Novel Treatment Strategy for Leg and Sternal Wound Complications After Coronary Artery Bypass Graft Surgery: Bioengineered Apligraf

Novel Treatment Strategy for Leg and Sternal Wound Complications After Coronary Artery Bypass Graft Surgery: Bioengineered Apligraf David E. Allie, MD, Chris J. Hebert, RT, RCIS, Mitchell D. Lirtzman, MD, Charles H. Wyatt, MD, V. Antoine Keller, MD, Stella M. Souther, RN, Adam A. Allie, MS, Elena V. Mitran, MD, and Craig M. Walker, MD Cardiovascular Institute of the South and Southwest Medical Center, Lafayette, Louisiana Purpose. To demonstrate that bioengineered Apligraf improves time to wound healing in sternal and leg wound complications after coronary artery bypass surgery. Description. Between 1998 and 2001, 1,550 patients underwent coronary artery bypass surgery utilizing saphenous vein. In 45 (2.9%) of 1,550 patients, leg wound complications developed (group A); and in 15 (0.9%) of 1,550 patients, sternal wound complications developed (group B). Apligraf was utilized as the primary treatment for 30 (66%) of 45 leg wounds and for 9 (60%) of 15 sternal wounds. Traditional wound care included debridement and daily wet-to-dry dressings. Evaluation. Time to wound healing ranged from 26 to 72 days (mean, 46) for Apligraf group A and from 34 to 120 days (mean, 84) for traditional wound care group A. The time to wound healing ranged from 21 to 80 days (mean, 39) for Apligraf group B, and from 36 to 110 days (mean, 62) for traditional care group B. Apligraf treatment was simpler, with less time and resource utilization than traditional wound care. Conclusions. Apligraf significantly improves time to wound healing in patients with leg and sternal wound complications and offers an attractive new treatment alternative to traditional wound care. (Ann Thorac Surg 2004;78:673 8) 2004 by The Society of Thoracic Surgeons Wound complications continue to substantially contribute to morbidity, mortality and costs after coronary artery bypass graft (CABG) surgery. Complications such as superficial and deep wound infections, wound dehiscence, cellulitis, lymphangitis, chronic edema, and delayed wound healing are reported in 1% to 24.3% of patients [1]. Sternal wound complications are reported at rates of 0.5% to 5.0% with deep chest infections and mediastinitis still reporting 7% to 40% mortality with significant morbidity [2]. A consensus treatment remains undefined. Apligraf (Novartis Pharmaceuticals Corporation, East Hanover, NJ) a new bioengineered, bilayered, living tissue skin equivalent, has shown statistical significance in improved wound healing of chronic venous leg ulcers and diabetic foot ulcers [3, 4]. Apligraf promotes wound healing by angiogenesis, increasing growth factors and cytokines, stimulating cell-to-cell vessel reactions, increasing matrix proteins, and providing a physical and Accepted for publication Sept 4, 2003. Address reprint requests to Dr Allie, Cardiothoracic and Vascular Surgery, 2730 Ambassador Caffery Pkwy, Lafayette, LA 70506; e-mail: david.allie@cardio.com. biological barrier against wound infection and desiccation [3 5]. Patients and Methods Between 1998 and 2001, 1,550 patients underwent CABG surgery utilizing greater saphenous vein. A continuous longitudinal incision was used in all greater saphenous vein harvests. Superficial and deep wound dehiscence with or without infection, occurred in 45 (2.9%) of 1,550 leg wounds (group A) and in 15 (0.9%) of 1,550 sternal wounds (group B). A retrospective multiple variable chart analysis forms the basis of this report. Data collection was performed under the Code of Federal Regulations (45 CFR 46.101[b] exemption no. 4 for retrospective studies and the HIPAA Privacy Rule). Retrospective data collection was approved by the Institutional Review Board. Apligraf was utilized as primary wound treatment in 30 (66%) of 45 leg wound complications and 9 (60%) of 15 sternal wound complications. The wound characteristics were similar for both groups. All wounds were evaluated weekly, and wound sizing was accomplished by comput- 2004 by The Society of Thoracic Surgeons 0003-4975/04/$30.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2003.09.032

674 ALLIE ET AL Ann Thorac Surg BIOENGINEERED APLIGRAF 2004;78:673 8 erized planimetry of surface acetate wound tracing and recorded by serial photographs. The time to wound healing was defined as time to complete epithelialization of the wound requiring no dressing changes. Early primary wound treatment consisted of sharp debridement of all devitalized tissue and appropriate systemic antibiotic treatment until the wound bed was free of active infection and demonstrated granulation. Traditional wound care consisted of a rigid protocol using a single home health care team, daily wet-to-dry saline gauze dressing changes, patient and family wound care education, and weekly physician wound evaluation. Minor local wound debridement was done in the office. The optimal Apligraf preparation and application technique has been previously discussed by Moneta and associates [6], and several points deserve mention. Optimizing wound bed preparation is of vital importance and should include control of local infection, aggressive necrotic tissue debridement, and treatment of edema until no exudation occurs. Apligraf looks, feels, and handles very much like an autologous split-thickness skin graft and requires attention to detail for optimal results [6]. Fenestrations are recommended to allow egress of fluid and air under the Apligraf that could interfere with wound bed adherence. Traditional split-thickness skin graft meshing techniques or scalpel slits at one slit per cm 2 of surface area are acceptable. The Apligraf should be removed from the tray by hand and placed on the wound bed within 30 minutes of opening the package, preferably within 10 minutes. The shiny dermal base should always be in direct contact with the wound bed and a 1.0 cm wound edge overlap has been shown to improve Apligraf adherence and stimulate edge wound epithelialization [6]. Sterile strips are recommended for optimal Apligraf wound anchoring. The standard dressing is a nonadherent primary dressing covered by sterile Vaseline gauze and an absorbent cotton wrap bolster. It is important to secure the dressing in place, but compression more than 30 mm is not recommended. The original dressing should be left in place for 1 week, then a weekly dressing change protocol is instituted. Achieving Apligraf wound bed adherence is tenuous for the first 2 weeks, but optimal results will occur if Apligraf adherence remains intact for 3 to 4 weeks. The Apligraf does not take in the same fashion as a split-thickness skin graft; therefore, the wound bed Apligraf appearance may vary greatly for the first several weeks. A foulsmelling greenish to yellowish gooey mass often occurs on the wound bed surface and must not be removed as this does not represent an infection but is indicative of Apligraf degradation and wound healing. After approximately 3 weeks, the mass will disappear and the wound bed dramatically reveals epithelization. Apligraf reapplication should not be considered for at least 6 weeks and should be considered only for the most recalcitrant wounds [5, 6]. Fig 1. Three-week postoperative sternal wound dehiscence. Patient 1 A 65-year-old diabetic man presented to the emergency room after a witnessed cardiac arrest and successful resuscitation. A cardiac catheterization was performed and an emergency left main coronary artery percutaneous transluminal balloon angioplasty (PTCA) and stent was placed for a critical 80% stenosis. The right coronary artery was nondominant. The patient recovered slowly, requiring a tracheostomy at 12 weeks. Recurrent chest pain at 12 weeks after PTCA required repeat cardiac catheterization, which revealed a 70% recurrent left main artery lesion with a complex dissection. Quadruple CABG surgery was performed with the left internal mammary artery and three greater saphenous veins used as conduits. At postoperative day 10, the superficial distal sternal wound drained and dehisced, requiring multiple wound debridements. The sternum remained stable. Bilateral pectoralis muscle flaps and a split-thickness skin graft were placed on postoperative day 21. The split-thickness skin graft failed, resulting ina6cm 4cm 2 cm chronic nonhealing superficial wound. After the wound bed was debrided of all devitalized tissue, active infection controlled, and early granulation appeared, an Apligraf was applied. The Apligraf remained adherent to the wound bed for 3 weeks, and the time to wound healing was 45 days after Apligraf application. The wound has remained healed, and the patient has fully recovered at 24 months after operation (Figs 1, 2, and 3). Patient 2 A 68-year-old diabetic woman presented with unstable angina after six prior PTCA procedures and stent placement. Cardiac catheterization revealed a 70% left main artery stenosis and a significant in-stent restenosis at four previous PTCA sites. Quadruple CABG surgery was performed with the left internal mammary artery, and three greater saphenous veins were used as conduits. Erythema and drainage of upper and lower left leg wounds were noted on postoperative day 5 with complete dehiscence of both sites occurring 48 hours later, leaving large gaping wounds. Multiple Apligrafs were applied to both leg wounds after wound bed preparation.

Ann Thorac Surg ALLIE ET AL 2004;78:673 8 BIOENGINEERED APLIGRAF 675 Fig 2. Apligraf application. Outpatient weekly dressing changes, wound evaluations, and wound care were instituted. At 6 weeks, approximately 50% wound healing had occurred and wound depth had significantly decreased. The patient received a second Apligraf application at 7 weeks. Total time to wound healing was 68 days for the upper leg wound and 72 days for the lower leg wound. The wounds have remained healed at 18 months. (Figs 4, 5, and 6). Results The preoperative clinical characteristics, demographics, and intraoperative patient characteristics were similar in both groups and are detailed in Table 1. The sternal and leg wound size and locations were similar in the Apligraf and traditional wound care groups with a trend toward larger wounds in the Apligraf group A(Tables 2 and 3). The time to wound healing was significantly less in Apligraf leg wound group A (26 to 72 days; mean, 46) as compared with traditional wound care group A (34 to 120 days; mean, 84). Likewise, the time to wound healing was decreased in Apligraf sternal wound group B (21 to 80 days; mean, 39) as compared with traditional wound care group B (36 to 110 days; mean, 62; Table 3). Fig 4. Two-week postoperative complete leg wound dehiscence. Comment Despite recent advances in surgical technique, antibiotic therapy, and adoption of less invasive surgical techniques, leg and sternal wound complications are reported as 1% to 24.3% after CABG [1, 2]. Endoscopic greater saphenous vein harvesting has been associated with fewer wound complications than the traditional longitudinal incision, but this less invasive technique has not been universally accepted. Endoscopic greater saphenous vein harvesting limitations include costs, increased harvest time, learning curve, and potential for greater saphenous vein trauma. Preoperative risk factors associated with wound complications have been well documented and include advanced age, female sex, diabetes, anemia, hypoalbuminemia, obesity, peripheral vascular disease, and corticosteroid therapy [1, 2]. Tissue edema, deconditioning, poor nutritional status, worsening anemia, and a relatively immunosuppressed state are common after CABG and are not conducive to the increased granulation and epithelialization demands of the compromised wound. The size and often ovoid shape of most superficial sternal and leg wounds are not conducive to primary wound approximation and therefore require healing by Fig 3. Sternal wound healed at 2 months. Fig 5. Apligraf application at 1 month.

676 ALLIE ET AL Ann Thorac Surg BIOENGINEERED APLIGRAF 2004;78:673 8 Table 2. Leg Wound and Wound Healing Characteristics (Group A) Apligraf n 30 Traditional n 15 Fig 6. Complete wound healing at 3 months. secondary intention. That requires an optimal granulation and epithelialization environment for new tissue formation and wound healing often a very slow and arduous process after CABG. Basic research has demonstrated that wound repair and closure and dermal regeneration can be both improved and hastened with newly developed bioengineered living skin products [7, 8]. The mechanisms involved in wound healing have been classically described and more recently reviewed [5]. The overlapping phases of wound healing include inflammation, tissue formation, and tissue remodeling. During inflammation, neutrophils, macrophages, and monocytes are recruited to the wound. Multiple growth factors and cytokines, including vascular and endothelial growth factor, transforming growth factor-beta and plateletderived growth factor, are then released, activating fibroblasts that promote the formation of granulation tissue and stimulate angiogenesis [7, 8]. The tissue formation phase, granulation, is characterized by the recruitment of keratinocytes, which produce a monolayer of epidermal cells that proliferate and differentiate forming a new stratified epidermis [5, 8]. The Leg (n 45) Location Upper leg 19 (63.3%) 10 (66.7%) Lower leg 9 (30%) 5 (33.3%) Both 2 (6.7%) 0 Wound Size Median wound area (cm 2 ) 4.3 (2.0 15.4) 3.1 (1.4 6.8) Small 4.0 cm 2 10 (33.3%) 11 (73.3%) Large 4.0 cm 2 20 (66.7%) 4 (26.7%) Time to wound healing (days) 46 (26 72) 84 (34 120) Outpatient treatment 28 (93.3%) 15 (100%) remodeling phase is characterized by collagen production, wound contraction, and increased collagen bundle cross-linking, which strengthens the new tissue. In biologically compromised environments, all three phases of wound healing are negatively affected, resulting in poor wound healing. Bioengineered substances are being developed as living skin tissue replacement to potentiate wound healing by increasing growth factors and cytokines, stimulating angiogenesis, and increasing matrix proteins all necessary for proper wound healing [5 8]. Apligraf is a bilayered living tissue engineered skin equivalent product grown in vitro from human neonatal foreskin. It contains a dermal layer containing fibroblasts anchored to a bovine collagen matrix and an epidermal layer containing live keratinocytes, which reproduce and form a stratified epidermis, including a welldifferentiated stratum corneum [5 8]. Apligraf has shown no evidence of rejection or sensitization to human construct antigens [6]. Apligraf does not contain melanocytes, blood cells, hair follicles, sweat glands, blood vessels, or Langerhan s cell, thus explaining the lack of rejection [7, 8]. Table 1. Preoperative and Intraoperative Patient Characteristics Group A Leg (n 45) Group B Sternal (n 15) n 30 n 15 n 9 n 6 Age (y) 66 8.3 68 7.9 68.9 66 6.5 Sex (male:female) 20:10 9:6 6:3 4:2 Obesity (%) 10 (33) 4 (26) 3 (33) 2 (33) Urgent operation (%) 1 (3) 1 (6) 1 (11) 1 (17) Congestive heart failure (%) 5 (16) 3 (20) 2 (22) 2 (33) Chronic obstructive pulmonary disease (%) 8 (27) 4 (26) 3 (33) 2 (33) Diabetes mellitus (%) 10 (33) 5 (33) 3 (33) 2 (33) Hypertension (%) 17 (57) 10 (67) 6 (67) 4 (67) Anemia (%) 5 (16) 2 (13) 2 (22) 2 (33) Hypoalbuminemia (%) 6 (20) 4 (27) 3 (33) 2 (33) Peripheral vascular disease (%) 14 (46) 8 (53) 6 (67) 4 (67) Cardiopulmonary bypass time (min) 97 21 94 20 99 25 94 27 Number of vein grafts 3.2 1.0 3.1 1.0 2.9 1.0 3.1 1.0

Ann Thorac Surg ALLIE ET AL 2004;78:673 8 BIOENGINEERED APLIGRAF 677 Table 3. Sternal Wound and Wound Healing Characteristics (Group B) Apligraf n 9 Traditional n 6 Sternal (n 15) Wound size Median wound area (cm 2 ) 3.1 (1.6 4.2) 2.4 (1.5 3.4) Time to wound healing (days) 39 (21 80) 62 (36 110) Outpatient treatment 7 (77.8%) 6 (100%) Apligraf accelerates wound healing by multiple mechanisms including increasing multiple growth factor and cytokines, promoting angiogenesis, increasing important matrix proteins vital to wound healing, and providing a physical and biological barrier against wound infection and desiccation [7, 8]. Fibronectin, tenascin, procollagen, and keratin 16 are extracellular matrix proteins vital to all phases of wound healing. Immunohistologic staining has shown markedly elevated levels of these matrix proteins produced by the living Apligraf fibroblasts as opposed to normal skin fibroblasts [7, 8]. Defensins are a group of proteins exhibiting a broad spectrum of antimicrobial activities against bacteria, yeasts, and fungi and have been isolated from macrophages, epithelial cells, and neutrophil granules. Human beta defensin-2 has been found to be highly effective in killing yeasts and both gram-negative and gram-positive bacteria [7, 8]. Human beta defensin-2 is not detectable in normal skin but is produced in high levels by living keratinocytes in Apligraf, providing protection from infection by opportunistic microorganisms [9]. Apligraf is currently approved for treating diabetic foot ulcers and for treating partial-thickness and fullthickness skin loss due to venous leg ulcers. Apligraf has been reported effective in treating other chronic and acute wounds including epidermolysis bullosa, thermal injuries, pressure ulcers, surgical excision sites, donor site wounds, and keloids [4, 5]. These clinical reports, the described wound healing attributes of Apligraf, and our experience in treating venous leg ulcers and diabetic foot ulcers led to our adoption of Apligraf as our primary wound treatment for patients with leg and sternal wound complications after CABG. Apligraf substantially decreased the time to wound healing in our patients with both leg wound complications (mean, 84 versus 39 days) and sternal wound complications (mean, 39 versus 62 days). There were no Apligraf-associated wound infections. Another important clinical advantage of Apligraf was improved pain control. The exact mechanism of enhanced pain control remains undefined and may reflect fewer dressing changes or early wound healing. The surgical technique of Apligraf application is relatively simple to perform, is conducive to outpatient treatment, and can be an office procedure. The onceweekly dressing change protocol for the first 3 weeks has obvious potential advantages that include simplicity; improved patient satisfaction; less pain, health care resource utilization, and time; and lower cost. There are several limitations of this study: the nonrandomized retrospective methodology of the study carries the bias inherent in a retrospective analysis; the relatively small sample size, the short follow-up, and the objective evaluation of wound healing preclude definitive recommendations. Biases were minimized by the use of computerized planimetry of surface acetate wound tracings for wound sizing and serial photographs, both of which are well-accepted objective measurements of wound care and wound healing. A detailed cost analysis was not performed, but recent reports have shown improved cost analysis with Apligraf treatment of diabetic foot ulcers [10] and venous leg ulcers [11]; therefore, a cost reduction seems feasible with post-cabg wound healing complications. Lower extremity peripheral vascular disease and diabetes are common among CABG patients and occurred in more than 60% of our leg wound healing complications (group A). It is recommended that an ankle-brachial index be performed on all patients with leg wound healing complications, and that clinicians have a low threshold for performing angiography on any patient with delayed wound healing. Four (9%) of 45 of patients in group A required angiography and a percutaneous iliofemoral revascularization to attain total wound healing. A policy to avoid incisions below the midcalf has been adopted for patients with preoperative peripheral vascular disease, diminished pedal pulses, obesity, or diabetes. In conclusion, Apligraf significantly decreases time to wound healing in patients with post-cabg sternal and leg wound complications. Apligraf treatment offers other potential advantages as compared with traditional wound care: these include simplicity, less pain, improved patient satisfaction, and more efficient use of health care resources. A larger, randomized, prospective multicenter trial is warranted. Disclosures and Freedom of Investigation The source of funds used to perform the evaluation was Novartis. The new method of treatment is our original contribution. The Apligraf was purchased. The authors had full control of the design of the study, methods used, outcome indicators, analysis of data, and production of the written report. References 1. L Ecuyer PB, Murphy D, Little JR, Frayser VJ. The epidemiology of chest and leg wound infections following cardiothoracic surgery. Clin Infect Dis 1996;22:424 9. 2. El Oakley RM, Wright JE. Postoperative mediastinitis: classification and management. Ann Thorac Surg 1996;61: 1030 6. 3. Atillasoy E. The safety and efficacy of Apligraf (Apligraf) in the treatment of venous leg ulcers: a multicenter, randomized, clinical trial. Wounds 2000;12(Suppl A):20 6. 4. Falanga V, Margolis D, Alvarez O, et al. Rapid healing of venous ulcers and lack of clinical rejection with an allogenic cultured human skin equivalent. Arch Dermatol 1998;134: 293 300.

678 ALLIE ET AL Ann Thorac Surg BIOENGINEERED APLIGRAF 2004;78:673 8 5. Sabolinski ML, Alvarez O, Auletta M, et al. Cultured skin as a smart material for healing wounds: expereince in venous ulcers. Biomaterials 1996;17:311 20. 6. Moneta G. Graftskin (Apligraf) in the treatment of venous leg ulcer: optimal application techniques. Wounds 2000; 12(Suppl A):27 32. 7. Falanga V. Apligraf treatment of venous ulcers and other chronic wounds. J Dermatol 1998;25:812 7. 8. Milstone L, Asgari M, Schwartz P, Hardin-Young J. Growth factor expression, healing, and structural characteristics of Graftskin (Apligraf). Wounds 2000;12(Suppl A):12 9. 9. Harder J, Bartels J, Christophers E, Schröder JM. A peptide antibiotic from the human skin. Nature 1997;387:861. 10. Curran MP, Plosker GL. Bilayered bioengineered skin substitute (Apligraf). A review of its use in the treatment of INVITED COMMENTARY I very much enjoyed reading about this novel therapy for wounds after cardiac surgery. The bioengineered Apligraf seems to work very well in this retrospective study. Certainly the accompanying pictures are most impressive. Usually cardiac surgeons do not like to talk about a wound complication. We try to pretend they do not occur and do our best to relegate care to the plastic surgeons. Unfortunately, they are a major cause of morbidity and mortality and are really the bane of the existence of the cardiac surgeon. This novel therapy is most interesting. According to Allie and colleagues this is a living tissue equivalent that allegedly promotes wound healing by increasing vascularity and potentially increasing growth factors. Allie is venous leg ulcers and diabetic foot ulcers. Biodrugs 2002;16: 439 55. 11. Kirsner RS, Fastenau J, Falabella A, Valencia I, Long R, Eaglestein WH. Clinical and economic outcomes with Graftskin for hard-to-heal venous leg ulcers. A single-center experience. Dermatol Surg 2002;28:81 2. Disclaimer The Society of Thoracic Surgeons, the Southern Thoracic Surgical Association, and The Annals of Thoracic Surgery neither endorse nor discourage use of the new technology described in this article. also most impressive in his retrospective study. Clearly, a cost analysis should be done and obviously there is room for a randomized study here. However, this is novel therapy and we would certainly be most interested in trying this at our institution. Irving L. Kron, MD TCV Surgery UVA Medical Center Lee St, Room 2753 Box 800679 Charlottesville, VA 22908 e-mail: ikron@virginia.edu 2004 by The Society of Thoracic Surgeons 0003-4975/04/$30.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2003.09.031