Effect of functional MRI–guided navigation on surgical outcomes: a prospective controlled trial in patients with arteriovenous malformations

View More View Less
  • 1 Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University;
  • 2 China National Clinical Research Center for Neurological Diseases;
  • 3 Center of Stroke, Beijing Institute for Brain Disorders;
  • 4 Beijing Key Laboratory of Translational Medicine for Cerebrovascular Disease;
  • 5 Medical Imaging Center, The 306th Hospital of PLA, Beijing; and
  • 6 Department of Neurosurgery, The First Affiliated Hospital of Fujian Medical University, Fujian, People's Republic of China
Full access

OBJECTIVE

The impact of functional MRI (fMRI)–guided navigation on the surgical outcome of patients with arteriovenous malformations (AVMs) is undetermined. This large, randomized controlled trial (RCT) was designed to determine the safety and efficacy of fMRI-guided microsurgery of AVMs. This paper reports the preliminary results of the interim analysis.

METHODS

Between September 2012 and June 2015, eligible patients were randomized to the standard microsurgery group (control group) or the fMRI-guided surgery group (experimental group) in a 1:1 ratio. Patients in the control group underwent conventional digital subtraction angiography and MRI before surgery. The surgery was performed according to the standard procedure. However, patients in the experimental group underwent blood oxygen level–dependent (BOLD) fMRI and diffusion tensor imaging within 1 week before surgery. Moreover, preoperative eloquent brain tissue mapping and intraoperative fMRI navigation were performed in addition to the standard procedure. The preliminary end points were the total removal rate of AVMs and postoperative surgical complications. The primary end points were modified Rankin Scale (mRS) score (favorable: mRS Score 0–2; poor: mRS Score 3–6) and surgery-related permanent functional deficits (S-PFD) at the last clinic visit (≥ 6 months). Statistical analysis was performed using the statistical package from SPSS.

RESULTS

The interim analysis included 184 participants (93 in the experimental group and 91 in the control group). Patients were equally distributed between the 2 groups. Neither the preliminary nor the primary end points, including postoperative complications (p = 0.781), residual AVM (p = 1.000), last mRS score (p = 0.654), and S-PFD (p = 0.944) showed any significant difference between the control and experimental group. According to the results of the univariate analysis, eloquent adjacent brain tissue (OR 0.14; 95% CI 0.06–0.32; p < 0.001), large size of the nidus (OR 1.05; 95% CI 1.02–1.08; p = 0.002), or diffuse nidus (OR 3.05; 95% CI 1.42–6.58; p = 0.004) were all significantly associated with S-PFD. Additionally, a high Spetzler-Martin score (OR 3.54; 95% CI 2.08–6.02; p < 0.001), no previous hemorrhage (OR 2.35; 95% CI 1.00–5.54; p = 0.05), or a low preoperative mRS score (OR 0.42; 95% CI 0.17–1.00; p = 0.049) were also significantly associated with S-PFD. Multivariate analysis revealed that independent factors correlated with S-PFD were eloquent adjacent brain tissue (OR 0.17; 95% CI 0.04–0.70; p = 0.014) and low preoperative mRS score (OR 0.22; 95% CI 0.07–0.69; p = 0.009).

CONCLUSIONS

This preplanned interim analysis revealed no significant differences in the primary end points between the experimental and control group, prompting an early termination of this RCT. The preliminary data indicated that the additional intervention of fMRI navigation is not associated with a more favorable surgical outcome in patients with AVMs. The results indicated that eloquent adjacent brain tissue and a low preoperative mRS score are independent risk factors for S-PFD.

Clinical trial registration no.: NCT01758211 (clinicaltrials.gov)

ABBREVIATIONS AVM = arteriovenous malformation; BOLD = blood oxygen level dependent; DSA = digital subtraction angiography; DTI = diffusion tensor imaging; fMRI = functional MRI; FOV = field of view; mRS = modified Rankin Scale; PRC = People's Republic of China; RCT = randomized controlled trial; ROIs = regions of interest; SM = Spetzler-Martin; S-PFD = surgery-related permanent functional deficits; TOF-MRA = time-of-flight MR angiography.

OBJECTIVE

The impact of functional MRI (fMRI)–guided navigation on the surgical outcome of patients with arteriovenous malformations (AVMs) is undetermined. This large, randomized controlled trial (RCT) was designed to determine the safety and efficacy of fMRI-guided microsurgery of AVMs. This paper reports the preliminary results of the interim analysis.

METHODS

Between September 2012 and June 2015, eligible patients were randomized to the standard microsurgery group (control group) or the fMRI-guided surgery group (experimental group) in a 1:1 ratio. Patients in the control group underwent conventional digital subtraction angiography and MRI before surgery. The surgery was performed according to the standard procedure. However, patients in the experimental group underwent blood oxygen level–dependent (BOLD) fMRI and diffusion tensor imaging within 1 week before surgery. Moreover, preoperative eloquent brain tissue mapping and intraoperative fMRI navigation were performed in addition to the standard procedure. The preliminary end points were the total removal rate of AVMs and postoperative surgical complications. The primary end points were modified Rankin Scale (mRS) score (favorable: mRS Score 0–2; poor: mRS Score 3–6) and surgery-related permanent functional deficits (S-PFD) at the last clinic visit (≥ 6 months). Statistical analysis was performed using the statistical package from SPSS.

RESULTS

The interim analysis included 184 participants (93 in the experimental group and 91 in the control group). Patients were equally distributed between the 2 groups. Neither the preliminary nor the primary end points, including postoperative complications (p = 0.781), residual AVM (p = 1.000), last mRS score (p = 0.654), and S-PFD (p = 0.944) showed any significant difference between the control and experimental group. According to the results of the univariate analysis, eloquent adjacent brain tissue (OR 0.14; 95% CI 0.06–0.32; p < 0.001), large size of the nidus (OR 1.05; 95% CI 1.02–1.08; p = 0.002), or diffuse nidus (OR 3.05; 95% CI 1.42–6.58; p = 0.004) were all significantly associated with S-PFD. Additionally, a high Spetzler-Martin score (OR 3.54; 95% CI 2.08–6.02; p < 0.001), no previous hemorrhage (OR 2.35; 95% CI 1.00–5.54; p = 0.05), or a low preoperative mRS score (OR 0.42; 95% CI 0.17–1.00; p = 0.049) were also significantly associated with S-PFD. Multivariate analysis revealed that independent factors correlated with S-PFD were eloquent adjacent brain tissue (OR 0.17; 95% CI 0.04–0.70; p = 0.014) and low preoperative mRS score (OR 0.22; 95% CI 0.07–0.69; p = 0.009).

CONCLUSIONS

This preplanned interim analysis revealed no significant differences in the primary end points between the experimental and control group, prompting an early termination of this RCT. The preliminary data indicated that the additional intervention of fMRI navigation is not associated with a more favorable surgical outcome in patients with AVMs. The results indicated that eloquent adjacent brain tissue and a low preoperative mRS score are independent risk factors for S-PFD.

Clinical trial registration no.: NCT01758211 (clinicaltrials.gov)

ABBREVIATIONS AVM = arteriovenous malformation; BOLD = blood oxygen level dependent; DSA = digital subtraction angiography; DTI = diffusion tensor imaging; fMRI = functional MRI; FOV = field of view; mRS = modified Rankin Scale; PRC = People's Republic of China; RCT = randomized controlled trial; ROIs = regions of interest; SM = Spetzler-Martin; S-PFD = surgery-related permanent functional deficits; TOF-MRA = time-of-flight MR angiography.

The annual hemorrhage rate is 2.2% for unruptured arteriovenous malformations (AVMs) and 4.5% for ruptured AVMs.10 Although relatively rare, hemorrhage caused by AVMs is the leading cause of hemorrhagic stroke in young individuals.2 Therefore, when it is appropriate, early management can be more effective in eliminating the cumulative lifetime risk of intracranial hemorrhage.6 In the ARUBA (A Randomized trial of Unruptured Brain Arteriovenous Malformations) study, the data suggest that in patients with unruptured AVMs, medical management alone might be superior to medical management combined with interventional therapy in the prevention of stroke and/or death.21 However, the results remain controversial due to bias in case selection and the short follow-up period of the study. Currently, resection remains the optimal treatment, with a higher complete obliteration rate and lower rehemorrhage rate compared with embolization and stereotactic radiosurgery.29 The disadvantage of surgical treatment is the higher rate of permanent neurological defects or death in comparison with other methods.29

Preoperative identification and intraoperative preservation of the eloquent brain areas are the key points to improving the surgical outcome in surgically treated patients with AVMs. In the last 20 years, surgical navigation based on functional MRI (fMRI), including blood oxygen level–dependent (BOLD) fMRI and diffusion tensor imaging (DTI) has been one of the most important techniques and concepts in neurosurgery.1,2,25 The most frequently reported indication for fMRI-guided intracranial surgery is cerebral glioma resection.1,3,15,23,33 The use of fMRI guidance has been shown to help in providing data on the extent of resection, localization of eloquent brain areas, and on tumor remnants, thus improving surgical outcomes. However, the literature regarding the application of fMRI-guided microsurgery in patients with AVMs is limited to a few case reports and small retrospective series.2,8,12,13,17,22 Therefore, there is insufficient evidence in support of the positive impact of fMRI navigation on the outcome of AVM surgery.

First-level evidence for the clinical application of fMRI-guided microsurgery in patients with AVMs is still lacking. The cost-benefit analysis of such expensive and complex facilities is controversial. Is the additional intervention of fMRI navigation indeed associated with a higher rate of complete AVM resection and functional preservation? If yes, would this result in improving the quality of life after surgery? Finally, if it does improve surgical outcomes, then would it be beneficial to mandate the widespread application of fMRI-guided microsurgery in patients with AVMs? It was the aim of this study to address these questions in an attempt to provide strong support for the clinical application of fMRI-guided microsurgery in patients with AVMs. Therefore, we designed this large, randomized controlled trial (RCT) to determine the safety and efficacy of fMRI-guided microsurgery of AVMs. This paper reports the preliminary results of the interim analysis.

Methods

Study Population

This study was a prospective, assessor-blinded, RCT. It was performed at Tiantan Hospital (China National Clinical Research Center for Neurological Diseases), which is the largest neurosurgical center in China. The selected participants fulfilled all of the inclusion criteria and none of the exclusion criteria as established in the registered study protocol (www.clinicaltrials.gov; NCT01758211 [Functional Magnetic Resonance Imagine Navigation in Intracranial Arteriovenous Malformation Surgery {FMRINAVMS}]).34 The inclusion criteria were as follows: 1) patient age ranging from 12 to 60 years; 2) a definitive diagnosis of AVMs; 3) no contraindications to performing fMRI examination; and 4) informed consent. The exclusion criteria were as follows: 1) patients initially undergoing endovascular and radiosurgery; 2) patients undergoing surgical treatment within 1 month of initial hemorrhage; 3) patients undergoing emergency removal of intracranial hematomas due to AVMs; 4) contraindications for general anesthesia and surgery; and 5) pregnancy and breastfeeding. This study adhered to good clinical practice and the ethical principles of the Declaration of Helsinki, and was approved by the Institutional Review Board of Beijing Tiantan Hospital, Capital Medical University (register no.: ky2012–016–02). Written informed consent was obtained from participants or their family members. For patients younger than 18 years, informed consent was obtained from their guardians. The patients could withdraw from the study at any time.

Randomization and Blinding

Eligible patients were randomly assigned to the standard microsurgery group (control group) or the fMRI-guided surgery group (experimental group) in a 1:1 ratio. We used stratified blocked randomization and minimization methods. Stratification factors included the sex and age of the patients, eloquence of adjacent brain areas, and the Spetzler-Martin (SM) grade of the lesions. Only the outcome assessors were blinded because it was impossible to blind the participants and the surgeons. Moreover, assessors were not involved in patient management. The allocation sequence was generated by an independent third party of the National Center for Cardiovascular Diseases of the People's Republic of China (PRC). Furthermore, to prevent bias, the third party was also responsible for patient randomization.

Acquisition and Processing of the fMRI Data Set

All patients in the fMRI-guided group underwent sagittal T1-weighted anatomy imaging, BOLD-fMRI, DTI, and time-of-flight MR angiography (TOF-MRA) within 1 week before their surgery. We used a 3-T MR system (SIEMENS Trio) for the fMRI studies. The sagittal T1-weighted anatomy images were acquired using a gradient-echo sequence and the following parameters: TR 2300 msec, TE 2.98 msec, slice thickness 1 mm, 176 slices, field of view (FOV) 256 mm, flip angle 9°, matrix 64 × 64, voxel size 1 × 1 × 1 mm3, and bandwidth 240. The BOLD-fMRI was collected using standard echo planar imaging and the following settings: TR 3000 msec, TE 30 msec, matrix 64 × 64, and 30 axial slices (isotropic resolution 3 mm) that included all cerebral areas. Based on the AVM location, different BOLD-fMRI tasks (language, motor, and visual) were used to generate functional maps of the patient's brain. The task and the area activated are as follows: 1) verb generation and picture naming for language area stimulation, 2) finger tap movement for motor area, and 3) the black-and-white checkerboard for visual area stimulation. The DTI was acquired using the diffusion-weighted echo planar imaging technique with the following settings: TR 6100 msec, TE 93 msec, slice thickness 3 mm, 45 slices, FOV 230 × 230 mm2, matrix 128 × 128, and a motion-probing gradient in 30 orientations. Axial TOF-MRA was acquired using a 3D TOF gradient-echo acquisition sequence and the following parameters: TR 22 msec, TE 3.86 msec, slice thickness 1 mm, 36 × 4 slices, FOV 220 × 220 mm2, flip angle 120°, and 512 × 512 matrix.

The generated image sets were processed on the iPlan 3.0 workstation (Brainlab). All image sets were automatically coregistered with each other and then fused to the anatomical images using an automatic rigid registration. A significance threshold of p < 0.001 was established for the identification of activated clusters. The anatomical locations of the activation for each paradigm were documented by 2 neuroradiologists (B.Z., X.Z.T.) using consensus; they were not involved in patient management. Three major fiber tracts were selected for evaluation: 1) corticospinal tract; 2) arcuate fasciculus; and 3) optic radiation. We selected the regions of interest (ROIs) based on anatomical knowledge and previous DTI studies. To track the corticospinal tract, we used two ROIs delineated in the precentral gyrus (seed) and pons (target).4 To track the arcuate fasciculus, we used a large half-moon-shaped region defined on the most dorsal part of the arcuate, usually 1 or 2 slices above the body of the corpus callosum. According to traditional definitions of the arcuate fasciculus, fiber tracts not connecting the frontal lobe and temporal lobe are not considered to be part of the arcuate fasciculus—therefore we removed them by using the “NOT” operation.5 To reconstruct optic radiation tractography, the first ROI was placed on the lateral geniculate body (seed), and the second ROI was placed on the occipital lobe, including both sides of the calcarine sulcus.31 We selected a default fractional anisotropy threshold of 0.20 and a minimum fiber length of 70 mm. The 2 neuroradiologists, with consensus, also documented the locations of the ROIs and the tracked fibers. The processed data sets were then incorporated into the neuronavigation platform for intraoperative navigation.

Interventions According to Group

Control Group

Patients in the control group underwent conventional digital subtraction angiography (DSA) and MRI before surgery. Operations were performed by experienced vascular neurosurgeons (S.W. and Y.C.). Standard microsurgical treatment included temporary clipping of the feeding arteries, careful dissection of the AVM from the adjacent brain tissue, and complete resection of the AVM. Motor evoked potential and somatosensory evoked potential techniques were available for intraoperative monitoring. In addition, intraoperative ultrasound and indocyanine green videoangiography were also available to guide the resection of the AVMs and to detect residual AVMs as needed during surgery. After the operation, patients were maintained in a hypotensive state and treated with mannitol, barbiturates, and dexamethasone therapy. If any neurological symptoms were observed, then a postoperative CT scan was immediately performed to rule out intracranial hemorrhage. Five days later, a second DSA was performed to confirm the complete obliteration.

Experimental Group

For the patients in the experimental group, a preoperative plan was created using the fMRI data, including BOLD-fMRI and DTI. The operations were then performed by the same neurosurgeons (S.W. and Y.C.) who conducted the surgical procedures in the control group. Craniotomy was performed according to the preoperative plan and neuronavigation based on the fMRI scans (Brainlab). After the cortex was exposed, the surgeons used a navigator's probe to identify the feeding arteries, nidus, and draining veins of the AVM as well as the eloquent brain areas (if involved). During each AVM resection, the neuronavigation system was instrumental in ensuring the preservation of the functional cortex and white matter fiber tracts. All the skills, techniques, and the postoperative care protocol performed in the standard surgery group were also used in the fMRI-guided group.

Data Collection

Data were prospectively collected using an electronic case report form through a study website that required a login and password. During the clinical trial, the security of the data was monitored by the National Center for Cardiovascular Diseases of the PRC. Patient demographic data such as age, sex, and chief complaint were collected. For the angioarchitectural features of each AVM (such as size, diffuseness, and deep draining veins), the eloquence of adjacent brain areas and the SM grade were determined from the preoperative angiograms and traditional MRI scans. The preliminary end points were the total removal rate of AVMs and postoperative surgical complications. The surgical complications in this study included intracranial infection, rehemorrhage, and cerebral infarction. We measured the modified Rankin Scale (mRS) score and the functional deficit of the patients at pretreatment, posttreatment, discharge, and at follow-ups (1, 3, and 6 months). An mRS score of 0–2 was classified as a favorable surgical outcome, whereas an mRS score of 3–6 was a poor outcome. The primary end points were the mRS scores and surgery-related permanent functional deficits (S-PFD) at the last clinical visit (at least 6 months after surgery).

Statistical Analysis

To detect a difference of 10% between the study arms for the primary end points in the full analysis set, we estimated that a sample size of 600 participants was needed. This estimate is based on a given of 90% complete power, with an experiment-wise Type I error of 0.05.34 After inclusion of one-third of the intended sample size (originally 200 subjects), an interim analysis of the primary end point data was conducted to verify the original estimates, ensure the quality of the clinical trial, and to allow for premature study termination. Data were reported as the mean and SD for continuous variables or as frequency for categorical variables. The differences in clinical variables and outcomes between the 2 groups were analyzed using the t-test, chi-square test, Fisher's exact test, or the Mann-Whitney U-test. Associations of variables were identified using a univariate and multivariate analysis. The results were reported as the odds ratio, 95% confidence interval, and p value. All statistical analyses were performed at the National Center for Cardiovascular Diseases of the PRC, using SPSS software (version 20.0.0, IBM Corp.).

Results

The results comprise 5 aspects: 1) the balance of the prognostic and predictive factors between the 2 groups; 2) the statistical differences of the preliminary end points between the 2 groups; 3) the statistical differences in primary end points between the 2 groups; 4) the risk factors for S-PFD in all patients; and 5) the differences in S-PFD between the control group and the experimental group in subgroups of patients stratified by the risk factors for S-PFD. The last aspect was necessary in determining the effect of fMRI navigation on surgical outcomes in particular patients.

During the study period, all of the patients admitted were enrolled in this clinical trial, with the exception of 33 patients who needed emergency surgery, 13 patients who had been previously treated (surgery, embolization, or radiosurgery), 20 patients older than 60 years or younger than 12 years, and 1 patient with concurrent glioma. In total, 206 participants were enrolled in the study between February 2013 and August 2015. One hundred four participants were randomly assigned to the experimental group, and 102 participants were assigned to the control group. Three patients in the experimental group and 5 in the control group were excluded from this study because they refused to accept the scheduled surgical treatment after randomization. Two patients in the experimental group were excluded because of navigation failure, and 1 in the control group was excluded for anesthesia intolerance. Another 11 participants (5 patients in the control group and 6 in the experimental group) were excluded from the interim analysis because they had < 6 months of follow-up visits. Thus, 184 participants (experimental group 93 and control group 91) were included in the interim analysis. No patients were lost to follow-up. Figure 1 delineates the trial profile.

FIG. 1.
FIG. 1.

Flow chart depicting the trial profile. M = months.

Balance of the Factors Between the Groups

Patients were equally distributed between the 2 groups in regard to the following factors: sex (p = 0.130), age (p = 0.705), lesioned hemisphere (p = 0.888), eloquence of the adjacent brain tissue (p = 0.786), nidus size (p = 0.824), nidus diffuseness (p = 0.458), SM grade (p = 0.822), nidus deep venous drainage (p = 0.207), previous hemorrhage (p = 0.788), preoperative mRS score (p = 0.787), preoperative functional deficits (p = 0.225), and follow-up duration (p = 0.118) (Table 1). The mean SM grade was 2.3 ± 0.86 and 2.3 ± 0.91 in the study and control group, respectively (Fig. 2). Each group had 1 patient with a poor mRS score before surgery (Fig. 3).

TABLE 1.

The balance of the prognostic and predictive factors between the groups of patients with AVMs

FactorControl Group, No. (%)Exp Group, No. (%)p Value
No. of pts91 (49.5)93 (50.5)NA
  Sex0.130
    Male56 (61.5)67 (72.0)
    Female35 (38.5)26 (28.0)
  Age in yrs, ± SD28.4 ± 12.329.1 ± 11.60.705
  Side0.888
    Lt47 (51.6)49 (52.7)
    Rt44 (48.4)44 (47.3)
  Eloquence*0.786
    Eloquent34 (37.4)38 (40.9)
    Noneloquent57 (62.6)55 (59.1)
  DV drainage0.207
    Yes12 (13.2)7 (7.5)
    No79 (86.8)86 (92.5)
  Nidus size in mm, ± SD38.8 ± 13.238.4 ± 84.40.824
  Diffuseness of nidus0.458
    Diffuse29 (31.9)25 (26.9)
    Compact62 (68.1)68 (73.1)
  SM score0.822
    116 (17.6)17 (18.3)
    240 (44.0)38 (40.9)
    329 (31.9)31 (33.3)
    45 (5.5)5 (5.4)
    51 (1.1)2 (2.2)
  Hemorrhage Hx0.788
    Yes36 (39.6)35 (37.6)
    No55 (60.4)58 (62.4)
  Preop mRS score, ± SD1.1 ± 0.61.1 ± 0.40.787
  Preop FD0.225
    Yes13 (14.3)8 (8.6)
    No78 (85.7)85 (91.4)
  Follow-up in mos, ± SD17.7 ± 7.819.5 ± 7.70.118

DV = deep venous; Exp = experimental; FD = functional deficit; Hx = history; NA = not applicable; pts = patients.

Denotes the eloquence of adjacent brain tissue.

FIG. 2.
FIG. 2.

Bar graph showing the SM grading scores of the AVMs in patients in the control group and in the experimental group. The SM scores were equally distributed between the 2 groups. Figure is available in color online only.

FIG. 3.
FIG. 3.

Bar graph showing the preoperative mRS score of the patients in the control group and in the experimental group. Each group had 1 patient with a poor mRS score before surgery. Figure is available in color online only.

Preliminary End Points

In the experimental group, 11 of 93 patients (11.8%) suffered from surgical complications, including 3 patients with postoperative rehemorrhage, 3 patients with brain infarction, and 7 patients with intracranial infection. In the control group, 12 patients of 91 (13.2%) experienced surgical complications, including 4 patients with brain infarction and 8 patients who developed intracranial infections. There was no significant difference in surgical complications between the 2 groups (p = 0.781) (Table 2). The postoperative DSA revealed that 1 patient (1.1%) in the control group and 2 patients (2.2%) in the experimental group had residual AVM. There was no significant difference between the 2 groups (p = 1.000).

TABLE 2.

Differences in preliminary end points between control and experimental groups of patients with AVMs

VariableControl Group, No. (%)Exp Group, No. (%)p Value
No. of pts91 (100.0)93 (100.0)NA
Complications12 (13.2)11 (11.8)0.781
  Rehemorrhage0 (0.0)3 (3.2)0.252
  Infarction4 (4.4)3 (3.2)0.977
  Infection8 (8.8)7 (7.5)0.754
AVM residual1 (1.1)2 (2.2)1.000

Primary End Points

In the control group, 3 patients (3.3%) had poor surgical outcomes (mRS Score 3–6), whereas 88 patients (96.7%) had favorable surgical outcomes (mRS Score 0–2) (Table 3 and Fig. 4). Similarly, in the experimental group, 5 patients (5.4%) had poor outcomes and 88 patients (94.6%) had favorable outcomes. The difference in mRS scores at the last clinical visit was not statistically significant between the 2 groups (p = 0.654). Both the control (18.7%) and experimental group (18.3%) had 17 patients with S-PFD at the last clinical visit. Furthermore, there was no statistically significant difference in S-PFD at the last clinical visit between the 2 groups (p = 0.944).

TABLE 3.

Differences in preoperative mRS score, last mRS score, and S-PFD between control and experimental groups of patients with AVMs

FactorControl Group, No. (%)Exp Group, No. (%)p Value
Preop mRS score0.651
  09 (9.9)5 (5.4)
  170 (76.9)77 (82.8)
  211 (12.1)10 (10.8)
  30 (0.0)1 (1.1)
  41 (1.1)0 (0.0)
  50 (0.0)0 (0.0)
  60 (0.0)0 (0.0)
Last mRS score*0.654
  048 (52.7)51 (54.8)
  115 (16.5)19 (20.4)
  225 (27.5)18 (19.4)
  32 (2.2)4 (4.3)
  41 (1.1)0 (0.0)
  50 (0.0)0 (0.0)
  60 (0.0)1 (1.1)
S-PFD0.944
  Yes17 (18.7)17 (18.3)
  No74 (81.3)76 (81.7)

The mRS score at the last clinical visit.

FIG. 4.
FIG. 4.

Bar graph showing the mRS score of the patients at their last clinical visit (at least 6 months after surgery). Three patients (3.3%) had poor surgical outcomes (mRS Score 3–6) in the control group and 5 patients (5.4%) had poor outcomes in the experimental group. Figure is available in color online only.

According to the univariate analysis, eloquent adjacent brain tissue (OR 0.14, 95% CI 0.06–0.32; p < 0.001), large nidus size (OR 1.05, 95% CI 1.02–1.08; p = 0.002), and diffuse nidus (OR 3.05, 95% CI 1.42–6.58; p = 0.004) were all significantly associated with S-PFD (Table 4). In addition, a high SM score (OR 3.54, 95% CI 2.08–6.02; p < 0.001), no previous hemorrhage (OR 2.35, 95% CI 1.00–5.54; p = 0.05), and a low preoperative mRS score (OR 0.42, 95% CI 0.17–1.00; p = 0.049) were also all significantly associated with S-PFD. However, the use of fMRI-based navigation had no significant influence on findings of S-PFD (p = 0.944). Multivariate analysis revealed that independent factors correlated with S-PFD were eloquent adjacent brain tissue (OR 0.17, 95% CI 0.04–0.70; p = 0.014) and low preoperative mRS score (OR 0.22, 95% CI 0.07–0.69; p = 0.009) (Table 5).

TABLE 4.

Univariate logistic regression analysis to test the association of each predictor with the S-PFD at the last clinical visit

VariableOR95% CIp Value*
Navigation1.030.49–2.160.944
Sex0.560.26–1.200.136
Age1.000.97–1.030.902
Side0.580.27–1.240.158
Eloquence0.140.06–0.32<0.001
DV drainage0.440.16–1.270.128
Size1.051.02–1.080.002
Diffuseness3.051.42–6.580.004
SM score3.542.08–6.02<0.001
Hemorrhage2.351.00–5.540.050
Preop mRS score0.420.17–1.000.049
Preop FD0.000.00–∞0.998
Complication0.460.17–1.230.121
Residual AVM0.450.04–5.070.515
Follow-up0.990.94–1.040.643

Boldface type indicates statistical significance.

Denotes fMRI-guided navigation.

TABLE 5.

Multivariate logistic regression analysis to test the association of significant predictors (derived from univariate logistic regression analysis) with the S-PFD at the last clinical visit

VariableOR95% CIp Value*
Navigation0.980.40–2.380.965
Eloquence0.170.04–0.700.014
Size1.010.96–1.060.683
Diffuseness1.840.66–5.120.246
SM score1.670.63–4.440.304
Hemorrhage1.670.62–4.550.313
Preop mRS score0.220.07–0.690.009

Boldface type indicates statistical significance.

According to the risk factors for S-PFD derived from the multivariate analysis, patients were stratified into subgroups (Table 6). The statistical analysis was performed in the subgroups to determine the effect of fMRI navigation on function protection in particular groups of patients. However, there was no significant association between fMRI navigation and the incidence of S-PFD in patients with eloquent-area AVMs (p = 0.891) or a low preoperative mRS score (0–1) (p = 0.753).

TABLE 6.

The differences in S-PFD between the control and experimental groups in subgroups of patients with AVMs*

SubgroupNo. of Pts w/S-PFD (%)No. of Pts w/o S-PFD (%)p Value
TotalControl GroupExp GroupTotalControl GroupExp Group
Eloquence
  Eloquent2612/34 (35.3)14/38 (36.8)4622/34 (64.7)24/38 (63.2)0.891
  Noneloquent85/57 (8.8)3/55 (5.5)10452/57 (91.2)52/55 (94.5)0.753
Preop mRS score
  0–13317/79 (21.5)16/82 (19.5)12862/79 (78.5)66/82 (80.5)0.753
  2–610/12 (0.0)1/11 (9.1)2212/12 (100.0)10/11 (90.9)0.965

Numbers in parentheses represent the percentages of the patients with or without S-PFD in control and experimental groups from each subgroup.

Discussion

In this prospective RCT, the baseline characteristics were equally represented between the control and experimental groups, thus validating the effectiveness of the dynamic allocation algorithm for randomization. Although the experimental group patients underwent an fMRI scan within 1 week before surgery, the decision to perform surgery was not influenced by the findings of the preoperative fMRI. In our hospital, selection of a treatment method for cerebral AVMs is determined by several factors: the clinical presentation, the patient's condition, the patient's preference, and the potential risks associated with the surgery. Surgery is recommended only when patients exhibit the following symptoms: a serious headache, intractable seizure, progressive neurological deficit, and/or previous hemorrhage. However, if the symptoms are managed by medication or the lesion is graded 5–6 on the SM grading system,27 then a conservative treatment option or radiosurgery is recommended. In this study, a group of independent neurosurgery residents evaluated the surgical outcomes. Furthermore, the outcome assessors received training before the RCT study commenced, were blinded to the patient allocation, and were not involved in the surgical treatment. Consequently, we precluded the heterogeneities in known prognostic factors that could have influenced or biased the evaluation of the surgical outcomes.

Our preplanned interim analysis revealed that there were no significant differences in the primary end points between the experimental group and the control group, resulting in the early termination of this RCT. Initially, we had estimated that favorable outcomes (mRS score of 0–2) would account for approximately 80% in control group patients, and approximately 90% in the experimental group. The difference in the primary end point (last mRS score) between the 2 groups would be 10%. Thus, we had determined that a sample size of 300 patients in each group, with a 2-tailed significance level of 5%, a complete power of 80%, and a dropout rate of 10% were necessary to detect the significant differences between the 2 groups. However, according to the results of the interim analysis, the actual favorable outcome rates were higher than we had estimated; 94.6% in the experimental group and 96.7% in the control group. Furthermore, the difference in the favorable outcome rate between the 2 groups was not significant (p = 0.654), calculated as 2.1%, which is much smaller than the estimated 10%. Under the same significance level (5%), complete power (80%), and dropout rate (10%), the reestimated sample size would need to expand to 1643 cases in each group to achieve a significant difference. The reestimated sample size is too large to successfully recruit and complete. Given the same favorable outcome rates in future enrolled patients in the 2 groups, obtaining a significant difference between the groups with the previously estimated sample size (300 per group), reduces the complete power to only 24% (Fig. 5). Thus, it was necessary to terminate this RCT early.

FIG. 5.
FIG. 5.

Graph showing estimation of the complete power. Given the same favorable outcome rates in future enrolled patients of the control group (P1) and the experimental group (P2), obtaining a significant difference between the 2 groups with the previously estimated sample size (N1, N2 denote the number in control and experimental group, respectively; 300 patients each), reduces the complete power to only 24%. A = 0.05; a 2-tailed significance level of 5%. Zp test = Z test with pooled variance. Figure is available in color online only.

The preliminary and primary end points reveal that the addition of fMRI navigation is not associated with a more favorable surgical outcome in an unselected patient population. Neither the preliminary nor the primary end points, including postoperative complications (p = 0.781), residual AVM (p = 1.000), last mRS score (p = 0.654), and S-PFD (p = 0.944) show any significant difference between the control and the experimental group. In the literature, it has been demonstrated that neuronavigation based on preoperative structural and functional image data sets contributes to the maximal safe resection of cerebral gliomas and thus increases the survival rate for patients with glioma.32 However, the literature regarding the application of fMRI-guided microsurgery in patients with AVMs is limited to a few case reports and retrospective small series.2,8,12,13,17,22 Therefore, the impact of fMRI navigation on the outcomes of AVM surgery is indeterminate. According to our results, the positive impact of fMRI navigation on surgical outcomes is undetected in patients with AVMs, including the complete resection rate and intraoperative preservation of neurological function. With the numerous variables driving AVM surgical outcomes, it is not surprising that fMRI navigation would be of marginal value in an unselected patient population.

There are several reasons that may account for the mixed effect of fMRI navigation on the surgical outcomes in patients with AVMs. First, there are disadvantages to using fMRI-based navigation in these patients, as follows. 1) The presence of high-flow, hypervascularized, and large vessels in cortical AVMs can disrupt the BOLD/DTI signal and thus reduce the accuracy of the functional mapping. 2) The silent loss of neuronavigation accuracy and brain shift can compromise the accuracy of neuronavigation based on preoperative images.23,26,28,30 Thus, at the late stage of the surgery, intraoperative ultrasound navigation19,24 and intraoperative indocyanine green angiography11 seem to be more reliable than fMRI navigation to identify the extent of resection and the position of the residual AVMs. 3) Navigation based on fMRI can also prolong the duration of the operation and the duration of anesthesia, which may increase anesthetic risk and the possibility of infection. Second, AVM operations are special in that palliative resection increases the risk of intra- and postoperative hemorrhage, and complete resection has always been required for surgical management of intracranial AVMs.20 However, gliomas could undergo partial resection to avoid postoperative neurological deficits.22 During surgery, even though fMRI navigation and intraoperative electrical stimulation may indicate that we are nearing eloquent brain tissue, we cannot immediately stop the surgery because diffuse and ectatic blood vessels complicate hemostasis, which poses a high risk of causing postoperative hemorrhage. Third, there are limitations inherent to the trial design that may also account for the results. In this study, it was impossible to blind the participants and the surgeons; only the outcome assessors were blinded. Thus, there existed the potential for underpowering and the placebo effect that could have influenced the results. For example, because the surgeons were not blinded to patient groups, this may have caused treatment bias. The reliance of the trial arm on fMRI navigation (the do-no-harm principle) could result in a more cautious resection to protect adjacent eloquent brain tissue, whereas in the control group the patients could be prone to a more aggressive resection to prevent postoperative AVM remnants and have positive results quickly. Additionally, because the patients were not treated in a blinded fashion, there may be a placebo effect. Patients in the experimental group may be prone to report a more favorable result. These factors taken together could result in a mixed effect for fMRI navigation and a limited effect in an unselected patient population.

In this patient cohort, eloquent adjacent brain tissue, large nidus size, high SM score, diffuse nidus, no previous hemorrhage, and a low preoperative mRS score were associated with S-PFD. The S-PFD findings represent the long-term functional deficits caused by surgery. Thus, determining the risk factors for S-PFD is crucial to identifying the best candidates for surgical treatment. According to the results of univariate analysis, 2 of 3 variables in the SM grading system, including nidus size (p = 0.002), eloquence of adjacent brain tissue (p < 0.001),27 and the SM grading score itself (p < 0.001), were significantly associated with the incidence of S-PFD. In an attempt to stratify the prediction of surgical risk, several grading systems have been proposed. The most popular of these grading systems for brain AVMs is the SM grading system.9 The results of this surgical series validate the ability of the SM score to predict functional deficits caused by surgery.

Diffuseness (p = 0.004) was also correlated with S-PFD. In the literature, Lawton and colleagues added diffuseness of the nidus in their supplementary grading scale to the SM score for selecting patients with AVMs for surgery.14,16 Diffuse AVMs with ragged borders and intermixed brain force the neurosurgeon to establish dissection planes that can extend into normal brain, whereas compact AVMs with tightly woven arteries and veins often have distinct borders that separate cleanly from the adjacent brain.7 In addition, no previous hemorrhage (p = 0.050) and a low preoperative mRS score (0–1) (p = 0.049) were the other 2 predictive factors for S-PFD in this study. Patients with a high preoperative mRS score often suffered from preoperative functional deficits attributed to a previous hemorrhage or ischemia. Previous hemorrhage may facilitate surgery. Hematomas help separate AVMs from adjacent brain; evacuation of the hematoma creates a working space around the AVM that can minimize violation of normal brain or provides access to a deep nidus that might otherwise have been unreachable.16

In the multivariate analysis, eloquent adjacent brain tissue (p = 0.014) and low preoperative mRS score (p = 0.009) were independent risk factors for S-PFD (Table 5). Because there was no protective effect associated with fMRI navigation, a statistical analysis was performed to determine the effect of fMRI navigation on functional protection in subgroups of patients with or without one of the particular risk factors (Table 6). However, there was still no significant association between fMRI navigation and the incidence of S-PFD in the patient subgroups, such as patients with eloquent-area AVMs (p = 0.891) or with low preoperative mRS score (0–1) (p = 0.753). In our previous studies,18 we have determined that preoperative functional findings derived from BOLD-fMRI and DTI are predictive of surgical outcomes in patients with AVMs. The use of DTI tractography allows assessment of important white matter tracts, and BOLD-fMRI enables functional mapping of at-risk cortex in the vicinity of an AVM.2,25 Given that the eloquent adjacent brain area of the nidus is one of the most important risk factors for S-PFD, pretreatment knowledge of the eloquent areas (cortex and fiber tracts) is particularly important for AVM surgery. Considering the negligible impact of navigation-based fMRI on surgical outcomes and the importance of fMRI in preoperative planning in patients with AVMs, we propose that preoperative fMRI images are more suitable for planning rather than guiding the resection of the AVMs.

Limitations of the Study

This is a report of a single-center prospective RCT of fMRI use in the microsurgical resection of brain AVMs. Patients with previous and/or emergency AVM surgery, radiosurgery, and embolization were screened. Additionally, because only the surgical outcome assessors were blinded, it is difficult to avoid selection bias and information bias. Also, the prematurely terminated study may have insufficient power to detect differences in subgroup patients with AVMs. Now that this RCT of an unselected patient population has ended, we will continue to enroll the specific patients with risk factors to detect differences in subgroups of patients with AVMs.

Conclusions

Additional intervention of fMRI navigation is not associated with a more favorable surgical outcome in patients with AVMs. Eloquent adjacent brain tissue and low preoperative mRS scores are independent risk factors for S-PFD.

Acknowledgments

This study was supported by a “National Science and Technology Support Plan” grant from the Ministry of Health of China (No. 2011BAI08B08, principal investigator Professor Shuo Wang). It was also supported by the “973 National Key Basic Research Development Plan” grant from the Ministry of Science and Technology of China (No. 2012CB720704, principal investigator Professor Zhen Jin).

References

  • 1

    Abd-El-Barr MM, Saleh E, Huang RY, Golby AJ: Effect of disease and recovery on functional anatomy in brain tumor patients: insights from functional MRI and diffusion tensor imaging. Imaging Med 5:333346, 2013

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Bendok BR, El Tecle NE, El Ahmadieh TY, Koht A, Gallagher TA, Carroll TJ, : Advances and innovations in brain arteriovenous malformation surgery. Neurosurgery 74:Suppl 1 S60S73, 2014

    • Search Google Scholar
    • Export Citation
  • 3

    Castellano A, Bello L, Michelozzi C, Gallucci M, Fava E, Iadanza A, : Role of diffusion tensor magnetic resonance tractography in predicting the extent of resection in glioma surgery. Neuro Oncol 14:192202, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Catani M, Thiebaut de Schotten M: A diffusion tensor imaging tractography atlas for virtual in vivo dissections. Cortex 44:11051132, 2008

  • 5

    Chang EF, Gabriel RA, Potts MB, Berger MS, Lawton MT: Supratentorial cavernous malformations in eloquent and deep locations: surgical approaches and outcomes. Clinical article. J Neurosurg 114:814827, 2011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Davidson AS, Morgan MK: How safe is arteriovenous malformation surgery? A prospective, observational study of surgery as first-line treatment for brain arteriovenous malformations. Neurosurgery 66:498505, 2010

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Du R, Keyoung HM, Dowd CF, Young WL, Lawton MT: The effects of diffuseness and deep perforating artery supply on outcomes after microsurgical resection of brain arteriovenous malformations. Neurosurgery 60:638648, 2007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Ellis MJ, Rutka JT, Kulkarni AV, Dirks PB, Widjaja E: Corticospinal tract mapping in children with ruptured arteriovenous malformations using functionally guided diffusion-tensor imaging. J Neurosurg Pediatr 9:505510, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9

    Frerichs K, Steig P, Friedlander R, Classification and grading systems. Steig P, Batjer HH, Samson DS: Intracranial Arteriovenous Malformations New York, Informa Healthcare, 2007

    • Search Google Scholar
    • Export Citation
  • 10

    Gross BA, Du R: Natural history of cerebral arteriovenous malformations: a meta-analysis. J Neurosurg 118:437443, 2013

  • 11

    Holling M, Brokinkel B, Ewelt C, Fischer BR, Stummer W: Dynamic ICG fluorescence provides better intraoperative understanding of arteriovenous fistulae. Neurosurgery 73:1 Suppl Operative ons93ons99, 2013

    • Search Google Scholar
    • Export Citation
  • 12

    Itoh D, Aoki S, Maruyama K, Masutani Y, Mori H, Masumoto T, : Corticospinal tracts by diffusion tensor tractography in patients with arteriovenous malformations. J Comput Assist Tomogr 30:618623, 2006

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Kikuta K, Takagi Y, Nozaki K, Hashimoto N: Introduction to tractography-guided navigation: using 3-tesla magnetic resonance tractography in surgery for cerebral arteriovenous malformations. Acta Neurochir Suppl 103:1114, 2008

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14

    Kim H, Abla AA, Nelson J, McCulloch CE, Bervini D, Morgan MK, : Validation of the supplemented Spetzler-Martin grading system for brain arteriovenous malformations in a multicenter cohort of 1009 surgical patients. Neurosurgery 76:2533, 2015

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15

    Kumar A, Chandra PS, Sharma BS, Garg A, Rath GK, Bithal PK, : The role of neuronavigation-guided functional MRI and diffusion tensor tractography along with cortical stimulation in patients with eloquent cortex lesions. Br J Neurosurg 28:226233, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Lawton MT, Kim H, McCulloch CE, Mikhak B, Young WL: A supplementary grading scale for selecting patients with brain arteriovenous malformations for surgery. Neurosurgery 66:702713, 2010

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Lepski G, Honegger J, Liebsch M, Sória MG, Narischat P, Ramina KF, : Safe resection of arteriovenous malformations in eloquent motor areas aided by functional imaging and intraoperative monitoring. Neurosurgery 70:2 Suppl Operative 276289, 2012

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18

    Lin F, Zhao B, Wu J, Wang L, Jin Z, Cao Y, : Risk factors for worsened muscle strength after the surgical treatment of arteriovenous malformations of the eloquent motor area. J Neurosurg [epub ahead of print December 4, 2015. DOI: 10.3171/20156.JNS15969]

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    Mathiesen T, Peredo I, Edner G, Kihlström L, Svensson M, Ulfarsson E, : Neuronavigation for arteriovenous malformation surgery by intraoperative three-dimensional ultrasound angiography. Neurosurgery 60:4 Suppl 2 345351, 2007

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Miyamoto S, Hashimoto N, Nagata I, Nozaki K, Morimoto M, Taki W, : Posttreatment sequelae of palliatively treated cerebral arteriovenous malformations. Neurosurgery 46:589595, 2000

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Mohr JP, Parides MK, Stapf C, Moquete E, Moy CS, Overbey JR, : Medical management with or without interventional therapy for unruptured brain arteriovenous malformations (ARUBA): a multicentre, non-blinded, randomised trial. Lancet 383:614621, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Okada T, Miki Y, Kikuta K, Mikuni N, Urayama S, Fushimi Y, : Diffusion tensor fiber tractography for arteriovenous malformations: quantitative analyses to evaluate the corticospinal tract and optic radiation. AJNR Am J Neuroradiol 28:11071113, 2007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Orringer DA, Golby A, Jolesz F: Neuronavigation in the surgical management of brain tumors: current and future trends. Expert Rev Med Devices 9:491500, 2012

  • 24

    Peredo-Harvey I, Lilja A, Mathiesen T: Post-craniotomy neuronavigation based purely on intraoperative ultrasound imaging without preoperative neuronavigational planning. Neurosurg Rev 35:263268, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25

    Pillai JJ: The evolution of clinical functional imaging during the past 2 decades and its current impact on neurosurgical planning. AJNR Am J Neuroradiol 31:219225, 2010

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    Shamir RR, Joskowicz L, Spektor S, Shoshan Y: Localization and registration accuracy in image guided neurosurgery: a clinical study. Int J CARS 4:4552, 2009

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27

    Spetzler RF, Martin NA: A proposed grading system for arteriovenous malformations. J Neurosurg 65:476483, 1986

  • 28

    Stieglitz LH, Fichtner J, Andres R, Schucht P, Krähenbühl AK, Raabe A, : The silent loss of neuronavigation accuracy: a systematic retrospective analysis of factors influencing the mismatch of frameless stereotactic systems in cranial neurosurgery. Neurosurgery 72:796807, 2013

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    van Beijnum J, van der Worp HB, Buis DR, Al-Shahi Salman R, Kappelle LJ, Rinkel GJ, : Treatment of brain arteriovenous malformations: a systematic review and meta-analysis. JAMA 306:20112019, 2011

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30

    Wang MN, Song ZJ: Classification and analysis of the errors in neuronavigation. Neurosurgery 68:11311143, 2011

  • 31

    Wostrack M, Shiban E, Harmening K, Obermueller T, Ringel F, Ryang YM, : Surgical treatment of symptomatic cerebral cavernous malformations in eloquent brain regions. Acta Neurochir (Wien) 154:14191430, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32

    Wu JS, Gong X, Song YY, Zhuang DX, Yao CJ, Qiu TM, : 3.0-T intraoperative magnetic resonance imaging-guided resection in cerebral glioma surgery: interim analysis of a prospective, randomized, triple-blind, parallel-controlled trial. Neurosurgery 61:Suppl 1 145154, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33

    Wu JS, Zhou LF, Tang WJ, Mao Y, Hu J, Song YY, : Clinical evaluation and follow-up outcome of diffusion tensor imaging-based functional neuronavigation: a prospective, controlled study in patients with gliomas involving pyramidal tracts. Neurosurgery 61:935949, 2007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34

    Zhao B, Cao Y, Zhao Y, Wu J, Wang S: Functional MRI-guided microsurgery of intracranial arteriovenous malformations: study protocol for a randomised controlled trial. BMJ Open 4:e006618, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

Disclosures

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

Author Contributions

Conception and design: Wang, Lin, Wu, Jin, Cao. Acquisition of data: Lin, Jiao, Wu, Zhao, Tong, Cao. Analysis and interpretation of data: Wang, Lin, Jiao, Wu, Zhao, Tong, Cao. Drafting the article: Lin, Jiao, Cao. Critically revising the article: Wang, Lin, Jiao, Cao. Reviewed submitted version of manuscript: Wang, Lin, Jiao, Wu, Jin, Cao. Approved the final version of the manuscript on behalf of all authors: Wang. Statistical analysis: Lin, Jiao, Wu, Zhao, Jin, Cao. Administrative/technical/material support: Wang, Jin, Cao. Study supervision: Wang, Jin, Cao.

If the inline PDF is not rendering correctly, you can download the PDF file here.

Contributor Notes

INCLUDE WHEN CITING Published online July 1, 2016; DOI: 10.3171/2016.4.JNS1616.

Drs. Lin and Jiao contributed equally to this work.

Correspondence Shuo Wang, Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, No. 6 Tiantanxili, Dongcheng District, Beijing 100050, China. email: captainwang9858@vip.sina.com.
  • View in gallery

    Flow chart depicting the trial profile. M = months.

  • View in gallery

    Bar graph showing the SM grading scores of the AVMs in patients in the control group and in the experimental group. The SM scores were equally distributed between the 2 groups. Figure is available in color online only.

  • View in gallery

    Bar graph showing the preoperative mRS score of the patients in the control group and in the experimental group. Each group had 1 patient with a poor mRS score before surgery. Figure is available in color online only.

  • View in gallery

    Bar graph showing the mRS score of the patients at their last clinical visit (at least 6 months after surgery). Three patients (3.3%) had poor surgical outcomes (mRS Score 3–6) in the control group and 5 patients (5.4%) had poor outcomes in the experimental group. Figure is available in color online only.

  • View in gallery

    Graph showing estimation of the complete power. Given the same favorable outcome rates in future enrolled patients of the control group (P1) and the experimental group (P2), obtaining a significant difference between the 2 groups with the previously estimated sample size (N1, N2 denote the number in control and experimental group, respectively; 300 patients each), reduces the complete power to only 24%. A = 0.05; a 2-tailed significance level of 5%. Zp test = Z test with pooled variance. Figure is available in color online only.

  • 1

    Abd-El-Barr MM, Saleh E, Huang RY, Golby AJ: Effect of disease and recovery on functional anatomy in brain tumor patients: insights from functional MRI and diffusion tensor imaging. Imaging Med 5:333346, 2013

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Bendok BR, El Tecle NE, El Ahmadieh TY, Koht A, Gallagher TA, Carroll TJ, : Advances and innovations in brain arteriovenous malformation surgery. Neurosurgery 74:Suppl 1 S60S73, 2014

    • Search Google Scholar
    • Export Citation
  • 3

    Castellano A, Bello L, Michelozzi C, Gallucci M, Fava E, Iadanza A, : Role of diffusion tensor magnetic resonance tractography in predicting the extent of resection in glioma surgery. Neuro Oncol 14:192202, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Catani M, Thiebaut de Schotten M: A diffusion tensor imaging tractography atlas for virtual in vivo dissections. Cortex 44:11051132, 2008

  • 5

    Chang EF, Gabriel RA, Potts MB, Berger MS, Lawton MT: Supratentorial cavernous malformations in eloquent and deep locations: surgical approaches and outcomes. Clinical article. J Neurosurg 114:814827, 2011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Davidson AS, Morgan MK: How safe is arteriovenous malformation surgery? A prospective, observational study of surgery as first-line treatment for brain arteriovenous malformations. Neurosurgery 66:498505, 2010

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Du R, Keyoung HM, Dowd CF, Young WL, Lawton MT: The effects of diffuseness and deep perforating artery supply on outcomes after microsurgical resection of brain arteriovenous malformations. Neurosurgery 60:638648, 2007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Ellis MJ, Rutka JT, Kulkarni AV, Dirks PB, Widjaja E: Corticospinal tract mapping in children with ruptured arteriovenous malformations using functionally guided diffusion-tensor imaging. J Neurosurg Pediatr 9:505510, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9

    Frerichs K, Steig P, Friedlander R, Classification and grading systems. Steig P, Batjer HH, Samson DS: Intracranial Arteriovenous Malformations New York, Informa Healthcare, 2007

    • Search Google Scholar
    • Export Citation
  • 10

    Gross BA, Du R: Natural history of cerebral arteriovenous malformations: a meta-analysis. J Neurosurg 118:437443, 2013

  • 11

    Holling M, Brokinkel B, Ewelt C, Fischer BR, Stummer W: Dynamic ICG fluorescence provides better intraoperative understanding of arteriovenous fistulae. Neurosurgery 73:1 Suppl Operative ons93ons99, 2013

    • Search Google Scholar
    • Export Citation
  • 12

    Itoh D, Aoki S, Maruyama K, Masutani Y, Mori H, Masumoto T, : Corticospinal tracts by diffusion tensor tractography in patients with arteriovenous malformations. J Comput Assist Tomogr 30:618623, 2006

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Kikuta K, Takagi Y, Nozaki K, Hashimoto N: Introduction to tractography-guided navigation: using 3-tesla magnetic resonance tractography in surgery for cerebral arteriovenous malformations. Acta Neurochir Suppl 103:1114, 2008

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14

    Kim H, Abla AA, Nelson J, McCulloch CE, Bervini D, Morgan MK, : Validation of the supplemented Spetzler-Martin grading system for brain arteriovenous malformations in a multicenter cohort of 1009 surgical patients. Neurosurgery 76:2533, 2015

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15

    Kumar A, Chandra PS, Sharma BS, Garg A, Rath GK, Bithal PK, : The role of neuronavigation-guided functional MRI and diffusion tensor tractography along with cortical stimulation in patients with eloquent cortex lesions. Br J Neurosurg 28:226233, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Lawton MT, Kim H, McCulloch CE, Mikhak B, Young WL: A supplementary grading scale for selecting patients with brain arteriovenous malformations for surgery. Neurosurgery 66:702713, 2010

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Lepski G, Honegger J, Liebsch M, Sória MG, Narischat P, Ramina KF, : Safe resection of arteriovenous malformations in eloquent motor areas aided by functional imaging and intraoperative monitoring. Neurosurgery 70:2 Suppl Operative 276289, 2012

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18

    Lin F, Zhao B, Wu J, Wang L, Jin Z, Cao Y, : Risk factors for worsened muscle strength after the surgical treatment of arteriovenous malformations of the eloquent motor area. J Neurosurg [epub ahead of print December 4, 2015. DOI: 10.3171/20156.JNS15969]

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    Mathiesen T, Peredo I, Edner G, Kihlström L, Svensson M, Ulfarsson E, : Neuronavigation for arteriovenous malformation surgery by intraoperative three-dimensional ultrasound angiography. Neurosurgery 60:4 Suppl 2 345351, 2007

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Miyamoto S, Hashimoto N, Nagata I, Nozaki K, Morimoto M, Taki W, : Posttreatment sequelae of palliatively treated cerebral arteriovenous malformations. Neurosurgery 46:589595, 2000

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Mohr JP, Parides MK, Stapf C, Moquete E, Moy CS, Overbey JR, : Medical management with or without interventional therapy for unruptured brain arteriovenous malformations (ARUBA): a multicentre, non-blinded, randomised trial. Lancet 383:614621, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Okada T, Miki Y, Kikuta K, Mikuni N, Urayama S, Fushimi Y, : Diffusion tensor fiber tractography for arteriovenous malformations: quantitative analyses to evaluate the corticospinal tract and optic radiation. AJNR Am J Neuroradiol 28:11071113, 2007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Orringer DA, Golby A, Jolesz F: Neuronavigation in the surgical management of brain tumors: current and future trends. Expert Rev Med Devices 9:491500, 2012

  • 24

    Peredo-Harvey I, Lilja A, Mathiesen T: Post-craniotomy neuronavigation based purely on intraoperative ultrasound imaging without preoperative neuronavigational planning. Neurosurg Rev 35:263268, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25

    Pillai JJ: The evolution of clinical functional imaging during the past 2 decades and its current impact on neurosurgical planning. AJNR Am J Neuroradiol 31:219225, 2010

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    Shamir RR, Joskowicz L, Spektor S, Shoshan Y: Localization and registration accuracy in image guided neurosurgery: a clinical study. Int J CARS 4:4552, 2009

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27

    Spetzler RF, Martin NA: A proposed grading system for arteriovenous malformations. J Neurosurg 65:476483, 1986

  • 28

    Stieglitz LH, Fichtner J, Andres R, Schucht P, Krähenbühl AK, Raabe A, : The silent loss of neuronavigation accuracy: a systematic retrospective analysis of factors influencing the mismatch of frameless stereotactic systems in cranial neurosurgery. Neurosurgery 72:796807, 2013

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    van Beijnum J, van der Worp HB, Buis DR, Al-Shahi Salman R, Kappelle LJ, Rinkel GJ, : Treatment of brain arteriovenous malformations: a systematic review and meta-analysis. JAMA 306:20112019, 2011

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30

    Wang MN, Song ZJ: Classification and analysis of the errors in neuronavigation. Neurosurgery 68:11311143, 2011

  • 31

    Wostrack M, Shiban E, Harmening K, Obermueller T, Ringel F, Ryang YM, : Surgical treatment of symptomatic cerebral cavernous malformations in eloquent brain regions. Acta Neurochir (Wien) 154:14191430, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32

    Wu JS, Gong X, Song YY, Zhuang DX, Yao CJ, Qiu TM, : 3.0-T intraoperative magnetic resonance imaging-guided resection in cerebral glioma surgery: interim analysis of a prospective, randomized, triple-blind, parallel-controlled trial. Neurosurgery 61:Suppl 1 145154, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33

    Wu JS, Zhou LF, Tang WJ, Mao Y, Hu J, Song YY, : Clinical evaluation and follow-up outcome of diffusion tensor imaging-based functional neuronavigation: a prospective, controlled study in patients with gliomas involving pyramidal tracts. Neurosurgery 61:935949, 2007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34

    Zhao B, Cao Y, Zhao Y, Wu J, Wang S: Functional MRI-guided microsurgery of intracranial arteriovenous malformations: study protocol for a randomised controlled trial. BMJ Open 4:e006618, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

Metrics