AJR 2002; 178:705-710
© American Roentgen Ray Society
Irradiated Carcinoma of the Tongue
Correlation of MR Imaging Findings with Pathology
Noriaki Tomura1,
Osamu Watanabe1,
Koki Kato1,
Satoshi Takahashi1,
Jiro Watarai1,
Masato Sageshima2 and
Michinori Yokomizo3
1
Department of Radiology, Akita University School of Medicine, 1-1-1, Hondo,
Akita City, Akita, 010-8543, Japan.
2
Department of Pathology, Akita University School of Medicine, Akita City,
Akita, 010-8543, Japan.
3
Department of Otorhinolaryngology, Akita University School of Medicine, Akita
City, Akita, 010-8543, Japan.
Received April 23, 2001;
accepted after revision September 14, 2001.
Address correspondence to N. Tomura.
Abstract
OBJECTIVE. MR imaging was prospectively correlated with pathologic
findings to study whether MR imaging can differentiate viable from nonviable
tumor tissue in the irradiated carcinoma of the tongue.
SUBJECTS AND METHODS. MR examinations were performed after radiation
therapy in 21 patients with carcinoma of the tongue. All patients underwent
either a total glossectomy or hemiglossectomy after radiation therapy.
Specimens were examined microscopically. Radiation changes were histologically
graded into four groups (I, minimal cellular changes; II, presence of cellular
changes and partial destruction of the tumor; III, only nonviable tumor cells;
IV, no tumor cells). MR examinations included T2-weighted imaging, unenhanced
T1-weighted imaging, dynamic contrast-enhanced imaging, and contrast-enhanced
T1-weighted imaging.
RESULTS. On unenhanced T1-weighted images, the lesion was
hypointense, except for two patients with histologic grade III. On T2-weighted
images, the lesion appeared hyperintense in 12 of 14 patients with viable
tumor cells (grades I and II); however, the lesion was hypointense in four,
and isointense in two of seven patients with nonviable tumor cells (grades III
or IV). Contrast-enhanced T1-weighted images showed that the degree of
contrast enhancement of the lesion was equal to or lower than that of a normal
salivary gland in 18 of 21 patients. For the time of maximal enhancement of
the lesion on dynamic imaging, there was no substantial difference between
viable (grades I and II) and nonviable (grades III and IV) tumor tissue.
CONCLUSION. The present study shows that T2-weighted imaging is
feasible for differentiating viable from nonviable tumor tissue in irradiated
carcinoma of the tongue.
Introduction
The first choice of treatment for carcinoma of the oral tongue is surgery
[1,
2]. Recently, preoperative
chemotherapy and radiotherapy
[3,4,5,6]
have been introduced to make partial resection possible. Partial glossectomy
is recommended for a better quality of life for the patients. In our hospital,
patients with carcinoma of the oral tongue are treated either by preoperative
radiation therapy plus surgery or by surgery alone. In carcinomas of the
tongue, MR imaging is superior to CT in depicting the tumor and its
surrounding structures [7]. MR
imaging is primarily performed for assessment of carcinomas of the tongue.
Degenerated tumors or the absence of tumor after irradiation would change MR
imaging findings. Radiologists should be familiar with the changes in MR
findings after irradiation of carcinomas of the tongue. In particular, the
absence of a residual tumor in MR imaging after radiation therapy is
clinically important because surgery may be avoided. In this prospective
study, MR imaging findings were compared with pathologic findings in the
irradiated tongue to assess the value of MR imaging in differentiating
radiation-induced changes from residual tumors in patients who underwent
radiation therapy for carcinoma of the oral tongue. Dynamic MR imaging was
also performed to evaluate the changes of hemodynamics after irradiation.
Subjects and Methods
From December 1996 to January 2000, 28 consecutive patients with carcinoma
of the tongue were treated with radiation therapy. Twenty-one patients with
resectable disease underwent MR imaging before surgery. All 21 patients had
biopsy-proven squamous cell carcinoma of the oral tongue. The patients
included 16 men and five women ranging in age from 45 to 77 years (average
age, 59.6 years). Clinical TNM staging was determined according to the
criteria of the International Union Against Cancer
[8]. Four patients were
diagnosed as T1, 11 as T2, three as T3, and three as T4. The N factor was
determined as N0 in 10 patients, N1 in four, N2b in three, and N2c in four.
All patients were free of distant metastasis. Before surgery, all patients
were irradiated with a 6-MV linear accelerator using planned radiotherapy with
once-daily fractionation. A total dose of 27-59.4 Gy (average, 40.3 ±
5.4 Gy) was given to the primary site.
MR imaging was performed with a 1.5-T unit within 1 week from the end of
radiation therapy. Axial and coronal T1-weighted imaging with a spin-echo
sequence (TR range/TE range, 320-680/9-14; number of excitations, 2) and
T2-weighted imaging with a fast spin-echo sequence (3000-5000/98-108; number
of excitations, 2) were performed before the administration of contrast
material. Dynamic imaging was performed in the coronal plane in 10 patients.
After rapid injection of contrast material (gadopentetate dimeglumine, 0.1
mmol/kg), eight to 12 sections were obtained every 19-29 sec for 114-189 sec
using fast spoiled gradient-recalled acquisition in the steady state sequence
(110-160/2.3-4.2; number of excitations, 1; flip angle, 80-90°). The
region of interest was set within the tumor, and the curve of signal intensity
was established over time. Contrast-enhanced T1-weighted MR imaging with a
spin-echo sequence in the axial and coronal planes was performed in all
patients. In 18 of 21 patients, a frequency-selective fat-suppression
technique was used for contrast-enhanced T1-weighted images. The field of view
was 18 cm for all images, with a matrix of 256 x 192, 256 x 224,
or 256 x 256; a slice thickness of 4 or 5 mm; and an interslice gap of 1
or 2 mm.
A total glossectomy (n = 10) or hemiglossectomy (n = 11)
was performed 10-30 days after MR imaging. Specimens obtained at glossectomy
were examined microscopically by one pathologist using H and E staining.
Special attention was focused on radiation changes. Histologic grading by
Shimosato [9] and Shimosato et
al. [10] was applied (0, no
changes; I, minimal cellular changes; IIa, in spite of the presence of
cellular changes and partial destruction of the tumor, viable tumor cells are
present in more than a quarter of the tumor mass; IIb, viable tumor cells are
present in less than a quarter of the tumor mass; III, only nonviable tumor
cells are present; IV, no tumor cells). The histologic grades were determined
without being aware of each patient's MR imaging finding. MR imaging findings
were determined before surgery. MR imaging findings, such as the signal
intensity of the tumor in unenhanced T1-weighted and T2-weighted imaging and
the degree of contrast enhancement in contrast-enhanced T1-weighted imaging,
were evaluated by three of the researchers, and a decision was made by
consensus. The time of maximal enhancement in the dynamic imaging was
determined on the time-intensity curve. The signal intensity and the degree of
contrast enhancement of the tumor were compared with histologic grading. One
radiologist and one pathologist retrospectively correlated the signal
abnormalities on the coronal MR imaging sections with lesions in coronal
histologic sections.
Results
One tumor was classified as histologic grade I, six as IIa, seven as IIb,
six as III, and one as IV. Three of four tumors with T1 stage showed
histologic grades IIa or IIb. Seven of 11 tumors with T2 stage showed
histologic grades IIa or IIb. There was no correlation between the
International Union Against Cancer T staging and the histologic grading for
radiation changes.
In the unenhanced T1-weighted image, two tumors with histologic grade III
were isointense compared with the signal intensity of the contralateral side
of the tongue. Nineteen tumors showed a hypointense mass. The sensitivity and
specificity of isointensity on T1-weighted images for histologic grades III
and IV were 29% (2/7) and 100% (14/14), respectively.
In T2-weighted images, four tumors appeared hypointense, four appeared
isointense, and 13 appeared hyperintense compared with the signal intensity of
the contralateral side of the tongue (Table
1). Four tumors with hypointense masses were histologic grades III
(three tumors) or IV (one tumor) (Fig.
1A,1B,1C,1D,1E).
Six of seven tumors with histologic grades III or IV showed isointense (two
tumors) or hypointense masses (four tumors). The sensitivity and specificity
of isointensity and hypointensity on T2-weighted images for histologic grades
III and IV were 86% (6/7) and 86% (12/14).
View larger version (128K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A. —36-year-old man with carcinoma of oral tongue classified as
T2. Patient underwent radiation with total dose of 39.6 Gy. T2-weighted MR
image obtained after irradiation shows lesion (arrows) to be slightly
hypointense compared with signal intensity of contralateral side.
|
|
View larger version (119K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —36-year-old man with carcinoma of oral tongue classified as
T2. Patient underwent radiation with total dose of 39.6 Gy. T1-weighted MR
image shows lesion (arrows) to be hypointense compared with signal
intensity of contralateral side.
|
|
View larger version (138K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C. —36-year-old man with carcinoma of oral tongue classified as
T2. Patient underwent radiation with total dose of 39.6 Gy. Contrast-enhanced
MR image shows same degree of enhancement of tumor (arrows) as
salivary glands.
|
|
View larger version (87K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1D. —36-year-old man with carcinoma of oral tongue classified as
T2. Patient underwent radiation with total dose of 39.6 Gy.
Time—intensity curve derived from dynamic MR image shows maximal
enhancement of tumor 54-81 sec after infusion of contrast material. 1 = tumor,
2 = contralateral side of tumor, 3 = digastric muscle.
|
|
View larger version (175K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1E. —36-year-old man with carcinoma of oral tongue classified as
T2. Patient underwent radiation with total dose of 39.6 Gy. Photomicrograph of
histopathologic specimen shows no tumor cells (grade IV). (H and E,
x25)
|
|
In the contrast-enhanced T1-weighted images, two tumors with histologic
grade III could not be detected. The degree of contrast enhancement was
stronger in one tumor, the same in 13, and lower in five, comparable with that
of the salivary gland (Fig.
2A,2B,2C,2D,2E).
In seven tumors with histologic grades III or IV, contrast-enhanced
T1-weighted images showed no abnormalities in two tumors, the same degree of
contrast enhancement in three tumors, and lower enhancement in two tumors,
compared with the contrast enhancement of the salivary gland. The degree of
contrast enhancement did not correlate with histologic grading.
View larger version (147K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A. —48-year-old man with carcinoma of oral tongue classified as
T3. Patient underwent radiation with total dose of 39.6 Gy. T2-weighted MR
image after irradiation shows lesion (arrows) to be hyperintense
compared with signal intensity of contralateral side.
|
|
View larger version (142K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B. —48-year-old man with carcinoma of oral tongue classified as
T3. Patient underwent radiation with total dose of 39.6 Gy. T1-weighted MR
image shows lesion (arrows) to be hypointense compared with signal
intensity of contralateral side.
|
|
View larger version (150K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C. —48-year-old man with carcinoma of oral tongue classified as
T3. Patient underwent radiation with total dose of 39.6 Gy. Contrast-enhanced
T1-weighted MR image with fat saturation shows moderate enhancement of tumor
(arrows).
|
|
View larger version (100K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D. —48-year-old man with carcinoma of oral tongue classified as
T3. Patient underwent radiation with total dose of 39.6 Gy. Photomicrographs
of histopathologic specimens show viable tumor cells (grade IIb)
(arrows and arrowheads). (H and E, x1 [D];
x50 [E])
|
|
View larger version (158K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2E. —48-year-old man with carcinoma of oral tongue classified as
T3. Patient underwent radiation with total dose of 39.6 Gy. Photomicrographs
of histopathologic specimens show viable tumor cells (grade IIb)
(arrows and arrowheads). (H and E, x1 [D];
x50 [E])
|
|
In dynamic imaging, two tumors with histologic grade III showed no abnormal
findings. In eight tumors that were detectable in dynamic imaging, the time of
maximal enhancement was less than 40 sec in one tumor, 40-120 sec in two, and
120 sec or more in five (Fig.
3A,3B,3C,3D).
No definite relationship was observed between the time of maximal enhancement
and histologic grading.
View larger version (151K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A. —68 year-old man with carcinoma of oral tongue classified as
T2. Patient underwent radiation with total dose of 39.6 Gy. Dynamic MR image
22-44 sec after infusion of contrast material.
|
|
View larger version (148K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B. —68 year-old man with carcinoma of oral tongue classified as
T2. Patient underwent radiation with total dose of 39.6 Gy. Dynamic MR image
44-66 sec after infusion of contrast material shows maximal contrast
enhancement (arrows).
|
|
View larger version (91K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C. —68 year-old man with carcinoma of oral tongue classified as
T2. Patient underwent radiation with total dose of 39.6 Gy.
Time—intensity curve derived from dynamic MR image shows maximal
enhancement of tumor at third scan. 1 = tumor, 2 = contralateral side of
tumor, 3 = geniohyoid muscle.
|
|
View larger version (183K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3D. —68 year-old man with carcinoma of oral tongue classified as
T2. Patient underwent radiation with total dose of 39.6 Gy. Photomicrograph of
histopathologic specimen shows viable tumor cells (grade IIb). (H and E,
x50)
|
|
Discussion
Radiation-induced changes in various organs are well documented
[11,12,13].
The early phase of radiation changes is seen some weeks after completion of
treatment. The late phase of radiation changes is related to the development
of late fibrosis. The patients in our series were examined within 1 week after
completion of radiation therapy, and MR imaging was performed to detect early
postradiation changes. Changes in tumors induced by irradiation are caused by
the direct effect of irradiation on the tumor cells, impairment of the blood
supply, and fibrosis and hyalinization of the stroma. Shimosato
[9] studied histologic changes
induced by irradiation in lung tumors. In his study, patients received a total
dose of 8-122.5 Gy. Duration from completion of radiotherapy to surgery was
from 1 to 75 days but within 14 days in most instances. He noted that a
histiocytic reaction with proliferation of fibroblasts was a prominent
feature, which was provoked by injured cells or cellular debris or both, along
with disappearance of the tumor cells. A marked increase in elastic fibers was
often observed. In the present study, six of seven tumors with histologic
grades III or IV appeared as hypointense (four tumors) or isointense masses
(two tumors) on T2-weighted images. These four tumors, which appeared
hypointense on T2-weighted images, histologically showed fibrotic changes with
nonviable tumor cells or without tumor cells. Signal intensity patterns and
changes in longitudinal relaxation time in fibrotic tissues have not yet been
reported for the tongue. To the best of our knowledge, the present study may
be the first report correlating the signal on T2-weighted imaging with
fibrotic changes in the oral tongue. Some authors have reported differences in
signal intensity patterns between recurrent tumors and radiation fibrosis
[11,12,13,14,15,16,17].
Prolonged longitudinal relaxation times have been shown in many malignant
tumors, and these findings have been ascribed to an increased intracellular
water content in neoplastic tissues. Glazer et al.
[11] found statistically
significant differences in signal intensity between recurrent tumors and
radiation fibrosis on T1-weighted imaging using spin-echo sequences (TR/TE,
500/30 or 500/60), but they failed to find statistically significant
differences on T2-weighted imaging (1500/60 or 1500/90). Ebner et al.
[15] studied signal intensity
patterns of recurrent tumors in the female pelvis and the value of signal
intensity patterns in differentiating active tumors from localized fibrotic
tissues. They compared signal intensities of recurrent tumors with fibrotic
tissues using T1-weighted imaging (TR range/TE, 400-800/20 or 25), moderately
T2-weighted imaging (2000-2500/20 or 25), and heavily T2-weighted imaging
(2000-2500/70 or 80). Statistically significant differences in signal
intensity between recurrent tumors and late fibrosis (
1 year after
treatment) were found on heavily T2-weighted imaging sequences. Their study
showed a broad distribution of signal intensity values in early fibrosis, less
than 6 months after therapy, which may reflect an inflammatory reaction after
surgery. The present study was different from the previously mentioned study.
In the present study, patients received radiation therapy alone before MR
imaging. Signal intensities of the lesion seem to reflect early changes
because of irradiation. It is suggested that the accumulation of collagen and
dense connective tissue might be partly responsible for the hypointense mass
on T2-weighted imaging seen in patients with histologic grades III or IV.
Hyperintensity on T2-weighted imaging in histologic grades I or II may be
caused by an increased intracellular water content of tumor cells and tumoral
edema.
Takashima et al. [18]
correlated MR imaging with pathology in 10 patients with tongue cancer. They
failed to differentiate viable tumor from postradiation scar tissue in one
patient by spin-echo T2-weighted imaging. They used a TR of 1800 msec and TE
of 90 msec on a magnet operating at 0.1 T, which was not a heavily T2-weighted
image. Arakawa et al. [19]
also correlated MR imaging with pathology in 11 patients with tongue cancer.
The study by Arakawa et al. included nine patients who underwent radiation or
chemotherapy or both before surgery. These researchers suggested that dynamic
and T2-weighted imaging were considerably superior to contrast-enhanced
T1-weighted imaging in revealing the lesion. In two of their patients who
microscopically showed no viable tumor cells after treatment, no abnormal
lesions were noted in any of the MR imaging sequences, including the dynamic
imaging. In their series, there were no cases with hypointensity of the lesion
after treatment. The acquisition times and total scanning times of dynamic MR
imaging used in their study were almost equal to those of the dynamic imaging
used in the present study. Two patients with histologic grade III in the
present study appeared isointense, and the lesions were also not detected by
dynamic imaging. The total scanning time of dynamic imaging was 2-3 min after
rapid injection of contrast material, which was not long enough to analyze the
washout rate of contrast material from the lesions. Thus, we evaluated the
time of maximal enhancement of the lesion but could not differentiate viable
(grades I and II) from nonviable tumor tissue (grades III and IV). Contrast
enhancement on dynamic imaging and contrast-enhanced T1-weighted imaging may
represent a histiocytic reaction or an inflammatory reaction in nonviable
tumor tissue. As seen in a microscopic examination, this MR finding also
indicates that the vascularity of the lesions was not decreased. This may be
due to the contraction of the tumor, resulting in a relatively increased
number of blood vessels per area. Dao et al.
[12] reported that short
inversion recovery imaging and spin-echo T2-weighted imaging could not
differentiate recurrent tumors from radiation fibrosis in the irradiated
breast, but that differentiation was possible with dynamic imaging.
Postirradiation changes, such as inflammatory reaction and edema, depend on
the site of the tumor, the dose of irradiation, and the duration from the end
of irradiation. With regard to dynamic imaging, further studies should be
performed, and an improved technique of dynamic imaging with shorter
acquisition time and longer scanning time is needed.
Our study has some limitations. The number of subjects studied was too
small to estimate a statistical significance. The sensitivity and specificity
of T1-weighted and T2-weighted imaging were evaluated; however, a statistical
estimate in a large number of subjects is necessary for T1-weighted and
T2-weighted images and contrast-enhanced studies, including the dynamic scan.
Although MR imaging was performed within 1 week after radiation therapy, the
time interval between MR imaging and surgery was not short (range, 10-30
days). This time interval should be shorter because the correlation between MR
imaging findings and pathology depends on this time.
In summary, this prospective study was preliminary and performed in a small
population. However, our results provide new information concerning the role
of MR imaging with postir-radiated carcinoma of the tongue. Correlated with
pathologic findings, the signal intensity of T2-weighted imaging was the most
helpful in distinguishing viable from nonviable tumor tissue. The present
study suggests that MR imaging after radiation therapy has an important role
in the treatment selection because surgery may be omitted for patients with
histologic grades III and IV.
References
-
Spiro RH, Spiro JD, Strong EW. Surgical approach to squamous
carcinoma confined to the tongue and the floor of the mouth. Head
Neck Surg 1986;9:27
-31[Medline]
-
Hicks WL Jr, North JH Jr, Loree TR, et al. Surgery as a single
modality therapy for squamous cell carcinoma of the oral tongue. Am
J Otolaryngol 1998;19:24
-28[Medline]
-
Shirinian MH, Weber RS, Lippman SM, et al. Laryngeal preservation
by induction chemotherapy plus radiotherapy in locally advanced head and neck
cancer: the M.D. Anderson Cancer Center experience. Head
Neck 1994;16:39
-44[Medline]
-
Mendenhall WM, Parsons JT, Stringer SP, Classisi NJ, Million RR. T2
oral tongue carcinoma treated with radiotherapy: analysis of local control and
complications. Radiother Oncol
1989;16:275
-281[Medline]
-
Korb LJ, Spaulding CA, Constable WC. The role of definitive
radiation therapy in squamous cell carcinoma of the oral tongue.
Cancer
1991;67:2733
-2737
-
Fein DA, Mendenhall WM, Parsons JT, et al. Carcinoma of the oral
tongue: a comparison of results and complications of treatment with
radiotherapy and/or surgery. Head Neck
1994;16:358
-365[Medline]
-
Sigal R, Zagdanski AM, Schwaab G, et al. CT and MR imaging of
squamous cell carcinoma of the tongue and floor of the mouth.
RadioGraphics
1996;16:787
-810[Abstract]
-
Sobin LH, Wittekind C. (International Union Against Cancer [UICC]),
eds. TNM classification of malignant tumours, 5th ed.
Baltimore: Wiley-Liss, 1997:20
-24
-
Shimosato Y. Histopathological studies on irradiated lung tumors.
Gann
1964;55:521
-535
-
Shimosato Y, Oboshi S, Baba K. Histological evaluation of effects
of radiotherapy and chemotherapy for carcinomas. Jpn J Clin
Oncol 1971;1:19
-35
-
Glazer HS, Lee JKT, Levitt RG, et al. Radiation fibrosis:
differentiation from recurrent tumor by MR imaging.
Radiology
1985;156:721
-726[Abstract/Free Full Text]
-
Dao TH, Rahmouni A, Campana F, Laurent M, Asselain B, Fourquet A.
Tumor recurrence versus fibrosis in the irradiated breast: differentiation
with dynamic gadolinium-enhanced MR imaging. Radiology
1993;187:751
-755[Abstract/Free Full Text]
-
Hricak H, Swift PS, Campos Z, Quivey JM, Gildengorin V, Goranson H.
Irradiation of the cervix uteri: value of unenhanced and contrast-enhanced MR
imaging. Radiology
1993;189:381
-388[Abstract/Free Full Text]
-
Glazer HS, Levitt RG, Lee JK, Emami B, Gronemeyer S, Murphy WA.
Differentiation of radiation fibrosis from recurrent pulmonary neoplasm by
magnetic resonance imaging. AJR
1984;143:729
-730[Free Full Text]
-
Ebner F, Kressel HY, Mintz MC, et al. Tumor recurrence versus
fibrosis in the female pelvis: differentiation with MR imaging at 1.5 T.
Radiology
1988;166:333
-340[Abstract/Free Full Text]
-
Rafto SE, Amendola MA, Gefter WB. MR imaging of recurrent
colorectal carcinoma versus fibrosis. J Comput Assist
Tomogr 1988;12:521
-523[Medline]
-
Vanel D, Shapeero LG, De Baere T, et al. MR imaging in the
follow-up of malignant and aggressive soft tissue tumors: results of 511
examinations. Radiology
1994;190:263
-268[Abstract/Free Full Text]
-
Takashima S, Ikezoe J, Harada K, et al. Tongue cancer: correlation
of MR imaging and sinology with pathology. AJNR
1989;10:419
-424[Abstract]
-
Arakawa A, Tsuruta J, Nishimura R, et al. Lingual carcinoma:
correlation of MR imaging with histopathological findings. Acta
Radiol 1996;37:700
-707[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
P. Lam, K. M. Au-Yeung, P. W. Cheng, W. I. Wei, A. P.-W. Yuen, N. Trendell-Smith, J. H. C. Li, and R. Li
Correlating MRI and Histologic Tumor Thickness in the Assessment of Oral Tongue Cancer
Am. J. Roentgenol.,
March 1, 2004;
182(3):
803 - 808.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
