Intraoperative perception and estimates on extent of resection during awake glioma surgery: overcoming the learning curve

Restricted access


There is ample evidence that extent of resection (EOR) is associated with improved outcomes for glioma surgery. However, it is often difficult to accurately estimate EOR intraoperatively, and surgeon accuracy has yet to be reviewed. In this study, the authors quantitatively assessed the accuracy of intraoperative perception of EOR during awake craniotomy for tumor resection.


A single-surgeon experience of performing awake craniotomies for tumor resection over a 17-year period was examined. Retrospective review of operative reports for quantitative estimation of EOR was recorded. Definitive EOR was based on postoperative MRI. Analysis of accuracy of EOR estimation was examined both as a general outcome (gross-total resection [GTR] or subtotal resection [STR]), and quantitatively (5% within EOR on postoperative MRI). Patient demographics, tumor characteristics, and surgeon experience were examined. The effects of accuracy on motor and language outcomes were assessed.


A total of 451 patients were included in the study. Overall accuracy of intraoperative perception of whether GTR or STR was achieved was 79.6%, and overall accuracy of quantitative perception of resection (within 5% of postoperative MRI) was 81.4%. There was a significant difference (p = 0.049) in accuracy for gross perception over the 17-year period, with improvement over the later years: 1997–2000 (72.6%), 2001–2004 (78.5%), 2005–2008 (80.7%), and 2009–2013 (84.4%). Similarly, there was a significant improvement (p = 0.015) in accuracy of quantitative perception of EOR over the 17-year period: 1997–2000 (72.2%), 2001–2004 (69.8%), 2005–2008 (84.8%), and 2009–2013 (93.4%). This improvement in accuracy is demonstrated by the significantly higher odds of correctly estimating quantitative EOR in the later years of the series on multivariate logistic regression. Insular tumors were associated with the highest accuracy of gross perception (89.3%; p = 0.034), but lowest accuracy of quantitative perception (61.1% correct; p < 0.001) compared with tumors in other locations. Even after adjusting for surgeon experience, this particular trend for insular tumors remained true. The absence of 1p19q co-deletion was associated with higher quantitative perception accuracy (96.9% vs 81.5%; p = 0.051). Tumor grade, recurrence, diagnosis, and isocitrate dehydrogenase-1 (IDH-1) status were not associated with accurate perception of EOR. Overall, new neurological deficits occurred in 8.4% of cases, and 42.1% of those new neurological deficits persisted after the 3-month follow-up. Correct quantitative perception was associated with lower postoperative motor deficits (2.4%) compared with incorrect perceptions (8.0%; p = 0.029). There were no detectable differences in language outcomes based on perception of EOR.


The findings from this study suggest that there is a learning curve associated with the ability to accurately assess intraoperative EOR during glioma surgery, and it may take more than a decade to be truly proficient. Understanding the factors associated with this ability to accurately assess EOR will provide safer surgeries while maximizing tumor resection.

ABBREVIATIONS ALA = aminolevulinic acid; CI = confidence interval; EOR = extent of resection; IDH = isocitrate dehydrogenase; iMRI = intraoperative MRI; OR = odds ratio; OS = overall survival; PFS = progression-free survival.

Article Information

Correspondence Mitchel S. Berger, Department of Neurological Surgery, University of California, San Francisco, 505 Parnassus Ave., M779, San Francisco, CA 94143. email:

INCLUDE WHEN CITING Published online July 21, 2017; DOI: 10.3171/2017.1.JNS161811.

Disclosures The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

© AANS, except where prohibited by US copyright law.



  • View in gallery

    Line graph of gross perception of EOR for awake glioma surgery. Gross perception was defined as whether GTR or STR was achieved compared with postoperative MRI measurements. Over the course of 17 years, there was a significant trend for improvement in accuracy of gross perception from 70.0% to 85.7% (p = 0.049). Subgroups based on 4 time periods also showed improvement in accuracy: 1997–2000 (72.6%), 2001–2004 (78.5%), 2005–2008 (80.7%), and 2009–2013 (84.4%; p = 0.175).

  • View in gallery

    Line graph of quantitative perception of EOR for awake glioma surgery. Quantitative perception was defined as whether the percentage of resection achieved was within 5% of postoperative MRI measurements. Accuracy of quantitative perception significantly improved from 77.8% to 100% over the 17-year time period (p = 0.015). Subgroup analysis showed significant improvement as well: 1997–2000 (72.2%), 2001–2004 (69.8%), 2005–2008 (84.8%), and 2009–2013 (93.4%; p < 0.001).





All Time Past Year Past 30 Days
Abstract Views 202 202 12
Full Text Views 245 245 6
PDF Downloads 189 189 6
EPUB Downloads 0 0 0


Google Scholar