- Email: [email protected]
The Value of Intraoperative and Early Postoperative Magnetic Resonance Imaging in Low-Grade Glioma Surgery: A Retrospective Study
Original Article
The Value of Intraoperative and Early Postoperative Magnetic Resonance Imaging in Low-Grade Glioma Surgery: A Retrospective Study Andrej Pala1, Christine Brand1, Thomas Kapapa3, Michal Hlavac1, Ralph Ko¨nig1, Bernd Schmitz2, Christian Rainer Wirtz1, Jan Coburger1
BACKGROUND: The presence of residual tumor is crucial in decision-making for low-grade gliomas (LGGs), because patients older than 40 years of age with residual tumor are considered for adjuvant treatment. There are hints that early postoperative fluid-attenuated inversion recovery (FLAIR) and T2 (within 48 hours) may overestimate residual tumor volume in LGG. Intraoperative magnetic resonance imaging (MRI) without subsequent resection or ultra-early postoperative MRI may assess the amount of residual tumor more adequately. To evaluate the utility of postoperative imaging in LGG, we volumetrically analyzed intraoperative, early, and late (3e4 months after surgery) postoperative MRIs of LGGs.
-
PATIENTS AND METHODS: A total of 33 patients with LGG were assessed retrospectively. Residual tumor was defined as signal-enhanced tissue in T2 and FLAIR. Volumetric assessment was performed with intraoperative, early, and late postoperative T2/FLAIR via Brainlab-iPlan 3.0. Wilcoxon and c2 tests were used for statistical analysis.
postoperative FLAIR/T2 mean volume [ 5.560 cm3 and 2.370 cm3, P [ 0.520, P [ 0.398), whereas a significant difference was detected between early and late postoperative images (FLAIR, P < 0.0001; T2, P < 0.00001). CONCLUSION: Intraoperative MRI without further resection or ultra-early postoperative MRI seems to reflect the actual volume of residual tumor in LGG more precisely compared with early postoperative MRI and therefore seems to be more useful regarding decisions for adjuvant therapy.
-
-
RESULTS: A significant difference of FLAIR/T2 abnormalities was found in intraoperative and early postoperative MRIs (FLAIR mean volume [ 5.433 cm3, T2 mean volume [ 3.374 cm3 vs. FLAIR mean volume [ 14.090 cm3, P [ 0.002, T2 mean volume [ 7.597 cm3, P [ 0.006). There was no significant difference between intraoperative and late postoperative FLAIR/T2 abnormalities (late
-
Key words FLAIR - Intraoperative MRI - Low-grade glioma - T2 -
Abbreviations and Acronyms DWI: Diffusion-weighted imaging EoR: Extent of resection FLAIR: Fluid-attenuated inversion recovery iMRI: Intraoperative magnetic resonance image LGG: Low-grade glioma OS: Overall survival
WORLD NEUROSURGERY 93: 191-197, SEPTEMBER 2016
INTRODUCTION
L
ow-grade gliomas (LGGs) are slow-growing, infiltrative lesions that despite all therapeutic efforts inevitably recur and transform to more malignant grades.1,2 Additional intraoperative imaging techniques, such as high-field intraoperative magnetic resonance imaging (iMRI), detect tumor remnants and increase the extent of resection (EoR) in LGG surgery.3,4 There is increasing evidence that EoR is an important positive prognostic factor for longer progression-free survival (PFS) and overall survival (OS) of patients with LGG.1,2,4,5 Early postoperative (within 48 hours) fluid-attenuated inversion recovery (FLAIR) and T2 sequences delineate tumor remnants and help to define EoR. A precise definition of residual tumor has an impact on further adjuvant therapy planning in patients with LGG.
PFS: Progression-free survival WHO: World Health Organization From the Departments of 1Neurosurgery and 2Neuroradiology, University of Ulm, Ulm; and 3 Department of Neurosurgery, University of Ulm, Günzburg, Germany To whom correspondence should be addressed: MUDr. Andrej Pala, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2016) 93:191-197. http://dx.doi.org/10.1016/j.wneu.2016.04.120 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2016 Elsevier Inc. All rights reserved.
www.WORLDNEUROSURGERY.org
191
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
There is increasing evidence that the early postoperative MRI overestimates the tumor borders and does not correlate with the extent of the disease.6,7 The use of iMRI after final tumor resection might amend the early postoperative MRI and define EoR more precisely. To date, no study has evaluated signal abnormalities between intraoperative and early and late postoperative T2/FLAIR images (34 months). To evaluate the utility of postoperative imaging in LGG, we compared the volume of signal enhancement in T2 and FLAIR sequences in intraoperative, early, and late postoperative MRI scans using Brainlab iPlan 3.0 (Brainlab AG, Feldkirchen, Germany) in patients with LGG. PATIENTS AND METHODS Patients and Follow-Up Assessment Only patients treated with World Health Organization (WHO) grade II LGG tumors between 2009 and 2015 were assessed retrospectively. Follow-up assessment including clinical status and MRI was performed at 34 months after surgery. Preoperative MRIs were not older than 14 days and included T1-weighted spin echo, T2-weigted turbo spin echo, FLAIR, and diffusion-weighted imaging (DWI) sequences. Operating Room Setup and MRI An intraoperative 1.5-T MRI Espree scanner is available (Espree, Siemens AG, Erlangen, Germany) at our department as a oneroom solution since October 2008. During surgery, an iMRI scan was performed according to surgeon’s decision. iMRI sequences included T1-weighted spin echo, T2-weigted turbo spin echo, FLAIR, and DWI sequences. In 8 cases, postoperative MRI was performed directly after surgery. Early postoperative MRI was performed within 48 hours after surgery, and late postoperative MRI was obtained 34 months after surgery. Pre- and
Figure 2. Axial T2 image of diffuse low-grade glioma.
postoperative MRIs were performed either with the intraoperative scanner or with 1.5-T MRI Symphony system (Siemens AG, Erlangen, Germany). MRI Volumetric Assessment Tumor volume was measured after image fusion via iPlan 3.0 (Brainlab AG). Volumetric assessment was performed based on
Table 1. Patient Characteristics Characteristic
n%
n
33
Age, years Median
38.7
Minemax
1161
Sex Female ratio
64%
Subtype Diffuse astrocytoma
46%
Oligodendroglioma
39%
Oligoastrocytoma
15%
Presenting symptoms
Figure 1. Axial T2 image of circumscribed low-grade glioma.
192
www.SCIENCEDIRECT.com
Seizure
64%
Headache
12%
Incidental
12%
Progress
9%
Aphasia
3%
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.04.120
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
50
40
Volume
30
20
10
0 intraoperative T2
early postoperative T2
T2 late postoperative
Figure 3. Volume of T2 abnormalities in intraoperative, early postoperative and late postoperative magnetic resonance imaging.
manual sequencing of T2 and FLAIR images in preoperative, intraoperative, early, and late postoperative T2 and FLAIR sequences. Circumscribed and diffuse growing lesions were distinguished according to FLAIR/T2 tumor boundaries (Figures 1 and 2). Circumscribed LGG was defined as a compact lesion without visible spread along white matter fibers.1 In addition, we created 2 groups of patients based on their morphology according to volume of signal abnormalities in late postoperative T2/FLAIR. The threshold for FLAIR was 1.5 cm3 and for T2 1 cm3. Furthermore, we analyzed the dependence of this tumor morphology on signal changes in late postoperative images. We didn’t apply the method used by Ius et al.1; the difference between circumscribed and diffuse lesions was based on the macroscopic infiltration of white matter tracts.
Data Analysis The data of 33 patients were evaluated retrospectively. At 34 months’ follow-up, tumor volume was evaluated. Statistical analysis was performed with SPSS 21.0 (Lead Technologies, INC, Charlotte, North Carolina, USA). Wilcoxon and c2 tests were used for the analysis. The study was conducted according to the international Declaration of Helsinki.
WORLD NEUROSURGERY 93: 191-197, SEPTEMBER 2016
RESULTS Patient Characteristics A total number of 33 patients treated for WHO grade II LGGs were assessed. The most common histologic subtype of LGG was diffuse astrocytoma (n ¼ 15, 46%), followed by oligodendroglioma (n ¼ 13, 39%) and oligoastrocytoma (n ¼ 5, 15%). A mean age of 38.7 (range 11e61) years was noted. The majority of patients were female (64%, n ¼ 21). The demographic data are summarized in Table 1. Seizure was the most common presenting symptom (n ¼ 21, 64%). Recurrent tumor was treated in 12% (n ¼ 4). Volumetric and Statistical Analysis The mean preoperative tumor volume was 33.766 cm3 in T2 and 39.193 cm3 in FLAIR. iMRI was available for all patients, whereas no subsequent resection was performed in 22 cases. The mean volume of intraoperative T2 and FLAIR without subsequent resection was 3.374 and 5.433 cm3, respectively. Eight patients obtained “ultra-early” postoperative MRI directly after wound suturing and were evaluated together with the iMRI cohort without further resection. An early postoperative T2 was obtained in 26 and FLAIR in 25 cases. The mean volume of early postoperative T2 and FLAIR was 7.597 cm3 and 14.090 cm3, retrospectively. A significant difference of T2 and FLAIR abnormalities was found in
www.WORLDNEUROSURGERY.org
193
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
100
80
Volume
60
40
20
0 intraoperative FLAIR
early postoperative FLAIR
Figure 4. Volume of fluid-attenuated inversion recovery abnormalities in intraoperative, early postoperative, and
intraoperative and early postoperative MRI (P ¼ 0.006, P ¼ 0.002) in Wilcoxon test (Figures 3 and 4). Late postoperative FLAIR was available for all patients. In 1 patient, no T2 sequences were available in the follow-up. No significant difference was noted between intraoperative and late postoperative FLAIR and T2 abnormalities (late postoperative FLAIR and T2 mean volume ¼ 5.560 cm3 and 2.370 cm3, P ¼ 0.520, P ¼ 0.398, Figures 3 and 4). In contrast, a significant difference was detected between early and late postoperative images (T2, P < 0.00001, FLAIR, P < 0.0001). According to the tumor’s FLAIR and T2 infiltration pattern of white matter pathways, 19 circumscribed (Figure 1) and 14 diffuse (Figure 2) gliomas were distinguished. Subgroup analysis of both morphologic entities showed similar results and significant differences between intraoperative and early postoperative FLAIR/T2 except for T2 of diffuse lesions (diffuse LGG, FLAIR P ¼ 0.043, T2 P ¼ 0.116, circumscribed LGG, FLAIR P ¼ 0.021, T2 P ¼ 0.015) as well as significant differences in signal abnormalities between early and late postoperative FLAIR/T2 (diffuse LGG, FLAIR P ¼ 0.025, T2 P ¼ 0.012, circumscribed LGG FLAIR P ¼ 0.0003, T2 P ¼ 0.0003). No significant difference was noted between intraoperative and late postoperative FLAIR and T2 abnormalities in the subgroup analysis (diffuse LGG, FLAIR P ¼ 0.799, T2 P ¼ 0.441, circumscribed LGG FLAIR P ¼ 0.477, T2 P ¼ 0.799). Fourteen
194
www.SCIENCEDIRECT.com
late postoperative FLAIR
late postoperative magnetic resonance imaging.
patients were identified in the late postoperative FLAIR with residual signal abnormalities less than 1.5 cm3, and 16 patients had signal changes with the volume less than 1 cm3 in T2. FLAIR signal abnormalities with volume less than 1.5 cm3 and T2 abnormalities with volume less than 1 cm3 were significantly more often associated with morphologically circumscribed lesions in late postoperative images (FLAIR, P ¼ 0.026, T2, P ¼ 0.016 with c2 test). Illustrative Case After an episode of generalized seizure and slight short-term memory disturbances on neurologic examination, a 46-year-old female patient presented with a T2 hyperintense right frontal mass lesion on MRI without gadolinium enhancement (Figure 1). Functional MRI confirmed the activation of primary motor region directly behind the mass lesion on the right side. Surgery with direct cortical stimulation, continuous monitoring of motor-evoked potentials, and iMRI was performed. After intraoperative impression of gross total resection, the surgeon indicated an iMRI scan. A potential small tumor remnant was detected directly in the precentral gyrus on iMRI (Figure 5). Because of the location in the primary motor region verified by direct cortical and subcortical stimulation, no further resection was performed. The histopathologic report confirmed
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.04.120
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
oligoastrocytoma WHO grade II. No transient or permanent focal neurologic deficit was found 3 months after the surgery. The early postoperative MRI showed much more pronounced T2 signal abnormalities (Figure 6) compared with iMRI (Figure 5). The exact distinction between tumor remnant (Figure 6, white arrow) and surgically induced artifacts (Figure 6, black arrow) was not possible in the early phase after the surgery. In the late postoperative MRI, T2 tumor depiction in the precentral gyrus (Figure 7, white arrow) was similar to the iMRI. No surgically induced changes are detectable in the late postoperative MRI. DISCUSSION LGGs are considered as WHO grade II infiltrative neoplasms that despite their slowly growing pattern recur and inevitably deteriorate to malignant phenotype.1,8 FLAIR and T2 sequences has been shown to play an important role in follow-up and in further management of LGG.7 EoR becomes an important positive prognostic factor for PFS and OS, and the proof of tumor remnants in postoperative MRI imaging influences further adjuvant therapy.1,2,4,5 FLAIR sequences delineate tumor boundaries more precisely than T2, especially in tumors located near cerebrospinal fluid space and enables better distinguishing between tumor and brain edema.9 Residual proteins and blood degradation products, such as methemoglobin, influence the relaxation time of cerebrospinal fluid in MRI. FLAIR consequently is not able to optimize inversion time and leads to artifacts in early postoperative images7 and might obscure possible tumor remnant.6 The routine application of a final iMRI scan or an ultra early MRI after resection might minimize the artifacts and define tumor borders more precisely than early postoperative MRI. We conducted a retrospective assessment of all LGG surgeries performed at our center from 2009 to 2015 and evaluated the
Figure 5. Intraoperative axial T2 image of low-grade glioma with suspected tumor remnant in the precentral gyrus (white arrow).
WORLD NEUROSURGERY 93: 191-197, SEPTEMBER 2016
Figure 6. Early postoperative axial T2 image with advanced surgical-induced T2 enhancement (black arrow) and suspected tumor remnant with pronounced T2 abnormalities (white arrow) in the precentral gyrus.
Figure 7. Late postoperative axial T2 image with suspected tumor remnant (white arrow) in the precentral gyrus.
www.WORLDNEUROSURGERY.org
195
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
volume of T2/FLAIR abnormalities in intraoperative, early, and late postoperative images. We found a significant difference between T2/FLAIR abnormalities in intraoperative and early postoperative MRI, whereas no significant difference was seen between intraoperative and late postoperative T2/FLAIR images. The routine MRI scan in early postoperative phase might be redundant, if no further resection was performed after iMRI scan. Belhawi et al.6 found in T2 22% and in FLAIR 49% smaller residual tumor volume when they compared early and late postoperative MRI, confirming that early postoperative MRI overestimates the tumor borders in glioma patients with no gadolinium enhancement and that the late postoperative MRI delineate tumor remnants more precisely.6 In our study, we evaluated a more precise cohort of only WHO grade II tumors compared with Belhawi et al.,6 who evaluated gliomas in general. We did not analyze the DWI and did not exclude the typical ischemic T2/FLAIR abnormalities.6 Despite this fact we found a significant difference between intraoperative and early postoperative T2/FLAIR and between early and late postoperative T2/FLAIR images. Such a rapid tumor regression seems to be unrealistic, even if some ischemic-dependent tumor shrinkage is possible. Nevertheless, the intraoperative T2/FLAIR abnormalities have shown no significant difference in volume compared with late postoperative images and seem to be a better alternative to early postoperative MRI, if no subsequent resection has been performed. In this concern, the precise delineation of LGG remnants is an important aspect for further follow-up, treatment, and definition of the velocity of tumor progress. Moreover, this approach could help to identify high-risk patients because residual tumor and age older than 40 years of age are considered for adjuvant treatment.10 LGG is a chronic disease with a silent progress that is difficult to quantify with common MRI.11 Pallud et al.12 have shown that LGGs infiltrate brain tissue beyond MRI defined boundaries. It is probably not possible to detect complete tumor extension of LGG but according to our data, iMRI could amend the early postoperative MRI, which seems to overestimate the extent of LGG and hereby help to identify visible tumor remnants more precisely. Moreover, the presence of residual tumor and the correct identification of tumor progress might become clearer at 3 months’ MRI follow-up. According to FLAIR images, we distinguished between wellcircumscribed and diffuse LGGs. The subgroup analysis of both
REFERENCES 1. Ius T, Isola M, Budai R, Pauletto G, Tomasino B, Fadiga L, et al. Low-grade glioma surgery in eloquent areas: volumetric analysis of extent of resection and its impact on overall survival. A single-institution experience in 190 patients. J Neurosurg. 2012;117:1039-1052. 2. McGirt MJ, Chaichana KL, Attenello FJ, Weingart JD, Than K, Burger PC, et al. Extent of surgical resection is independently associated with survival in patients with hemispheric
196
www.SCIENCEDIRECT.com
morphologic entities showed significant differences between intraoperative or ultra-early postoperative MRI and early postoperative MRI as well as between early and late postoperative signal abnormalities. In the case of a circumscribed lesion, the difference was even more apparent. As expected, diffuse lesions were significantly more often associated with FLAIR abnormalities with volume more than 1.5 cm3 and T2 more than 1 cm3 in late postoperative MRI images. Skrap et al.13 described the infiltrative and proliferative growth pattern. Infiltrative tumors spread in diffuse manner without distinct borders and according to the study from Ius et al.1 are associated with shorter OS and PFS. In our study, we did not define the infiltrative growth according to the method described by Skrap et al.13 as the difference between preoperative T2 and postcontrast T1-weigted tumor volume. We distinguished between compact and diffuse lesions based on the appearance and tumor spread along white matter fibers in FLAIR and T2. The exact assignment might be biased in borderline cases with our method. On the contrary, it is a simple method that could be a relevant and useful prognostic factor for EoR and hereby for PFS and OS. According to our knowledge, this is the first report comparing intraoperative, early, and late postoperative tumor volume in LGG. Our study encompasses several biases. The small number of patients might have influenced data negatively. Because of the retrospective character of our study, it is not possible to confirm our statement and more studies with prospective design are needed to define the role of intraoperative and “ultra-early” postoperative imaging using iMRI in LGG surgery. CONCLUSION According to our results, iMRI without subsequent resection T2/ FLAIR shows to be superior to early postoperative imaging with regard to postoperative FLAIR/T2 abnormalities. Postoperative MRI within 48 hours overestimates the amount of residual tumor. In a setting with iMRI, final imaging without subsequent resection should be considered as a reference to evaluate residual tumor in LGG. Alternatively, ultra-early MRI might show similar results. Well-circumscribed LGG without apparent diffuse growth pattern in MRI is associated with less T2/FLAIR changes in 3-month postoperative follow-up MRI.
infiltrating low-grade gliomas. Neurosurgery. 2008; 63:700-707.
intraoperative magnetic resonance Neurosurgery. 2016;78:775-786.
imaging.
3. Hatiboglu MA, Weinberg JS, Suki D, Rao G, Prabhu SS, Shah K, et al. Impact of intraoperative high-field magnetic resonance imaging guidance on glioma surgery: a prospective volumetric analysis. Neurosurgery. 2009;64:1073-1081.
5. Pala A, König R, Hlavac M, Wirtz CR, Coburger J. Does the routine use of intraoperative MRI prolong progression free survival in low-grade glioma surgery? A retrospective study. Innovative Neurosurg. 2015;3:67-74.
4. Coburger J, Merkel A, Scherer M, Schwartz F, Gessler F, Roder C, et al. Low-grade glioma surgery in intraoperative magnetic resonance imaging. Results of a multicenter retrospective assessment of the German Study Group for
6. Belhawi SMK, Hoefnagels FW, Baaijen JC, Aliaga ES, Reijneveld JC, Heimans JJ, et al. Early postoperative MRI overestimates residual tumour after resection of gliomas with no or minimal enhancement. Eur Radiol. 2011;21:1526-1534.
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.04.120
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
7. Essig M, Metzner R, Bonsanto M, Hawighorst H, Debus J, Tronnier V, et al. Postoperative fluidattenuated inversion recovery MR imaging of cerebral gliomas: initial results. Eur Radiol. 2001;11: 2004-2010. 8. McGirt MJ, Chaichana KL, Gathinji M, Attenello FJ, Than K, Olivi A, et al. Independent association of extent of resection with survival in patients with malignant brain astrocytoma. J Neurosurg. 2009;110:156-162. 9. Essig M, Hawighorst H, Schoenberg SO, Engenhart-Cabillic R, Fuss M, Debus J, et al. Fast fluid-attenuated inversion-recovery (FLAIR) MRI in the assessment of intraaxial brain tumors. J Magn Reson Imaging. 1998;8:789-798.
10. Van Den Bent MJ. Practice changing mature results of RTOG study 9802: another positive PCV trial makes adjuvant chemotherapy part of standard of care in low-grade glioma. Neuro Oncol. 2014;16:1570-1574. 11. Pallud J, Capelle L, Taillandier L, Badoual M, Duffau H, Mandonnet E, et al. The silent phase of diffuse low-grade gliomas. is it when we missed the action? Acta Neurochir (Wien). 2013;155:2237-2242.
moral resection, clinical outcome, and survival in a consecutive series of 66 cases. Neurosurgery. 2012; 70:1081-1093.
Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Received 17 March 2016; accepted 28 April 2016
12. Pallud J, Varlet P, Devaux B, Geha S, Badoual M, Deroulers C, et al. Diffuse low-grade oligodendrogliomas extend beyond MRI-defined abnormalities. Neurology. 2010;74:1724-1731. 13. Skrap M, Mondani M, Tomasino B, Weis L, Budai R, Pauletto G, et al. Surgery of insular nonenhancing gliomas: volumetric analysis of tu-
Citation: World Neurosurg. (2016) 93:191-197. http://dx.doi.org/10.1016/j.wneu.2016.04.120 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2016 Elsevier Inc. All rights reserved.
Photo by Leonidas Quintana, MD. “Sunset at Tahai Ahu e Sunset at Tahai altar.”
WORLD NEUROSURGERY 93: 191-197, SEPTEMBER 2016
www.WORLDNEUROSURGERY.org
197
The Value of Intraoperative and Early Postoperative Magnetic Resonance Imaging in Low-Grade Glioma Surgery: A Retrospective Study Andrej Pala1, Christine Brand1, Thomas Kapapa3, Michal Hlavac1, Ralph Ko¨nig1, Bernd Schmitz2, Christian Rainer Wirtz1, Jan Coburger1
BACKGROUND: The presence of residual tumor is crucial in decision-making for low-grade gliomas (LGGs), because patients older than 40 years of age with residual tumor are considered for adjuvant treatment. There are hints that early postoperative fluid-attenuated inversion recovery (FLAIR) and T2 (within 48 hours) may overestimate residual tumor volume in LGG. Intraoperative magnetic resonance imaging (MRI) without subsequent resection or ultra-early postoperative MRI may assess the amount of residual tumor more adequately. To evaluate the utility of postoperative imaging in LGG, we volumetrically analyzed intraoperative, early, and late (3e4 months after surgery) postoperative MRIs of LGGs.
-
PATIENTS AND METHODS: A total of 33 patients with LGG were assessed retrospectively. Residual tumor was defined as signal-enhanced tissue in T2 and FLAIR. Volumetric assessment was performed with intraoperative, early, and late postoperative T2/FLAIR via Brainlab-iPlan 3.0. Wilcoxon and c2 tests were used for statistical analysis.
postoperative FLAIR/T2 mean volume [ 5.560 cm3 and 2.370 cm3, P [ 0.520, P [ 0.398), whereas a significant difference was detected between early and late postoperative images (FLAIR, P < 0.0001; T2, P < 0.00001). CONCLUSION: Intraoperative MRI without further resection or ultra-early postoperative MRI seems to reflect the actual volume of residual tumor in LGG more precisely compared with early postoperative MRI and therefore seems to be more useful regarding decisions for adjuvant therapy.
-
-
RESULTS: A significant difference of FLAIR/T2 abnormalities was found in intraoperative and early postoperative MRIs (FLAIR mean volume [ 5.433 cm3, T2 mean volume [ 3.374 cm3 vs. FLAIR mean volume [ 14.090 cm3, P [ 0.002, T2 mean volume [ 7.597 cm3, P [ 0.006). There was no significant difference between intraoperative and late postoperative FLAIR/T2 abnormalities (late
-
Key words FLAIR - Intraoperative MRI - Low-grade glioma - T2 -
Abbreviations and Acronyms DWI: Diffusion-weighted imaging EoR: Extent of resection FLAIR: Fluid-attenuated inversion recovery iMRI: Intraoperative magnetic resonance image LGG: Low-grade glioma OS: Overall survival
WORLD NEUROSURGERY 93: 191-197, SEPTEMBER 2016
INTRODUCTION
L
ow-grade gliomas (LGGs) are slow-growing, infiltrative lesions that despite all therapeutic efforts inevitably recur and transform to more malignant grades.1,2 Additional intraoperative imaging techniques, such as high-field intraoperative magnetic resonance imaging (iMRI), detect tumor remnants and increase the extent of resection (EoR) in LGG surgery.3,4 There is increasing evidence that EoR is an important positive prognostic factor for longer progression-free survival (PFS) and overall survival (OS) of patients with LGG.1,2,4,5 Early postoperative (within 48 hours) fluid-attenuated inversion recovery (FLAIR) and T2 sequences delineate tumor remnants and help to define EoR. A precise definition of residual tumor has an impact on further adjuvant therapy planning in patients with LGG.
PFS: Progression-free survival WHO: World Health Organization From the Departments of 1Neurosurgery and 2Neuroradiology, University of Ulm, Ulm; and 3 Department of Neurosurgery, University of Ulm, Günzburg, Germany To whom correspondence should be addressed: MUDr. Andrej Pala, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2016) 93:191-197. http://dx.doi.org/10.1016/j.wneu.2016.04.120 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2016 Elsevier Inc. All rights reserved.
www.WORLDNEUROSURGERY.org
191
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
There is increasing evidence that the early postoperative MRI overestimates the tumor borders and does not correlate with the extent of the disease.6,7 The use of iMRI after final tumor resection might amend the early postoperative MRI and define EoR more precisely. To date, no study has evaluated signal abnormalities between intraoperative and early and late postoperative T2/FLAIR images (34 months). To evaluate the utility of postoperative imaging in LGG, we compared the volume of signal enhancement in T2 and FLAIR sequences in intraoperative, early, and late postoperative MRI scans using Brainlab iPlan 3.0 (Brainlab AG, Feldkirchen, Germany) in patients with LGG. PATIENTS AND METHODS Patients and Follow-Up Assessment Only patients treated with World Health Organization (WHO) grade II LGG tumors between 2009 and 2015 were assessed retrospectively. Follow-up assessment including clinical status and MRI was performed at 34 months after surgery. Preoperative MRIs were not older than 14 days and included T1-weighted spin echo, T2-weigted turbo spin echo, FLAIR, and diffusion-weighted imaging (DWI) sequences. Operating Room Setup and MRI An intraoperative 1.5-T MRI Espree scanner is available (Espree, Siemens AG, Erlangen, Germany) at our department as a oneroom solution since October 2008. During surgery, an iMRI scan was performed according to surgeon’s decision. iMRI sequences included T1-weighted spin echo, T2-weigted turbo spin echo, FLAIR, and DWI sequences. In 8 cases, postoperative MRI was performed directly after surgery. Early postoperative MRI was performed within 48 hours after surgery, and late postoperative MRI was obtained 34 months after surgery. Pre- and
Figure 2. Axial T2 image of diffuse low-grade glioma.
postoperative MRIs were performed either with the intraoperative scanner or with 1.5-T MRI Symphony system (Siemens AG, Erlangen, Germany). MRI Volumetric Assessment Tumor volume was measured after image fusion via iPlan 3.0 (Brainlab AG). Volumetric assessment was performed based on
Table 1. Patient Characteristics Characteristic
n%
n
33
Age, years Median
38.7
Minemax
1161
Sex Female ratio
64%
Subtype Diffuse astrocytoma
46%
Oligodendroglioma
39%
Oligoastrocytoma
15%
Presenting symptoms
Figure 1. Axial T2 image of circumscribed low-grade glioma.
192
www.SCIENCEDIRECT.com
Seizure
64%
Headache
12%
Incidental
12%
Progress
9%
Aphasia
3%
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.04.120
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
50
40
Volume
30
20
10
0 intraoperative T2
early postoperative T2
T2 late postoperative
Figure 3. Volume of T2 abnormalities in intraoperative, early postoperative and late postoperative magnetic resonance imaging.
manual sequencing of T2 and FLAIR images in preoperative, intraoperative, early, and late postoperative T2 and FLAIR sequences. Circumscribed and diffuse growing lesions were distinguished according to FLAIR/T2 tumor boundaries (Figures 1 and 2). Circumscribed LGG was defined as a compact lesion without visible spread along white matter fibers.1 In addition, we created 2 groups of patients based on their morphology according to volume of signal abnormalities in late postoperative T2/FLAIR. The threshold for FLAIR was 1.5 cm3 and for T2 1 cm3. Furthermore, we analyzed the dependence of this tumor morphology on signal changes in late postoperative images. We didn’t apply the method used by Ius et al.1; the difference between circumscribed and diffuse lesions was based on the macroscopic infiltration of white matter tracts.
Data Analysis The data of 33 patients were evaluated retrospectively. At 34 months’ follow-up, tumor volume was evaluated. Statistical analysis was performed with SPSS 21.0 (Lead Technologies, INC, Charlotte, North Carolina, USA). Wilcoxon and c2 tests were used for the analysis. The study was conducted according to the international Declaration of Helsinki.
WORLD NEUROSURGERY 93: 191-197, SEPTEMBER 2016
RESULTS Patient Characteristics A total number of 33 patients treated for WHO grade II LGGs were assessed. The most common histologic subtype of LGG was diffuse astrocytoma (n ¼ 15, 46%), followed by oligodendroglioma (n ¼ 13, 39%) and oligoastrocytoma (n ¼ 5, 15%). A mean age of 38.7 (range 11e61) years was noted. The majority of patients were female (64%, n ¼ 21). The demographic data are summarized in Table 1. Seizure was the most common presenting symptom (n ¼ 21, 64%). Recurrent tumor was treated in 12% (n ¼ 4). Volumetric and Statistical Analysis The mean preoperative tumor volume was 33.766 cm3 in T2 and 39.193 cm3 in FLAIR. iMRI was available for all patients, whereas no subsequent resection was performed in 22 cases. The mean volume of intraoperative T2 and FLAIR without subsequent resection was 3.374 and 5.433 cm3, respectively. Eight patients obtained “ultra-early” postoperative MRI directly after wound suturing and were evaluated together with the iMRI cohort without further resection. An early postoperative T2 was obtained in 26 and FLAIR in 25 cases. The mean volume of early postoperative T2 and FLAIR was 7.597 cm3 and 14.090 cm3, retrospectively. A significant difference of T2 and FLAIR abnormalities was found in
www.WORLDNEUROSURGERY.org
193
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
100
80
Volume
60
40
20
0 intraoperative FLAIR
early postoperative FLAIR
Figure 4. Volume of fluid-attenuated inversion recovery abnormalities in intraoperative, early postoperative, and
intraoperative and early postoperative MRI (P ¼ 0.006, P ¼ 0.002) in Wilcoxon test (Figures 3 and 4). Late postoperative FLAIR was available for all patients. In 1 patient, no T2 sequences were available in the follow-up. No significant difference was noted between intraoperative and late postoperative FLAIR and T2 abnormalities (late postoperative FLAIR and T2 mean volume ¼ 5.560 cm3 and 2.370 cm3, P ¼ 0.520, P ¼ 0.398, Figures 3 and 4). In contrast, a significant difference was detected between early and late postoperative images (T2, P < 0.00001, FLAIR, P < 0.0001). According to the tumor’s FLAIR and T2 infiltration pattern of white matter pathways, 19 circumscribed (Figure 1) and 14 diffuse (Figure 2) gliomas were distinguished. Subgroup analysis of both morphologic entities showed similar results and significant differences between intraoperative and early postoperative FLAIR/T2 except for T2 of diffuse lesions (diffuse LGG, FLAIR P ¼ 0.043, T2 P ¼ 0.116, circumscribed LGG, FLAIR P ¼ 0.021, T2 P ¼ 0.015) as well as significant differences in signal abnormalities between early and late postoperative FLAIR/T2 (diffuse LGG, FLAIR P ¼ 0.025, T2 P ¼ 0.012, circumscribed LGG FLAIR P ¼ 0.0003, T2 P ¼ 0.0003). No significant difference was noted between intraoperative and late postoperative FLAIR and T2 abnormalities in the subgroup analysis (diffuse LGG, FLAIR P ¼ 0.799, T2 P ¼ 0.441, circumscribed LGG FLAIR P ¼ 0.477, T2 P ¼ 0.799). Fourteen
194
www.SCIENCEDIRECT.com
late postoperative FLAIR
late postoperative magnetic resonance imaging.
patients were identified in the late postoperative FLAIR with residual signal abnormalities less than 1.5 cm3, and 16 patients had signal changes with the volume less than 1 cm3 in T2. FLAIR signal abnormalities with volume less than 1.5 cm3 and T2 abnormalities with volume less than 1 cm3 were significantly more often associated with morphologically circumscribed lesions in late postoperative images (FLAIR, P ¼ 0.026, T2, P ¼ 0.016 with c2 test). Illustrative Case After an episode of generalized seizure and slight short-term memory disturbances on neurologic examination, a 46-year-old female patient presented with a T2 hyperintense right frontal mass lesion on MRI without gadolinium enhancement (Figure 1). Functional MRI confirmed the activation of primary motor region directly behind the mass lesion on the right side. Surgery with direct cortical stimulation, continuous monitoring of motor-evoked potentials, and iMRI was performed. After intraoperative impression of gross total resection, the surgeon indicated an iMRI scan. A potential small tumor remnant was detected directly in the precentral gyrus on iMRI (Figure 5). Because of the location in the primary motor region verified by direct cortical and subcortical stimulation, no further resection was performed. The histopathologic report confirmed
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.04.120
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
oligoastrocytoma WHO grade II. No transient or permanent focal neurologic deficit was found 3 months after the surgery. The early postoperative MRI showed much more pronounced T2 signal abnormalities (Figure 6) compared with iMRI (Figure 5). The exact distinction between tumor remnant (Figure 6, white arrow) and surgically induced artifacts (Figure 6, black arrow) was not possible in the early phase after the surgery. In the late postoperative MRI, T2 tumor depiction in the precentral gyrus (Figure 7, white arrow) was similar to the iMRI. No surgically induced changes are detectable in the late postoperative MRI. DISCUSSION LGGs are considered as WHO grade II infiltrative neoplasms that despite their slowly growing pattern recur and inevitably deteriorate to malignant phenotype.1,8 FLAIR and T2 sequences has been shown to play an important role in follow-up and in further management of LGG.7 EoR becomes an important positive prognostic factor for PFS and OS, and the proof of tumor remnants in postoperative MRI imaging influences further adjuvant therapy.1,2,4,5 FLAIR sequences delineate tumor boundaries more precisely than T2, especially in tumors located near cerebrospinal fluid space and enables better distinguishing between tumor and brain edema.9 Residual proteins and blood degradation products, such as methemoglobin, influence the relaxation time of cerebrospinal fluid in MRI. FLAIR consequently is not able to optimize inversion time and leads to artifacts in early postoperative images7 and might obscure possible tumor remnant.6 The routine application of a final iMRI scan or an ultra early MRI after resection might minimize the artifacts and define tumor borders more precisely than early postoperative MRI. We conducted a retrospective assessment of all LGG surgeries performed at our center from 2009 to 2015 and evaluated the
Figure 5. Intraoperative axial T2 image of low-grade glioma with suspected tumor remnant in the precentral gyrus (white arrow).
WORLD NEUROSURGERY 93: 191-197, SEPTEMBER 2016
Figure 6. Early postoperative axial T2 image with advanced surgical-induced T2 enhancement (black arrow) and suspected tumor remnant with pronounced T2 abnormalities (white arrow) in the precentral gyrus.
Figure 7. Late postoperative axial T2 image with suspected tumor remnant (white arrow) in the precentral gyrus.
www.WORLDNEUROSURGERY.org
195
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
volume of T2/FLAIR abnormalities in intraoperative, early, and late postoperative images. We found a significant difference between T2/FLAIR abnormalities in intraoperative and early postoperative MRI, whereas no significant difference was seen between intraoperative and late postoperative T2/FLAIR images. The routine MRI scan in early postoperative phase might be redundant, if no further resection was performed after iMRI scan. Belhawi et al.6 found in T2 22% and in FLAIR 49% smaller residual tumor volume when they compared early and late postoperative MRI, confirming that early postoperative MRI overestimates the tumor borders in glioma patients with no gadolinium enhancement and that the late postoperative MRI delineate tumor remnants more precisely.6 In our study, we evaluated a more precise cohort of only WHO grade II tumors compared with Belhawi et al.,6 who evaluated gliomas in general. We did not analyze the DWI and did not exclude the typical ischemic T2/FLAIR abnormalities.6 Despite this fact we found a significant difference between intraoperative and early postoperative T2/FLAIR and between early and late postoperative T2/FLAIR images. Such a rapid tumor regression seems to be unrealistic, even if some ischemic-dependent tumor shrinkage is possible. Nevertheless, the intraoperative T2/FLAIR abnormalities have shown no significant difference in volume compared with late postoperative images and seem to be a better alternative to early postoperative MRI, if no subsequent resection has been performed. In this concern, the precise delineation of LGG remnants is an important aspect for further follow-up, treatment, and definition of the velocity of tumor progress. Moreover, this approach could help to identify high-risk patients because residual tumor and age older than 40 years of age are considered for adjuvant treatment.10 LGG is a chronic disease with a silent progress that is difficult to quantify with common MRI.11 Pallud et al.12 have shown that LGGs infiltrate brain tissue beyond MRI defined boundaries. It is probably not possible to detect complete tumor extension of LGG but according to our data, iMRI could amend the early postoperative MRI, which seems to overestimate the extent of LGG and hereby help to identify visible tumor remnants more precisely. Moreover, the presence of residual tumor and the correct identification of tumor progress might become clearer at 3 months’ MRI follow-up. According to FLAIR images, we distinguished between wellcircumscribed and diffuse LGGs. The subgroup analysis of both
REFERENCES 1. Ius T, Isola M, Budai R, Pauletto G, Tomasino B, Fadiga L, et al. Low-grade glioma surgery in eloquent areas: volumetric analysis of extent of resection and its impact on overall survival. A single-institution experience in 190 patients. J Neurosurg. 2012;117:1039-1052. 2. McGirt MJ, Chaichana KL, Attenello FJ, Weingart JD, Than K, Burger PC, et al. Extent of surgical resection is independently associated with survival in patients with hemispheric
196
www.SCIENCEDIRECT.com
morphologic entities showed significant differences between intraoperative or ultra-early postoperative MRI and early postoperative MRI as well as between early and late postoperative signal abnormalities. In the case of a circumscribed lesion, the difference was even more apparent. As expected, diffuse lesions were significantly more often associated with FLAIR abnormalities with volume more than 1.5 cm3 and T2 more than 1 cm3 in late postoperative MRI images. Skrap et al.13 described the infiltrative and proliferative growth pattern. Infiltrative tumors spread in diffuse manner without distinct borders and according to the study from Ius et al.1 are associated with shorter OS and PFS. In our study, we did not define the infiltrative growth according to the method described by Skrap et al.13 as the difference between preoperative T2 and postcontrast T1-weigted tumor volume. We distinguished between compact and diffuse lesions based on the appearance and tumor spread along white matter fibers in FLAIR and T2. The exact assignment might be biased in borderline cases with our method. On the contrary, it is a simple method that could be a relevant and useful prognostic factor for EoR and hereby for PFS and OS. According to our knowledge, this is the first report comparing intraoperative, early, and late postoperative tumor volume in LGG. Our study encompasses several biases. The small number of patients might have influenced data negatively. Because of the retrospective character of our study, it is not possible to confirm our statement and more studies with prospective design are needed to define the role of intraoperative and “ultra-early” postoperative imaging using iMRI in LGG surgery. CONCLUSION According to our results, iMRI without subsequent resection T2/ FLAIR shows to be superior to early postoperative imaging with regard to postoperative FLAIR/T2 abnormalities. Postoperative MRI within 48 hours overestimates the amount of residual tumor. In a setting with iMRI, final imaging without subsequent resection should be considered as a reference to evaluate residual tumor in LGG. Alternatively, ultra-early MRI might show similar results. Well-circumscribed LGG without apparent diffuse growth pattern in MRI is associated with less T2/FLAIR changes in 3-month postoperative follow-up MRI.
infiltrating low-grade gliomas. Neurosurgery. 2008; 63:700-707.
intraoperative magnetic resonance Neurosurgery. 2016;78:775-786.
imaging.
3. Hatiboglu MA, Weinberg JS, Suki D, Rao G, Prabhu SS, Shah K, et al. Impact of intraoperative high-field magnetic resonance imaging guidance on glioma surgery: a prospective volumetric analysis. Neurosurgery. 2009;64:1073-1081.
5. Pala A, König R, Hlavac M, Wirtz CR, Coburger J. Does the routine use of intraoperative MRI prolong progression free survival in low-grade glioma surgery? A retrospective study. Innovative Neurosurg. 2015;3:67-74.
4. Coburger J, Merkel A, Scherer M, Schwartz F, Gessler F, Roder C, et al. Low-grade glioma surgery in intraoperative magnetic resonance imaging. Results of a multicenter retrospective assessment of the German Study Group for
6. Belhawi SMK, Hoefnagels FW, Baaijen JC, Aliaga ES, Reijneveld JC, Heimans JJ, et al. Early postoperative MRI overestimates residual tumour after resection of gliomas with no or minimal enhancement. Eur Radiol. 2011;21:1526-1534.
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.04.120
ORIGINAL ARTICLE ANDREJ PALA ET AL.
INTRAOPERATIVE AND EARLY POSTOPERATIVE MRI IN LOW-GRADE GLIOMA
7. Essig M, Metzner R, Bonsanto M, Hawighorst H, Debus J, Tronnier V, et al. Postoperative fluidattenuated inversion recovery MR imaging of cerebral gliomas: initial results. Eur Radiol. 2001;11: 2004-2010. 8. McGirt MJ, Chaichana KL, Gathinji M, Attenello FJ, Than K, Olivi A, et al. Independent association of extent of resection with survival in patients with malignant brain astrocytoma. J Neurosurg. 2009;110:156-162. 9. Essig M, Hawighorst H, Schoenberg SO, Engenhart-Cabillic R, Fuss M, Debus J, et al. Fast fluid-attenuated inversion-recovery (FLAIR) MRI in the assessment of intraaxial brain tumors. J Magn Reson Imaging. 1998;8:789-798.
10. Van Den Bent MJ. Practice changing mature results of RTOG study 9802: another positive PCV trial makes adjuvant chemotherapy part of standard of care in low-grade glioma. Neuro Oncol. 2014;16:1570-1574. 11. Pallud J, Capelle L, Taillandier L, Badoual M, Duffau H, Mandonnet E, et al. The silent phase of diffuse low-grade gliomas. is it when we missed the action? Acta Neurochir (Wien). 2013;155:2237-2242.
moral resection, clinical outcome, and survival in a consecutive series of 66 cases. Neurosurgery. 2012; 70:1081-1093.
Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Received 17 March 2016; accepted 28 April 2016
12. Pallud J, Varlet P, Devaux B, Geha S, Badoual M, Deroulers C, et al. Diffuse low-grade oligodendrogliomas extend beyond MRI-defined abnormalities. Neurology. 2010;74:1724-1731. 13. Skrap M, Mondani M, Tomasino B, Weis L, Budai R, Pauletto G, et al. Surgery of insular nonenhancing gliomas: volumetric analysis of tu-
Citation: World Neurosurg. (2016) 93:191-197. http://dx.doi.org/10.1016/j.wneu.2016.04.120 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2016 Elsevier Inc. All rights reserved.
Photo by Leonidas Quintana, MD. “Sunset at Tahai Ahu e Sunset at Tahai altar.”
WORLD NEUROSURGERY 93: 191-197, SEPTEMBER 2016
www.WORLDNEUROSURGERY.org
197