Ann Thorac Surg 2004;77:1033-1037
© 2004 The Society of Thoracic Surgeons
Original article: general thoracic
Preoperative imaging of the lung sentinel lymphatic basin with computed tomographic lymphography: a preliminary study
Kazuhiro Ueda, MDa*,
Kazuyoshi Suga, MDb,
Yoshikazu Kaneda, MDa,
Tao-Sheng Li, MDa,
Katsuhiko Ueda, RTb,
Kimikazu Hamano, MDa
a First Department of Surgery, Yamaguchi University School of Medicine, Ube Yamaguchi, Japan
b Department of Radiology, Yamaguchi University School of Medicine, Ube Yamaguchi, Japan
Accepted for publication September 10, 2003.
* Address reprint requests to Dr Kazuhiro Ueda, First Department of Surgery, Yamaguchi University School of Medicine, 1-1-1 Minami-Kogushi, Ube Yamaguchi 755-8505, Japan
e-mail: kaueda{at}yamaguchi-u.ac.jp
 |
Abstract
|
|---|
BACKGROUND: Preoperative localization of the sentinel node basin would guide selective lymph node dissection. We tried to identify these nodal stations with indirect computed tomographic lymphography using a conventional extracellular contrast agent, iopamidol.
METHODS: Eleven consecutive patients scheduled to undergo anatomic resection of suspected lung cancer, without lymphadenopathy, were given a peritumoral injection of undiluted iopamidol under computed tomography guidance, and lymphatic migration was assessed by multidetector-row helical computed tomography.
RESULTS: There were no complications such as bleeding, pneumothorax, or allergic reactions. Enhanced nodes were detected in all but 1 patient who had diffuse lymph nodal calcification. Enhanced nodes were identified at 32 ipsilateral intrathoracic nodal stations (20 hilar stations and 12 mediastinal stations). The average length of the longer axis of the enhanced nodes was 4.8 mm (range, 3 to 8 mm), and the average attenuation of the enhanced nodes was 132 (range, 46 to 261) Hounsfield units. In 9 patients with confirmed lung cancer, enhanced nodes appeared at 26 nodal stations, and all apparent enhanced nodes were identified as actual lymph nodes at appropriate position during lymphadenectomy. None of the resected lymph nodes had metastatic involvement.
CONCLUSIONS: Indirect computed tomographic lymphography with the peritumoral injection of iopamidol effectively depicts the drainage nodes unless they are diffusely calcified. Although further study is required, this method could guide selective lymph node dissection.
|
Introduction
|
|---|
Sentinel lymph node assessment for patients with lung cancer is now in the validation phase [1–6]. Liptay and colleagues [7] reported that lymph node mapping using a radioactive tracer improves the detection of nodal micrometastasis. However, it remains unclear whether this procedure can guide so-called sentinel node biopsy because sentinel lymph nodes are hidden by the injected isotope (shine-through phenomenon) and by the airway isotope [6]. Consequently, sentinel lymph nodes are not able to be identified without lymph node dissection. This specific pitfall is attributable to the fact that the colloid particles instilled within the peripheral lung tissue may initially migrate into the tracheobronchus rather than the lymphatics as it is physiologically difficult for colloid particles to penetrate alveolar epithelial cells to enter the interstitium, slowing transport of the particles into the lymphatics [8]. In contrast, soluble materials injected into the alveolar lumen are rapidly transported into the lung interstitium by means of the alveolar epithelial cells and are cleared by means of both the circulation and the lymphatics [8]. Therefore, we hypothesized that computed tomographic (CT) scan could show the successive migration of injected soluble contrast material into the drainage lymph nodes without obstruction by the injected materials or their migration into the airway. Furthermore, accurate and reproducible definition of anatomic landmarks for every intrathoracic nodal station should allow for correspondence between the CT image and the intraoperative view during lymph node dissection [9]. Thus, we conducted a trial using indirect CT lymphography with iopamidol as extracellular contrast medium to identify lymphatic drainage nodes, which may guide selective lymph node dissection.
|
Patients and methods
|
|---|
Patients
This study was approved by the Yamaguchi University Hospital Institutional Review Board in November 2002. From November 2002 to April 2003, a total of 11 consecutive patients scheduled to undergo anatomic resection of clinically suspected N0 lung cancer, none of whom had received prior adjuvant therapy or mediastinoscopy, were enrolled in this study after providing informed consent. The study group comprised 4 men and 7 women with a mean age of 63 years (range, 48 to 84 years). Lesions were located in the right upper lobe (n = 2), right middle lobe (n = 1), right lower lobe (n = 3), left upper lobe (n = 3), or left lower lobe (n = 2; Table 1). The mean tumor diameter was 19 mm (range, 8 to 40 mm).
Indirect computed tomographic lymphography
The CT lymphography was performed using a multidetector-row helical CT scanner (Siemens Volume Zoom; Siemens-Asahi Medical, Tokyo, Japan). The injection site was determined according to the position of the tumor. After the local anesthesia was achieved with a total of 5 mL of 1% lidocaine hydrochloride, one injection of 1.5 to 2 mL of iopamidol (Iopamiron 300; Nippon Shering, Osaka, Japan), targeted to the peripheral outer border of the tumor, was given through a 23-gauge needle attached to a 2.5-mL syringe. This contrast agent, which is conventionally used as an extracellular contrast agent for angiography and enhancement CT, has an iodine concentration of 300 mg/mL, an osmolarity of 585 mOsm/kg, a viscosity of 4.4 mPa/s, and pH of 6.5 to 7.5. Contiguous 2-mm-thick CT images from the upper thorax to the lower mediastinum were obtained once before injection of the contrast agent, in addition to successively at 30 seconds and 1, 3, and 5 minutes after the injection. The CT scanner was operated at 140 kV and 100 mA, with a 45-cm field of view, a 512 x 512 matrix, a detector of 0.5 mm x 2 rows, section spacing of 5 mm, and table speed of 1.53 mm/0.5 s. The number of sections, adapted on a case-by-case basis to ensure coverage of the lung and mediastinum, ranged from 95 to 102; the acquisition time ranged from 25 to 30 seconds. During CT image acquisition, breath-hold was performed to a tidal inspiration level.
Image interpretation
The CT images were analyzed on a workstation connected to the CT unit. Enhanced lymph nodes (ELNs) in each patient were prospectively reported by two radiologists who were blinded to the experimental protocol. Quantitative analysis included measurement of each lymph node depicted in the field of view and assessment of the time course of the enhancement effect in hilar lymph nodes, mediastinal lymph nodes, and chest wall muscles. Lymph node size was determined by measurement of the longer and shorter diameters of each node on the consecutive postcontrast CT images. Because small contrasted structures could be lymphatic vessels, only structures with a shorter diameter of more than 3 mm were considered lymph nodes. To assess the enhancement effects, operator-defined regions of interest were drawn on the consecutive CT images for all lymph nodes. The size of each region of interest was adapted to encompass as much of the lymph node as possible with the maximum enhancement. Each lymph node imaged was measured, and measurements were obtained at the same location for the series of images. The postcontrast images facilitated identification of the nonenhanced lymph nodes on the precontrast images, and the same region of interest size was used on precontrast and postcontrast images of structures. The areas of the regions of interest ranged from 6.1 to 26.1 mm2 for lymph nodes and 8.6 to 109 mm2 for muscles. A lymph node was confirmed to be an ELN if the attenuation was more than 30 Hounsfield units (HU) greater on postcontrast images than on precontrast images.
Correspondence between computed tomographic image and intraoperative view
The location of every ELN on the CT image was classified into regional lymph node stations according to the definitions of the American Joint Committee on Cancer and the Union Internationale Contre le Cancer [9]. Anatomic resection of the tumor from all patients with pathologically diagnosed lung cancer included standard lymph node dissection. To confirm that each ELN was indeed a lymph node, every one was intraoperatively examined to see that it corresponded to the actual lymph node of compatible size and position on the CT image.
Pathologic nodal assessment and data analysis
Formalin-fixed and paraffin-embedded sections of the dissected lymph nodes were examined histologically by hematoxylin and eosin staining.
Values are expressed as means with standard deviations. The significance of the differences in the data comparisons was assessed by paired or unpaired Student's t test. A p value of less than 0.05 was considered significant.
|
Results
|
|---|
Pathologic study showed 9 of the 11 tumors to be non–small-cell lung cancers (7 adenocarcinomas, 2 squamous cell carcinomas). Two of the tumors were benign. The patients with lung cancer underwent lobectomy (n = 5) or segmentectomy (n = 4) with standard lymph node dissection. The 2 patients with benign disease underwent pulmonary wedge resection without lymphadenectomy. Postoperative pathologic examinations confirmed that all patients were free of visceral pleural invasion. The pathologic disease stages for patients with lung cancer were stage IA (n = 7), and stage IB (n = 2).
There was no complication, such as bleeding, pneumothorax, or allergic reaction, related to the injection of iopamidol. All patients received accurately delivered injections without alveolar structural changes (Fig 1).
View larger version (63K):
[in this window]
[in a new window]
|
Fig 1. Chest computed tomographic scan of the primary lung tumor (moderately differentiated adenocarcinoma) before (Pre, left) and after injection of iopamidol. The iopamidol is injected to the border of the lung tumor (1 min, center) and completely cleared at 20 hours (20 hr, right) after injection without any structural change of the injection site except for the formation of minimal free air space.
|
|
Enhanced lymph nodes were identified in all but 1 patient who showed diffuse lymph node calcification (Fig 2). Enhanced lymph nodes were localized to a total of 32 ipsilateral intrathoracic nodal stations: 20 hilar stations and 12 mediastinal stations (Table 1). In the 9 patients with lung cancer, ELNs appeared at 26 nodal stations, and all ELNs were identified as actual lymph nodes during surgery. Average length of the longer axis of all ELNs was 4.8 mm (range, 3 to 8 mm). Throughout the 5 minutes of evaluation, the attenuation value of both the hilar and mediastinal ELNs was significantly greater than that of either the precontrast lymph nodes (paired Student's t test, p < 0.01) or chest wall muscle (unpaired Student's t test, p < 0.02; Fig 3). Maximal attenuation of ELNs was obtained at 1 minute after injection; average attenuation of the pulmonary ELNs was 146 ± 49 HU, and average attenuation of the mediastinal ELNs was 119 ± 51 HU (Fig 3).
View larger version (67K):
[in this window]
[in a new window]
|
Fig 2. Intrathoracic lymph nodes on precontrast (Pre, top) and postcontrast (Post, bottom) chest computed tomographic scans. Long arrows indicate the enhanced intrathoracic lymph nodes, and short arrows indicate the site of iopamidol injection.
|
|
Serial pathologic examination of the resected lymph nodes showed that no ELN stations or non-ELN stations had metastatic involvement.
|
Comment
|
|---|
The assessment of intrathoracic lymph nodes by indirect lymphography is limited because insoluble colloid particles instilled into the alveolar space cannot penetrate into the interstitium easily [8, 10, 11]. Ketai and associates [12] instilled iodinated nanoparticles into the alveolar lumen to perform indirect CT lymphography in dogs. They found no significant accumulation of contrast material in the tracheobronchial lymph nodes even 2 days after instillation, and noted maximal enhancement occurred up to 9 days after instillation because the nanoparticles were slowly transported to the interstitium mainly by macrophages. We believe that these properties of colloid particles in the alveolar space make them impractical for use in the preoperative assessment of lung lymphatics. Thus, we conducted this preliminary study on indirect lymphography with a local injection of iopamidol and found, as expected, that iopamidol instilled into the alveolar lumen is rapidly absorbed and cleared from the site of injection, like other extracellular, water-soluble, and low-molecular solutes. Although some of the injected iopamidol may be cleared through the blood vessels, there was no noticeable enhancement in the blood vessels and muscles in this study, which may be explained by the dilution and rapid elimination of the small volume of contrast agent in the pulmonary blood vessels with much faster flow (approximately 5 to 6 L/min) and greater volume than the lymphatics [13]. The systemic recirculation of the contrast agent is also considerably reduced by rapid elimination through the kidneys.
Currently, we also reported a study on indirect breast lymphography with a local injection of iopamidol, which clearly showed the breast lymphatic pathways and their connections to the sentinel lymph nodes [14]. In the present study on indirect lung lymphography, we could not visualize such lymphatic pathways clearly because pulmonary lymphatic drainage from the lung parenchyma is complex and variable. Previous investigators reported that pulmonary lymphatics drain the connective tissue spaces of the peripheral lung down to the level of the primary lobules, including the interlobular, subpleural, peribronchial, and perivascular spaces, and that they run proximally from the level of the alveolar ducts and subsequently drain into nodes that lie along the course of the bronchi and at the hilum of the lung [15]. The migration of the imaging agent from the injection sites toward the hilum appears to reflect this physiologic lymphatic flow. Therefore, plural sentinel node stations, which directly drain the lymphatics from a solitary lung tumor through these variable lymphatic routes, can be seen, as supported by the recent radiocolloids lymph node mapping study [4, 6]. However, even in the radiocolloids study, the exact criteria for a sentinel lymph node have not been defined because they may depend on the imaging agent and the timing of sentinel node detection [16]. The sentinel lymph node is usually regarded as the first lymph node that the imaging agent reaches from the injected site. Therefore, with indirect lymphography using extracellular contrast media, the first lymph node should have an earlier enhancement peak than more distant nodes. Our results indicated that there could be a time point, approximately 1 minute after the injection, when maximal iopamidol entered the first lymph node. Further analysis of the time course of enhancement of each lymph node on dynamic CT images with adequately fractioned intervals could help to identify the first lymph node station.
After the injected iopamidol reaches the lung interstitium, then the lymphatics, it distributes rapidly, not only to the sentinel node but also to distant nodes, without becoming entrapped in the sentinel node. However, in this study, the iopamidol was concentrated only in a limited area of ELNs; on average, 2.0 hilar and 1.2 mediastinal lymph node stations per patient. This could be because the ELNs are supplied by lymphatics predominantly from the injected site, whereas the distant, nonenhanced nodes are supplied not only by the lymphatics passing through the ELNs, but also by lymphatics from noninjected areas. Therefore, ELNs could be considered as the specific lymph node stations that are directly or closely linked to the lymphatics from the primary injected site (sentinel lymphatic basin).
Lung lymphography could be extremely useful for the preoperative identification of specific mediastinal lymph node stations that would be the first N2 stations of metastatic involvement. This identification could improve the accuracy of staging mediastinal nodal status under mediastinoscopy or mediastinal lymph node sampling with minimal lymphadenectomy. Assuming that the mediastinal ELN can be identified even in the contralateral mediastinum, pathologic examination of this lymph node under mediastinoscopy could effectively determine not only accurate mediastinal nodal status, but also the best therapeutic strategy. However, further investigation on larger series, including node-positive disease, is required to confirm whether the imaged nodes can accurately predict nodal histology in lung cancer, which could validate the above speculations.
Even without any labeling of the ELN using radiocolloid or vital dye, they were easily found in the appropriate position during surgery, using the surrounding anatomic structures such as the bronchi or vessels as a guide. This result was in accordance with that reported by Suga and coworkers [17], who visualized pulmonary lymphatic drainage with magnetic resonance lymphography in dogs. They accurately detected the imaged lymph node in exactly the same site as seen on the image, and the size of the resected node corresponded well with that of the image. These investigators therefore suggested that this imaging method could be a viable approach to guide lymph node sampling or selective lymph node dissection, which may reduce the extent of lymph node dissection in lung cancer patients.
Our method does not appear to carry a risk of complications, although the possible side effects of the intrapulmonary injection of iopamidol were considered; for example, pneumothorax, intrathoracic or intrapulmonary hemorrhage, allergic reaction induced by iodine, and intravascular air embolism. Furthermore, this preoperative method did not prolong surgery or require any special instrument such as a 
-detecting probe because it does not involve any intraoperative procedures. The mean time required to complete this preoperative lymphography procedure was 28 minutes (range, 25 to 35 minutes). Thus, we think that it is feasible and cost-effective.
In conclusion, indirect CT lymphography with peritumoral injection of iopamidol successfully depicts the drainage nodes unless they are diffusely calcified. Although further study is required for validation, this method could be a useful approach to guide selective lymph node dissection.
|
References
|
|---|
- Liptay M.J., Masters G.A., Winchester D.J., et al. Intraoperative radioisotope sentinel lymph node mapping in non-small cell lung cancer. Ann Thorac Surg 2000;70:384-389.[Abstract/Free Full Text]
- Melfi F.M., Chella A., Menconi G.F., et al. Intraoperative radioguided sentinel lymph node biopsy in non-small cell lung cancer. Eur J Cardiothorac Surg 2003;23:214-220.[Abstract/Free Full Text]
- Schmidt F.E., Woltering E.A., Webb W.R., Garcia O.M., Cohen J.E., Rozans M.H. Sentinel nodal assessment in patients with carcinoma of the lung. Ann Thorac Surg 2002;74:870-874.[Abstract/Free Full Text]
- Nomori H., Horio H., Naruke T., Orikasa H., Yamazaki K., Suemasu K. Use of technetium-99m tin colloid for sentinel lymph node identification in non-small cell lung cancer. J Thorac Cardiovasc Surg 2002;124:486-492.[Abstract/Free Full Text]
- Sugi K., Fukuda M., Nakamura H., Kaneda Y. Comparison of three tracers for detecting sentinel lymph nodes in patients with clinical N0 lung cancer. Lung Cancer 2003;39:37-40.[Medline]
- Ueda K, Suga K, Kaneda Y, et al. Radioisotope lymph node mapping in non-small cell lung cancer: can it be applicable for sentinel node biopsy? Ann Thorac Surg. In press
- Liptay M.J., Grondin S.C., Fry W.A., et al. Intraoperative sentinel lymph node mapping in non-small-cell lung cancer improves detection of micrometastases. J Clin Oncol 2002;20:1984-1988.[Abstract/Free Full Text]
- Lippmann M., Yeates D.B., Albert R.E. Deposition, retention, and clearance of inhaled particles. Br J Ind Med 1980;37:337-362.[Medline]
- Mountain C.F., Dresler C.M. Regional lymph node classification for lung cancer staging. Chest 1997;111:1718-1723.[Abstract/Free Full Text]
- Matthay M.A., Landolt C.C., Staub N.C. Differential liquid and protein clearance from the alveoli of anesthetized sheep. J Appl Physiol 1982;53:96-104.[Abstract/Free Full Text]
- Havill A.M., Gee M.H. Role of interstitium in clearance of alveolar fluid in normal and injured lungs. J Appl Physiol 1984;57:1-6.[Abstract/Free Full Text]
- Ketai L.H., Muggenberg B.A., McIntire G.L., et al. CT imaging of intrathoracic lymph nodes in dogs with bronchoscopically administered iodinated nanoparticles. Acad Radiol 1999;6:49-54.[Medline]
- West J.B. Structure and function—how the architecture of the lung subserves its function. In: West J.B., ed. Respiratory physiology—the essentials, 5th ed Baltimore: Williams & Wilkins, 1995:1-10.
- Suga K., Ogasawara N., Okada M., Matsunaga N. Interstitial CT lymphography-guided localization of breast sentinel lymph node. Preliminary results. Surgery 2003;133:170-179.[Medline]
- Lauweryns J.M., Baert J.H. Alveolar clearance and the role of the pulmonary lymphatics. Am Rev Respir Dis 1977;115:625-683.[Medline]
- Boxen I., McCready D., Ballinger J.R. Sentinel node detection and definition may depend on the imaging agent and timing. Clin Nucl Med 1999;24:390-394.[Medline]
- Suga K, Yuan Y, Ogasawara N, Okada M, Kawakami Y, Matsunaga N. Potential of magnetic resonance lymphography with intrapulmonary injection of gadopentetate dimeglumine for visualization of the pulmonary lymphatic basin in dogs: preliminary results. Invest Radiol 2003;38:679–89
This article has been cited by other articles:
|
|
|
|
 
K. Ueda, Y. Kaneda, M. Sudo, M. Jinbo, K. Suga, and K. Hamono
Cerebral Air Embolism During Imaging of a Sentinel Lymphatic Drainage in the Respiratory Tract
Ann. Thorac. Surg.,
February 1, 2006;
81(2):
721 - 723.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. Suga, K. Shimizu, Y. Kawakami, A. Tangoku, M. Zaki, N. Matsunaga, and M. Oka
Lymphatic Drainage from Esophagogastric Tract: Feasibility of Endoscopic CT Lymphography for Direct Visualization of Pathways
Radiology,
December 1, 2005;
237(3):
952 - 960.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
), enhanced mediastinal lymph nodes (
), and muscles of the chest wall (
). Bar indicates a standard deviation of the mean. (CT = computed tomographic; HU = Hounsfield units; Pre = before injection.)