Optimizing Outcomes With Maximal Surgical Resection of Malignant Gliomas

Stephen J. Hentschel, MD, Raymond Sawaya, MD


Cancer Control. 2003;10(2) 

In This Article

Reasons for the Controversy

Extensive reviews have failed to find a statistical correlation between extent of surgical resection and survival in malignant gliomas.[13,14,15] Why is the concept of aggressive resection for glioblastoma still controversial, given the above arguments and the more recent studies? Several factors that surround the controversy have been identified:

Despite recent surgical and technological advances, the prognosis for GBM has changed little over the last 20 years. The best average survival for all patients with GBM is still 16 to 18 months, which is similar to the survival average reported 20 years ago.[2,16,17] However, average survival as long as 3 years may occur in some patients who are younger, have a good Karnofsky performance status, and have lesions in favorable locations.[2] Bucy and colleagues[18] reported a case of a patient who underwent surgery in 1958 for a GBM who was alive 26 years later.

Studies have shown that tumor cells may invade far beyond the main tumor mass into the brain due to the infiltrating nature of malignant gliomas and to the impossibility of removing all of the tumor cells.[19] Early studies have shown that even hemispherectomy is inadequate for the control of malignant gliomas. In the 1980s, Kelly et al[20] reported no survival advantage using volumetric laser resection but showed that patients with deep tumors could have the same survival as those with more superficial and more easily resectable tumors. While still controversial, this was the beginning of the concept that maximal resection could improve prognosis for malignant gliomas, an idea that has taken 15 more years to statistically demonstrate in a select group of patients.

It is a falsely held belief by some that radical resection of malignant gliomas is associated with an increased incidence of neurologic morbidity compared with lesser resections, even when considering lesions in eloquent areas.[1,7] Fadul et al[8] showed that patients undergoing radical resections were at no greater risk of being neurologically impaired at 1 week following surgery and had fewer acute neurologic complications than patients undergoing lesser resections. When examining the Karnofsky performance status in these two groups of patients,Ammirati and colleagues[7] found that the scores improved by a mean of 6.8 (P<.006) compared to preoperative scores in the completely resected group, while there was no improvement in the scores in the subtotally resected group (P<.002).

Other than the recent report from the M. D. Anderson Cancer Center, 2 no other studies have examined the relationship between the extent of surgical resection and survival using rigorous preoperative and postoperative imaging analysis in a prospective fashion. In most studies, the extent of resection was not confirmed with postoperative imaging[5,6] or patients had undergone relatively few true maximal resections,[7,21] and only one study was prospective.[21] The opinion of the surgeon at the time of surgery may be remarkably inaccurate, with only a 30% correlation with MRI.[21] Therefore, postoperative imaging is key in determining the extent of resection. In addition to not confirming the extent of resection on imaging, investigators for many of the studies considered a 90% resection to be gross total. This "dilutes" the category of complete resections and thus renders the study unable to identify the group with a better prognosis -- that being the ≥98% resection group. Because of these drawbacks, two review studies of the literature prior to 1990[13,14] and another after 1990[15] determined that "...little scientifically credible evidence is available to support the assertion that aggressive surgical resection prolongs survival."[15] Many authors have called for a prospective, randomized trial to clarify this issue,[9,13,14,15] but it is unlikely that such a study will be conducted because a significant number of patients are required to determine a difference statistically and because many surgeons would not allow their patients to be randomized to such a trial.