|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Medizinische Klinik II, Städtisches Krankenhaus Bad Reichenhall (Drs. Hauch and Lambeck), and Pulmonary Division, 1. Medizinische Klinik, Klinikum rechts der Isar and Deutsches Herzzentrum (Dr. Peltz) and Institut für Pathologie (Drs. Barbur and Werner), Technische Universität, Munich, Germany.
Correspondence to: Rainer W. Hauck, MD, Assistant Professor, Head of Heart and Pulmonary Division, Städtisches Krankenhaus Bad Reichenhall, Riedelstrasse 5, D-83435 Bad Reichenhall, Germany; e-mail: hauck{at}krankenhaus-bad-reichenhall.de
| Abstract |
|---|
|
|
|---|
Patients and methods: In 17 patients, intraluminal tumor tissue was histologically investigated before and 1 week after stent implantation. The amount of intact tumor cells (ITCs) was compared to necrotic and nontumor cells. The result in patency was proved by fiberoptic bronchoscopy.
Results: Initially, stenoses in all patients were > 75%. Before stent implantation, biopsy samples in seven patients showed > 67% ITCs, and five patients had 34 to 67% ITCs. Five patients had 1 to 33% ITCs, and no patients had 0% ITCs. One week after stent implantation, the cellular aspect of the biopsy samples had changed significantly (p < 0.03): two patients had > 67% ITCs, one patient had 34 to 67% ITCs, and seven patients had 1 to 33% ITCs. Seven patients had no ITCs at all. Endoscopically, patency increased significantly (prestent, 10 ± 14.1%; poststent, 90.6 ± 14.3% [mean ± SD]; p < 0.0001).
Conclusion: Pressure exerted by the stent on adjacent tumor tissue may cause a profound reduction in the amount of ITCs, most probably caused by radial and shear stress forces that compromise blood supply and nutrients of the tumor stroma.
Key Words: bronchial stenosis bronchoscopy histology lung cancer necrosis stent
| Introduction |
|---|
|
|
|---|
| Materials and Methods |
|---|
|
|
|---|
|
In addition, one patient was included in the study who refused initially surgical resection of his stage II squamous cell lung cancer. This tumor caused a longer stenosis (> 75%) of the distal right main and intermediate bronchus. Therefore, stent implantation had to be performed to restore patency of the right bronchial system. Three weeks later, the patients physical condition had improved, and a curative operation was requested. The surgically resected bronchial specimen was examined.
Self-expanding, uncovered metal stents were implanted in all patients (Ultraflex; Boston Scientific; Watertown, MA). The stents were made of nitinol, an alloy of nickel and titanium with thermal as well as shape memory, and ranged in diameter from 12 to 14 mm and were 40 mm in length. Bronchoscopy and stent implantation were performed as described.5 7 Written and informed consent was obtained from all patients for the procedure.
Tissue Preparation for Histologic Investigation
Biopsy samples were fixed in 4% buffered formaldehyde solution for 24 h. The samples were routinely washed with graded ethanols and xylole and embedded in paraffin. Subsequently, eight serial sections were cut with a rotation microtome (Microm; Walldorf, Germany). The 2 µm-thick slices were subsequently stained with hematoxylin-eosin. From the surgically obtained specimen, several tissue sections from the stent area were obtained and prepared by the same method. After preparation, all slides were analyzed by light microscopy (Zeiss; Gottingen, Germany) [magnification, x12.5 to x400]. Cells were counted as percentage of all cells per slide. The pathologist did not know patients name and the date when the biopsy samples were harvested.
Samples were classified into four groups according to the percentages of intact tumor cells (ITCs) vs necrosis and/or nontumor cells (N/NTCs). Nontumor cells consisted of fibroblasts, endothelial cells, various inflammatory cells, and nonneoplastic epithelium (Table 2 , Fig 1 ). Patients in category 1 had no ITCs in tissue biopsy samples. If biopsy samples had a low number (1 to 33%) of ITCs, they were classified as category 2. If samples contained 34 to 67% of ITCs, they were classified as category 3. Category 4 included samples with a high number of ITCs (> 67%). In the surgically obtained specimens, cell composition was analyzed in the bronchial wall and the surrounding tissue of the region where the stent had been implanted.
|
|
| Results |
|---|
|
|
|---|
|
|
Histologic follow-up 7 days after stent implantation revealed two patients (12%) with > 67% ITCs, one patient (6%) with 34 to 67% ITCs, and seven patients (41%) with 1 to 33% ITCs. Thus, in only three cases (18%), cellular composition of the tumor in the stent area 1 week after the intervention consisted of 34 to 100% ITCs (Table 3) .
In most of the samples (n = 16), in-stent tissue contained granulation tissue with varying amounts of fibroblasts, small capillaries, and inflammatory cells. In five samples, metaplastic squamous epithelium was seen. Occasionally, fibrinous secretions were observed.
Matched-pair analysis of the cellular composition of biopsy samples from each of the 17 patients before and 7 days after stent implantation revealed a decrease in ITCs in 12 patients, and no difference in grade in 2 patients. Three patients showed an increase in ITC grade (Fig 3 ). The increase in N/NTCs after stent implantation was significantly higher than the amount of N/NTCs that were obtained prior to stent implantation (p < 0.03).
|
| Discussion |
|---|
|
|
|---|
Metallic stents are increasingly being used in the palliation of malignant bronchial stenosis; their clinical efficacy has been proved in many studies.1 5 8 9 10 However, the use of uncovered stents in tracheobronchial exophytic lesions remains controversial because of the potential risk for bronchial restenosis by tumor tissue ingrowth into the stent lumen.4 11 Because tumors are frequently located at bronchial branch points, implantation of a covered stent in these areas would occlude the exit of the segmental airway involved, thus potentially allowing atelectasis and pneumonia to develop.12 13 Therefore, uncovered metal stents are frequently used for recanalization and stabilization of exophytic tumor stenoses.
Until now, little data has been published concerning the histologic composition of material that penetrates the wire mesh of airway stents. Hence, in this study, our interest focused on the mechanical influence of self-expanding metal stents on tumor tissue. Since these prostheses exert permanent radial stent forces on superficial tumor parts, they compress localized tumor cells, impair supply with blood and nutrients, and thus can impact viability. Possibly, even influences of metal ions liberated by the presence of the stent could gain some relevance.
Most studies in this field have been performed in the GI tract. Bethge et al14 looked for stent influences in surgically and postmortem-obtained specimens from the esophagus and bile duct. They found that tumor protrusion through the stent mesh occurs in all cases within 1 month. In patients with longer survival time, the stent became embedded in collagen or fibrous tissue, but stent occlusion by ingrowth of viable tumor tissue was not seen in any case. Histologically, inflammatory cells (ie, granulocytes and lymphocytes) were found, as well as fibrosis and necrotic tissue. These findings correspond with data from Hausegger et al15 from the biliary tract. They demonstrated that no proliferation of solid intraluminal tumor was seen within uncovered metal stents. In-stent tissue consisted mainly of necrotic tumor cells and granulation tissue. This was considered a consequence of radial stent forces on the tumor surface, which can compromise blood supply of the tumor tissue and in consequence reduce tumor cell viability.15 Similar findings were suggested by Grewe et al6 in the human bronchus. They observed that tumor growth appears to be much slower at the site of the stent implantation than in the adjacent bronchus.
Endoscopic analysis showed comparable results in prospectively analyzed patients with lung cancer treated with uncovered metal stents for exophytic lesions.5 In the majority of these patients, the stents were penetrated but not compromised in their patency to a functionally relevant degree (> 75%). During a mean follow-up of 122 days, tumor stenosis > 75% occurred in only approximately 20% of the patients.
Our study investigated the influence of pressure forces on tumor tissue. As mostly no circumscript tumor growth was underlying, laser resection or argon plasma coagulation was not used. However, forceps debulking was performed in most of the cases. The harvested tissue was used for the histologic investigations. Since necrosis commonly occurs in malignant tumors, biopsy samples obtained directly before stent implantation were compared to those harvested after the stent had been in place for 7 days. The follow-up window was kept so narrow in order to minimize influences of tumor aging and effects of tumor therapy on the cellular composition of the biopsy samples. In all patients, tumor therapy was kept constant 2 weeks prior to and within the follow-up week. However, it can be argued that the main influence of stent forces on the tissue beneath the stent will occur within this time. In these comparative investigations, we showed that before the intervention ITCs dominated the cellular aspect of the biopsy samples. In more than two thirds of the patients, ITCs in the biopsy samples were > 66%. Interestingly, this ratio was significantly different from the poststent studies, where less than one third of the patients showed such a high percentage of ITCs. In contrast, the percentage of ITCs decreased significantly in the 7 days after the stent had been implanted.
Comparably, the effect of stent forces on ITCs in stent-penetrating tissue was also illustrated by the surgically resected lung specimen. Histologic sections obtained from the stent area showed that, close to the stent, necrotic tumor cells dominated the cellular pattern, whereas with increasing distance from the stent (ie, with decreasing stent forces), ITCs were seen more frequently.
As in our former investigation5 excellent endoscopic restoration of bronchial patency after stent implantation was maintained in nearly all cases after 1 week of follow-up. This, however, was not unexpected at a 7-day follow-up time.
When self-expanding metal stents are deployed in stenotic bronchial lesions, the stent first expands and induces expansive and shear-stress forces on the surrounding tissue. As the pressure continues, blood supply in the tumor decreases, and pressure necrosis in the compressed tumor tissue occurs.14 Depending on the force required to distend the tumor-related stenosis, the degree of shear stress can vary. Additionally, stretching of the airway lumen can evoke shear stress and tangential forces in the stent area, contributing to tumor cell death. If the pressure continues, necrosis beneath the wire mesh leads to migration of the stent deeper into the tumor.14 This delayed expansion of the stent is a well-known characteristic; it precludes accurate estimation of final stent expansion size for 24 to 48 h after stent implantation.16
In this study, most of the treated stenoses were caused by squamous cell carcinomas. Interestingly, no relationship between the amount of vital tumor cells in the stent area and the histologic type of the tumor was found. In contrast, in the GI tract, more ingrowth has been found in cell-rich or poorly differentiated anaplastic carcinomas.15 This tendency could not be demonstrated even by endoscopic means, and thus it does not seem to be relevant in judging the potential risk of restenosis in uncovered stents.
The data obtained do not allow conclusions concerning the influence of expansile forces on bronchial structures in benign stenoses. The formation of granulation tissue at the stent endings, which has been reported for both metallic and silicone stents, must be regarded as a different cellular response.17 18 It is triggered by shear-stress forces at the stent edges, where a more localized contact between stent and bronchial mucosa exists.19 20 In one experimental study21 of the biliary tract, marked hyperplasia of the mucosa and submucosal fibrosis was seen, particularly at the proximal and distal stent edges. These alterations varied from animal to animal, and thus suggested an individualized pattern of response. Another study22 performed in dogs showed that these tissue responses were quite exuberant, and led to compromise in the luminal stent diameter, a finding that thus far has not been reported with metallic stents used in inflammatory strictures.3 23 Within the stent body, in contrast, force is distributed more homogenously; consequently, formation of granulation tissue has not yet been observed in this region. In a study by Susanto et al,24 who used metal stents for the treatment of bronchial stenoses and tracheomalacia in seven patients after lung transplantation, the stents were embedded in the mucosa and epithelialization began within a few weeks. No granuloma formation or newly developing necrosis was reported in the stent area.24 In human subjects, epithelialization of the stent occurs within 3 to 6 weeks after implantation, usually without compromise in stent patency.25 26 Localized chronic or mild inflammatory reaction of the stent penetrating tissue is found to a variable degree in both experimental and in vivo studies,14 25 and was the case in the majority of samples in this study as well. Of special interest is the fact that in all studies involving benign lesions, no necrotic cells were observed in the follow-up period of up to 3 months. Even in the operative specimen from this study, no necrotic changes were seen in epithelial cell layers that were not infiltrated by neoplastic cells. This suggests that normal bronchial tissue has a different susceptibility, and differing cellular responses, to pressure forces generated by a stent than is the case with neoplastic tissue.
In conclusion, this study demonstrates that the amount of ITCs decreases in neoplastic tissue located next to implanted metal stents. Radial stent and shear-stress forces seem to be causative by decreasing cell viability in superficial tumor areas. Such a favorable side effect of metal stents serves to help maintain stent patency. However, conclusions on long-term results cannot be completely based on these data.
| Acknowledgements |
|---|
| Footnotes |
|---|
Received for publication August 29, 2001. Accepted for publication April 17, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. H. Grewe, K. M. Muller, M. Lindstaedt, A. Germing, A. Muller, A. Mugge, and T. Deneke Reaction Patterns of the Tracheobronchial Wall to Implanted Noncovered Metal Stents Chest, August 1, 2005; 128(2): 986 - 990. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |