| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thoracic Imaging |
1 From the Departments of Radiology (S.D., S.K., K.Y., H.M., S.A., M.K, Y.H.) and Surgery II (A.A., N.S.), Okayama University Medical School, 2-5-1 Shikatacho, Okayama 700-8558, Japan; and Department of Radiology, Fukuyama National Hospital, Japan (T.H., Y.K.). Received June 11, 2001; revision requested August 3; final revision received February 18, 2002; accepted March 20. Address correspondence to S.K. (e-mail: susumu@cc.okayama-u.ac.jp).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Percutaneous localization of 168 lesions was performed with computed tomographic (CT) guidance in 150 patients. Patients were classified into three groups: a 3-year early-learning experience of treatment of 40 lesions mainly in one institution (group A1), a more recent 4-year experience of treatment of 88 lesions in the same institution (group A2), and the roughly synchronous recent 3-year experience of treatment of 40 lesions in a different hospital (group B).
RESULTS: The hook wire was successfully placed without dislodgment in 146 patients, accounting for 164 (97.6%) of 168 lesions. Group A2 showed a success rate of 100%. There was no difference in patients among the three groups in regard to size of lesions or their distance from the pleural surface. In patients in groups A2 and B, the proportion of nodules with ground-glass opacity and primary lung carcinoma at CT was significantly greater than that in patients in group A1. In 168 placements, nonsymptomatic pneumothorax cases were observed in 54 (32.1%), hemorrhages into the lung were observed in 25 (14.9%), and hemorrhage into the pleural space was observed in one (0.6%). No patient complained of notable pain during or after the procedure, and no serious complication was experienced. Unsuccessful placement was caused by too shallow a puncture with the introducer needle.
CONCLUSION: This system with a flexible suture for preoperative localization has a high success rate.
© RSNA, 2002
Index terms: Lung, ground-glass opacification, 60.322 • Lung, nodule, 60.28 • Lung neoplasms, CT, 60.1211, 60.32 • Lung neoplasms, surgery, 60.45 • Video systems
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
When the lesion appears as a nodule at CT, the surgeon’s requirement for localization usually is that the diameter is less than 10 mm and/or the distance from the pleural surface is more than 10 mm. Surgeons request localizations for lesions more than 10 mm in diameter in cases of nodules with ground-glass opacity that are separated from the pleural surface. In this study, the patients were 71 men and 79 women, and the age range was 25–87 years, with a mean age of 61.8 years. Sixty patients had known malignancies as follows: 25 lung cancers, eight colon cancers, six breast cancers, three renal cancers, two thyroid cancers, two esophageal cancers, two gastric cancers, two hepatocellular carcinomas, one cholangiocellular carcinoma, one choriocarcinoma, one gingival cancer, one prostatic cancer, one laryngeal cancer, one mucoepidermoid cancer, one malignant fibrous histiocytoma, one ovarian leiomyosarcoma, one meningioma, and one phyllode tumor of the breast.
Procedure
The short hook wire and suture system (Guiding-Marker System; Hakko, Tokyo, Japan) includes a stainless steel hook wire and its introducer system (4) (Fig 1). The hook wire is 10 mm long, and it has a 30-cm-long nylon suture firmly attached to the funnel at the proximal end of the wire. The two-layer introducer system includes a 21-gauge 10-cm-long cannula with an inner 24-gauge 10-cm-long hollow pusher.
|
|
|
|
|
Each patient was placed on the CT table in such a position as to allow the shortest possible direct access route for hook wire placement. After administration of a local anesthetic, the cannula tip was advanced with CT guidance to within 0–2 cm of the lesion so as not to injure the pathologic specimen of the lesion (Fig 2). Two early failures with hook wire placement led us to establish the following ground rules: We inserted the tip of the cannula to at least 1 cm beneath the pleural surface. Only after optimal placement of the cannula tip was confirmed with CT scanning was the pusher fully advanced to eject the hook wire. Once the hook wire was just free from the cannula, the introducer system was carefully withdrawn. A CT scan was again obtained to confirm that the hook wire was placed in the planned position (Fig 2). The portion of the suture outside the patient was loosely coiled and kept covered with sterile gauze.
|
|
|
|
Evaluations
The 168 procedures performed with this hook wire placement were classified into three groups, to allow comparison of the results over different periods at the same institution and at different institutions at similar overlapping periods. Group A1 included patients at institution A in whom procedures were performed in 40 lesions between September 1993 and October 1996, and these were reported as early experiences with this system (5). Group A2 included patients at institution A in whom procedures were performed in 88 lesions between November 1996 and September 2000. Group B included patients at institution B in whom procedures were performed in 40 lesions between February 1998, when this system was first used, and August 2000.
In all patients, we evaluated the following: the successful or unsuccessful placement and retrieval of the hook wire, the distance between hook wire and lesion, the number of lesions for which placement was suggested, the position of the patient on the CT table, the time for the procedure, the time patients waited before the thoracoscopy was performed, the visibility or palpability of the lesions at removal, the performance of lobectomy or nodal resection with VATS, the lesion size and location, the distance from the pleural surface, the lesion characteristics at CT, the number of primary lung cancers or other lesions, the complications, and the presence of an emphysematous lung.
We evaluated these items by review of patients’ records and images. Three of the authors (S.D., T.H., S.K.) undertook this review. We compared the results among these three groups. Basically, the results in patients in group A1 were compared with those in patients in group A2 for different time comparisons—early and recent—at the same institution, and the results in patients in group A2 were compared with those in patients in group B for synchronous comparison at different institutions. However, the results in patients in groups A1 and B were also compared to determine possible significant differences between them. Furthermore, in patients with unsuccessful placement, we summarized the results to clarify the reasons for such unsuccessful placement.
Statistical Analysis
All calculations and statistical tests were performed by using statistical software (StatView; SAS Institute, Cary, NC) that was based on a standard personal computer system (Macintosh; Apple Computer, Cupertino, Calif). For measurement of data regarding incidence or proportion, the 
2 test was performed, and if necessary, the Fisher exact probability test was performed; the results were adjusted by using Bonferroni statistics. For continuous variables, the multiple comparison procedure of Dunnet was applied. P values less than .05 were considered to indicate a significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In all patients in whom placement of the system was successful, the suture served as a clear guide at thoracoscopy. All hook wires were retrieved with resected specimens and disposed of. The distance between the hook wire and the lesion in 164 lesions was 0–20 mm, and the mean distance was 4.8 mm. In patients in group A1, the distance was 0–20 mm, and the mean distance was 5.4 mm ± 4.2 (SD); in patients in group A2, the distance was 1–18 mm, and the mean distance was 5.4 mm ± 4.4; in patients in group B, the distance was 1–14 mm, and the mean distance was 4.1 mm ± 3.3. The data did not show a significant difference among the three groups (groups A1 and A2, P > .05; groups A2 and B, P > .05). These observations were confirmed with a CT scan, and they were again confirmed at observation of the resected specimens. In group A2, in two patients with three pulmonary lesions each, simultaneous placement was successfully performed for all three lesions in each patient.
Double hook wire placement in 14 patients (group A1, one patient; group A2, 12 patients; group B, one patient) also was successful (Fig 2). Of these, 13 patients had two lesions, but one patient had three, two of which were located closely enough for one hook wire localization. One hundred thirty of 150 patients underwent successful single placement. In three patients in group A2, hook wire placements for bilateral pulmonary lesions were successfully performed. In one patient, the position on the CT table was changed twice, and hook wire placement was performed three times without dislodgment. In four patients with two lesions, the position on the CT table was changed once, and there was no dislocation of hook wire. Table 1 includes the position of the patients on the CT table. In two patients in groups A1 and B in whom a transfissural approach was used, the suture at the interlobar fissure was used as a guide during thoracoscopic resection.
|
In 45 patients with no functional contraindications (pulmonary functions such that the patient would be unable to tolerate lobectomy according to the surgeon’s criteria), further lobectomy and lymph node resection were performed with VATS immediately after intraoperative frozen pathologic examination, and results showed that their lesions were primary lung carcinoma.
Lesion Characteristics
In patients in group A1, the diameter of the 40 lesions was 4–23 mm, and the mean diameter was 9.4 mm ± 6.8; in those in group A2, the diameter of the 88 lesions was 5–29 mm, and the mean diameter was 11.0 mm ± 4.8; and in those in group B, the diameter of the 40 lesions was 2–30 mm, and the mean diameter was 10.6 mm ± 6.5. There was no statistically significant difference (groups A1 and A2, P > .05; groups A2 and B, P > .05), nor was there such a difference in the location of the lesions in the three groups (Table 2). The distance from the nearest pleural surface of each lesion was 0–30 mm, with a mean of 10.3 mm ± 6.0, in group A1; 0–38 mm, with a mean of 8.5 mm ± 7.5, in group A2; and 1–38 mm, with a mean of 7.4 mm ± 7.8, in group B. There was no statistically significant difference (groups A1 and A2, P > .05; groups A2 and B, P > .05), and the overall distance from the pleural surface was 0–38 mm, with a mean of 8.2 mm.
|
|
|
|
Unsuccessful Placement
As described before, the hook wire was dislodged at thoracoscopy in four lesions in four patients. In patients in group A1, three lesions were two metastatic nodules and one carcinoma nodule, and in those in group B, the other lesion was a carcinoma nodule with ground-glass opacity. In group A1, dislodgment occurred in two patients in the earliest 10 attempts, at which time CT confirmation of appropriate positioning of the hook wire in the lung after insertion had not yet been standardized.
In addition to this lack of standardized CT confirmation, the introducer needle had not been inserted to a minimum depth of 1 cm from the pleural surface; it had been inserted to only 5 mm. After these failures, we inserted the tip of the cannula to at least 1 cm beneath the pleural surface. By review of CT scans obtained in the latter two patients in whom dislodgment occurred, we confirmed that insertion of the introducer needle was too shallow. Shallowness of insertion was the only conceivable cause of the hook wire dislodgment that was observed at thoracoscopy. The distance between the inserted cannula tip and the pleural surface was only 6 mm in each of the latter two placements.
Procedural results and lesion characteristics in these unsuccessful placements were not different from those in successful ones (Table 6). Thus, unsuccessful placement was thought to be caused by a puncture with the introducer needle that was too shallow. No definite complications, except for one pneumothorax, were observed during the hook wire placement procedures in four patients. These four patients underwent thoracoscopy within 1 hour after hook wire placement. Pulmonary lesions were successfully resected by performing VATS after detection of localized hemorrhage on the lung surface. This hemorrhage was caused by puncture. None of the lesions themselves were visible or palpable during resection. The dislodged hook wires were thoracoscopically retrieved in all of the previously mentioned patients.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
We now believe that the necessity for preoperative localization will probably not decrease. As was shown in patients in our groups A2 and B, VATS is being performed more often in lesions with ground-glass opacity, as shown on thin-section CT scans, that cannot be palpated easily during thoracoscopy. These lesions are often atypical adenomatous hyperplasia or primary adenocarcinoma, including bronchioloalveolar carcinoma (8,9). Whenever they are proved to be primary lung malignancy by means of intraoperative frozen pathologic examination, lobectomy with lymph node dissection also can be performed with VATS (10). The wider the application of VATS for pulmonary lesions becomes, the more preoperative localization will be needed.
We can classify the localization of lesions into three types. The first type is localization with imaging modalities during thoracoscopy. This includes intraoperative ultrasonography (US) (11) and CT fluoroscopy (12). The second type is preoperative localization with injection of dyes (13), contrast media (14), radionuclides (15), or colored adhesive agents (16,17). The third type is preoperative localization with hook wire placement (7,18–22). The latter two types of preoperative localization usually are performed with CT guidance. To establish the ideal form of localization, clarification is needed as to the advantages and disadvantages of these reported methods.
Although localization with intraoperative US is noninvasive, it has several limitations (11,23). Visualizing pulmonary lesions at US requires complete collapse of the lung, which is often impossible in patients with obstructive disease such as centrilobular emphysema. Formless abnormalities may be particularly difficult to visualize. Even when preoperative localization was used, a success rate of only 76% was reported (23), and the time required for preparation and in the surgical suite was 30–150 minutes (11). Our success rate and the time required for localization are superior to these data. Recent experience with thoracoscopic resection by using CT fluoroscopy is limited (12), and this procedure cannot be performed in all hospitals.
Dye infusion with CT guidance is a feasible technique for localization. We note, however, three problems regarding this approach: possible risk of shock after injection of dye, difficulty in seeing during thoracoscopic surgery in patients with extensive anthracotic pigmentation, and diffusion of dye across the lung surface (13,20). Contrast material localization with CT guidance is a simple and more accurate technique than dye infusion (14); 2–3 mL of nonsoluble (barium or iodized oil) or water-soluble contrast material is injected within or near the nodule. Iodized oil is preferred because of problematic early diffusion and absorption with water-soluble contrast material and difficulty in preparing thin aseptic barium (14).
Though the general results of iodized oil injection seem good, we observed some drawbacks, including the necessity for fluoroscopy during surgery and the possibility of destruction of the specimen of a lesion with ground-glass opacity with direct injection. Although contrast material cannot act as a retraction device during thoracoscopic surgery, a wire itself or, as in our system, suture to wire can (22). It is also not exactly clear how long iodized oil remains within or near the lesion. With our system, hook wires in 38 patients who waited 4–14 hours for the start of thoracoscopic surgery did not dislodge. Localization with injection of radionuclides (15) and colored adhesive agents (16,17) has been limited and seems to have most of the same problems as that with contrast material injection, that is, the need for gamma probe during thoracoscopic surgery, no means of retraction, and unknown retention time.
Localization with hook wire placement is probably the oldest method (18). The conventional mammographic needle and wire system is most commonly used. Unfortunately, this popular method has problems, which include wire dislodgment and pleural pain during and after insertion (19–21). Thaete et al (6) revealed that the wire became dislodged during the interval between CT placement and thoracoscopic insertion in 22 of 100 patients in whom wire placement with CT guidance was successful. These failures included that in a patient undergoing simultaneous triple lesion localization, whereas we always succeeded in such cases. The success rate (78%) in their series was lower than that (97.6%) in ours. They also noted a high incidence of wire dislodgment with pneumothorax in their series. They presumed that a mechanism was present whereby the coefficient of friction of the wire being pulled through the chest wall is greater than that of the hook wire inserted into the lung parenchyma. As the pneumothorax expands, the barb of the hook wire is pulled out of the lung. With our device, the friction was less than that of a conventional hook wire because we used a thin flexible nylon suture in place of a rigid wire.
According to Mullan et al (22), a wire is generally dislodged at one of three times: during transportation of the patient to the surgical suite; during surgical deflation of the lung; or during resection, when the surgeon will often apply gentle retraction on the wire. Our flexible suture, which protrudes from the skin and is covered with gauze, evinces a reliable tendency not to pull or push the hook wire embedded in the lung. Patients were moved to the surgical suite with no special handling, since the exposed suture was entirely covered with gauze and was flat enough to enable the patients to roll around on the stretcher. Moreover, simultaneous triple-wire placement was successfully performed in two patients and double-wire placement was performed in 14. Although five patients needed to change their position once or twice on the CT table during the placement procedure, no dislodgment was experienced.
This flexibility of the suture, which, to our knowledge, has not been demonstrated in any other reported device, may be one of the biggest advantages of our system, in our opinion. Mullan et al (22) reported three kinds of custom-made wire in which a cloverleaf design showed good anchoring power in the lung parenchyma. This elaborate wire gave surgeons "moderately strong" retraction without dislodgment. As they described it, the anchoring power of the hook was caused by the cross-sectional area of the hook in the plane of tension, and their device may be superior to ours in this respect alone. However, when we reviewed our 168 procedures with only four cases of dislodgment (mainly in the early days), their limited five successful experiences failed to persuade us of any general advantage. Cost problems and applicability to simultaneous double- or triple-wire placement so far remain unaddressed with regard to their device.
We emphasize that our system can be considered standardized, because in patients in all three groups there was a high success rate. In addition, understanding of the necessity for CT confirmation before and after deployment of the hook wire and insertion of the tip of the cannula more than 1 cm from the pleural surface led to the 100% rate of success in patients in group A2. We experienced no serious complications. The incidence of pneumothorax or hemorrhage into the lung is not any higher than that in other reports of localization or CT-guided needle biopsy (6,24). In our study, the reason for the higher incidence of pneumothorax in patients in groups A2 and B than in those in group A1 may be attributable to an increase in the risk caused by the treatment of simultaneous multiple lesions in the former groups. Likewise, the reason for higher incidence of pulmonary hemorrhage in patients in group A2 than in those in group A1 may be the same as that for the higher incidence of pneumothorax.
A limitation of our device in weaker lung tissue is that it may provide less anchoring power, as Mullan et al (22) pointed out. Surgeons can retract "gently" but not "strongly" during surgery, but it should be emphasized that this is a problem with the tissue, as the rigorously tested suture will not pull away from the hook wire in normal use. Though resection with our system resulted in a high success rate, surgeons will be more confident with greater anchoring power.
In conclusion, after review of 168 procedures, the usefulness of the short hook wire and suture system was affirmed in terms of its high success rate, applicability to multiple lesions, ability to allow long waiting times for surgery, up to 14 hours so far, and patient comfort. We believe the need for preoperative localization will remain as long as the application of thoracoscopic surgery increases.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, S.K.; study concepts and design, S.K.; literature research, S.D., S.K.; clinical studies, S.D., S.K., A.A., T.H., Y.K., K.Y., H.M., N.S.; data acquisition, S.D., A.A.; data analysis/interpretation, S.D., S.K.; statistical analysis, S.D., M.K.; manuscript preparation and definition of intellectual content, S.D., S.K.; manuscript editing, S.K., S.A., M.K., A.A., N.S., Y.H.; manuscript revision/review, S.D., S.K., N.S., Y.H.; manuscript final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Ikeda, H. Nomori, T. Mori, H. Kobayashi, K. Iwatani, K. Yoshimoto, and K.-i. Kawanaka Impalpable Pulmonary Nodules With Ground-Glass Opacity: Success for Making Pathologic Sections With Preoperative Marking by Lipiodol Chest, February 1, 2007; 131(2): 502 - 506. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-i. Watanabe, H. Nomori, T. Ohtsuka, M. Kaji, T. Naruke, and K. Suemasu Usefulness and complications of computed tomography-guided lipiodol marking for fluoroscopy-assisted thoracoscopic resection of small pulmonary nodules: Experience with 174 nodules. J. Thorac. Cardiovasc. Surg., August 1, 2006; 132(2): 320 - 324. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Sortini, C. V. Feo, P. Carcoforo, G. Carrella, E. Pozza, A. Liboni, and A. Sortini Thoracoscopic Localization Techniques for Patients With Solitary Pulmonary Nodule and History of Malignancy Ann. Thorac. Surg., January 1, 2005; 79(1): 258 - 262. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Ciriaco, G. Negri, A. Puglisi, and P. Zannini Reply to Carcoforo et al. Eur. J. Cardiothorac. Surg., July 1, 2004; 26(1): 234 - 235. [Full Text] [PDF]
|
|
|
|
|
|
P. Carcoforo, D. Sortini, E. Pozza, and A. Sortini Necessity of needle wire localization during video assisted thoracic surgery for patients with solitary pulmonary nodule Eur. J. Cardiothorac. Surg., July 1, 2004; 26(1): 233 - 234. [Full Text] [PDF]
|
|
|
|
|
|
S. Yamada and T. Kohno Video-assisted thoracic surgery for pure ground-glass opacities 2 cm or less in diameter Ann. Thorac. Surg., June 1, 2004; 77(6): 1911 - 1915. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Watanabe, T. Watanabe, H. Ohsawa, T. Mawatari, Y. Ichimiya, N. Takahashi, H. Sato, and T. Abe Avoiding chest tube placement after video-assisted thoracoscopic wedge resection of the lung Eur. J. Cardiothorac. Surg., May 1, 2004; 25(5): 872 - 876. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Chen, C. D. Lehman, and K. E. Dee MRI-Guided Breast Biopsy: Clinical Experience with 14-Gauge Stainless Steel Core Biopsy Needle Am. J. Roentgenol., April 1, 2004; 182(4): 1075 - 1080. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Sortini, D. Sortini, and G. Carrella Preoperative localization techniques during thoracoscopic operations J. Thorac. Cardiovasc. Surg., August 1, 2003; 126(2): 608 - 609. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sortini, D. Sortini, E. Pozza, G. Carrella, S. Dendo, and S. Kanazawa Intrathorascopic US: Usefulness in Localization of Solitary Pulmonary Nodules [letter] * Drs Dendo and Kanazawa respond: Radiology, July 1, 2003; 228(1): 291 - 292. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
