|
|
||||||||
Ann Thorac Surg 2003;75:184-189
© 2003 The Society of Thoracic Surgeons
a Heart Care Associates, Milwaukee Heart Institute, Milwaukee, Wisconsin, USA
b Lenox Hill Heart and Vascular Institute, Cardiovascular Research Foundation, New York, New York, USA
Accepted for publication July 26, 2002.
* Address reprint requests to Dr Kipshidze, Lenox Hill Heart and Vascular Institute, 130 East 77th St, Black Hall, 9th Floor, New York, NY 10021, USA.
e-mail: nkipshidze{at}lenoxhill.net
| Abstract |
|---|
|
|
|---|
METHODS: The latissimus dorsi muscle (LDM) was mobilized in 20 adult sheep. Two separate pockets were created in each sheep. Autologous fibrin sealant with or without 100 mg/mL of deferoxamine mesylate (10 pockets) was added to the pockets. Deferoxamine mesylate alone was also applied to another 10 pockets, whereas the 10 other pockets served as controls.
RESULTS: Conventional, indirect immunofluorescent en-face staining showed that in nonmobilized, nonischemic LDM the capillary density was 196 ± 14 capillaries/mm2 in the distal and 207 ± 19 capillaries/mm2 in the middle part. After severe ischemic shock (subtotal mobilization), the muscle did not recover completely even after 2 months (149 ± 15 capillaries/mm2 in the distal part and 177 ± 16 capillaries/mm2 in the middle part of the LDM). Fibrin application only increased muscle neovascularization. The number of capillaries per mm2 of muscle increased to 250 ± 25 in the distal part and to 271 ± 24 in the middle part of the LDM. However, when fibrin was applied with added deferoxamine mesylate, the capillary density increased to 361 ± 25 capillaries/mm2 in the distal part (p < 0.05 vs fibrin only; controls) and to 401 ± 20 capillaries/mm2 in the middle part of the LDM (p < 0.05 vs fibrin only and p < 0.001 vs controls). The data are concordant with the blood flow estimation before and after mobilization (severe ischemic shock) in the different parts of the LDM.
CONCLUSIONS: Local application of deferoxamine mesylate enhances neovascularization and recovery of surgically induced skeletal muscle ischemia in a sheep model.
| Introduction |
|---|
|
|
|---|
Iron chelators have been reported to interfere with inflammatory cells and possibly enhance vascular growth factor production [2]. We hypothesized that if the adverse influence of iron cations and free-radicals could be decreased or eliminated, endothelial cells would be better equipped to overcome ischemia and produce their own growth factors to promote angiogenesis and vasculogenesis in areas of high ischemic risk.
On the other hand, it was also necessary to devise a method for the prolonged release of deferoxamine mesylate into the ischemic tissue. Studies from several laboratories [3, 4] including our own [5] have shown that fibrin can be used as a carrier for the application and prolonged delivery of different pharmaceuticals, genetic products, and living cells. A fibrin network is critical for effective wound healing, and is biodegradable through routine tissue fibrinolysis. Because the fibrin sealant is slowly lysed, it can serve as a vehicle to deliver agents that may act to help heal wounds [68], to promote new vessel growth [3, 9], or to store and slowly release any therapeutic agent [6]. Recently we demonstrated the efficacy of fibrin sealant to deliver aprotinin and pyrrolostatin to ischemic skeletal muscle in order to enhance local neovascularization [10, 11].
Therefore the aims of this study were to investigate the efficacy of deferoxamine mesylate to facilitate neo-angiogenesis and enhance blood flow in ischemic skeletal muscle using the experimental sheep model.
| Material and methods |
|---|
|
|
|---|
All sheep underwent general anesthesia for each surgical procedure. The animals were pre-medicated with diazepam (Elkins-Sinn, Cherry Hill, NJ; 5 mg/kg IV) and anesthetized with thiopental (Abbot Laboratories, North Chicago, IL; 20 to 25 mg/kg IV). They were then intubated, placed on a Draeger (North American Draeger, Telford, PA) ventilator, and maintained on Halothane gas anesthesia (1% to 2% with 4.0 L O2). Oxygen saturation levels and heart rate were monitored by a pulse oximeter placed on the animals tongue. Strict sterile technique was followed at all times to reduce the potential for infection. The animals were started on a postoperative regimen of antibiotic therapy (Amoxicillin, 15 mg/kg IM once a day for 5 to 7 days), and monitored twice daily for signs of infection.
Surgical procedure
A 25-cm longitudinal skin incision was made in the left midaxillary line to expose the left latissimus dorsi muscle (LDM) of 20 animals. A flap of subcutaneous adipose tissue (approximately 6 x 16 cm) was dissected leaving one proximal section hinged to the remaining adipose tissue. The anterior border of the LDM was completely mobilized with dissection of all vessels from intercostal arteries. The posterior border of the LDM was left untouched.
Two different parts to the LDM (according to blood supply) were then created: the in situ posterior section with its undisturbed vascular supply and the anterior section in an ischemic state. Therefore the LDM anterior section consisted of the lateral segment of the LDM, and the posterior section consisted of the oblique and transverse LDM segments. Next, these sections were separated from one another (Fig 1A). The ischemic anterior portion was than dissected from its distal intercostal connection, thereby leaving it subtotally mobilized with the only blood supply originating from a branch of the thoracodorsalis artery, which normally supplies the proximal part of the LDM only. The flap of adipose tissue was positioned over the in situ LDM creating a barrier between the nonischemic LDM and the ischemic part of the LDM, which was positioned over the adipose tissue flap (Fig 1B). The tissue layers were sutured together with two pockets created (Fig 1C). Therefore, each pocket consisted of nonischemic and ischemic muscle. In 10 animals, one pocket fibrin sealant was injected, whereas other pockets served as a control (Fig 2). In another 10 animals, one pocket mixture of fibrin sealant and deferoxamine mesylate was introduced in one pocket, and in another pocket was the contents of only deferoxamine mesylate (100 mg/mL).
|
|
Blood flow analysis
Temporary catheters were placed in all animals into the femoral artery and through the subclavian artery into the thoracodorsalis artery, and seven and a half million microspheres of different colors were injected directly into the thoracodorsalis artery to determine collateral blood flow in the LDM before and 1 hour after muscle mobilization.
Two months later the procedure was repeated, and after euthanization the muscle was removed. Samples of LDM were taken to determine blood flow.
Histology
Transverse sections were made for conventional histologic (hematoxylin and eosin) staining and for subsequent evaluation. Multiple slides were made of each biopsy sample. Histologic data were submitted to an independent observer for interpretation. Particular attention was paid to evidence of muscle regeneration, thickness and composition of the reparative response, the density of neovascularization, the presence of large bore vessels, and margination of leukocytes.
Biopsy specimens for histologic determination of capillary density were taken from the ischemic and nonischemic flap of the LDM from each pocket in the same area in each animal. These samples were frozen for 30 seconds in a bath of liquid nitrogen. Multiple frozen sections were cut (10 µm thickness) on a cryostat and placed on glass slides. Then the samples were stained for alkaline phosphatase using the indoxyl-terazolin method. A pathologist who was blinded to the series of experiments counted the number of capillaries less than 20 x objectives in a chosen field (a total of 10 fields from two different sections of the muscle tissue per sample). Capillary density was calculated as the number of capillaries per square millimeter of muscle.
Data analysis
Values are reported as mean ± standard deviation of the mean. Data were compared using an analysis of variance and a subsequent Student Newman-Keuls t test if applicable. A p value of less than 0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
|
|
|
|
| Comment |
|---|
|
|
|---|
The present study indicates that ex-vivo therapy, consisting of deferoxamine mesylate application, may be used to successfully promote neovascularization. This also previously demonstrated that normal activity of basic fibroblast growth factor (bFGF) and vascular endothelial growth factor (VEGF) in endothelial cells did not change under the influence of deferoxamine mesylate. However the activity of both growth factors increased up to 5 to 10 times (deferoxamine mesylate concentration at 1 mmol/L) in the fibroblasts and smooth muscle cells [12]. We confirmed that the iron chelator triggers an angiogenic program in the in vitro fibroblasts and smooth muscle cells by releasing at least two growth factors (bFGF and VEGF). When released together from fibroblasts and smooth muscle cells, bFGF and VEGF are able to produce an angiogenic response from endothelial cells far greater and also with greater than either mitogen alone. Further growth factors will recruit more quiescent endothelial cells in the LDM to amplify and create new capillaries.
In 1996, Beerepoot and colleagues [2] showed that deferoxamine mesylate, at concentrations achievable in humans, induced a threefold to fivefold increase in mRNA expression. These findings suggest that VEGF may act as a mediator of the side effects induced by iron chelation therapy and iron chelators should be given with caution to cancer patients.
In a previous study [10, 11], we also noted that fibrin sealant alone, when applied to ischemic muscle, stops bleeding by forming a layer around the traumatized LDM, thereby protecting it from severe damage. On day 14 after LDM mobilization we noted considerably less leukocyte migration, fibrosis, and calcified necrosis in tissue treated with fibrin sealant than in nontreated ischemic LDM. Moreover, transmission electron microscopy showed newly formed capillaries by day 56 in the interlayer between ischemic and nonischemic muscles when fibrin sealant was applied. In controls there was no capillary growth from either ischemic or nonischemic muscle by day 56. However, when some pharmaceutical agents were added to fibrin sealant (aprotinin, a proteinase inhibitor, or pyrrolostatin, a free radical scavenger) the damage of endothelial cells was attenuated and neovascularization of ischemic tissue was greater than with fibrin sealant alone.
In the present study we investigated the possibility of deferoxamine mesylate improving muscle neovascularization. Deferoxamine is known to reduce the iron-dependent generation of toxic oxygen-derived radicals during ischemia [13]. It was demonstrated that endothelial cells from several species generate a burst of free radical generation upon reoxygenation, which obligately accompanied revascularization during clinical treatment of ischemic or infarcted tissue [12]. The authors also showed that deferoxamine mesylate decreased the measured radical levels by 40%.
Several experimental and clinical investigations have shown the positive effects of iron chelators [14] and demonstrated that treatment with polymeric iron chelators significantly attenuates systemic oxidant injury, with the degree of protection being more impressive in the lung and kidney in cases of septic shock. Katoh and colleagues [15] and Morris and colleagues [16] showed that deferoxamine mesylate reduces myocardial injury and free radical generation in isolated neonatal rabbit hearts subjected to global ischemia-reperfusion.
Our investigations also showed that a combination of autologous fibrin and iron chelator decreases the level of ischemia of mobilized LDM and increases angiogenic potential. Our data correlate with findings of other investigators. Fantini and Yoshioka [17] showed that high-grade partial ischemia in skeletal muscle is accompanied by iron-dependent lipid peroxidation through a mechanism that persists and accelerates reoxygenation lipid peroxidation impact on functional membrane integrity during the reperfusion phase only. Hickley and colleagues [18] showed that administration of free radical scavengers, immediately before and during the early minutes of reperfusion, improved muscle survival 24 hours after the ischemic period. The dose effect of deferoxamine mesylate treatment in attenuation of ischemia induced reperfusion injury in the skin and muscle of latissimus dorsi myocutaneous flaps were demonstrated by Morris and colleagues [16] and Patterson [19]. During our investigation, we also showed that application of deferoxamine mesylate within fibrin matrix to the ischemic LDM increased blood flow to the muscle more than twofold compared with nontreated mobilized muscle. Damages of endothelium, which have a place after onset of acute ischemia, cannot be fixed during the short period of time. From the gap into the endothelium, the erythrocytes continue to leave the capillaries for several weeks and there is continuous releasing of the hemoglobin and iron ion from them. So, after one injection, deferoxamine mesylate will chelate only those ions that formatted before injection, but after that another ion will be formatted and produce the same damage effect on endothelium. To prevent this continuous damage effect it is necessary to administrate deferoxamine mesylate every day during several weeks, or use fibrin sealant for its step-by-step delivery.
Several investigators have also shown that deferoxamine mesylate enhanced myocardial protection during ischemia [20] helped to avoid myocardial injury and dysfunction after cardiopulmonary bypass [21], and reduced leukocyte uptake and free radical formation during coronary artery occlusion [19]. Spencer and colleagues [22] recently showed that transition metals such as iron are present in the myocardium and can act as catalysts for the formation of oxygen-free radicals during reperfusion after myocardial ischemia. These data determine the rationale for deferoxamine mesylate use for myocardial protection.
In our previous investigation [23, 24] we demonstrated that increased vascularity results in increased function of the muscle. The LDM after subtotal mobilization had better fatigue resistance, stronger contractility force, and a shorter period of time for recovery after the fatigue test if the area occupied by capillaries is greater. For this investigation we used a different regimen of electrical stimulation for 2 months that resulted in considerable decreased muscle vascularity if 60 contractions per minute was applied and increased vascularity if 15 contractions per minute was used. In our future investigation we plan to study the influence of deferoxamine mesylate in fibrin sealant on contractile force of ischemic muscle, but we cannot speculate that the effect will be positive.
Study limitations
One of the limitations of this study is the animal model. It is not possible to study LDM function in this model. Additional studies are needed to address this issue. Another limitation of the study was that we did not measure the tissue levels of VEGF and FGF. However the concept of enhancing neovascularization and blood flow to limit surgically induced ischemia of LDM remains appealing.
Clinical relevance
Presently several methods to preserve LDM function such as vascular delay, different patterns of electrical stimulation, and novel methods of harvesting are used; however their efficacy is somewhat modest [24]. Therefore, pharmacological support of LDM to enhance blood flow and recovery from surgical trauma after mobilization to preserve muscle function may greatly improve results of various forms of cardiac bio-assist.
Conclusion
The iron chelator deferoxamine mesylate enhances neovascularization and blood flow and decreases or prevents surgically induced LDM ischemia in an experimental sheep. Human studies are warranted to investigate the clinical angiogenic potential of deferoxamine mesylate.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
I. Sheikh, G. Tchekanov, D. Krum, J. Hare, M. Djelmami-Hani, R. Maddikunta, M. E. Mortada, P. Karakozov, I. Baibekov, J. Hauck, et al. Effect of Electrical Stimulation on Arteriogenesis and Angiogenesis after Bilateral Femoral Artery Excision in the Rabbit Hind-Limb Ischemia Model Vascular and Endovascular Surgery, May 1, 2005; 39(3): 257 - 265. [Abstract] [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 |