CLINICAL STUDY
CT-Guided Transthoracic Needle Aspiration Biopsy of Subsolid Lung Lesions Aaron W.P. Maxwell, BSc, Jeffrey S. Klein, MD, Kossivi Dantey, MD, Sharon L. Mount, MD, Kelly J. Butnor, MD, and Gladwyn Leiman, MD
ABSTRACT Purpose: To assess the diagnostic performance of computed tomography (CT)–guided transthoracic needle aspiration biopsy (TNAB) in the evaluation of persistent subsolid lung lesions. Materials and Methods: A retrospective review of all CT-guided TNABs performed at a single institution from January 2002 to November 2012 was conducted to identify patients with persistent subsolid lung lesions. The diagnostic performance of CTguided TNAB was assessed through comparison of cytologic diagnoses with core needle biopsy, surgical resection, or imaging and clinical follow-up. The cytologic, histologic, and imaging features of each lesion were characterized, and CT-guided TNAB complications were recorded. Results: In 32 patients, a diagnosis of benign or malignant disease was identified through evaluation of pathologic or follow-up data. There were 18 men and 14 women, with a mean age of 67.1 years ⫾ 9.6 (range, 52–86 y). The mean lesion diameter was 21 mm ⫾ 11 (range, 8–62 mm). A final diagnosis of malignancy was made in 28 cases (87.5%); four benign lesions were also diagnosed. The overall sensitivity of CT-guided TNAB in the evaluation of these lesions was 89.2%, and the specificity and positive predictive value were 100%. Two pneumothoraces (6.3%) were identified. Conclusions: Among patients with subsolid lung lesions, CT-guided TNAB is safe and shows high sensitivity. The high specificity and positive predictive value of the procedure allow for definitive treatment decisions to be made for most patients.
ABBREVIATIONS GG = ground-glass, TNAB = transthoracic needle aspiration biopsy
Although there has been considerable interest in thinsection computed tomography (CT) characterization of subsolid lung nodules (defined as spherical lesions o 3 cm in diameter with a component that is less than soft tissue density) over the past 10 years (1), particularly in light of increasing detection rates on multidetector CT performed for lung cancer screening (2) or in the
From the Department of Radiology, The University of Vermont College of Medicine (A.W.P.M.); and Departments of Radiology (J.S.K.) and Pathology (K.D., S.L.M., K.J.B., G.L.), Fletcher Allen Health Care, 111 Colchester Avenue, Burlington, VT 05401. Received September 7, 2013; final revision received November 2, 2013; accepted November 27, 2013. Address correspondence to J.S.K.; E-mail: [email protected] None of the authors have identified a conflict of interest. Figures E1 and E2 are available online at www.jvir.org. & SIR, 2014 J Vasc Interv Radiol 2014; 25:340–346 http://dx.doi.org/10.1016/j.jvir.2013.11.037
evaluation of chest disease, accurate diagnosis of these lesions remains difficult. This difficulty is due to the overlap in appearance on thin-section CT of preinvasive conditions, such as atypical adenomatous hyperplasia and adenocarcinoma in situ, with invasive adenocarcinoma as well as with inflammatory processes that may have a nodular subsolid appearance during the course of resolution (3,4). Although there is increasing recognition that a significant percentage of stable, subsolid lesions represent preinvasive or invasive adenocarcinoma (2,3), the accurate preoperative diagnosis of subsolid malignant lesions is important, particularly because some lesions are not readily amenable to thoracoscopic biopsy, and a substantial percentage of patients with subsolid malignant lesions are poor candidates for surgical resection because of comorbidities. There is scant information on the role of minimally invasive techniques such as CT-guided transthoracic needle aspiration biopsy (TNAB) cytology in the pathologic characterization of these lesions. The purpose of this study
Volume 25
’
Number 3
’
March
’
2014
was to assess the diagnostic performance of CT-guided TNAB in the diagnosis of persistent subsolid lung lesions.
MATERIALS AND METHODS This study, which was compliant with the Health Insurance Portability and Accountability Act, was approved by our institutional review board with a waiver for informed patient consent. A retrospective search of all CT-guided biopsy procedures performed at our institution from January 2002 to November 2012 was conducted using Current Procedure Terminology codes. Results were filtered by the key word “nodule” to yield a preliminary list, and from this list potentially relevant cases were identified via manual search of the associated procedure report for the terms “subsolid” and “groundglass” (GG). When identified, all such cases were transferred into a database, and any associated chest CT images obtained before biopsy were analyzed on thinsection series to determine the presence or absence of persistent, solitary subsolid lesions. All cases featuring lesions not meeting criteria as “subsolid” were excluded from analysis. For the purposes of this investigation, lesion “persistence” was defined as repeated CT demonstration of the lesion under evaluation for at least 3 months after initial detection. All identified subsolid nodules were characterized as to lobe within the lung, maximum lesion diameter in any plane, and estimated percent density (pure GG, 4 50% GG component [predominant GG], or o 50% GG component [predominant solid]). Also recorded for each case was the presence or absence of biopsy-associated complications, in particular, pneumothoraces requiring inhospital observation or treatment and bleeding requiring hospital admission. All patients underwent CT-guided TNAB with or without concomitant core needle biopsy. The CTguided TNAB procedure was performed by two boardcertified thoracic radiologists with 24 years and 8 years of experience with CT-guided TNAB as well as cardiothoracic radiology fellows and a physician’s assistant under direct supervision. All biopsies were performed under conscious sedation using CT fluoroscopy with a step-and-shoot mode controlled by the radiologist performing the biopsy. The patient was placed in a recumbent position on the CT table, with a needle approach chosen that used the shortest vertical pathway to the lesion to be sampled. A coaxial needle (Cook, Inc, Bloomington, Indiana) placement technique was used with a combination of an ultra-thin-walled needle placed via a single pleural puncture during patient breath hold in normal end expiration (ie, at functional residual capacity), with the outer needle placed to the edge of the lesion, followed by at least three aspiration biopsy specimens obtained during patient breath hold using a 22-gauge Chiba needle (INRAD Inc, Kentwood,
341
Michigan). A rotatory and to-and-fro motion with aspiration using an attached 10- to 12-mL syringe was used to obtain specimens, which were transferred onto glass slides for fixation while the remaining aspirate was rinsed into CytoLyt (Hologic, Marlborough, Massachusetts) solution to enable cell block formation for immunocytologic or molecular studies. Rapid on-site assessments and preliminary diagnoses were rendered by departmental cytopathologists or fellows. All CT-guided TNAB specimens were subsequently reviewed for this study by a board-certified cytopathologist. In a few patients for whom a cytologic diagnosis was not evident by on-site rapid staining using toluidine blue and light microscopic examination, one to three core biopsy specimens were attempted using a 20-gauge Temno automated cutting biopsy needle (CareFusion, San Diego, California), with the specimens placed into formalin for histologic analysis. Core biopsy and surgical resection specimens were signed out routinely by surgical pathologists and reviewed for this study by a board-certified surgical pathologist with fellowship training in pulmonary pathology. Review of both cytologic and surgical specimens was conducted with reviewers blinded to the original diagnoses. Standard morphologic criteria were used to evaluate both the cytologic and the histologic specimens. Histologic diagnoses based on core biopsy and surgical resection specimens were made according to the terminology recommended by the International Association for the Study of Lung Cancer (5). Cytologic diagnoses included “positive for malignancy” (eg, adenocarcinoma), “suspicious for malignancy,” “atypical,” and “negative for malignancy.” In accordance with cytopathology convention, both “positive for malignancy” and “suspicious for malignancy” were considered positive CT-guided TNAB results, whereas “atypical” and “negative for malignancy” were considered negative CT-guided TNAB results. In instances in which evidence of malignancy was not identified on cytology, a general diagnosis of “negative for malignancy” was provided, unless additional features suggestive of a specific diagnosis were identified (eg, fungal hyphae). When a tissue diagnosis was unavailable, repeat chest CT studies were reviewed in conjunction with clinical follow-up to determine the final diagnosis as malignant or benign. Imaging findings used to characterize a lesion as malignant include progressive lesion growth, development of metastatic disease, and increase in lesion density. Imaging features used to characterize a lesion as benign include progressive decrease in size over time, lesion stability over a minimum of 3 years of follow-up, and complete lesion resolution. The rates of true-positive, true-negative, false-positive, and false-negative diagnoses were determined by comparing the CT-guided TNAB diagnosis with the final diagnosis as determined by either histology (core biopsy or surgical specimens) or radiologic or clinical follow-up.
342
’
CT-Guided TNAB of Subsolid Nodules
Maxwell et al
For the purposes of this study, all cases in which a diagnosis of “positive for malignancy” or “suspicious for malignancy” was made by CT-guided TNAB and in which this diagnosis was confirmed by histologic or radiologic or clinical follow-up were categorized as truepositive cases, regardless of agreement between cancer subtypes (eg, adenocarcinoma by CT-guided TNAB and squamous cell carcinoma by surgical pathology). Truenegative cases included cases that were diagnosed as “negative for malignancy” and “atypical” by CT-guided TNAB and found on follow-up to be benign. Falsepositive cases were defined as cases with a diagnosis of malignancy by CT-guided TNAB in which there was no evidence of malignancy on tissue pathology or radiographic or clinical follow-up. Cases were categorized as false-negative cases when CT-guided TNAB failed to identify a malignancy that was discovered by subsequent histologic evaluation or radiologic or clinical follow-up. Using these categories, the following diagnostic indices were calculated: diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. Together, these measures were used to quantify the utility of CT-guided TNAB in the evaluation of subsolid lung lesions. During the specified period of investigation, a preliminary list of 1,365 CT-guided procedures was compiled, with entries filtered according to the key word “nodule.” From this list, 68 cases were identified that specifically referenced the terms “subsolid” or “ground glass.” On review of the pertinent CT study, 28 of these cases were determined not to meet full inclusion criteria (eg, purely solid lesions) and were excluded from analysis. In eight cases, no final diagnosis could be identified, yielding a final cohort of 32 patients with persistent subsolid lung lesions (Fig 1). The final group was composed of 18 men and 14 women with a mean age of 67.1 years ⫾ 9.6 (range, 52– 86 y). The mean time between chest CT scan performed before biopsy and CT-guided TNAB was 23.8 days ⫾ 68 subsolid nodules in 68 paents Review of CT studies subsolid nodule?
No
28 paentsexcluded from analysis
Yes
40 paents
Final diagnosis?
Yes
32 paentsStudy group
No 8 paentsexcluded from analysis
Figure 1. Flow chart depicting determination of the final patient cohort.
’
JVIR
12.7 (range, 7–62 d). The mean lesion diameter was 21 mm ⫾ 11 (range, 8–62 mm). The subsolid lesions were predominantly located in the upper lobes (23 cases; 71.9%). With regard to density, 5 lesions (15.6%) were classified as pure GG, 18 lesions (56.3%) were classified as GG-predominant (4 50% GG), and 9 lesions (28.1%) were classified as solid-predominant (o 50% GG). Biopsy complications were noted in two cases (6.3%), both of which were small pneumothoraces not requiring treatment (Society of Interventional Radiology Minor Complications Type A). No bleeding complications were identified.
RESULTS A CT-guided TNAB diagnosis was rendered in all 32 cases, and this diagnosis was confirmed by histology in 27 cases (84.4%). Of these, 22 cases (68.8%) represented tissue from resected surgical specimens, and 5 cases (15.6%) represented tissue from core needle biopsies collected at the time of CT-guided TNAB. In two cases, both a resection specimen and a core needle biopsy sample were obtained; in these cases, the resection specimens were used to render a final diagnosis. The remaining five diagnoses were made by radiologic or clinical follow-up. Details regarding all cases—including pathologic diagnoses—are provided in Table 1. Among the 32 cases, 28 malignant lesions were identified (Fig 2, Fig E1a, b [available online at www. jvir.org]). Agreement between CT-guided TNAB and tissue pathology or follow-up was demonstrated in 25 of these cases (true-positive cases). In the remaining three cases of malignancy, a nonmalignant diagnosis was rendered by CT-guided TNAB on the available material obtained (false-negative cases). These three patients all had a nonspecific cytologic result and were given the option for close follow-up or surgical resection and opted for surgical resection of their lesions. In all three of these cases, the subsolid lesions were r 15 mm in diameter with 4 50% GG component (GG-predominant lesion). Four benign lesions were identified among the 32 cases. In three of these cases, complete lesion resolution was noted after CT-guided TNAB, and no further evaluation was undertaken. In the fourth case, a diagnosis of benign alveolar tissue was rendered by core needle biopsy, and lesion resolution was subsequently noted on follow-up chest CT scan 6 months after CT-guided TNAB. Agreement between CT-guided TNAB and tissue pathology or follow-up was noted in all four cases (true-negative diagnoses). A malignant diagnosis was not rendered for a benign process (false-positive diagnosis) in any of the cases. Using the diagnostic frequencies and pathologic criteria detailed previously, the overall diagnostic accuracy of CT-guided TNAB in the evaluation of subsolid lung
Volume 25
Table 1 . Subsolid Nodules That Underwent CT-Guided TNAB
65/M 74/M 67/F
1.9 3.1 2.5
GG o 50% GG o 50% GG 4 50%
NSCC AC, MD AC, MD
8
62/M
2.1
GG o 50%
AC, MD
9
73/F
1.1
GG 4 50%
10
86/M
4.2
GG 4 50%
11
54/F
1.3
GG 4 50%
12
62/M
1.4
GG 4 50%
13
79/M
2.4
GG 4 50%
14
59/M
1.2
GG 4 50%
15 16 17
55/F 63/M 74/M
2.6 1.8 4.0
GG 4 50% GG 4 50% GG o 50%
18
63/F
1.5
GG 4 50%
19
70/F
3.2
GG 4 50%
AC, MD
20
72/M
1.7
GG o 50%
AC, WD
21 22 23 24
79/M 51/M 67/F 70/F
2.0 1.2 0.9 0.8
GG 4 50% Pure GG GG 4 50% Pure GG
AC, WD AC, MD Suspicious for AC
25
66/M
1.5
Pure GG
Suspicious for AC, WD
50% 50% 50% 50%
Malignant AC, MD NSCC AC, MD AC, MD
Benign
Negative for malignant cells Suspicious for AC, WD Rare atypical cells Suspicious for AC, WD AC, WD Suspicious for AC, WD AC, WD AC, WD AC, MD/PD
Negative for malignant cells
Malignant Benign ADSCC, MD SqCC, MD AC, MD, INV, acinar pattern AC, MD, INV, papillary pattern SqCC, MD/PD AC, WD, INV, acinar pattern AC, WD/MD, INV, acinar pattern AC, MD/PD, INV, papillary pattern AC, MD/PD, INV, acinar pattern
Surgical Surgical Surgical Surgical
resection resection resection resection
Diagnostic Category TP X X X X
Surgical resection Surgical resection Surgical resection
X X X
Surgical resection
X
Surgical resection
Surgical resection
Nonmucinous MIA, WD/MD
Surgical resection
AC, WD/MD, INV, acinar pattern AC, MD, INV, papillary pattern AC, MD, INV, acinar pattern
Surgical resection
X
Surgical resection
X
Surgical resection
X
Mucinous AC, INV Nonmucinous MIA AC, MD/PD, INV, acinar, papillary, and micropapillary patterns Nonmucinous AIS
Surgical resection Surgical resection Surgical resection
X X X
AC, MD/PD, INV, acinar, papillary, and micropapillary patterns AC, MD, INV, papillary pattern Nonmucinous MIA, WD AC, PD, INV, acinar pattern AC, WD, lepidic pattern
Surgical resection
X
Surgical resection
X
AC, WD, lepidic pattern
FN
X X
Surgical resection
Benign alveolar tissue
FP
X
AC, PD, INV, acinar pattern
Rare atypical cells
TN
2014
5 6 7
4 o 4 4
Diagnostic Method
’
GG GG GG GG
Final Diagnosis
March
1.4 1.6 2.3 2.7
CT-Guided TNAB Diagnosis
’
63/M 80/M 55/F 77/M
Lesion Density
Number 3
1 2 3 4
Case No.
’
Age (y)/ Sex
Maximum Lesion Diameter (cm)
X
Surgical resection X Surgical resection X Core needle biopsy X Core needle biopsy
X
Core needle biopsy X
343
(Continued)
344 ’
Case No.
Age (y)/ Sex
Maximum Lesion Diameter (cm)
Lesion Density
CT-Guided TNAB Diagnosis Malignant
1.6 1.6
Pure GG Pure GG
Benign
26 27
54/F 54/F
28
74/F
1.5
GG 4 50%
AC, MD Suspicious for AC, WD AC, MD/PD
29
86/F
2.3
Pure GG
AC, MD
30
71/M
1.1
GG o 50%
Fungal hyphae
31
54/F
1.9
GG o 50%
32
66/M
6.2
GG 4 50%
Negative for malignant cells Negative for malignant cells
Final Diagnosis Malignant
Diagnostic Method
Benign
Diagnostic Category TP
AC, WD, lepidic pattern AC, WD, lepidic pattern
Core needle biopsy X Core needle biopsy X
New metastatic disease to brain Increase in density from pure GG to 4 50% solid component
Clinical/imaging follow-up Clinical/imaging follow-up Subsequent lesion resolution Subsequent lesion resolution Subsequent lesion resolution
TN
FP
FN
CT-Guided TNAB of Subsolid Nodules
Table 1 . Subsolid Nodules That Underwent CT-Guided TNAB (Continued)
X X
Clinical/imaging follow-up
X
Clinical/imaging follow-up
X
Clinical/imaging follow-up
X
AC ¼ adenocarcinoma; ADSCC ¼ adenosquamous carcinoma; AIS ¼ adenocarcinoma in situ; CT ¼ computed tomography; GG ¼ ground-glass; F ¼ female; INV ¼ invasive; M ¼ male; MD ¼ moderately differentiated; MIA ¼ minimally invasive adenocarcinoma; NSCC ¼ non–small cell carcinoma; PD ¼ poorly differentiated; SCC ¼ squamous cell carcinoma; TNAB ¼ transthoracic needle aspiration biopsy; WD ¼ well-differentiated.
Maxwell et al ’
JVIR
Volume 25
’
Number 3
’
March
’
2014
Figure 2. Adenocarcinoma in part-solid nodule diagnosed by CT-guided TNAB. Axial thin-section CT image of the chest through the upper lobes at lung windows in a 51-year-old man shows a 12-mm GG-predominant nodule (arrow) in the right upper lobe.
lesions was 90.6%. The sensitivity of CT-guided TNAB was 89.3%, with three false-negative diagnoses (Fig 3, Fig E2a, b [available online at www.jvir.org]). The specificity and positive predictive value of CT-guided TNAB for the evaluation of subsolid lesions were both 100% because all patients with positive cytology by CTguided TNAB were subsequently given a diagnosis of malignant disease. The negative predictive value of CTguided TNAB was 57.1%.
DISCUSSION Subsolid lung nodules, defined as spherical lesions o 3 cm in diameter with a component that is less than soft tissue density, are common findings on thin-section CT of the lungs. Based on results from CT screening studies and careful radiologic and pathologic correlative studies, it is now recognized that most persistent subsolid lung nodules 4 8 mm in diameter that show either pure GG attenuation or mixed solid and GG attenuation (ie, part-solid nodules) (6) as seen on thin-section CT scan reflect preinvasive or invasive adenocarcinoma (3,7). In patients undergoing low-dose CT screening for lung cancer, the incidence of solitary subsolid nodules of pure GG or part-solid attenuation has been shown to be approximately 48% of screening-detected cancers (8). Henschke et al (2) showed 63% of part-solid nodules and 18% of pure GG nodules detected on CT screening were malignant. Malignancy is more likely in subsolid lesions if the lesion is new or increasing in size or density or developing lobulated or spiculated margins compared with prior thin-section CT studies (9). Although guidelines for the appropriate management of subsolid lung nodules have been published more recently (10,) these guidelines primarily address the follow-up of lesions thought likely to be benign. Although biopsy or resection is recommended for growing lesions or lesions with a solid component 4 5 mm in diameter, there is
345
Figure 3. False-negative CT-guided TNAB in a 54-year-old woman. Axial thin-section CT image of the chest through the upper lobes at lung windows shows a 13-mm GG-predominant nodule (arrow) in the right upper lobe.
limited guidance on optimal methods for biopsy of subsolid lesions thought likely to reflect preinvasive or invasive malignancy. CT-guided TNAB has been shown to have a high sensitivity and positive predictive value 4 90% for sampling of malignant solid solitary pulmonary nodules (11). However, for subsolid nodules, the reported sensitivity is approximately 65% (12,13). Our results suggest that CT-guided TNAB has a high sensitivity for the diagnosis of persistent subsolid nodules found to reflect malignancy, higher than that found by Shimizu et al (12) and similar to the published sensitivity of transthoracic needle biopsy for solid malignant nodules. The three patients in our series with false-negative results at CT-guided TNAB all had small lesions (o 1.5 cm in diameter) with each showing o 50% solid attenuation on thin-section CT. These findings are consistent with the results of Shimizu et al (12), who found that in resected malignant subsolid nodules, larger lesion diameter and greater relative percentage of solid attenuation in part-solid nodules were directly associated with higher sensitivity on preoperative CT-guided needle biopsy. Given that two of the three lesions in our series were found to represent adenocarcinoma in situ and minimally invasive adenocarcinoma at resection, it is unlikely that a delay in diagnosis owing to a falsenegative CT-guided TNAB diagnosis would have led to a deleterious effect on staging or management given the typically indolent nature of these lesions. The 100% positive predictive value of CT-guided TNAB for persistent subsolid nodules in our series has important implications for patient treatment because all patients determined to have malignant subsolid nodules based on cytologic analysis of CT-guided TNAB specimens were correctly characterized as having malignancy and were offered appropriate treatment for localized lung cancer. Our low rate of biopsy-related pneumothorax (6.3%) confirms that CT-guided TNAB of subsolid lesions is a safe, minimally invasive procedure with a complication
346
’
CT-Guided TNAB of Subsolid Nodules
rate similar to the lowest published rates for CT-guided TNAB of solid lung nodules. Either CT-guided TNAB or core specimens could be available for molecular markers should resection not be performed. Because it has been suggested that patients with subsolid nodules found to reflect adenocarcinoma in situ and minimally invasive adenocarcinoma have an almost 100% 5-year survival after limited resection (14), the accurate identification of malignant cells in these lesions, which are typically subsolid in attenuation at thin-section CT, allows for direct therapeutic wedge resection of the lesion as definitive management without the need for intraoperative frozen section analysis. For patients who are poor candidates for surgical resection because of comorbid disease or lesion inaccessibility to thoracoscopic resection or patients who refuse surgical management, local treatment using stereotactic or conventional radiation therapy or thermal ablation techniques including radiofrequency ablation, cryotherapy, and microwave ablation may be considered for achieving local control of disease. In our study, 6 of 28 patients (21%) found ultimately to have malignancy did not undergo surgical resection, and in these patients, CT-guided TNAB was the only method by which a diagnosis of malignancy was established and treatment recommendations could be rendered. Our study has several limitations. The first is its retrospective nature, which may have introduced a degree of selection bias. The second limitation is the small number of patients evaluable, although the number is comparable to other published studies on the utility of TNAB in the evaluation of subsolid lung lesions. Finally, our results reflect those at a single academic medical center and may not be widely applicable to centers not experienced in CT-guided TNAB and cytopathologic analysis of lung biopsy specimens. Nevertheless, despite these limitations, the results suggest a role for TNAB with cytologic analysis in the evaluation of subsolid lung lesions. In conclusion, although future studies should provide better information regarding the natural history of malignant subsolid lesions in specific patient cohorts,
Maxwell et al
’
JVIR
the current standard of care indicates that patients with malignant subsolid lesions are best offered treatment. Our findings indicate that CT-guided TNAB is a sensitive and safe minimally invasive procedure that provides for the correct characterization of malignant subsolid nodules, guiding appropriate treatment in affected patients.
REFERENCES 1. Austin JH, Garg K, Aberle D, et al. Radiologic implications of the 2011 classification of adenocarcinoma of the lung. Radiology 2013; 266:62–71. 2. Henschke CI, Yankelevitz DF, Mirtcheva R, et al. CT screening for lung cancer: frequency and significance of part-solid and nonsolid nodules. AJR Am J Roentgenol 2002; 178:1053–1057. 3. Kim HY, Shim YM, Lee KS, Han J, Yi CA, Kim YK. Persistent pulmonary nodular ground-glass opacity at thin-section CT: histopathologic comparisons. Radiology 2007; 245:267–275. 4. Yang PS, Lee KS, Han J, Kim EA, Kim TS, Choo IW. Focal organizing pneumonia: CT and pathologic findings. J Korean Med Sci 2001; 16: 573–578. 5. 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 2011; 6:244–285. 6. Hansell DM, Bankier AA, Macmahon H, Mcloud TC, Müller NL, Remy J. Fleischner Society: glossary of terms for thoracic imaging. Radiology 2008; 246:697–722. 7. Nakata M, Saeki H, Takata I, et al. Focal ground-glass opacity detected by low-dose helical CT. Chest 2002; 121:1464–1467. 8. Hasegawa M, Sone S, Takashima S, et al. Growth rate of small lung cancers detected on mass CT screening. Br J Radiol 2000; 73: 1252–1259. 9. Lee HJ, Goo JM, Lee CH, et al. Predictive CT findings of malignancy in ground-glass nodules on thin-section chest CT: the effects on radiologist performance. Eur Radiol 2009; 19:552–560. 10. 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 2013; 266:304–317. 11. Stanley JH, Fish GD, Andriole JG, et al. Lung lesions: cytologic diagnosis by fine-needle biopsy. Radiology 1987; 162:389–391. 12. Shimizu K, Ikeda N, Tsuboi M, Hirano T, Kato H. Percutaneous CTguided fine needle aspiration for lung cancer smaller than 2 cm and revealed by ground-glass opacity at CT. Lung Cancer 2006; 51:173–179. 13. Hur J, Lee HJ, Nam JE, et al. Diagnostic accuracy of CT fluoroscopyguided needle aspiration biopsy of ground-glass opacity pulmonary lesions. AJR Am J Roentgenol 2009; 192:629–634. 14. Sakao Y, Miyamoto H, Sakuraba M, et al. Prognostic significance of a histologic subtype in small adenocarcinoma of the lung: the impact of nonbronchioloalveolar carcinoma components. Ann Thorac Surg 2007; 83:209–214.
Volume 25
’
Number 3
’
March
’
2014
346.e1
Figure E1. Adenocarcinoma in part-solid nodule diagnosed by CT-guided TNAB. (a) Cytologic smear from CT-guided TNAB shows a loosely cohesive cluster of tumor cells (arrow) with eccentrically placed nuclei, prominent nucleoli, and delicate foamy cytoplasm, findings consistent with adenocarcinoma. (Papanicolaou, 20.) (b) Histologic section of lung wedge resection shows adenocarcinoma with a solid architectural pattern and invasion of adjacent normal lung parenchyma (arrow). (Hematoxylin-eosin, 20.)
Figure E2. False-negative CT-guided TNAB in a 54-year-old woman. (a) Cytologic smear from CT-guided TNAB shows histiocytes (arrow), rare reactive bronchial cells (arrowhead), and obscuring blood. (Papanicolaou, 40.) (b) Histologic section from pulmonary wedge resection shows minimally invasive nonmucinous adenocarcinoma with a lepidic growth (arrow) pattern. (Hematoxylin-eosin, 10.)
CT-Guided Transthoracic Needle Aspiration Biopsy of Subsolid Lung Lesions Aaron W.P. Maxwell, BSc, Jeffrey S. Klein, MD, Kossivi Dantey, MD, Sharon L. Mount, MD, Kelly J. Butnor, MD, and Gladwyn Leiman, MD
ABSTRACT Purpose: To assess the diagnostic performance of computed tomography (CT)–guided transthoracic needle aspiration biopsy (TNAB) in the evaluation of persistent subsolid lung lesions. Materials and Methods: A retrospective review of all CT-guided TNABs performed at a single institution from January 2002 to November 2012 was conducted to identify patients with persistent subsolid lung lesions. The diagnostic performance of CTguided TNAB was assessed through comparison of cytologic diagnoses with core needle biopsy, surgical resection, or imaging and clinical follow-up. The cytologic, histologic, and imaging features of each lesion were characterized, and CT-guided TNAB complications were recorded. Results: In 32 patients, a diagnosis of benign or malignant disease was identified through evaluation of pathologic or follow-up data. There were 18 men and 14 women, with a mean age of 67.1 years ⫾ 9.6 (range, 52–86 y). The mean lesion diameter was 21 mm ⫾ 11 (range, 8–62 mm). A final diagnosis of malignancy was made in 28 cases (87.5%); four benign lesions were also diagnosed. The overall sensitivity of CT-guided TNAB in the evaluation of these lesions was 89.2%, and the specificity and positive predictive value were 100%. Two pneumothoraces (6.3%) were identified. Conclusions: Among patients with subsolid lung lesions, CT-guided TNAB is safe and shows high sensitivity. The high specificity and positive predictive value of the procedure allow for definitive treatment decisions to be made for most patients.
ABBREVIATIONS GG = ground-glass, TNAB = transthoracic needle aspiration biopsy
Although there has been considerable interest in thinsection computed tomography (CT) characterization of subsolid lung nodules (defined as spherical lesions o 3 cm in diameter with a component that is less than soft tissue density) over the past 10 years (1), particularly in light of increasing detection rates on multidetector CT performed for lung cancer screening (2) or in the
From the Department of Radiology, The University of Vermont College of Medicine (A.W.P.M.); and Departments of Radiology (J.S.K.) and Pathology (K.D., S.L.M., K.J.B., G.L.), Fletcher Allen Health Care, 111 Colchester Avenue, Burlington, VT 05401. Received September 7, 2013; final revision received November 2, 2013; accepted November 27, 2013. Address correspondence to J.S.K.; E-mail: [email protected] None of the authors have identified a conflict of interest. Figures E1 and E2 are available online at www.jvir.org. & SIR, 2014 J Vasc Interv Radiol 2014; 25:340–346 http://dx.doi.org/10.1016/j.jvir.2013.11.037
evaluation of chest disease, accurate diagnosis of these lesions remains difficult. This difficulty is due to the overlap in appearance on thin-section CT of preinvasive conditions, such as atypical adenomatous hyperplasia and adenocarcinoma in situ, with invasive adenocarcinoma as well as with inflammatory processes that may have a nodular subsolid appearance during the course of resolution (3,4). Although there is increasing recognition that a significant percentage of stable, subsolid lesions represent preinvasive or invasive adenocarcinoma (2,3), the accurate preoperative diagnosis of subsolid malignant lesions is important, particularly because some lesions are not readily amenable to thoracoscopic biopsy, and a substantial percentage of patients with subsolid malignant lesions are poor candidates for surgical resection because of comorbidities. There is scant information on the role of minimally invasive techniques such as CT-guided transthoracic needle aspiration biopsy (TNAB) cytology in the pathologic characterization of these lesions. The purpose of this study
Volume 25
’
Number 3
’
March
’
2014
was to assess the diagnostic performance of CT-guided TNAB in the diagnosis of persistent subsolid lung lesions.
MATERIALS AND METHODS This study, which was compliant with the Health Insurance Portability and Accountability Act, was approved by our institutional review board with a waiver for informed patient consent. A retrospective search of all CT-guided biopsy procedures performed at our institution from January 2002 to November 2012 was conducted using Current Procedure Terminology codes. Results were filtered by the key word “nodule” to yield a preliminary list, and from this list potentially relevant cases were identified via manual search of the associated procedure report for the terms “subsolid” and “groundglass” (GG). When identified, all such cases were transferred into a database, and any associated chest CT images obtained before biopsy were analyzed on thinsection series to determine the presence or absence of persistent, solitary subsolid lesions. All cases featuring lesions not meeting criteria as “subsolid” were excluded from analysis. For the purposes of this investigation, lesion “persistence” was defined as repeated CT demonstration of the lesion under evaluation for at least 3 months after initial detection. All identified subsolid nodules were characterized as to lobe within the lung, maximum lesion diameter in any plane, and estimated percent density (pure GG, 4 50% GG component [predominant GG], or o 50% GG component [predominant solid]). Also recorded for each case was the presence or absence of biopsy-associated complications, in particular, pneumothoraces requiring inhospital observation or treatment and bleeding requiring hospital admission. All patients underwent CT-guided TNAB with or without concomitant core needle biopsy. The CTguided TNAB procedure was performed by two boardcertified thoracic radiologists with 24 years and 8 years of experience with CT-guided TNAB as well as cardiothoracic radiology fellows and a physician’s assistant under direct supervision. All biopsies were performed under conscious sedation using CT fluoroscopy with a step-and-shoot mode controlled by the radiologist performing the biopsy. The patient was placed in a recumbent position on the CT table, with a needle approach chosen that used the shortest vertical pathway to the lesion to be sampled. A coaxial needle (Cook, Inc, Bloomington, Indiana) placement technique was used with a combination of an ultra-thin-walled needle placed via a single pleural puncture during patient breath hold in normal end expiration (ie, at functional residual capacity), with the outer needle placed to the edge of the lesion, followed by at least three aspiration biopsy specimens obtained during patient breath hold using a 22-gauge Chiba needle (INRAD Inc, Kentwood,
341
Michigan). A rotatory and to-and-fro motion with aspiration using an attached 10- to 12-mL syringe was used to obtain specimens, which were transferred onto glass slides for fixation while the remaining aspirate was rinsed into CytoLyt (Hologic, Marlborough, Massachusetts) solution to enable cell block formation for immunocytologic or molecular studies. Rapid on-site assessments and preliminary diagnoses were rendered by departmental cytopathologists or fellows. All CT-guided TNAB specimens were subsequently reviewed for this study by a board-certified cytopathologist. In a few patients for whom a cytologic diagnosis was not evident by on-site rapid staining using toluidine blue and light microscopic examination, one to three core biopsy specimens were attempted using a 20-gauge Temno automated cutting biopsy needle (CareFusion, San Diego, California), with the specimens placed into formalin for histologic analysis. Core biopsy and surgical resection specimens were signed out routinely by surgical pathologists and reviewed for this study by a board-certified surgical pathologist with fellowship training in pulmonary pathology. Review of both cytologic and surgical specimens was conducted with reviewers blinded to the original diagnoses. Standard morphologic criteria were used to evaluate both the cytologic and the histologic specimens. Histologic diagnoses based on core biopsy and surgical resection specimens were made according to the terminology recommended by the International Association for the Study of Lung Cancer (5). Cytologic diagnoses included “positive for malignancy” (eg, adenocarcinoma), “suspicious for malignancy,” “atypical,” and “negative for malignancy.” In accordance with cytopathology convention, both “positive for malignancy” and “suspicious for malignancy” were considered positive CT-guided TNAB results, whereas “atypical” and “negative for malignancy” were considered negative CT-guided TNAB results. In instances in which evidence of malignancy was not identified on cytology, a general diagnosis of “negative for malignancy” was provided, unless additional features suggestive of a specific diagnosis were identified (eg, fungal hyphae). When a tissue diagnosis was unavailable, repeat chest CT studies were reviewed in conjunction with clinical follow-up to determine the final diagnosis as malignant or benign. Imaging findings used to characterize a lesion as malignant include progressive lesion growth, development of metastatic disease, and increase in lesion density. Imaging features used to characterize a lesion as benign include progressive decrease in size over time, lesion stability over a minimum of 3 years of follow-up, and complete lesion resolution. The rates of true-positive, true-negative, false-positive, and false-negative diagnoses were determined by comparing the CT-guided TNAB diagnosis with the final diagnosis as determined by either histology (core biopsy or surgical specimens) or radiologic or clinical follow-up.
342
’
CT-Guided TNAB of Subsolid Nodules
Maxwell et al
For the purposes of this study, all cases in which a diagnosis of “positive for malignancy” or “suspicious for malignancy” was made by CT-guided TNAB and in which this diagnosis was confirmed by histologic or radiologic or clinical follow-up were categorized as truepositive cases, regardless of agreement between cancer subtypes (eg, adenocarcinoma by CT-guided TNAB and squamous cell carcinoma by surgical pathology). Truenegative cases included cases that were diagnosed as “negative for malignancy” and “atypical” by CT-guided TNAB and found on follow-up to be benign. Falsepositive cases were defined as cases with a diagnosis of malignancy by CT-guided TNAB in which there was no evidence of malignancy on tissue pathology or radiographic or clinical follow-up. Cases were categorized as false-negative cases when CT-guided TNAB failed to identify a malignancy that was discovered by subsequent histologic evaluation or radiologic or clinical follow-up. Using these categories, the following diagnostic indices were calculated: diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. Together, these measures were used to quantify the utility of CT-guided TNAB in the evaluation of subsolid lung lesions. During the specified period of investigation, a preliminary list of 1,365 CT-guided procedures was compiled, with entries filtered according to the key word “nodule.” From this list, 68 cases were identified that specifically referenced the terms “subsolid” or “ground glass.” On review of the pertinent CT study, 28 of these cases were determined not to meet full inclusion criteria (eg, purely solid lesions) and were excluded from analysis. In eight cases, no final diagnosis could be identified, yielding a final cohort of 32 patients with persistent subsolid lung lesions (Fig 1). The final group was composed of 18 men and 14 women with a mean age of 67.1 years ⫾ 9.6 (range, 52– 86 y). The mean time between chest CT scan performed before biopsy and CT-guided TNAB was 23.8 days ⫾ 68 subsolid nodules in 68 paents Review of CT studies subsolid nodule?
No
28 paentsexcluded from analysis
Yes
40 paents
Final diagnosis?
Yes
32 paentsStudy group
No 8 paentsexcluded from analysis
Figure 1. Flow chart depicting determination of the final patient cohort.
’
JVIR
12.7 (range, 7–62 d). The mean lesion diameter was 21 mm ⫾ 11 (range, 8–62 mm). The subsolid lesions were predominantly located in the upper lobes (23 cases; 71.9%). With regard to density, 5 lesions (15.6%) were classified as pure GG, 18 lesions (56.3%) were classified as GG-predominant (4 50% GG), and 9 lesions (28.1%) were classified as solid-predominant (o 50% GG). Biopsy complications were noted in two cases (6.3%), both of which were small pneumothoraces not requiring treatment (Society of Interventional Radiology Minor Complications Type A). No bleeding complications were identified.
RESULTS A CT-guided TNAB diagnosis was rendered in all 32 cases, and this diagnosis was confirmed by histology in 27 cases (84.4%). Of these, 22 cases (68.8%) represented tissue from resected surgical specimens, and 5 cases (15.6%) represented tissue from core needle biopsies collected at the time of CT-guided TNAB. In two cases, both a resection specimen and a core needle biopsy sample were obtained; in these cases, the resection specimens were used to render a final diagnosis. The remaining five diagnoses were made by radiologic or clinical follow-up. Details regarding all cases—including pathologic diagnoses—are provided in Table 1. Among the 32 cases, 28 malignant lesions were identified (Fig 2, Fig E1a, b [available online at www. jvir.org]). Agreement between CT-guided TNAB and tissue pathology or follow-up was demonstrated in 25 of these cases (true-positive cases). In the remaining three cases of malignancy, a nonmalignant diagnosis was rendered by CT-guided TNAB on the available material obtained (false-negative cases). These three patients all had a nonspecific cytologic result and were given the option for close follow-up or surgical resection and opted for surgical resection of their lesions. In all three of these cases, the subsolid lesions were r 15 mm in diameter with 4 50% GG component (GG-predominant lesion). Four benign lesions were identified among the 32 cases. In three of these cases, complete lesion resolution was noted after CT-guided TNAB, and no further evaluation was undertaken. In the fourth case, a diagnosis of benign alveolar tissue was rendered by core needle biopsy, and lesion resolution was subsequently noted on follow-up chest CT scan 6 months after CT-guided TNAB. Agreement between CT-guided TNAB and tissue pathology or follow-up was noted in all four cases (true-negative diagnoses). A malignant diagnosis was not rendered for a benign process (false-positive diagnosis) in any of the cases. Using the diagnostic frequencies and pathologic criteria detailed previously, the overall diagnostic accuracy of CT-guided TNAB in the evaluation of subsolid lung
Volume 25
Table 1 . Subsolid Nodules That Underwent CT-Guided TNAB
65/M 74/M 67/F
1.9 3.1 2.5
GG o 50% GG o 50% GG 4 50%
NSCC AC, MD AC, MD
8
62/M
2.1
GG o 50%
AC, MD
9
73/F
1.1
GG 4 50%
10
86/M
4.2
GG 4 50%
11
54/F
1.3
GG 4 50%
12
62/M
1.4
GG 4 50%
13
79/M
2.4
GG 4 50%
14
59/M
1.2
GG 4 50%
15 16 17
55/F 63/M 74/M
2.6 1.8 4.0
GG 4 50% GG 4 50% GG o 50%
18
63/F
1.5
GG 4 50%
19
70/F
3.2
GG 4 50%
AC, MD
20
72/M
1.7
GG o 50%
AC, WD
21 22 23 24
79/M 51/M 67/F 70/F
2.0 1.2 0.9 0.8
GG 4 50% Pure GG GG 4 50% Pure GG
AC, WD AC, MD Suspicious for AC
25
66/M
1.5
Pure GG
Suspicious for AC, WD
50% 50% 50% 50%
Malignant AC, MD NSCC AC, MD AC, MD
Benign
Negative for malignant cells Suspicious for AC, WD Rare atypical cells Suspicious for AC, WD AC, WD Suspicious for AC, WD AC, WD AC, WD AC, MD/PD
Negative for malignant cells
Malignant Benign ADSCC, MD SqCC, MD AC, MD, INV, acinar pattern AC, MD, INV, papillary pattern SqCC, MD/PD AC, WD, INV, acinar pattern AC, WD/MD, INV, acinar pattern AC, MD/PD, INV, papillary pattern AC, MD/PD, INV, acinar pattern
Surgical Surgical Surgical Surgical
resection resection resection resection
Diagnostic Category TP X X X X
Surgical resection Surgical resection Surgical resection
X X X
Surgical resection
X
Surgical resection
Surgical resection
Nonmucinous MIA, WD/MD
Surgical resection
AC, WD/MD, INV, acinar pattern AC, MD, INV, papillary pattern AC, MD, INV, acinar pattern
Surgical resection
X
Surgical resection
X
Surgical resection
X
Mucinous AC, INV Nonmucinous MIA AC, MD/PD, INV, acinar, papillary, and micropapillary patterns Nonmucinous AIS
Surgical resection Surgical resection Surgical resection
X X X
AC, MD/PD, INV, acinar, papillary, and micropapillary patterns AC, MD, INV, papillary pattern Nonmucinous MIA, WD AC, PD, INV, acinar pattern AC, WD, lepidic pattern
Surgical resection
X
Surgical resection
X
AC, WD, lepidic pattern
FN
X X
Surgical resection
Benign alveolar tissue
FP
X
AC, PD, INV, acinar pattern
Rare atypical cells
TN
2014
5 6 7
4 o 4 4
Diagnostic Method
’
GG GG GG GG
Final Diagnosis
March
1.4 1.6 2.3 2.7
CT-Guided TNAB Diagnosis
’
63/M 80/M 55/F 77/M
Lesion Density
Number 3
1 2 3 4
Case No.
’
Age (y)/ Sex
Maximum Lesion Diameter (cm)
X
Surgical resection X Surgical resection X Core needle biopsy X Core needle biopsy
X
Core needle biopsy X
343
(Continued)
344 ’
Case No.
Age (y)/ Sex
Maximum Lesion Diameter (cm)
Lesion Density
CT-Guided TNAB Diagnosis Malignant
1.6 1.6
Pure GG Pure GG
Benign
26 27
54/F 54/F
28
74/F
1.5
GG 4 50%
AC, MD Suspicious for AC, WD AC, MD/PD
29
86/F
2.3
Pure GG
AC, MD
30
71/M
1.1
GG o 50%
Fungal hyphae
31
54/F
1.9
GG o 50%
32
66/M
6.2
GG 4 50%
Negative for malignant cells Negative for malignant cells
Final Diagnosis Malignant
Diagnostic Method
Benign
Diagnostic Category TP
AC, WD, lepidic pattern AC, WD, lepidic pattern
Core needle biopsy X Core needle biopsy X
New metastatic disease to brain Increase in density from pure GG to 4 50% solid component
Clinical/imaging follow-up Clinical/imaging follow-up Subsequent lesion resolution Subsequent lesion resolution Subsequent lesion resolution
TN
FP
FN
CT-Guided TNAB of Subsolid Nodules
Table 1 . Subsolid Nodules That Underwent CT-Guided TNAB (Continued)
X X
Clinical/imaging follow-up
X
Clinical/imaging follow-up
X
Clinical/imaging follow-up
X
AC ¼ adenocarcinoma; ADSCC ¼ adenosquamous carcinoma; AIS ¼ adenocarcinoma in situ; CT ¼ computed tomography; GG ¼ ground-glass; F ¼ female; INV ¼ invasive; M ¼ male; MD ¼ moderately differentiated; MIA ¼ minimally invasive adenocarcinoma; NSCC ¼ non–small cell carcinoma; PD ¼ poorly differentiated; SCC ¼ squamous cell carcinoma; TNAB ¼ transthoracic needle aspiration biopsy; WD ¼ well-differentiated.
Maxwell et al ’
JVIR
Volume 25
’
Number 3
’
March
’
2014
Figure 2. Adenocarcinoma in part-solid nodule diagnosed by CT-guided TNAB. Axial thin-section CT image of the chest through the upper lobes at lung windows in a 51-year-old man shows a 12-mm GG-predominant nodule (arrow) in the right upper lobe.
lesions was 90.6%. The sensitivity of CT-guided TNAB was 89.3%, with three false-negative diagnoses (Fig 3, Fig E2a, b [available online at www.jvir.org]). The specificity and positive predictive value of CT-guided TNAB for the evaluation of subsolid lesions were both 100% because all patients with positive cytology by CTguided TNAB were subsequently given a diagnosis of malignant disease. The negative predictive value of CTguided TNAB was 57.1%.
DISCUSSION Subsolid lung nodules, defined as spherical lesions o 3 cm in diameter with a component that is less than soft tissue density, are common findings on thin-section CT of the lungs. Based on results from CT screening studies and careful radiologic and pathologic correlative studies, it is now recognized that most persistent subsolid lung nodules 4 8 mm in diameter that show either pure GG attenuation or mixed solid and GG attenuation (ie, part-solid nodules) (6) as seen on thin-section CT scan reflect preinvasive or invasive adenocarcinoma (3,7). In patients undergoing low-dose CT screening for lung cancer, the incidence of solitary subsolid nodules of pure GG or part-solid attenuation has been shown to be approximately 48% of screening-detected cancers (8). Henschke et al (2) showed 63% of part-solid nodules and 18% of pure GG nodules detected on CT screening were malignant. Malignancy is more likely in subsolid lesions if the lesion is new or increasing in size or density or developing lobulated or spiculated margins compared with prior thin-section CT studies (9). Although guidelines for the appropriate management of subsolid lung nodules have been published more recently (10,) these guidelines primarily address the follow-up of lesions thought likely to be benign. Although biopsy or resection is recommended for growing lesions or lesions with a solid component 4 5 mm in diameter, there is
345
Figure 3. False-negative CT-guided TNAB in a 54-year-old woman. Axial thin-section CT image of the chest through the upper lobes at lung windows shows a 13-mm GG-predominant nodule (arrow) in the right upper lobe.
limited guidance on optimal methods for biopsy of subsolid lesions thought likely to reflect preinvasive or invasive malignancy. CT-guided TNAB has been shown to have a high sensitivity and positive predictive value 4 90% for sampling of malignant solid solitary pulmonary nodules (11). However, for subsolid nodules, the reported sensitivity is approximately 65% (12,13). Our results suggest that CT-guided TNAB has a high sensitivity for the diagnosis of persistent subsolid nodules found to reflect malignancy, higher than that found by Shimizu et al (12) and similar to the published sensitivity of transthoracic needle biopsy for solid malignant nodules. The three patients in our series with false-negative results at CT-guided TNAB all had small lesions (o 1.5 cm in diameter) with each showing o 50% solid attenuation on thin-section CT. These findings are consistent with the results of Shimizu et al (12), who found that in resected malignant subsolid nodules, larger lesion diameter and greater relative percentage of solid attenuation in part-solid nodules were directly associated with higher sensitivity on preoperative CT-guided needle biopsy. Given that two of the three lesions in our series were found to represent adenocarcinoma in situ and minimally invasive adenocarcinoma at resection, it is unlikely that a delay in diagnosis owing to a falsenegative CT-guided TNAB diagnosis would have led to a deleterious effect on staging or management given the typically indolent nature of these lesions. The 100% positive predictive value of CT-guided TNAB for persistent subsolid nodules in our series has important implications for patient treatment because all patients determined to have malignant subsolid nodules based on cytologic analysis of CT-guided TNAB specimens were correctly characterized as having malignancy and were offered appropriate treatment for localized lung cancer. Our low rate of biopsy-related pneumothorax (6.3%) confirms that CT-guided TNAB of subsolid lesions is a safe, minimally invasive procedure with a complication
346
’
CT-Guided TNAB of Subsolid Nodules
rate similar to the lowest published rates for CT-guided TNAB of solid lung nodules. Either CT-guided TNAB or core specimens could be available for molecular markers should resection not be performed. Because it has been suggested that patients with subsolid nodules found to reflect adenocarcinoma in situ and minimally invasive adenocarcinoma have an almost 100% 5-year survival after limited resection (14), the accurate identification of malignant cells in these lesions, which are typically subsolid in attenuation at thin-section CT, allows for direct therapeutic wedge resection of the lesion as definitive management without the need for intraoperative frozen section analysis. For patients who are poor candidates for surgical resection because of comorbid disease or lesion inaccessibility to thoracoscopic resection or patients who refuse surgical management, local treatment using stereotactic or conventional radiation therapy or thermal ablation techniques including radiofrequency ablation, cryotherapy, and microwave ablation may be considered for achieving local control of disease. In our study, 6 of 28 patients (21%) found ultimately to have malignancy did not undergo surgical resection, and in these patients, CT-guided TNAB was the only method by which a diagnosis of malignancy was established and treatment recommendations could be rendered. Our study has several limitations. The first is its retrospective nature, which may have introduced a degree of selection bias. The second limitation is the small number of patients evaluable, although the number is comparable to other published studies on the utility of TNAB in the evaluation of subsolid lung lesions. Finally, our results reflect those at a single academic medical center and may not be widely applicable to centers not experienced in CT-guided TNAB and cytopathologic analysis of lung biopsy specimens. Nevertheless, despite these limitations, the results suggest a role for TNAB with cytologic analysis in the evaluation of subsolid lung lesions. In conclusion, although future studies should provide better information regarding the natural history of malignant subsolid lesions in specific patient cohorts,
Maxwell et al
’
JVIR
the current standard of care indicates that patients with malignant subsolid lesions are best offered treatment. Our findings indicate that CT-guided TNAB is a sensitive and safe minimally invasive procedure that provides for the correct characterization of malignant subsolid nodules, guiding appropriate treatment in affected patients.
REFERENCES 1. Austin JH, Garg K, Aberle D, et al. Radiologic implications of the 2011 classification of adenocarcinoma of the lung. Radiology 2013; 266:62–71. 2. Henschke CI, Yankelevitz DF, Mirtcheva R, et al. CT screening for lung cancer: frequency and significance of part-solid and nonsolid nodules. AJR Am J Roentgenol 2002; 178:1053–1057. 3. Kim HY, Shim YM, Lee KS, Han J, Yi CA, Kim YK. Persistent pulmonary nodular ground-glass opacity at thin-section CT: histopathologic comparisons. Radiology 2007; 245:267–275. 4. Yang PS, Lee KS, Han J, Kim EA, Kim TS, Choo IW. Focal organizing pneumonia: CT and pathologic findings. J Korean Med Sci 2001; 16: 573–578. 5. 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 2011; 6:244–285. 6. Hansell DM, Bankier AA, Macmahon H, Mcloud TC, Müller NL, Remy J. Fleischner Society: glossary of terms for thoracic imaging. Radiology 2008; 246:697–722. 7. Nakata M, Saeki H, Takata I, et al. Focal ground-glass opacity detected by low-dose helical CT. Chest 2002; 121:1464–1467. 8. Hasegawa M, Sone S, Takashima S, et al. Growth rate of small lung cancers detected on mass CT screening. Br J Radiol 2000; 73: 1252–1259. 9. Lee HJ, Goo JM, Lee CH, et al. Predictive CT findings of malignancy in ground-glass nodules on thin-section chest CT: the effects on radiologist performance. Eur Radiol 2009; 19:552–560. 10. 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 2013; 266:304–317. 11. Stanley JH, Fish GD, Andriole JG, et al. Lung lesions: cytologic diagnosis by fine-needle biopsy. Radiology 1987; 162:389–391. 12. Shimizu K, Ikeda N, Tsuboi M, Hirano T, Kato H. Percutaneous CTguided fine needle aspiration for lung cancer smaller than 2 cm and revealed by ground-glass opacity at CT. Lung Cancer 2006; 51:173–179. 13. Hur J, Lee HJ, Nam JE, et al. Diagnostic accuracy of CT fluoroscopyguided needle aspiration biopsy of ground-glass opacity pulmonary lesions. AJR Am J Roentgenol 2009; 192:629–634. 14. Sakao Y, Miyamoto H, Sakuraba M, et al. Prognostic significance of a histologic subtype in small adenocarcinoma of the lung: the impact of nonbronchioloalveolar carcinoma components. Ann Thorac Surg 2007; 83:209–214.
Volume 25
’
Number 3
’
March
’
2014
346.e1
Figure E1. Adenocarcinoma in part-solid nodule diagnosed by CT-guided TNAB. (a) Cytologic smear from CT-guided TNAB shows a loosely cohesive cluster of tumor cells (arrow) with eccentrically placed nuclei, prominent nucleoli, and delicate foamy cytoplasm, findings consistent with adenocarcinoma. (Papanicolaou, 20.) (b) Histologic section of lung wedge resection shows adenocarcinoma with a solid architectural pattern and invasion of adjacent normal lung parenchyma (arrow). (Hematoxylin-eosin, 20.)
Figure E2. False-negative CT-guided TNAB in a 54-year-old woman. (a) Cytologic smear from CT-guided TNAB shows histiocytes (arrow), rare reactive bronchial cells (arrowhead), and obscuring blood. (Papanicolaou, 40.) (b) Histologic section from pulmonary wedge resection shows minimally invasive nonmucinous adenocarcinoma with a lepidic growth (arrow) pattern. (Hematoxylin-eosin, 10.)
