| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Neuroradiology |
1 From the Department of Radiology, Division of Neuroradiology (J.R.P., L.M.S., J.S.P.), Brain Imaging and Analysis Center (K.M.H., J.T.V.), and Department of Surgery, Division of Neurosurgery (A.H.F., T.M.G., J.H.S.), Duke University Medical Center, Box 3808, Durham, NC 27710-3808. Received July 8, 2005; revision requested September 12; revision received October 5; accepted November 4; final version accepted December 19. Address correspondence to J.R.P. (e-mail: jeffrey.petrella{at}duke.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Materials and Methods: The Institutional Review Board approved this HIPAA-compliant study, and each patient gave written informed consent. Thirty-nine consecutive patients (19 male, 20 female; mean age, 42.2 years) referred for functional MR imaging for possible tumor resection were prospectively evaluated. A preoperative diagnosis of brain tumor was made in all patients. Sentence completion and bilateral hand squeeze tasks were used to map language and sensory motor areas. Neurosurgeons completed questionnaires regarding the proposed treatment plan before and after functional MR imaging and after surgery. They also gave confidence ratings for functional MR imaging results and estimated the effect on surgical time, extent of resection, and surgical approach. The effect of functional MR imaging on changes in treatment plan was assessed with the Wilcoxon signed rank test. Differences in confidence ratings between altered and unaltered treatment plans were assessed with the Mann-Whitney U test. The estimated influence of functional MR imaging on surgical time, extent of resection, and surgical approach was denoted with summary statistics.
Results: Treatment plans before and after functional MR imaging differed in 19 patients (P < .05), with a more aggressive approach recommended after imaging in 18 patients. There were no significant differences in confidence ratings for functional MR imaging between altered and unaltered plans. Functional MR imaging resulted in reduced surgical time (estimated reduction, 15–60 minutes) in 22 patients who underwent surgery, a more aggressive resection in six, and a smaller craniotomy in two.
Conclusion: Functional MR imaging enables the selection of a more aggressive therapeutic approach than might otherwise be considered because of functional risk. In certain patients, surgical time may be shortened, the extent of resection increased, and craniotomy size decreased.
© RSNA, 2006
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
In addition to providing information regarding the feasibility of resection, functional MR imaging may provide additional information regarding surgical approach and the selection of patients for invasive functional mapping (1). The results of numerous studies have shown agreement between functional MR imaging localization of sensory motor function and localization by means of invasive neurosurgical methods (13–15). Functional MR imaging has been shown to have high sensitivity (81%–92%) for the mapping of language areas in comparison with intraoperative electrocortical stimulation (16) and generates highly reproducible results for global and regional language lateralization, particularly in patients with epilepsy (17).
Despite demonstration of the technical and diagnostic efficacy of functional MR imaging compared with more invasive reference standards, definitive evidence that functional MR imaging substantially affects therapeutic decision making and patient outcome has not been fully established. Consequently, functional MR imaging has not been fully integrated into the broader clinical practice of neuroradiology. Thus, the purpose of our study was to prospectively evaluate the effect of preoperative functional MR imaging localization of language and motor areas on therapeutic decision making in patients with potentially resectable brain tumors.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
Paradigm
All paradigms were controlled by using the CIGal software program (developed by J.T.V. [18]) operating on a Windows XP platform (Microsoft, Redmond, Wash). The software provided real-time video and audio stimulus presentation, behavioral response monitoring, and accurate synchronization with the MR imager. For language tasks, patients were shown sentences with blanks and were asked to silently fill in the blanks. For example, "The American flag is red, white and ." This was followed by the control condition for which patients were shown scrambled letters parsed into groups to simulate the appearance of text and were instructed to guide their eyes over the pseudo-text. The following is an example of the pseudo-text used for the control condition: "Mry Ghtpieuw ghyr mo ywu, qoiem chg ." The tasks and control conditions were matched for sentence length. Patients were also asked to perform a motor task that involved alternating hand squeezing. Both of these tasks (language and motor tasks) have been shown to induce reliable activation in language areas (16,19) and hand sensory motor areas (20,21). All patients were trained in verbal and motor tasks 10 minutes prior to examination. Stimuli were projected onto a translucent screen that was positioned alongside the MR imager. Stimuli were viewed by way of a mirror that was mounted on the head coil.
There were no failures due to technical difficulties or to an inability to perform tasks or obtain usable data. The time involved in performing an entire preoperative functional MR imaging session, which included complete patient training and the acquisition of structural images, was less than 1 hour, with approximately 30 minutes for actual imaging.
Image Analysis
Functional MR imaging data were analyzed by using the fScan image analysis program (developed by J.T.V. [18,22]), and t test activation maps were generated to reveal the degree to which variations in MR signal intensity corresponded to fluctuations in task stimulus and performance. These voxel-wise assays were created in real time, which enabled immediate identification of imaging problems and interactive determination of when enough data had been collected to produce adequate mapping results (J.T.V., 10 years of experience in functional MR imaging). Maps were convolved with a 10-mm smoothing kernel and were thresholded to a t value, which varied from 1.5 to 3.0 depending on the threshold for optimal localization of the activation foci. Statistical data were overlaid in color on a registered set of coplanar conventional dual-echo long-repetition-time spin-echo MR images.
Image Interpretation
Images were reviewed in consensus by two neuroradiologists (J.R.P. and L.M.S., with 10 years and 1 year of experience in functional MR imaging, respectively) who rendered a written interpretation. This interpretation consisted of an annotated set of T2-weighted MR images with thresholded activation maps superimposed as a color-coded overlay. The report included the identification of active areas, the presumed function that the active areas corresponded to, and the proximity of these areas to the tumor.
Effect of Assessment on Treatment Plan
The referring neurosurgeons were given questionnaires regarding their treatment plan before and after receiving the functional MR imaging interpretation. The preoperative questionnaire, which was given to neurosurgeons before functional MR imaging, included questions regarding the indication for functional MR imaging, the preoperative treatment plan, and additional diagnostic testing (eg, Wada testing) that would be performed if functional MR imaging were not available. Plan options, which were listed in order of increasingly aggressive therapy and surgical risk, included (a) no surgery, (b) biopsy, (c) craniotomy with intraoperative cortical mapping, performed with sedation, and (d) craniotomy without mapping, performed with a general anesthetic.
The functional MR imaging questionnaire, which was given to neurosurgeons after functional MR imaging, included questions regarding the preoperative plan after taking into account the functional MR imaging results. The neurosurgeons were asked to rate on a scale of 1–5 their confidence in the functional MR imaging results for the lateralization and 1 localization of expressive speech, receptive speech, and motor function (0, no confidence in results; 1, results suggestive of lateralization; 2, results could be used as a rough guide to surgical planning; 3, lateralization and localization are likely but need confirmation; 4, fairly confident in results; and 5, very confident in results).
A postoperative questionnaire was used in patients who underwent surgery to establish the type of surgical procedure that was actually used (plan options b–d) and to determine how (if at all) functional MR imaging affected surgical time, the extent of surgical resection, and surgical approach.
Statistical Analysis
A Wilcoxon signed rank test was used to assess any significant changes between treatment planning before and after functional MR imaging, equally weighting incremental changes in the plan ranked on a scale of 1–4 from least to most aggressive. Differences in confidence regarding the functional MR imaging result for lateralization of speech and localization of speech and motor function, between patients in whom the therapeutic plan was altered and those in whom it was not altered, were assessed by using the Mann-Whitney U test. The estimated influence of functional MR imaging on surgical time, extent of resection, and surgical approach was denoted with summary statistics. Statistical tests were performed by using a commercially available software program (SPSS, version 12.2; SPSS, Chicago, Ill), and a P value of .05 was considered to indicate a statistically significant difference. Trends in the estimated reduction in surgical time over the course of the study were assessed graphically in 3-month intervals.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
|
|
|
|
|
|
Surgical Time, Resection, and Approach
In addition to the surgical treatment plan, functional MR imaging also affected surgical time, the extent of surgical resection, and surgical approach, as determined by means of the postoperative questionnaire. By facilitating the localization of motor and language areas, functional MR imaging helped to shorten surgical time by an estimated 15–60 minutes in 22 (60%; 95% confidence interval: 43.6%, 75.3%) of 37 patients. There was an increasing trend in this regard for the first three quarters of the study (Fig 7). In six (16%; 95% confidence interval: 4.3%, 28.1%) of 37 patients who underwent surgery, the neurosurgeon specifically commented that the functional MR imaging results enabled a more complete resection than otherwise would have been achieved. Surgical approach was also affected, with a smaller craniotomy performed in two (5%; 95% confidence interval: 1.9%, 12.7%) of 37 patients.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
For patients in whom the treatment plan was not altered, functional MR imaging likely contributed confidence to the neurosurgeons' decision to proceed with the planned treatment approach. The neurosurgeons' confidence ratings for these patients were not significantly different from those for patients in whom the treatment plan was altered. Of note, there were two patients in whom the preoperative plan of no surgery was confirmed by means of functional MR imaging.
Studies in the literature corroborate the lack of postoperative neurologic deficits in patients who have undergone preoperative functional MR imaging (11,23). The influence of functional MR imaging on therapy has also been suggested in the literature (13). In the study by Wilkinson et al (11) it was shown that, when functional MR imaging data were incorporated into neuronavigation-guided surgical approaches, functional MR imaging facilitated safe tumor resection in anatomic areas that were previously thought to be high risk, without intraoperative cortical mapping, by means of conventional MR imaging. In a retrospective study, Lee et al (1) demonstrated that functional MR imaging was complementary to invasive functional mapping. The effect of functional MR imaging on therapeutic planning has been documented thus far in only one prospective study in which investigators demonstrated considerable influence on diagnostic and therapeutic decision making in patients with epilepsy (24).
Despite the increasing clinical use of functional MR imaging as a preoperative planning tool, there are several limiting factors that are intrinsic to the method itself. First, functional MR imaging is an indirect activation modality that shows some (but not necessarily all) of the brain areas that participate in the execution of a particular task. In particular, functional MR imaging cannot differentiate between activation in brain areas that are simply correlated with a particular function and activation in brain areas that are truly essential for performing that function. Second, the pathologic features of tissue can alter signal intensity on blood oxygen level–dependent functional MR images in a manner that has not been fully characterized (25,26). Third, accuracy in mapping the spatial extent of the activated voxels is limited and depends on the statistical threshold chosen. Individuals vary in their functional MR activation levels for a given task. Fitzgerald et al (16) described variable activation in individuals across a variety of task paradigms in many brain regions. Finally, functional MR imaging requires active participation by a cooperative patient, a condition that is not always feasible in the neuro-oncologic population.
There are multiple steps that can be taken to reduce the effect of these limitations. In our study, we used a standardized imaging and analysis protocol, consistent neuroradiologists, and a small group of neurosurgeons to minimize the variability in methods across the patient population. Variations in the pathologic characteristics of the tumors, the perceived tumor extent, and the surrounding edema may account for some of the variability in the neurosurgeons' confidence in the accuracy of the spatial extent of localization.
To minimize task performance variability across different patients, we used real-time functional MR imaging analysis and head-motion plots to allow the operator to assess the quality of the data at the time of imaging. The operator views functional MR imaging activation maps and motion plots in real time and can assess an active area and plot fluctuations of task-related blood oxygen level–dependent signal intensity over time. In this way, head motion or task performance problems can be quickly identified and corrected so that new activation maps can be generated. In rendering a clinical interpretation for functional MR imaging, the neuroradiologist uses the combination of activation maps, activation time course plots, and expected anatomic location of functional areas.
Given that the small number of neurosurgeons who referred patients already had some confidence in functional MR imaging, another limitation of our study is referral bias, and the patients who were referred were those for whom the localization of eloquent speech and sensorimotor cortex might affect treatment planning. This approach, however, has an advantage in that, because it is used in clinical practice, it provides a realistic assessment of therapeutic options and the role of diagnostic functional MR imaging. Because the referring neurosurgeon also completed the preoperative, functional MR imaging, and postoperative questionnaires, this prospective study focused on the treatment effect of functional MR imaging for patients in whom functional MR imaging might be ordered, without attempting to assess its role across the entire population of potential patients.
Our study has addressed the issue of therapeutic planning effect without attempting to assess the patient outcome effect of functional MR imaging (27). For the latter, one would need to perform a randomized trial with a control population in whom functional MR imaging information was requested but not provided. To do so, however, would have ethical implications.
Because of rapid technologic advancements in medicine—particularly in the specialty of radiology—it is often difficult to establish the actual benefits of new techniques with a traditional hierarchical scientific assessment, which can take a relatively long period of time compared with the rapid evolution of new technology. Thus, new technologies are often implemented on the basis of subjective expectations of usefulness and/or experience with a limited number of patients (28). In fact, it has been suggested that all that may be needed to decide to implement a new technology is to demonstrate that the new diagnostic strategy facilitates clinical decision making without detriment to patient care (28). The goal of our study therefore was to establish if and how this new modality (ie, functional MR imaging) would affect therapeutic planning in the routine clinical environment.
Because it can be fully implemented in a busy clinical environment, functional MR imaging has a broad role in neurosurgical planning, including patient selection, assessment of the feasibility of surgery, and surgical planning. In the appropriate setting, it can be easily added to conventional MR imaging because of the speed of the study. Functional MR imaging can take less than 10 minutes to complete, which includes the time needed to perform online statistical analysis (29). This immediate feedback allows the radiologist to decide whether a given task needs to be repeated to ensure that the appropriate functional area is localized. We were successful in our study in patients with varying clinical functional status. There were no failures due to technical difficulties or to an inability to perform tasks or obtain usable data. The time involved in performing an entire preoperative functional MR imaging session, complete with structural image acquisition, was less than 1 hour, with approximately 10 minutes prior to the study for instruction on the tasks and approximately 30 minutes for actual imaging.
Future studies focusing on the outcome of patients who undergo preoperative functional MR imaging are still needed, though such studies will be difficult to perform. A randomized controlled trial in a large number of patients will be needed to show an effect because of the indirect relationship between the diagnostic test and patient outcome. Other issues warranting investigation include disease-specific effects on the measured blood oxygen level–dependent response, effects of individual variations in cerebrovascular anatomy, and effects due to the presence of incidental pathologic abnormalities (eg, atherosclerotic disease).
In conclusion, functional MR imaging has a significant effect on therapeutic planning in patients with potentially resectable brain tumors and enables the selection of therapeutic options that might otherwise not have been considered because of functional risk. In certain patients, surgical time may be shortened, the extent of resection may be larger, and craniotomy size may be smaller.
|
ADVANCES IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, J.R.P.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, J.R.P., L.M.S., J.S.P., J.T.V.; clinical studies, all authors; statistical analysis, J.R.P., L.M.S., K.M.H., J.T.V.; and manuscript editing, J.R.P., L.M.S., A.H.F., T.M.G., J.S.P., J.T.V.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A M BLAMIRE The technology of MRI -- the next 10 years? Br. J. Radiol., August 1, 2008; 81(968): 601 - 617. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Bizzi, V. Blasi, A. Falini, P. Ferroli, M. Cadioli, U. Danesi, D. Aquino, C. Marras, D. Caldiroli, and G. Broggi Presurgical Functional MR Imaging of Language and Motor Functions: Validation with Intraoperative Electrocortical Mapping Radiology, August 1, 2008; 248(2): 579 - 589. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Smits, M.W. Vernooij, P.A. Wielopolski, A.J.P.E. Vincent, G.C. Houston, and A. van der Lugt Incorporating Functional MR Imaging into Diffusion Tensor Tractography in the Preoperative Assessment of the Corticospinal Tract in Patients with Brain Tumors AJNR Am. J. Neuroradiol., August 1, 2007; 28(7): 1354 - 1361. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
