Diffusion tensor imaging and tractography in diffuse intrinsic pontine glioma – A major determinant of resectability and description of a new subtype
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.253625
Source of Support: None, Conflict of Interest: None
Diffuse infiltrative pontine gliomas (DIPGs) form about 80% of all pediatric brain stem gliomas. They are diagnosed by their clinical features and by conventional computed tomography (CT) and magnetic resonance imaging (MRI) scans. DIPGs are subjected to direct radiation therapy, as most authors feel that the potential risk of destroying important fiber tracts and the resultant morbidity outweighs the therapeutic benefit of tumor debulking.,, More recently, more patients are being subjected to biopsy of the lesion, as there has been inconsistency in diagnosis and the complete reliability on radiology for prognosis has been questioned by many.,,,,, There is also a feeling that there has not been much tangible benefit from either radiation or chemotherapy, and there is a need to look at the probability and benefit of surgery in these patients, while trails are on for targeted therapies.,,,,
The use of diffusion tensor imaging (DTI) in the management of supratentorial lesions has prompted many to look at its use in brain stem lesions., More importantly, the use of quantitative fractional anisotropy (FA) and apparent diffusion coefficient (ADC) values of the lesion and the involvement of the major fiber tracts in the pons have been studied and have been found to be helpful in deciding the prognosis and the role of surgery in intrinsic brain stem lesions.,, We present a child with a diffuse pontine lesion, where we have shown for the first time in literature, displacement and lack of infiltration or destruction of major fiber tracts of the pons on MRI, suggesting a subtype of these lesions where the tumor is noninfiltrative and, therefore, amenable to surgery.
An 8-year-old girl presented with progressive weakness of the right-sided limbs over 4 months, slurring of speech, and difficulty in swallowing for the past 2 months. There were no features of raised intracranial pressure nor any seizures. There was no preceding history of fever or trauma. On examination, she was conscious and alert with dysarthric speech and a hypertonic 4/5 right hemiparesis. Her extraocular movements and lower cranial nerves were normal. She had a hemiparetic gait. Her MRI of the brain showed an expansile pontine lesion which was hypointense on T1, hyperintense on T2, with no true diffusion restriction, obliterating the peripontine and infrapontine cisterns as well as the suprasellar cistern [Figure 1]. The lesion was seen to engulf the basilar artery. There was no evidence of hemorrhage, necrosis, or calcification within the lesion. Small focal areas of enhancement with corresponding increased perfusion were present on the left side near the level of the internal auditory meatus. The fourth ventricle was compressed and displaced posteriorly, with no ventriculomegaly. On DTI, the ADC value at the center of the lesion was 2.05 (×10−3 mm 2/s) while the FAL was 0.103. The corticospinal, transverse pontine, and medial lemniscal fiber tracts were tracked and color-coded. The corticospinal tract, being the major fiber tract, was studied bilaterally at the upper pontine, mid pontine, and pontomedullary levels, while the transverse pontine and medial lemnisci were studied at the mid pontine level. The FA values of the corticospinal tracts on both sides were found to be decreased at all levels [Table 1]. The FA values of the transverse pontine and medial lemnisci were within the normal range [Table 1]. Studying the anatomical location of the tracts on DTI and with color-coded tractograghy, the corticospinal tracts were seen to be predominantly displaced posteriorly, as were the central transverse pontine fibers and the medial lemniscus, with neither evidence of destruction nor infiltration [Figure 2] and [Figure 3].
The child underwent surgery using a navigation-guided retromastoid approach with intraoperative monitoring [Figure 4]. The brain stem was seen bulging between the Vth and the VII–VIII nerve complex. Confirming our position with navigation, the brain stem was entered at an avascular area in the peritrigeminal dorsal root entry zone. The tumor was encountered just beneath the pia and was greyish, soft, and suckable. Subtotal excision was done, taking care to proceed in an anteromedial direction away from the displaced fiber tracts. The basilar artery was recognized at the end point of surgery, as we did not want to cross the midline. There were no changes in the intraoperative neuromonitoring throughout the surgery. The child was ventilated overnight and extubated the next morning. There were no new deficits. Aggressive rehabilitation therapy was initiated and the child made a good progress and attained normal neurological status in 1 week. The pathology showed a moderately cellular tumor in a fibrillary matrix. The tumor cells showed mild-to-moderate pleomorphism with clumped chromatin. Focal microcystic changes were seen with only occasional mitosis and no endothelial proliferation. Isocitrate dehydrogenase-1 was negative, ATRX was positive, and there was a Ki67 index of 10%–15% [Figure 5]. The child underwent radiation therapy which has been completed successfully. She is now on a regular follow-up for the last 4 months.
Diffuse pontine gliomas occur more commonly in children and comprise approximately 80% of all pediatric brain stem gliomas. The appearance of a diffuse expansion of the pons, hypointense on T1W and hyperintense on T2W, with increased diffusion, with inconsistent contrast enhancement in a patient with rapid progression of symptoms is considered pathognomonic of DIPG. However, complete reliability on radiology has been called into question by many authors.,,, Hankinson et al., concluded from their online survey of MRI images of pediatric pontine lesions that there was considerable inconsistency in diagnosis as far as individual patients were concerned and that the practice of diagnosing DIPG based on imaging characteristics and clinical history alone does not reach the appropriate threshold to be considered a standard of care.
Novel MRI imaging techniques have been used to improve the diagnosis and the prognostication of intra-axial tumors in the brain, and more recently, of brain stem gliomas. DTI evaluates the molecular motion of water along nerve fiber tracts and this has been found to be abnormal in and around brain tumors. Currently, it has become an essential MRI tool in the surgical planning of tumors located in eloquent and deep areas of the brain, especially as it is able to quantitatively evaluate ADC and the FA values. These values have been used to characterize tumors and suggest prognosis and recovery following treatment., It provides a better visualization of white matter tracts and is a promising method for evaluating the involvement of fiber tracts by the lesion, to improve diagnosis and surgical planning.,,,, Field et al., reported preoperative DTI color mapping to augment surgical planning in nine adult patients with supratentorial tumors., They examined the relationships between the primary tumor and the adjacent fiber tracts and compared them with the contralateral normal hemisphere. They then characterized the tract involvement into four patterns depending on the FA and ADC values as well as the location and possible displacement of the tracts. The authors reported that intact white matter tracts, including those displaced by the tumor but having normal FA values, are present in regions of the brain that appear abnormal on conventional MR images and that the study substantially improved preoperative planning. These and other articles on supratentorial tumors were taken forward by Helton et al., who studied the involvement of tracts in diffuse brain stem gliomas using DTI and divided their findings of involvement of fiber tracts into five types (normal, displaced, edematous, infiltrated, and disrupted) based on the FA and ADC values and the appearance on T2-weighted MRI images.,,, They concluded that DTI analysis can delineate tract invasion and displacement and that these would help to better discriminate between diffuse and focal brain stem tumors in the future and may be useful for guiding surgical biopsies.
Phillips et al., in 2005 described the use of DTI in the management of focal low-grade brain stem gliomas to better delineate the peritumoral tracts and demarcate the borders between the lesion and the normal brain stem. In their opinion, accurate localization of the tumor in relation to important white matter tracts could affect the decision of whether to operate or not, besides contributing valuable information to decide on the surgical resectability and planning. DTI has since been used in the study of brain stem gliomas not only as a differentiation point between focal and diffuse lesions but also as a prognosticator of focal lesions, as an indicator of extrapontine extention of lesions and as a tool to evaluate treatment protocols of these lesions.,,,,, Lober et al., used ADC values from diffusion imaging to distinguish two subtypes of DIPG with a distinct prognosis. Yang et al., reporting on seven adults who underwent excision of intrinsic nonexophytic pontine gliomas. They were of the opinion that the possibility of surgical intervention in pontine gliomas exists as long as there is a proper preoperative workup performed. All their patients underwent DTI preoperatively, which showed infiltration in only one of their seven patients, while in the other six, the tracts were “pushed.” This finding, as per the authors, along with image guidance and intraoperative monitoring, allowed them to choose the ideal trajectory and allowed for maximal resection of the lesions. Their report, however, excluded diffuse gliomas without focal enhancement. Chen et al., looked at the ADC and FA values in nine patients with DIPG. They found that the mean ADC in the lesion was 1.14 ± 0.18 (×10−3 mm 2/s), which was higher than the mean ADC of normal brain stem controls 0.75 ± 0.06 (×10−3 mm 2/s). The mean FA in the lesions of DIPG was 0.24 ± 0.04, which was significantly lower than the mean FA of normal brain stem in the controls (0.43 ± 0.02). Our patient had an ADC of 2.05 (×10−3 mm 2/s) and an FA of 0.103. These high ADC and low FA values were consistent with the values in DIPG with a poor prognosis.
The role of surgery in DIPG has been minimal with most authors and institutions believing that the clinical and radiological presentations of the patient are enough to warrant further management without the need for any surgery or a biopsy. However, this thought has been brought into question with data showing the difficulty in distinguishing typical from atypical lesions. The development of refined neurosurgical techniques has increased the frequency of performing biopsies, with low complications and high success rates.,,,, Wang et al., in their retrospective study of 15 cases of DIPG, have shown that biopsy of DIPG did not adversely affect the patients. None of the 15 children suffered any surgery-related mortality or permanent deficit. While the prognosis of DIPG has remained dismal despite advances in radiation and chemotherapy, there still remains a small subgroup of patients whose survival rate provides hope.,
We have studied DTI in patients with focal brain stem gliomas (FSBGs) where surgery was indicated. In patients with FBSG, the displacement of the fiber tracts is common and we have used this as a factor in determining the resectability and the route of approach, as has been the practice by other authors as well.
DTI was performed in the patient reported as a part of our current imaging protocol of intra-axial brain tumors. The findings in our patient differ strikingly from all other reports in literature in that on DTI color coding of the tracts, the corticospinal, transverse pontine, and medial lemnisci were all predominantly and significantly displaced posteriorly, without evidence of infiltration or destruction. We compared the FA values of the displaced tracts with controls [Figure 6] and [Figure 7]. We looked at the corticospinal tract (CST) at three levels, namely, the upper pons, mid pons, and lower pons or pontomedullary junction. The FA values were taken from both sides at each level. The FA values of the CST were decreased by an average of 40%, 42%, and 38%, respectively, at the upper, mid, and lower pontine levels. This decrease, while being more than the 25%, as defined by Field et al., was associated with a gross displacement of the tract without fiber loss. This finding meant that the pathology in our patient was significantly of Pattern 1 of Field et al., and that of “displacement” of Helton et al.,, The significance of decreased FA values could probably be explained by edema at the interface between the tumor and the tracts, especially as the values of the transverse pontine fibers and the medial lemnisci were normal on both sides across the length of the pons, despite also being displaced along with the CST. Yao et al., looking at the prognostic value of the diffusion values of the CST in brain stem surgery, felt that morphological characteristics and diffusion parameters at the lesion level cannot accurately predict a poor prognosis. This is borne out by the fact that our patient improved in neurology after the surgery despite an approximate 40% reduction in FA value of the corticospinal tracts at the level of the lesion.
In neurosurgery, we regularly strive to remove compressive elements and reduce the patient's tumor load in an effort to improve the patient's quality of life and to improve survival chances. The finding in our patient that displacement of the fiber tracts was the significant pathology, and that neither infiltration nor destruction of the tracts was present, helped us take a decision to undertake surgery. We considered surgery as a means to not only get a proper biopsy of the lesion but also affect a decompression of the displaced and compressed fiber tracts. The posterior displacement of the tracts shown by DTI made us choose the retromastoid approach. Using image guidance, we debulked the tumor, keeping the trajectory away from the posterior third of the pons and toward the basilar artery. We were wary of crossing the midline as we did not want unwarranted morbidity. However, with the experience of this patient behind us, we are now confident that using DTI and image guidance, one can effectively perform a subtotal to a near-total excision of similar lesions. The result of the surgery reaffirmed the concept that when the fiber tracts are merely displaced and not infiltrated and destroyed, maximal safe surgical excision will not cause a major disability or morbidity, as was once believed. In fact, surgery actually improves the patient's neurological status, as was the case with our patient.
This case has, therefore, brought into question how we have been looking at DIPG. Until now, there has been a nihilistic approach to these patients, with direct radiation and chemotherapy being the main management tools. Both these have not shown any significant benefit. However, with a recent increase in the frequency of biopsy of these lesions, there is hope that there will be more knowledge gleaned as to the varied biology of these lesions and, subsequently, effective targeted therapies will be developed.,,,,, We believe that the option of safe surgical excision should be included in the management algorithm of DIPG, with DTI being the major determinant factor in the surgical decision making. DTI and color-coded tractography will allow surgeons to decide “safe zones” from which biopsies can be taken; while a combination of FA values and the fiber tract anatomy will allow decisions regarding “maximally safe excision.”,,,,
We postulate that DIPGs are of two types – the classical DIPG with infiltration of the tracts and lowered FA values, and the “atypical DIPG” or “noninfiltrative” DIPG, where the tracts are merely displaced with normal or slightly decreased FA values. The latter would be candidates for excisional surgery. It is, therefore, paramount that the radiological assessment of DIPG should include DTI and color-coded tractography, for their accurate diagnosis and the development of management strategy. Routine DTI study of patients with suspect DIPG would give us more information on the prevalence of the “atypical DIPG,” which in turn would create a paradigm shift in our thinking of these lesions.
Helton et al., mentioned that “DTI provides a wealth of information that, until now, has not been available in surgical planning for patients with pontine tumours.” The time has come to change and allow for progress in the management of these lesions with hitherto dismal prognosis.
DIPGs have been classically diagnosed using clinical and conventional CT and MRI radiology. DTI has recently been used to study brain stem lesions. We have shown in our patient that the major fiber tracts are neither infiltrated nor destroyed but grossly displaced posteriorly, allowing for safe excisional surgery. DTI is an essential radiological tool in the management protocol of an intrinsic pontine lesion. We believe that there is a subset of patients with DIPG who could undergo safe maximal resection, where the tumor predominantly displaces the major fiber tracts and does not infiltrate or destroy them. The algorithm for the management of DIPG should change to include this significant finding, so that with new knowledge and new ideas, patients with DIPG can be offered some hope in the future.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]