| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Thoracic Imaging |
1 From the Department of Thoracic Oncology, National Cancer Center Hospital East, 6-5-1 Kashiwa-no-ha, Kashiwa 277-8577, Japan (R.K., H.O., K.N., Y.N.); the Departments of Endoscopy (M.K.), Surgery (T.N.), and Radiology (N.M.), National Cancer Center Hospital, Tokyo, Japan; the National Shikoku Cancer Center Hospital, Matsuyama, Japan (K.E.); and the Japan Anti-Tuberculosis Association, Shizuoka Branch, Japan (A.S.). Received April 3, 1998; revision requested June 5; revision received September 16; accepted December 9. Supported in part by a grant-in-aid for cancer research from the Ministry of Health and Welfare, by a grant for scientific research expenses for health and welfare programs from the Foundation for the Promotion of Cancer Research, and by the New 10-year Strategy for Cancer Control (researching new technology for diagnostic purposes). Address reprint requests to R.K. (e-mail: rkaki@east.ncc.go.jp).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: The findings from seven patients with lung cancer that was missed at the initial spiral CT screening were reviewed. Retrospective CT findings, time to detection, cell type, and pathologic stage were evaluated.
RESULTS: Minute lung cancers missed at early spiral CT included a nodule among the shadows of old tuberculosis (n = 2), a faint nodule with high attenuation in the center of the nodule (n = 1), an increase in attenuation just adjacent to an axial peripheral pulmonary vessel (n = 1) and adjacent to a craniocaudal peripheral pulmonary vessel (n = 1), and a minute faint nodule (n = 2). The time to detection ranged from 6 to 18 months. At pathologic examination, six cancers were stage I, and one was stage II.
CONCLUSION: Minute nodules of lung cancer that are near the threshold of detectability may be missed at spiral CT screening. It is important to examine noncalcified nodules with thin-section CT even when lesions from prior disease, such as those from old tuberculosis, exist and to evaluate the shadows of pulmonary vessels carefully. A follow-up examination is highly recommended.
Index terms: Cancer screening, 60.321 • Diagnostic radiology, observer performance, 60.321 • Lung neoplasms, CT, 60.12115, 60.12118 • Lung neoplasms, diagnosis, 60.3211, 60.3212
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The purpose of this study was to clarify the CT findings and the progression of minute lung cancers that were missed at early spiral CT screening, thereby improving the diagnostic criteria for detecting lung cancer.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The CT screening system used was a TCT-900S Superhelix (Toshiba Medical, Tokyo, Japan). The scanning parameters were 120 kVp, 50 mA, 10-mm collimation, one rotation of the x-ray tube per second, and a table speed of 20 mm/sec (pitch, 2:1). A thoracic area spanning 30 cm was scanned from a level 2 cm superior to the apex to the level of the diaphragm with a 15-second breath hold. Image reconstruction was performed with 180° linear interpolation at intervals of 1 cm.
All CT images were reviewed independently by four diagnostic experts (M.K., K.E., H.O., R.K.) who used both the cathode-ray–tube monitor and the hard-copy display. Display conditions were a window width of 2,000 HU and a window level of -700 HU. As part of the screening protocol, the diagnostic experts classified the images into four categories: (a) suspected lung cancer, (b) a benign tumor or an active inflammatory disease, (c) a scar lesion caused by a previous inflammatory episode, and (d) normal.
Thin-section CT was performed with the same CT scanner whenever lesions seen on CT scans, radiographs, or both were suggestive of lung cancer, or whenever the lesion seen was a solitary nodule without calcification, a nodule larger than 4.9 mm in diameter, or an area of localized opacification increasing in size with sequential comparison. The scanning parameters for thin-section CT were 120 kVp, 250 mA, 2-mm collimation, one rotation of the x-ray tube per second, a table speed of 2 mm/sec (pitch, 1:1), and a thin-section algorithm. Areas of opacification with calcification, multiple lesions confined to the same segment, linear shadows of focal collapsed lung, and shadows without any progressive change were considered benign (1).
Four authors (R.K., H.O., M.K., K.E.) together reviewed the CT images of the 22 patients with lung cancer to determine retrospectively when the cancers were first visible. Consensus opinion was used. Retrospective CT findings, the time to detection, and the pathologic stage of the lung cancer at detection were evaluated. The diameters of the nodules at the initial screening and at detection were assessed from the standpoint of conspicuity.
We classified nodules into two categories: conspicuous and inconspicuous. Conspicuous nodules were easily seen because they were well defined and had high attenuation. Inconspicuous nodules were not only smaller but also less well defined and had low attenuation. The diameters of the nodules were measured with a built-in program in the CT console by using digital data from the screening CT image stored on optical disks.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The clinical and imaging characteristics of the seven patients are listed in the Table. In reviewing the original diagnoses from initial CT images, the nodule in patient 1 (Fig 1) was misdiagnosed as a lesion of old pulmonary tuberculosis because of an area of opacification that resembled calcification and pleural thickening. In patient 2, the nodule was also diagnosed as old pulmonary tuberculosis because of the coexistence of multiple lesions confined to the same segment and areas of opacification with calcification. In patient 3, the nodule was diagnosed as a granuloma (Fig 2) because of its high attenuation and relatively smooth edge.
|
|
|
|
|
|
|
|
|
|
|
|
|
The diagnoses at detection of lung cancer were classified into two groups. In the first group (patients 1 and 4–6), two diagnostic experts agreed that a nodule was highly suggestive of lung cancer. In the second group (patients 2, 3, and 7), one diagnostic expert diagnosed each nodule as highly suggestive of lung cancer, but the other diagnostic expert diagnosed the findings in patients 2 and 3 as scar lesions caused by previous inflammatory episodes and failed to detect the lesion in patient 7. In the second group, two lesions (in patients 2 and 7) were diagnosed on the cathode-ray–tube monitor as highly suggestive of lung cancer, and one lesion (in patient 3) was so diagnosed on the hard-copy display.
At the time of detection, the pathologic stages of the missed lung cancers were as follows: the six cancers detected only with spiral CT (ie, five adenocarcinomas [patients 1–3, 6, and 7] [Figs 1, 2] and one squamous cell carcinoma [patient 4]) were stage I, and the second squamous cell carcinoma (patient 5) (Fig 3), which was detected with frontal chest radiography and spiral CT, was stage II.
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Saida et al (12) reported that more than half of the early squamous cell carcinoma lesions that showed polypoid growth or that reached the cartilaginous layer of the bronchial wall could be seen at CT with 5.0-mm, 3.0-mm, or 1.5-mm sections. However, in our study of spiral CT screening, we were not able to acquire images of peripheral bronchi in detail because the pitch was 2:1 and because image reconstruction was performed at intervals of 1 cm. We thought that in its early stages, the spread of squamous cell carcinoma to the outside of the bronchial wall looked like a slight increase in attenuation just adjacent to the peripheral pulmonary vessel, which led to the missed diagnosis of early squamous cell carcinoma. In patients 1 and 2, the reason why adenocarcinoma was missed was probably the presence of shadows of old pulmonary tuberculosis.
White et al (6) also reported that the presence of unrelated major abnormalities on CT images may contribute to diagnostic failure. In patient 3, the reason why the malignant nodule was missed was its close resemblance to a minute granuloma. However, the follow-up CT examinations revealed an increase in attenuation around the granuloma. The area of increased attenuation appeared as an area of ground-glass opacity at thin-section CT.
Pathologic examination of the specimen from patient 3 revealed that the granuloma-like appearance was caused by a fibrotic focus. In patient 3, this minute lung cancer already had fibrosis in the nodule. Jang et al (13) reported that focal areas of ground-glass opacity on thin-section CT images could be an early sign of localized bronchioloalveolar carcinoma. Yamada et al (14) reported that conventional and thin-section CT revealed "air-density lesions" in each of five resected minute lung cancers that measured less than 6 mm in diameter. Air density is a synonym for ground-glass opacity.
In our patients with inconspicuous adenocarcinoma (patients 6 and 7), differences between localized faint increased attenuation and the attenuation of the surrounding lung were not clear enough to permit detection of these lesions with any degree of certainty. Furthermore, if these lesions were detected, it would be difficult to differentiate these lesions from other causes of a localized faint increase in attenuation, such as localized fibrosis, localized emphysema, or artifacts of spiral CT. Artifacts on spiral CT images are faint trapezoid-like increases in attenuation just inside the ribs and faint round or oval increases in attenuation between pulmonary vessels that run parallel.
Gurney (7) reported that all peripheral tumors that were missed were smaller than 3 mm in diameter and that such failure of detection was attributable primarily to the poor conspicuity of lesions. Naidich et al (15) reported that observers identified 1%, 48%, 82%, and 91% of the nodules that were less than 1.5 mm, 3.0 mm, 4.5 mm, and 7.0 mm in diameter, respectively, on 1.5-mm sections in a three-dimensional computer simulation. From these results, Gurney (7) proposed that 3 mm might be a threshold size at which an observer had an approximately even chance of detecting a nodule.
However, of the lung cancers that were missed in our study, the diameter of conspicuous nodules was at least 9 mm, whereas the diameter of inconspicuous nodules was at most 7 mm. In our initial experience with the use of spiral CT screening, a threshold size at which radiologists had an approximately even chance of detecting a nodule was between 7 and 9 mm.
Davis (16) suggested that the frequency of missed diagnoses cannot be accurately determined without prospective analysis but is undoubtedly higher than the 0.045% rate (14 patients underwent 17 examinations out of a total of 37,500 CT examinations) that can be estimated from the results of the study by White et al (6). In retrospect, as of April 1996, our results showed that the frequency of missed cancers during spiral CT screening was 0.26% (seven patients with missed lung cancer underwent a total of 14 CT examinations out of 5,418 CT examinations).
Despite the fact that lung cancer was missed in seven patients by using spiral CT screening, six of these cancers were pathologic stage I at detection, and one was pathologic stage II. Therefore, these cancers had the potential for cure.
At spiral CT, the pitch used was 2:1 (ie, 10-mm collimation and a table speed of 20 mm/sec), and the image reconstruction was performed at intervals of 1 cm. We previously evaluated the feasibility of using a 2.5:1 pitch on solitary and metastatic pulmonary nodules (17). Mori et al (18) reported that the optimal table speed and reconstruction image interval for helical CT that are needed to keep the detection sensitivity comparable with that of conventional CT are 20 mm/sec and 10-mm images. However, Wright et al (19) reported that the risk of understaging disease in patients with solitary nodules is greater with increasing pitch, and pitch should therefore be limited to no greater than 1.5 for initial staging of pulmonary metastatic disease.
Buckley et al (20) compared spiral CT with interscan spacing of 4 to 5 mm versus spacing of 8 to 10 mm with regard to the detection rate and the level of confidence in the diagnosis of pulmonary nodules. These investigators (20) concluded that increased reconstruction frequency of spiral CT volume data sets improves the detection of pulmonary nodules and enhances confidence in the diagnosis.
With regard to reading methods, we read static images on the cathode-ray–tube monitor and film display. Seltzer et al (21) compared cine viewing with film-based viewing at spiral CT of the chest and concluded that cine viewing of spiral CT images of the chest improves the ability of the radiologist to detect nodules.
Given these preliminary results, further research is needed to determine the most appropriate conditions for spiral CT screening, including pitch, viewing method, and reconstruction interval. We are now developing a computer-aided lung cancer diagnosis system that is based on spiral CT images (22,23) as a counterpart to the double-reading technique now used in our spiral CT screening program.
In conclusion, minute nodules of lung cancer that are near the threshold of detectability may be overlooked at spiral CT screening. It is important to examine noncalcified nodules with thin-section CT even when prior lesions, such as those of old tuberculosis, exist. It is also important to evaluate the shadows of pulmonary vessels carefully. To establish appropriate diagnostic criteria for minute lung cancer, we need to analyze the progression of many such minute lung cancers. Therefore, follow-up CT examination is highly recommended.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Author contributions: Guarantors of integrity of entire study, R.K., H.O., M.K., K.E., T.N., A.S., N.M.; study concepts and design, R.K., H.O., M.K., K.E.; literature research, R.K.; clinical studies, R.K., H.O., M.K., K.E.; data acquisition and analysis, R.K., H.O., M.K., K.E.; manuscript preparation, R.K., H.O., M.K., K.E.; manuscript editing, H.O., M.K., K.E.; manuscript review, K.N., Y.N.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Girvin and J. P. Ko Pulmonary Nodules: Detection, Assessment, and CAD Am. J. Roentgenol., October 1, 2008; 191(4): 1057 - 1069. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U Shaikh and H. Lewis-Jones VOMIT victim of medical investigative technology BMJ, January 5, 2008; 336(7634): 8 - 8. [Full Text] [PDF]
|
|
|
|
 |
|
 C. Wittram, A. C. Waltman, J.-A. O. Shepard, E. Halpern, and L. R. Goodman Discordance between CT and Angiography in the PIOPED II Study Radiology, September 1, 2007; 244(3): 883 - 889. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Das, G. Muhlenbruch, A. H. Mahnken, T. G. Flohr, L. Gundel, S. Stanzel, T. Kraus, R. W. Gunther, and J. E. Wildberger Small Pulmonary Nodules: Effect of Two Computer-aided Detection Systems on Radiologist Performance. Radiology, November 1, 2006; 241(2): 564 - 571. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Haller, C. Kaiser, P. Buser, G. Bongartz, and J. Bremerich Coronary artery imaging with contrast-enhanced MDCT: extracardiac findings. Am. J. Roentgenol., July 1, 2006; 187(1): 105 - 110. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Lionakis, J. Lahdenranta, J. Sun, W. Liu, R. E. Lewis, N. D. Albert, R. Pasqualini, W. Arap, and D. P. Kontoyiannis Development of a Ligand-Directed Approach To Study the Pathogenesis of Invasive Aspergillosis Infect. Immun., November 1, 2005; 73(11): 7747 - 7758. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Li, H. Arimura, K. Suzuki, J. Shiraishi, Q. Li, H. Abe, R. Engelmann, S. Sone, H. MacMahon, and K. Doi Computer-aided Detection of Peripheral Lung Cancers Missed at CT: ROC Analyses without and with Localization Radiology, November 1, 2005; 237(2): 684 - 690. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Peldschus, P. Herzog, S. A. Wood, J. I. Cheema, P. Costello, and U. J. Schoepf Computer-Aided Diagnosis as a Second Reader: Spectrum of Findings in CT Studies of the Chest Interpreted as Normal Chest, September 1, 2005; 128(3): 1517 - 1523. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. D. Rubin, J. K. Lyo, D. S. Paik, A. J. Sherbondy, L. C. Chow, A. N. Leung, R. Mindelzun, P. K. Schraedley-Desmond, S. E. Zinck, D. P. Naidich, et al. Pulmonary Nodules on Multi-Detector Row CT Scans: Performance Comparison of Radiologists and Computer-aided Detection Radiology, January 1, 2005; 234(1): 274 - 283. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G. Armato III, G. McLennan, M. F. McNitt-Gray, C. R. Meyer, D. Yankelevitz, D. R. Aberle, C. I. Henschke, E. A. Hoffman, E. A. Kazerooni, H. MacMahon, et al. Lung Image Database Consortium: Developing a Resource for the Medical Imaging Research Community Radiology, September 1, 2004; 232(3): 739 - 748. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-H. Yu, W.-H. Hsu, N.-Y. Hsu, T.-Y. Lin, T.-C. Hsia, S.-S. Sun, and C.-H. Kao The Use of Dual Phase 201Tl SPECT for Differentiating Pulmonary Malignancies from Benign Lesions Jpn. J. Clin. Oncol., August 1, 2004; 34(8): 445 - 451. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Berlin Potential Legal Ramifications of Whole-Body CT Screening: Taking a Peek into Pandora's Box Am. J. Roentgenol., February 1, 2003; 180(2): 317 - 322. [Full Text] [PDF]
|
|
|
|
|
|
P. B. Bach, M. J. Kelley, R. C. Tate, and D. C. McCrory Screening for Lung Cancer: A Review of the Current Literature Chest, January 1, 2003; 123(1_suppl): 72S - 82S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Li, S. Sone, H. Abe, H. MacMahon, S. G. Armato III, and K. Doi Lung Cancers Missed at Low-Dose Helical CT Screening in a General Population: Comparison of Clinical, Histopathologic, and Imaging Findings Radiology, December 1, 2002; 225(3): 673 - 683. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Garg, R. L. Keith, T. Byers, K. Kelly, A. L. Kerzner, D. A. Lynch, and Y. E. Miller Randomized Controlled Trial with Low-Dose Spiral CT for Lung Cancer Screening: Feasibility Study and Preliminary Results Radiology, November 1, 2002; 225(2): 506 - 510. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Gruden, S. Ouanounou, S. Tigges, S. D. Norris, and T. S. Klausner Incremental Benefit of Maximum-Intensity-Projection Images on Observer Detection of Small Pulmonary Nodules Revealed by Multidetector CT Am. J. Roentgenol., July 1, 2002; 179(1): 149 - 157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sobue, N. Moriyama, M. Kaneko, M. Kusumoto, T. Kobayashi, R. Tsuchiya, R. Kakinuma, H. Ohmatsu, K. Nagai, H. Nishiyama, et al. Screening for Lung Cancer With Low-Dose Helical Computed Tomography: Anti-Lung Cancer Association Project J. Clin. Oncol., February 15, 2002; 20(4): 911 - 920. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. SWENSEN, J. R. JETT, J. A. SLOAN, D. E. MIDTHUN, T. E. HARTMAN, A.-M. SYKES, G. L. AUGHENBAUGH, F. E. ZINK, S. L. HILLMAN, G. R. NOETZEL, et al. Screening for Lung Cancer with Low-Dose Spiral Computed Tomography Am. J. Respir. Crit. Care Med., February 15, 2002; 165(4): 508 - 513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Schaefer-Prokop and M. Prokop New imaging techniques in the treatment guidelines for lung cancer Eur. Respir. J., February 1, 2002; 19(35_suppl): 71S - 83s. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.J. van Klaveren, J.D.F. Habbema, J.H. Pedersen, H.J. de Koning, M. Oudkerk, and H.C. Hoogsteden Lung cancer screening by low-dose spiral computed tomography Eur. Respir. J., November 1, 2001; 18(5): 857 - 866. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. R. Hirsch, W. A. Franklin, A. F. Gazdar, and P. A. Bunn Jr. Early Detection of Lung Cancer: Clinical Perspectives of Recent Advances in Biology and Radiology Clin. Cancer Res., January 1, 2001; 7(1): 5 - 22. [Abstract] [Full Text]
|
|
|
|
|
|
J. P. Ko and M. Betke Chest CT: Automated Nodule Detection and Assessment of Change over Time--Preliminary Experience Radiology, January 1, 2001; 218(1): 267 - 273. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
