Tissue engineered prefabricated vascularized flaps

ORIGINAL ARTICLE

TISSUE ENGINEERED PREFABRICATED VASCULARIZED FLAPS Kenneth Kian Kwan Oo, MBBS,1 Wei Chen Ong, MBBS,2 Annette Hui Chi Ang, MBBS,1 Dietmar W. Hutmacher, PHD,3 Luke Kim Siang Tan, MD1 1

Department of Otolaryngology, Head and Neck Surgery, National University of Singapore, Singapore, Singapore. E-mail: [email protected] 2 Division of Plastic Surgery, Department of Surgery, National University Hospital, Singapore, Singapore 3 Division of Bioengineering, Faculty of Engineering, National University of Singapore, Singapore, Singapore Accepted 10 August 2006 Published online 1 February 2007 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/hed.20546

Abstract: Background. Microvascular free tissue transfer has become increasingly popular in the reconstruction of head and neck defects, but it also has its disadvantages. Tissue engineering allows the generation of neo-tissue for implantation, but these tissues are often avascular. We propose to combine tissue-engineering techniques together with flap prefabrication techniques to generate a prefabricated vascularized soft tissue flap. Methods. Human dermal fibroblasts (HDFs) labeled with fluorescein diacetate were static seeded onto polylactic-co-glycolic acid-collagen (PLGA-c) mesh. Controls were plain PLGA-c mesh. The femoral artery and vein of the nude rat was ligated and used as a vascular carrier for the constructs. After 4 weeks of implantation, the constructs were assessed by gross morphology, routine histology, Masson trichrome, and cell viability determined by green fluorescence. Results. All the constructs maintained their initial shape and dimensions. Angiogenesis was evident in all the constructs with neo-capillary formation within the PLGA-c mesh seen. HDFs proliferated and filled the interyarn spaces of the PLGA-c mesh, while unseeded PLGA-c mesh remained relatively acellular. Cell tracer study indicated that the seeded HDFs remained viable and closely associated to remaining PLGA-c fibers. Collagen formation was more abundant in the constructs seeded with HDFs.

Correspondence to: K. K. K. Oo Winner of the Resident’s Research presentation at the Singapore Society of Otolaryngology Annual meeting, Singapore, November 2004. C V

2007 Wiley Periodicals, Inc.

458

Tissue Engineered Prefabricated Vascularised Flaps

Conclusions. PLGA-c, enveloped by a cell sheet composed of fibroblasts, can serve as a suitable scaffold for generation of a soft tissue flap. A ligated arteriovenous pedicle can serve as a vascular carrier for the generation of a tissue engineered vascularC 2007 Wiley Periodicals, Inc. Head Neck 29: 458– ized flap. V 464, 2007 Keywords: tissue engineering; prefabricated; vascularized; polylactic-co-glycolic acid (PLGA)

In

current clinical practice, there is a constant need for autologous tissue for reconstruction. The pectoralis major flap,1,2 radial forearm flap,3,4 and osteocutaneous fibula flap5 are some of the more commonly used flaps for head and neck reconstruction. The drive by reconstructive surgeons to provide the best functional and aesthetic outcome for their patients has made microvascular free tissue transfer the preferred reconstructive method.6 These flaps have superior versatility when compared with regional pedicled flaps, and their reliability7 makes them increasingly popular. Advances in microsurgical techniques have allowed a wider repertoire of autologous tissue transfer. In recent years, perforator flaps like the anterolateral thigh flap8 have gained popular-

HEAD & NECK—DOI 10.1002/hed

May 2007

ity. However, a major drawback of using these autologous tissues is donor site morbidity. A steep learning curve often limits its use, especially in the case of perforator flaps. Furthermore, the harvested flap often requires some surgical manipulation and adjustments in spatial positioning in order to obtain the desired shape to fit the defect. Prefabrication can address this problem by creating a 3-dimensional (3D) construct that matches the defect before it is implanted. Currently, in vitro engineered tissue that has been re-implanted in vivo is either avascular or thin enough to obtain sufficient nutrients by diffusion from surrounding vessels.9 Tissues that need to be transferred from 1 part of the body to another as 3-dimensional constructs with their own vascular supply need to be prefabricated by either wrapping vascularized soft tissue around the construct10 or placing a vascular pedicle within the tissue to be transferred11 so that angiogenic outgrowth may allow successful microvascular transfer subsequently.12 Therefore, a distinct alternative would be to use tissue engineering techniques consisting of a scaffold and mesenchymal stem cells, with a vascular supply. The generated prefabricated flap would address the many pressing problems that harvesting autologous tissue entails and exploit the potential of tissue engineering. Current techniques have shown that generating vascularized soft tissue is possible by either using an arteriovenous (AV) shunt loop or an arteriovenous bundle as a vascular carrier.13,14 Polylacticco-glycolic acid (PLGA) has been shown to be a suitable matrix for seeding human dermal fibroblasts (HDFs) in in vitro studies carried out in our laboratory.15 PLGA, in the form of Dermagraft (Advanced Tissue Sciences, La Jolla, CA), has also been used clinically to treat diabetic foot ulcers.16,17 Therefore, it is possible that by combining a ligated arteriovenous bundle with PLGA seeded with fibroblasts, a vascularized soft tissue flap can be produced. This experiment aimed to generate a prefabricated vascularized soft tissue flap in vivo by providing vascular supply through an arteriovenous bundle. Specifically, we want to evaluate PLGA, enveloped by a cell sheet composed of fibroblasts, as a suitable scaffold for a prefabricated flap and document angiogenesis and proliferation of seeded cells within the scaffolds. With this model, we hope to be able to tissue engineer a prefabricated vascularized flap without the use of intrinsic soft tissue. Eventually, larger and more complex prefabricated flaps may be generated with novel tissue engineering techniques.

Tissue Engineered Prefabricated Vascularised Flaps

MATERIALS AND METHODS

The PLGA-10/90 fiber was supplied by Shanghai Tianqing Biomaterials Company, Shanghai, China. The PLGA mesh was knitted from continuous fiber yarns of PLGA as described: 24 continuous PLGA single fibers of 20 lm diameter were combined to form a PLGA yarn, which was knitted into a mesh using a Silver Reed SK270 knitting machine. The mesh was folded and fused by heat along the edges to prevent fraying. The double-layered mesh was approximately 2 mm thick. They were stored in a dry cabinet at 17% relative humidity prior to use. Rat tail collagen was extracted from fresh animal carcasses and stored in lyophilized form until use. The collagen is reconstituted (1.3 mg/mL) in 0.05% acetic acid and polymerized onto the PLGA meshes by neutralizing with sodium bicarbonate (71.2 mg/mL) in the ratio of 100:9. PLGA-collagen (PLGA-c) meshes were left at room temperature for 30 minutes for complete polymerization, deep frozen at 808C, and freeze dried overnight. PLGA-c was stored in a dry cabinet at 17% relative humidity and sterilized under UV irradiation prior to use. The meshes were flexible and highly porous, with loosely bundled fibers not more than 100 lm apart. Interyarn spaces ranged from 300 to 800 lm wide. Pieces of mesh (8 3 10 mm2) were used for the study.

Scaffold Preparation.

Polycaprolactone (PCL) was purchased from Sigma-Aldrich. It is a semi-crystalline and biodegradable polymer belonging to a group of aliphatic ester. PCL film was first processed using standard solution casting method, which gave a thickness of approximately 80 to 100 lm. The film was then heat pressed and quenched in ice water and biaxially stretched in a temperature controlled environment, as described by Ng et al.18 This ultimately produced a relatively thin impervious membrane with a thickness of approximately 10 to 15 lm. The PCL film was cut into 1 cm 3 2 cm strips and sterilized in 70% ethanol for 1 hour prior to use.

Polycaprolactone Film Preparation.

HDFs were derived from enzymatic digestion of human skin samples with 2 mg/mL collagenase type I (Roche, Switzerland), overnight at 378C. Isolated cells were subsequently cultured in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, NY) supplemented with 10% fetal bovine serum (FBS, Hyclone, UT) and 1% penicillin-streptomycin solution (Gibco, NY) in culture flasks. For routine culture and experi-

Cell Culture and Seeding.

HEAD & NECK—DOI 10.1002/hed

May 2007

459

The experiment was carried out on 14- to 16-week-old nude rats weighing 200 to 250 g. The operative technique is as follows. Each animal is anesthetized and placed supine. Through a skin incision parallel to the inguinal ligament on one side, the femoral vessels are fully exposed and dissected from the inguinal ligament proximally to the origin of the inferior epigastric vessels. The femoral artery and vein were used as an arteriovenous bundle and were ligated proximal to the origin of the inferior epigastric vessels. The PLGA-c/HDF constructs were then used to sandwich the arteriovenous bundle. PCL film was then wrapped around this entire complex to isolate it from surrounding tissue to prevent vascular in-growth from the surrounding tissue (Figure 2), hence ensuring that the arteriovenous bundle remains as the only vascular supply. Entire complex was anchored to the upper thigh using 8/0 ethilon sutures. Control group had unseeded PLGA-c mesh sandwiched around the vascular pedicle as in the experimental group. Four weeks after implantation, the constructs were harvested for analysis.

In Vivo Implantation.

FIGURE 1. Polylactic-co-glycolic acid-collagen/human dermal fibroblast (PLGA-c/HDF) construct prior to implantation. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

ments, all cells were maintained in a self sterilizable incubator (WTB Binder, Tuttlingen, Germany) at 378C in 5% CO2, 95% air, and 99% relative humidity, with medium changed every 2 days. HDF at their 7th passage were used for the entire study. Cell viability assay using fluorescein diacetate (Molecular probes, OR) was used. Pelleted HDF at their 7th passage were resuspended in prewarmed PBS containing 5 lm of the probe. Cells were then incubated for 15 minutes at 378C and re-pelleted by centrifugation at 1000 rpm 3 10 minutes and resuspended in complete DMEM. As previously described,19,20 the cells were then counted with a hemocytometer and static seeded at 100,000 cells onto each mesh in 6-well plates. Additionally, HDFs were seeded at a density of 50,000 per cm2, over the entire well surface. The well plates were transferred into the incubator and left for 90 minutes to allow for cell attachment. Each well was then filled with 3 mL complete DMEM supplemented with 50 lg/mL L-ascorbic acid (Sigma, St. Louis). The samples were cultured for 10 days, with medium change done every 2 days. On the day of implantation, the confluent HDF sheets were mechanically peeled off using fine forceps and folded over the matrices to form a PLGA-c/HDF construct (Figure 1). Each construct formed half of the ‘‘sandwich’’ scaffold to be implanted. A total of 5 pairs of constructs were implanted. Three pairs were PLGA-c/HDF constructs, while 2 plain pairs of PLGA-c mesh served as controls.

460

Tissue Engineered Prefabricated Vascularised Flaps

Histology. For histology, specimens were fixed in 10% buffered formaline, dehydrated in ethanol, embedded in paraffin wax, and routinely sectioned and stained. Sections were stained with hematoxylin–eosin and Masson’s Trichrome to demonstrate collagen formation.

FIGURE 2. Polylactic-co-glycolic acid-collagen (PLGA-c) mesh ‘‘sandwiched’’ over femoral artery and vein. Polycaprolactone film envelopes the entire construct and edges secured with running ethilon suture. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

HEAD & NECK—DOI 10.1002/hed

May 2007

remained intact. All vascular pedicles remained pulsatile and bled upon truncation. Histology. The PLGA-c/HDF constructs showed a relatively homogeneous distribution of HDFs. Most of the PLGA-c fibers were surrounded by neotissue and had greater number of cells when compared with the controls. These cells had abundant cytoplasm when compared with the control scaffolds. PLGA-c fibers seeded with HDF appeared to undergo greater degradation when compared with controls (Figure 4A).

FIGURE 3. (A) Gross appearance of plain polylactic-co-glycolic acid-collagen (PLGA-c) generated flap. The construct appears thin and similar to its preimplantation dimensions. (B) Gross appearance of the human dermal fibroblast (HDF)/PLGA-c generated flap. It appears thicker and more opaque than the unseeded controls. Dimensions remained unchanged after 4 weeks of implantation. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] RESULTS

After 4 weeks, the implanted scaffolds were examined for external size, shape, and patency of the pedicle. All the implants showed no significant contraction and their square shape remained (Figures 3A and 3B). The PLGA-c/HDF constructs appeared thicker and more opaque when compared with the unseeded PLGA constructs. The PCL sheet surrounding the construct

Gross Morphology.

Tissue Engineered Prefabricated Vascularised Flaps

FIGURE 4. (A) Histology slide (hematoxylin-eosin [H&E]) of plain PLGA-c construct. There are fewer cells within the construct and the cells appear to have less cytoplasm than the seeded constructs. PLGA fibers can clearly be seen (original magnification 340). (B) Histology slide (H&E) of human dermal fibroblast (HDF)/PLGA-c construct. A much more cellular appearance is noted with greater degradation of PLGA fibers seen. There appears to be more cytoplasm within these cells as evident by the eosinophilc cytoplasm (original magnification 340). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

HEAD & NECK—DOI 10.1002/hed

May 2007

461

teriovenous bundle. No new capillaries were seen sprouting from the arterial side of the pedicle. Cell tracer studies showed that the labeled HDFs that were used to seed the PLGA-c scaffolds at the start of the experiment remained viable and were adjacent to the remaining scaffolds as evident by their fluorescence (Figures 7A and 7B).

DISCUSSION

Tissue engineering provides a prospect of generating tissue to replace or restore defects that result from surgical extirpation, disease, or congenital anomaly. It allows us to alleviate the problem associated with lack of suitable autologous tissue or donor site morbidity. Over the years we have evolved from just filling a defect to replacing like-for-like tissue, allowing for true functional reconstruction. With improvement in microsurgical techniques, prefabrication of flaps for use at a distant site has become a reality. Therefore, this experiment aims to combine the promise of tissue engineering together with flap prefabrication techniques to produce a tissue engineered prefabricated vascularized flap. This tissue engineered flap would require little intrinsic tissue and hence minimal donor site morbidity. The flap has the potential to

FIGURE 5. (A) Histology slide (Masson Trichrome) of plain polylactic-co-glycolic acid-collagen (PLGA-c) constructs showing less pronounced green staining by Masson Trichrome (original magnification 340). (B) Histology slide (Masson Trichrome) of human dermal fibroblast (HDF)/PLGA-c construct showing abundant collagen fibers stained green within the cellular field (original magnification 340). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

In contrast, plain PLGA-c constructs consisted mainly of PLGA-c fibers with little tissue formation or cells in between the fibers. The PLGA-c fibers remained largely intact with little degradation observed (Figure 4B). Masson Trichrome stains demonstrated a greater amount of collagen formation in the HDF seeded constructs when compared with the unseeded constructs (Figures 5A and 5B). Angiogenesis was also evident in both the control and PLGA-c/HDF constructs. Neocapillaries were seen sprouting from the lumen of the femoral vein (see Figure 6). These capillaries were noted within the interyarn spaces and neotissue of both the PLGA-c/HDF as well as the control constructs. New capillaries were also found away from the ar-

462

Tissue Engineered Prefabricated Vascularised Flaps

FIGURE 6. Histology slide (hematoxylin-eosin) showing luminal sprouting and capillary ingrowth (original magnification 340). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

HEAD & NECK—DOI 10.1002/hed

May 2007

FIGURE 7. (A) Histology image of cell seeded polylactic-co-glycolic acid-collagen (PLGA-c) mesh. PLGA-c fibers are the blue colored inclusions over the lower half of the image (original magnification 340). (B) Corresponding image of the above slide under fluorescence demonstrating viable human dermal fibroblasts (HDFs) that were originally labeled with FDA prior to implantation. The viable HDFs are adjacent to the PLGA-c fibers (original magnification 340). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

be transferred to a distant site through microsurgical techniques and can be designed to a predetermined shape and size to fit the defect. Prefabrication of a tissue engineered soft tissue flap requires a scaffold that is capable of supporting the development of dermal-like neotissue. The scaffold should also allow the homogenous proliferation of cells and yet be able to withstand cell contraction forces. Previous in vitro studies15 have demonstrated that PLGA has properties that allow relatively homogenous distribution and proliferation of cells within the interyarn spaces when compared with Alloderm. Furthermore, contraction and distortion of PLGA was minimal

Tissue Engineered Prefabricated Vascularised Flaps

when compared with TissuFleece. In this in vivo model, results after 4 weeks of implantation showed that the PLGA-c/HDF and plain PLGA-c constructs maintained their original shape and size regardless of whether they were seeded with HDF or not. Effects of capsular contraction were avoided possibly because of the duration of implantation (more extensive contraction might have been seen if implantation was for a longer duration), and the ‘‘splinting’’ effect of the PCL film that was used to isolate the construct from the surrounding tissue. This quality is crucial for prefabricated flaps as contraction and distortion would lead to loss of definition, scarring, and contracture. Scaffolds must therefore possess mechanical properties that would be able to resist contraction forces of both the seeded cells within the scaffold as well as cells surrounding the construct. One of the greatest advantages of using synthetic materials as scaffolds is the possibility of using various processing techniques to fabricate matrices of desired form, mechanical strength, and degradation profiles to achieve specific tissue engineering goals. Future studies should assess the performance of this construct in an immunocompetent model as capsular formation and inflammation may affect the results. The PLGA-c fibers served as suitable scaffold for the continued survival and proliferation of HDFs. These fibroblasts that proliferated within the construct were able to induce more collagen production than the unseeded PLGA-c constructs. The HDFs were also generally homogenously distributed within the construct with high cell counts per high-powered field when compared with unseeded scaffolds. Continued proliferation and survival of HDFs in PLGA-c scaffold makes it a suitable cell carrier and model for a resorbable scaffold that would be replaced by proliferating autologous cells. This generated construct may therefore serve as a dermal substitute. Current in vitro studies in our laboratory have seen some success in adding an epithelial layer over the dermal construct and further studies in an in vitro model will hopefully lead to the development of a more complex soft tissue flap. Tissue engineered constructs are often limited by their intrinsic lack of blood supply. Most constructs either rely on surrounding tissue for blood supply or are avascular. Recent work suggests that by combining a vascular supply, either in the form of a pedicle or an arteriovenous loop, neotissue can be generated. It is a natural extension, therefore, to try to combine this dermal substitute with a vas-

HEAD & NECK—DOI 10.1002/hed

May 2007

463

cular supply such that a potentially transferable tissue engineered prefabricated flap is produced. This experiment demonstrated luminal sprouting and vascularization of the constructs. This phenomenon was also present in the PLGA-c constructs that were unseeded. Hence, it shows that a ligated pedicle is a suitable model for providing a vascular supply to an otherwise avascular construct. More importantly, vascularization would allow for larger and more complex constructs, make these vascularized constructs ‘‘transferable’’ to a distant site, and enhance a construct’s viability by possessing its own blood supply. CONCLUSION

PLGA-c enveloped by a cell sheet composed of fibroblasts can serve as a suitable scaffold for generating a soft tissue flap. Documented angiogenesis and proliferation of seeded cells within this scaffolds show that a ligated arteriovenous pedicle can serve as a suitable vascular carrier. It is hoped that through such novel techniques, we may move closer to developing a prefabricated vascularized soft tissue flap for clinical use. Ultimately, with advancement in microsurgical techniques and further progress in tissue engineering, these tissue engineered prefabricated vascularized flaps may help overcome donor site morbidity, enhance reconstructive options, and potentially overcome technical difficulties in raising free flaps. Acknowledgments. The authors would like to acknowledge the help of Mr. K. W. Ng for the advice rendered during the course of the experiment and Mr. Anandkumar for help in data collection.

REFERENCES 1. Ariyan S. The pectoralis major myocutaneous flap. A versatile flap for reconstruction in the head and neck. Plast Reconstr Surg 1979;63:73–81. 2. Ariyan S. Further experiences with the pectoralis major myocutaneous flap for the immediate repair of defects from excisions of head and neck cancers. Plast Reconstr Surg 1979;64:605–612.

464

Tissue Engineered Prefabricated Vascularised Flaps

3. Soutar DS, McGregor IA. The radial forearm flap in intraoral reconstruction: the experience of 60 consecutive cases. Plast Reconstr Surg 1986;78:1–8. 4. Moscoso JF, Urken ML. Radial forearm flaps. Otolaryngol Clin North Am 1994;27:1119–1140. 5. Wei FC, Seah CS, Tsai YC, Liu SJ, Tsai MS. Fibula osteoseptocutaneous flap for reconstruction of composite mandibular defects. Plast Reconstr Surg 1994;93:294–304. 6. Chepeha DB, Teknos T. Microvascular free flaps in head and neck reconstruction. In: Bailey BJ, editor. Head and neck surgery-otolaryngology, 3rd ed. Philadelphia: Lippincott Williams & Wilkins; 2001. p 2045. 7. Schusterman MA, Horndeski G. Analysis of the morbidity associated with immediate microvascular reconstruction in head and neck cancer patients. Head Neck 1991; 13:51–55. 8. Wei FC, Jain V, Celik N, Chen HC, Chuang DCC, Lin CH. Have we found an ideal soft-tissue flap? An experience with 672 anterolateral thigh flaps. Plast Reconstr Surg 2002;109:2219–2226. 9. Haisch A, Groger A, Gebert C, et al. Creating artificial perichondrium by polymer complex membrane macroencapsulation: immune protection and stabilization of subcutaneously transplanted tissue-engineered cartilage. Eur Arch Otorhinolaryngol 2005;262:338–344. 10. Tan BK, Chen HC, He TM, Song IC. Flap prefabrication – the bridge between conventional flaps and tissue-engineered flaps. Ann Acad Med Singapore 2004;33:662–666. 11. Staudenmaier R, Hoang TN, Kleinsasser N, et al. Flap prefabrication and prelamination with tissue engineered cartilage. J Reconstr Microsurg 2004;20:555–564. 12. Hofer SP, Knight KM, Cooper-White JJ, et al. Increasing the volume of vascularised tissue formation in engineered constructs: an experimental study in rats. Plast Reconstr Surg 2003;111:1186–1192. 13. Mian R, Morrison WA, Hurley JV, et al. Formation of new tissue from an arteriovenous loop in the absence of added extracellular matrix. Tissue Eng 2000;6:595–603. 14. Tanaka Y, Sung KC, Tsutusmi A, Ohba S, Ueda K, Morrison WA. Tissue engineering skin flaps: which vascular carrier, arteriovenous shunt loop or arteriovenous bundle, has more potential for angiogenesis and tissue generation? Plast Reconstr Surg 2003;112:1636–1644. 15. Ng KW, Khor HL, Hutmacher DW. In vitro characterization of natural and synthetic dermal matrices cultured with human dermal fibroblasts. Biomaterials 2004;25:2807– 2818. 16. Gentzkow GD, Iwasaki SD, Hershon KS, et al. Use of dermagraft, a cultured human dermis, to treat diabetic foot ulcers. Diabetes Care 1996;4:350–354. 17. Naughton G, Mansbridge J, Gentzkow G. A metabolically active human dermal replacement for the treatment of diabetic foot ulcers. Artif Organs 1997;21:1203–1210. 18. Ng CS, Teoh SH, Chung TS, Hutmacher DW. Simultaneous biaxial drawing of poly(e-caprolactone) membranes. Polymer 2000;41:5855–5864. 19. Ng KW, Louis J, Ho BST, et al. Characterization of a novel bioactive poly(lactic-co-glycolic acid) and collagen hybrid matrix for dermal regeneration. Polymer International 2005;54:1449–1457. 20. Ng KW, Tham WR, Lim TC, Hutmacher DW. Assimilating cell sheets and hybrid matrices for dermal tissue engineering. J Biomed Mater Res Part A 2005;75:425–438.

HEAD & NECK—DOI 10.1002/hed

May 2007