Keywords
combined MRI - consistency - meningioma consistency - preoperative MRI - prediction
- sequences
Introduction
Meningiomas are one of the most common intracranial extra-axial tumors found in clinical
practice in most neurosurgical departments. Meningiomas vary in many aspects including
pathologic, imaging appearance, neurovascular involvements, tumor consistency, and
treatment outcome. Many factors including peritumoral arachnoid plane and brain invasion,[1] major neurovascular involvement,[2] and tumor consistency may affect the completeness of tumor removal and surgical
outcome. Tumor consistency is one of the critical factors affecting the completeness
of surgical removal and treatment outcome[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11] especially in the case of neurovascular involvement. Recognition of tumor consistency
constitutes valuable data to preoperative planning for safe surgical approaches and
improved outcomes.[5]
Water and collagen content of the tumor are important determinants of meningioma consistency.[7] Soft consistency tumors harbor more water and less collagen content for which T2-weighted
image (T2WI) on magnetic resonance imaging (MRI) sequences could be detected as hyperintensity.
High signal intensity on T2WI and fluid-attenuated inversion recovery (FLAIR) image
is due to long T2WI relaxation time of the material which is related to the amount
of water content and impurity of the water within the material. For tumor signal intensity
on the T2WI and FLAIR, the higher signal intensity correlates with greater amount
of tumor water content and vascularity which prior studies have shown to correlate
with the angioblastic, meningothelial, or syncytial histopathologic subtype and tended
to be softer tumors. Meningioma with high collagen content (fibroblastic subtype),
has less amount of water molecule within the tumor, is shown by lower T2WI signal,
and the tumor consistency tended to be firmer.[12]
[13]
[14] Diffusion-weighted imaging (DWI) represent the random Brownian movement of water
molecule within the object in which tumor with high cellularity or dense collagen
matrix (fibroblastic subtype) will restrict the movement of water molecule within
the tumor, causing high signal intensity on the DWI.
Many studies reported the benefits of preoperative MRI using different techniques
to establish the consistency of the tumor including both qualitative and quantitative
methods. Related studies have investigated the routine sequences of MRI intensity
including T1-weighted image (T1WI), T2WI, and FLAIR to predict meningioma consistency.[8] Among the quantification methods, many techniques were reported including using
magnetic resonance fingerprinting quantification of T1 and T2 relaxometry[3] and fractional anisotropy (FA) value calculated from preoperative magnetic resonance
diffusion tensor imaging to predict meningioma consistency.[2]
[5] Currently, the absence remains of a definite method in meningioma consistency prediction.
Many studies reported hyperintensity of meningioma on T2WI was associated with soft
consistency of meningioma.[7]
[8]
[15]
[16] However, using the combination of MRI sequences which may add additional predictive
value has not yet been studied well. This study used T1WI, T2WI, FLAIR, and DWI sequences
of a preoperative MRI to predict the consistency of the tumor and study the combination
of these MRI sequences to evaluate the value in meningioma consistency prediction.
Methods
The prospective study included all the patients with a diagnosis of an intracranial
meningioma undergoing tumor evaluation on two 1.5T MRI systems (Signa Excite HD Software,
Versions 14 and 16; GE Medical Systems) using an 8-channel head coil. Preoperative
MRI pulse sequence protocol was the standard used to evaluate intracranial tumor (T1WI,
T2WI, FLAIR, DWI, gradient recalled echo T2*, and post-gadolinium T1W with fat suppression).
The signal intensity of the tumor was evaluated blindly by an experienced neuroradiologist,
comparing the tumor signal with the signal intensity of cerebral cortical grey matter
in vicinity to the tumor. The tumor signal intensity pattern was described as hypointensity,
isointensity, and hyperintensity ([Fig. 1]). The patients were operated on by one of the five attending neurosurgeons at the
Faculty of Medicine, Chiang Mai University Hospital from July 1, 2012 to June 8, 2020.
All 287 patients included were pathologically confirmed as meningiomas. The data collected
included sex, age, tumor location, intraoperative consistency, and reported pathology.
Intraoperative consistency was categorized in four grades according to the method
of surgical removal. The grades were identified as the methods of the major part of
the tumor (> 75%) removed including suction, the intensity of the ultrasonic aspirator
(Söring: SONOCA 300), sharp cut, and cautery loop. Grade 1 was defined as soft tumor,
which was easily removed by suction. Grade 2 was defined as quite soft tumor, some
fibrous stroma with capsule which could be removed by ultrasonic aspirator with power
intensity less than 50%. Grade 3 was defined as average consistency tumor, denser
stroma with capsule whose removal required more than 50% of ultrasonic aspirator power
intensity. Grade 4 was defined as hard or calcified tumor requiring sharp cut or cutting
cautery loop for removal. Grades 1 and 2 consistencies were considered to be soft,
while grades 3 and 4 were considered to be hard. The intraoperative findings regarding
consistency correlated with preoperative MRI sequences which were interpreted blindly
by an experienced neuroradiologist. Statistical analysis was performed using STATA
Program, Version 11 (serial number 40110561832). For statistical analysis, we collected
the data and analyzed using percent and mean (standard deviation). Univariable risk
regression was used to identify the risk for soft and hard consistency. A p-value of 0.05 or less was considered statistically significant. The study was reviewed
and approved by the Research Ethics Committee, Faculty of Medicine, Chiang Mai University,
Chiang Mai, Thailand (ID266/2557).
Fig. 1 (A) T1-weighted image (T1WI) = 0, T2-weighted image (T2WI) = 2, fluid-attenuated inversion
recovery (FLAIR) = 2, diffusion-weighted image (DWI) = 2, T1WI with gadolinium. (B) T1WI = 0, T2WI = 0, FLAIR = 0, DWI = 0, T1WI with gadolinium. Magnetic resonance
imaging (MRI). T1WI, T2WI, FLAIR, DWI, and T1WI with gadolinium sequences in orders.
While 0 = hypointensity, 1 = isointensity, 2 = hyperintensity.
Result
Of 287 patients enrolled in the study, 29 (10.10%) were male and 258 (89.90%) female.
Their ages ranged from 22 to 83 years. In total, 189 tumors were situated in the supratentorial
space including 26 convexity meningiomas, 25 parasagittal, and 138 anterior cranial
fossa and sphenoid wing meningiomas. In all, 98 were in the middle fossa including
cavernous sinus meningiomas and those in the infratentorial location. Details of tumor
location are shown in [Table 1]. In total, 187 were World Health Organization (WHO) grade I, 95 were WHO grade II,
and 5 were WHO grade III in which WHO grades were not significantly associated with
meningioma consistency as shown in [Table 2]. Altogether, 125 tumors were found to be of soft consistency (grades 1, 2) and 162
tumors of hard consistency (grades 3, 4). The consistency was not significantly associated
with tumor location (p = 0.151) as shown in [Table 2]. T1WI signal intensity was not significantly associated with meningioma's consistency.
Hyperintensity signals on T2WI, FLAIR, and DWI were significantly associated with
soft consistency of meningiomas (relative risk [RR] 2.02, 95% confidence interval
[CI] 1.35–3.03, p = 0.001, RR 2.19, 95% CI 1.43–3.35, p < 0.001, and RR 1.47, 95% CI 1.02–2.11, p = 0.037, respectively). Chance to be soft consistency significantly increased when
two and three hyperintensity signals were combined (RR 2.75, 95% CI 1.62–4.65, p ≤ 0.001 and RR 2.79, 95% CI 1.58–4.93, p < 0.001, respectively), as shown in [Table 3]. Hypointensity signals on T2WI, FLAIR, and DWI were significantly associated with
hard consistency of meningiomas (RR 1.82, 95% CI 1.18–2.81, p = 0.007, RR 1.80, 95% CI 1.15–2.83, p = 0.010, and RR 1.67, 95% CI 1.07–2.59, p = 0.023, respectively) and chance to be hard consistency significantly increased
when three hypointensity signals were combined (RR 1.82, 95% CI 1.11–2.97, p = 0.017) as shown in [Table 4].
Table 1
Characteristic of patients
|
Parameters
|
N (%)
(n = 287)
|
|
Sex, n (%)
|
|
• Female
|
258 (89.90)
|
|
• Male
|
29 (10.10)
|
|
Age (y)
|
|
Mean (SD)
|
51.72 (9.87)
|
|
Min-max
|
22–83
|
|
≤ 60 y
|
235 (81.88)
|
|
> 60 y
|
52 (18.12)
|
|
Location, n (%)
|
|
• Anterior cranial fossa
|
138 (48.08)
|
|
• Middle cranial fossa
|
37 (12.89)
|
|
• Posterior cranial fossa
|
61 (21.25)
|
|
• Convexity
|
26 (9.06)
|
|
• Parasagittal
|
25 (8.71)
|
|
Consistency grading, n (%)
|
|
• Suction
|
105 (36.59)
|
|
• < 50% power
|
20 (6.97)
|
|
• > 50% power
|
82 (28.57)
|
|
• Scissors
|
80 (27.87)
|
Abbreviation: SD, standard deviation.
Table 2
Demographic data and patients' parameter (MRI findings, location, pathology) and tumor
consistency
|
Parameters
|
Soft
|
Hard
|
p-Value
|
|
N = 125
|
N = 162
|
|
Age, year, n (%)
|
0.443
|
|
• ≤ 60
|
105 (84)
|
130 (80.25)
|
|
•> 60
|
20 (16)
|
32 (19.75)
|
|
Gender, n (%)
|
0.559
|
|
• Male
|
11 (8.80)
|
18 (11.11)
|
|
• Female
|
114 (91.20)
|
144 (88.89)
|
|
Location, n (%)
|
0.151
|
|
• Anterior cranial fossa
|
56 (44.80)
|
82 (50.62)
|
|
• Middle cranial fossa
|
23 (18.40)
|
14 (8.64)
|
|
• Posterior cranial fossa
|
23 (18.40)
|
38 (23.46)
|
|
• Convexity
|
11 (8.80)
|
15 (9.26)
|
|
• Parasagittal
|
12 (9.60)
|
13 (8.02)
|
|
T1WI signal, n (%)
|
0.347
|
|
• Hypointensity
|
22 (17.60)
|
22 (13.58)
|
|
• Isointensity
|
88 (70.40)
|
126 (77.78)
|
|
• Hyperintensity
|
15 (12)
|
14 (8.64)
|
|
T2WI signal, n (%)
|
< 0.001
|
|
• Hypointensity
|
1 (0.80)
|
24 (14.81)
|
|
• Isointensity
|
31 (24.80)
|
62 (38.27)
|
|
• Hyperintensity
|
93 (74.40)
|
76 (46.91)
|
|
FLAIR, n (%)
|
< 0.001
|
|
• Hypointensity
|
1 (0.80)
|
22 (13.58)
|
|
• Isointensity
|
26 (20.80)
|
59 (36.42)
|
|
• Hyperintensity
|
98 (78.40)
|
81 (50)
|
|
DWI, n (%)
|
< 0.001
|
|
• Hypointensity
|
3 (2.40)
|
23 (14.29)
|
|
• Isointensity
|
76 (60.80)
|
103 (63.98)
|
|
• Hyperintensity
|
46 (36.80)
|
35 (21.74)
|
|
Consistency grading, n (%)
|
< 0.001
|
|
• Suction
|
105 (84)
|
0
|
|
• < 50% ultrasonic aspirator power
|
20 (16)
|
0
|
|
•> 50% ultrasonic aspirator power
|
0
|
82 (50.62)
|
|
• Sharp cut and cautery loop
|
0
|
80 (49.38)
|
|
WHO
|
0.422
|
|
• Grade I
|
79 (63.20)
|
108 (66.67)
|
|
• Grade II
|
45 (36.0)
|
50 (30.86)
|
|
• Grade III
|
1 (0.8)
|
4 (2.47)
|
Abbreviations: DWI, diffusion-weighted image; FLAIR, fluid-attenuated inversion recovery;
MRI, magnetic resonance imaging; T1WI, T1-weighted image; T2WI, T2-weighted image;
WHO, World Health Organization.
Table 3
Chance for soft consistency (univariable analysis)
|
Signal
|
RR
|
95% CI
|
p-Value
|
|
T1WI hyperintensity
|
1.21
|
0.71–2.08
|
0.483
|
|
T2WI hyperintensity
|
2.02
|
1.35–3.03
|
0.001
|
|
FLAIR hyperintensity
|
2.19
|
1.43–3.35
|
< 0.001
|
|
DWI hyperintensity
|
1.47
|
1.02–2.11
|
0.037
|
|
Combination of either hyperintensity T2WI, FLAIR, DWI
|
|
- Hyperintensity in 1 phase
|
2.19
|
1.05–4.56
|
0.035
|
|
- Hyperintensity in 2 phase
|
2.75
|
1.62–4.65
|
< 0.001
|
|
- Hyperintensity in 3 phase
|
2.79
|
1.58–4.93
|
< 0.001
|
Abbreviations: CI, confidence interval; DWI, diffusion-weighted image; FLAIR, fluid-attenuated
inversion recovery; RR, relative risk; T1WI, T1-weighted image; T2WI, T2-weighted
image.
Table 4
Chance for hard consistency (univariable analysis)
|
Signal
|
RR
|
95% CI
|
p-Value
|
|
T1W1 hypointensity
|
0.87
|
0.55–1.36
|
0.537
|
|
T2W1 hypointensity
|
1.82
|
1.18–2.81
|
0.007
|
|
FLAIR hypointensity
|
1.80
|
1.15–2.83
|
0.010
|
|
DWI hypointensity
|
1.67
|
1.07–2.59
|
0.023
|
|
Combination of either hypointensity T2WI, FLAIR, DWI
|
|
- Hypointensity in 1 phase
|
1.37
|
0.56–3.34
|
0.490
|
|
- Hypointensity in 2 phase
|
1.92
|
0.79–4.68
|
0.153
|
|
- Hypointensity in 3 phase
|
1.82
|
1.11–2.97
|
0.017
|
Abbreviations: CI, confidence interval; DWI, diffusion-weighted image; FLAIR, fluid-attenuated
inversion recovery; RR, relative risk; T1WI, T1-weighted image; T2WI, T2-weighted
image.
Discussion
Meningiomas are globally one of the most common intracranial tumors, more than 90%
being pathologically benign[17] The management strategy goal is to keep the patient fully functional and provide
long-term relief or prevent intracranial tumor growth-associated problems.[18] Extent and completeness of surgical resection are clearly related to recurrence
rate.[18] Many factors including deep-seated location of the tumors, adjacent structure involvement,
and major neurovascular encasement are all influential and can preclude safe total
surgical removal of the meningioma.
Consistency of the meningioma is one of the most important factors influencing the
completeness of surgical excision and surgical risks especially in situations with
neurovascular involvement. Meningiomas with soft consistency are simply removable
by suction and are quite safe to achieve total removal even in situations of major
neurovascular encasement.[19] However, hard consistency tumors, where tumor removal requires a sharp cut or cutting
cautery loop, may harbor much higher risks of neurovascular injury while attempting
total surgical removal. In this situation maximum safe resection should be considered
to avoid devastating complications from iatrogenic neurovascular injury. Hence, preoperative
recognition of tumor consistency represents valuable information to optimize surgical
approach selection and good surgical planning to enable achieving the optimal goal
for the individual patient.
Tumor consistency grading systems were proposed by some studies. Zada et al[11] reported five consistency score grades according to the method used in the tumor
debulking process including suction, ultrasonic aspirator, sharp dissection, and loop
cautery. Smith et al[9] reported three tumor consistency grading scales based on Cavitron ultrasonic surgical
aspirator intensity used in tumor removal. Currently, tumor consistency grading systems
remain inconclusive. Tumor consistency may differ in parts of the tumor. In this study,
we graded tumor consistency by methods used to remove the major part of tumor (> 75%)
including suction, ultrasonic aspirator power less than 50%, ultrasonic aspirator
power greater than 50%, and sharp cut or cautery loop which is more clearly defined
and less subjective as shown in [Table 5].
Table 5
Meningioma consistency grading system
|
Consistency grade
|
General description
|
Instrument(s) used for internal debulking for more than 75% of tumor volume
|
|
1
|
Soft
|
Suction
|
|
2
|
Quite soft, some fibrous stroma with capsule
|
Ultrasonic aspirator power < 50%
|
|
3
|
Average consistency, denser stroma with capsule
|
Ultrasonic aspirator power > 50%
|
|
4
|
Hard or calcified tumor
|
Sharp cut with scissor or cautery loop
|
MRI is an important imaging tool for preoperative evaluation in most intracranial
brain tumors, providing essential details of soft tissue components and related adjacent
structure. The hyperintensity of soft meningiomas on T2WI are thought to be related
to higher water content of the tumor, while hypointensity on T2WI of hard tumors might
be due to less water, more collagen and calcium content, and hypercellularity of the
tumors.[7] Benign meningiomas typically showed isointensity to slightly hypointensity to the
gray matter on T1WI and iso- to slightly hyperintensity on T2WI.[20] Meningiomas have a wide spectrum of histologic appearances in which some features
cause distinct MRI signal patterns. Microcystic variant shows predominant uniform
hypointensity on T1WI and hyperintensity on T2WI[21]; lipoblastic meningioma shows hyperintensity of intratumoral fat on T1WI,[22] and sclerotic meningioma (densely packed hyalinized collagen bundles and abundant
tumor calcification) shows hypointensity on T1WI and marked hypointensity on T2WI.[23]
Many earlier published studies endeavored to predict meningioma consistency using
several methods including quantitative measurement and conventional MRI sequence qualitative
methods, but a definite conclusion remains unavailable.[2]
[5]
[7]
[8]
[9]
[10]
[16]
[24]
[25]
Regarding quantitative studies, MRI calculation of FA values of hard fibroblastic
meningiomas were reported significantly higher than those of soft meningothelial meningiomas
(p = 0.002).[5] FA value of more than 0.3 reported a significant predictive value of hard consistency
of meningiomas.[2] Ortega-Porcayo et al[24] reported FA values for hard tumors were not significantly higher than those for
soft tumors (p = 0.115) and concluded FA value was not an independent predictor for tumor consistency.
Among conventional MRI sequence qualitative studies, mainly based on detecting free
water content of the tumor,[2] hyperintensity on protein density and that T2WI reported signs of a soft tumor.[16] Signal intensity of T2WI and FLAIR significantly correlated with the tumor consistency;
hypointense tumors on T2WI and FLAIR sequences tended to be hard, whereas tumors showing
hyperintensity of T2WI and FLAIR were associated with soft consistency.[8] These findings were also reported in related studies.[7]
[8]
[16]
In the present study, we found hyperintensity solely on T2WI, FLAIR, and DWI was significantly
associated with meningiomas of soft consistency and when two and three hyperintensity
of T2WI, FLAIR, and DWI were combined the predictive value of soft consistency significantly
increased as shown in [Table 3]. Also, hypointensity solely on T2WI, FLAIR, and DWI was significantly associated
with hard consistency and chance to be hard consistency significantly increased when
hypointensity was revealed in all three sequences as shown in [Table 4]. However, while directly comparing the relative risk of each parameter with the
combinations, due to limited number of sample size in each study arm, no statistical
significance was observed. Further study with larger sample size needs to be carried
out to better prove this finding.
Tumor size and location were also reported related to tumor consistency, meningioma
size (diameter > 2 cm), and supratentorial or sphenoidal wing location and were more
frequently associated with hard consistency meningiomas (p < 0.05).[2] However, in our study the location of meningiomas was not significantly associated
with tumor consistency (p = 0.151) as shown in [Table 2].
Conclusion
Each of T2WI, FLAIR, and DWI hyperintensity signal of the meningiomas on MRI was solely
and significantly associated with soft consistency and predictive value that significantly
increased when two and three hyperintensities of T2WI, FLAIR, and DWI were combined.
Each hypointensity signal of the meningiomas concerning T2WI, FLAIR, and DWI on MRI
was solely and significantly associated with hard consistency of tumors and chance
to be hard consistency significantly increased when hypointensity was found in all
three sequences. Combined intensity findings on T2WI, FLAIR, and DWI added more predictive
value of meningioma consistency prediction, further study with more sample size may
benefit to confirm this finding.
