When gross total resection of a glioblastoma is possible, how much resection should be achieved?

Kaisorn L Chaichana; Eibar Ernesto Cabrera-Aldana; Ignacio Jusue-Torres; Olindi Wijesekera; Alessandro Olivi; Maryam Rahman; Alfredo Quinones-Hinojosa

doi:10.1016/j.wneu.2014.01.019

When gross total resection of a glioblastoma is possible, how much resection should be achieved?

World Neurosurg. 2014 Jul-Aug;82(1-2):e257-65. doi: 10.1016/j.wneu.2014.01.019. Epub 2014 Feb 6.

Authors

Kaisorn L Chaichana¹, Eibar Ernesto Cabrera-Aldana², Ignacio Jusue-Torres², Olindi Wijesekera², Alessandro Olivi², Maryam Rahman², Alfredo Quinones-Hinojosa²

Affiliations

¹ Department of Neurosurgery, Johns Hopkins University School of Medicine, Neuro-Oncology Outcomes Laboratory, Baltimore, Maryland, USA. Electronic address: Kaisorn@jhmi.edu.
² Department of Neurosurgery, Johns Hopkins University School of Medicine, Neuro-Oncology Outcomes Laboratory, Baltimore, Maryland, USA.

PMID: 24508595
DOI: 10.1016/j.wneu.2014.01.019

Abstract

Objective: The efficacy of extensive resection on prolonging survival for patients with glioblastoma (GBM) is controversial because prior studies have included tumors with dissimilar resection capabilities. The true isolated effect of increasing resection on survival for GBM therefore remains unclear.

Methods: Adult patients who underwent surgery of an intracranial newly diagnosed GBM at an academic tertiary-care institution from 2007 to 2011 were reviewed. Preoperative images were reviewed by 3 neurosurgeons independently. Tumors considered amenable to gross total resection based on preoperative imaging by all neurosurgeons were included. Multivariate proportional hazards regression analysis was used to identify if an association existed between residual volume (RV) and extent of resection (EOR) with survival.

Results: Of the 292 patients with newly diagnosed GBM, 84 (29%) were amenable to gross total resection. The median (interquartile range) pre and postoperative tumor volumes were 27 (13.8-54.4) and 0.9 (0-2.7) cm(3), respectively. The mean percent resection was 91.7% ± 1.3%. In multivariate analysis, after controlling for age, functional status, and adjuvant therapies, RV (hazards ratio [HR] [95% confidence interval (CI)] = 1.114 [1.033-1.193], P = 0.006) and EOR (HR [95% CI] = 0.959 [0.934-0.985], P = 0.003) were each independently associated with survival. The RV and EOR with the greatest reduction in the risk of death was <2 cm(3) and >95%, respectively. Likewise, RV (HR [95% CI] = 1.085 [1.010-1.178], P = 0.01) and EOR (HR [95% CI] = 0.962 [0.930-0.998], P = 0.04) each remained independently associated with recurrence.

Conclusion: This is the first study to evaluate RV and EOR in a more uniform population of patients with tumors of similar surgical capabilities. This study shows that achieving a decreased RV and/or an increased EOR is independently associated with survival and recurrence in those patients with tumors with similar resection capacities.

Keywords: Extent of resection; GBM; Glioblastoma; Residual; Surgery; Survival; Volumetric.

Publication types

Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't

MeSH terms

Aged
Aged, 80 and over
Brain Neoplasms / pathology
Brain Neoplasms / surgery*
Chemoradiotherapy
Contrast Media
Evoked Potentials, Somatosensory / physiology
Female
Glioblastoma / pathology
Glioblastoma / surgery*
Humans
Image Processing, Computer-Assisted
Magnetic Resonance Imaging
Male
Neoplasm Recurrence, Local / epidemiology
Neoplasm Recurrence, Local / therapy
Neoplasm, Residual / pathology
Neoplasm, Residual / therapy
Neurosurgical Procedures / methods*
Observer Variation
Patient Care Team
Patient Selection
Reproducibility of Results
Retrospective Studies
Survival Analysis
Treatment Outcome

Substances

Contrast Media