Currently, lung carcinoma is the malignant tumor with the highest incidence (11.6% of all cases) and mortality (18.4% of all cancer deaths),1–4 because most patients with lung carcinoma are in intermediate and advanced stages at the time of treatment.5,6 Surgical resection for early lung carcinoma is the most effective therapy,5,7 the overall postoperative 5-year survival rate and 5-year relapse-free survival rate for which may reach 80%.8,9 The wide use of computed tomography (CT) imaging in the clinic has largely increased the detection rate of small lung nodules,10,11 and the lobulation and spiculation are common findings on imaging to differentiate benign or malignant nodules. However, these 2 signs have little effects on the prediction of staging and prognosis of malignant nodules. Pleural tail sign is a finding that has been observed between a nodule and the pleura for a long time. A line of tail sign may be single or multiple, and vary in length and thickness. Earlier researchers often judged whether lung cancers invaded the pleura according to the morphological features and presence of the pleural tail sign.12,13 Studies have shown that this sign is valuable for detection of pleural invasion. After maximal intensity projection (MIP) reconstruction, the pleural tail sign can be shown as a 5-line sign.14 On the basis of previous observations, this article attempts to judge the invasiveness of peripheral lung cancer according to the clarity of the 5-line sign on MIP images and to predict the progression and prognosis of the cancer.
MATERIALS AND METHODS
This study was approved by the institutional review boards First Affiliated Hospital of Hainan Medical College and Haikou People's Hospital with a waiver of informed consent.
One hundred fifty-one cases confirmed as peripheral lung carcinoma between December 2007 and December 2014 were collected. Among these cases, 110 were from the First Affiliated Hospital of Hainan Medical University (between December 2007 and December 2014); 41 cases were from Haikou People's Hospital (between February 2012 and December 2014). Inclusion criteria were the following: (1) the lung neoplasm was first discovered with definite pathological diagnosis; (2) there was no history of other tumors in the recent 5 years; (3) cases underwent a plain scan; and (4) the slice thickness of the original scan image was 1 or 1.25 mm, and the image was reconstructed with a standard algorithm. According to the inclusion criteria, 19 cases were excluded. One hundred thirty-two cases were included (93 men and 39 women with average age of 61.37 ± 11.25 years and an age range of 27–82 years). The average size of nodules/lumps was 3.22 ± 1.60 cm (from 0.98 to 8.75 cm). There were 133 nodules (masses) in total (2 nodules existed in left and right lung, respectively, of 1 patient). Among the nodules, 4 were partially solid nodules, and the others were solid nodules (or masses). According to the pathological diagnosis, 93 cases were adenocarcinoma, 16 cases were squamous carcinoma, 5 cases were large cell carcinoma, 2 cases were small cell neuroendocrine carcinoma, 1 case was carcinosarcoma, 1 case was mucoepidermoid carcinoma, and 15 cases were unspecified carcinoma. The pathological information comes from various ways (Table 1).
A GE (GE Healthcare, Chicago, Illinois) LightSpeed volume CT scanner with an AW 4.3 workstation was used with the following scanning parameters: 120 kV; 89 to 301 mAs (with tube current modulation active); reconstructive slice thickness, 1.25 mm; interval, 1.25 mm; and slab thickness, 6.25 mm for axial MIP. The parameters for the Philips (Philips Healthcare, Amsterdam, Holland), Brilliance iCT (Cleveland, Ohio), and Siemens (Siemens Healthcare, Erlangen, Germany) Somatom Definition Flash (Munich, Germany) were the following: 120 kV; 180 to 200 mAs (with tube current modulation active on the 16-slice CT unit); rebuilding slice thickness of 1.0 mm; space, 1.0 mm; and slab thickness of 5.0 mm for axial MIP, using a Philips workstation (Brilliance Workspace Portal) for both MIPs rebuilding based on thin section image from either Siemens or Philips.
The T, N, and M staging in this article is based on the TNM stages of lung cancer from the International Union Against Cancer and the American Joint Committee on Cancer, seventh edition, 2009; pieces of evidence of the imaging and/or pathological data were allowed for staging. When both imaging and pathological data were available, the pathological data were adopted.
T staging was based on the size of the nodule (mass) measured on the imaging.
N staging was based on pieces of evidence of pathology, or image if no pathological data. On the image, a short diameter of the axial lymph node of ≥10 mm was judged to be lymph node enlargement.15
For M staging, distant metastasis was determined by imaging data (metastases of the lung and other organs) and pathological data (positive pleural effusion and metastatic tumors). No metastasis was found in M0. Metastases were classified into M1a and M1b according to the seventh edition of TNM staging of lung cancer in 2009 by the International Union Against Cancer and the American Joint Committee on Cancer.
Diameter Measurements of the Lung Nodules (Masses)
The axial lung window with a slice thickness of ≤5 mm was taken, showing the maximal nodule section. The long diameter was measured on a single CT image that showed the maximal length of a nodule; the wide diameter, defined as the longest perpendicular to the length, was measured on the same CT image. The average of the long and wide diameters was the diameter of the lung nodule (mass).
Grading Method of 5-Line Signs
The single line of the normal interlobar fissure shown on a thin section image can be reconstructed as a 5-line sign on axial MIP.16 The line between the lung nodule and the pleura is called the pleural tail sign on thin image. On the axial MIP, it can also be reconstructed as the 5-line sign or fewer than 5 lines.14 The pleural tail sign can derive from the margin of a nodule as single or multiple lines extending in different directions and be observed in axial, coronary, or sagittal positions, on which the longest line is chosen as a baseline to reconstruct a 5-line sign on axis position in this research (Fig. 1).
Maximal intensity projection reconstruction of all cases was performed by one doctor A. Y. with more than 20 years of experience in chest CT diagnosis. If there was an unclear or no 5-line sign at the edge of the nodule on the axial MIP, rotating the reconstruction tangent lines clockwise or counterclockwise on the sagittal or coronal plane of the workstation interface was tried, to observe whether the 5-line sign would appear or improve the clarity after adjusting the angle,14 and improved sharpness image was retained for grading.
Five-line signs on the margin of the lung nodule (mass) were classified into 5 grades according to their sharpness (Fig. 2), as follows:
0: There were no 5-line signs. On the axial, coronal, and sagittal images, the nodule (mass) showed no line shadow between the nodule margin and pleura and no adhesion between the nodule (mass) and pleura.
- 1: 5-line sign disappeared. A band shadow was connected to the pleura, and/or multiple adhesions between the nodule (mass) and the pleura were observed.
- 2: Most 5-line sign disappeared. (a) The reconstructed line shadow only showed 2 or 3 lines. (b) The 5 lines were partly fused as high-density shadow, or disappeared (low density); the length of the changed part was more than half of the whole line.
- 3: The partial 5-line sign disappeared. (a) The number of the main 5 lines was reduced to 4 lines. (b) The merging or disappearing part of the main 5 lines was larger than one-quarter of the line length and less than half of the line length.
- 4: The 5-line signs were clear or relatively so. Clear: The lines of the 5 lines were uniform in line thickness and clear in space. Relatively clear: The lines of the 5 lines were not so uniform in line thickness and were relatively obscure in space; the unclear part was equal to or shorter than one-quarter of the 5-line length.
When nodules were small in the early stage of tumors, the length of axial 5-line sign was usually greater than or equal to the diameter of nodules. As lung cancer nodules grew and gradually approached the pleura, the pleural tail sign and the axis 5-line sign based on it also became short. When the length of the axis 5-line sign is less than the diameter of the nodule (mass) but longer than one-half of that of the tumor, the grade of the 5-line sign is reduced from grade 4 to grade 3, even it is clarity. If the length of the 5-line sign is less than one-half of the diameter of the tumor, it should be reduced to grade 2.
The SPSS 16.0 software package was used for statistical analysis. An ordered logistic regression model was used to incorporate tumor size and length of tail sign with the grading of 5-line sign to predict T, N, and TNM staging. Receiver operating characteristic (ROC) curve analysis was used to evaluate the value of 5-line sign grading in predicting the progression and prognosis of lung cancer. It was considered that the diagnostic accuracy was low if the area under the ROC curve was less than 0.7, moderate if it was 0.7 to 0.9, and high if that was more than 0.9. P < 0.05 indicated that the results had statistical significance.
Correlation Between 5-Line Signs and Tumor TNM Staging
Three cases of grade 0 of 5-line sign were not included in the statistics because there were not enough cases. One case had 2 nodules located, respectively, in the left lung (part-solid nodule) with grade 4 of 5-line sign and in the right lung (solid nodule) with grade 2 of 5-line sign. Based on the pathological diagnosis from needle biopsy, the nodule in the left lung was considered a low-grade malignant mucoepidermoid carcinoma, and the nodule in the right lung was a papillary adenocarcinoma, which was high invasion.7 The purpose of this article is to evaluate the effect of 5-line sign on assessing lung carcinoma progression. So only the nodule in the right lung with a lower grade of 5-line sign was selected to enter in the data analysis, as the tumor was enlarged and metastasized after 20 months of follow-up. Meanwhile, the nodule in the left lung had the same size as it was in the initial examination. Thus, 129 nodules are included in Tables 2 and 3.
Ordered logistic regression analysis showed that destruction of the 5-line sign was a risk factor for the increase of N stage (“N0 to N1–N3” or “N0–N2 to N3”) and TNM stage (“I A to I B–IV” or “I A–III B to IV”), but it had no effect on T stage and M stage. Tumor enlargement was a risk factor for T, N, and TNM staging, but its size did not affect M staging. The length of pleural tail sign was not a factor influencing T, N, M, and TNM stages of lung cancer (Tables 2, 3).
Small lung nodules (≤3 cm) with 5-line sign 4 or 3 grade showed negative metastasis (Figs. 2E, 3, 4), but those with 5-line sign 2 or 1 grade showed various metastases (Table 4; Figs. 5, 6).
In lung lumps larger than 3 cm in diameter, 5-line often showed different destructions, and more metastasis were seen (Table 5, Fig. 2A).
Correlation of the 5-Line Signs and Lung Carcinoma Progression
Sixty-three patients in 132 cases were followed up by imaging examination for 2 years and 10 months to 6 years (Table 6). Other cases did not get follow-up data for various reasons.
The ROC curve showed that 5-line sign grading and N stage were valuable (medium to high) on predicting progression of lung cancer. The sensitivity and specificity were 0.563 and 0.933 for 5-line sign grading and 0.646 and 0.967 for N stage. We noted that such a trend also existed in small lung nodules (≤3 cm; Fig. 7). That means the lower the sharpness of the 5-line signs, the greater the tumor progression. In this group, 5-line sign and N staging showed a better predictive value than did TNM staging (moderate) and T staging (moderate to low).
Correlation Between 5-Line Signs and Tumor Prognosis
A statistical analysis was conducted in the 53 cases with follow-up data for the prognosis of lung carcinoma. The remaining 72 cases were not included because they had no survival information (Table 7).
The statistical analyses on prognosis of lung carcinoma were performed for a period of 3 years. According to the ROC curve, the prognostic value of 5-line sign grading for lung cancer was moderately low, with sensitivity and specificity of 0.611 and 0.824, respectively (Fig. 8). N staging and TNM staging were not so valuable (moderately high).
The axial MIP reconstruction of normal pleura may form 5-line signs. The number of lines is equal to the ratio of MIP thickness and original slice thickness. This phenomenon may be the same as the beaded manifestations shown by the MIP reconstruction of the small vessels of the lung.14,16 This phenomenon of beaded vessels (Fig. 1C), which was first described as stair-step vessels by Napel et al,17 results when smaller vessels, particularly those with dimensions that approximate the section width, pass obliquely through the scan plane because of partial volume averaging. As a result, artifacts are generated (Fig. 1B).
The pleural tail sign18,19 represented thickened septae histologically.18,20,21 The 5-line sign was formed by MIP reconstruction on the basis of the pleural tail sign at the margin of the tumor. The thickness of pleural tail sign rather than its length is the most important condition for the formation of a 5-line sign. In this study, a small number of nodules with the tail sign on thin section images showed a clear 5-line sign by MIP rebuilding, suggesting that it was a relatively smooth and regular plane,12,14,20 which was difficult to be determined whether it is smooth and regular by conventional multiplanar or thin-section reconstruction. Carefully observed, these lines were thicker and denser than those of normal interlobular fissures (Fig. 2D). We observed that the 5-line sign on MIP showed to be unclear or disappearing in most nodules; we speculated that was for more thickened, rough and irregular tail sign,14,18,20 which reflected tumor's damage and invasion of the septa12,18,20,21 (Figs. 4, 6), whereas disappearance (low density) of the 5-line sign may be due to local edema20 (Fig. 2C).
By reconstruction, the pleura tail sign showed a 5-line sign on 6.25-mm MIP thickness; that is, the pleural tail is seen on 5 contiguous slices of 1.25 mm; in preexploration by the GE CT scanner, 3.75- and 8.75-mm MIPs were used to produce multilines. Three lines by 3.75-mm MIP and 7 lines by 8.75-mm MIP were produced. Relative to 5-line sign from 6.25-mm MIP, 3-line sign was clearer in grading but may cover up some clue, whereas 7-line sign was difficult to analyze for too much lines.
There was no correlation between the length of tail sign and T, N, and TNM staging, suggesting that only the length provided too few information on tumor invasion. Based on the artificial image of pleural tail sign, 5-line sign presented more information for analysis, such as the number of lines, and the clarity and length and width of 5-line sign (Figs. 3A–E). By comprehensive analyzing these phenomena, a prediction of tumor progression can be made.
Multivariate logical regression statistical analysis showed that tumor size positively correlated with T, N, and TNM staging of the tumor in this group. Although 5-line sign had no relation to T staging, it positively correlated with N and TNM staging.
Up to now, tumor size is still regarded as a key factor for evaluating lung cancer staging.22 However, Ohta et al23 considered that tumor size was not a reliable guide for limited surgical intervention for possibility of micro–lymph node metastasis and pleural invasion in a small nodule, especially in small adenocarcinoma (≤2 cm).23–25 The data from this group showed that a small nodule with good 5-line sign (4 or 3 grade) had no any metastasis, but those with bad 5-line sign (2 or 1 grade) were confirmed to have metastasis by various ways, most of that were lymph node metastasis, which implied the significance of 5-line clarity on predicting metastasis in a small malignant lung nodule. A positive correlation also existed between 5-line sign and TNM staging, which indicated a dominant position of N staging in TNM staging.
Multivariable data analysis of the ROC curve showed that the sensitivity and specificity of 5-line sign in predicting tumor progression were 56.3% and 93.3%, respectively. It was also effective in small lung malignant nodule (<3 cm). As a good example, 1 patient in this study had 2 nodules with different grades of 5-line signs in the left (4 grade) and right (2 grade) lungs. Follow-up showed an indolent nodule in the left lung but active growth in the right one, which implied that the lower the sharpness of the 5-line signs, the faster the tumor progression (Fig. 4).
Five-line sign was correlated with prognosis of tumors to some degree (sensitivity, 0.611; specificity, 0.824), but N staging was valuable (sensitivity, 0.778; specificity, 0.941). N staging is considered as the most important factor in lung cancer prognosis.25–27 Five-line sign was useful in predicting lymph node metastasis, so it was indirectly helpful in predicting a patient's prognosis.
From our data, beside tumor size, ground-glass opacity (GGO) component, and other ways,28–30 5-line sign might be another parameter in predicting metastasis, progression, and prognosis of lung carcinoma, particularly in small lung carcinoma (≤3 cm).
The GGO component in a nodule is an important index of prognosis of lung carcinoma,31,32 but it cannot predict the prognosis of solid lung carcinoma. In our group, most small lung carcinomas were solid nodules (41/43), so GGO is useless to these cases. Five-line sign might offset this shortage.
Limitations of the Study
The final survival of a great many cases in this study was not followed up because of various reasons. A 3-year follow-up is not sufficiently long. The samples falling into grades 3 and 4 of 5-line signs are not enough, so the correlation between 5-line signs and lung carcinoma staging, progression, and prognosis has to be proven by using larger samples. Unlike interlobular fissure, the extending direction of pleural tail signs at the margin of the tumor is random, so it is not comprehensive enough to observe the 5-line signs only by rotating the reconstructed tangent line at the coronal or sagittal position on the workstation interface. Forty-seven cases were staged according to tumor biopsy pathology and did not undergo surgery because of the patients' option. Their N staging was performed based on imaging manifestations.
The MIP-rendered 5-line signs at the margin of peripheral lung carcinomas may be a partial volume effect. The sharpness of the 5-line signs might have an effect on the prediction of invasion, progression, and prognosis of lung carcinomas, especially on small (<3 cm) lung carcinoma.
We deeply express our gratitude to Jiangjun Qin and Beihai Wu for advises and suggestions, as well as Baozhong Wu, Miaoqiong Zheng, Anlun Yu, and Zhongxin Hu for contributions to this work.
1. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin
2. Jemal A, Bray F, Center MM, et al. Global cancer statistics. CA Cancer J Clin
3. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA Cancer J Clin
4. Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics, 2002. CA Cancer J Clin
5. Kauczor HU, Bonomo L, Gaga M, et al. ESR/ERS white paper on lung cancer screening. Eur Respir J
6. Bach PB, Mirkin JN, Oliver TK, et al. Benefits and harms of CT screening for lung cancer: a systematic review. JAMA
7. Travis WD, Brambilla E, Noguchi M, et al. International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society International Multidisciplinary Classification of Lung Adenocarcinoma. J Thorac Oncol
8. Watanabe H, Saito H, Yokose T, et al. Relation between thin-section computed tomography and clinical findings of mucinous adenocarcinoma. Ann Thorac Surg
9. Shi CL, Zhang XY, Han BH, et al. A clinicopathological study of resected non–small cell lung cancers 2 cm or less in diameter: a prognostic assessment. Med Oncol
10. International Early Lung Cancer Action Program Investigators, Henschke CI, Yankelevitz DF, et al. Survival of patients with stage I lung cancer detected on CT screening. N Engl J Med
11. Aberle DR, DeMello S, Berg CD, et al. Results of the two incidence screenings in the National Lung Screening Trial. N Engl J Med
12. Hsu JS, Han IT, Tsai TH, et al. Pleural tags on CT scans to predict visceral pleural invasion of non–small cell lung cancer that does not abut the pleura. Radiology
13. Ebara K, Takashima S, Jiang B, et al. Pleural invasion by peripheral lung cancer: prediction with three-dimensional CT. Acad Radiol
14. Yu A, Li Q, Zhan Y, et al. Comparison of the rebuilding effects of different computed tomography scanners and reconstructive settings for the 5-line sign in normal interlobular fissures. European Radiology Experimental
15. Silvestri GA, Gonzalez AV, Jantz MA, et al. Methods for staging
non-small cell lung cancer: diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest
. 2013;143(suppl 5):e211S–e250S.
16. Yu A, Li Q, He J, et al. Conditions and mechanism for the appearance of interlobar fissures as 5-Line signs in axial maximum intensity projection images. J Comput Assist Tomogr
17. Napel S, Rubin GD, Jeffrey RB Jr. STS-MIP: a new reconstruction technique for CT of the chest. J Comput Assist Tomogr
18. Webb WR. The pleural tail sign. Radiology
19. Hill CA. “Tail” signs associated with pulmonary lesions: critical reappraisal. AJR Am J Roentgenol
20. Gruden JF. What is the significance of pleural tags? AJR Am J Roentgenol
21. Yang X, Soimakallio S. Pleural signs of small peripheral pulmonary masses: pathologic correlation with chest radiographs and diagnostic value. Eur J Radiol
22. Kligerman S, Abbott G. A radiologic review of the new TNM classification for lung cancer. AJR Am J Roentgenol
23. Ohta Y, Oda M, Wu J, et al. Can tumor size be a guide for limited surgical intervention in patients with peripheral non–small cell lung cancer? Assessment from the point of view of nodal micrometastasis. J Thorac Cardiovasc Surg
24. Watanabe S, Oda M, Tsunezuka Y, et al. Peripheral small-sized (2 cm or less) non-small cell lung cancer with mediastinal lymph node metastasis; clinicopathologic features and patterns of nodal spread. Eur J Cardiothorac Surg
25. Hashizume T, Yamada K, Okamoto N, et al. Prognostic significance of thin-section CT scan findings in small-sized lung adenocarcinoma. Chest
26. Bayanati H, E Thornhill R, Souza CA, et al. Quantitative CT texture and shape analysis: can it differentiate benign and malignant mediastinal lymph nodes in patients with primary lung cancer? Eur Radiol
27. Stewart DJ, Nunez MI, Behrens C, et al. Membrane carbonic anhydrase IX expression and relapse risk in resected stage I–II non–small-cell lung cancer. J Thorac Oncol
28. Kojima Y, Saito H, Sakuma Y, et al. Correlations of thin-section computed tomographic, histopathological, and clinical findings of adenocarcinoma with a bubblelike appearance. J Comput Assist Tomogr
29. Yanagawa M, Tanaka Y, Leung AN, et al. Prognostic importance of volumetric measurements in stage I lung adenocarcinoma. Radiology
30. Ikehara M, Saito H, Yamada K, et al. Prognosis
of small adenocarcinoma of the lung based on thin-section computed tomography and pathological preparations. J Comput Assist Tomogr
31. Goo JM, Park CM, Lee HJ. Ground-glass nodules on chest CT as imaging biomarkers in the management of lung adenocarcinoma. AJR Am J Roentgenol
32. Naidich DP, Bankier AA, MacMahon H, et al. Recommendations for the management of subsolid pulmonary nodules detected at CT: a statement from the Fleischner Society. Radiology