical sites) and systemic complications (medical) were more prevalent among older patients (age >60 yr) with lower preoperative Karnofsky Performance Scale scores (550) and posterior fossa masses. We showed how our data can be used to predict the total risk of surgical morbidity for a given patient, to facilitate patient counseling and surgical decision-making. CONCLUSION: The finding that gross total resections could be performed in eloquent brain regions with an acceptable level of neurological impairment suggested that the mere presence of a tumor in eloquent brain does not automatically contraindicate surgery. Our results have practical risk-predictive value, and they should aid in the construction of subsequent outcome studies, because we have identified the key areas to monitor.