The cause of death in patients with glioblastoma is multifactorial: clinical factors and autopsy findings in 117 cases of supratentorial glioblastoma in adults
- PMID: 1654403
- DOI: 10.1007/BF00146880
The cause of death in patients with glioblastoma is multifactorial: clinical factors and autopsy findings in 117 cases of supratentorial glioblastoma in adults
Abstract
To delineate the causes of death (COD) in adults with supratentorial glioblastoma multiforme (GM) we reviewed 117 consecutive cases examined at autopsy over a nineteen year period at the University of Washington. Twenty cases (17%) had expired unexpectedly without ante mortem diagnosis, 5 patients (4%) had been diagnosed as having lower grade astrocytomas prior to death. Other than the 20 patients without ante mortem diagnosis, all patients had a surgical procedure for treatment and/or diagnosis (biopsy 10%, craniotomy 90%). Postsurgical therapy varied, but there was no significant difference in median length of survival among the different treatment groups. Factors considered as potential COD were: herniation (axial, transtentorial, subfalcine, tonsillar), surgical complications (death within thirty days of surgery secondary to cerebral hemorrhage and/or edema), severe systemic illness, brainstem invasion by tumor, and neutron-induced cerebral injury (cerebral and brainstem gliosis were evident in these cases). A potential COD could be identified in 93% of patients. Patients with no ante mortem diagnosis were likely to have herniated (p = 0.01), whereas patients who underwent neutron irradiation were unlikely to have herniated (p = 0.001). No other variables were statistically significant predictors of herniation, including multifocal tumors (20 patients), and brainstem invasion by tumor (18 patients). No patients died as a result of treatment except those who underwent neutron radiotherapy and those who died postoperatively. Although significant mass effect, as evidenced by herniation, was apparent in 61% of patients, most of these patients had an additional identifiable COD. We conclude that the COD in patients with GM varies and is multifactorial.
References
Publication types
MeSH terms
Grants and funding
LinkOut - more resources
Medical