Original papers
2012, Vol. 14, no. 2, 115-119Abstract
Aim: a comparative evaluation of lung lesions in contact with the thoracic wall by transthoracic ultrasonography (US) and computed tomography (CT). Material and method: A retrospective analysis of the US and CT results was performed in 131 patients with thoracic masses. In all of these cases the histological diagnosis was established after a US guided transthoracic biopsy. A comparative analysis of the two methods was realized regarding the following items: lesion dimensions, presence of tumoral wall invasion, peritumoral atelectasis, presence of pleural effusion and incidence of post-biopsy complications.
Results: A number of 17 patients from the study group were diagnosed with benign lesions and 114 (87%) with malignancies.
US showed signs of wall invasion in 78 patients (68.42%), whereas CT revealed it in 83 of the patients (72.8%) from the ma- lignancies group. Intratumoral necrosis was diagnosed by US in 100 patients (87.71%) and by CT in 83 patients (72,8%), also from the 114 patients with malignant lesions. US found peritumoral atelectasis in 33 cases and CT in 38 cases. All parameters had good and very good correlation indexes between the methods (kappa =0,8; 0,6; 0,72; p < 0,001 in all cases). Conclusions:
In patients with thoracic masses in contact with the thoracic wall there is an overlap in the diagnosis performances of US and CT. The sensitivity and specificity of US in diagnosing wall invasion of malignant lesions is superior to that of CT.
Keywords:
Intrathoracic tumors in contact with the chest wall – ultrasonographic and computed tomography comparative evaluation
Romeo Chira, Alexandra Chira, Petru Adrian Mircea
Ist Medical Clinic, “Iuliu Haţieganu” University of Medicine and Pharmacy Cluj-Napoca, Romania
Received 02.03.2012 Accepted 08.04.2012 Med Ultrason
2012, Vol. 14, No 2, 115-119
Corresponding author: Romeo Ioan Chira Ist Medical Clinic 3-5 Clinicilor str
Cluj-Napoca 400006, Romania Phone: 0040742025557 Fax: 0040264590899 Email: [email protected]
Introduction
Thoracic masses with wall contact represent a fre- quent pathology that requires complex imaging studies, and often interventional procedures, in order to reach the complete diagnosis. In most cases, after a thoracic lesion is found on a thoracic X Ray, the next step is to perform a bronchoscopy and/or a CT exam. But mediastinal and pulmonary lesions often call for additional investiga- tions. Therefore, transthoracic ultrasonography (US) per- mits visualization of these lesions, their structural char- acterization, while offering suggestive elements for their malignant nature and for the differential diagnosis [1-3].
Furthermore US allows percutaneous guided biopsies with lower risks compared with the radiological guiding methods (fluoroscopy and CT) [4-6]. US advantages are numerous – accesibility (including bedside exams), low- er costs, no radiation exposure, and shorter biopsy time.
Regardless of these advantages, the use of US in ma- lignant lesions of the thorax is still low, in many centers the CT guided biopsy being preferred.
The purpose of this study was to realize a compara- tive analysis of US and CT in diagnosing thoracic lesions with parietal contact. In order to do this the following parameters were considered: lesions size, presence of tu- moral wall invasion, intratumoral necrosis, peritumoral atelectasis, and presence of pleural effusion, in a group of patients that benefited from histopathological diagnosis after US guided transthoracic biopsy.
Material and method
A retrospective analysis of the files of 131 patients (35 females and 96 males, ages between 20 and 82 years
old) with thoraco-pulmonary peripheral lesions, demon- strated by CT, was performed. The patients were hospi- talized in ‘Leon Daniello’ Pneumophtisiology Hospital or in the Ist Medical Clinic, Cluj-Napoca between 2004 and 2010. Some of the patients underwent bronchoscopy with or without biopsy, depending on the direct and indi- rect signs of the tumors. Subsequently the patients were examined through transthoracic US and US guided bi- opsy of the lesions. Before biopsy all patients signed an informed consent form.
The ultrasonographic exam and the biopsy were per- formed using one of the following equipments – Toshiba SSH 140 or General Electric LOGIQ S6 – by the same examiner (RC). The intrathoracic lesions were initially evaluated with a 3.5 MHz convex transducer or with a variable frequency transducer (4-5.5 MHz). In case of thoracic wall invasion the examination was completed with a 7.5 MHz or a 10-13 MHz linear transducer. For the biopsies a Bard Biopsy-Gun system was used, with Tru-cut needles, with diameters between 16 and 18 G.
The needle was chosen depending of the charactersit- ics of the lesion: for lesions with signs of wall invasion or larger than 4 cm higher caliber needles were usually used. In 127 patients a single passage was performed and in 4 patients two passages were needed. The length of the obtained fragments was measured and the histopatologi- cal examination was performed in the Pathology Lab of the Emergency County Hospital, Cluj-Napoca. After the biopsy the patients were monitored for at least 24 hours and in cases where complications were suspected the US was repeated or a thoracic x-ray was done.
The initial US assessment consisted of measurements of the largest diameters of the lesions, evaluation for in- tratumoral necrosis, peritumoral atelectasis, wall inva- sion and presence of pleural effusions and mediastinal lymphadenopathies. All this parameters were analyzed on the CT examination as well.
The invading character of the tumoral masses in the thoracic wall was established by US, based on the fol- lowing signs: absence of lesion mobility with respiratory cycles (absence of the lung sliding sign), an obtuse angle between a subpleural lesion and the parietal pleura (fig 1), and evidence of tumor invasion within the muscle lay- ers, bone or hypoderm [1,2,6-8]. Intratumoral necrosis was diagnosed based on the presence of transonic or hy- poechoic fluid areas within the lesion, with no evidence of vascular signal on the Doppler examination (fig 2) or based on the presence of cavities with or mixed content (fig 3). The corresponding aspect of the intratumoral necrosis as it seen on CT is presented in fig 4.
Peritumoral atelectasis was demonstrated in two ways: as areas of parenchyma consolidation, with trian-
Fig 1. Hypoechoic, homogeneous, ill circumscribed lung mass with wall invasion and obtuse angle with the parietal pleura – non-small cell carcinoma.
Fig 2. Pulmonary mass with unhomogeneous struc- ture and ill circumscribed hypoechoic areas inside – adenocarcinoma.
Fig 3. Left upper lobe central pulmonary mass with necrosis and mixed content, fluid and gas (hyper- echoic images with posterior shadowing in the up- per zone).
gular shape and the apex towards the hilum, where an obstructive tumoral nodule was visualized or as areas of consolidation adjacent to a tumoral mass in contact with the wall, usually also of triangular shape. Fluid broncho- grams and pulmonary vessels with perpendicular distri- bution towards the thoracic wall and radiating towards the atelectasis area were observed in both situations (fig 5) [1-3].
In many cases several types of alterations were asso- ciated: intratumoral necrosis and peritumoral atelectasis (fig 6), necrosis and wall invasion.
The data obtained by US were compared with those from the CT exam, which was considered the reference method, using the following statistics tools: Chi square test, T test, and Cohen test (for the evaluation of param- eters correlation between the two techniques).
Results
Seventeen (13%) patients from the study group were diagnosed with benign lesions and 114 (87%) with can- cers. The mean age of the patients with benign lesions was 54.53+/-13.6 years old and that of the patients with malignant lesions was 62.08+/- 10.13 years old.
The mean value of the maximum length of the lesions as measured by US was 7.38 cm, with limits of 2.8 cm and 11.5 cm. These findings were similar to those de- termined by CT – mean value 7.53 cm with limits of 3 cm and 12 cm – the resulted correlation kappa index was 0.888.
The sites of the lesions were: pulmonary – 106 cases (80.9%), mediastinal – 14 cases (10.7%) and pleuro-pari- etal – 11 cases (8.4%) (table I).
Thoracic biopsy was followed by a reduced number of complications. Thus, there where 3 patients (2.29%) who required analgesics for pain management, one pa- tient (0.76%) presented hemoptysis, and one patient (0.76%) developed hemothorax that necessitated drain- age tube placement. There were no cases of pneumot- horax clinically manifested or confirmed by CT or US.
The overall percentage of complications was 3.81 % (5 cases).
The histological diagnosis established after percuta- neous biopsy revealed the histopathological types pre- sented in table II. In the ‘other types of cancers’ category are included lymphomas, pleural mesotheliomas and me- diastinal tumors (thymomas, lymphomas).
The mean size of the benign lesions was 5.11 +/- 1.71 cm, and that of the malignant lesions was 7.71 +/- 1.85 cm.The US evaluation revealed signs of wall invasion in 78 (68.42 %) of the 114 patients from the malignant
Fig 4. Right upper lobe pulmonary mass with cen- tral necrosis – CT aspect – non-small cell carci- noma.
Fig 5. Peritumoral atelectasis with radiating, nor- mal vascularity distribution, adjacent to a hypoe- choic, unhomogeneous tumor that is hypovascular when compared with the surrounding parenchyma.
Fig 6. Tumoral mass with central necrosis (T) and peritumoral atelectasis (a) – triangular shape, well- circumscribed; Ao – thoracic aorta.
lesions group. The CT exam found this feature in 83 pa- tients (72.8%). The information obtained through the two techniques had a kappa correlation index of 0.8; p<0.001.
Tumoral necrosis was diagnosed by US in 100 pa- tients (87.71%) out of the 114 patients with malignant lesions. On the CT exam, necrosis was evident in 95 pa- tients (83.33%), the correlation index for this parameter being kappa=0.6; p<0.001.
US found peritumoral atelectasis in 33 patients, while CT showed it in 38 cases. There was a good correlation of the data given by the two methods – kappa = 0.672;
p<0.001.
Pleural effusions were diagnosed by US in 10 cases and by CT in 8 cases, out of the 131 patients from the study group. The correlation index between the two im- aging methods in this case is also very good (p<0.001).
Discussions
The sizes of the thoracic lesion as they were meas- ured by US are overlapping those obtained by CT. Our data revealed a very good correlation (kappa=0.888;
p<0.001) between the two imaging techniques. It is well known that just considering the intra and interobserver variations, in evaluating diameters and volumes, these
can reach significant differences as up to 7% interobserv- er and up to 4.8% intraobserver [9].
Between the patients with benign lesions and those with malignant lesions there are differences regarding the size of the masses, malignant lesions being significantly larger in size (p<0.001).
A particular aspect is that of the wall invasion deter- mined by the tumoral masses. The performance of US in diagnosing wall invasion is recognized (89-100%), being at least equal with that of CT (42-68%) [7,8,10,11]. In the present study the percentage of cases with wall invasion suggested by CT was higher than the one diagnosed by US, but on the US exam some of these tumors presented clear respiratory mobility, a sign that excludes invasion of the parietal pleura (7 cases). Therefore, the sensitiv- ity and specificity of US remain superior to those of CT, which may give a percentage of false positive results.
In our study group intratumoral necrosis was very frequently detected by both imaging techniques (over 80%). This high percentage is probably due to the large size of the tumoral masses that were examined (the mean size higher than 7 cm).
The sensitivity of US in diagnosing small amounts of fluid within the pleural space is excellent. In the study group US found a higher number of cases with pleural effusion than CT. The presence of pleural effusion in a patient with lung cancer is equivalent with pleural inva- sion and corresponds to a T4 mass [11]. Rarely pleural effusions may be of paraneoplastic nature, caused by ob- struction of the lymphatic drainage, atelectasis or hypo- proteinemia [12].
Within the study group, the complication rate after bi- opsy was low, especially that of pneumothorax, because of the wall contact of the tumoral masses and also prob- ably because of the large sizes of the lesions, which were over 2.8 cm. Literature data report higher percentages of pneumotorax occurence, especially in patients who un- derwent CT guided biopsies [4,6]. A higher rate of pneu- mothorax may also be caused by the higher number of passages, a higher distance for the needle to go through lung parenchyma to reach the lesion (especially over 40 mm) and a value of FEV1 < 70% in patients with chronic obstructive pulmonary disease.
As a result of evaluating US and CT it was conclud- ed that there is an agreement between the two imaging methods regarding tumor invasion, intratumoral necrosis and peritumoral atelectatsis in patients with thoracic le- sions in contact with the thoracic wall. In this category of patients the US visualisation of the lesions is very ac- curate providing structural details, bringing important elements for tumor staging, allowing a quick differential diagnosis with other types of lesions (pneumonia, atel- Table I. Topographic distribution of thoracic lesions in contact
with the thoracic wall
Lesion localisation Patients
Right upper lobe 29 (22.1%)
Right lower lobe 24 (18.3%)
Left upper lobe 28 (21.4%)
Left lower lobe 25 (19.1%)
Mediastinum 14 (10.7%)
Other localisations 11 (8.4%)
Total 131 (100%)
Table II. Histopathological types of thoracic lesions in contact with the thoracic wall (US guided biopsy).
Types of lesions Patients (number)
Benign 17 (13%)
Non small cell carcinoma 75 (57.3%)
Small cell carcinoma 5 (3.8%)
Other types of cancer 26 (19.8%)
Metastasis 8 (6.1%)
Total 131 (100%)
ectasis, pulmonary embolism, cystic lesions) [1-3]. The sensitivity and specificity of US in detecting lung cancer wall invasion is superior to that of CT, a finding that is in agreement with literature data [7,8].
The limits of the US examination are determined by the impossibility to visualize the pleura entirely, due to the presence of the skeletal components of the thorax, and the necessity to have a rather large wall contact in order to completely analyze a lesion or to perform a US guided biopsy. The limits of our study are the small num- bers of tumors (especially cancer) and the retrospective design.
Overall, US and CT offer comparable information, US keeping its well known advantages (lower cost, no radiation exposure, bed side examination possible). Ad- ditionally, US guided biopsies of lesions in contact with the thoracic wall are followed by a lower number of ad- verse reactions, with lower costs and a shorter execution time.
Conflict of interest: none
References
1. Koh DM, Burke S, Davies N, Padley SP. Transthoracic US of the chest: clinical uses and applications. Radiographics 2002; 22: e1.
2. Sartori S, Tombesi P. Emerging roles for transthoracic ultra- sonography in pulmonary diseases. World J Radiol 2010; 2:
203-214.
3. Beckh S, Bolcskei PL, Lessnau KD. Real-time chest ultra- sonography: a comprehensive review for the pulmonolo- gist. Chest 2002; 122: 1759-1773.
4. Sheth S, Hamper UM, Stanley DB, Wheeler JH, Smith PA.
US guidance for thoracic biopsy: a valuable alternative to CT. Radiology 1999; 210: 721-726.
5. Liao WY, Chen MZ, Chang YL, et al. US-guided transtho- racic cutting biopsy for peripheral thoracic lesions less than 3 cm in diameter. Radiology 2000; 217: 685-691.
6. Geraghty RP, Kee ST, McFarlane G, Razavi MK, Sze DY, Dake MD. CT-guided transthoracic needle aspiration bi- opsy of pulmonary nodules: needle size and pneumothorax rate. Radiology 2003; 229: 475–481.
7. Suzuki N, Saitoh T, Kitamura S. Tumor invasion of the chest wall in lung cancer: diagnosis with US. Radiology 1993; 187: 39-42.
8. Bandi V, Lunn W, Ernst A, Eberhardt R, Hoffmann H, Herth FJ. Ultrasound vs. CT in detecting chest wall invasion by tumor: a prospective study. Chest 2008; 133: 881-886.
9. Erasmus JJ, Gladish GW, Broemeling L, et al. Interobserver and intraobserver variability in measurement of non–small- cell carcinoma lung lesions: implications for assessment of tumor response. J Clin Oncol 2003; 21: 2574-2582.
10. Ohno Y, Hatabu H, Takenaka D, et al. CT-guided transtho- racic needle aspiration biopsy of small (< or = 20 mm) soli- tary pulmonary nodules. AJR Am J Roentgenol 2003; 180:
1665–1669.
11. Nakano N, Yasumitsu T, Kotake Y, Morino H, Ikezoe J.
Preoperative histological diagnosis of chest wall invasion by lung cancer using ultrasonically guided biopsy. J Thorac Cardiovasc Surg 1994; 107: 891-895.
12. Prosch H, Mathis G, Mostbeck GH. Percutaneous ultra- sound in diagnosis and staging of lung cancer. Ultraschall Med 2008; 29: 466-478.
