AJR 2001; 176:1571-1575
© American Roentgen Ray Society
Contrast-Enhanced Helical CT of the Head and Neck
Improved Conspicuity of Squamous Cell Carcinoma on Delayed Scans
Reinhard Groell1,
Otto Doerfler1,
Gottfried J. Schaffler1 and
Walter Habermann2
1
Department of Radiology, University Hospital Graz, Auenbruggerplatz 9, A-8036
Graz, Austria.
2
Department of Otorhinolaryngology, University Hospital Graz, A-8036 Graz,
Austria.
Received October 3, 2000;
accepted after revision November 9, 2000.
Address correspondence to R. Groell.
Abstract
OBJECTIVE. We evaluated the impact of delayed scans on the
conspicuity of squamous cell carcinoma in helical CT of the head and neck.
SUBJECTS AND METHODS. Twenty-seven patients with biopsy-proven
squamous cell carcinoma of the head and neck underwent dual-phase helical CT
examinations using 100 mL of nonionic contrast material. In all patients, the
early phase started 30 sec after the commencement of injection. The patients
were assigned to one of two groups in which the delayed phase started either
180 sec (group A, n = 13) or 300 sec (group B, n = 14) after
the start of injection. The overall image quality, including vascular
opacification and the quality of lesion conspicuity, was determined according
to a three-point scoring system.
RESULTS. Overall image quality scored better on the early scans
(score, 1.4 ± 0.5) than on the late scans with a 180-sec (score, 1.6
± 0.6; p = 0.03) or a 300-sec delay (score, 2.4 ± 0.5;
p = 0.002). Tumor conspicuity scored better on scans with a 180-sec
delay (score, 1.4 ± 0.5) than on the scans with a 30-sec delay (score,
2.3 ± 0.7; p = 0.02) or the scans with a 300-sec delay (score,
2.3 ± 0.7; p = 0.03). In eight (62%) of 13 patients in group A
and in six (43%) of 14 patients in group B, the tumor was better delineated on
the late scans than on the early scans.
CONCLUSION. Although early scans provide optimal vascular
enhancement and are therefore necessary for helical CT studies of the head and
neck, additional delayed scans may improve lesion detection in patients with
squamous cell carcinoma of the head and neck.
Introduction
With the introduction of helical technology, the scanning time for CT has
gradually decreased during the last decade, which has enabled the performance
of CT studies of the entire head and neck with sufficient vessel opacification
[1,
2]. Sufficient contrast
enhancement of neck vessels is important for adequate interpretation of head
and neck scans, and adequate vessel attenuation is often used as a parameter
of image quality
[2,3,4,5,6,7].
Only a limited number of reports have evaluated soft-tissue enhancement
during contrast-enhanced CT of the head and neck. Harris et al.
[8] described several cases of
squamous cell carcinoma, lymphadenopathy, or pleomorphic adenoma of the
parotid gland that showed improved conspicuity on CT performed several minutes
after the injection of contrast material. On the other hand, Conrad et al.
[9] proposed arterial phase
imaging for the detection of squamous cell carcinoma of the head and neck
because they observed higher contrast between squamous cell carcinoma and
soft-tissue structures on scans obtained 20 sec after the start of injection
than on scans performed after 70 sec.
Thus, the impact of delayed scans remains uncertain for helical CT studies
of the head and neck. This uncertainty prompted us to design a prospective
study of patients with squamous cell carcinoma of the head and neck and to
evaluate the impact of early and delayed scans on tumor conspicuity.
Subjects and Methods
The study population consisted of 27 patients (six women, 21 men) who were
27-85 years old (mean, 60 ± 12 years) with biopsy-proven primary or
recurrent squamous cell carcinoma of the head and neck. The carcinomas
originated in the pharynx in 13 patients, in the larynx in seven patients, in
the tongue in two, in the esophagus in one, in the tonsils in one, in the soft
palate in one, in the skin in one, and in the external auditory canal in one
patient. Ten of the 27 patients additionally had cervical lymph node
metastases that were either surgically or biopsy-proven or that fulfilled CT
criteria of infiltration as described in the literature (> 1.5 cm, necrotic
areas, or rim enhancement)
[10]. The other 17 patients
were classified as N0 (no lymph node metastases).
CT was performed in the helical mode with one of two CT scanners in our
department (Somatom Plus 4, Siemens, Erlangen, Germany; or Light-Speed Qx/i,
General Electric Medical Systems, Milwaukee, WI). The scanning volume covered
the area from the skull base to the thoracic inlet in the cephalocaudal
direction using axial slices. The patients were lying in the supine position.
All studies were performed in the helical mode with 3- or 3.75-mm slice
thickness using a pitch of 1.5, and all images were reconstructed every 3 or
3.75 mm, respectively. The average scan duration was approximately 30 sec. In
both scanners, tube current and voltage were 220 mA and 120 kVp, respectively.
In both scanners a 512 x 512 matrix was used, and the field of view
varied from 20 to 30 cm.
Each patient received a total of 100 mL of nonionic contrast material (300
mg I/mL of Ultravist [iopromide]; Schering, Berlin, Germany) at a uniphasic
flow rate of 2 mL/sec administered by power injector (MCT; Medrad, Pittsburgh,
PA) through an IV cannula located in an antecubital vein. In all patients, the
early scans started after a delay of 30 sec after the commencement of
injection. The late phase started either 180 sec (group A, n = 13) or
300 sec (group B, n = 14) after the commencement of injection using
the same image acquisition parameters as for the early scans. No additional
contrast material was administered for the late scans. During scanning the
patients were instructed to breathe normally but not to swallow.
Image analysis was performed on a digital image workstation (Sienet Magic
View; Siemens). The attenuation of the carotid artery, the internal jugular
vein, and the sternocleidomastoid muscle was evaluated at three cervical
levels (below the scull base and at the levels of the submandibular and
thyroid glands). When the sternocleidomastoid muscle was not visible or was
too small for the calculations at the most cephalic level, the temporal muscle
was used for muscle evaluation. In all patients the left-sided carotid artery,
jugular vein, and muscle were measured, except for certain patients in whom
the right side of the neck was used for the calculation because, for example,
the left internal jugular vein was too small or was absent or the patient had
undergone left-sided neck dissection. Moreover, attenuation measurements were
performed in three thyroid nodules (visible in three patients) and in 19
morphologically nonsuspicious cervical lymph nodes (in 12 patients whose
disease was staged as N0). These nonsuspicious lymph nodes also fulfilled the
criteria published in the literature that were mainly based on size, shape,
and homogeneity of the nodes
[10]. Finally, attenuation was
measured in the tumors and in lymph nodes that were known to be infiltrated.
When the tumors were partly necrotic, attenuation measurements were performed
in the solid area. The measurements were performed using a circular
region-of-interest cursor that varied in size to match the different tissues,
and care was taken to keep sufficient distance from contour borders to
minimize possible partial volume effects.
All images were displayed using a standard soft-tissue window setting
(center, 70 H; width, 250 H), and the interpreting radiologists could change
the window settings as they thought necessary. Each study was interpreted in
consensus by two radiologists who analyzed the early and delayed studies for
each patient at the same time. The radiologists scored the overall image
quality (including vascular enhancement, topographic delineation of soft
tissues, general assessment of lymph nodes) and tumor conspicuity according to
a three-point scoring system (1 = good, 2 = medium, and 3 = poor).
Statistical comparisons of attenuation values between the results of the
early and the late scans and between groups A and B were performed using the
Student's t test for paired and unpaired samples, respectively. The
nonparametric scoring results were analyzed using the Wilcoxon's and
Mann-Whitney tests. A p value of 0.05 was considered statistically
significant. All statistical analyses were performed with a standard PC
software package (Statistical Package for the Social Sciences, version 10;
SPSS, Chicago, IL).
Results
The mean attenuation values on early and late scans for the tissues
investigated are listed in Table
1. Vascular opacification was best on the early scans and
decreased with increasing delay. The reported overall image quality score
(Table 2) was significantly
better on early scans than on scans with a 180-sec (p = 0.03) and a
300-sec delay (p = 0.002). Moreover, the overall image quality score
was significantly better on the scans having a 180-sec delay than on those
with a 300-sec delay (p = 0.04).
Squamous cell carcinoma revealed prolonged contrast material uptake (Figs.
1A,1B,2A,2B,3A,3B,4A,4B,5A,5B)
when compared with the sternocleidomastoid muscle, with the submandibular and
thyroid glands, and with nonsuspicious lymph nodes. The contrast between tumor
and sternocleidoid muscle (as background) was significantly greater on delayed
than on early scans (3 ± 13 H vs. 14 ± 10 H; p <
0.01). No significant difference between attenuation of primary carcinoma and
infiltrated cervical lymph nodes was present on early or delayed scans.
View larger version (132K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B. —85-year-old woman with metastases from squamous cell
carcinoma of hypopharynx. On delayed (300-sec delay) helical CT scan,
metastases (solid arrows) show higher enhancement than on A.
Note carotid artery calcifications (open arrow) are visible on
delayed scan.
|
|
View larger version (117K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A. —65-year-old man with metastatic adenopathy caused by squamous
cell carcinoma of hypopharynx. Early (30-sec delay) helical CT scan shows weak
enhancement of carcinoma (short arrows) and lymph node metastasis
(long arrows).
|
|
View larger version (118K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B. —65-year-old man with metastatic adenopathy caused by squamous
cell carcinoma of hypopharynx. Delayed (180-sec delay) helical CT scan shows
greater attenuation of carcinoma (short arrows) and metastasis
(long arrows) than A, with better demarcation of metastasis
from sternocleidomastoid muscle.
|
|
View larger version (131K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A. —47-year-old man with necrotic lymph node metastases
originating from squamous cell carcinoma of soft palate. Early (30-sec delay)
helical CT scan shows weak necrotic rim enhancement (arrows).
|
|
View larger version (139K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B. —47-year-old man with necrotic lymph node metastases
originating from squamous cell carcinoma of soft palate. On delayed (180-sec
delay) helical CT scan, necrotic rims (arrows) are better delineated
than on A.
|
|
Tumor discrimination scored significantly better on the late scans with a
180-sec delay than on the scans with a 300-sec delay (p = 0.03) or on
the early scans (p = 0.02). However, no significant difference was
seen in the quantitative evaluation of tumor-to-muscle contrast between scans
with a 180-sec delay and those with a 300-sec delay.
In eight (62%) of 13 patients in group A (180-sec delay) and in six (43%)
of 14 patients in group B (300-sec delay), the carcinoma was better visualized
on the late than on the early scans (Figs.
1A,1B,2A,2B,3A,3B,4A,4B,5A,5B).
In no patient of group A (180-sec delay) and in two (14%) of 14 patients of
group B (300-sec delay), the carcinoma was better demarcated on the early
scans than on the late scans.
Morphologically nonsuspicious lymph nodes were generally better demarcated
from vessels on the early scans because of a greater vessel-to-lymph node
contrast (Figs.
1A,1B,2A,2B,3A,3B,4A,4B,5A,5B;
Table 1).
The attenuation of the submandibular gland was not significantly different
between the early and the delayed scans, whereas the attenuation of the
thyroid gland was greater on the early scans (p < 0.001). In each
of the three patients with thyroid nodules, the nodules were better demarcated
on the early scans (Fig.
6A,6B)
because of a high nodule-to-thyroid contrast of 91 ± 9 H. On the
delayed scans, the attenuation of the nodules was comparable to that of the
surrounding thyroid gland parenchyma (mean difference, 8 ± 18 H). In
one patient, the nodule was not seen at all on the late scans; and in one
patient the thyroid nodule was visible on the late scan only because of the
presence of peripheral calcifications.
Discussion
In this study, approximately 50% of squamous cell carcinomas of the head
and neck were better delineated on the late scans (scan delay, 180 sec or 300
sec) than on the early scans (scan delay, 30 sec), which resulted in better
determination of tumor size and tumor discrimination against adjacent
soft-tissue structures such as musculature, pharyngeal and laryngeal wall, and
cutis. Squamous cell carcinoma revealed increased contrast material uptake on
the late scans, with a maximum uptake on scans obtained 180 sec after the
start of injection (Table 1).
The conspicuity score was significantly better for the scans using a delay of
180 sec than for those using a delay of 300 sec and for the early scans. That
is why we conclude that for the optimal delineation of squamous cell carcinoma
of the head and neck, delayed images obtained 180 sec after the start of
contrast injection are superior to early images and to images obtained 300 sec
after the start of injection.
Cervical vessels were generally better opacified on the early scans; their
attenuation decreased with increasing delay. Thus, early images seem to be
necessary, especially to differentiate nonsuspicious lymph nodes from vessels.
Moreover, when imaging squamous cell carcinoma, early images provide optimal
vascular enhancement that shows the relation of tumors to the adjacent
cervical vessels. In the three patients with thyroid nodules, the nodules were
also generally better seen on the early scans. Finally, in two (14%) of 14
patients in group B (300-sec delay), the tumors were better demarcated on the
early scans than on the late scans.
To combine the merits of early vascular and delayed tumor enhancement,
Harris et al. [8] proposed the
administration of a second bolus of approximately 50 mL of contrast agent
during the delayed scanning. However, the administration of a second bolus
would increase the amount of contrast material needed for helical CT studies
of the head and neck in patients with squamous cell carcinoma. A possible
future alternative might be digital image fusion of early and delayed scans,
which could combine adequate opacification of vessels and tumors without
additional contrast medium administration. To our knowledge, such studies have
not been performed; we plan to evaluate this in the future.
Conrad et al. [9] described
the superiority of scans obtained 20 sec after the start of injection over
scans acquired 70 sec after the start. Those researchers showed that the
contrast between squamous cell carcinoma of the head and neck and surrounding
soft-tissue structures was significantly better on the arterial phase scans
(20-sec delay) than on the later scans (70-sec delay). However, between the
scans at 20 sec and those at 70 sec, they observed only a minimal increase in
tumor enhancement (from 107 ± 23 H to 108 ± 21 H). In our study,
between early and late scans (combining the scans with 180- and 300-sec delay)
the tumors revealed a mean enhancement of 19 ± 13 H. This finding is
also supported by the images shown in the report of Harris et al.
[8], although quantitative data
were not given in their article. Therefore, we consider that a scan delay of
70 sec after the start of injection might be too early to achieve optimal
tumor-to-background contrast in patients with squamous cell carcinoma of the
head and neck.
In conclusion, using early scanning for helical CT of the head and neck is
necessary to achieve adequate vessel opacification. In patients with squamous
cell carcinoma of the head and neck, improved lesion conspicuity may be
achieved when additional delayed scans are obtained approximately 180 sec
after the start of contrast material injection.
References
-
Suojanen JN, Mukherji SK, Dupuy DE, Takahashi JH, Costello P.
Spiral CT in evaluation of head and neck lesions: work in progress.
Radiology
1992;183:281
-283[Abstract/Free Full Text]
-
Spreer J, Krahe T, Jung G, Lackner K. Spiral versus conventional CT
in routine examinations of the neck. J Comput Assist
Tomogr 1995;19:905
-910[Medline]
-
Yoon DH, Chang KH, Han MH, et al. Re-evaluation of optimal dose of
contrast medium for vascular enhancement in CT of the head and neck.
Neuroradiology
1997;39:30
-34[Medline]
-
Ehritt-Braun C, Ferstl F, Brehm A, Krause W, Langer M. Optimizing
intravenous bolus contrast medium injection in cervical CT diagnosis [in
German]. Aktuelle Radiol
1994;4:222
-224[Medline]
-
Feuerbach S, Lorenz W, Klose KJ, et al. Contrast medium
administration in spiral computed tomography: the results of a consensus
conference [in German]. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb
Verfahr 1996;164:158
-165[Medline]
-
Gay SB, Pevarski DR, Phillips CD, Levine PA. Dynamic CT of the
neck. Radiology
1991;180:284
-285[Abstract/Free Full Text]
-
Groell R, Willfurth P, Schaffler GJ, et al. Contrast-enhanced
spiral CT of the head and neck: effect of contrast material injection rate.
AJNR
1999;20:1732
-1736[Abstract/Free Full Text]
-
Harris EW, LaMarca AJ, Kondroski EM, Murtagh FR, Clark RA. Enhanced
CT of the neck: improved visualization of lesions with delayed imaging.
AJR
1996;167:1057
-1058[Free Full Text]
-
Conrad R, Pauleit D, Layer G, et al. Spiral CT of the head-neck
area: the advantages of the early arterial phase in the detection of
squamous-cell carcinomas [in German]. Rofo Fortschr Geb Rontgenstr
Neuen Bildgeb Verfahr
1999;171:15
-19[Medline]
-
Steinkamp HJ, Hosten N, Richter C, Schedel H, Felix R. Enlarged
cervical lymph nodes at helical CT. Radiology
1994;191:795
-798[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
R. Groell, M. Wiltgen, G. Wolf, and W. Habermann
Digital Image Fusion of Early and Delayed CT Scans: How to Achieve Optimal Opacification of Vessels and Squamous Cell Carcinomas of the Head and Neck
Am. J. Roentgenol.,
January 1, 2002;
178(1):
211 - 213.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
