AJR Women's Imaging Online
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
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 Kono, R.
Right arrow Articles by Takamori, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kono, R.
Right arrow Articles by Takamori, S.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
DOI:10.2214/AJR.05.1446
AJR 2007; 188:26-36
© American Roentgen Ray Society

Original Research

Dynamic MRI of Solitary Pulmonary Nodules: Comparison of Enhancement Patterns of Malignant and Benign Small Peripheral Lung Lesions

Rei Kono1, Kiminori Fujimoto1, Hiroshi Terasaki1, Nestor L. Müller2, Seiya Kato3, Junko Sadohara1, Naofumi Hayabuchi1 and Shinzo Takamori4

1 Department of Radiology, Kurume University School of Medicine, 67 Asahi-machi, Kurume, Fukuoka 830-0011, Japan.
2 Department of Radiology, Vancouver General Hospital and University of British Columbia, Vancouver, BC, V5Z 1M9, Canada.
3 Department of Pathology, Kurume University School of Medicine, Kurume, Japan.
4 Department of Surgery, Kurume University School of Medicine, Kurume, Japan.

Received August 17, 2005; accepted after revision December 7, 2005.

 
Address correspondence to K. Fujimoto (kimichan{at}med.kurume-u.ac.jp).


Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
OBJECTIVE. The purpose of this study was to compare the dynamic contrast-enhanced MRI enhancement characteristics of malignant and benign solitary pulmonary nodules.

MATERIALS AND METHODS. The characteristics of 202 solitary pulmonary nodules (diameter, 1-3 cm; 144 cases of primary lung cancer, 31 cases of focal organizing pneumonia, 15 tuberculomas, 12 hamartomas) were reviewed retrospectively. In all cases dynamic MR images were obtained before and 1, 2, 3, 4, 5, 6, and 8 minutes after bolus injection of gadopentetate dimeglumine. Maximum enhancement ratio, time at maximum enhancement ratio, slope of time-enhancement ratio curves, and washout ratio were assessed. Statistical analyses were performed with the Kruskal-Wallis test with Bonferroni correction, chi-square test, and receiver operating characteristic curves.

RESULTS. For 122 (85%) of 144 primary lung cancers, time at maximum enhancement ratio was 4 minutes or less. For all tuberculomas and hamartomas, time at maximum enhancement ratio was greater than 4 minutes or gradual enhancement occurred without a peak time (p < 0.0001). Lung cancers had different maximum enhancement ratios and slopes than benign lesions (all p < 0.005). With 110% or lower maximum enhancement ratio as a cutoff value, the positive predictive value for malignancy was 92%; sensitivity, 63%; and specificity, 74%. With 13.5%/min or greater slope as a cutoff value, sensitivity, specificity, positive predictive value, and negative predictive value for malignancy were 94%, 96%, 99%, and 74%, respectively.

CONCLUSION. Dynamic contrast-enhanced MRI is helpful in differentiating benign from malignant solitary pulmonary nodules. Absence of significant enhancement is a strong predictor that a lesion is benign.

Keywords: chest • dynamic MRI • infectious disease • lung disease • oncologic imaging


Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Lung cancer is the most common cause of cancer death and the second most common cause of new cancer cases among men and women in the United States. In 2004, there were an estimated 173,770 new cases and 160,440 deaths of lung cancer in the United States [1]. At diagnosis of lung cancer, most patients are older than 65 years and have stage 3 or 4 disease [2].

A solitary pulmonary nodule (SPN) is defined as an approximately round lesion less than 3 cm in diameter that is completely surrounded by pulmonary parenchyma without other pulmonary abnormalities [3]. An SPN is found on 0.09-0.20% of all chest radiographs, and an estimated 150,000 such nodules are identified each year in the United States [3]. Differentiation of malignant from benign lung nodules is a common clinical problem. Although small SPNs are evaluated with modern techniques such as thin-section CT, transbronchial biopsy, and transthoracic needle biopsy, as many as 30% of benign SPNs are resected unnecessarily [4-6]. On FDG PET, malignant nodules can be differentiated from benign nodules with an accuracy of approximately 90% [7]. The sensitivity of FDG PET in the detection of cancer decreases considerably if the lesions are less than 2 cm in diameter [8]. Furthermore, in differentiation of malignant from benign SPNs many false-positive diagnoses are made among patients with an active infection or inflammatory lesion and tuberculoma [9, 10]. False-negative diagnoses are made among patients with bronchioloalveolar carcinoma and a carcinoid tumor [11, 12].

Several studies have shown that enhancement of malignant pulmonary tumors is greater than that of benign lung nodules on angiography [13], contrast-enhanced conventional tomography [14], FDG PET [15, 16], Doppler sonography [17], contrast-enhanced CT [18-21], and contrast-enhanced MRI [22-29]. These studies, however, were conducted with small numbers of patients with acute inflammatory lesions or infection, such as round pneumonia associated with active infection, focal organizing pneumonia [30-32], and focal granulomatous disease. Because they are associated with increased blood flow and vessel permeability, acute inflammatory lesions and infections are characterized by increased accumulation of contrast material on dynamic CT [6, 20, 21] and MRI [26, 27]. Therefore it is difficult to differentiate inflammatory lesions from malignant neoplasms by use of analysis of perfusion characteristics [6, 19-21, 27].


Figure 1
View larger version (41K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 1 —66-year-old man with adenocarcinoma of the lung. Solitary nodule measuring 15 mm in diameter is present in right middle lobe. Dynamic MR images obtained before and after IV injection of gadopentetate dimeglumine show rapid and relatively heterogeneous enhancement that continues to late phase. Maximum enhancement ratio, 68%; time at maximum enhancement ratio, 2 minutes; slope, 34%/min; washout ratio, 4% (prior). Subsequent panels (left to right) show dynamic MR images obtained at times noted. Last panel shows placement of region of interest (ROI).

 
Dynamic MRI has been used to assess tumor vascularity (microvessel counts) and interstitium (degree of elastic and collagen fibers) and to predict outcome among patients with peripheral small lung carcinoma [29]. Results of several studies of MRI of pulmonary lesions have suggested that the kinetic indexes and morphologic parameters of dynamic MRI allow accurate differentiation of malignant and benign lesions [22-28]. These studies, however, were conducted with a relatively small number of cases (total, {approx} 50 cases) and a small number of benign inflammatory lesions and infections. The purpose of our study was to compare the enhancement characteristics of malignant with those of benign SPNs, including the nodules of focal organizing pneumonia, on dynamic contrast-enhanced MRI.


Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Patients
We retrospectively reviewed the medical records of all 421 patients who had undergone gadopentetate dimeglumine-enhanced dynamic MRI for further evaluation of pulmonary nodular lesions at our hospital between 1992 and 2001. Among these 421 patients, we selected all patients who fulfilled the inclusion criteria for the current study: presence of SPN without evidence of calcification or fat attenuation, SPN diameter of 1-3 cm, histologic proof of diagnosis, and dynamic MRI performed within 2 weeks before surgical resection or biopsy. Our institutional review board approved this study, and all patients gave informed consent for review of their records, files, and images.

Two hundred two patients met the inclusion criteria for the study. In all patients, an SPN was discovered on a routine chest radiograph or chest CT scan. All benign nodular lesions were difficult to differentiate from lung cancer at this stage. The age range of the selected patients was 30-87 years (median, 67 years); 152 patients were men, and 50 were women. The 202 SPNs were 1.0-2.9 cm (median, 2.2 cm) in largest diameter.

In all cases, definitive diagnosis was made at surgical resection, transbronchial lung biopsy, or CT-guided fine-needle biopsy. The final diagnoses were confirmed with microbiologic and histopathologic examinations of specimens obtained by CT-guided transthoracic needle biopsy, transbronchial biopsy, videotape-assisted thoracoscopic surgery, or surgical resection. The 202 SPNs were classified into the following four groups on the basis of the final diagnosis proved at pathologic analysis: 144 cases of primary non-small cell lung cancer (89 adenocarcinomas and 55 squamous cell carcinomas; 110 men, 34 women; median age, 66 years), 31 cases of focal organizing pneumonia (23 men, eight women; median age, 60 years), 15 inactive tuberculomas (12 men, three women; median age, 57 years), and 12 hamartomas (seven men, five women; median age, 58 years). In this study, focal organizing pneumonia was defined as the presence of a focal (localized) round opacity on chest radiograph or CT scan and as pathologically focal distribution of a nonspecific inflammatory lesion or organizing pneumonia pattern, excluding tuberculous granuloma and underlying secondary causative factors, such as connective tissue disease, hematologic malignant neoplasm, drug reaction, and radiation exposure known to be associated with organizing pneumonia [30, 31]. The criteria for histologic diagnosis of focal organizing pneumonia were presence of an organization of intraalveolar exudate, chronic inflammatory cell infiltration, and fibrotic change in the alveolar septa and peribronchovascular interstitium [30-32]. Nine cases of focal organizing pneumonia encompassing 31 lesions were proved at culture to be caused by bacterial infection, including three cases of Staphylococcus pneumoniae infection, two cases of Streptococcus pneumoniae infection, two cases of Mycobacterium avium-intracellulare infection, and two cases of Cryptococcus neoformans infection. The inactive tuberculomas were differentiated from active tuberculosis on the basis of the following criteria: no evidence of change in size at follow-up CT performed every 6 months for more than 2 years and no evidence of the presence of Mycobacterium tuberculosis at microbiologic examination [27].

Dynamic MR Studies
All MR images were obtained with a 0.5-T superconducting system (Magnex 50HP, Shimadzu). The studies were performed with a T1-weighted spin-echo sequence (TR/TE, 150/10; one signal acquired) during breath-hold at full inspiration. To avoid cardiac and arterial motion artifacts, three oblique sagittal or transverse images that included the center of the lesion and excluded the heart and great vessels were chosen for the dynamic MRI. A section thickness of 8 mm and a gap of 2 mm were chosen to maintain a sufficient signal-to-noise ratio for a 256 x 204 rectangular matrix, 30 x 30 cm field of view, and one signal acquisition. Sampling time per image was 16 seconds. After the first spin-echo sequence, a manual IV injection of a bolus of 0.1 mmol/kg body weight gadopentetate dimeglumine (Magnevist, Schering) was administered over 10 seconds with a 21-gauge butterfly infusion set (Hakko Shoji). The range of dose volumes administered was 10-16 mL (median, 12 mL). The injection was administered through a small syringe, and a stopwatch was used to ensure even injection of contrast material over 10 seconds. Contrast-enhanced dynamic MR images were obtained 1, 2, 3, 4, 5, 6, and 8 minutes after completion of the IV injection (Fig. 1).


Figure 2
View larger version (139K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2A —66-year-old man with inactive tuberculoma in left upper lobe of lung. Dynamic MR image obtained before bolus injection of gadopentetate dimeglumine shows solitary nodule measuring 25 mm in diameter with low signal intensity.

 


Figure 3
View larger version (147K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2B —66-year-old man with inactive tuberculoma in left upper lobe of lung. Dynamic MR image obtained 6 minutes after contrast injection shows lesion in A as being peripheral rim enhancement. Outer rim of lesion shows gradual enhancement; most of central area shows no enhancement (thin-rim enhancement).

 


Figure 4
View larger version (154K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2C —66-year-old man with inactive tuberculoma in left upper lobe of lung. Photograph of cut surface shows thin-rim fibrous capsule, epithelioid granulomas (arrows) on periphery, and areas of caseous necrosis with scattered anthracosis in central portion.

 


Figure 5
View larger version (141K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2D —66-year-old man with inactive tuberculoma in left upper lobe of lung. Photomicrograph shows small congestive capillaries (arrows) scattered in border between fibrous rim and epithelioid granuloma with scarring. (H and E, x100)

 
Image Analysis
Two observers who had no knowledge of the patients' clinical or histopathologic data reviewed all images and reached a decision by consensus. All morphologic evaluations were performed on the center slice of each SPN. SPN enhancement on the dynamic spin-echo images was measured with stand-alone software (Time-Intensity Curve, Shimadzu). The signal intensity of SPNs was measured on circular operator-defined regions of interest with an electronic cursor on each spin-echo image (Fig. 1). All regions of interest were placed by the same two observers by consensus. Large regions of interest were chosen to incorporate solid-appearing parts of a tumor and to exclude obvious cystic or necrotic areas, as described by Yamashita et al. [18] and Swensen et al. [19].

Curves of time-enhancement ratios were obtained by calculating the percentage increase in signal intensity at any given time compared with the signal intensity before injection of contrast material and plotting this percentage increase in signal intensity against time. Maximum enhancement ratio was the peak of the time-enhancement ratio curve. When the rate of increase in enhancement ratio toward the peak was less than 3%, maximum enhancement ratio was defined as the peak of the time-enhancement ratio curve, or before the point of the peak [29, 33]. The time at which maximum enhancement ratio occurred was determined for each image. The slope of the time-enhancement ratio curve was calculated as the maximum enhancement ratio divided by the baseline value per minute. Washout ratio was calculated as the percentage decrease in enhancement ratio between the maximum enhancement ratio and the enhancement ratio 3 minutes after time at maximum enhancement ratio (ERwr) with the following equation: washout ratio = (maximum enhancement ratio - ERwr)/maximum enhancement ratio. When the time at maximum enhancement ratio of the time-enhancement ratio curve was 4 minutes, the ERwr was obtained directly from the time-enhancement ratio curve because an image was not obtained 7 minutes after injection. When the maximum enhancement ratio occurred 6 or 8 minutes after injection, the washout ratio could not be calculated according to the definition of washout ratio. The washout ratio in such cases therefore was considered zero. Thus the enhancement characteristics were determined as enhancement ratio at each time point, maximum enhancement ratio, time at maximum enhancement ratio, slope, and washout ratio.

The time-enhancement ratio curves were classified into three major types as follows: Type A had an early peak (time at maximum enhancement ratio within 4 minutes); type B had a late peak (time at maximum enhancement ratio later than 4 minutes); and type C had a gradually increasing pattern but without a peak (time at maximum enhancement ratio, 8 minutes; washout ratio, zero). Type A curves were divided into two subtypes. Type A1 had a higher washout ratio (> 10%), and type A2 had a relatively low washout ratio (≤ 10%).


Figure 6
View larger version (77K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3A —30-year-old woman with hamartoma of middle lobe of lung. Images show marked rim enhancement of solitary nodule measuring 29 mm in diameter and mixture of gradually heterogeneous and irregular linear enhancement and lack of enhancement (network enhancement) in central area. Dynamic MR image obtained before contrast injection.

 


Figure 7
View larger version (77K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3B —30-year-old woman with hamartoma of middle lobe of lung. Images show marked rim enhancement of solitary nodule measuring 29 mm in diameter and mixture of gradually heterogeneous and irregular linear enhancement and lack of enhancement (network enhancement) in central area. Dynamic MR image obtained 3 minutes after contrast injection.

 


Figure 8
View larger version (78K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3C —30-year-old woman with hamartoma of middle lobe of lung. Images show marked rim enhancement of solitary nodule measuring 29 mm in diameter and mixture of gradually heterogeneous and irregular linear enhancement and lack of enhancement (network enhancement) in central area. Dynamic MR image obtained 8 minutes after contrast injection.

 


Figure 9
View larger version (83K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3D —30-year-old woman with hamartoma of middle lobe of lung. Images show marked rim enhancement of solitary nodule measuring 29 mm in diameter and mixture of gradually heterogeneous and irregular linear enhancement and lack of enhancement (network enhancement) in central area. Loupe magnification shows hamartoma well circumscribed with islands of cartilage and fat. Outer fibrous rim of tumor exhibits invagination (arrows) toward central cartilaginous tissue.

 


Figure 10
View larger version (148K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3E —30-year-old woman with hamartoma of middle lobe of lung. Images show marked rim enhancement of solitary nodule measuring 29 mm in diameter and mixture of gradually heterogeneous and irregular linear enhancement and lack of enhancement (network enhancement) in central area. Low-power photomicrograph shows tumor components. Cleftlike space in invaginated stromal tissue is lined by respiratory epithelium (arrows). M = primitive mesenchymal tissue, F = fat, C= cartilage. (H and E, x20).

 
The static enhancement pattern was evaluated on the image at the time point of maximum enhancement ratio and was classified into four patterns: homogeneous if enhancement of the nodule was uniform or homogeneous, heterogeneous if enhancement of the nodule included areas with no or little enhancement, peripheral if contrast enhancement was predominately or exclusively in the periphery of a round or ovoid lesion, and no enhancement if a nodule had no visually apparent enhancement. Specific other enhancement patterns of thin-rim enhancement [34-36] (Figs. 2A, 2B, 2C, and 2D) and network enhancement (Figs. 3A, 3B, 3C, 3D, and 3E) also were recorded. Thin-rim enhancement was considered present when a nodular lesion had an area of enhancement limited to 2 mm or less of the outer rim of the lesion (Figs. 2A, 2B, 2C, and 2D). The network pattern was considered present when enhancement of the nodular lesion was heterogeneous with areas of irregular linear enhancement and areas of no enhancement (Figs. 3A, 3B, 3C, 3D, and 3E).

Statistical Analysis
Normality of distribution was evaluated with the Shapiro-Wilk test. Because none of the continuous variables was normally distributed, all statistical analyses were performed with nonparametric methods, and skewed data were summarized with median and range from 27th-75th percentile of the interquartile range (IQR).

The enhancement characteristics (i.e., enhancement ratio at each time, maximum enhancement ratio, time at maximum enhancement ratio, slope, and washout ratio) of the time-enhancement ratio curves on dynamic MRI of primary lung cancers were compared with those of benign lesions. Statistical analyses of the correlation of each enhancement characteristic were performed with the Kruskal-Wallis test. Multiple comparisons of each coupled combination were analyzed with Bonferroni correction after a Mann-Whitney test. The relations between static enhancement pattern and each SPN and between types of time-enhancement ratio curves and each SPN were analyzed with a chi-square test. For cases of primary lung cancer, statistical analyses between adenocarcinoma and squamous cell carcinoma were performed with Mann-Whitney and chi-square tests.

Receiver operating characteristic (ROC) analysis was performed to evaluate the usefulness of maximum enhancement ratio, slope, and washout ratio as markers for differentiating primary lung cancer from focal organizing pneumonia and for differentiating lung cancer from other benign SPNs. Area under the ROC curve (Az) was calculated and ranged from 0.5 to 1.0, increasing when diagnostic performance approached that of the reference standard (in this case, determination of malignancy). Sensitivity, specificity, positive predictive value, and negative predictive value were calculated with standard formulas according to the values of these indexes and by varying the index values that indicated positive differentiation (i.e., threshold value) [37]. Feasible threshold dynamic MRI parameters were tested for capability to enable differentiation between lung cancer and focal organizing pneumonia and between lung cancer and other benign SPNs.

A p value less than 0.05 was considered statistically significant. When Bonferroni correction was used to adjust for multiple comparisons, a p value less than 0.0083 (0.05/6, when 6 means 4C2) was considered to indicate a statistically significant difference. The Shapiro-Wilk test and ROC analysis were performed with statistical software (JMP version 4.0, SAS). All other statistical analyses were performed with StatView version 5.0 software (Abacus Concepts) and a Macintosh computer (Apple).


Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Patient Age and Sex and SPN Diameter
Comparisons of patient age and sex and largest diameter of SPNs, including primary lung cancer, focal organizing pneumonia, tuberculoma, and hamartoma are summarized in Table 1. Bonferroni correction showed that the only parameter with a statistically significant difference was patient age for lung cancer and tuberculoma (p < 0.001). In patients with primary lung cancer, there were no statistically significant differences between adenocarcinoma and squamous cell carcinoma for age, sex, or enhancement characteristics.


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

  		TABLE 1: Comparisons of Patient Characteristics for Small Lung Cancer and Benign Nodular Lesions

 

Enhancement Characteristics (Parameters of Dynamic Study) for Primary Lung Cancer and Benign Nodules
The enhancement ratios for SPNs are summarized in Figures 4A, 4B, 4C, and 4D and Table 2. The Kruskal-Wallis test showed significant differences between primary lung cancer and benign nodules in maximum enhancement ratio, time at maximum enhancement ratio, slope, and washout ratio (all comparisons, p < 0.0001).


Figure 11
View larger version (18K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4A —Graphs show curves of time-enhancement ratios. Dotted lines connect medians of enhancement ratios at each time point for each solitary pulmonary nodule. Horizontal bars indicate medians; vertical bars, ranges. Horizontal boundaries of boxes represent 25th and 75th percentiles of interquartile range. Lung cancer.

 

Figure 12
View larger version (21K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4B —Graphs show curves of time-enhancement ratios. Dotted lines connect medians of enhancement ratios at each time point for each solitary pulmonary nodule. Horizontal bars indicate medians; vertical bars, ranges. Horizontal boundaries of boxes represent 25th and 75th percentiles of interquartile range. Focal organizing pneumonia.

 

Figure 13
View larger version (16K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4C —Graphs show curves of time-enhancement ratios. Dotted lines connect medians of enhancement ratios at each time point for each solitary pulmonary nodule. Horizontal bars indicate medians; vertical bars, ranges. Horizontal boundaries of boxes represent 25th and 75th percentiles of interquartile range. Tuberculoma.

 

Figure 14
View larger version (16K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4D —Graphs show curves of time-enhancement ratios. Dotted lines connect medians of enhancement ratios at each time point for each solitary pulmonary nodule. Horizontal bars indicate medians; vertical bars, ranges. Horizontal boundaries of boxes represent 25th and 75th percentiles of interquartile range. Hamartoma.

 

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

  		TABLE 2: Comparison of Enhancement Characteristics of Small Lung Cancer and Benign Nodular Lesions

 

Maximum Enhancement Ratio
Bonferroni correction showed statistically significant differences in maximum enhancement ratio between lung cancer (median, 101%; IQR, 79-124%) and focal organizing pneumonia (median, 128%; IQR, 113-160%; p < 0.005), between lung cancer and tuberculoma (median, 36%; IQR, 16-65%; p < 0.001), and between lung cancer and hamartoma (median, 68; IQR, 54-88%; p < 0.05).

Time at Maximum Enhancement Ratio
Bonferroni correction showed statistically significant differences in time at maximum enhancement ratio between lung cancer (median, 3 minutes; IQR, 2-4 minutes) and tuberculoma (median, 8 minutes; IQR, 6-8 minutes; p < 0.001) and between lung cancer and hamartoma (median, 8 minutes; IQR, 6-8 minutes; p < 0.001). However, there was no significant difference between lung cancer and focal organizing pneumonia.

Slope
Primary lung cancer had a greater slope than tuberculoma and hamartoma. Focal organizing pneumonia also had a greater slope; however, the slope at 1 minute (i.e., median of enhancement ratio at 1 minute) of focal organizing pneumonia was steeper than that of primary lung cancer (Figs. 4A and 4B). Moreover, focal organizing pneumonia had more rapid and stronger enhancement in the first transit of bolus injection than did primary lung cancer. Bonferroni correction showed statistically significant differences in slope between lung cancer (median, 36%/min; IQR, 24-50%/min) and focal organizing pneumonia (median, 50%/min; IQR, 30-83%/min; p < 0.001), between lung cancer and tuberculoma (median, 5%/min; IQR, 2-11%/min; p < 0.001), and between lung cancer and hamartoma (median, 9%/min; IQR, 7-13%/min; p < 0.005).

Washout Ratio
Bonferroni correction showed significant difference in washout ratio between lung cancer (median, 6; IQR, 3-12%) and tuberculoma (median, zero; IQR, 0-8%; p < 0.05) and between lung cancer and hamartoma (median, zero; IQR, 0-6%; p < 0.05).

Types of Time-Enhancement Ratio Curves
The time-enhancement ratio curves were different for lung cancer, focal organizing pneumonia, tuberculoma, and hamartoma (Figs. 4A, 4B, 4C, and 4D). The time-enhancement ratio curve for lung cancer usually showed a moderate and progressive increase to peak height, where it maintained a plateau. The time-enhancement ratio curves for tuberculoma and hamartoma showed some, gradual, or no increase after injection of gadopentetate dimeglumine. Conversely, the time-enhancement ratio curve for focal organizing pneumonia showed a rapid increase in enhancement and a gradual decrease in signal intensity after reaching peak height. The most common types of time-enhancement ratio curve for lung cancer were type A2 (time at maximum enhancement ratio ≤ 4 minutes and washout ratio < 10%) (53%, 77/144) and type A1 (time at maximum enhancement ratio ≤ 4 minutes and washout ratio ≥ 10%) (31%, 45/144) (Table 2). Type B occurred in 12% and type C in 4% of lung cancers. The most common types of time-enhancement ratio curve for focal organizing pneumonia were type A1 (62%, 19/31) and type A2 (32%, 10/31). Type B occurred in 10% and type C in 6% of cases of focal organizing pneumonia. The most common types of time-enhancement ratio curve for tuberculoma and hamartoma were type B (40% of tuberculomas, 42% of hamartomas) and type C (60% of tuberculomas, 58% of hamartomas). There was no case of type A curve for tuberculoma or hamartoma.

Enhancement Pattern
The differences in static enhancement patterns within the four groups were statistically significant (p < 0.001) (Table 2). Of the lung cancers (Fig. 1), 81 (56%) of 144 tumors had heterogeneous enhancement, 60 (42%) of the tumors had completely homogeneous enhancement, and only three (2%) of the tumors had peripheral enhancement owing to central necrosis. Nineteen (62%) of the 31 nodular lesions of focal organizing pneumonia (Figs. 5A, 5B, 5C, and 5D) had homogeneous, 10 (32%) had heterogeneous, and two (6%) had peripheral enhancement. In contrast, none of the tuberculomas or hamartomas had homogeneous enhancement. The tuberculomas and hamartomas were more likely than lung cancers to have patterns of peripheral enhancement or no enhancement. Furthermore, tuberculomas had a higher prevalence of thin-rim enhancement (eight of 15 tuberculomas) (Figs. 2A, 2B, 2C, and 2D) than did lung cancer (none of 144 tumors). Hamartomas had a higher prevalence of network enhancement (six of 12 hamartomas) (Figs. 3A, 3B, 3C, 3D, and 3E) compared with the heterogeneous enhancement pattern of other SPNs (none of the other SPNs had network enhancement).


Figure 15
View larger version (77K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5A —68-year-old woman with focal organizing pneumonia in right middle lobe of lung. Thin-section CT scan shows solitary nodule measuring 20-mm in diameter with irregular margin. Pleural tag is visible in subpleural region.

 

Figure 16
View larger version (121K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5B —68-year-old woman with focal organizing pneumonia in right middle lobe of lung. Dynamic MR image before IV injection of gadopentetate dimeglumine.

 

Figure 17
View larger version (132K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5C —68-year-old woman with focal organizing pneumonia in right middle lobe of lung. Dynamic MR image obtained 3 minutes after IV injection of gadopentetate dimeglumine shows slightly heterogeneous enhancement that is strongest in early phase (3 minutes = time at maximum enhancement ratio). Signal intensity before and 3 and 6 minutes (time at maximum enhancement ratio + 3) after contrast injection was calculated by each region of interest. Maximum enhancement ratio, 126%; slope, 42%/min; washout ratio, 15%.

 

Figure 18
View larger version (152K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5D —68-year-old woman with focal organizing pneumonia in right middle lobe of lung. Photomicrograph shows focal organizing pneumonia lesion characterized by patchy interstitial inflammation and air-space organizing granulation tissue. Dilated small vessels (arrows) are highlighted with congestion and intraalveolar hemorrhage. (H and E, x100)

 

ROC Analysis
Parameters helpful in differentiating the lesions of lung cancer from those of focal organizing pneumonia included maximum enhancement ratio (Az, 0.72; 95% CI, 0.62-0.81) and slope (Az, 0.65; 95% CI, 0.54-0.77) (Fig. 6A). To differentiate lung cancer from focal organizing pneumonia, threshold levels of 110% for maximum enhancement ratio and 37.5%/min for slope were found suitable. Table 3 summarizes the diagnostic characteristics according to threshold values on the basis of each ROC curve. The sensitivity was 63% for maximum enhancement ratio and 55% for slope. The specificity was 84% for maximum enhancement ratio and 71% for slope. The highest positive predictive value was 95% for maximum enhancement ratio; however, the negative predictive value was only 29%.


Figure 19
View larger version (10K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 6A —Graphs show receiver operating characteristic curve of enhancement characteristics. Fraction of true-positive results (sensitivity) and false-positive results (1 minus specificity) for maximum enhancement ratio (thick solid line), slope (thin solid line), and washout ratio (dashed line) as markers of malignancy. Value of 0.5 is no better than expected by chance, and value of 1.0 reflects perfect indicator. Graph shows receiver operating characteristic curve for differentiating lung cancer from focal organizing pneumonia. Calculated area under curve is 0.72 for maximum enhancement ratio, 0.65 for slope, and 0.54 for washout ratio.

 

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

  		TABLE 3: Diagnostic Rates of Maximum Enhancement Ratio According to Threshold of Cutoff Value for Differentiation of Lung Cancer and Focal Organizing Pneumonias

 

Parameters helpful in differentiating lung cancer from benign SPNs other than those of focal organizing pneumonia included maximum enhancement ratio (Az, 0.87; 95% CI, 0.79-0.94), slope (Az, 0.97; 95% CI, 0.95-0.99), and washout ratio (Az, 0.75; 95% CI, 0.63-0.87) (Fig. 6B). To differentiate lung cancer from benign SPNs (tuberculoma and hamartoma), threshold levels of 75% for maximum enhancement ratio, 13.5%/min for slope, and 2% for washout ratio were found suitable. Table 4 summarizes the diagnostic characteristics according to threshold values on the basis of each ROC curve. The sensitivity was 81% for maximum enhancement ratio, 94% for slope, and 83% for washout ratio. The specificity was 81% for maximum enhancement ratio, 96% for slope, and 63% for washout ratio. The highest positive predictive value and negative predictive value were 99% and 74%, respectively, for slope.


Figure 20
View larger version (9K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 6B —Graphs show receiver operating characteristic curve of enhancement characteristics. Fraction of true-positive results (sensitivity) and false-positive results (1 minus specificity) for maximum enhancement ratio (thick solid line), slope (thin solid line), and washout ratio (dashed line) as markers of malignancy. Value of 0.5 is no better than expected by chance, and value of 1.0 reflects perfect indicator. Graph shows receiver operating characteristics curve for differentiating lung cancer from benign solitary pulmonary nodules (tuberculoma and hamartoma). Calculated area under curve is 0.87 for maximum enhancement ratio (thick solid line), 0.97 for slope (thin solid line), and 0.75 for washout ratio (dashed line).

 

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

  		TABLE 4: Diagnostic Rates of Enhancement Characteristics According to Thresholds of Cutoff Values for Differentiation of Lung Cancer and Tuberculoma or Hamartoma

 


Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Differentiation of benign from malignant SPNs with a noninvasive method such as CT, MRI, or FDG PET has been considered an important goal of diagnostic radiology. Results of several studies have suggested a potential role of contrast-enhanced dynamic MRI with gadopentetate dimeglumine in differentiating benign from malignant nodules [22-28]. The biodistribution of gadopentetate dimeglumine is nonspecific and is determined by the relative level of vascular perfusion of different tissues and their capillary permeability. These characteristics allow diffusion into the extracellular space [14, 29, 38, 39]. In general, tumors have a large extracellular space and exhibit greater enhancement than does normal tissue [14]. Fujimoto et al. [29] reported that the parameters of the first half of the time-enhancement ratio curve correlated with tumor angiogenesis (microvessel count) and that those of the latter half correlated with tumor interstitium (degree of elastic and collagen fibers). These findings suggest that the intratumoral circulation of gadopentetate dimeglumine depends on the quantity and distribution of microvessels, elastic fibers, and collagen fibers.

Studies have shown that enhancement of malignant SPNs after IV administration of contrast material is greater than that of benign SPNs. However, some studies excluded nodules with active inflammation or infection. In our study, gadopentetate dimeglumine enhancement of the lesions of focal organizing pneumonia was greater than that of lung cancer. In contrast, enhancement of other benign SPNs (i.e., tuberculoma and hamartoma) was less than that of lung cancer. Ohno et al. [27] evaluated 58 SPNs, including malignant and benign lesions and active infections, and found that the mean relative enhancement ratio and mean slope of enhancement for the active infection group were higher than those for the malignant SPN group.

Focal organizing pneumonia is difficult to differentiate from lung cancer [30-32]. Our results indicate that lung cancer has a significantly lower maximum enhancement ratio and slope than focal organizing pneumonia. ROC curve analysis showed that SPNs with a maximum enhancement ratio less than 110% or slope less than 37.5%/min were more likely to be lung cancer (positive predictive values, 95% and 89%, respectively). The sensitivity and specificity of maximum enhancement ratio, however, were relatively low at 63% and 84%, respectively. The overlap of enhancement characteristics of dynamic MRI between focal organizing pneumonia and lung cancer is not surprising because malignant neoplasms and tissues with acute inflammatory lesions or infection have increased blood flow, perfusion, and capillary permeability [27]. It should be noted, however, that the slope at 1 minute of the time-enhancement ratio curve (i.e., the median of enhancement ratio at 1 minute) for focal organizing pneumonia was steeper than that of lung cancer. This finding suggests that initial slope and peak enhancement may be different with if higher temporal resolution (i.e., acquisition of images every second) were possible. Differentiation between acute inflammatory lesions and malignant tumors needs further evaluation.

As in previous studies, we found significant differences in maximum enhancement ratio, time at maximum enhancement ratio, slope, and washout ratio between primary lung cancer and tuberculoma and hamartoma [22-28]. Enhancement parameters (maximum enhancement ratio, slope, and washout ratio) had high sensitivity, specificity, and positive predictive value for diagnosis of lung cancer and differentiation from benign SPNs other than focal organizing pneumonia. Both tuberculoma and hamartoma were more likely than primary lung cancer to have a late enhancement peak (type B) or gradually increased enhancement (type C) on the time-enhancement ratio curve. Twelve (80%) of 15 tuberculomas had a peripheral enhancement pattern or no enhancement, findings strongly suggestive of benignity. Furthermore, eight of 15 tuberculomas had thin-rim enhancement, and none of the cases of primary lung cancer did. The thin-rim enhancement pattern of tuberculoma on gadopentetate dimeglumine-enhanced MR images or iodinated contrast medium-enhanced CT scans has been previously reported [34-36]. At histopathologic examination, the enhancing rim at the periphery of the masses has been shown to correspond to fibrous tissue surrounding epithelioid (or tuberculous) granulomas, and the area of central portion without contrast enhancement to correspond to caseous necrosis or scarring [34, 36].

Calcification and a fat component in hamartomas are common findings; however, as many as two thirds of hamartomas do not have foci of calcification or fat density evident on CT or MRI and are difficult to differentiate from malignant SPNs [40]. In our study, 11 (92%) of 12 hamartomas had heterogeneous enhancement or peripheral enhancement. Furthermore, six (50%) of 12 hamartomas had network enhancement (heterogeneous enhancement representing a mixture of areas of irregular linear enhancement and areas of no enhancement) (Figs. 3A, 3B, 3C, 3D, and 3E). Areas with no enhancement have been shown to correspond to core cartilaginous tissue and septa, and areas with irregular linear enhancement to cleftlike branching mesenchymal connective tissue dipped into the cartilaginous core [41]. Network enhancement was not seen in any of the malignant pulmonary nodules.

Yi et al. [21] reported that dynamic enhancement on MDCT showed high sensitivity and negative predictive value for diagnosis of malignant nodules but low specificity because of the presence of highly enhancing benign nodules. Those authors did not find a significant difference in extent of microvessel density between malignant and benign lesions.

Our study had several limitations. Data acquisition was retrospective and needs further validation. Data analyses were by consensus interpretation, thus interobserver variability was not assessed. Selection bias might have occurred because the study included only SPNs measuring 1-3 cm in diameter and excluded a small number of various histologic types, such as small cell carcinoma and metastasis from other organs in the malignant SPN group, and other benign SPNs, such as sclerosing hemangioma and intrapulmonary lymph node. The cost-effectiveness of our approach has yet to be assessed.

In conclusion, gadopentetate dimeglumine-enhanced dynamic MRI (analysis of enhancement characteristics and types of time-enhancement ratio curve) may be helpful for differentiating primary malignant lung tumors from benign SPNs, especially inactive tuberculoma and hamartoma. Absence of significant enhancement (slope < 13.5%/min; maximum enhancement ratio < 75%) and the presence of peripheral enhancement and no enhancement are strong predictors that an SPN is benign.


References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Jemal A, Tiwari RC, Murray T, et al. Cancer statistics, 2004. CA Cancer J Clin2004; 54:8 -29[Abstract/Free Full Text]
  2. Hurria A, Kris MG. Management of lung cancer in older adults. CA Cancer J Clin2003; 53:325 -341[Abstract/Free Full Text]
  3. Ost D, Fein AM, Feinsilver SH. The solitary pulmonary nodule. N Engl J Med2003; 348:2535 -2542[Free Full Text]
  4. Siegelman SS, Khouri NF, Leo FP, Fishman EK, Braverman RM, Zerhouni EA. Solitary pulmonary nodules: CT assessment. Radiology1986; 160:307 -312[Abstract/Free Full Text]
  5. Ginsberg MS, Griff SK, Go BD, Yoo HH, Schwartz LH, Panicek DM. Pulmonary nodules resected at video-assisted thoracoscopic surgery: etiology in 426 patients. Radiology1999; 213:277 -282[Abstract/Free Full Text]
  6. Swensen SJ, Viggiano RW, Midthun DE, et al. Lung nodule enhancement at CT: multicenter study. Radiology2000; 214:73 -80[Abstract/Free Full Text]
  7. Kostakoglu L, Agress H Jr, Goldsmith SJ. Clinical role of FDG PET in evaluation of cancer patients. RadioGraphics2003; 23:315 -340[Abstract/Free Full Text]
  8. Mastin ST, Drane WE, Harman EM, Fenton JJ, Quesenberry L. FDG SPECT in patients with lung masses. Chest1999; 115:1012 -1017
  9. Goldsmith SJ, Kostakoglu L. Nuclear medicine imaging of lung cancer. Radiol Clin North Am2000; 38:511 -524[CrossRef][Medline]
  10. Goo JM, Im JG, Do KH, et al. Pulmonary tuberculoma evaluated by means of FDG PET: findings in 10 cases. Radiology2000; 216:117 -121[Abstract/Free Full Text]
  11. Higashi K, Nishikawa T, Seki H, et al. Fluorine-18-FDG PET imaging is negative in bronchioloalveolar lung carcinoma. J Nucl Med 1998;39:1016 -1020[Abstract/Free Full Text]
  12. Cheran SK, Nielsen ND, Patz EF Jr. False-negative findings for primary lung tumors on FDG positron emission tomography: staging and prognostic implications. AJR2004; 182:1129 -1132[Abstract/Free Full Text]
  13. Ney FG, Feist JH, Altemus LR, Ordinario VR. The characteristic angiographic criteria of malignancy. Radiology1972; 104:567 -570[Medline]
  14. Littleton JT, Durizch ML, Moeller G, Herbert DE. Pulmonary masses: contrast enhancement. Radiology1990; 177:861 -871[Abstract/Free Full Text]
  15. Patz EF Jr, Lowe VJ, Hoffman JM, et al. Focal pulmonary abnormalities: evaluation with F-18 fluoro-deoxyglucose PET scanning. Radiology1993; 188:487 -490[Abstract/Free Full Text]
  16. Gupta NC, Frank AR, Dewan NA, et al. Solitary pulmonary nodule: detection of malignancy with PET with 2-[F-18]-fluoro-2-deoxy-D-glucose. Radiology1992; 184:441 -444[Abstract/Free Full Text]
  17. Hsu WH, Ikezoe J, Chen CY, et al. Color Doppler ultrasound signals of thoracic lesions: correlation with resected histologic specimens. Am J Respir Crit Care Med1996; 153:1938 -1951[Abstract]
  18. Yamashita K, Matsunobu S, Tsuda T, et al. Solitary pulmonary nodule: preliminary study of evaluation with incremental dynamic CT. Radiology1995; 194:399 -405[Abstract/Free Full Text]
  19. Swensen SJ, Brown LR, Colby TV, Weaver AL, Midthun DE. Lung nodule enhancement at CT: prospective findings. Radiology1996; 201:447 -455[Abstract/Free Full Text]
  20. Zhang M, Kono M. Solitary pulmonary nodules: evaluation of blood flow patterns with dynamic CT. Radiology1997; 205:471 -478[Abstract/Free Full Text]
  21. Yi CA, Lee KS, Kim EA, et al. Solitary pulmonary nodules: dynamic enhanced multi-detector row CT study and comparison with vascular endothelial growth factor and microvessel density. Radiology2004; 233:191 -199[Abstract/Free Full Text]
  22. Fujimoto K, Edamitsu O, Meno S, Abe T, Uchida M, Nishimura H. Gd-DTPA-enhanced dynamic MR imaging in pulmonary disease: evaluation of usefulness in differentiating benign from malignant disease. Radiology1993; 189[suppl]:438
  23. Kono M, Adachi S, Kusumoto M, Sakai E. Clinical utility of Gd-DTPA-enhanced magnetic resonance imaging in lung cancer. J Thorac Imaging 1993;8:18 -26[Medline]
  24. Hittmair K, Eckersberger F, Klepetko W, Helbich T, Herold CJ. Evaluation of solitary pulmonary nodules evaluated with dynamic contrast-enhanced MR imaging: a promising technique? Magn Reson Imaging 1995;13:923 -933[CrossRef][Medline]
  25. Gückel C, Schnabel K, Deimling M, Steinbrich W. Solitary pulmonary nodules: MR evaluation of enhancement patterns with contrast-enhanced dynamic snapshot gradient-echo imaging. Radiology1996; 200:681 -686[Abstract/Free Full Text]
  26. Kono R, Fujimoto K, Terasaki H, Sadohara J, Nishimura H, Hayabuchi N. Gd-DTPA-enhanced dynamic MR imaging in 210 pulmonary diseases: evaluation of usefulness in differentiating lung cancer from benign nodular lesion. Radiology2001; 221[suppl]:203
  27. Ohno Y, Hatabu H, Takenaka D, et al. Solitary pulmonary nodules: Potential role of dynamic MR imaging in management: initial experience. Radiology2002; 224:503 -511[Abstract/Free Full Text]
  28. Schaefer JF, Vollmar J, Schick F, et al. Solitary pulmonary nodules: dynamic contrast-enhanced MR imaging: perfusion differences in malignant and benign lesions. Radiology2004; 232:544 -553[Abstract/Free Full Text]
  29. Fujimoto K, Abe T, Müller NL, et al. Small peripheral pulmonary carcinoma evaluated with dynamic MR imaging: correlation with tumor vascularity and prognosis. Radiology2003; 227:786 -793[Abstract/Free Full Text]
  30. Kohno N, Ikezoe J, Johkoh T, et al. Focal organizing pneumonia: CT appearance. Radiology1993; 189:119 -123[Abstract/Free Full Text]
  31. Lohr RH, Boland BJ, Douglas WW, et al. Organizing pneumonia: features and prognosis of cryptogenic, secondary, and focal variants. Arch Intern Med1997; 157:1323 -1329[Abstract]
  32. Oikonomou A, Hansell DM. Organizing pneumonia: the many morphological faces. Eur Radiol2002; 12:1486 -1496[CrossRef][Medline]
  33. Verstraete KL, De Deene Y, Roels H, Dierick A, Uyttendaele D, Kunnen M. Benign and malignant musculoskeletal lesions: dynamic contrast-enhanced MR imaging: parametric "first-pass" images depict tissue vascularization and perfusion. Radiology1994; 192:835 -843[Abstract/Free Full Text]
  34. Sakai F, Sone S, Maruyama A, et al. Thin-rim enhancement in Gd-DTPA-enhanced magnetic resonance imaging of tuberculoma: a new finding of potential differential diagnostic importance. J Thorac Imaging 1992;7:64 -69[Medline]
  35. Kusumoto M, Kono M, Adachi S, et al. Gadopentetate-dimeglumine-enhanced magnetic resonance imaging for lung nodules: differentiation of lung cancer and tuberculoma. Invest Radiol 1994;29[suppl 2]:S255 -S256
  36. Maruyama S, Murakami J, Hashimoto S, et al. Noncalcified pulmonary tuberculosis: CT enhancement pattern with histological correlation. J Thorac Imaging1995; 10:91 -95[Medline]
  37. Gupta NC, Aloof J, Gunnel E. Probability of malignancy in solitary pulmonary nodules using fluorine-18-FDG and PET. J Nucl Med 1996;37:943 -948[Abstract/Free Full Text]
  38. Strich G, Hagen PL, Gerber KH, Slutsky RA. Tissue distribution and magnetic resonance spin lattice relaxation effects of gadolinium-DTPA. Radiology1985; 154:723 -726[Abstract/Free Full Text]
  39. Weinmann HJ, Brasch RC, Press WR, Wesbey GE. Characteristics of gadolinium-DTPA complex: a potential NMR contrast agent. AJR 1984;142:619 -624[Abstract/Free Full Text]
  40. Siegelman SS, Khouri NF, Scott WW Jr, et al. Pulmonary hamartoma: CT findings. Radiology1986; 160:313 -317[Abstract/Free Full Text]
  41. Sakai F, Sone S, Kiyono K, et al. MR of pulmonary hamartoma: pathologic correlation. J Thorac Imaging1994; 9:51 -55[Medline]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
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 Artic