| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;80:2063-2069
© 2005 The Society of Thoracic Surgeons
a Department of Cardiothoracic Surgery, Medical University Vienna, Vienna, Austria
b Department of General Surgery and Transplantology, Medical University Vienna, Vienna, Austria
c Department of Anatomy and Cell Biology, Medical University Vienna, Vienna, Austria
Accepted for publication May 17, 2005.
* Address correspondence to Dr Mueller, Department of Cardiothoracic Surgery, Medical University Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria (Email: michael.mueller{at}meduniwien.ac.at).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentNotice From the American...AcknowledgmentsReferences |
|---|
METHODS: Eighteen juvenile pigs (median weight 40.6 kg) were randomly assigned to three groups. In all animals left pneumonectomy was performed and the stapled bronchial stump (median length 3.8 cm) carefully denuded. Group I animals received coverage of the stump by intercostal flap. In group II, the stump was covered with TachoComb, an impermeable hemostatic fleece; and group III served as a control without any coverage of the stump. Animals were sacrificed at day 14 after surgery. Vascular density was evaluated in serial histologic sections at multiple levels stained with CD-31 antibody. One-way analysis of variance and the Wilcoxon test were used for data analysis.
RESULTS: At autopsy, stumps of group III animals were totally covered by adjacent mediastinal structures. In group I, intercostal flaps were viable and completely healed to the bronchial stumps. There were no signs of infection or stump insufficiency in these groups. In all group II animals, empyema developed, and stumps were found necrotic at macroscopic and histologic evaluation. Statistical analysis revealed significantly lower vascular density of mature vessels in the area of the bronchial stump in group II compared with both other groups.
CONCLUSIONS: Reconstitution of microcirculation of the denuded bronchial stump after pneumonectomy takes place in a centripetal way from adjacent viable tissue. Hence, the purpose of covering the bronchial stump is the improvement of blood supply rather than mechanical reinforcement.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentNotice From the American...AcknowledgmentsReferences |
|---|
To reduce the incidence of this life-threatening complication, many authors recommend bronchial stump coverage with viable autologous tissue, such as intercostal muscle [2], pleura or pericardium [3], pericardial fat pad [1, 4], diaphragm [2, 5], or omentum [6]. The lowest incidence rates have been reported in the series of two different single surgeons [1, 7]. In the Goldstraw series (al-Kattan K et al [7]), bronchial stumps were hand sutured and not covered resulting in an incidence of 0.5% (2 of 410) stump insufficiencies. Klepetko and colleagues [1] report 0.8% (1 of 129) using coverage of the stapled bronchial stump with different viable tissues. Hence, it remains unclear whether coverage is mandatory and which role the different flap techniques play in the prevention of postpneumonectomy bronchial stump fistulas [1].
To facilitate procedures and reduce operation time, surgeons might be tempted to use nonviable biodegradable material like hemostatic fleece. However, such material has never been evaluated for this purpose systematically. Hemostatic fleece leads to an air- and fluid-tight sealing of tissues and stops formation of adhesions at its external surface. Hence, growing in of granulation tissue from adjacent mediastinal structures is prevented and may hamper restoration of blood supply of the bronchial stump. On the other hand, negative clinical experience with this approach has not been reported so far.
We studied healing of the denuded bronchial stump after pneumonectomy in an experimental animal model. The aim of this study was to investigate the process of reconstitution of blood supply of the bronchial stump after systematic removal of the adventitial layer to mimic the clinical situation of radical surgery, especially after induction therapy.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentNotice From the American...AcknowledgmentsReferences |
|---|
Experimental Setting and Surgical Technique
Eighteen juvenile pigs with a median weight of 40.6 kg (range, 35 to 45 kg) were planned for left pneumonectomy and randomly assigned to three groups of 6 animals each: (1) group I: coverage of the bronchial stump with viable autologous tissue; (2) group II: coverage of the bronchial stump with hemostatic fleece (TachoComb; Nycomed International, Roskilder, Denmark); and (3) group III: control group without coverage of the bronchial stump.
Owing to the existence of a so-called pig bronchus for the right upper lobe directly deriving from the distal trachea, we decided to perform pneumonectomies on the left side. Pigs have a quite long left main stem bronchus of about 4 cm and a shallow aortopulmonary window, resulting in a long free portion of the stump after pneumonectomy.
All procedures were carried out under general anesthesia (premedication with 1.0 mg atropinsulfate intramuscularly, introduction with 20 mg/kg thiopental intravenously; maintenance with orotracheal intubation, volume-controlled ventilation with oxygen-NO2 mixture and 1 to 2 vol/100 mL isofluran; 0.1 mg fentanyl as a bolus/h). After skin preparation with an aqueous iodine solution and surrounding of the operation field with sterile drapes, left pneumonectomy was performed through lateral thoracotomy, analogous to the corresponding procedure in humans.
After mobilization of the left lung, the lower lobe vein was identified and dissected using ligation. The upper lobe vein and left pulmonary artery were divided in the same way. The left main stem bronchus was then pulled and exposed until its origin from the trachea, using a silk string. Radical lymphadenectomy was carried out, including the infracarinal and lower left paratracheal groups. Bronchial arteries were identified and closed as high as possible using ligatures. The adventitia of the bronchus was then thoroughly removed in a standardized fashion using dissection and a periosteal elevator. We did not experience any difficulties with this procedure. The bronchus was closed with a 30-mm bronchial stapler (AutoSuture Premium; Tyco Healthcare Group LP, Mansfield, MA) and cut at the level of the bifurcation of the left main stem bronchus. After further treatment of the denuded bronchial stump according to the referring experimental group (I, II, III), the surgical procedure was completed and the chest closed in a standard way without placing a chest tube.
In group I animals, a pediculated intercostal flap was prepared before spreading the ribs. After completion of the whole procedure, the well-perfused flap was attached to the anterior and posterior aspect of the denuded bronchial stump using interrupted monofilic 4.0 PDS (Ethicon, Somerville, New Jersey) stitches, resulting in a complete coverage of the bronchus.
In group II animals, one unit of hemostatic fleece (TachoComb, 9.5 x 4.8 cm) was divided in half, briefly moistened in saline solution, and applied to the denuded stump in a two-step approach. One half was used in an anterior-posterior direction and the second half across to the first one, covering the edges of the bronchus.
TachoComb is a combination that consists of a patch of dry porous collagen with one side impregnated with fibrinogen and thrombin coagulation factors. The mode of action replicates the last step of the blood coagulation cascade. Thrombin transforms fibrinogen into a fibrin monomer by splitting off fibrin peptides. This enables the collagen to strongly conglutinate with the wound surface to create an air and liquid tight seal that also improves the tensile strength of the tissue. (Within the member states of the European Union, TachoComb was substituted in 2004 by its successor TachoSil. As has been shown by biomechanical and clinical studies, TachoSil's characteristics are comparable with those of TachoComb.)
In group III animals, no further treatment was applied to the denuded bronchial stump.
Wound closure was carried out with 1.0 silk, 2.0 Vicryl, 2.0 silk. Buprenorphin 0.3 mg was administered subcutaneously every 8 hours for 3 days after the procedure.
Autopsy and Histologic Examination
Animals were sacrificed at day 14 after surgery by means of thiopental and potassium chloride overdose. During autopsy, the distal trachea and both mainstem bronchi were excised en bloc. Full-thickness stripes of bronchial wall together with the adjacent structures were prepared for histologic evaluation.
The areas of interest were defined as proximal, intermediate, and distal levels. The proximal level comprises the distal trachea with its original adventitial layer; the intermediate level represents the proximal part; and the distal level, the distal part of the bronchial stump. Specimens were taken from both flanks (a, b) of the bronchial stump or distal trachea at each level.
Specimens were immersion-fixed in 5% formaldehyde and embedded in paraffin. Serial sections were prepared from each segment at 7 µm.
To evaluate the microvascular density, sections were stained with a CD31 (endothelial cell-specific marker) mouse monoclonal antibody (Serotec, Oxford, United Kingdom [dilution 1:50]). The primary antibody was recognized by staining with a secondary rabbit anti-mouse antibody (Dako, Glostrup, Denmark). The specific staining was developed with benzoyl peroxide. Positively stained vessels were classified and allocated to three different types, namely, capillaries (5 to 12 µm), arterioles (13 to 500 µm), and arteries (greater than 500 µm). The classification of vessels was based on the vascular diameter scoring that correlates with the vascular wall architecture (concerning the muscularity and pericyte composition), as described by our group and others [8–10].
Two slides of each specimen were analyzed (total number of specimens, n = 108; total number of slides, n = 216). The specimens were screened first at x10 magnification. Then, three high-power microscopic fields that did not overlap were studied in each slide, and the vascular density was determined. Vascular density was expressed as number of vessels in 1.5 mm2 (Tables 1, 2, and 3). 
The pathologist was masked as to which groups the slides belonged.
|
|
|
Randomization of animals was performed according to the method of randomly permuted blocks using the free online program available at www.randomization.com.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentNotice From the American...AcknowledgmentsReferences |
|---|
No wound infections were observed in any of the animals. None of the animals had fever or showed any other signs of infections in the postoperative period. Animals of groups I and III gained weight like normal nonoperated pigs. Animals of group II had a tendency toward gaining weight slower compared with the other groups; however, this difference was not statistically significant (median of weekly increase in weight was 4.5 kg in groups I and III, and 4 kg in group II; p = 0.336).
Autopsy
At autopsy, all animals showed perfectly healed thoracotomies 14 days after surgery. Median length of left bronchial stump from the carina to the staple line was 3.8 cm. In group III animals, the bronchial stump was totally covered by adjacent mediastinal structures and young granulation tissue. There was no sign of an inflammatory process in any of these animals. In group I, the intercostal flaps were viable and completely healed to the bronchial stumps. There was no sign of infection in animals of this group as well.
In all animals of group II, putrid infection with abscesses and fibrinous membranes (postpneumonectomy empyema) was observed at autopsy. All bronchial stumps were found macroscopically necrotic and not covered by any material except loose fibrin masses. In 1 animal, a partial stump insufficiency at the top of the stump had already developed. Bacteriologic cultures revealed beta-hemolysing Streptococcus organisms in all cases.
Histologic Findings
The effectiveness of denudation of the bronchial stump was confirmed by histologic examination of the left bronchial stump in the 1 animal (group III) that died early after surgery. In all animals of group I, a large number of newly formed mature vessels of all three categories were found. There was no sign of inflammation, and all anatomical structures of the bronchial wall were found in proper condition (Fig 1). Animals of group III revealed a similar histology compared with group I, except for a slightly reduced density of mature arteries and a higher percentage of arterioles (Fig 2). In contrast, among group II animals, almost no mature vessels could be detected in the area of the bronchial stump. There were large numbers of immature sinusoids and capillaries, associated with inflammatory cells and lymphocytes (Fig 3). Accordingly, the bronchial wall was severely damaged, as indicated by the lysis of the bronchial cartilage. Among the existing microvessels surrounded by the inflammatory cells, 52% had a discontinuous endothelial lining with signs of extravasation, and 37% were occluded by thrombi. These immature sinusoid vessels were excluded from statistical analysis.
|
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentNotice From the American...AcknowledgmentsReferences |
|---|
From basic surgical knowledge, it may be derived that improving local perfusion may contribute to avoid stump insufficiencies after pneumonectomy. However, just a few publications support this hypothesis and put it on a more scientific basis. A potential role of mechanical reinforcement of the staple line has not been studied as well. Reviewing the literature, we have found three relevant papers from the 1940s and 1950s. Modern experimental work simply does not exist, to our knowledge. According to one of the earliest observations published in 1942 by Rienhoff and colleagues [11], the main point of healing of the bronchus is at the cut end. Every effort should be made to preserve the viability of this portion, not only by gentle handling, which avoids any form of trauma such as crushing, cauterization, or suturing, but also by preserving the circulation in the bronchial artery. The use of occluding nonabsorbable sutures is advocated as a means of closing the bronchial tube for a sufficient length of time to permit healing of the cut ends. The bronchial stump should be brought into apposition with nearby or contiguous viable tissue, preferably mediastinal pleura.
In a German paper from 1951, Holle and Viereck [12] reported on experimental study in 10 dogs. The authors explain the process of healing of the bronchial stump through the development of purulent inflammation. In this study, the use of free periostal flaps to cover the bronchial stump proved to be a very reliable method, offering a quite good prevention of bronchial stump insufficiency even after expectoration of the necrotic stump. Free periostal flaps proved to be durable enough to cover the abscess area, maintain a local inflammation, and prevent free perforation.
In contrast, based on an autopsy series including 20 patients a few hours for as long as 479 days after pneumonectomy, Schütz [13] in 1955 stated that the development of an abscess of the bronchial stump was a rare finding and might represent the onset of bronchial fistula. In addition to other recommendations concerning handling of the bronchial stump, he postulated avoiding the use of nonpediculated tissues, as such material may only provide mechanical reinforcement of sutures but would not contribute to biological sealing of the bronchial stump by means of formation of granulation tissue. Besides this paper lacking an experimental design, the negative effects of using freely transplanted tissues like fascia lata or periostium at the bronchial stump has not been proven.
Surgical techniques and suture materials have improved significantly in the last 50 years and hardly can be compared with modern thoracic surgery, making the interpretation of these postmortem findings even more difficult. Coughing out the sutures was considered part of the regular healing process at the bronchial stump in those days. Hence the literature does not provide sufficient information regarding the value of either free or pediculated flaps in preventing postpneumonectomy bronchial stump insufficiency.
Clinical experience with patients undergoing extensive mediastinal lymphadenectomy after induction radiotherapy [14] suggests that healing of the bronchial stump mainly depends on the reconstitution of disturbed microcirculation through newly built vessels from adjacent structures. The time until restoration of sufficient blood supply is to be considered a crucial phase. This concept has been supported by the fact that bronchial stump insufficiency occurs significantly more often on the right side [15]. It can be supposed that, owing to a more extensive mediastinal lymphadenectomy, the right bronchial stump is denuded to a greater extent compared with the left side. It seems impossible to achieve the same degree of denudation of the left main stem bronchus because of its anatomical localization inside the aorto-pulmonary window, thus microcirculation disorders in the bronchial stump after left pneumonectomy are probably less severe in comparison with the right side.
In choosing the suitable model for our study, we first considered the setting used in the experimental work conducted by Morgan and associates [16] in the preclinical phase of lung transplantation. In their model, successful revascularization of totally ischemic bronchial autografts was achieved by the use of omental pedicle flaps. In our study, however, restoration of blood supply coming from the distal trachea should not be prevented, and therefore stumps were not cut and reanastomosed.
To isolate the denuded bronchial stump from adjacent mediastinal tissues we used hemostatic fleece (TachoComb). The physical and biochemical properties of this material enable quick and strong attachment to various surfaces and result in an air- and fluid-tight sealing. The ability of TachoComb to completely isolate bronchial stumps from adjacent mediastinal structures was used in our study to investigate alternative ways of restoration of blood supply.
Although fibrin mediates acute inflammatory responses to biomaterials [17], biocompatibility studies of TachoComb have shown good histocompatibility, moderate biodegradability, and lack of toxicity of collagen carriers in combination with fibrin glues [18]. Moreover, TachoComb induces mitogenic and chemotactic processes and thereby improves wound healing. Hence, hemostatic fleece has not been demonstrated to cause negative effects in tissues with regular blood supply. Inoculation of bacteria with the hemostatic fleece may be excluded, as this material was sterile and identical to that used in the operating room in human recipients.
Coverage of the bronchial stump by muscle flap (group I) provided early restoration of blood supply in the denuded bronchial stump. Direct contact of the bronchial stump to adjacent mediastinal tissue allowed proper restoration of microcirculation even without additional coverage (group III). To the contrary, isolation of the denuded bronchial stump by an impermeable hemostatic fleece (group II) almost completely prevented revascularization resulting in necrosis of the bronchial stump and secondary pleural empyema in all cases.
If the adventitial layer of the bronchial wall and its vessels had been systematically removed and revascularization from surrounding tissues prevented in group II animals, how can the high density of immature sinusoid vessels be explained in this group? The submucosal capillary bed of the bronchus with its transmural anastomoses to the adventitial vessels alone seems to be insufficient for the survival of the bronchial wall. Ischemia and incipient necrosis lead to increased permeability of submucosal vessels, inflammation and migration of bacteria. As a result, many sinusoid capillaries of a granulation reaction occur and occlude parallel with the decline of the bronchial wall.
We conclude that reconstitution of the microcirculation of the denuded bronchial stump after pneumonectomy takes place in a centripetal way from adjacent viable tissue. Hence, the purpose of covering the bronchial stump is the improvement of microcirculation rather than mechanical reinforcement.
|
Notice From the American Board of Thoracic Surgery Regarding Trainees and Candidates for Certification Who Are Called to Military Service Related to the War on Terrorism |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentNotice From the American...AcknowledgmentsReferences |
|---|
Timothy J. Gardner,MD
Chairman
The American Board of Thoracic Surgery
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentNotice From the American...AcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentNotice From the American...AcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. G. Saueressig, A. M. Neto, E. A.F. Fortis, D. Westphal, M. I.A. Edelweiss, L. Meurer, and U. Matte Vascular endothelial growth factor gene therapy induces early re-establishment of canine bronchial circulation Eur. J. Cardiothorac. Surg., April 1, 2008; 33(4): 717 - 722. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. Sturridge, M. R. Mueller, and T. Treasure Blood Supply of the Trachea and Proximal Bronchi Ann. Thorac. Surg., August 1, 2007; 84(2): 675 - 675. [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 |
