Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print April 18, 2006, 10.1148/radiol.2393050222
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2393050222v1
239/3/777    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Umeoka, S.
Right arrow Articles by Imamura, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Umeoka, S.
Right arrow Articles by Imamura, M.
(Radiology 2006;239:777-783.)
© RSNA, 2006

Gastrointestinal Imaging

Esophageal Cancer: Evaluation with Triple-Phase Dynamic CT—Initial Experience1

Shigeaki Umeoka, MD, Takashi Koyama, MD, PhD, Kaori Togashi, MD, PhD, Tsuneo Saga, MD, PhD, Go Watanabe, MD, PhD, Yutaka Shimada, MD, PhD and Masayuki Imamura, MD, PhD

1 From the Department of Diagnostic Imaging and Nuclear Medicine (S.U., K.T., T.S.) and Department of Surgery and Surgical Basic Science (G.W., Y.S., M.I.), Graduate School of Medicine, Kyoto University, Kyoto, Japan; and Department of Radiology, Kyoto University Hospital, 54 Kawahara-cho, Shogoin, Sakyo, Kyoto, 606-8507, Japan (T.K.). From the 2003 RSNA Annual Meeting. Received February 9, 2005; revision requested April 8; revision received April 28; accepted June 3; final version accepted August 11. Address correspondence to T.K. (e-mail: montpeti{at}kuhp.kyoto-u.ac.jp).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCE IN KNOWLEDGE
 References
 
Purpose: To prospectively assess which phase of a triple-phase dynamic contrast material–enhanced multi–detector row computed tomography (CT) protocol is optimal for visualization of esophageal cancer.

Materials and Methods: The study was supported by the local ethical committee; all patients gave written informed consent. Thirty-one lesions in 28 consecutive patients (26 men, two women; mean age, 65 years; range, 53–87 years) with histopathologically confirmed esophageal cancer were evaluated with triple-phase dynamic CT performed at 5, 35, and 65 seconds (first arterial, second arterial, and venous phases) after attenuation of 200 HU was obtained at the descending aorta. Qualitative image analysis was performed to assess appearance and conspicuity of the tumor. Appearances of all 31 lesions were classified into three categories—not identifiable, focal enhancement with or without minimal (<1 cm) wall thickening, and focal mass lesion or obvious (>1 cm) wall thickening. Results were compared with surgical or endoscopic ultrasonographic findings. Quantitative assessment included regions-of-interest measurement of the tumor and normal esophageal wall and the difference between those measurements. A paired t test was used to determine which phase showed the highest tumor attenuation and tumor-to–normal esophageal wall attenuation differences.

Results: At visual assessment, 30 lesions were identified in the second arterial phase. Of these 30 lesions, eight were focal enhancements; the best conspicuity was during the second arterial phase. Furthermore, seven of these eight lesions were T1 cancers. The remaining 22 lesions were enhanced masses or wall thickening. Twenty-one of these 22 tumors also showed best conspicuity in the second arterial phase. The greatest attenuation of tumors in the second arterial phase was 130.0 HU, and the difference in attenuation between tumor and normal esophageal wall was 50.6 HU in the second arterial phase, which were significantly higher than those in the other two phases (P < .01, each).

Conclusion: The second arterial phase of dynamic CT is the optimal phase for visualization of esophageal cancer.

© RSNA, 2006


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCE IN KNOWLEDGE
 References
 
Esophageal cancer is one of the common malignant neoplasms in the world. Every year, approximately 13 900 cases of esophageal cancer are diagnosed in the United States, and 13 000 Americans die of it (1). Patients with esophageal cancer generally present with progressive dysphagia, malnutrition, and weight loss. As with other malignant tumors, accurate TNM staging and localization of the esophageal cancer are important parameters for selection of the optimal treatment and for prediction of patients' prognoses.

Small polypoid lesions, plaquelike lesions, focal irregularities of the esophageal wall, and spreading superficial lesions are common findings of early esophageal cancer (2,3). Because some benign esophageal tumors such as squamous papillomas can reveal similar findings, subsequent biopsy with an upper-esophageal endoscopic technique is required to confirm malignancy. A barium swallow examination typically reveals mucosal irregularity or stricture or ulceration of the esophagus. In cases of early and advanced stage esophageal cancer, the TNM stage is determined after histopathologic diagnosis in order to devise therapeutic strategies based on results of multiple imaging studies, including upper gastrointestinal endoscopy, endoscopic ultrasonography (US), computed tomography (CT), and fluorine 18 fluorodeoxyglucose positron emission tomography (4).

Because endoscopic US can depict the normal esophageal wall as a five-layer structure, it can be used to evaluate the depth of tumor extension (5). Although CT has been used for preoperative evaluation of esophageal cancer, the major role of CT has been the depiction of lymph nodes, distant metastases, or both rather than the evaluation of the local status of esophageal cancer. Several studies attempted to use conventional CT to stage esophageal cancer, only to find that CT was useful for evaluating T4 lesions (6–8). The sensitivity of conventional CT protocols in localizing esophageal cancer, especially early stage cancer, is not satisfactory, perhaps because conventional CT cannot afford optimal conspicuity of esophageal cancers against the normal esophageal wall.

Several reports describe potential advantages of CT images obtained during the arterial phase after the administration of contrast material; these images depict gastrointestinal cancers in other organs (9–11). To our knowledge, however, this imaging technique has not been applied to esophageal cancer, as the esophagus is too long to be imaged entirely during the arterial phase at conventional CT. Multi–detector row CT has markedly improved time resolution and permits acquisition of arterial phase images of the entire esophagus during a single breath hold. Thus, the goal of our study was to prospectively assess which phase of a triple-phase dynamic contrast material–enhanced multi–detector row CT protocol is optimal for visualization of esophageal cancer.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCE IN KNOWLEDGE
 References
 
Patients
Our study population included 31 lesions in 28 consecutive patients (26 men, two women; age range, 53–87 years; mean age, 65 years) with esophageal cancer histopathologically proved at endoscopic biopsy who were referred to our institution for possible surgical treatment from June 2002 to March 2003. Three patients had double primary lesions. All 28 patients underwent CT for local staging of the tumors and for evaluation of lymph nodes and distant metastases. Informed consent, which included regard to radiation dose, was obtained from all 28 patients prior to performance of triple-phase dynamic contrast-enhanced CT, in accordance with a protocol approved by the ethical committee at our institution. Anatomic subsite was divided according to Union Internationale Contre le Cancer classification: cervical esophagus, from the lower border of the cricoid cartilage to the superior thoracic aperture; upper thoracic portion, from the superior thoracic aperture to the level of tracheal bifurcation; midthoracic portion as the proximal half between the tracheal bifurcation and the esophagogastric junction; and lower thoracic portion, as the distal half between the tracheal bifurcation and the esophagogastric junction. The esophageal cancers were in the cervical esophagus in two lesions, the upper thoracic portion in five lesions, the midthoracic portion in eight lesions, both the upper thoracic and midthoracic portions in three lesions, the lower thoracic portion in five lesions, and both the midthoracic and lower thoracic portions in eight lesions. Histopathologic diagnoses were squamous cell carcinoma (30 lesions, 27 patients) and adenocarcinoma arising in a Barrett esophagus (one lesion, one patient). Treatments were the following: surgical resection for 15 lesions in 14 patients, radiation therapy or combined chemotherapy and radiation therapy for seven lesions in five patients, endoscopic mucosal resection for one lesion in one patient, and combination of surgical resection with radiation therapy, chemotherapy, or both in eight lesions in eight patients. The local extent of the primary tumor (T classification) was determined with histopathologic assessment of the resected specimens (23 lesions) according to the TNM classification. For the eight lesions without surgical interventions, results from endoscopic US were used as the reference standard for diagnosis. The criteria for staging with endoscopic US evaluation were based on the classification system for the esophagus as proposed by the Japanese Society for Esophageal Diseases (12,13).

Dynamic CT Protocol
All examinations were performed with the same multi–detector row CT scanner (Aquilion M8; Toshiba Medical, Tokyo, Japan). CT scans were obtained after intravenous injection of 100 mL of nonionic contrast medium (350 or 370 mg of iodine per milliliter) (Iomeprol, Eisai, Tokyo, Japan; Iopamidol, Nihon Schering, Osaka, Japan) at a rate of 3 mL/sec. Triple-phase imaging with a single breath hold for each phase was automatically performed 5, 35, and 65 seconds (first arterial phase, second arterial phase, and venous phase) after the attenuation of the descending aorta reached 200 HU. The first and second arterial phase acquisitions covered from the neck to the level of the esophagogastic junction, whereas the venous phase acquisitions covered the whole abdominal and pelvic region, in addition to the neck and chest region. Thirty-second intervals were allowed between each phase (15 seconds to obtain images of the entire length of the esophagus and another 15 seconds for patients to recover their breath). Oral esophageal contrast material was not used in this study, because esophageal enhancement may be confused with tumor enhancement on images.

Images were acquired with 1- or 2-mm section thickness, beam collimation of 8, 556-msec rotation time, beam pitch of 1.7 or 0.875, 120 kVp, 300 mA per rotation, and 14-mm table feed per rotation. The thin-section CT data were transferred to a workstation (M900 Quadra; Ziosoft, Tokyo, Japan), and transverse 7.0-mm-thick sections were reconstructed at 7.0-mm intervals. The reconstruction field of view was 320 mm for each section. Multiplanar reformatted images were also reconstructed by one of the authors (S.U., with approximately 5 years of experience in thoracic and abdominal CT) with a 1.0-mm section thickness in the oblique sagittal plane, which included the tumor and either the trachea (in cases of cervical or upper thoracic esophageal cancers) or the descending aorta (in cases of midthoracic or lower thoracic esophageal cancers). Standard mediastinal window images (window width, 400 HU; window level, 60 HU) were used for displaying the images.

Qualitative Analysis
Qualitative image analysis was prospectively performed on both transverse CT images and multiplanar reformatted CT images by two radiologists (T.K. and T.S., each with approximately 9 years of experience in thoracic and abdominal CT) who were aware of the histopathologic diagnosis of esophageal cancer but had no information on numbers and location of the tumor described in the surgical and US endoscopic findings. These readers assessed the images for the presence of the tumor, which is noted as focal enhancement. When the tumor was identified, each reader recorded the location and thickness of the lesion. Discrepancies in assessment were solved by consensus. For each imaging phase, appearances of all 31 lesions were classified into the following three categories: not identifiable (type 1), focal enhancement with or without minimal (<1 cm) wall thickening (type 2), and focal mass lesion or obvious (>1 cm) wall thickening (type 3).

Visual assessment of the phase that best showed the tumor against normal esophageal wall was also performed by using a five-point scale: 1, nonidentifiable; 2, hardly identifiable; 3, adequate; 4, good; and 5, excellent. Conspicuity of a type 1 lesion was equivalent to a score of 1. Conspicuity of type 2 or 3 lesions was ranked according to scores 2–5. Although histopathologic confirmations were not obtained, linear or punctuate enhancements along the surface mucosa were considered to be physiologic mucosal enhancement or vessels. These enhanced structures were carefully excluded. High attenuation due to beam hardening artifacts at the interface between the esophageal lumen and the esophageal wall were also carefully excluded. The locations of the esophageal cancers at CT assessment were finally compared with surgical or endoscopic findings by another author (S.U.).

Quantitative Analysis
In parallel with qualitative analyses, quantitative analyses were performed by another radiologist (S.U.) who was aware of the tumor locations from endoscopic US or surgical results. The mean attenuations of the esophageal tumor and of the normal esophageal wall were measured in Hounsfield units within three circular regions of interest (ROIs), which were created as large as possible on both the cancerous and the normal esophageal wall on the transverse image. The average of the three measurements was calculated as attenuation of the tumor and attenuation of the normal esophageal wall. When defining all ROIs, the radiologist paid special attention not to include necrosis within the tumor, linear enhancement along the surface mucosa, esophageal contents, fat tissue, or surrounding vessels. If the tumor was not identifiable during one or two imaging phases, the ROI was defined by referring to images of another phase that showed the location of the lesion.

Data and Statistical Analysis
Tumor locations and appearances interpreted from the CT images were compared with surgical and endoscopic US findings. The differences in CT attenuation between the tumor and the normal esophageal wall were calculated as an indicator of visual conspicuity of the lesion. Both attenuation of the tumors and subtracted tumor-to–normal esophageal wall attenuation differences were compared between images from different phases. Paired t tests were performed with statistical software (Stat View version 5.0.1; SAS Institute, Cary, NC). Because data were accumulated over time, a repeated measurement distribution analysis was carried out. Compensation of the multiplex nature was performed by the Tukey method. A P value of less than .05 was considered to indicate a statistically significant difference.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCE IN KNOWLEDGE
 References
 
Staging
The local endoscopic US or surgical staging of the 31 lesions included T1 cancer in nine lesions (one T1a, eight T1b), T2 cancer in five lesions, T3 cancer in 12 lesions, and T4 cancer in five lesions.

Qualitative Analysis
The presence of all esophageal cancers depicted at CT had been correctly determined from the surgical specimens or endoscopic US. At CT, thirty lesions (97%), which included three skip lesions, were identifiable in the second arterial phase, whereas 22 lesions (71%) were identifiable in the first arterial phase and 24 lesions (77%) were identifiable in the venous phase (Table 1). Only one lesion (type 1) was not identifiable at CT during any of the three phases; this lesion was an adenocarcinoma arising in a Barrett esophagus. At histopathologic examination, the tumor was confined to the mucosal layer within the Barrett epithelium (T1a cancer). The mean wall thickness of the 30 identifiable tumors was 12 mm.


View this table:
[in this window]
[in a new window]

  		Table 1. Appearance of 31 Esophageal Tumors on Triple-Phase CT Images, according to Imaging Phase

 
The appearances of eight lesions were interpreted as type 2 cancers on second arterial phase images. Seven of these eight lesions (88%) were histopathologically diagnosed as T1b tumors, whereas the remaining lesion was diagnosed as T2 cancer. Of these eight lesions, none were identifiable in the first arterial phase, and only two lesions were depicted in the venous phase. All eight lesions were depicted most conspicuously in the second arterial phase (Table 2), and six lesions were identifiable only in the second arterial phase (Fig 1).


View this table:
[in this window]
[in a new window]

  		Table 2. Best Esophageal Tumor Conspicuity on Triple-Phase CT Images, according to Imaging Phase

 

Figure 1
View larger version (108K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a: Oblique sagittal reformatted CT images of 72-year-old man with submucosal esophageal cancer (T1b cancer) 20 x 15 mm in diameter in midthoracic portion. (a) First arterial phase image shows smooth esophageal walls without any abnormal enhancement. (b) Second arterial phase image reveals focal enhancement (arrow) about 8 mm in diameter and 15 mm in length without any wall thickening (approximately 9 mm thick). Image provides best conspicuity among three phases. Subcarinal adenopathy is seen (*). (c) Venous phase image does not show abnormality as well. Average attenuation of tumor is 82 HU in first arterial phase, 129 HU in second arterial phase, and 97 HU in venous phase. Difference in attenuation between tumor and esophageal wall is 10 HU in first arterial phase, 49 HU in second arterial phase, and 5 HU in venous phase.

 

Figure 1
View larger version (113K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b: Oblique sagittal reformatted CT images of 72-year-old man with submucosal esophageal cancer (T1b cancer) 20 x 15 mm in diameter in midthoracic portion. (a) First arterial phase image shows smooth esophageal walls without any abnormal enhancement. (b) Second arterial phase image reveals focal enhancement (arrow) about 8 mm in diameter and 15 mm in length without any wall thickening (approximately 9 mm thick). Image provides best conspicuity among three phases. Subcarinal adenopathy is seen (*). (c) Venous phase image does not show abnormality as well. Average attenuation of tumor is 82 HU in first arterial phase, 129 HU in second arterial phase, and 97 HU in venous phase. Difference in attenuation between tumor and esophageal wall is 10 HU in first arterial phase, 49 HU in second arterial phase, and 5 HU in venous phase.

 

Figure 1
View larger version (119K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1c: Oblique sagittal reformatted CT images of 72-year-old man with submucosal esophageal cancer (T1b cancer) 20 x 15 mm in diameter in midthoracic portion. (a) First arterial phase image shows smooth esophageal walls without any abnormal enhancement. (b) Second arterial phase image reveals focal enhancement (arrow) about 8 mm in diameter and 15 mm in length without any wall thickening (approximately 9 mm thick). Image provides best conspicuity among three phases. Subcarinal adenopathy is seen (*). (c) Venous phase image does not show abnormality as well. Average attenuation of tumor is 82 HU in first arterial phase, 129 HU in second arterial phase, and 97 HU in venous phase. Difference in attenuation between tumor and esophageal wall is 10 HU in first arterial phase, 49 HU in second arterial phase, and 5 HU in venous phase.

 
Twenty-two lesions had type 3 appearances and were identifiable in all three phases (Fig 2). These 22 lesions consisted of one T1b lesion, four T2 lesions, 12 T3 lesions, and five T4 lesions. In 21 of 22 lesions (95%), the best conspicuity between the tumor and the normal esophageal wall was obtained in the second arterial phase (Table 2). The mean wall thickness of eight type 2 lesions was 6.3 mm, whereas that of the 22 type 3 lesions was 13 mm.


Figure 2
View larger version (96K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a: Oblique sagittal reformatted CT images of 55-year-old man with T3 tumor of 5.5 cm in diameter in midthoracic portion (small arrows) and T3 tumor of 7.0 cm in diameter in upper thoracic portion (arrowheads). (a) First arterial phase image demonstrates tumors in upper and midthoracic portions. (b) Second arterial phase image displays an early attenuation of lesions and provides best conspicuity of tumor among the three phases. Lower edge of tumor is clearly identifiable on image (large arrow). (c) Venous phase image shows tumors with less conspicuous margins compared with second arterial phase image. Average attenuations of upper thoracic and midthoracic tumors are 83 HU and 89 HU in first arterial phase, 129 HU and 116 HU in second arterial phase, and 116 HU and 112 HU in venous phase, respectively. Tumor-to–normal esophageal wall attenuation differences are 28 HU and 34 HU in first arterial phase, 64 HU and 66 HU in second arterial phase, and 44 HU and 40 HU in venous phase, respectively.

 

Figure 2
View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b: Oblique sagittal reformatted CT images of 55-year-old man with T3 tumor of 5.5 cm in diameter in midthoracic portion (small arrows) and T3 tumor of 7.0 cm in diameter in upper thoracic portion (arrowheads). (a) First arterial phase image demonstrates tumors in upper and midthoracic portions. (b) Second arterial phase image displays an early attenuation of lesions and provides best conspicuity of tumor among the three phases. Lower edge of tumor is clearly identifiable on image (large arrow). (c) Venous phase image shows tumors with less conspicuous margins compared with second arterial phase image. Average attenuations of upper thoracic and midthoracic tumors are 83 HU and 89 HU in first arterial phase, 129 HU and 116 HU in second arterial phase, and 116 HU and 112 HU in venous phase, respectively. Tumor-to–normal esophageal wall attenuation differences are 28 HU and 34 HU in first arterial phase, 64 HU and 66 HU in second arterial phase, and 44 HU and 40 HU in venous phase, respectively.

 

Figure 2
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c: Oblique sagittal reformatted CT images of 55-year-old man with T3 tumor of 5.5 cm in diameter in midthoracic portion (small arrows) and T3 tumor of 7.0 cm in diameter in upper thoracic portion (arrowheads). (a) First arterial phase image demonstrates tumors in upper and midthoracic portions. (b) Second arterial phase image displays an early attenuation of lesions and provides best conspicuity of tumor among the three phases. Lower edge of tumor is clearly identifiable on image (large arrow). (c) Venous phase image shows tumors with less conspicuous margins compared with second arterial phase image. Average attenuations of upper thoracic and midthoracic tumors are 83 HU and 89 HU in first arterial phase, 129 HU and 116 HU in second arterial phase, and 116 HU and 112 HU in venous phase, respectively. Tumor-to–normal esophageal wall attenuation differences are 28 HU and 34 HU in first arterial phase, 64 HU and 66 HU in second arterial phase, and 44 HU and 40 HU in venous phase, respectively.

 
Quantitative Analysis
In 28 of 31 lesions, attenuation was measured. In the remaining three lesions, attenuation could not be measured because one lesion was not identifiable in all three phases and the remaining two lesions were too small to define ROIs within the tumors. The mean attenuation of the esophageal tumor was statistically highest in the second arterial phase (P < .01 compared with both the first arterial and the venous phases) (Table 3). With regard to the attenuation difference between the tumor and normal esophagus wall, the second arterial phase also showed significantly better conspicuity than both the first arterial and the venous phases (P < .01). The mean attenuation of the esophageal tumor showed a peak in the second arterial phase, while the attenuation of the normal esophageal wall tended to gradually enhance (Fig 3). The mean tumor-to–normal esophageal wall attenuation difference was greatest in the second arterial phase.


View this table:
[in this window]
[in a new window]

  		Table 3. Mean Attenuation of 28 Esophageal Tumors, Normal Esophageal Walls, and the Difference between Them

 

Figure 3
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3: Graph of mean attenuation of normal esophageal wall ({blacksquare}) and tumor ({diamondsuit}). Note that attenuation of tumor and tumor-to–normal esophageal wall attenuation differences ({blacktriangleup}) peak in second arterial phase, whereas attenuation of esophageal wall elevates gradually.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCE IN KNOWLEDGE
 References
 
At conventional CT with contrast material enhancement, images are acquired in the venous phase, and esophageal cancer can be demonstrated in approximately 80% of lesions (14). This frequency of demonstration is nearly equivalent to that in the venous phase in our study (24 of 31 [77%]). Missed cancers at CT may result from poor delineation of early stage cancers without obvious wall thickening. Yoon et al reported that 69.2% of T1 lesions were not demonstrated at CT (14). At conventional CT, identification of esophageal cancers usually depends on wall thickening of the esophagus (15). Early stage cancers, however, may show minimal or no wall thickening. Wu et al (16) proposed a classification of tumor staging: Cancers with thickness greater than 5 mm and less than 15 mm are considered modified T2, those with thickness greater than 15 mm and with an irregular outer margin are considered T3, and tumor invasion of adjacent organs is considered T4. However, Wu et al had no criteria for T1 lesions because T1 lesions cannot be delineated by the thickness of the wall alone.

Our results showed that with dynamic CT, the second arterial phase is the optimal phase for visualization. Cancers in the second arterial phase can be identified as enhanced foci, even if the cancer does not cause wall thickening. As for advanced stage cancers with wall thickening, the second arterial phase most clearly depicts the tumors. These results are also supported by quantitative analysis of the difference in attenuation between the tumor and normal esophageal wall. Esophageal cancer shows a peak of enhancement around the second arterial phase, whereas normal esophageal wall shows gradual enhancement. Therefore, the best conspicuity of esophageal cancer against the esophageal wall can be obtained at this time.

Although the venous phase images seem inferior to the second arterial phase images with regard to the representation of esophageal cancer, venous phase images may be useful in evaluating mediastinal lymphadenopathy. In addition, by analyzing the venous phase images of the esophagus, multi–detector row CT can allow rapid imaging of the abdomen for the purpose of screening distant metastases and lymphadenopathy. In the evaluation of liver metastasis, liver parenchyma is not enhanced enough at the arterial phases and may not be adequate for depicting lesions other than hypervascular tumors, such as hepatocellular carcinoma (17). The first arterial phase images may have no clinical value in the evaluation of esophageal cancer and metastatic lesions. Ideally, both the second arterial and the venous phase images are required for the evaluation of esophageal cancers; however, increasing the number of scans causes increased radiation dose.

Improved delineation of esophageal cancer at multi–detector row CT can provide several potential clinical benefits. Precise localization of esophageal cancer is useful for planning radiation therapy or surgical resection. Once chemotherapy or radiation therapy is chosen, clear delineation of the tumor as an enhanced lesion may be useful for evaluating the therapeutic outcome. Moreover, clear delineation of esophageal cancer may improve recognition of relationships to the adjacent mediastinal organs and thus improve accuracy of local staging of the tumors.

There were some limitations in our study. First, the number of adenocarcinomas and T1a cancers was very limited in our population. The only case of adenocarcinoma in our study was a mucosal cancer (T1a cancer), and it was not demonstrated at CT. T1a cancer may not be delineated, even with dynamic CT. In such instances, the presence of enhanced lesions within nonthickened walls may become a useful radiologic feature that allows the differentiation of T1b cancers from T1a cancers, although further evaluation is required with early stage cases. Esophageal adenocarcinomas of advanced stages may show radiologic features and enhancing patterns that are different from those of squamous cell carcinomas. Although the prevalence of adenocarcinomas in Japan is increasing, the proportion of adenocarcinomas to squamous cell carcinomas is still low and differs from that in the United States (18). Further evaluation is required to investigate whether this technique is applicable for patients with adenocarcinoma.

Second, our study population included only patients with esophageal cancer and did not consist of any healthy subjects because of radiation exposure. The readers were not blinded for the presence of the tumor. Hence, it is questionable whether this technique is useful in screening for esophageal cancer. In patients with inflammatory diseases, including gastroesophageal reflux disease or Barrett esophagus, CT appearances of these diseases may resemble early stage esophageal cancers (19). Distribution of inflammatory lesions, however, would be usually more superficial and diffuse than that of cancers.

Third, qualitative evaluation of CT images from all three phases was performed on the same day in our study, which may cause bias. We are convinced, however, that the results from quantitative analysis support reproducibility of qualitative analysis.

In conclusion, the results of our study show that the second arterial phase of dynamic CT is the optimal phase for visualization of esophageal cancer.


   ADVANCE IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCE IN KNOWLEDGE
References
 


   ACKNOWLEDGMENTS
 
The authors thank Hiroto Hatabu, MD, PhD, Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Mass, for his guidance in the preparation of the manuscript.


   FOOTNOTES
 

Abbreviations: ROI = region of interest

Authors stated no financial relationship to disclose.

Author contributions: Guarantor of integrity of entire study, T.K.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, S.U., T.K., K.T.; clinical studies, all authors; statistical analysis, S.U.; and manuscript editing, S.U., T.K., K.T., T.S., M.I.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCE IN KNOWLEDGE
 References
 

  1. Jemal A, Tiwari RC, Murray T, et al. Cancer statistics, 2004. CA Cancer J Clin 2004;54:8–29.[Abstract/Free Full Text]
  2. Levine MS. Esophageal cancer: radiologic diagnosis. Radiol Clin North Am 1997;35:265–279.[Medline]
  3. Kumbasar B. Carcinoma of esophagus: radiologic diagnosis and staging. Eur J Radiol 2002;42:170–180.[CrossRef][Medline]
  4. Rankin SC, Taylor H, Cook GJ, Mason R. Computed tomography and positron emission tomography in the pre-operative staging of oesophageal carcinoma. Clin Radiol 1998;53:659–665.[CrossRef][Medline]
  5. Whyte RI. Advances in the staging of intrathoracic malignancies. World J Surg 2001;25:167–173.[CrossRef][Medline]
  6. Quint LE, Glazer GM, Orringer MB, Gross BH. Esophageal carcinoma: CT findings. Radiology 1985;155:171–175.[Abstract/Free Full Text]
  7. Picus D, Balfe DM, Koehler RE, Roper CL, Owen JW. Computed tomography in the staging of esophageal carcinoma. Radiology 1983;146:433–438.[Abstract/Free Full Text]
  8. Vilgrain V, Mompoint D, Palazzo L, et al. Staging of esophageal carcinoma: comparison of results with endoscopic sonography and CT. AJR Am J Roentgenol 1990;155:277–281.[Abstract/Free Full Text]
  9. Lee JH, Jeong YK, Kim DH, et al. Two-phase helical CT for detection of early gastric carcinoma: importance of the mucosal phase for analysis of the abnormal mucosal layer. J Comput Assist Tomogr 2000;24:777–782.[CrossRef][Medline]
  10. Mani NB, Suri S, Gupta S, Wig JD. Two-phase dynamic contrast-enhanced computed tomography with water-filling method for staging of gastric carcinoma. Clin Imaging 2001;25:38–43.[CrossRef][Medline]
  11. Hundt W, Braunschweig R, Reiser M. Evaluation of spiral CT in staging of colon and rectum carcinoma. Eur Radiol 1999;9:78–84.[CrossRef][Medline]
  12. Dittler HJ, Pesarini AC, Siewert JR. Endoscopic classification of esophageal cancer: correlation with the T stage. Gastrointest Endosc 1992;38:662–668.[Medline]
  13. Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma: 2nd English edition. Gastric Cancer 1998;1:10–24.[Medline]
  14. Yoon YC, Lee KS, Shim YM, Kim BT, Kim K, Kim TS. Metastasis to regional lymph nodes in patients with esophageal squamous cell carcinoma: CT versus FDG PET for presurgical detection prospective study. Radiology 2003;227:764–770.[Abstract/Free Full Text]
  15. Koehler RE, Memel DS, Stanley RJ. Gastrointestinal tract. 3rd ed. In: Lee JKT, Sagel SS, Stanley RJ, Heiken JP, eds. Computed tomography with MRI correlation. Philadelphia, Pa: Lippincott-Raven, 1998; 637–700.
  16. Wu LF, Wang BZ, Feng JL, et al. Preoperative TN staging of esophageal cancer: comparison of miniprobe ultrasonography, spiral CT and MRI. World J Gastroenterol 2003;9:219–224.[Medline]
  17. Hori M, Murakami T, Kim T, Tomoda K, Nakamura H. CT scan and MRI in the differentiation of liver tumors. Dig Dis 2004;22:39–55.[CrossRef][Medline]
  18. Hongo M. Review article: Barrett's oesophagus and carcinoma in Japan. Aliment Pharmacol Ther 2004;20(suppl 8):50–54.
  19. Berkovich GY, Levine MS, Miller WT Jr. CT findings in patients with esophagitis. AJR Am J Roentgenol 2000;175:1431–1434.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2393050222v1
239/3/777    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Umeoka, S.
Right arrow Articles by Imamura, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Umeoka, S.
Right arrow Articles by Imamura, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE