Int J Clin Exp Med 2016;9(10):19480-19490
www.ijcem.com /ISSN:1940-5901/IJCEM0019733
Original Article
Comparison of magnetic resonance
spectroscopy and diffusion weighted
imaging in the differentiation of glioma
recurrence from radiation necrosis: meta-analysis
Wen-Fei Li, Yu-Chen Zhang, Chen Niu, Li-Ping Guo, Feng-Li Liang, Xiao Ling, Xuan Niu, Ming Zhang
Department of Radiology, First Affiliated Hospital, School of Medicine, Xi’an Jiao Tong University, Xi’an, Shaanxi,
People’s Republic of China
Received November 13, 2015; Accepted April 8, 2016; Epub October 15, 2016; Published October 30, 2016
Abstract: Aim: The purpose of this study was to compare the diagnostic accuracy of diffusion-weighted magnetic
resonance imaging (DWI) and magnetic resonance spectroscopy (MRS) for the diagnosis of differentiating glioma
recurrence from radiation necrosis through a meta-analysis. Methods: We performed a meta-analysis of all available
studies of the diagnostic performance of DWI and MRS for differentiating glioma recurrence from radiation necrosis. PubMed, Web of Science, and Chinese Biomedical databases (CBM) were searched for initial studies. Pooled
sensitivity (SEN), specificity (SPE), negative likelihood ratio (NLR), positive likelihood ratio (PLR), and diagnostic odds
ratio (DOR) were calculated. Results: 5 studies involving 112 patients (116 lesions) met all inclusion and exclusion
criteria. For DWI, the pooled sensitivity of Apparent Diffusion Coefficient (ADC) was 0.80 and specificity was 0.81.
Overall, PLR was 3.90, NLR was 0.28 and DOR was 17.53. For MRS, the pooled sensitivity of Cho/Cr was 0.86 and
specificity was 0.79. PLR was 3.37, NLR was 0.22, and DOR was 21.04. The pooled sensitivity of Cho/NAA was 0.75
and specificity was 0.88, PLR was 4.57, NLR was 0.27, and DOR was 25.83. Without statistically significant differences except for the fact that the pooled sensitivity of Cho/NAA and Cho/Cr, and Cho/NAA also displayed higher
area under the curve and Q *index compared with ADC (P<0.05). Conclusion: Both DWI and MRS were accurate
tools for detecting glioma recurrence. Although MRS seemed to be superior to DWI, the latter could also be used to
differentiate glioma recurrence from radiation necrosis when the former was unavailable. However, MRS and DWI
could play different roles in differentiating glioma recurrence from radiation necrosis. Because of significant publication bias, pooled diagnostic measures might be overvalued. Large-sample randomized controlled studies were
needed to establish its value for differentiating glioma recurrence from radiation necrosis.
Keywords: DWI, MRS, glioma recurrence, radiation necrosis
Introduction
Primary tumor of the central nervous system
had an annual age-adjusted incidence rate of
28/100 000 in adults. Gliomas accounted 28%
of intracranial tumors but 80% of malignant
tumors [1]. So far, the current standard treatment was surgery followed by adjunctive radiotherapy and chemotherapy [2]. However, differentiating glioma recurrence from radiation
necrosis remained a great challenge. To solve
the problem, numerous innovative magnetic
resonance imaging technologies were focusing
on metabolic, structural and hemodynamic
properties of tissues, like proton magnetic res-
onance spectroscopic imaging (1H-MRS), diffusion weighted imaging (DWI), perfusion-weighted imaging (PWI) and dynamic susceptibility
contrast magnetic resonance imaging [3-8].
DWI was a magnetic resonance imaging (MRI)
technique, which based on the imaging of the
molecular mobility of water [9]. Apparent diffusion coefficient (ADC) was the most widely used
parameter, it could be used to grading gliomas
easily and quickly [10]. Previous studies had
investigated the diagnostic value of ADC in distinguishing glioma recurrence from radiation
necrosis, but the findings had been incongruent
[11, 12].
Differentiating glioma recurrence from radiation necrosis
Figure 1. Flow diagram of the
study selection process.
MRS provided information about metabolic tissue composition, advanced spectroscopic methods had been used to quantify markers of
brain tumor metabolism, (choline [Cho]), (creatine [Cr]), (N-acetyl-aspartate [NAA]), (lactate
[Lac] or lipids) were the most commonly used
parameters [13]. Results were usually expressed as ratios between brain metabolites.
There were some studies have evaluated the
diagnostic role of MRS for distinguishing glioma
recurrence from radiation necrosis [14-16],
nevertheless, the sensitivity and specificity of
each research were different. Although many
studies compared the effect between the DWI
and MRS, but there were no uniform results
[3-8]. Matsusue [5] reported that DWI had
equivalent diagnostic efficiency with MRS and
19481
perfusion MRI. In contrast, Fink [3] reported that a perfusion MR and multi-voxel MRS
superior than DWI for distinguishing glioma
recurrence from post-treatment effects. Thus,
the aim of this meta-analysis was to compare the diagnostic value of DWI and MRS for
differentiating glioma recurrence from radiation necrosis.
Materials and methods
Search strategy
A systematic literature search was performed
to identify studies assessing the diagnostic
value of DWI and MRS for differentiating glioma
recurrence from radiation necrosis. We systemInt J Clin Exp Med 2016;9(10):19480-19490
Differentiating glioma recurrence from radiation necrosis
Table 1. Characteristics of studies included in the meta-analysis
Study name
Year Country
Study
No. of No. of Mean age
Language
design
patient lesion (yr) (range)
M/F
Histology
Reference Imaging field
standard
strength
MRS metabolite
ratio
ADC
ratio
QUADAS
scores
NA
23
DiCostanzo et al.
2014
Italy
R
English
29
29
63 (38-74) 18/11
HGG (29)
His + Cli
3T
Cho/Cr and Cho/NAA
Fink et al.
2012
USA
R
English
38
39
48 (28-70) 20/18
NA
His + Cli
3T
Cho/Cr and Cho/NAA <1.28
Meng et al.
2011
China
R
Chinese
22
22
45 (13-73) 10/12
NA
His + Cli
3T
Cho/Cr and Cho/NAA <1.66
17
Matsusue et al.
2010
USA
R
English
15
15
47 (30-64)
9/6
LGG (7), HGG (6)
His + Cli
1.5T/3T
Cho/Cr and Cho/NAA <1.30
24
R
English
8
11
23-68
3/5
HGG (8)
His + Cli
3T
BobekBillewicz et al. 2010 Poland
Cho/Cr
<1.59
23
25
ADC, Apparent diffusion coefficient; Cho, Choline; Cli, Clinical; Cr, Creatine; M, Male; F, Female; HGG, High grade glioma; His, Histology; LGG, Low grade glioma; MRS, Magnetic resonance spectroscopy; NA,
Not available; NAA, N-acetyl-aspartate; R, Retrospective.
19482
Int J Clin Exp Med 2016;9(10):19480-19490
Differentiating glioma recurrence from radiation necrosis
Table 2. The diagnostic test parameter of MRS and ADC for differentiating glioma recurrence from radiation necrosis
Study name
Year
DiCostanzo et al.
2014
Fink et al.
2012
Meng et al.
2011
Matsusue et al.
2010
BobekBillewicz et al. 2010
Metabolite
ratio
Cho/Cr
Cho/NAA
ADC ratio
MRS and DWI
Cho/Cr
Cho/NAA
ADC ratio
Cho/Cr
Cho/NAA
ADC ratio
Cho/Cr
Cho/NAA
MRS
ADC ratio
Cho/Cr
Cho/NAA
ADC ratio
Cut-off
NA
NA
NA
NA
≥1.54
≥1.05
<1.28
≥1.51
≥1.42
<1.66
≥1.29
≥1.06
Cho/Cr≥1.29,
Cho/NAA≥1.06
<1.30
≥1.35
≥2.11
<1.59
TP FP FN TN
17
10
17
18
22
21
21
10
12
14
10
9
9
2 4 6
0 11 8
1 4 7
1 3 7
1 1 5
1 2 5
2 8 8
1 5 6
1 3 6
2 1 5
1 0 2
1 1 2
1 1 2
9
4
2
3
1
1
1
1
1
0
1
2
4
3
2
5
NA, Not available; Cho, Choline; Cli, Clinical; Cr, Creatine; NAA, N-acetyl-aspartate.
atically searched PubMed, Web of Science, and
Chinese Biomedical databases (CBM) to identify relevant published articles until April 20,
2014. A search algorithm was used by the following terms (both as text and MeSH term):
“Diffusion MR” or “diffusion weighted imaging”
or “DWI” or “MRS” or “magnetic resonance
spectroscopy” AND (“glioma” or “brain neoplasm”) AND (“recurrence” or “recurrent”) AND
(“sensitivity” OR “specificity” OR “false-negative” OR “false-positive” OR “diagnosis” OR
“accuracy”). Additionally, the reference lists of
retrieved articles were also scrutinized by handsearching to identify other potentially eligible
studies that have not been identified as
aforementioned.
Literature selection
Two investigators, who were blinded to the journal, author, institution and date of publication,
independently evaluated all eligible studies.
The inclusion criteria were: (1) patients were
histopathologically diagnosed as glioma, and
has a history of treatment with surgery or radiotherapy or chemotherapy; (2) DWI and MRS was
used to differentiate glioma recurrence from
19483
radiation necrosis; (3) The
values of true positive (TP),
false positive (FP), false
negative (FN), true negative (TN) should be calculated from the data reported; (4) the effective
case number was greater
than 8 in selected studies;
(5) only publications in
English and Chinese were
included. Abstracts presented at conferences, letters, unpublished data, reviews, case reports, editorials and comments were
excluded. Initial searches identified 619 articles.
According to the inclusion
and exclusion criteria mentioned above, Only 5 (1
Chinese article and 4
English articles) of these
articles were included this
analysis which contained
the appropriated data.
Quality assessment
The same two investigators who performed the
database searches also performed the related
data extraction independently. Relevant studies were further examined with QUADAS criteria
[17]. Basal characteristics (authors, year of
publication), and patients’ demographic characteristics (mean age, sex, and number) and
technical aspects (imaging field strength, b
value, quantitative parameter, cut-off value,
reference standard, TP, FP, FN and TN value)
were extracted.
Statistical analysis
Meta-disc software (Version 1.4) and Stata11.0
software (Stata, College Station, TX) was performed to analysis our result. The sensitivity,
specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio
(DOR) and their 95% confidence intervals (CIs)
were calculated from the original data. We also
obtained the summary receiver operating characteristic (SROC) curve and the area under the
curve (AUC). We used I2 tests to test heterogeneity. An I2 value ranges from 0% to 40% was
regarded as “mild heterogeneity”, value ranges
Int J Clin Exp Med 2016;9(10):19480-19490
Differentiating glioma recurrence from radiation necrosis
Table 3. Parameters and patterns of MRS in all studies
Study name
Di Costanzo et al.
Fink et al.
Meng et al.
Matsusue et al.
BobekBillewicz et al.
Pattern
NA
SVS/MVS
SVS
MVS
SVS
SVS (1.5T)
SVS (3T)
Slice thickness
10 mm
NA
NA
NA
NA
NA
NA
TR (ms)
1,500
2000
2000
2,000
2,000
1300/1500
1083
TE (ms)
144
144/288
144
144/288
36
135/30
288/35s
FOV
24 cm
24 cm
NA
24 cm
NA
NA
NA
Acquisition time
6 min 53 s
NA
NA
5-7 min
NA
NA
NA
MVS, Multi-voxel spectroscopic; SVS, Signal-voxel spectroscopic; NA, Not available.
Table 4. Parameters and patterns of DWI in all studies
Study name
Di Costanzo et al.
Fink et al.
Meng et al.
Matsusue et al.
BobekBillewicz et al. 1.5T
3T
Slice thickness
(mm)
5
4
6
4
5
4
Gap
(mm)
1
1
1
1
1
1
TR
(ms)
11,000
5,210
5000
5,210
3100
3080
TE
(ms)
66.6
53
65
53
99
70
Slice
number
NA
30
NA
28
NA
NA
Matrix
164×192
112×89
NA
112×89
192×192
112×256
Acquisition
time
44 s
NA
18 s
32 s
NA
NA
NA, Not available.
from 30% to 60% was treated as “moderate
heterogeneity”, value ranges from 50% to 90%
was regarded as “substantial heterogeneity”,
and I2 ranges from 75% to 100% was considered as “considerable heterogeneity” [18]. A
random effects model was used when P<0.05,
if not, a fixed effects model was used when
P>0.05.
Ethical approval
The ethical approval was not necessary for the
reason that our study type was systematic
review and meta-analysis.
Results
Literature selected
A total of 619 potentially relevant studies were
initially selected. Five eligible studies were
identified [3-7]. A flowchart summarizing the
selection process of the finally included studies
was shown in Figure 1. Study design was
described as retrospective in all studies. The
characteristics of the included studies were
presented in Tables 1 and 2. QUADAS scores
were shown in Table 1.
19484
Spectroscopy and diffusion weighted imaging:
acquisition technique
In the majority (n = 5) of the eligible studies,
spectroscopic data were acquired with 3T magnets except one study examined MR spectroscopy at 1.5T and 3T. Single-voxel spectroscopy
was used in 3 studies and two applying the
multi-voxel spectroscopy. Water saturation was
applied in all studies. Four studies provided
information on spectroscopic voxel size (Table
3).
Demographic data
In these 5 included studies, a total of 112
patients and 116 lesions, of which 60 were
men and 52 were women, were included. Age
distribution was heterogeneously reported in
Table 1, while mean age was reported in all
studies (38-74, 28-70, 13-73, 30-64, 23-68).
Final determination between tumor recurrence
and radiation necrosis was decided either histologically or through combined clinical and
imaging follow-up in all include articles, however, clinical follow-up criteria were imparity. MR
protocols also different from each other, scan
parameter and pattern of DWI and MRS were
shown in Tables 3 and 4, respectively.
Int J Clin Exp Med 2016;9(10):19480-19490
Differentiating glioma recurrence from radiation necrosis
Figure 2. Forest plot of the sensitivity and specificity of the meta-analysis. The pooled sensitivity and specificity of ADC in differentiation of recurrent glioma from
radiation necrosis (A); The pooled sensitivity and specificity of Cho/Cr in differentiation of recurrent glioma from radiation necrosis (B); The pooled sensitivity and
specificity of Cho/NAA in differentiation of recurrent glioma from radiation necrosis (C).
19485
Int J Clin Exp Med 2016;9(10):19480-19490
Differentiating glioma recurrence from radiation necrosis
Figure 3. Summary receiver-operating characteristic curve of ADC (A) and MRS (B) for the differential diagnosis of
glioma recurrence from radiation necrosis. The AUC and Q *index of MRS were significantly higher than those of ADC
(P<0.01). AUC = area under the curve, SE = standard error.
Diagnostic accuracy of MRS and DWI
Assessment of publication bias
The diagnostic data of DWI and MRS were performed pooled analysis, respectively. The
pooled sensitivities and specificities of these
methods were shown in Figure 2. The pooled of
ADC sensitivity was 0.80 (95% CI: 0.70, 0.88)
and specificity was 0.81 (95% CI: 0.64, 0.92).
Overall, PLR was 3.90 (95% CI: 2.00, 7.61) and
NLR was 0.28 (95% CI: 0.18, 0.45) (see in
Figure 2A). With MR spectroscopic data, two
variables (Cho/NAA and Cho/Cr ratios) were
shown to differentiate recurrent glioma from
radiation necrosis. The pooled sensitivity of
Cho/Cr was 0.86 (95% CI: 0.76, 0.93) and specificity was 0.79 (95% CI: 0.59, 0.92), PLR was
3.37 (95% CI: 1.77, 6.40) and NLR was 0.22
(95% CI: 0.11, 0.46) (see in Figure 2B). The
pooled sensitivity of Cho/NAA was 0.75 (95%
CI: 0.64, 0.85) and specificity was 0.88 (95%
CI: 0.68, 0.97), PLR was 4.57 (95% CI: 1.78,
11.72) and NLR was 0.27 (95% CI: 0.09, 0.67)
(see in Figure 2C). Three variables (Cho/NAA,
Cho/Cr, and ADC ratio) diagnostic odds ratio
(DOR) were 25.83 (95% CI: 6.51, 102.5), 21.04
(95% CI: 6.5, 68.13), 17.3 (95% CI: 6.08,
50.56), respectively. The SROC curves with Q
*index for DWI and MRS were shown in Figure
3A and 3B respectively. The AUC and Q *index
of MRS were significantly higher than those of
ADC (P<0.01).
The presence of publication bias was confirmed
by the Egger test (P<0.001). Because the small
sample size, our conclusion was adopted cautiously. Multimodal imaging trials and multicenter trials should be implemented in the
future.
19486
Discussion
With the advanced of the glioma therapy, radiotherapy had become an indispensable ways of
intracranial malignant tumor treatment [19].
However, clinical and imaging manifestations
of radiation necrosis liked tumor recurrence.
Accurate diagnosis of these post-treatment
lesions, either radiation necrosis or tumor
recurrence, was vital to the patient’s prognosis.
Whether a mass was a recurred glioma was
usually determined by histopathology. Nevertheless, histopathological examination could
not be applied to every patient due to ethical
reasons; it was unreasonable to perform a
biopsy or surgical resection in patients who
showed absence of clinical or radiological progression. Therefore, the current study used histopathological results and/or clinical and/or
radiological follow-up as the reference standard. Conventional MRI with gadolinium contrast couldn’t distinguish between tumor recurInt J Clin Exp Med 2016;9(10):19480-19490
Differentiating glioma recurrence from radiation necrosis
rence and radiation necrosis accurately, owing
to their similar imaging manifestations in gadolinium-enhanced MRI [20]. Modern advancements with PET scans, DWI and MRS had
shown prospect for differentiating tumor recurrence from radiation necrosis [21]. Previous
studies demonstrated that PET methods appeared relatively specific instrument as to
evaluation of the response to treatment and
differentiation between recurrent tumor tissue
and radiation necrosis [22, 23]. FDG uptake
suggested the presence of glioma recurrence,
whereas absence of FDG uptake suggested
the presence of necrosis [24]. Voges et al [25]
reported 11C-methionine superior to FDG in
illustrating residual of recurrent tumor tissue.
While 11C-methionine uptake may be increased when disruption of the blood brain
barrier, such as cerebral hematoma or even
necrotic areas resulted from radiotherapy [26].
PET scan was an expensive examination, and
had certain radiation effect to human body as
well. MRI was a noninvasive and economic
methods, the most important advantage was
no radiation to human. Besides, only a few
studies with small patient populations reported comparison of DWI and MRS in differentiating glioma recurrence from radiation necrosis in humans directly.
For decades, MRS had been used as a value
imaging tool to reveal the metabolism of brain
tumor. The metabolic profiles of tumor recurrence and radiation necrosis had been widely
studied, but there was still no consensus about
which spectroscopic parameter(s) and/or which
threshold level(s) ensure the better distinction
accuracy [3, 4, 8]. “Nonnormalized” ratios had
been largely used and several studies have
reported that recurrent tumours had higher
Cho/NAA and Cho/Cr than radiation necrosis
[3, 4, 8]. Taking into account the threshold values that maximized the discriminative power,
the reported diagnostic accuracy of Cho/NAA
and/or Cho/Cr ranged between 86 and 91%
[3-6, 8]. This study had demonstrated that MRS
exhibited a high diagnostic accuracy for the
detection of glioma recurrence, with the high
sensitivity of the Cho/Cr (0.86) and specificity
of the Cho/NAA (0.88).
The diagnostic value of DWI in differentiating
tumor recurrence from radiation necrosis had
not been fully reported, nor had a threshold of
19487
ADC been established that provided the best
diagnostic accuracy [27, 28]. In fact, comparing
diffusion features between recurrent tumor
and radiation necrosis groups, some authors
[11, 12] measured significantly higher ADC values in radiation necrosis group, while others
[29] found significantly higher ADC values in the
recurrent tumor and others [30] still did not find
any significant differences in mean ADC values.
In the current study, DWI showed a high diagnostic accuracy in detecting glioma recurrence,
with an overall sensitivity was 0.80 (95% CI:
0.70, 0.88) and specificity was 0.81 (95% CI:
0.64, 0.92). The AUC and Q *index of MRS were
higher than those of DWI (P<0.05).
In the presented meta-analysis, we demonstrated the slightly advantage of MRS over DWI
for differentiating glioma recurrence from radiation necrosis. Various studies had reported on
the broad range of the diagnosis accuracy of
this imaging technique. The current study
showed that the pooled sensitivity of DWI was
82%. Because ADC influenced by many factors,
such as the field strength, sequence, b value,
previous studies though a higher b-value provided better contrast and higher ADC value
theoretically [31]. Hein et al showed that ADC
ratios (1.43 vs. 1.82*10 -3 mm2/s) and mean
ADC values (1.18 vs. 1.4*10 -3 mm2/s) in the
recurrence group was significantly lower than
those of the radiation necrosis group [32].
Similar results presented by these studies [8,
11, 33]. MRS reflected the information of brain
metabolism more than the ADC, not only could
reflect the cell intensity, and also could reflect
the damage of neurons. Previous studies suggested that tumor recurrence might be characterized by high Cho and low NAA levels [3, 4, 16,
34]. In this meta-analysis, the pooled sensitivity of Cho/NAA was higher than DWI, and the
pooled specificity was closer to DWI (79-81%).
Substantial data in the literature also demonstrated Cho/NAA diagnosis accuracy was better than the ADC, this results was consistent
with previous articles [4, 35]. Moreover, in
SROC analyses, the AUC of Cho/NAA was higher than that of ADC (see Figure 3). Some studies considered combining 3T DWI and MRS
diagnostic results using multiparametric scoring system had potential to improve overall
diagnostic accuracy in distinguishing glioma
progression from post-radiation change beyond
that of each technique alone [5, 7].
Int J Clin Exp Med 2016;9(10):19480-19490
Differentiating glioma recurrence from radiation necrosis
Our systematic review found only included 5
studies that directly compared DWI and MRS in
differentiating glioma recurrence from radiation necrosis. Therefore, our conclusion should
be interpreted cautionly because of some reasons. The five limited included papers varied a
lot in several aspects. As we known, different
field strength (1.5T and 3T) and different MRI
equipments with different methodology and
post-processing could result in unfavorable
results. Firstly, most included articles paid
attention to diagnostic accuracy of single
metabolite ratio, it is unknown if combined
metabolite ratios could improve diagnostic efficiency. Secondly, although the absence of heterogeneity in this meta-analysis, designs of the
included studies varied a lot, such as glioma
grades, sample size and cut off values. Thirdly,
this meta-analysis only included papers published in the English and Chinese language,
which may miss some eligible unpublished
studies or reported in other languages. Besides,
some inevitable publication bias existed in the
results as Egger’s regression test indicated in
the meta-analysis.
Conclusions
Our meta-analysis demonstrated that MRS
superiority over glioma recurrence from radiation necrosis than DWI. So far, this was the
first meta-analysis to compare the overall diagnostic value of MRS and DWI in detecting recurrent glioma from radiation necrosis. Although
MRS had an advantage over DWI in detecting
the recurrence of glioma, DWI as an alternative
tool when MRS is unavailable. Based on the
above results, we tentatively indicated that
MRS was a favorable imaging method in the
detection of tumor recurrence in glioma owing
to high sensitivity. Cho/Cr was highly specific
but had low sensitivity in recurrence diagnosis.
However, the results of the meta-analysis were
concluded from 5 limited studies. Multimodal
imaging trials and multicenter trials should
be required in the future. Using prospective
designs and large-sample randomized controlled studies are needed to establish its value
for differentiating glioma recurrence from radiation necrosis.
Disclosure of conflict of interest
None.
19488
Abbreviations
DWI, diffusion-weighted magnetic resonance
imaging; MRS, Magnetic Resonance Spectroscopy; ADC, Apparent Diffusion Coefficient;
CBM, Chinese Biomedical databases; SEN,
sensitivity; SPE, specificity; NLR, negative likelihood ratio; PLR, positive likelihood ratio; DOR,
diagnostic odds ratio.
Address correspondence to: Ming Zhang, Department of Radiology, First Affiliated Hospital, School
of Medicine, Xi’an Jiao Tong University, 277 Yanta
West Road, Xi’an 710061, Shaanxi, People’s Republic of China. E-mail: [email protected]
References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
Ostrom QT, Gittleman H, Liao P, Rouse C, Chen
Y, Dowling J, Wolinsky Y, Kruchko C and
Barnholtz-Sloan J. CBTRUS statistical report:
primary brain and central nervous system tumors diagnosed in the United States in 20072011. Neuro Oncol 2014; 16 Suppl 4: iv1-63.
Zhang H, Ma L, Shu C, Wang YB and Dong LQ.
Diagnostic accuracy of diffusion MRI with
quantitative ADC measurements in differentiating glioma recurrence from radiation necrosis. J Neurol Sci 2015; 351: 65-71.
Fink JR, Carr RB, Matsusue E, Iyer RS, Rockhill
JK, Haynor DR and Maravilla KR. Comparison
of 3 Tesla proton MR spectroscopy, MR perfusion and MR diffusion for distinguishing glioma recurrence from posttreatment effects. J
Magn Reson Imaging 2012; 35: 56-63.
Mingzhu M, Shengli Z, Zhongchang M and
Gang Y. MR DWI combined with ~1H-MRS
analysis of multi-parameter in distinction of
brain glioma recurrence and post-radiation
brain injuries. Chin J Med Imaging Technol
2011; 27: 905-909.
Matsusue E, Fink JR, Rockhill JK, Ogawa T and
Maravilla KR. Distinction between glioma progression and post-radiation change by combined physiologic MR imaging. Neuroradiology
2010; 52: 297-306.
Bobek-Billewicz B, Stasik-Pres G, Majchrzak H
and Zarudzki L. Differentiation between brain
tumor recurrence and radiation injury using
perfusion, diffusion-weighted imaging and MR
spectroscopy. Folia Neuropathol 2010; 48: 8192.
Di Costanzo A, Scarabino T, Trojsi F, Popolizio
T, Bonavita S, de Cristofaro M, Conforti R,
Cristofano A, Colonnese C, Salvolini U and
Tedeschi G. Recurrent glioblastoma multiforme versus radiation injury: a multiparametric 3-T MR approach. Radiol Med 2014; 119:
616-624.
Int J Clin Exp Med 2016;9(10):19480-19490
Differentiating glioma recurrence from radiation necrosis
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
Zeng QS, Li CF, Liu H, Zhen JH and Feng DC.
Distinction between recurrent glioma and radiation injury using magnetic resonance spectroscopy in combination with diffusion-weighted imaging. Int J Radiat Oncol Biol Phys 2007;
68: 151-158.
Schaefer PW, Ozsunar Y, He J, Hamberg LM,
Hunter GJ, Sorensen AG, Koroshetz WJ and
Gonzalez RG. Assessing tissue viability with
MR diffusion and perfusion imaging. AJNR Am
J Neuroradiol 2003; 24: 436-443.
Kang Y, Choi SH, Kim YJ, Kim KG, Sohn CH,
Kim JH, Yun TJ and Chang KH. Gliomas:
Histogram analysis of apparent diffusion coefficient maps with standard- or high-b-value diffusion-weighted MR imaging--correlation with
tumor grade. Radiology 2011; 261: 882-890.
Alexiou GA, Zikou A, Tsiouris S, Goussia A,
Kosta P, Papadopoulos A, Voulgaris S, Tsekeris
P, Kyritsis AP, Fotopoulos AD and Argyropoulou
MI. Comparison of diffusion tensor, dynamic
susceptibility contrast MRI and (99m)TcTetrofosmin brain SPECT for the detection of
recurrent high-grade glioma. Magn Reson
Imaging 2014; 32: 854-859.
Xu JL, Li YL, Lian JM, Dou SW, Yan FS, Wu H
and Shi DP. Distinction between postoperative
recurrent glioma and radiation injury using MR
diffusion tensor imaging. Neuroradiology
2010; 52: 1193-1199.
Tsougos I, Svolos P, Kousi E, Fountas K,
Theodorou K, Fezoulidis I and Kapsalaki E.
Differentiation of glioblastoma multiforme
from metastatic brain tumor using proton magnetic resonance spectroscopy, diffusion and
perfusion metrics at 3T. Cancer Imaging 2012;
12: 423-436.
Ando K, Ishikura R, Nagami Y, Morikawa T,
Takada Y, Ikeda J, Nakao N, Matsumoto T and
Arita N. [Usefulness of Cho/Cr ratio in proton
MR spectroscopy for differentiating residual/
recurrent glioma from non-neoplastic lesions].
Nihon Igaku Hoshasen Gakkai Zasshi 2004;
64: 121-126.
Palumbo B, Lupattelli M, Pelliccioli GP, Chiarini
P, Moschini TO, Palumbo I, Siepi D,
Buoncristiani P, Nardi M, Giovenali P and
Palumbo R. Association of 99mTc-MIBI brain
SPECT and proton magnetic resonance spectroscopy (1H-MRS) to assess glioma recurrence after radiotherapy. Q J Nucl Med Mol
Imaging 2006; 50: 88-93.
Seeger A, Braun C, Skardelly M, Paulsen F,
Schittenhelm J, Ernemann U and Bisdas S.
Comparison of three different MR perfusion
techniques and MR spectroscopy for multiparametric assessment in distinguishing recurrent high-grade gliomas from stable disease.
Acad Radiol 2013; 20: 1557-1565.
19489
[17] Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM
and Kleijnen J. The development of QUADAS: a
tool for the quality assessment of studies of
diagnostic accuracy included in systematic reviews. BMC Med Res Methodol 2003; 3: 25.
[18] Higgins JP, G. S.: Cochrane Handbook for
Systematic Reviews of Interventions Version
5.1.0.
[19] Yaman E, Buyukberber S, Benekli M, Oner Y,
Coskun U, Akmansu M, Ozturk B, Kaya AO,
Uncu D and Yildiz R. Radiation induced early
necrosis in patients with malignant gliomas receiving temozolomide. Clin Neurol Neurosurg
2010; 112: 662-667.
[20] Hygino da Cruz LC Jr, Rodriguez I, Domingues
RC, Gasparetto EL and Sorensen AG.
Pseudoprogression and pseudoresponse: imaging challenges in the assessment of posttreatment glioma. AJNR Am J Neuroradiol
2011; 32: 1978-1985.
[21] Yang I and Aghi MK. New advances that enable
identification of glioblastoma recurrence. Nat
Rev Clin Oncol 2009; 6: 648-657.
[22] Schaller BJ, Modo M and Buchfelder M.
Molecular imaging of brain tumors: a bridge
between clinical and molecular medicine? Mol
Imaging Biol 2007; 9: 60-71.
[23] Schaller BJ, Cornelius JF, Sandu N and
Buchfelder M. Molecular imaging of brain tumors personal experience and review of the
literature. Curr Mol Med 2008; 8: 711-726.
[24] Di Chiro G, Oldfield E, Wright DC, De Michele D,
Katz DA, Patronas NJ, Doppman JL, Larson
SM, Ito M and Kufta CV. Cerebral necrosis after
radiotherapy and/or intraarterial chemotherapy for brain tumors: PET and neuropathologic
studies. AJR Am J Roentgenol 1988; 150: 189197.
[25] Voges J, Herholz K, Holzer T, Wurker M, Bauer
B, Pietrzyk U, Treuer H, Schroder R, Sturm V
and Heiss WD. 11C-methionine and 18F-2fluorodeoxyglucose positron emission tomography: a tool for diagnosis of cerebral glioma
and monitoring after brachytherapy with 125I
seeds. Stereotact Funct Neurosurg 1997; 69:
129-135.
[26] Dethy S, Goldman S, Blecic S, Luxen A, Levivier
M and Hildebrand J. Carbon-11-methionine
and fluorine-18-FDG PET study in brain hematoma. J Nucl Med 1994; 35: 1162-1166.
[27] Siu A, Wind JJ, Iorgulescu JB, Chan TA, Yamada
Y and Sherman JH. Radiation necrosis following treatment of high grade glioma--a review of
the literature and current understanding. Acta
Neurochir (Wien) 2012; 154: 191-201; discussion 201.
[28] Jain R, Narang J, Sundgren PM, Hearshen D,
Saksena S, Rock JP, Gutierrez J and Mikkelsen
T. Treatment induced necrosis versus recur-
Int J Clin Exp Med 2016;9(10):19480-19490
Differentiating glioma recurrence from radiation necrosis
[29]
[30]
[31]
[32]
rent/progressing brain tumor: going beyond
the boundaries of conventional morphologic
imaging. J Neurooncol 2010; 100: 17-29.
Sundgren PC, Fan X, Weybright P, Welsh RC,
Carlos RC, Petrou M, McKeever PE and
Chenevert TL. Differentiation of recurrent brain
tumor versus radiation injury using diffusion
tensor imaging in patients with new contrastenhancing lesions. Magn Reson Imaging
2006; 24: 1131-1142.
Asao C, Korogi Y, Kitajima M, Hirai T, Baba Y,
Makino K, Kochi M, Morishita S and Yamashita
Y. Diffusion-weighted imaging of radiation-induced brain injury for differentiation from tumor recurrence. AJNR Am J Neuroradiol 2005;
26: 1455-1460.
DeLano MC, Cooper TG, Siebert JE, Potchen
MJ and Kuppusamy K. High-b-value diffusionweighted MR imaging of adult brain: image
contrast and apparent diffusion coefficient
map features. AJNR Am J Neuroradiol 2000;
21: 1830-1836.
Hein PA, Eskey CJ, Dunn JF and Hug EB.
Diffusion-weighted imaging in the follow-up of
treated high-grade gliomas: tumor recurrence
versus radiation injury. AJNR Am J Neuroradiol
2004; 25: 201-209.
19490
[33] Perez-Torres CJ, Engelbach JA, Cates J, Thotala
D, Yuan L, Schmidt RE, Rich KM, Drzymala RE,
Ackerman JJ and Garbow JR. Toward distinguishing recurrent tumor from radiation necrosis: DWI and MTC in a Gamma Knife--irradiated
mouse glioma model. Int J Radiat Oncol Biol
Phys 2014; 90: 446-453.
[34] Prat R, Galeano I, Lucas A, Martinez JC, Martin
M, Amador R and Reynes G. Relative value of
magnetic resonance spectroscopy, magnetic
resonance perfusion, and 2-(18F) fluoro-2-deoxy-D-glucose positron emission tomography
for detection of recurrence or grade increase
in gliomas. J Clin Neurosci 2010; 17: 50-53.
[35] Zhang H, Ma L, Wang Q, Zheng X, Wu C and Xu
BN. Role of magnetic resonance spectroscopy
for the differentiation of recurrent glioma from
radiation necrosis: a systematic review and
meta-analysis. Eur J Radiol 2014; 83: 21812189.
Int J Clin Exp Med 2016;9(10):19480-19490