|
|
||||||||
Ann Thorac Surg 2004;77:1756-1762
© 2004 The Society of Thoracic Surgeons
a Department of Surgery, Charlottesville, VA, USA
b Department of Radiology, Charlottesville, VA, USA
c Department of Physics, University of Virginia Health System, Charlottesville, Virginia, USA
Accepted for publication October 24, 2003.
* Address reprint requests to Dr Daniel, Department of Surgery, Box 800679, University of Virginia Health System, Charlottesville, VA 22908-0679, USA.
e-mail: tmd5m{at}virginia.edu
| Abstract |
|---|
|
|
|---|
METHODS: This study consisted of two parts: a laboratory study in rats to determine the most suitable radiotracer, and a pilot study in humans to determine the feasibility of radiotracer lung nodule localization. The right lung of 12 rats was injected with a technetium 99m (Tc 99m) based radiotracer solution: 4 each with macroaggregated albumin (MAA), unfiltered sulfur colloid (SC), and pertechnetate (TcO4). Serial imaging was performed using a small animal gamma camera for 4 hours following injection. In 13 patients, computed tomographic (CT) guided injection of Tc 99m MAA solution was made into or adjacent to a pulmonary nodule suspicious for primary lung cancer. Gamma probe localization of the nodule was performed during subsequent thoracoscopic surgery.
RESULTS: In the animal model, MAA provided more precise localization than SC or TcO4 and was selected for the human study. In the human series, all 13 patients had successful gamma probe localization of their lesion. There were no radiologic or surgical complications. Seven of 13 nodules were malignant, and five of these nodules were stage IA primary lung carcinomas.
CONCLUSIONS: Computed tomographic-guided radiotracer localization of small or ill-defined pulmonary nodules using Tc 99 m MAA before thoracoscopic excisional biopsy is feasible and may make excisional biopsy the preferred management strategy for the management of small pulmonary nodules in patients at high risk for malignancy.
| Introduction |
|---|
|
|
|---|
Several lung nodule localization techniques have been previously reported to facilitate the successful thoracoscopic localization and excision of the above-described lesions. These include CT-guided methylene blue pleural localization [4, 5], preoperative percutaneously placed hooks of various design [68], tomographic-guided preoperative bronchoscopically placed barium markers with subsequent intraoperative fluoroscopically guided biopsy [9], preoperative CT-guided percutaneous placement of cyanoacrylate with subsequent intraoperative fluoroscopic-guided biopsy [10], and preoperative CT-guided placement of radioopaque Lipiodol (Mitsui Pharmaceutical Co, Tokyo, Japan) with subsequent fluoroscopic-guided biopsy [11, 12]. Chella [13] has described the preoperative percutaneous placement of a technetium radiotracer in or near the lung lesion using CT fluoroscopic guidance with subsequent thoracoscopic localization using an endoscopic radioprobe. Burdine [14] and Sugi [15] have described similar radiotracer localization techniques using different radiotracer solutions. The radiotracer solutions used by these three investigators are either unavailable routinely in North America or dissipated from the localization site too quickly to allow practical coordination with the subsequent surgical procedure.
The purpose of this work was to develop a reliable technique for the localization of small lung nodules and subsequent thoracoscopic excisional biopsy. The present study was composed of two parts. First we evaluated three readily available radiotracer solutions in a labora''tory animal lung to determine which had the most precise and sustained localization area over time. Next, using the radiotracer that tested best in the laboratory study, we performed CT-guided radiotracer injection for nodule localization with subsequent thoracoscopic excision of small or ill-defined lesions in patients felt to be at risk for a primary lung cancer. The results from the first 13 patients treated at our institution are presented and discussed.
| Material and methods |
|---|
|
|
|---|
Sequential coregistered anterior-posterior radiographs and scintigrams of the chest and abdomen were obtained at 1 minute, 30 minutes, and then 1, 2, 3, and 4 hours postinjection (Fig 1). Each scintigram was analyzed as follows. A region of interest (ROI) was drawn containing the lungs. To measure the area of the injection site, pixels within the lung having counts more than 10% of peak count in the image were identified, and these pixels were considered to be the injection site. The injection site area was determined by summing the area of these pixels. Decay corrected activity within the injection site was measured at the 30 minutes, and then 1, 2, 3, and 4 hours postinjection time points. This study was approved by our institutional animal care review board, and all animals received humane care in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (National Institutes of Health publication 85 to 23, revised 1985).
|
|
|
All patients subsequently underwent general anesthesia with single lung ventilation of the contralateral lung. Patients were placed in the lateral decubitus position. Three small thoracoscopic incisions were made: one for a 10-mm diameter 30-degree telescope, one for an endoscopic grasper, and one for dual use to insert a sterile gamma probe and an endoscopic stapler for subsequent lung resection.
All procedures involved the following sequence. After endoscopic visualization of the lung surface, a gamma radioprobe (C-Trak; Care Wise Medical Products Corp., Morgan Hill, CA) was inserted and radioactivity in counts per second (CPS) was determined over the general area where the lesion was believed to be. This was based on the CT scan and the information supplied by the site of the needle insertion in the chest wall. The lung surface exhibiting maximum radioactivity was then grasped by an endoscopic forceps. The probe then more precisely queried the area under the grasper to determine the maximum signal source using different probe angles. Early in the study a straight (0 degree) endoscopic gamma probe was used. The collimator aperture had a diameter of 10 mm and the overall probe length was 47 cm. Subsequently, a 37-cm long thoracic probe with a 30-degree bend in it was used. For this probe, the collimator aperture size was 15 mm. Following localization, the probe was removed and replaced by an articulating endostapler for excisional biopsy of the area of maximum radioactive signal.
Frozen section pathologic examination of all specimens was performed intraoperatively to make a diagnosis and confirm the removal of a lesion conforming to the CT abnormality. All patients found to have primary lung cancers underwent immediate lobectomies and nodal staging by muscle-sparing vertical axillary incisions.
| Results |
|---|
|
|
|---|
Part 2: radiotracer lung nodule localization in human patients
All 13 patients undergoing preoperative radiotracer marking of their solitary lung nodule or ill-defined mass had their lesions successfully localized (Table 1). In several patients the radiotracer allowed rapid identification of the approximate nodule location. This was followed by instrumental or digital palpation to confirm the exact localization of the nodule before excisional biopsy. All resection margins were microscopically clear. In 1 patient (patient 7), a dense pleurodesis required abandonment of the thoracoscopic approach. An open thoracotomy was done and the small nonpalpable nodule was localized by the radioprobe and successfully excised. In another patient (patient 12), the lesion was located 3.5-cm deep in the lobe and adjacent to major bronchovascular hilar structures and was successfully localized with the radioprobe. Thoracoscopic excision was attempted but abandoned due to the inability of the endoscopic stapling device to perform deep biopsies. Conversion to an open thoracotomy led to successful excisional biopsy. All other nodules in patients were localized using the radioprobe and successful thoracoscopic excisional biopsy was performed.
The range of time from the radiotracer injection to the initial surgical skin incision was 76 minutes to 8.1 hours with a mean time of 186 minutes. Use of the Tc 99m MAA radiotracer in humans confirmed the animal laboratory finding of maintenance of a small localized injection site size for up to 8 hours. Because surgical case start times are often unpredictable, the stability of the radiotracer in vivo after CT-guided placement greatly enhanced its adaptation to the operating suite realities.
Lesion characteristics in the study were as follows. The average distance from the chest wall pleural surface to the outer edge of the lesion on CT was 12.7 mm with a range of 1 to 47 mm. CT morphologic characteristics were 7 solid lesions, 4 ground-glass, and 2 mixed solid and ground-glass. Lesion size on pathologic description ranged from 4 to 23 mm with a median size of 10 mm. Seven of 13 lesions were malignant. Five of these were primary lung cancers, all of which were stage IA adenocarcinomas or bronchioloalveolar cell carcinomas (BAC). Two were solitary metastatic nodules (melanoma and renal cell) in patients who had a history of previous extrathoracic malignancies and who also had significant smoking histories. The average age of all patients was 59 years old with a range of 48 to 77 years old. The average age of the patients found to have primary lung cancers was 64.8 years old with a range of 52 to 77 years old.
There were no complications related to the radiotracer localization procedure, and there were no surgical complications. The average length of stay (LOS) for patients undergoing wedge excision was 2 days (range 1 to 4 days). The average LOS for patients undergoing biopsy and subsequent lobectomy was 3.9 days (range 3 to 4 days).
| Comment |
|---|
|
|
|---|
Bronchoscopic placement of a barium marker has been described by Okumura [9] with subsequent intraoperative use of fluoroscopy to locate the radioopaque substance placed near the lung nodule. The bronchoscopic skill required to do such localization is not readily available at most institutions. In addition, the requirement for fluoroscopy in the operating room and the limited viewing angles available with a patient in the lateral decubitus position make this technique cumbersome.
Localization involving CT-fluoroscopic injection of Lipiodol into or near the nodule has been described by Nomori [12]. Successful localization and excision was achieved in all patients, but possible systemic embolization of the marking solution and intraoperative need for fluoroscopy limit this technique's usefulness.
Transthoracic placement of a cyanoacrylate material using CT guidance was reported by Yoshida [10]. A firm, readily visible mass was readily found at thoracoscopy. However, two serious limitations were noted: the hardness of the localization material made pathologic sectioning for examination of the specimen difficult; the concern about possible embolization of the material into the systemic circulation.
Computed tomographic guided placement of a radiotracer for lung nodule localization was first described by Chella [13]. All 39 patients underwent successful localization using a solution of Tc 99m-labeled human serum albumin microspheres. However, this radiotracer is not available in the United States. Burdine [14] subsequently reported a radiotracer localization technique using unfiltered Tc 99m sulfur colloid, which is the same substance used for breast sentinel node detection. Loss of signal was an apparent problem with this radiotracer in the lung environment because patients were "transported immediately to the operating room" with an average time between CT injection and operating room localization of 67 minutes. Burdine's technique thus requires immediate access to the surgical suite, which is a significant limitation in most institutions. Sugi [15] described the successful use of Tc 99m tin colloid and Tc 99m phytate to guide thoracoscopic biopsy of lesions averaging 13.8 mm in size. Neither the tin colloid nor phytate suspension are available in the U.S. However, they successfully localized small pulmonary nodules in 25 patients without complication from the radiology procedure or surgery. The injected dose ranged from 2 to 4 mCi, approximately 10 times the dose used in our study. Sugi [15] used special radiation precautions in the handling of the surgical specimens. The injected dose in our study, 0.3 mCi, is slightly less than typically used for sentinel lymph node mapping. No special radiation protection precautions in the operating room or in handling the surgical specimens are thought to be required for sentinel lymph node mapping and, thus, none were used in the current study of lung nodules.
The results of our animal study suggest that either Tc 99m-labeled MAA or SC could be used for radiotracer-guided lung nodule localization, and both demonstrated good retention in the rat lung for 4 hours following direct intraparenchymal injection. The Tc 99m-labeled MAA had a smaller injection site diameter, which may provide more precise localization and permit smaller wedge resections. The particle size of Tc 99m-labeled MAA is more than 10 times greater than Tc 99m-labeled SC. This large particle size may impede particle diffusion within the lung and may contribute to the smaller injection site diameter of Tc 99m-labeled MAA.
The results of our feasibility study in humans using a solution of Tc 99m-labeled MAA revealed the following advantages when compared to previous techniques. Because no needle is left in place, dislodgement and trans-fissural positioning cannot occur. The substance used is readily available. Because it is routinely injected intravascularly to perform lung perfusion scans the risk of radiotracer embolization is associated with neglible morbidity. The procedure for Tc 99m MAA placement requires skill with CT-guided injection that is readily available in most institutions. The injection requires much less of a radiologist's time than attempts to aspirate diagnostic material from these small lung nodules. The procedure does not require intraoperative fluoroscopy, but can be done with a gamma probe available in any operating unit that performs sentinel node surgery. The endoscopic straight probe or the thorascopic 30-degree probe provide additional flexibility for thoracoscopic procedures. Frozen and permanent pathologic examination of the specimen are not affected by the localization technique and there is no significant risk to operating room and pathology department personnel from the low-dose gamma radiation. Our laboratory studies suggested (and our subsequent clinical experience has confirmed) that Tc 99m-labeled MAA produces a small intraparenchymal injection site that maximizes accuracy of lung nodule localization while minimizing tissue removed for the biopsy. Our animal data and clinical experience demonstrate that Tc 99m MAA is quite stable in the lung for up to 8 hours.
Our study did indicate some limitations. The biggest limitation remains that of the inherent difficulty in coordinating an operation with any preoperative same-day procedure. It is conceivable that radiotracer placement the evening before surgery might demonstrate enough sustained activity the following day to permit radiotracer guided thoracoscopic biopsy. Any thoracoscopic procedure is handicapped by the intraoperative discovery of a dense pleurodesis. In the one patient in our study that required conversion to an open thoracotomy because of a pleurodesis, the presence of the radiotracer still helped in the subsequent intraoperative localization of the small, nonpalpable nodule. In one other patient in the present study that required conversion to an open thoracotomy for biopsy is instructive in that the lesion was well-localized by the radiotracer marker but it lay deep and adjacent to peribronchial structures where thoracoscopic instruments could not successfully be used to excise it.
In summary, preoperative radiotracer lung nodule localization is feasible and a promising technique for assisting excisional diagnostic biopsy of possible early lung cancer presenting as a small pulmonary nodule or an ill-defined lesion. If further clinical use of this technique confirms its reliability and low morbidity and mortality, then, as discussed by Ost and coworkers [17] in a recent clinical practices review, "the strategy of proceeding directly to video-assisted thoracoscopic surgery becomes more effective than other diagnostic approaches" for the management of solitary pulmonary nodules.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
20 mm) solitary pulmonary nodules. AJR Am J Roentgenol 2003;180:1665-1669.This article has been cited by other articles:
![]() |
E. L. Grogan, G. J. Stukenborg, A. S. Nagji, W. Simmons, B. D. Kozower, D. R. Jones, and T. M. Daniel Radiotracer-Guided Thoracoscopic Resection is a Cost-Effective Technique for the Evaluation of Subcentimeter Pulmonary Nodules Ann. Thorac. Surg., September 1, 2008; 86(3): 934 - 940. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. L. Grogan, D. R. Jones, B. D. Kozower, W. D. Simmons, and T. M. Daniel Identification of Small Lung Nodules: Technique of Radiotracer-Guided Thoracoscopic Biopsy Ann. Thorac. Surg., February 1, 2008; 85(2): S772 - S777. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. M. Stiles, T. A. Altes, D. R. Jones, K. R. Shen, G. Ailawadi, S. B. Gay, J. Olazagasti, P. K. Rehm, and T. M. Daniel Clinical experience with radiotracer-guided thoracoscopic biopsy of small, indeterminate lung nodules. Ann. Thorac. Surg., October 1, 2006; 82(4): 1191 - 1197. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Sortini, C. V. Feo, P. Carcoforo, K. Maravegias, E. Pozza, A. Liboni, and A. Sortini Present and Future Applications of Radio-Guided Technique Ann. Thorac. Surg., June 1, 2005; 79(6): 2197 - 2197. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |