Navigable 3D-Ultrasound Facilitates Supra-Radical Resections beyond the Contrast-Enhancing Boundaries in Malignant Gliomas

Dear Editor,

Intraoperative ultrasound (US) has been shown to improve visualization and resection of gliomas. Studies have demonstrated high gross total resection (GTR) rates in gliomas.[1] [2] Studies dealing with extent of resection (EOR) in gliomas are hampered by the lack of uniform radiologic criteria accurately defining the extent of the disease. For malignant gliomas, it is widely acknowledged that the tumor extends beyond the contrast-enhancing margins visible on magnetic resonance imaging (MRI). This includes T2 signal abnormalities for varying distances beyond the rim of contrast enhancement. Whereas it is easy (and reliable) to delineate the volume of contrast-enhancing tumor, the same cannot be said regarding the nonenhancing infiltrating zone. Direct measurement of T2 abnormalities are limited by a lack of clarity in distinguishing nonenhancing tumor from edema. In the postoperative setting, ischemic and other tissue changes imparted by surgical handling further add to the problems of assessing residual disease. Acknowledging this limitation, the Response Assessment in Neuro-Oncology (RANO) group has suggested using the term CRET, or complete resection of enhancing tissue, to denote a radical resection.[3] What this fails to convey, however, is the volume of nonenhancing tumor that may be resected over and above the CRET. Disregarding the additional volume resected may misrepresent the actual EOR. Measuring this may be crucial for comparing tools where the end point is EOR. With improvements in imaging adjuncts (such as fluorescence guidance, US, and MRI) it is likely that achieving CRET is going to be easier. In that case deciding between techniques would require more sensitive measures of EOR. Even otherwise, quantifying the actual EOR would be preferable in clinical trials where EOR is an important prognostic factor.

Schucht et al described an indirect method to measure this.[4] They essentially described a way to calculate the volume of cerebral tissue before and after the resection, taking into account the brain shift. The difference between these volumes would provide the total volume of resected brain. Knowing the preoperative contrast-enhancing tumor volume, the excess tissue resected can be calculated. We used this method in our cases. Five patients diagnosed with potentially resectable contrast-enhancing high-grade gliomas were analyzed. They were part of an ongoing clinical study approved by the hospital ethics committee evaluating the role of navigable ultrasound (NUS) in the resection of malignant gliomas. In this study, following consent, patients were randomized to either conventional navigation or NUS. The present report only describes the results in five of the patients in the NUS group. We used the SonoWand system (SonoWand, Trondheim, Norway) for performing the NUS as we previously described.[2] Preoperative and postoperative MRI scans were performed using a uniform scanning protocol for volumetric sequences on a 3-T system (Signa HDxT, General Electric, Milwaukee, Wisconsin, United States). Pre- and post-gadolinium T1 3D fast spoiled gradient-recalled or brain volume imaging (BRAVO) sequences, and T2 CUBE sequences were used for accurately delineating the tumor and postoperative resection cavity. Volumetric assessments were subsequently performed using iPlan RT Image v.4.1. cranial software (Brainlab AG, Feldkirchen, Germany) as per the method described by Schucht et al. This essentially involved manual segmentation of the contrast-enhancing tumor with central necrosis in the preoperative scan, and of the resection cavity and contrast-enhancing residue in the postoperative scan, across all the slices. Pre- and postoperative cerebrum volume was calculated by autosegmentation and manual correction of the same. [Fig. 1] depicts typical snapshots from the software showing segmentation of the pre- and postoperative volumes. [Table 1] shows the volumes calculated. The additional non–contrast-enhancing tumor volume resected ranged from 3.8 to 37.9 cc with a mean of 17.5 cc. This means that in every case where we achieved a radiologic GTR, we were able to resect an average of 17.5 cc of additional tumor (not demonstrable by routine MR volumetry based on measurement of contrast-enhancing tumor).

Fig. 1 Snapshots of one patient depicting the performance of a typical segmentation. (A) Preoperative T1 volume of the tumor. (B) Postoperative images with resection cavity. Note that the cavity appears smaller than the preoperative tumor volume. However, the calculated tumor resection volume using the software was greater than the tumor volume.

Table 1 Summary of pre- and postoperative volumes computed for the five patients

Patient no.	Preoperative contrast-enhancing tumor volume, cc: A	Preoperative total cerebral volume, cc: B	Preoperative normal cerebral volume, cc: C = A − B	Postoperative resection cavity volume, cc: D	Postoperative residual contrast-enhancing volume, cc: E	Postoperative cerebral volume, cc: F	Postoperative normal cerebral volume, cc: G = F – (E + D)	Nonenhancing tumor volume resected, cc: H = C − G
1	34.4	1001	966.6	16.7	0	975.1	958.4	8.2
2	52.2	1020	967.8	44.1	0	1008	963.9	3.8
3	47.7	943.6	896	11.3	0	890.9	879.6	16.4
4	21.1	1009	988	27.8	0	977.9	950.1	37.9
5	82.6	1129	1046.4	14.9	0	1040	1025.2	21.2

One may question the benefit of this additional resection. Sufficient evidence has now accumulated to support radical resections in malignant gliomas.[5] Although some studies suggest that increasing the percentage resection (even if short of a GTR) provides for proportionately incremental benefits,[6] [7] achieving a CRET remains the surgical goal. It has also been shown that by using aminolevulinic acid, when resections extend beyond the contrast-enhancing tumor volume, the survival benefits are greater.[8] What this eventually implies is that there is no sacrosanct end point for resection. Hence to extract the maximum benefit, neurosurgeons should aim for the best possible resection, albeit keeping in mind the safety of the patient. We showed earlier that high rates of GTR up to 88% can be achieved using NUS in similarly enhancing, potentially resectable malignant gliomas.[2] In this study we document for the first time the volume of extended resections possible. A mean of 17.5 cc was resected beyond the contrast-enhancing tumor volume. Although the technique we used can have some limitations, as discussed by Schucht et al, the results show that using NUS, extended resections can be achieved. It has previously been shown that in gliomas, US correlates better with T2 images than with contrast T1 images.[9] Thus using NUS, one can theoretically resect tumor volume beyond the contrast-enhancing boundary. The problem with following the T2 abnormalities is that it is not always possible to discriminate the tumor infiltration from the edema surrounding it. US may be able to discriminate edema from tumor because of differing echogenic characteristics. Pretreatment with steroids can help reduce edema and improve the diagnostic accuracy of the US. Although the volumetric method used by us has inherent limitations as described by Schucht et al, it is still possibly the only available method to our knowledge to measure this reliably. Although measurement of the T2 abnormality or intraoperative US volumetry are more direct methods of documenting this, they are still not standardized enough for routine use.

• References

• 1 Solheim O, Selbekk T, Jakola AS, Unsgård G. Ultrasound-guided operations in unselected high-grade gliomas—overall results, impact of image quality and patient selection. Acta Neurochir (Wien) 2010; 152 (11) 1873-1886
• 2 Moiyadi AV, Shetty PM, Mahajan A, Udare A, Sridhar E. Usefulness of three-dimensional navigable intraoperative ultrasound in resection of brain tumors with a special emphasis on malignant gliomas. Acta Neurochir (Wien) 2013; 155 (12) 2217-2225
• 3 Vogelbaum MA, Jost S, Aghi MK , et al. Application of novel response/progression measures for surgically delivered therapies for gliomas: Response Assessment in Neuro-Oncology (RANO) Working Group. Neurosurgery 2012; 70 (1) 234-243 ; discussion 243–244
• 4 Schucht P, Knittel S, Slotboom J , et al. 5-ALA complete resections go beyond MR contrast enhancement: shift corrected volumetric analysis of the extent of resection in surgery for glioblastoma. Acta Neurochir (Wien) 2014; 156 (2) 305-312 ; discussion 312
• 5 Sanai N, Berger MS. Glioma extent of resection and its impact on patient outcome. Neurosurgery 2008; 62 (4) 753-764 ; discussion 264–266
• 6 Sanai N, Polley MY, McDermott MW, Parsa AT, Berger MS. An extent of resection threshold for newly diagnosed glioblastomas. J Neurosurg 2011; 115 (1) 3-8
• 7 Chaichana KL, Jusue-Torres I, Navarro-Ramirez R , et al. Establishing percent resection and residual volume thresholds affecting survival and recurrence for patients with newly diagnosed intracranial glioblastoma. Neuro Oncol 2014; 16 (1) 113-122
• 8 Aldave G, Tejada S, Pay E , et al. Prognostic value of residual fluorescent tissue in glioblastoma patients after gross total resection in 5-aminolevulinic acid-guided surgery. Neurosurgery 2013; 72 (6) 915-920 ; discussion 920–921
• 9 Unsgaard G, Selbekk T, Brostrup Müller T , et al. Ability of navigated 3D ultrasound to delineate gliomas and metastases—comparison of image interpretations with histopathology. Acta Neurochir (Wien) 2005; 147 (12) 1259-1269 ; discussion 1269