Abstracts-Part II

Annual Meeting, Koloa, Kauai, Hawaii

January 11-14, 2003

Analysis of Flow Hemodynamics and Vascular Resistance in a Distal Free Flap in a Rat Model. Vishal Sarwahi, Yelena Akelina, Ulugbek Ulugbek, Robert J. Strauch, and Melvin P. Rosenwasser. Columbia University Medical Center, New York, NY.

The objective of this study was to create an animal model and to study the hemodynamics and effects of increased vascular resistance in a long vein bypass graft with a flap.

Sixteen Wistar rats were used. A tail vein graft 7.5 cm long was harvested. A 2-×4-cm left groin flap was elevated pedicled on the femoral artery and maintaining its original venous outflow. The graft was anatomosed end-to-side with the right femoral artery, and end-to-end with the left femoral artery proximal to the flap pedicle. Thus, the right femoral artery provided the inflow to the right leg and through the graft to the left groin flap and the leg. The venous outflow was left undisturbed. Blood flow was measured immediately and 5 days later, using a Doppler microprobe. On day 5, the left femoral artery was clamped distal to the flap pedicle to increase the resistance.

One rat died and one had an early thrombosis. The remaining flaps survived, and all grafts were patent. Mean flow velocities were reported. The velocity and volume did not differ at the proximal or distal end of the graft. The mean velocity through the vein graft was significantly slower than in the femoral arteries. On day 5, the velocity in the left femoral artery was similar to that in the proximal right femoral artery, but was more than in the distal right femoral artery. Thrombosis was seen in all grafts within 10 min of clamping the left femoral artery distal to the pedicle.

Many diabetic patients have a single patent vessel at the ankle level. Limb salvage in these patients is usually attempted by a proximal end-to-end anastomosis, using a long vein graft. Failure of free flaps is not uncommon in distal anastomoses in the presence of poor collaterals. All the vessels in this study thrombosed due to increased resistance, showing that end-to-end anastomoses with high inflow can lead to graft failure. Conversely, an end-to-side anastomosis has been shown to increase the distal outflow bed, decreasing the resistance and increasing graft flow and survival.

Radial Artery Flow-Through (RAFT) Flap: A New Bypass Conduit with a Modulating Arteriovenous Fistula. Jin Chun, Lester Silver, Victoria J. Teodorescu, Michael L. Marin, and Michael Christy. Mount Sinai Medical Center, New York, NY.

The distal arterial bypass for an ischemic lower extremity with inadequate distal target vessels can be a technical challenge. High outflow resistance, combined with the difficulty of small vessel anastomosis, may lead to bypass graft failure.

The authors have used a radial artery flow-through (RAFT) flap to facilitate these difficult bypasses in 10 patients between November, 1999 and January, 2002. All patients had limb-threatening ischemia. All but one had non-healing ulcers. Seven had diabetes mellitus. Five had previously undergone a conventional bypass.

The bypass patency rate was 80%, with an overall limb salvage rate of 80%. There were no cardiac events or hand ischemia following the RAFT flap harvest in this small series. The authors hypothesized that the increased bypass patency was due primarily to the built-in modulating arteriovenous fistula of the flow-through flap, which maintains a high flow through the bypass graft with high distal outflow resistance. This hypothesis was tested in the rat epigastric flow-through flap model, demonstrating a four-fold increase of epigastric blood flow over a 3-hr period on tying off the outflow branches. The flow-through flap bypass was patent 70% of the time, compared to 0% for the control arterial interpositional graft in this model.

The authors believe that the RAFT flap bypass presents a new option in limb salvage procedures for peripheral arterial occlusive disease, especially when there are concomitant non-healing ulcers. They presented their experience and the complications encountered with this technique.

Preferential Use of the Slit Arteriotomy for End-to-Side Arterial Anastomoses in Free-Tissue Transfers to the Extremities. Bien-Keem Tan, Winston Chew, Soo-Wan Hong, and Kok-Chai Tan. Singapore General Hospital and Tan Tock Seng Hospital, Singapore.

The value of end-to-side arterial anastomoses for free-flap transfers to the extremities is well-established. However, it is still unclear whether a slit could be used in preference to the conventional elliptical arteriotomy. The slit arteriotomy is simpler to create and does not require excision of the vessel wall. The purpose of this clinical study was to demonstrate the reliability of the slit arteriotomy for end-to-side arterial anastomoses.

Between 1999 and 2002, 49 free flaps for resurfacing the upper and lower extremities were revascularized using this technique. Flaps used included the gracilis (n=18), latissimus dorsi (n=14), rectus abdominis (n=6), lateral arm (n=4), fibula/peroneal (n=3), radial forearm (n=3), and anterolateral thigh (n=1).

An end-to-side anastomosis with a 45-degree incident angle is created using an interrupted suture technique. A longitudinal slit arteriotomy is made with a 30-degree microknife. The salient features of the technique are: 1) use of a double microclamp to maintain slackness of the arterial segment bearing the slit. By approximating the clamps, the longitudinal slit is converted into an elliptical opening for better visualization. 2) Beginning at the heel, sutures are placed in such a way as to feed the redundancy of the donor artery wall toward the toe end of the anastomosis. The slit can be progressively lengthened to match the diameter of the expanded donor vessel end. 3) A slightly oversized slit is used to maintain the donor vessel on a stretch. Elastic recoil of the donor vessel will keep the slit patent. Following tourniquet release, the slit becomes an elliptical opening as blood flow resumes.

There was one arterial failure attributed to kinking of the end-to-side anastomosis (98% arterial patency rate). The overall success rate was 94% (one arterial failure and two venous failures).

The slit arteriotomy is simple, reliable, and can be used consistently for free-tissue transfers to the extremities. The anastomotic diameter is increased, as no portion of the recipient vessel wall is excised.

Endoscopic-Assisted Microsurgery and Realistic Expectations-Early Clinical Experience. Mohamed M. El-Shazly, Wei Chao Huang, Ming-Huei Cheng, and Hung-Chi Chen. Assiut University Hospital, Assiut, Egypt, and Chang-Gung Memorial Hospital, Taipei, Taiwan.

For successful microsurgical performance, the microsurgeon needs to see the field magnified with a good degree of resolution. It would also be preferable if optimal physical and visual comfort for the microsurgeon and the operating team could be achieved. These two requirements can be fulfilled with the aid of the endoscope. After a previous successful experimental application of the endoscope as a visual assistant in microsurgery, the question this reported study sought to answer was: can microvascular surgery be performed in clinical situations with the use of the endoscope? An initial experience in 10 cases was presented.

An endoscopic unit was used as a visual assistant in the operative field for microvascular anastomoses of the vascular pedicles of 10 free flaps (DIEP and ALT perforator flaps) to their corresponding recipient vessels (internal mammary and facial vessels), for breast and head and neck reconstructions, respectively. Technical aspects of concern were recorded, including handling of the instruments, eye-to-hand coordination, degree of magnification, working distance, diameter of the operative field, resolution and quality of the two-dimensional image, and the surgeon's physical condition.

The technical and clinical data from the 20 vascular anastomoses of the 10 flaps were collected, assessed, and compared with conventional use of the operating microscope. All microsurgical procedures were performed with maximum comfort, regarding dissection, clamp application, vascular end preparation, irrigation, and the microvascular anastomoses themselves. The surgeon's physical status was an excellent visual and physical comfort, compared to using the microscope. Handling of the instruments, adjustment of the zoom and focus, were done without any recorded difficulty. The endoscopic image, magnification, working distance, range of movement, and operative field were all very suitable.

The endoscope appears to be a promising tool in performing microsurgery with improved technical aspects. These authors attempted to establish a schematic algorithm for future routine use of an endoscopic microsurgical unit. They presented a concept of endoscopic-assisted microsurgery, which might be of interest to new generations of microsurgeons and in special operative situations.

Donor-Specific Tolerance and Chimerism in Limb Allografts. Maria Siemionow, Kagan Ozer, Maciej Zielinski, and Dariusz Izycki. The Cleveland Clinic Foundation, Cleveland, OH.

Induction of donor-specific tolerance in composite tissue allografts (CTAs) is the ultimate goal in transplantation. Recently, the authors induced donor-specific tolerance to rat hind-limb allografts across an MHC barrier in a 35-day course of cyclosporine A (CsA) and ab T-cell receptor (TCR) mAB therapy. In this report, they presented the efficacy of a shorter immunomodulating protocol and its effect on chimerism and tolerance induction.

Thirty-six hind-limb transplantations across an MHC barrier [Lewis-Brown-Norway (LBN, RT1l-n) to Lewis (LEW, RT1l)] were tested for potential to induce tolerance with 35, 21, 7, and 5-day protocols of combined CsA and ab TCR mAB therapy. There was no chronic immunosuppression once the protocol was stopped. Experimental groups included transplants under four different time/dose regimens and with monotherapy controls of CsA and TCR alone. The efficacy of immunomodulating therapy was measured by flow cytometry (FACS) levels of CD3, CD4, CD8, CD90, NK, and TCR. Clinical tolerance was tested by skin grafting from the donor (LBN), recipient (LEW), and third party (ACl, RT1a). Donor-specific chimerism was evaluated in the peripheral blood lymphocytes (PBL) and lymphoid organs by flow cytometry (9FC). The expression of RT1l relative to the MHC class 1 antigen of the LBN donor's CD4 and CD8 cells was assessed.

Without immunosuppression, limbs rejected at day 7. Monotherapy with CsA or ab-TCR prolonged survival up to 21 and 13 days, respectively (p<0.05). All transplants (n=10) under 35 days of CsA/TCR protocols survived rejection-free over 700 days (p<0.0001). Limb allografts under short protocols of 5, 7, and 14 days were rejection-free and still under evaluation at 110, 125, and 150 days, respectively. Clinical tolerance was confirmed by acceptance of skin grafts from the CTA donors (LBN). This correlated with PBL chimerism of 12.6% of CD4+ and 6.8% of CD8+ chimeric RT1n+ T-cell subpopulations. Evaluation of lymphoid organs confirmed higher levels of chimerism in splenic tissue (62.8% total, 11.8% being RT1n+/CD4+) and in para-aortic lymph nodes (14.7% total, 1.3% being RT1n+/CD4+), but not in thymic tissue.

To the best of the authors' knowledge, this is the first report documenting induction of tolerance across the MHC barrier in CTA transplants submitted to short protocols of immunomodulatory therapy without a need for chronic immunosuppression. The mechanism of tolerance induction seems to be directly associated with the development of donor-specific chimerism in peripheral blood and lymphoid organs. This protocol may have direct impact on management of clinical composite tissue allografts.

Assessment of Rejection and Graft-Versus-Host Disease after Hand Transplantation. Vijay S. Gorantla, Nestor R Gonzalez, Diaane J. Pidwell, Linda C. Cendales, Carolyn Burns, Allan D. Kirk, Darla K. Granger, and Warren C. Breidenbach. Christine M. Kleinert Institute, Jewish Hospital, Louisville, KY, National Institutes of Health, Bethesda, MD, and University of Louisville, Louisville, KY.

Much of the knowledge regarding composite tissue allografts (CTAs) is derived from research in animal-limb transplant models. The authors have demonstrated that CTAs like the rat hind-limb consist of multiple tissues, some of which (skin, muscle) are highly immunogenic and trigger rapid rejection, and others (lymph nodes and bone marrow) can initiate graft-vs.-host disease (GVHD) in the host. The highly successful results of the initial trial of human hand transplantation have allowed the differentiation of rejection and GVHD in transplanted hands, both clinically and by histopathology. In this study, the authors compared clinical signs and skin biopsies from the two Louisville recipients (at 3 and 1 year post- transplantation, respectively), with those from patients with GVHD following bone marrow allotransplantation. The aim was to determine if the clinical or histopathologic changes of rejection were similar to those of GVHD.

Clinically, early rejection manifested as diffuse or patchy erythema localized to the allograft in both patients. Episodes of mild acute rejection resolved promptly and completely with topical medications, without the need to increase systemic immunosuppression. No clinical evidence of GVHD was observed in both patients. Biopsies were taken from the site of intense dermatoerythema during episodes of acute rejection and from the distal forearm during the rejection-free period. Clinically, evident rejection was associated with a perivascular and periadnexal lymphocytic infiltrate in the superficial dermis. This was correlated with relatively little exocytosis of lymphocytes. There was no dyskeratosis of epidermis or hair follicle epithelium, and no apoptosis of keratinocytes. A few biopsies showed non-leucocytoclastic vasculitic changes in the small dermal arterioles with some destruction of basal epidermis. Some biopsies obtained more than 2 years after surgery showed epidermal atrophy with collagen sclerosis in the superficial dermis.

Taken together, these findings at this early stage indicate that rejection in CTAs like hand transplants is directed more at small vessels than at epithelial structures (epidermis and adnexa). In comparison, GVHD in bone marrow transplant patients is associated with lesser lymphocytic infiltration in this distribution and greater direct involvement of the overlying epidermis. To date, there have been no reports of acute GVHD in any of the hand transplants performed around the world. Further follow-up of hand transplants by clinical, histopathologic, and immunohistochemical studies will enable the establishment of criteria for assessment of acute and chronic rejection or GVHD in clinical CTAs like hand transplants.

A New Composite Tissue Transplantation Model in Mice. Kagan Ozer, Maciej Zielinski, and Maria Siemionow. The Cleveland Clinic Foundation, Cleveland, OH.

Initial manifestation of composite tissue allograft (CTA) rejection is reflected at the microcirculatory level. Intravital microscopy provides invaluable information regarding the mechanism of allograft rejection and the outcome of therapeutic modalities. With the development of a variety of inbred, recombinant, and mutant mice strains, a greater biologic understanding of allograft rejection is possible; however, the implementation of a mouse model in CTA transplantation studies is challenging, due to the technical difficulties. The purpose of this study was to develop a CTA transplantation model in mice, combining the advantages of intravital microscopy with the access to transgenic mice technology.

The mice cremaster muscle transplantation model was developed and tested in the following experimental groups. Isosgraft control transplantations (n=6) were performed between C57BL/6N mice. In the allograft (n=5), transplantations were performed across fully mismatched strains between C3H and C57BL/6N mice. The donor cremaster muscle allografts were dissected as a free flap, harvested on the common iliac artery and vein, and anastomosed to the recipient's ipsilateral carotid artery and external jugular vein, using standard end-to-end technique (11-0 Ethilon under ×40 magnification). After clamp release, the cremaster muscle flap was prepared for direct in vivo observation of the microcirculation. The following measurements were taken during the first and second hour after transplantation: red blood cell velocities (RBC) and vessel diameters of the first, second, and third order arterioles; functional capillary perfusion (FCP); the number of rolling, adhering, and transmigrating polymorphonuclear leukocytes and lymphocytes in the postcapillary venule; and endothelial edema index.

The average time for cremaster transplantation was 2 hr and the average ischemia time was 75 min, with a 95% success rate. RBC velocity and vessel diameters showed a normal pattern during the first 2 hr without significant differences between the groups. However, leukocyte-endothelial interactions were activated in the allograft group. The number of rolling (70.3±14), adhering (44.2±2.9) and transmigrating (7±1.5) leukocytes was significantly increased, when compared to the isograft control group (rolling: 14.8±2.2, adhering: 4.2±0.4, transmigrating: 1.96±1.3, p<0.001). Significant activation of the rolling and adhering lymphocytes was also observed in the allograft group (rolling: 5.6±0.6, adhering: 4.6±0.9), when compared to the isograft controls (rolling: 0.63±0.1, adhering: 0.66±0.3, p<0.05).

This study validated the feasibility of transplanting the cremaster muscle flap in mice, with a 95% success rate. To the authors' knowledge, this is the first report on a composite tissue transplantation model, which offers the possibility of combining transgenic mice technology with direct in vivo evaluation of transplant hemodynmamics via an intravital microscopy system.

Nitric Oxide Synthases Gene Expression in the Late Phase of Ischemic or Adenosine Preconditioning in Skeletal Muscle. Wei Z. Wang, Xin-Hua Fang, Linda L. Stepheson, Kayvan T. Khiabani, and William A. Zamboni. University of Nevada School of Medicine, Las Vegas, NV.

Previously, these authors have reported that the adenosine (ADO)-initiated microvascular protection in the late phase of ischemic preconditioning (IPC) can be blocked by an inhibitor of nitric oxide synthase. The purpose of the present study was to determine eNOS and iNOS gene expression and microvascular response in the cremaster muscle 24 hr after ischemic- or ADO-preconditioning.

The vascular isolated cremaster model in male SD rats was used. IPC consisted of 45 min of ischemia on day 1 and then 24 hr of reperfusion. To mimic the effects of IPC in the late phase, ADO (0.35 mg/100 g) was given via local intraarterial infusion at a rate of 0.1 ml/hr for 15 min. To block the effects of IPC in the late phase, 8-sulfophenyo-theophylline (SPT, a nonspecific ADO receptor blocker, 0.625 mg/100 g) was given in the same way prior to IPC.

Two parts of the study have been carried out. In the microcirculation aspect, cremaster muscles underwent 4 hr of warm ischemia and then 60 min of reperfusion on day 2. Microvascular responses in the cremaster muscle to IPC or pharmacological preconditioning were determined by measuring terminal arteriole diameter and capillary perfusion, using intravital microscopy and by the evaluation of the endothelium-dependent and -independent vasodilatation in the terminal arterioles. In the molecular biology aspect, the cremaster muscles were harvested on day 2 for measuring eNOS and iNOS gene expression by semi-quantitative reverse transcriptase polymerase chain reaction (QRT-PCR). Four groups of animals (12 rats/group) were used: Group 1 - sham IPC, Group 2 - IPC, Group 3 - ADO+sham IPC, Group 4 - SPT+IPC.

The average diameter of the terminal arterioles after 4 hr of ischemia/reperfusion was significantly larger in the IPC (121% of baseline) or ADO groups (113%) than in the sham IPC (71%) or SPT groups (79%) (p<0.05). The average capillary perfusion was also better in the IPC (96% of baseline) and ADO (105%) groups than in the sham IPC (54%) or SPT groups (65%) (p<0.05). Endothelium-dependent vasodilatation induced by acetylcholine (10-7M) was shown intact in the IPC or ADO pre-treated cremasters, but impaired in the sham IPC or SPT pre-treated cremasters. IPC or ADO preconditioning on day 1 significantly enhanced eNOS mRNA expression (30% or 15% increase from controls, respectively) and iNOS mRNA expression (15% or 20% increase from controls, respectively) on day 2 (p<0.05).

The results of these studies suggested that ischemic or ADO preconditioning induces late-phase microvascular protection in the skeletal muscle by a NOS-dependent mechanism.

Pre-Ischemic Application of the NO Donor Spermine/Nitric Oxide Complex Enhances Flap Survival in a Rat Model. Markus V. Kuentscher, Sebastian Juran, Detlev Erdmann, and Guenter Germann. University of Heidelberg, BG Trauma Center Ludwigshafen, Ludwigshafen, Germany.

Nitric oxide (NO) is reported to antagonize endotheline, resulting in protection from ischemia-reperfusion injury. Additionally, NO seems to play an important role in the phenomenon of ischemic preconditioning (IP). Previous studies regarding the role of NO in flap ischemia have focused on late IP, using sodium nitroprusside, or the role of L-arginine as a precursor of NO, and a substrate of the endothelial NO synthase. The purpose of this reported study was to determine whether flap survival could be improved by pre- or post-ischemic intravenous application of the new NO donor, spermine/nitric oxide complex (Sper/NO) with a long half life (39 min in 37 degrees C phosphate buffer pH 7.4), and a controlled biologic release of NO in vivo.

Thirty-seven male Wistar rats were divided into four experimental groups. An extended epigastric adipocutanaeous flap (6×10 cm) was raised in each animal. In one of the control groups (C, n=10), a 3-hr flap ischemia was induced by clamping the pedicle. Another group served as a non-ischemic control (CO, n=8). The animals in Group S1 (n=10) received 500 nmol/kg Sper/ NO intravenously 30 min prior to ischemia. In Group S2 (n=9), an equivalent dose of Sper/NO was administered 5 min prior to reperfusion. Flap ischemia was induced in Groups S1 and S2 as in control Group C. The mean flap necrosis area was assessed for all groups on the fifth postoperative day, using planimetry software.

The average flap necrosis area was 68.2% (±18.1%) in control Group C, and 29.7% (±13.3%) in the non-ischemic controls (CO). The group with pre-ischemic application of Sper/NO (S1) demonstrated an average flap necrosis of 11.2% (±5.9%), while it was 59.2% (±14.4%) in Group S2. Group S1 showed a significantly lower flap necrosis area than the Groups C, CO, and S2 (p<0.05). Group S2 demonstrated a significantly higher mean flap necrosis area than the non-ischemic controls (p<0.05). However, no statistically significant difference could be shown between Groups S2 and C.

The data demonstrated that pharmacologic ischemic preconditioning and enhancement of flap survival could be achieved by intravenous administration of Sper/NO. The application of Sper/NO at the end of the ischemic period does not provide protection from ischemia. The data suggested that acute pharmacologic ischemic preconditioning for improved flap survival could be provided simultaneously with or even prior to flap harvesting in the clinical setting.

Early Microsurgical Arteriolysis Decreases Limb Necrosis in Meningococcal Sepsis Infants. Willy D. Boeckx, Marleen Meesters Caberg, and Rene Van der Hulst. Maastricht University, Maastricht, The Netherlands.

Meningococcal sepsis in infants often results in devastating amputations of fingers, toes, upper or lower limbs. The authors' experience is that the extent of skin necrosis usually matches the muscle necrosis.

When Doppler evaluation of the major limb vessels shows an interruption of blood flow, the authors then immediately perform an escharotomy and fasciotomy added to an arteriolysis. This is done within the first few hours after admission of the patient to the pediatric intensive care unit. At that time, vasopressor drugs, such as adrenaline, are given at high concentrations. The arteriolysis consists of a liberation of the arteries from their surrounding fibrous sheets, connective tissue, sympathetic nerves, and comitant veins. Microsurgical magnification is mandatory, since the radial or tibial artery diameter in 1-year-old infants does not exceed 1 mm. Thrombus formation in the vessel lumen is very rare, although no blood flow may be present, due to extrinsic compression which may lead to ischemic necrosis. After arteriolysis, the extent of skin lesions in meningococcal sepsis no longer matches the occlusion of the major limb vessels and the accompanying muscle necrosis, which is more limited.

These techniques have been applied in eight infants. Arteriolysis finally resulted in the following permanent lesions: fingertip necrosis in 4/32 fingers in four patients. One FTG corrected a contracted index finger. Distal nail necrosis was found in one finger with total ischemia of the distal upper arm and forearm. One midfoot amputation occurred when arteriolysis was performed at 37 hr. One distal foot amputation was carried out after early arteriolysis down to the midfoot level and exposing the plantar artery. Three toes were lost after arteriolysis from mid-calf down to the plantar artery distal to the tarsal tunnel.

Early arteriolysis allows for the restoration of the blood supply to the musculo-skeletal system, with only terminal occlusion of the skin vessels, which is not reversible.

Use of Expanded Free Flaps for the Complex Reconstruction of Congenital Giant Pigmented Nevi. Hongshik Han, Neil A. Fine, Bruce S. Bauer, and Alex Margulis. Northwestern University, Chicago, IL.

At Children's Memorial Hospital in Chicago over the last 20 years, 245 patients have been treated with excision and reconstruction of congenital large and giant pigmented nevi. The method of reconstruction is a function of the size and location of the lesion, ranging from skin graft and serial excision to regional pedicle flap, with or without the use of tissue expanders. In some areas in which skin graft was the only available technique, due to the shortage of adjacent flap skin, and where grafts resulted in unacceptable functional and aesthetic outcomes, other approaches have been considered. In these complex cases, expanded free flaps have been utilized with excellent aesthetic and functional results.

Pre-expansion of free tissue transfers has been utilized for the reconstruction of large soft-tissue defects; the final outcome can be more superior aesthetic and functional results. Pre-expansion of the flap increases the dimensions and provides a “delay phenomenon” to the tissues by opening the choke vessels. However, the expanded free flap may be prone to venous congestion, epidermolysis, and subsequent distal flap tip loss after the transfer. The authors have adopted the technique of placing the tissue expander adjacent to the flap, to provide the necessary expansion for enabling primary donor-site closure, rather than to expand the flap itself. This has enabled them to obtain large flaps based on known vascular territories, without the undesirable consequences of expanded free flaps.

The authors presented eight patients with congenital giant pigmented nevi reconstructed with expanded free flaps between 1997 and 2001, and discussed their rationale for choice of the particular TE free flap used, the technical considerations, and the final outcome for each case. The choice of expanded free-tissue transfer included one parascapular flap for a complete forehead giant CPN, one anterolateral thigh flap to a circumferential upper arm lesion, and six TRAM flaps to the mid-upper back and neck, shoulder, and leg.

Defining a Patient Population for Microsurgery and Complex Reconstruction. Frank Louis Stile, Jay H. Schwartz, Sumona V. Smith, Wyvett Walker, and William C. Lineaweaver. University of Mississippi Medical Center, Jackson, MS.

The University of Mississippi Medical Center has created a facility dedicated to the treatment of complex wounds. In the authors' paradigm, a wound center is, by definition, multi-disciplinary in scope, with the primary treating service being reconstructive microsurgery, and with close interaction with nursing and social work, together with various dedicated medical specialty consultants. The center has been modeled after those designed specifically to treat patient populations with diabetes or cardiovascular illnesses.

Complex wound patients in the model are made up of a DRG grouping that has been validated to correspond to certain key ICD-9 and CPT codes. These codes were generated at the time of definitive treatment. The focal point of the center is a dedicated inpatient unit and specialty clinic. The objective is to create a centralized point of service for all complex wound patients in the state of Mississippi, and to promote research and education in the care of these patients.

The authors presented their data on the relative volume of complex wound patients in the 1999-2000 and 2000-2001 academic years seen at UMC vs. the other major revenue (or cost) centers by DRG grouping, e.g., cardiovascular, cancer, etc. For the 1999 academic year, patient admissions with complex wound DRGs accounted for 1142 patients. This represented 5.2 percent of that year's total 22,1000 admissions. Patients admitted with illnesses assigned cardiovascular disease or cancer DRGs totaled 6.7 and 9.0 percent, respectively, of total admissions for the same year. Patient admissions with complex wound DRGs for the following academic year numbered 1173. This represented 4.3 percent of that year's total 27,470 admissions. Cardiovascular admissions were 7.8 percent, while those admitted with cancer-related DRGs represented 5.5 percent of the total admissions for the same year.

The authors' model was described, along with the process used to validate it. They have used this method to identify a patient population not previously defined. Its magnitude is comparable to patient populations of cardiovascular and cancer admissions at their institution. They plan to study this population and its outcomes within the infrastructure of the specialty center.

No Tumor Recurrence at Donor Sites When Local Flaps Are Used for Immediate Coverage in Limb-Salvage Surgery for Sarcomas. E. Gene Deune, Maurice Nahabedian, Anthony P. Tufaro, Deborah Frassica, Steve Lietman, and Frank J. Frassica. Johns Hopkins School of Medicine, Baltimore, MD.

Wide tumor resection, sparing major nerves and vessels with immediate soft-tissue coverage and adjuvant therapy, has become the treatment modality for limb sarcomas. Choices of soft-tissue coverage include local, regional, or distant free flaps. Although easier, the use of local or regional flaps raises concerns about potentially seeding tumor into the flap donor bed. Meticulous handling of the tumor during dissection, to prevent spillage, and changing all drapes, instruments, gowns, and gloves prior to the resection, are important protocols to follow.

Seventy-four patients with limb sarcomas underwent limb salvage and immediate soft-tissue reconstruction between February, 1998 and May, 2002 (51 months). Of these, the authors analyzed those patients who were reconstructed with local flaps and who had at least 12 months of follow-up (mean: 19.1±6.2 months, range: 12-31 months). Twenty-one patients were identified (M=12, F=9), with a mean age of 54.0±18.2 years (range: 20 to 79 years). The most common tumor was MFH (n=9, 43%), followed by osteogenic sarcoma (n=2) and liposarcoma (n=2). Seventy-one percent of the tumors were high-grade (n=15). The thigh (n=5, 24%) and the upper arm (n=5, 24%) were the most common locations, followed by the lower leg (n=3, 14%). Three patients had alloplastic joint replacement and two had allograft bone replacement. Twenty-three pedicle flaps were used to cover 21 wounds. One wound required three flaps. The latissimus dorsi (n=6) and the rectus abdominis (n=5) were used most often.

All flaps survived. Complications were minor: delayed wound healing or seroma of the recipient sites. Eight patients had preoperative adjuvant therapy, 12 had postoperative therapy, including five who had brachytherapy. Two patients developed local recurrences in the original tumor bed (gluteal and upper arm), but both donor flap sites (vastus lateralis, latissimus) remained tumor-free. There has been, to date, no tumor recurrence in the donor sites of the flaps used to cover the wounds in these 21 patients. Six have developed metastases and two have died (mean survival time: 17.0±4.2 months). The most common location was the lung (n=3, 37%). Limb function was self-rated: 15 (71%) reported excellent motion with no or little restriction, compared to preoperative levels, and six (29%) reported moderate function with some limitation. No limbs have been amputated.

Successful flap coverage of complex wounds in limb-sparing surgery for sarcomas is vital for success. The concern for seeding tumor into the flap bed remains real. Based on the authors' experience, this concern can be lessened.

Vascularized Fibular Grafts for the Salvage of Failed Massive Bone Allografts. Steven L. Moran, Allen T. Bishop, and Michael B. Wood. Mayo Clinic, Rochester, MN.

The implantation of massive cadaveric allografts is an increasingly common means of restoring skeletal integrity, following resection of bone and musculoskeletal tumors. Unfortunately, the procedure has a high complication rate, ranging from 55 to 84%. Allograft failure can be a devastating complication and is a difficult reconstructive dilemma. The three most common complications following large allograft transplantation are infection, nonunion, and fracture. The current study reviewed the use of vascularized fibular grafts for limb salvage in cases of failed massive bone allografts.

A retrospective analysis was performed on all patients undergoing salvage of failed massive bone allografts over a 14-year period (July, 1983 to August, 2001). Operative time, hospital course, and postoperative complications were noted. Time to bony union was verified through radiographic evaluation. Vascularized fibular grafts were performed in 22 patients who had failed allograft reconstruction. There were 15 male and 7 female patients with a mean age of 29 years (range: 13-46 years). The cause of allograft failure was nonunion in 14 patients, infection in 5 patients, and fracture in 3 patients. Of the 22 patients, 17 vascularized fibular grafts were performed as a primary procedure, and five as a secondary procedure. The average length of the fibular graft was 20 cm (range: 11-30 cm). Of the 22 vascularized fibular grafts, 18 were free vascularized fibular transfers and four were pedicled fibular transfers. Three procedures were combined with a muscle flap to provide additional soft-tissue coverage to the area.

The average operating time was 9 hr. The average follow-up time was 30 months (range: 6-13 months). Limb salvage was possible in 95% of patients (21/22). Thirty-two percent of patients (7) developed a postoperative complication. Three of these were related to secondary fractures in the allograft following vascularized fibular transfer. Bony union was achieved in 70% of cases (15 patients) after a single operation; two patients required a second free fibular transfer to achieve bony union. The union time between the fibular graft and the native bone or allograft ranged from 1.5-6 months, with an average of 3.5 months.

Excellent limb salvage rates are possible with the use of vascularized fibular transfer for failed massive allograft reconstruction. Vascularized fibular grafts should be considered as a primary means of limb salvage in these difficult reconstructive settings.

Reconstruction of Lower Limb Bone Defects after Sarcoma Resection in Children and Adolescents Using Free Vascularized Fibular Transfer. Tarek Abdalla El-Gammal, Amr El-Sayed Ali, and Mohammed Mostafa Kotb. Assiut University, Assiut, Egypt.

This study included 11 patients with lower limb tumors who had wide local resection and reconstruction by vascularized fibula osteoseptocutaneous flap, and had their surgery performed at least 24 months before the end of the study. The average patient age at operation was 14.4 years. All tumors, except one, were high-grade sarcomas (stage IIA, IIB, and Ewing's sarcoma). The tumor volume averaged 200.6 (range: 41.5-400). The resulting defect after tumor resection averaged 14.8 cm (range: 9-20 cm). The fibula was inserted as a single strut in eight patients and as a double-barrel construct in three patients. Fixation was augmented by interlocking nail in four cases, bridge plate in five cases, and external fixator in two cases.

All flaps survived. All grafts united in an average time of 3.9 months (range: 3-8 months) after transfer. Full weight-bearing was possible after an average period of 6 months (range: 6-10 months). Significant hypertrophy (30% of the original fibular diameter) occurred in all patients after an average period of 10 months from the index operation. In the latest follow-up radiographs (mean: 38 months), the degree of hypertrophy averaged 96.6% (range: 30-200%). Graft fracture occurred in one patient after plate removal and healed conservatively. Shortening averaged 1 cm (range: 0-3). The Musculoskeletal Tumor Society Rating Score (MTSRS) averaged 21 points at the end of the first postoperative year, and 23 at the end of the study.

Soft-Tissue Coverage of Alloplastic Prostheses Used during Limb-Salvage Surgery for Sarcomas. E. Gene Deune, Anthony P. Tufaro, Steve Lietman, Deborah Frassica, and Frank Frassica. Johns Hopkins School of Medicine, Baltimore, MD.

Limb-salvage surgery for sarcomas requires wide surgical resection. When joint resection is required, the authors routinely reconstruct with alloplastic joints to maintain limb function. Reliable soft-tissue coverage of these joints is crucial, as implant infection and exposure may result in amputation and failure of the limb-salvage surgery.

Between February, 1998 and May, 2002 (51 months), 74 patients underwent limb-sparing surgery for sarcomas with immediate soft-tissue flap coverage. Of these, eight patients (M=5, F=3, mean age: 48.1±18.5 years, range: 26-71 years) required joint resection and insertion of alloplastic joints (shoulder, n=3; knee, n=5) with immediate flap coverage. The tumors were located in the proximal humerus (n=3, distal femur (n=1), and proximal tibia (n=4). Malignant fibrous histiocytoma was the most common (n=4), followed by osteosarcomas (n=2). Seven of the eight tumors were high-grade. There were 13 flaps (2 free, 11 pedicle) used. The most common were the gastrocnemius (n=4), latissimus dorsi (n=3), and rectus abdominis (n=2). One free flap failed from venous thrombosis, requiring a second rectus abdominis free flap. All pedicle flaps survived. Overall flap survival was 92.3%. Seven patients had adjuvant therapy (4 preoperatively and 3 postoperatively).

The mean follow-up has been 14.2±10.9 months. Sixty-three percent (n=5) of the patients had one or more postoperative complications: flap loss (n=1), delayed healing (n=3), latissimus donor-site seroma (n=1), periprosthetic infection (n=1). Two patients (flap loss, periprosthetic infection) required repeat surgery to successfully treat their complications. There have been no local recurrence or amputations. Two patients have developed metastasis (lung: 1, spine: 1, but all patients remain alive. All were referred to physical therapy as soon as their wounds had healed sufficiently. Postoperative function was graded as excellent, good, or poor. Thirty-eight percent (n=3) had excellent function, reporting preoperative function; 50% (n=4) had moderate function with some restrictions; and 13% (n=1) had poor function. Functional recovery was better for the lower extremity than for the upper. All five (100%) patients with knee replacements, while only two (66%) with shoulder replacements reported excellent or good function.

Sarcoma involvement of the joints should not be a contraindication to limb-sparing surgery. Joint resection and concurrent insertion of alloplastic implants with immediate flap coverage can successfully rehabilitate limbs that otherwise would have been amputated. Despite a high complication rate, all limbs have been salvaged and overall success has been high, with excellent-to-good limb function.

Use of Allografts with Intramedullary Free Vascularized Fibular Grafts for the Treatment of Large Bony Defects. Steven L. Moran and Allen T. Bishop. Mayo Clinic, Rochester, MN.

Cortical allografts and free vascularized fibular grafts are the most commonly used techniques for the reconstruction of large bony defects following tumor resection. Unfortunately, large bony allografts have a high complication rate. The three most common complications following large allograft transplantation are infection, nonunion, and fracture, all related to the avascular status of the bone. Free fibular grafts provide vascularized bone, but often lack the structural strength of large allografts. In addition, adequate fixation is difficult, as plate and screw fixation can compromise periosteal blood flow. To overcome these disadvantages, a new method of combining a structural allograft with an intramedullary free vascularized fibular graft was developed.

Six patients who had large bony defects following tumor resection were reconstructed with an allograft-vascular fibular construct. Initially, cortical allograft was fashioned to fit the resection defect. The allograft was then reamed so that the medullary segment would accept the fibular graft. The free fibular graft was then passed through the intramedullary canal and doweled into the proximal and distal ends of the patient's remaining bone. The fibular vessels were brought out through a window in the allograft to allow for anastomosis. Fixation was then applied to the allograft. Grafts were evaluated for viability with postoperative bone scans. Time to union was evaluated radiographically. Patients' charts were evaluated for postoperative complications.

There were two female and four male patients, with an average age of 11.5 years. Three cases involved femoral reconstruction and three involved tibial reconstruction. One case involved replacement of a failed allograft used for femoral reconstruction. Three patients underwent chemotherapy during the postoperative course. The average allograft length was 14.5 cm, while the average fibular length was 20.2 cm. The average follow-up time was 26 months. Bony union of the fibula to the allograft and native bone averaged 8 months. Free fibular grafts became incorporated into the allograft. There were no stress fractures or infections. Two patients required additional bone grafting to obtain union between the allograft and native bone. One patient developed a non-union at the donor leg syndesmosis site. Limb salvage was 100%. All patients returned to independent ambulation.

This new technique provides the reconstructive benefits of vascularized bone with the cortical rigidity of an allograft for large bony defects. The free fibular graft provides vascular ingrowth to the allograft, which accelerates bony healing. The technique allows for excellent functional limb salvage.

Hypertrophy after Free Vascularized Fibular Transfer to the Lower Limb. Tarek Abdalla El-Gammal, Amr El-Sayed Ali, and Mohammed Mostafa Kotb. Assiut University, Assiut, Egypt.

This study included 25 patients with lower-limb tumors who had reconstructions with vascularized fibula performed at least 24 months before the end of the study. The average patient age at surgery was 23.5 years (range: 11-50 years). The resulting defect after tumor resection averaged 16 cm (range: 9-20 cm).

The fibula was inserted as a single strut in 21 patients and as a double-barrel construct in 4 patients. Fixation of the graft was augmented by interlocking nails in 11 patients, bridge plates in 9 cases, and external fixators in 5 cases. The relative width of the fibular graft to the recipient bone was estimated by a modification of Deboer and Wood's formula.

Significant graft hypertrophy (defined as >30% of the original fibular width) was observed in 90% of the patients at an average follow-up of 27 months (range: 30-200%). The time to graft union (start of partial weight-bearing) correlated positively with the time to significant graft hypertrophy (r=0.9, p<0.01). The time to significant graft hypertrophy demonstrated no significant differences between patients regarding age, graft length, reconstruction site, fixation method, and chemotherapy administration. No significant increase in the amount of graft hypertrophy was observed after plate removal or dynamization of locked nails. The final amount of graft hypertrophy was affected by the age of the patient (p<0.01) and the length of follow-up (p<0.05).

Graft hypertrophy progressed at an average rate of 3.3% per month (range: 2.3 to 4.9%) until the end of the 30^th month; thereafter, little or no increase was observed. The rate showed two significant peaks at 6-12 months (p<0.001) and at 18-24 months (p<0.05). Patients younger than 20 years of age showed faster hypertrophy, with a peak at the 12^th month, while older patients showed a slower but steadier rate. Hypertrophy progressed more rapidly in patients who received chemotherapy until the 12th month, then declined sharply, compared to patients who did not receive chemotherapy. Otherwise, hypertrophy progressed at a similar rate, regardless of the patient's sex, graft length, reconstruction site, and fixation method.

To summarize, hypertrophy of the vascularized fibular graft is a time-related phenomenon, which is basically dependent on mechanical loading of the graft. It is generally affected by the age of the patient, as younger patients have a tendency toward earlier and faster graft hypertrophy that is probably related to the higher remodeling power and increased level of activity.

When Cookbook Amputations Do Not Fill the Bill: Evaluating the Effectiveness of Applying Modern Wound-Healing Principles and Plastic Surgery Soft-Tissue Techniques to Trans-Foot Amputations in Limb Salvage. Karen F. Kim, Michael Howard, Paul Cooper, and Christopher Attinger. Georgetown University Medical Center, Washington, D.C.

Because of the mortality and risk of contralateral limb loss with above-knee and below-knee amputations, the role of trans-foot amputations is critical in limb salvage. For these amputations to be successful, they have to be well-designed, biomechanically stable and, when healed, have to be adequately fitted by a pedorthetist. Frequently, textbook amputations cannot be performed because of the existing wound and infection. The patient then faces an immediate below-knee or above-knee amputation. The authors eschewed traditional amputation templates and applied wound-healing and soft-tissue reconstructive techniques to salvage these limbs. In the process, they always tried to maximize bone length and then to apply assorted soft-tissue techniques for coverage. They then evaluated their experience, to see how effective this approach was.

The authors reviewed the Georgetown limb-salvage registry to identify all patients between 1992 and 2000, who underwent modified trans-foot amputations (trans-metatarsal, Chopart's or Lisfranc's amputations). Trans-foot amputations were performed on 128 limbs of 116 patients in the time period mentioned. Success was measured by how long the patient was able to use the partially amputated limb, and failure, by when the case progressed to a below-knee amputation. Diabetes and/ or dialysis-dependent renal failure were present in 81 of 116 cases (70%) and 29 of 116 (25%) cases, respectively. Peripheral vascular disease (PVD) was present in 65 (56%) of all patients. The wound presentations were: frank gangrene/osteomyelitis (70%), ulcer (15%), trauma (10%), and unknown (5%).

Operatively, aggressive serial debridements were performed in conjunction with modern wound-healing techniques and revascularization until a clean, granulating, well-vascularized wound was achieved. The most distal amputation was then performed that the remaining soft-tissue envelope allowed. Skin graft, local and pedicled flaps were used when necessary. Tendon rebalancing was also utilized.

Overall, wound closure and long-term limb preservation were achieved in 114 patients (90%), while 14 patients went on to eventual below-knee amputations (9/93 trans-metatarsal amputations underwent a BKA 9 months after initial amputation, 1/19 Lisfranc amputations at 3 years, and 4/16 Chopart amputations at 11 months). Complications of the trans-foot amputation occurred in 32% of patients. These included dehiscence (13/116), partial flap loss (11/116), cellulitis (9/116), partial skin-graft loss (3/116), and death (1/116).

To summarize, the study clearly showed that applying modern wound-healing techniques and using plastic soft-tissue techniques to preserve foot length could achieve a high long-term salvage rate.

Microsurgical Free-Flap Transfer to Amputation Sites: Indications and Results. Burton M. Sundin, Detlev Erdmann, Michael S. Wong, and L. Scott Levin. Duke University Medical Center, Durham, NC.

A series of nine microsurgical free-flap reconstructions to amputation stumps of the upper and lower extremities was reviewed. Indications included preservation of length after trauma (n=6), cure of local infection (n=2), and soft-tissue reconstruction after tumor resection (n=1). Microvascular free flaps used were 4 scapular flaps, 2 fillet flaps from amputated extremities, 1 anterolateral thigh flap, and 1 lateral arm flap. Seven of nine patients were fitted with prostheses and underwent occupational therapy, resulting in functional status. Microvascular reconstruction is indicated in emergency settings, as well as for elective reconstruction of amputation sites. Using uninjured “spare parts” of the amputated extremity should be considered. Elective reconstruction is preferably performed with free flaps based on the subscapular vascular system.

Soft-Tissue Reconstruction during External Fixation for Charcot Foot and Ankle Disorders. Michael A. Howard, Chris Attinger, and Paul S. Cooper. Georgetown University, Washington, D.C.

This study evaluated the reconstructive ladder of soft-tissue reconstructive options in the setting of Ilizarov fixation for Charcot food and ankle disorders.

A retrospective study was conducted utilizing the Georgetown Limb Center Charcot database between 1997 and 2001. Sixty-four patients who received Ilizarov fixators were available for review. In this series, plastic surgery closure was required for 33 patients. Among these, 76 procedures (range: 1-9, average: 2.3) were performed in an effort to achieve final wound closure. Of these, 35 were staged debridements. In 28 patients (85%), definitive closure was achieved without complication. Procedures utilized to achieve closure included delayed primary closure with or without limited bone resection (12), skin graft alone (8), local flap closure +/- skin graft (5), pedicled flap +/- skin graft, and free-tissue transfer +/- skin graft (2).

Complications, including partial flap loss, dehiscence, and failure to heal occurred in 5 patients (15.2%). In these cases, secondary procedures were required to achieve final wound closure. Complete failure to heal occurred in 1 of 33 patients (3.0%).

The authors found that the key to reconstruction around an Ilizarov fixator is to initially cover exposed bone with local tissue. Residual soft-tissue can then be covered with a skin graft. Usually, an adjacent local or pedicled flap in the lower extremity can be found to cover bone exposure that measures less than 10 cm². For defects >10 cm², more distal tissue is usually required (e.g., from a free flap). The VAC has proved to be a very useful device to prepare a wound prior to definitive reconstruction. Indeed, it can decrease the size of the wound sufficiently so that a local or pedicled flap, rather than a free flap, will close the defect. Given this approach, the authors' ratio of local and pedicled to free flaps was 6:1. In the past, the ratio would have been the exact opposite. Space limitations imposed by the fixation device present significant challenges in selecting appropriate reconstructive options. Careful preoperative planning is required with the orthopedic surgeon, so that access for the planned reconstruction is possible. If a free flap is planned, then access to the recipient vessel must be maintained and definitive fixation may have to be delayed until after the free flap is done. With local or pedicled flaps, careful planning in pin placement usually provides sufficient access to the flap.

Radical Excision of Lymphedema of the Lower Extremity with Preservation of the Cutaneous Blood Supply Using Microscopic Dissection of the Skin Perforators: A New Method. Samir Mardini, Hung-Chi Chen, and Jag Chana. Chang Gung Memorial Hospital, Taipei, Taiwan.

Although much of the pathogenesis of lymphedema has been elucidated, the treatment of this physically limiting disease is still in its early stages of evolution. Several methods of treatment have been advocated, including complete excision of all suprafacial structures and skin grafting, staged excision of subcutaneous tissue, microvascular lymph-node transfer, as well as aggressive liposuctioning of affected areas. All of these methods have improved the quality of life of patients treated, when patient selection is strict, execution of surgery is precise, and postoperative care, including compression therapy is followed. The tradeoff of complete excision of lymphedematous tissue, which has very little recurrence, is a disfigured leg. With less radical excision, such as moderate liposuction or subcutaneous excision, recurrence is frequent. More radical subcutaneous excisions can result in significant skin loss.

Between August, 2001 and April, 2002, nine patients underwent surgical treatment of lymphedema using a new technique. All patients had radical excision with preservation of skin perforators. Microscopic dissection of the perforators was the technique used when direct radical excision was performed. Six patients underwent treatment of lower-leg lymphedema, one patient for the treatment of thigh lymphedema, and two for arm edema, using this method. Preoperative Doppler evaluation was used to locate the perforators in all patients.

All patients had a significant improvement in the size of the involved extremity, as measured by non-biased personnel. No recurrence of lymphedema was noted in the current follow-up period. All patients had significant improvement in quality of life. Complications included superficial slough of skin in two patients, and small full-thickness defects in two cases, requiring skin grafts. One patient with skin slough had a hematoma under the medial skin flap. The average number of perforators per patient was 2.4 per skin flap (range: 2 to 5).

Treatment of moderate-to-severe lymphedema requires aggressive excision of scarred and involved tissue. A better understanding of the microvascular anatomy of the lower extremity, with preservation of the cutaneous blood supply, along with the availability of Doppler ultrasonography, can allow for a more radical excision of lymphedematous tissue, while yielding a more aesthetically pleasing result and a low rate of complications. The described method of treatment has yielded promising results and has great potential as a treatment for lymphedema, which can combine a curative excision with an aesthetically acceptable result.

Reverse Sural Neurofasciocutaneous Flap: Versatile Alternative to Free-Tissue Transfer in Lower-Third Extremity Defects. Michael Klebuc, Larry Hollier, Jeffrey Friedman, and Saleh Shenaq. Baylor College of Medicine, Houston, TX.

Free-tissue transfer remains the preferred reconstructive method for many complex lower-extremity defects. Alternatives to microsurgical reconstruction may be sought for 1) individuals unable to withstand a lengthy procedure; 2) young children; 3) hypercoagulable states; 4) vascular trauma; 5) limited soft-tissue defects with simple underlying fracture patterns; and 6) salvage of failed free-tissue transfers. The purpose of this study was to demonstrate the versatility of the reversed sural neurofasciocutaneous flap as a reconstructive option in these difficult situations.

Broad, inferiorly-based fasciocutaneous flaps, incorporating the median sural artery, medial sural cutaneous nerve, lesser saphenous vein, and distal peroneal artery perforators, were utilized to cover 19 defects in 18 patients. The group was composed of 13 males and 5 females, with ages ranging from 2 to 69 years (average: 32 years). The etiology of the defects included 13 open fractures (blunt trauma) 1 gunshot wound, 2 lawn-mower injuries, 1 industrial crush, and 1 chronic venous ulcer. All defects were isolated to the lower third of the leg and heel. The largest skin island measured 10×22 cm, with an average dimension of 7×14 cm.

Flaps were utilized in a variety of ways, including a cross-leg flap for salvage of a failed free-tissue transfer, and for pediatric heel reconstruction. Delayed flaps were employed in 4 smokers with underlying vascular disease. The flap was used bilaterally in a single case, and in one patient with a chronic venous ulcer.

A closed wound was achieved in all patients; however, partial flap loss developed in 10% (2/19). Salvage was achieved utilizing local wound care and skin grafting. One partial loss was attributed to excessive subcutaneous tissue thickness and the presence of extensive lateral malleolar trauma. The second partial necrosis was attributed to the development of a yeast infection. An additional patient developed an enterococcal infection requiring removal of loose hardware. Unrestricted weight-bearing was achieved in 16/18 patients. One patient failed to ambulate as a result of an occluded vascular bypass graft during the 6^th postoperative week, requiring a BKA, while the other remained paraplegic following a spinal-cord injury.

The reverse sural neurofasciocutaneous flap can be utilized as a cross-leg flap, a delayed flap, and in pediatric reconstruction, providing a versatile alternative to free-tissue transfer in select cases.

Long-Term Functional Results Following Salvage of the Lower Extremity with Free Flaps. A.J. Khalil, C. Balakrishnan, and Nayiri Dodikin-Scaff. Wayne State University, Detroit, MI.

Severe crush injuries of the lower extremities are often associated with poor function, chronic pain, and even loss of the extremity. Although certain criteria can be used to determine the degree of salvage of the lower extremity, following severe crush injuries, the decision either to amputate or salvage is not easy. The aim of this study was to determine the results of salvaged lower extremities, using free flaps. The flaps used, technical problems, hospital stay, number of operative procedures, bone infection, non-union, need for bone grafting, length to bone union, and late amputation were assessed.

Over a 5-year period, 15 crush injuries of the lower extremities with soft-tissue and bone loss were treated. The ages of these patients ranged between 18 and 62 years. They had good vascularity of the extremity, with good sensation of the sole of the foot. The patients underwent repeated washout and bony fixation prior to flap closure. All patients were evaluated with angiograms, and muscle flaps were carried out at an average of 9 days from injury (range: 3 to 16 days). Rectus abdominis and latissimus dorsi flaps were used to cover the defects. Skin grafting was delayed in six patients, in five for 2 days, and in one for 4 days. Inflow problems were present in three patients, due to the zone of injury requiring intraoperative revision of the anastomoses.

All patients received similar postoperative care and were transferred to the rehabilitation institute or to home with outpatient rehabilitation. Three patients underwent amputation of the extremity at the end of the first week, due to problems with infection and the vascularity of the tibia. All other patients obtained soft-tissue coverage by about the 6^th week after injury. Bone fixation was either by intramedullary nail or by Ilizarov external fixator. A small external fixator or an extension to the frame corrected foot drop.

Although a challenge, all patients were satisfied to have the lower extremity salvaged. Function improved over time, with no late amputations.

Anterolateral Thigh Adipofascial Free Flap. Ching-Hua Hsieh, Seng-Feng Jeng, and Yur-Ren Kuo. Chang Gung Memorial Hospital at Kaohsiung, Kaosiung, Taiwan.

The anterolateral thigh flap is versatile and useful in reconstructing various types of defects. To repair defects that require a thin and pliable gliding coverage, the authors developed an anterolateral thigh adipofascial perforator flap from the anterolateral thigh region, which utilized the vascularized adipofascial layer, rather than the conventional skin portion of the conventional anterolateral thigh flap.

Eleven cases were reported, in which this flap was used for the reconstruction of extremity defects between 1999 and 2002. Six patients had lower-limb defects resulting from crushing injuries; three patients had finger defects caused by degloving injuries; one patient had a wrist palmar defect after a burn injury, and one diabetic patient had a pretibial chronic ulcer with tendon exposure. There were 6 male and 5 female patients, with ages ranging from 20 to 71 years. Only the deep fascia of the anterolateral thigh and a 2-3-mm thickness of subcutaneous fatty tissue above the fascia were harvested as a free perforator flap for extremity reconstruction. Immediate or delayed split-thickness skin grafting was used for flap resurfacing, and the donor site was closed directly.

The flaps ranged in size from 5-11 cm in length and 4-8 cm in width. A flap dimension as large as 8×22 cm² could be harvested successfully. Immediate split-thickness skin grafts were used to resurface these adipofascial flaps in eight patients; in another three patients, delayed skin grafts were done about 1 week after flap reconstruction. Bleeding from the edge of the deep fascia layer and its adipose surface, and subsequent successful skin-graft take confirmed the viability of the adipofascial flap. All donor sites were closed directly with minimal morbidity.

The virtue of the anterolateral thigh adipofascial flap is its ability to provide a vascularized, thin, pliable, and gliding coverage; in addition, the donor defect can be closed directly with a less noticeable scar and minimized morbidity. Safe elevation, a long and wide vascular pedicle, and a flow-through pattern to reconstruct major vessel defects simultaneously, are also advantages. The adipofascial flap also minimizes flap-surface breakdown by providing subcutaneous fat to cushion a skin graft. The main disadvantages include the need for a skin graft, with possible subsequent loss or hyperpigmentation, and the difficulty of postoperative flap monitoring.

Reconstruction of the Upper Limb with Congenital Differences Using Toe Joint Transfer. Minoru Shibata, Hideki Tsuyoshi, Hajime Suzuki, and Masaji Shirokura. Niigata University Hospital, Niigata, Japan.

Various prostheses can be selected for arthroplasty for adult patients with destruction of or missing joints in the upper limb. However, these procedures are not suitable for children, because of their limited durability and interference with growth. The authors used autologous toe-joint transfers to reconstruct congenitally missing or stiff joints in pediatric patients.

Fourteen joints of 12 patients were reconstructed, using toe-joint transfers. Reconstructed joints consisted of 4 PIP joints with true symphalangism, 1 elbow joint of a patient with Crouzon's disease, and 7 carpometacarpal (CMC) joints of seven hypoplastic thumbs (4 type IIIB and 3 type IV). One MP and one CMC joint of the thumb were reconstructed as salvage procedures for inappropriate previous surgeries for congenital problems. Patient age at operation ranged from 1 year 4 months to 16 years 9 months, with an average of 5 years 8 months.

The authors used the PIP joint of the second toe for finger PIP symphalangism; the second metatarso-phalangeal (MTP) joint for 7 hypoplastic thumbs; and the first MTP joint for an elbow joint. Either the PIP or MTP joint was elevated on the long vascular system of the first dorsal metatarsal or plantar artery. Multiple tendon transfers were combined for adduction, abduction, extension, and flexion of the reconstructed thumb, for the reconstruction of the hypoplastic thumb, as well as the elevation of dorsalis pedis skin for the coverage of the bulky MTP joint. An external fixator was applied for the stabilization of the transferred first MTP joint for elbow reconstruction.

The average follow-up was 11.6 years for Blauth type IIB, 13 months for Blauth type IV reconstruction, 11 years for salvage cases, and 2.3 years for elbow reconstruction. All transfers were successfully performed, and mobile painless joints were reconstructed. The epiphyses have remained open, as those in the other toes allowed for expected growth. Active range of motion (AROM) from 40 to 102 degrees was attained for the reconstructed PIP joint in true symphalangism. The total AROM of the IP and MP joints averaged 36 degrees for Blauth type IIB hypoplastic thumb. The AROM for the reconstructed elbow was 55 degrees, allowing for sucking of the thumb that was not demonstrable before surgery.

Mobile and painless joints can be reconstructed for congenitally defective or stiff joints, to attain useful hand function, by autologous toe-joint transfer. The authors support these procedures as one of the most useful joint reconstructions in children.

Free Multiple Toe-to-Hand Transfer for Mutilating Hand Injuries. Tsu-Min Tsai and Binu P. Thomas. Christine M. Kleinert Institute for Hand and Microsurgery, Louisville, KY.

Mutilating injuries of the hand result in prolonged physical and economic morbidity. Multiple toe-to-hand transfers are a good management option in the treatment of selected cases. The authors reported the use of multiple free toe transfers as salvage procedures, done either immediately for primary reconstruction or secondarily as reconstructive procedures for restoring sensibility. The sensory recovery following pulp resurfacing averaged 13 mm in the transferred pulps. The function of the foot was not compromised after removing two or more toes from the same foot. Harvesting multiple toes from the same foot localizes the potential morbidity to one foot only. Primary reconstruction is preferred whenever possible, as it reduces the duration of treatment and the number of procedures.

Microsurgical Applications for Reconstruction of the Burned Hand: Role of Toe-to-Hand Transfers. Tewodros Gedebou and Fu-Chan Wei. Chang Gung Memorial Hospital-Linkou, Taoyuan, Taiwan.

The purpose of this reported study was to assess the role of microsurgical toe transfers to reconstruct digits of the hand severely injured by burns.

A retrospective review was undertaken of all patients with burn injuries to their hands, who underwent toe-to-hand reconstruction between 1990 and 2000. Demographic information, indications, operative details, and functional results were evaluated over a period of 6 months to 1 year postoperatively. Seven males and 5 females with an average age of 28 years (range: 14-42 years) underwent toe-to-hand transfers following fourth-degree burn injury (thermal or electric) acutely (2 patients) or secondarily (10). Among these patients, a total of 26 toes were transferred from 18 feet to reconstruct 4 thumbs and 22 fingers. Five combined second and third toes, two big toes, ten second toes, and four third toes were taken for digit restoration of the hand.

The immediate postoperative recovery was unremarkable in 10 patients. Reexploration was undertaken in one digit, revealing arterial spasm, and reanastomosis of a thrombosed vein in another digit that was the only partial failure in this series. All donor sites healed without any complications. After healing was complete, six additional (minor) procedures were undertaken in three patients. Functional evaluation during follow-up revealed all digits recovering protective sensation as a minimum. Range of motion was restricted in 5 patients, but exhibited power grip strength of 10 kg on average. Among the remainder, recovery of range of motion of each digital joint and pulp-to-pulp pinch strength between the reconstructed digits were satisfactory.

The microsurgical transfer of toes to reconstruct digits of the hand, missing due to trauma or congenital absence, is standard. However, burn injury to the hand brings additional complexity due to variable injury to the surrounding skin cover, underlying vital tissues, and secondary problems arising from inadequate management during the initial injury period. If amputation is required either during the acute period or due to the useless “hook” finger that is otherwise unreconstructable, a “replica” is available from the foot with little donor-site morbidity. Meticulous surgical technique and dedicated physical therapy are essential for success during the acute and late periods, respectively.

Burn injury leading to the amputation of digits important for hand function is not a contraindication to the application of toe-to-hand transfer techniques for restoration of function and even form to the hand.

Microsurgical Indications for Aesthetic and Functional Nail Reconstruction. Yuichi Hirase, Tadao Kojima, Keizo Fukumoto, and Toshihito Yamaguchi. Saitama Hand Surgery Institute, Saitama Seikeikai Hospital, and Tokyo Hand Surgery and Sports Medicine Institute, Japan.

The hand is an essential organ in human daily life, and is generally naked and unprotected. If part, or all, of the digit is lacking, this creates a great functional disadvantage. In such cases, an adequate method should be selected considering the condition of the remaining finger(s). To select appropriate reconstructive methods, the authors have used the classification of amputation level: type A-the nail matrix is not injured and more than two-thirds of the distal phalangeal bone remains; type B-1-the nail matrix is not damaged but more than half of the distal phalangeal remains; type B-2-the nail matrix is not damaged but less than half of the distal phalangeal bone remains; type C-the nail matrix is lost or severely damaged but PIP joint function remains; type D-PIP joint function is lost.

Surgical procedures are selected based on the amputation classification: for type A, the combined method of finger island arterial flap and split-thickness nail-bed graft from the toe; for type B-1, a type A procedure and bone graft to the fingertip; for type B-2, vascularized nail bed graft from the big toe; for type C, vascularized nail graft, combined with finger arterial island flap; for type D, whole toe transfer. For aesthetic reconstruction of both the donor and recipient sites, minimum harvesting from the toe should be performed, considering combination with a finger flap at the recipient site. Thus, microsurgical transfer is indicated in cases of amputation proximal to type B-2.

Clinical Experience in the Coverage of Extensive Defects of the Upper Extremity. Moustapha Hamdi, Koenraad VanLanduyt, Phillip Blondeel, and Stan Monstrey. Gent University, Gent, Belgium.

Classic skin free flaps are rarely used to cover large skin defects of the upper extremity because of the limited donor size. At the authors' center, they have a wide experience with skin flaps. Using a muscle flap is limited to very specific indications, and the amount of muscle required is restricted to a minimum, in order to decrease donor-site morbidity.

They presented a series of 13 patients who were admitted with extensive defects of the upper extremity between December, 1999 and March, 2002. The mean defect was 30×14 cm. Thin flaps such as the thoracodorsal artery perforator (TAP) or lateral arm flap (LAF) were used for forearm and hand defects. The deep inferior epigastric perforator (DIEP) flap was indicated for deeper defects.

Free skin flaps (1 LAF, 2 TAP, 7 DIEP flaps) were used in nine patients. A chimera flap, which is a combination of two separated tissues based on one pedicle, was needed in 4 patients (scapular with TAP flap in one case, and TAP with partial latissimus dorsi muscle in 3 cases). Immediate defatting of the DIEP flap on the wrist was performed to achieve immediate thin coverage. The average operating time was 4 hr 20 min (range: 3-7 hr). All but one of the flaps survived completely. There was one chimera flap in which the cutaneous portion of the flap necrosed due to a postoperative hematoma. Adequate coverage and complete wound healing were obtained.

Skin flaps or perforator flaps can be used successfully to cover large defects in an extremity, with minimal donor-site morbidity. Chimera flaps provide a large amount of tissue, a wide range of mobility, and easy shaping, modeling, and defatting.

Upper Extremity Microvascular Limb Salvage for Advanced Sarcoma. Yohn Y.S. Kim, Vairavan Subramanian, Allen Rogers, and David Chang. M.D. Anderson Cancer Center and Baylor College of Medicine, Houston, TX.

Limb salvage for advanced upper extremity sarcoma has been demonstrated to be a viable option in select patients. In this setting, chemoradiation is often coupled with radical resection, making durable soft-tissue reconstruction a paramount concern. To this end, the authors investigated the potential utility of microvascular reconstruction in upper extremity limb salvage for advanced sarcoma. Their hypothesis was that upper extremity limb salvage following sarcoma extirpation can be safely and reliably performed using microvascular flaps.

A retrospective analysis was performed on all patients at the authors' institution (1992-2002), who underwent microvascular reconstruction following sarcoma extirpation. A total of 17 patients met the inclusion criteria. Mean follow-up was 38 months, and average age of patients was 52 years. The male:female distribution was 59%:41%. Location of the primary tumor was distal to the elbow in 59% of patients (n=10). Average defect size was 160 cm². Most cases (71%, n=12) were recurrent sarcomas at the time of evaluation. Malignant fibrous histiocytoma variants were the most common pathologic subtype (n=6), followed by leiomyosarcoma (n=3), osteosarcoma (n=2), liposarcoma (n=2), and synovial cell sarcoma (n=2). High-grade tumors comprised 71% (n=12) of the lesions. Most patients (59%, n=10) received preoperative chemotherapy, 76% (n=13) received preoperative radiation, and 12% (n=2) received postoperative radiation therapy.

Free flap survival was 100%. Two patients were taken back to the O.R. for venous congestion, and the flaps survived without sequelae following anastomotic revision. There was no flap infection or dehiscence and one donor-site hematoma. The rectus abdominis was the most common free flap used (41%, n=7), followed by the latissimus dorsi free flap (18%, n=3). Four patients (24%) had local recurrence, with three ultimately requiring amputations (with subsequent disease-free survival). One of the patients with local recurrence died of concomitant metastatic disease. The average modified Enneking score for limb function was 68% (n=14). Cumulative 5-year disease-specific survival was 50%; overall actuarial survival was 3.2 years.

In select patients undergoing upper extremity limb salvage for advanced sarcoma, microvascular flap reconstruction can provide reliable, safe coverage with maintenance of a functional upper extremity.

Radial Sensory Neurotization of the Thumb and Index Finger for Prehension after Proximal Median Nerve Injuries. Ivan Ducic and A. Lee Dellon, Georgetown University, Washington, D.C., and Johns Hopkins University, Baltimore, MD.

In the absence of war, there have been few patients requiring microsurgical reconstruction of injuries to the median nerve where the location of the injury is proximal to the elbow. Available microsurgical techniques for nerve reconstruction will permit the sensory component of the median to reach the thumb and index finger in about 3 years after reconstruction. This report described the successful neurotization of the thumb and index from the radial sensory nerve, in order to provide sensibility for pinch.

The first patient reported was a 21-year-old right-handed baseball pitcher, who had 23 cm of the right median nerve removed during a tendon reconstruction. The second patient was a 53-year-old right-handed man with a glass laceration of the median and ulnar nerves in the right axilla. The operative procedure transferred the dorsal sensory branches of the radial nerve into the ulnar digital nerve of the thumb and the radial digital nerve of the index finger. By 3 months after surgery, there was sufficient sensation in each patient to provide protection of these fingers. By 6 months, there was sufficient sensibility to permit active lateral key pinch. The results at 1 year for each patient were presented, along with the sensory rehabilitation protocol.

Radial Artery Fascial Free Flap with Vascular Donor Reconstruction: A Low-Cost, Versatile, Reconstructive Choice. Wyndell H. Merritt. Richmond, VA.

The use of radial artery free flaps for large defects has been denounced because of the unacceptable grafted donor defect. Use of radial artery fascial free flaps leaves only a palmar linear forearm scar, creates larger vascularized flaps, provides “macro-micro” easy vascular anastomoses for older surgeons and, with vein graft radial artery reconstruction, has a minimal donor cost. Nonetheless, temporalis, arm, and back fascial free flaps seem more popular. Five arm and leg cases were chosen to illustrate the ease, versatility, and technical suggestions for the author's method.

These cases included circumferential late wrist reconstruction with palmar contracture release, proximal carpectomy, and extensor tendon grafting following train injury; hand dorsum reconstruction with extensor tenolysis in a fisherman who fell into a chum bucket, with subsequent vibrio septic shock and necrosis; two Achilles tendon coverage cases; and one heel reconstruction over exposed calcaneal avulsion.

All the flaps survived, although one skin graft proved unstable and was regrafted. All reconstructed radial arteries remained patent beyond a 10-year follow-up, contrary to some speculation.

The radial artery fascia provides an underutilized free flap the size of the entire palmar forearm, particularly useful over tendon gliding surfaces of large defects. Reconstructed radial arteries remained patent and the grafts durable, even on the heel.

Temporary Innervation of a Primary Coverage Muscle: A New Technique to Optimize Function in a Subsequent Functional Microvascular Muscle Transplant. Darrell Brooks. Buncke Clinic, San Francisco, CA.

Functional microvascular muscle transplantation (FMMT) has met with less success in situations involving traumatic major soft-tissue and nerve loss, because of the less than optimal recipient wound bed and target motor nerve. The author reported an innovation to the staged reconstruction of a traumatized upper extremity with an FMMT. He also discussed how this innovation optimized the recipient bed and target motor nerve.

A 37-year-old man sustained a crush-degloving injury of his forearm. Surgical exploration revealed an 18×22-cm soft-tissue defect, which included the dorsal and palmar compartments, as well as the radial nerve proximal to the lateral epicondyle. The patient underwent latissimus muscle transplantation to obtain primary wound cover. Next, the thoracodorsal nerve, measuring 10 cm, was repaired to the mixed motor-sensory radial nerve stump, which was debrided 2 cm proximal to healthy-appearing fascicles well outside the zone of injury.

Between the 6^th and 8^th weeks after muscle transplantation, the latissimus began to contract at the proximal forearm. These contractions strengthened and, by the 13^th week, the patient could voluntarily contract the entire muscle. At that time, the patient had no active wrist or digit extension, and injury to the palmar forearm precluded tendon transfer. Eight months after latissimus transplantation, the contralateral gracilis muscle was harvested to restore finger and wrist extension. On elevation of the latissimus, an avascular septo-alveolar plane similar to that between two healthy muscle bellies, could be developed with gentle finger dissection. The latissimus was elevated as a bipedicle flap and the gracilis was drawn into this plane. The thoracodorsal-obturator nerve repair was performed as close as possible to the gracilis muscle.

At 8 weeks, the patient had regained active extension of his wrist and fingers. Seven months after FMMT, the patient had excellent extension of the wrist and digits. Strength was 4+, based on the Medical Research Council (MRC) grading scale.

The simple technique of innervating the muscle used for coverage of a large upper extremity wound helped overcome the potential limitations introduced by traumatic injury. The latissimus muscle excursion not only signaled the appropriateness of the target motor nerve, but also eventually created a recipient bed comparable to the plane between two healthy muscles. The thoracodorsal nerve provided enough graft length to allow initial proximal repair outside the zone of injury and eventual distal repair as close as possible to the functional muscle, decreasing reinnervation time.

Factors Associated with Late Vascular Thrombosis of Free Tissue Transfers. Roman Skoracki, C.E. Butler, and David W. Chang. M.D. Anderson Cancer Center, Houston, TX.

The vascular thrombosis rate of free tissue transfers has been reported to be 2 to 5%. Most thromboses occur within the first 48 hr postoperatively. On rare occasions, the authors have experienced late (postoperative day 5 or later) free flap vascular thromboses. To their knowledge, there have been no published reports which evaluate the potential etiologic or predictive factors associated with these cases. In an effort to identify potential factors associated with late free flap thrombosis, they retrospectively reviewed all free tissue transfers performed at a single institution over a 16-year period. They wished to quantify the incidence of late microvascular thromboses in a large number of elective free flaps used for oncologic reconstruction, and to identify specific factors that might be causative, predictive, or associated with late thrombosis.

All patients who underwent free tissue transfer between 4/1/86 and 4/1/02 at the University of Texas, M.D. Anderson Cancer Center, were reviewed. A comprehensive database was used to identify patients who developed vascular thrombosis on postoperative day 5 or later. A retrospective chart review was conducted to analyze multiple factors potentially associated with these late thromboses.

A total of 9 patients who developed a late free flap thrombosis were identified. Only one (11%) of the nine flaps could be salvaged at reexploration. No identified pre- or postoperative factors were associated with late flap thrombosis. Adverse intraoperative events, including severe vasospasm, poor arterial inflow requiring anastomotic revision, and requirement of a vein graft were noted in 6 of the 9 cases. Five of the 9 cases also required at least one anastomotic revision at the time of initial surgery. Smoking, atherosclerosis, and diabetes did not appear to affect late free flap thrombosis.

Late free flap vascular thrombosis is rare and the salvage rate is low. Certain adverse intraoperative events and the need for anastomotic revision are factors that may correlate with or predict late vascular thromboses.

Free Tissue Transfer for Large Lumbosacral Wound Reconstruction. Todd Michael Willcox and Craig Johnson. Mayo Clinic, Rochester, MN.

Soft tissue and skeletal reconstruction of lumbosacral wounds after tumor extirpation is a difficult problem. These defects are frequently large, irradiated, and have failed conservative management. They also frequently require fixation devices and hardware. Local flap options may not be available, or of sufficient size, requiring the reconstructive surgeon to consider free tissue transfer.

The authors reviewed their patient database over a 12-year period, and nine patients were identified with complex mid-back wounds treated with free-flap coverage. Seven different types of neoplasms were resected, and six (67%) had been irradiated. Seven (78%) patients required saphenous vein arteriovenous fistulas, and two (22%) received vascularization from local vessels. Four of the 7 (57%) vein loops were staged 48 hr prior to flap transfer. There were six latissimus dorsi muscle flaps, one latissimus dorsi musculocutaneous flap, one rectus abdominis muscle flap, and one rectus abdominis musculocutaneous flap used for reconstruction.

Six of the 9 (67%) remained viable and healed without incident. Three (33%) failed and required revision. Two of the three failed flaps were in irradiated tissue beds, and one had been transferred with a nonstaged vein loop.

Free tissue transfer for complex back wounds is a reasonable reconstructive option when loco-regional techniques have been exhausted. It is the authors' practice to perform a staged saphenous vein arteriovenous fistula 48 hr prior to tissue transfer. They believe that this provides reliable inflow and drainage for subsequent free tissue transfer. Preferably, one would go directly to the iliac vessels in an end-to-side fashion, but this was possible only in the minority of cases.

Effects of Renal Disease on Free Tissue Transfer. Steven L. Moran and Joseph M. Serletti. Mayo Clinic, Rochester, MN, and University of Rochester, Rochester, NY.

Soft-tissue reconstruction in patients with renal disease can pose great challenges to the reconstructive microsurgeon. The authors performed a retrospective analysis of all patients who underwent free tissue transfer with concomitant renal disease, to determine what effect renal disease had on flap survival and reconstructive success.

A retrospective analysis was performed on all free tissue transfers between January, 1991 and August, 2000. All patients with a history of renal insufficiency (creatine >1.6 mg/dL), renal failure requiring dialysis, and functional renal transplants were included in the study. Patients' charts were examined and telephone interviews were conducted to determine the postoperative course. Of the 1053 free flaps examined, renal disease was identified in 32 patients who underwent 33 free flaps. The average patient age was 57 years. Twenty-eight flaps were performed for complications of limb-threatening ischemia, four were for tumor reconstruction, and one was for trauma. Twelve patients were on chronic dialysis (ESRD), 15 patients had renal insufficiency (CRI), three had the diagnosis of ARF at the time of surgery. Three patients had a functioning renal transplant. Twenty-five patients had peripheral vascular disease, 28 suffered from diabetes. Nineteen patients underwent a concomitant vascular bypass procedure at the time of free tissue transfer.

The average follow-up was 19 months. Immediate postoperative complications occurred in 12 patients (36%). Overall perioperative mortality was 3%. Within the first 30 days, there were only two cases of primary flap loss. Two lower extremities were amputated because of vascular complications: one due to hemorrhage and one to vascular bypass ligation. This resulted in a total of five early reconstructive failures (15%). There were no primary flap failures after 30 days; however, within the first year following surgery, an additional six limbs were lost due to progressive ischemia or infection, and an additional five patients died. This resulted in a 67% incidence of major morbidity or mortality occurring in the first postoperative year, and a 52% reconstructive success rate at 1 year. No significant difference was seen in postoperative morbidity or mortality when comparing the ESRD group to the CRI group. Sixty-four percent of patients with diabetes experienced a postoperative complication at 1 year, compared to 20% of non-diabetic patients.

Renal disease, especially renal disease associated with diabetes, and critical lower limb ischemia, should be strong indicators of possible reconstructive failure. Surgeon and patient should be aware of the medical and surgical complications of this procedure prior to the outset.

Hypothermic Protection of Compromised Flaps. Michael Neumeister, Arian Mowlavi, Bradon Wilhelmi, Robert C. Russell, and Garth Meldrum. Southern Illinois University, Springfield, IL.

Approximately 3 to 10% of microsurgical procedures are subject to vascular compromise. The authors proposed that tissue loss is compounded not only by the increased ischemia time (and IR injury), but also by the increased (hyperthermic) ambient temperatures of normal body heat and the placement of postoperative free flap patients in warm rooms (80 degrees F). They hypothesized that compromised flaps maintained in a hypothermic environment will improve survival.

The rat rectus femoris muscle flap was elevated on its major pedicle and subjected to 4 hr of ischemia. The flaps were then exposed to either hypothermia (50 degrees F) or ambient temperatures of 72 degrees F during the early perfusion period. Following 1 hr of perfusion, complete venous occlusion of the muscle flap was accomplished by placing clamps across the venous outflow. Following 4 hr of venous occlusion, the clamps were removed. This simulated the clinical scenario of a salvage attempt of free flaps that develop subsequent venous occlusion and require reexploration. The flaps were then allowed to reperfuse at normal body temperature. Muscle necrosis and muscle edema were assessed following 24 hr of this reperfusion. A microarray cDNA analysis was performed on each group to observe changes in gene expression of the different muscle groups.

Hypothermic exposed muscles (n=4) demonstrated decreased muscle necrosis, compared to the room temperature-exposed muscles (n=4) following 4 hr of venous occlusion (9.3±5.7% vs. 93.6±8.9% necrosis, p<0.001). Hypothermic exposed muscles demonstrated a trend toward improved edema (1.79±0.36 vs. 1.89±0.38 wet to dry weight ratio, p=0.684). The gene expression analysis of the muscle groups revealed a significant number of up-regulated and down-regulated genes, most notably egr-1 which was up-regulated 104-fold.

These preliminary results suggested a role for exposure of elective free flaps to regional hypothermia during the early reperfusion period, to provide protection from IR injury associated with venous occlusion. The variations in gene expression were discussed.

Use of Interpositional Vein Grafts Does Not Contribute to Higher Flap Failure Rates When Planned Properly. Mark C. Tan and Howard M. Langstein. The University of Texas, M.D. Anderson Cancer Center, Houston, TX.

There is a perception that the use of interpositional vein grafts in microsurgery is associated with higher anastomotic and flap failure rates, but it is not clear whether failures are caused by poor planning or the emergent use of vein grafts. In an attempt to test the hypothesis that “well-planned” vein interposition grafts do not increase the risk of flap failure, a retrospective review of all vein grafts performed in the authors' institution over the last 10 years was undertaken.

A total of 112 vein grafts were performed in the course of 62 free tissue transfers. Most of these (99/112) occurred in head and neck reconstructions and were the focus of the study. Factors that might contribute to flap failure (radiation therapy, previous surgery, age, flap choice, recipient vessels, length and type of vein graft) were subjected to univariate and multivariate analyses to determine whether they were associated with failure. Vein graft use was also separated into a well-planned or anticipated group (defined by graft harvest before flap pedicle division, n=30), an unanticipated group (vein grafts harvested after flap pedicle division, n=50), and a reoperative salvage group (harvested following a return to the operating room, n=19). Vein grafts to arteries (n=49) were analyzed separately from vein grafts to veins (n=50).

Well-planned venous vein grafts had a statistically significantly (p<0.05) lower thrombosis rate (0/14, 0%) than those which were unanticipated (2/27, 7.4%) or used during reoperative salvage (4/9, 44.4%). Overall, the use of vein grafts was associated with a much higher flap success rate (91%) than has been previously reported. Those which were well-planned were associated with failure rates (1.8%) similar to non-vein grafted free flaps. Flap choice, the presence of radiation therapy, previous surgery, length of vein graft, recipient vessels, and age did not influence flap success or failure, either as a single factor or when analyzed by logistic regression.

Vein grafts were associated with a higher flap loss rate when the grafts were used in a salvage or emergent situation. In contrast, when the vein grafts were well-planned, flap success rates were similar to those in non-vein grafted flaps. This study demonstrated that, despite additional microanastomoses, interpositional vein grafts can be used with no increase in flap failure rate.

Contractile Function of Reperfused Skeletal Muscle is Protected by Complement Inhibition with C1-Esterase Inhibitor. Glen A. Toomayan, John Haixiang Jiang, Long-En Chen, Anthony V. Seaber, and James R. Urbaniak. Duke University Medical Center, Durham, NC.

Ischemia/reperfusion (I/R) injury is a pathologic phenomenon observed in microvascular surgery, which results in failure of revascularized tissue, despite the presence of a technically well-performed anastomosis. This study examined the role of complement in skeletal muscle I/R injury. Human C1-esterase inhibitor (C1-INH) is a naturally occurring protein that binds to and inhibits the activated C1r/C1s-esterase complex of the complement pathway. A synthetic 12 amino acid peptide with a similar inhibitory effect on the activated C1r/C1s complex was also examined in this study.

Eighteen rats weighing 125-150 g underwent 3-hr ischemia and 3-hr reperfusion of the right extensor digitorum longus (EDL) muscle. An intravenous infusion of C1-INH (100 mg/kg), peptide (5 mg/kg), or human serum albumin control (n=6 per group) lasting 10 min was started 10 min before reperfusion. After 3 hr of reperfusion, the EDL underwent in vitro contractile testing. Contractile forces were normalized to the contralateral untreated, non-ischemic EDL.

There was a significant overall increase in tetanic contractile force of the reperfused EDL in both C1-INH and peptide groups, compared to controls, across stimulation frequencies of 70 Hz, 100 Hz, and 120 Hz (p<.01). Maximum improvement occurred with peptide treatment at 120 Hz stimulation, with an increase in force from 38.4±10.9% of normal in controls to 52.4±8/9% of normal in peptide-treated rats (p<.05). There were no significant differences between C1-INH and peptide groups.

The results suggested that inhibition of complement has a beneficial effect on skeletal muscle contractile function in early reperfusion. Treatment with either an endogenous inhibitor (C1-INH) or a synthetic peptide inhibitor yields similar results. Inhibition of complement represents a potential therapeutic approach to preventing I/R injury.

Combination Treatment of Nitric Oxide Donor and Inducible Nitric Oxide Synthase Inhibitor Improves Microcirculation in Reperfused Skeletal Muscle. Pruk Chaiyakit, Long-En Chen, Anthony V. Seaber, and James R. Urbaniak. Duke University Medical Center, Durham, NC.

This reported study evaluated the effects of a combination of 1400W ([3(aminomethyl)benzyl]acetamidine), a highly selective iNOS (inducible NO synthase) inhibitor, and SNAC (S-nitroso-N-acetylcysteine), an NO donor, on microcirculation in reperfused skeletal muscle.

The left cremaster muscle of 87 rats weighing 90 to 100 g underwent 5 hr of ischemia followed by 90 min of reperfusion. The rats were divided into innervated (n=43) and denervated (n=44) groups. Each group was further divided into 3 subgroups treated with intravenous PBS, subcutaneous injection of 1400W (3 mg/kg), or a combination of subcutaneous 1400W (mg/kg) and intravenous SNAC (100 nmol/100 gm/min). The agents were administered 30 min before reperfusion, and SNAC infusion was continued throughout the reperfusion. At 10-min intervals during reperfusion, vessel diameters were measured by intravital microscopy and blood flow was measured by laser Doppler.

In the innervated model, treatment with both agents yielded a maximal flow of 137±27% of baseline at 40 min, compared to 127±22% at 80 min with 1400W and 102±23% at 90 min in controls. Time to return of baseline flow improved from 67.5±17.5 min in controls to 40.0±15.1 min for 1400W (p<0.01 compared to control), and 20.0±9.2 min for both agents (p<.01 compared to 1400W). Treatment with both agents led to greater vessel diameters than 1400W alone or control at nearly all time points. The greatest difference occurred at 20 min in 10 to 20-micron vessels (65±12% of baseline diameter for controls, 77±11% for 1400W, and 99±21% for both agents, p<.05 for both agents compared to 1400W).

In the denervated model, flow returned to baseline in 8 of 8 rats in an average of 35.6±13.3 min for both agents, whereas only 2 of 8 1499W-treated rats and 1 of 8 control rats returned to baseline flow. Vessel diameters were greater at nearly all time points in rats treated with both agents than 1400W or control groups.

The study demonstrated that a combination treatment of 1400W and SNAC restored microcirculation better than 1400W alone in both innervated and denervated reperfused muscle. The fact that a combination treatment of NO donor and iNOS inhibitor resulted in greater maximal flow and earlier restoration of baseline flow than iNOS inhibitor alone, suggests that supplementation of NO to I/R tissue is important for improvement of microcirculation and reduction of I/R injury. The ability of NO donor SNAC to reduce I/R injury appears to protect small arterioles from vasospasm.

Pulsed Magnetic Energy Supports the Prefabricated Flap Based on a Transplanted Rat Tail Arterial Loop. Renata V. Weber, June Wu, Han-Liang Yu, Jennifer Weintraub, and Berish Strauch. Albert Einstein College of Medicine and Montefiore Medical Center, Bronx, NY.

Angiogenesis occurs in arteries exposed to pulsed magnetic energy (PME). Using a rat groin cutaneous flap model, the authors created a neo-vasculature by anastomosing the tail artery to the femoral artery and placing it under the groin and abdominal skin. They proposed that transplanted arterial loops treated with PME would support the skin flaps, even when the superficial inferior epigastric vessels were divided.

Male Sprague-Dawley rats (6-7 per group) were divided into six groups. The tail artery (12 cm) was anastomosed end-to-end or end-to-side to the right femoral artery proximal to the superficial epigastric artery and placed under the skin. Experimental groups received 12 weeks of PME for 30 min twice a day. After 12 weeks, all groups had an 8×4-cm skin flap raised, and the superficial epigastric artery, or both superficial epigastric artery and vein were ligated. On day 5, the area of viable skin was traced onto velum paper, randomly renumbered, and the percentage of viable skin over total surface area calculated.

Dividing the superficial epigastric vein resulted in flap necrosis (p<0.01); however, all flaps with intact venous drainage had significant flap survival based solely on the transplanted arterial loop (p<0.05). In the end-to-side group, treatment with PME significantly improved flap survival (26% vs. 79%, p=0.01). Treating the end-to-end loop for 12 weeks did not appear to increase survival of the flap over controls.

Venous drainage is essential to flap survival. The cutaneous flap was able to survive purely on circulation from the transplanted arterial loop. PME treatment in the end-to-side group significantly improved survival of the flap, and it may have a role in the prefabrication of flaps.

Advanced Breast Cancer and the TRAM Flap. Amir Babak Behnam, M.D. Nguyen, S.L. Moran, and J.M. Serletti. University of Rochester Medical Center, Rochester, NY.

Many studies have demonstrated the benefits of post-mastectomy breast reconstruction with the transverse rectus abdominis myocutaneous (TRAM) flap. Initially, surgeons offered this reconstructive option to patients with early disease with good results. TRAM flaps in advanced cancer patients were often considered only for complex chest-wall defects following extensive resection. Good results in this select population without interference with adjuvant therapies (chemotherapy and radiation therapy) led surgeons to more frequently offer the TRAM flap to all patients, including those with advanced disease.

This study presented the authors' experience with TRAM flap breast reconstruction in patients with stage II-IV breast cancer over a 10-year period, including an informal survey of patient satisfaction. Twenty-one patients were identified, with a mean age of 48.6 years (range: 35-73 years). All patients had stage III disease, except one who had stage IV breast cancer. A total of 26 TRAM reconstructions were performed by the same surgeon.

Postoperative complications included fat necrosis (3), hematoma (1), cellulitis (1), poor donor-site healing (2), and seroma (1). Follow-up from time of TRAM reconstruction to survey averaged 6.5 years (range: 2-10 years). Of these 21 patients, 9 (42.9%) died primarily due to metastasis. Only one patient had definite local recurrence of tumor. The average interval of time between TRAM flap and death was 3.7 years (range: 1-6.5 years). The average recovery time before patients were able to return to normal activities or work was 10.6 weeks (range: 3-36 weeks). Eight patients and 3 surviving family members participated in the patient satisfaction survey, for a total of 11 completed surveys. Of the 11 patients or families that completed the survey, 11 (100%) would repeat the TRAM reconstruction.

The results of this study indicated that patients with locally advanced breast cancer would consider reconstruction, if offered the option again, and would have a lengthy disease-free interval following their reconstruction.

Utility of a Hand-Held Ultrasound Device for Planning and Monitoring of Microvascular Free Tissue Transfer. Alexes Hazen, Hooman Soltanian, Jamie P. Levine, and Geoffrey C. Gurtner. NYU Medical Center, New York, NY.

Monitoring of free tissue transfer is essential in the early postoperative period for attempts to salvage compromised flaps. Numerous non-invasive techniques are available for monitoring tissue viability, including clinical observation, hand-held Doppler, surface temperature probes, implantable venous Doppler, and color duplex. Evaluation of recipient vessels and identification of perforators are important for thorough preoperative planning. Visualization of recipient vessels is crucial in previously irradiated or operated sites. Commonly used non-invasive methods for assessment of recipient vessels and perforators have been limited to duplex examination with large ultrasound units. Traditional duplex machines are not readily available because they are difficult to transport and have to be operated by vascular technicians.

With the advent of high-resolution hand-held ultrasound units, the authors sought to evaluate the efficacy of a portable device for pre- and postoperative assessment of free flap vasculature. They used SonoSite 180^TM with a 5-10 mHz linear transducer. The transducer can visualize structures to a depth of 7 cm. The unit can color-code blood flow and measure the vascular diameter with 0.1 mm accuracy. Fluid collections are easily identified and distinguished from vascular structures.

A total of 13 patients were examined. Five were evaluated preoperatively. Four patients underwent free TRAM procedures. Dominant flap perforators, size, and location of recipient vessels were visualized. The findings correlated closely with intraoperative observations. One patient had a free parascapular flap transfer for facial asymmetry. The superficial temporal vessels were identified despite multiple previous surgeries and scarring in the preauricular region. Ten patients were evaluated following free flap procedures. Vascular pedicles were visualized in all cases. The portable ultrasound was used in two patients to identify fluid collection and to drain them percutaneously.

Among the non-invasive methods for monitoring and evaluation of free flaps, the hand-held ultrasound device is the most convenient, easy to use, and versatile. Ultrasonic evaluation of recipient vessels and perforators is accurate and reproducible. Actual vessel size corresponded to the size predicted by imaging. Identification of perforators is helpful in designing flaps and in pre-surgical planning. Use of the ultrasound for postoperative monitoring of free flaps yields more information than the Doppler, as both artery and vein can be visualized. The technique is useful in evaluation and treatment of fluid collection. The unit is readily available at any time and can be used by plastic surgery residents after a short training period.

Penile Revascularization in Select Cases of Prostatomembranous Urethral Disruption or Post-Traumatic Erectile Dysfunction. Daniel Rosenstein, Gerald Jordan, and Lawrence Colen. Eastern Virginia Medical School, Norfolk, VA.

At the time of traumatic prostatomembranous urethral disruption, the internal pudendal arteries suffer varying degrees of injury. These may put the patient at risk for ischemic stenosis following urethroplasty. Cavernous arterial inflow may also be severely decreased in some of these patients, thus limiting the capacity for both physiologic and pharmacologically-induced erections. This report reviewed the authors' experience with penile revascularization in a group of carefully selected trauma patients.

Between 1994 and 2001, penile revascularization was performed in 8 patients who had suffered prostatomembranous urethral disruptions. They were found by both Doppler ultrasound and highly selective angiography to have bilateral injury to the internal pudendal arteries, yielding abnormal vascular parameters. The patients underwent revascularization procedures utilizing end-to-side arterial anastomosis between one of their deep inferior epigastric arteries and one or both dorsal penile arteries at the base of the phallus. In one patient with a poor donor artery, a “Y” arterial graft of the thoracodorsal artery was interposed between the common femoral artery and the dorsal penile arteries. These patients were then systematically reevaluated for return of erectile function, for success of subsequent urethral reconstruction, and by duplex imaging that assessed improvement in vascular parameters.

In all cases, there was a significant increase in arterial inflow to normal by Doppler flow analysis. In these patients, a patent revascularization was demonstrated by duplex imaging. Patients who underwent subsequent urethral reconstructions suffered no incidence of ischemic stenosis. All patients with vasculogenic erectile dysfunction unresponsive to Sildenafil or intravenous injection therapy became responders to these agents following the procedure.

Penile revascularization should be strongly considered in carefully selected patients with documented post-traumatic vasculogenic erectile dysfunction, or prior to urethroplasty in cases of urethral circulatory compromise.

Using Nerve Conduits in End-to-Side Neurorrhaphy. Trevor Pickering, Long-En Chen, Anthony V. Seaber, and James R. Urbaniak. Duke University Medical Center, Durham, NC.

End-to-side neurorrhaphy (ESN) has reemerged as a potential alternative technique in nerve repair. This study sought to determine whether collateral axonal sprouting can occur across gaps in end-to-side neurorrhaphy and to compare the functional recovery of gap-repaired lesions to those repaired by direct suture.

The right peroneal nerve of 80 female Lewis rats (200 g ±25) was transected 3 mm distal to the bifurcation of the sciatic nerve. A perineurial window sized to match the distal peroneal stump was made in the tibial nerve of each rat. The animals were then divided into four groups of 24 each. In the direct suture group, ESN repair was carried out by suturing the distal stump to the epineurium surrounding the window with 12-0 nylon suture. The remaining three groups were treated with an ultrathin-walled PTFE plastic tube with a gap of 2, 4, and 6 mm. Results were assessed by walking-track test calculated by the peroneal functional index (PFI) and the tibial functional index (TFI); wet weight ratio of the extensor digitorum longus (EDL) muscle; contractile functional testing of the EDL; and histologic examination.

The PFI recovered to normal range in all groups by week 12, with the first significant improvement in the direct ESN group at week 4, followed by the 2-mm group at week 5, the 4-mm group at week 6, and the 6-mm group at week 8. From week 5 onward, there was no significant difference between the direct and 2-mm groups, but both these groups were significantly better than the 4-mm and 6-mm groups. By week 9, there was no significant difference in the PFI between the groups. The TFI demonstrated some initial impairment in donor-nerve function, but recovered to normal range by week 5. Muscle atrophy was noted in the EDL at 8 and 12 weeks. At week 12, there was a significant difference between the wet weight ratios of the EDLs, and the direct, 2-mm, 4-mm, and 6-mm groups. Contractile testing revealed greater muscle force in the direct and 2-mm groups at each stimulation frequency at each time period. Histology demonstrated sprouting axons in the tubes of each gap group, and regenerated axons in the distal peroneal nerve in all groups.

The study demonstrated successful nerve regeneration and considerable motor functional recovery across a conduit up to 6 mm in end-to-side repair, indicating a potential clinical alternative when direct suture is not possible.

Flap Prefabrication with Bioengineered Acellular Dermal Replacement (Integra^TM) in a Rat Model. Colin H.J. Tham, C. Song, R. Ng, I.C. Song, and I. Kee. Singapore General Hospital, Singapore.

The objectives of this study were to investigate the vascularization of a bioengineered acellular dermal replacement (Integra^TM) in a rat model, and subsequently to effect the microsurgical transfer of the neovascularized “Integra flap” in the same animal model. The findings of this study should form the foundation for further work on the prefabrication of free flaps, en route to the development of thin “designer flaps.”

The study was conducted utilizing a rat model in two separate phases. Phase one: Under anesthesia, the superficial inferior epigastric vessels were isolated and sandwiched within a 20x10-mm folded sheet of Integra. In the contralateral groin, a folded sheet of Integra without inlaid vessels acted as a control. Twenty rats were subdivided in four groups of 5 rats each, which were assessed at 24, 48, 72, and 96 hr, respectively, for evidence of neovascularization of the “Integra sandwich.” Assessment was conducted via stained histologic sections, plain, and intravital microscopy. Phase two: Twenty “Integra flaps” were prefabricated, as described in phase one. After a delay period of 96 hr, to allow sufficient neovascularization of the dermal replacement, microsurgical free-flap transfer was performed. Assessment of flap viability after 72 hr was conducted as in the first phase.

In phase one, the earliest evidence of neovascularization occurred within 24 hr and was detected on intravital microscopy and histologic examination. There was no evidence of neovascularization in the controls. In phase two, free microsurgical transfer of the flaps was successfully concluded with post-transfer flap viability confirmed on histologic examination, plain, and intravital microscopy.

Vascularization of Integra flaps in a rat model was demonstrated to consistently occur within 24 hr. Microsurgical transfer of the free Integra flap was successfully performed.

Acellular Dermal Composite Allograft Reconstruction of the Radial Forearm Donor Site: A Significant Improvement in Form and Function. Mark D. DeLacure, Norm Rowe, Jamie Levine, Abraham Kuriakose, and Hooman Soltanian. NYU School of Medicine, New York, NY.

Since its description in the 1970's, the radial forearm free flap has earned a clearly defined role in the armamentarium of reconstructive head and neck surgery. Three decades later, the donor site remains an intrinsic drawback, primarily due to its aesthetic impact, although functional morbidity is significant in a minority of patients. These aspects do not outweigh its significant advantages, but are occasionally reasons for the choice of alternative flaps. Modifications have evolved in an effort to improve the undesirable features, and include primary closure, rotation-advancement, proximal paddle placement, use of FTSGs, and suprafascial dissection.

Acellular dermal allografts (Alloderm-Branchburg, NJ) have been in broad clinical use since their introduction in 1993. Use as the basal layer of a composite graft (patient's STSG the outer finishing layer) has been described. From July, 1999 to May, 2002, 10 patients underwent radial forearm reconstruction of head and neck defects in which the donor site was closed with a composite technique. Control patients had donor sites closed by conventional STSG technique and were retrospectively selected for comparative defects. The contralateral upper extremities also served as controls. Standardized functional measures were obtained for donor and contralateral extremities.

In long-term follow-up (5/10 patients, minimum 1 year), superior aesthetic outcomes were truly remarkable. Objective functional outcome measures demonstrated a minor comparative advantage for the described technique.

A vast improvement in donor-site aesthetics was consistently reproducible, with minor improvements in functional outcome measures. The composite grafting technique significantly refined the state-of-the-art in donor-site reconstruction for patients undergoing transfer of the radial forearm free flap.

Evaluation of the Mechanism of Vascular Endothelium Growth Factor on Improving Ischemic Flap Survival in Rats. Tanya M. Oswald, Feng Zhang, Yi Pang, Man-Ping Lei, and William C. Lineaweaver. University of Mississippi Medical Center, Jackson, MS.

Partial necrosis of pedicle skin flaps, believed to be due to inadequate blood perfusion and ischemia-reperfusion injury, remains a significant problem in plastic and reconstructive surgery. Recently, the potential of therapeutic agents, such as a variety of growth factors, to improve the viability of ischemic skin flaps has aroused considerable interest. Vascular endothelial growth factor (VEGF), a heparin-binding glycoprotein, is a potent endogenous stimulator of both angiogenesis and vascular permeability. In this study, the authors examined the effect of exogenous VEGF on ischemic flap survival and on the regulation of cytokines in a rat skin flap model.

Fifty-eight male Sprague-Dawley rats were used. A caudally-based dorsal musculocutaneous flap, measuring 3×10 cm, was the flap model. The study was divided into two parts. In part 1, exogenous VEGF (mg/mL) was injected subdermally into the flap of 14 rats before the flaps were sutured back into place. The same number of rats with flaps received saline injection and served as controls. Skin paddle survival was measured 5 days postoperatively. In part 2, biopsies were taken from the flaps treated with VEGF and saline, and were obtained at 2.5, 5.5, and 8.5 cm from the distal edge at 12 and 24 hr after the flaps were sutured. Gene expressions of IGF-1, TGF-β, bfGF, PDGF, IL-1, TNF-α and inducible nitric oxide synthase (iNOS) were measured.

Subdermal injection of exogenous VEGF to the flap induced angiogenesis and significantly improved survival of the flap (89% of the total skin paddle, p<0.05), compared to the control group (64% survival, p<0.05). The expressions of TNF-α and iNOS in the distal part of the flap treated with VEGF were significantly decreased, compared to the control group at 12 and 24 hr postoperatively.

Administration of exogenous VEGF can significantly enhance survival of ischemic flaps. VEGF can also protect the flap from ischemia-reperfusion injury through the regulation of proinflammatory cytokines and inhibition of cytotoxic nitric oxide production.

Complex Facial Reconstruction with the Use of Prelaminated Free Flaps. George Psaras and Carla Norval. University of the Witwatersrand, Rivonia, South Africa.

Complex central facial defects are often difficult to manage due to the extent of the injury and the multi-planar nature of the deficit. The aim of this study was to show that such complex defects can be managed effectively with prelaminated free flaps such as the radial forearm and fibula flaps.

During the period from September 1999 to June 2001, four patients with severe and complex facial defects were treated. The causes of injury were gunshot (2) and electrocution and burns. The facial structures damaged included the mandible, nose, upper and lower lip, floor of the mouth, hard palate, forehead, and cheeks. Reconstruction was achieved with the use of prelaminated free flaps, mainly the radial forearm and fibula. The tissues used for the prelamination were skin graft and conchal cartilage (where appropriate). The techniques used in each case were illustrated. Experience derived from prelaminating fibula flaps was discussed.

All flaps survived and patients achieved satisfactory aesthetic results, taking into account the extent of their injuries. Reconstruction of structures like the nose and upper and lower lips requires bilaminar flaps (lining and cover). Prelaminating flaps such as the radial forearm and fibula (skin paddle) enables the creation of “thin” bilaminar flaps, ideal for such reconstructions. The need for a further flap (free or local) is thus eliminated.

Autogenous Ear Reconstruction: Prefabrication and Prelamination in a Rat Model. Michael W. Neumeister, Tammy Wu, and Bradon J. Wilhelmi. S.I.U. School of Medicine, Springfield, IL.

Prefabrication and prelamination of flaps and tissue engineering provide an enriched armamentarium in difficult reconstructions. This study employed these techniques to create a vascularized autogenous ear construct that is readily transferable on its own vascular pedicle.

Fifteen male Wistar rats (250 to 350 g) were anesthetized. An incision was made over the right lower abdominal wall. A pocket was formed by blunt dissection just below the panniculus carnosus. A separate incision was made over the right femoral vessels, which were then isolated and transected distally. The vessels were transposed in a subcutaneous plane to the abdominal wound. A silicone mold in the shape of an ear (2×1.5 cm) was placed over the transposed vessels in the abdominal wound pocket. The mold and the vessels were secured with 6-0 Vicryl sutures. Auricular cartilage was minced, washed, and cultured using standard chondrocyte culturing techniques. After 14 days, the chondrocyte culturing was complete, and a vascularized capsule based on the incorporated, transposed femoral vessels was formed. The abdominal incision was then reopened, an incision made in the lateral capsule, and the cultured chondrocytes introduced into the molded capsule. Study groups included capsules filled with chrondrocytes only, chondrocytes and a fibrin glue carrier, and fibrin glue only. The capsule was closed and the wounds sutured. The prefabricated, prelaminated construct was isolated on its vascular pedicle 14 days later. The pedicle was transected, the flap transferred to a heterotopic location, and the vessels microsurgically anastomosed. Blood flow to the ear construct was reestablished. Histologic analysis was carried out.

All 15 capsules were completely vascularized and could be reliably isolated on the transposed femoral vessels. The pedicle, incorporated directly into the capsule, provided the dominant blood supply to the construct. None of the capsules with fibrin glue only retained any shape and were devoid of cartilage. Similarly, there was no evidence of retained cartilage in the capsules filled with chondrocytes alone. All capsules with the chondrocytes and fibrin carrier had mature-shaped cartilage preserved.

The authors have shown that transposing a vascular pedicle to a subcutaneously placed silicone block will result in a vascular capsule that can be mobilized and transferred, based solely on the pedicle. While the capsule provides vascularity to the chondrocytes, the cultured cartilage will fill the shape of the silicone mold only if an appropriate carrier, such as fibrin glue, is employed.

Deep Inferior Epigastric Perforator Flaps in Head and Neck Reconstruction. Eamon S. Beausang, Erik Ang, Joan Lipa, Dale H. Brown, Jonathon Irish, Patrick J. Gullane, Peter Neligan, and Ralph Gilbert. Division of Plastic Surgery and Head and Neck Program, Toronto, Ontario, Canada.

Microvascular free tissue transfer has become a well-established technique in head and neck cancer reconstruction, offering a single-stage definitive procedure. Defects requiring soft-tissue bulk have been reconstructed in the past using a variety of flaps, including skin and subcutaneous tissue flaps, such as scapula and lateral arm flaps. Muscle flaps have also been used; however, the long-term results can be unpredictable, due to muscle atrophy.

The deep inferior epigastric perforator (DIEP) flap is a relatively new flap, developed as a modification of the TRAM flap, and has been used predominantly in breast reconstruction. It can be harvested in a vertical or horizontal direction and, unlike the TRAM flap, avoids sacrifice of the rectus muscle. Patient repositioning is not required, and flap harvest can be performed at the same time as tumor ablation. Dissection of the perforators can be technically difficult, and there is a definite learning curve with the technique. Despite this, a long pedicle can be harvested, thus facilitating flap inset in the head and neck region. The authors believe that the DIEP flap is a useful flap in the armamentarium for head and neck reconstruction. They described their experience with 12 cases.

Comparison of the Swallowing Function between Free Jejunum and Colon Flaps for Reconstruction of the Cervical Esophagus. Hung-Chi Chen, M.H. Cheng, and Yueh-bih Tang. Chang Gung Memorial Hospital and National Taiwan University Hospital, Taipei, Taiwan.

Following wide excision of cancers in the larynx and pharynx, there are two ways of reconstruction using free bowel transfers: microvascular transfer of the free jejunum flap for cervical esophagus with an accessory loop for voice reconstruction; and the free ileocolon flap with ascending colon for reconstruction of the pharynx and cervical esophagus, with a segment of ileum for voice reconstruction. The swallowing function is important to the patient's survival, and should be carefully evaluated to find guidelines for selection of the procedure.

From 1995 to 2002, 40 patients with cancers of the larynx or pharynx were studied. After total pharyngolaryngectomy, 30 of them were reconstructed with the first method, and 10 with the second method. The cases selected did not have recurrence with a least follow-up time of 10 months. The swallowing function was evaluated by the following methods: recording of nature of diet, time for each meal, and progression in a postoperative training course; weight gain; esophageal transit scan using radionuclides; and esophagograms.

All patients were instructed to start oral intake at 3 weeks following reconstruction. All the patients with jejunum transfers could eat a solid diet properly after training in swallowing for 2 weeks. However, in the colon group, it took 1 month of training on average, before they could take a semisolid diet. Initially, the colon group complained of food stasis in the neck after oral intake. This gradually disappeared, but the speed of eating was much slower than in the jejunum transfer group (42 min per meal in the colon group, compared to 28 min per meal in the jejunum group). Only four patients in the colon group could eventually take a solid diet. Esophagograms showed patency in both groups, except for one case of mild stricture at the junction of jejunum and thoracic esophagus.

The swallowing function was better in the jejunum group than in the colon group. (Although ileocolon transfer provides better patient satisfaction in voice function in another study of the patients.) However, if the resection of tumor is high up in the oropharynx, colon transfer is still a reasonable option because of a good size match between the oropharynx and colon.

Recipient Vessel Analysis for Microvascular Reconstruction of the Head and Neck. Maurice Nahabedian, Gene Deune, Navin Singh, and Anthony Tufaro. Johns Hopkins University, Baltimore, MD.

The selection of suitable recipient vessels for microvascular reconstruction of the head and neck region is important for flap survival. The purpose of this study was to evaluate a set of factors related to the recipient artery and vein, and to determine how these factors influence flap survival.

This was a retrospective review of 82 patients over a 4-year consecutive period. Included were 56 men and 26 women, with a mean age of 54 years (range: 10 to 85 years). Indications for microvascular reconstruction included tumor ablation (59), trauma (12), and other (11). Flaps included the latissimus dorsi (21), radial forearm (18), rectus abdominis (15), fibula (10), parascapular (6), lateral arm (5), and other (7). Recipient arteries included the facial (30), superficial temporal (19), superior thyroid (17), carotid (5), and other (11). Recipient veins included the jugular system (30), facial (24), superficial temporal (19), superior thyroid (6), and other (3). Factors considered relevant to the vessels were analyzed and included diabetes mellitus, tobacco use (10 cigarettes/day), prior irradiation, timing of the reconstruction, use of a vein loop or graft, and type of vascular anastomosis. Statistical analysis was performed with logistic regression analysis.

Factor analysis demonstrated diabetes mellitus in 7 (8.5%), tobacco use in 18 (22%), immediate reconstruction in 62 (76%), delayed reconstruction in 20 (24%), vein loop in 6 (7%), and vein graft in 2 (2.4%). The arterial anastomosis was end-to-end in 77 (94%) and end-to-side in 11 (14%). Total flap survival was obtained in 77 of 82 (94%). Flap loss was due to venous thrombosis in 4 and to arterial thrombosis in 1 case. Statistical analysis demonstrated that anastomotic failure was associated with a vein loop (2 of 5, p=0.03), venous end-to-side (1 of 5, p=0.004), and tobacco use (3 of 5, p=0.03). Anastomotic failure was not related to diabetes, mellitus, previous irradiation, method of arterial anastomosis, and timing of reconstruction.

Flap failure in the head and neck may be minimized by use of vein grafts when pedicle length is insufficient. Vein loops are associated with increased failure. Patency of the venous anastomosis is increased with end-to-end, rather than end-to-side technique. Tobacco use should be avoided prior to microvascular reconstruction of the head and neck.

Intraabdominal Prefabrication of Composite Colo-Jejunal or Gastro-Jejunal Conduits for Total Esophageal Reconstruction. William Samson, Manjit Bains, and Peter G. Cordeiro. Memorial Sloan Kettering Cancer Center, New York, NY.

It has been well-established that microvascular free jejunal transfer is the method of choice for reconstruction of the hypopharynx and the cervical esophagus. On the other hand, gastric pull-up and colon interposition are the preferred methods for reconstruction of the thoracic esophagus. There are rare instances when neither the stomach nor an adequate segment of colon is available to provide appropriate length for a tension-free proximal anastomosis. The purpose of this paper was to illustrate a technique of composite esophageal reconstruction that may be employed when conventional methods are inadequate.

Three cases of composite esophageal reconstruction were reviewed. In these difficult cases, the limited colon or gastric segment was used as a pedicle graft to reconstruct the distal, thoracic esophagus, while the more proximal reconstruction was composed of a free jejunal transfer. Two patients had colo-jejunal reconstruction, while one patient underwent esophagoplasty combining a jejunal free graft with a short gastric tube. In all of these cases, the esophageal reconstruction was simplified by performing the colo-jejunal or gastro-jejunal anastomosis intraabdominally, prior to division of the jejunal vascular pedicle. This key maneuver eliminates the challenge of performing a retro-sternal, intra-thoracic anastomosis during the ischemic period. The composite conduit can then be passed through the mediastinum as a single unit, and the proximal jejunal and vascular anastomoses may then be performed.

All patients ultimately progressed to a regular diet. There were no instances of flap loss and no evidence of stricture. One patient required reoperation in the early postoperative period for mechanical obstruction of the jejunum at the thoracic inlet. This was resolved with partial sternectomy. The same patient developed a cervical fistula that resolved spontaneously with expectant management. Another patient was admitted 3 months after surgery for nonoperative management of an intraabdominal partial small bowel obstruction.

The authors demonstrated that this one-staged, composite, retrosternal approach is a safe and reliable method for restoring esophageal continuity. Prefabrication of the composite colo-jeunal or gastro-jejunal conduit within the peritoneal cavity is the key maneuver. This minimizes ischemic time and allows for a technically easy distal anastomosis within the abdominal cavity, prior to retro-sternal transfer.

Adjunctive Gut Revascularization for Traditional Enteric Surgery. Navin Singh and Maurice Y. Nahabedian. Johns Hopkins University School of Medicine, Baltimore, MD.

Gut revascularization is a versatile technique that extends the reach and versatility of enteric bypass and establishment of enteric continuity. This was demonstrated in three patients.

C.M. was a 52-year-old man with esophageal discontinuity related to esophageal cancer resection and stricture formation. He had an enteric angiogram that revealed that the inferior mesenteric artery was occluded, thus excluding the possibility of a colon transposition. His stomach was so extensively scarred in previous surgical endeavors that a gastric pull-up was also not feasible. A long roux-en-y loop was constructed of the jejunum to reach the cervical esophagus. During this procedure, the distal third of the jejunal “flap” became venously congested. Through an open sternotomy, the internal mammary artery and vein were delivered down, and the gut was revascularized via a mesenteric arcade. Bowel peristalsis resumed on the table and the gut pinked up. A follow-up esophagram revealed enteric continuity.

D.F. was a 66-year-old woman in whom a gastric pull-up was performed, and the stomach became disvascular. It was revascularized into the internal mammary artery and vein via a gastro-epiploic arcade, with immediate salutary effects. Long-term follow-up was provided with a cine-swallow esophagram demonstrating normal patency, peristalsis, and enteric passage of the food bolus.

J.M. was a 52-year-old woman who underwent roux-en-y jejunal reconstruction for esophageal discontinuity. Based on the institutional experience of the first two patients, plastic surgery consultation was obtained prior to surgery, and a planned revascularization to the internal mammary vessels was undertaken.

The authors concluded that gut revascularization should remain in the armamentarium of plastic surgeons and should be discussed with the general and thoracic surgeons, so that they are aware of this possibility.

Role of the Microsurgeon in the Treatment of Patients with Rhabdomyosarcoma of the Head and Neck. Claire L.F. Temple and Howard N. Langstein. M.D. Anderson Cancer Center, Houston, TX.

Rhabdomyosarcoma (RMS) of the head and neck is characterized by skull-base involvement and a propensity for recurrence. Fortunately, with modern chemoradiation protocols, children enjoy a 70% 5-year survival without radical surgery. However, microsurgical reconstruction may be needed at some point in the care of RMS patients, particularly in those with recurrent tumors requiring radical resection, and in those suffering long-term consequences of prior radiotherapy. The purpose of this study was to critically evaluate a series of RMS patients, in order to define the role of the reconstructive plastic surgeon.

The prospective database revealed 17 patients with RMS of the head and neck seen in the plastic surgery clinic at M.D. Anderson Cancer center from 1990 to 2002. Charts were reviewed for demographic information, prior therapy, extent of surgical resection, type of flap(s) used, flap complications, number of recurrences, and overall survival.

Thirteen of the 17 patients were reconstructed after radical resection for extensive primary or recurrent RMS. The mean patient age at diagnosis was 14.6 years (range: 2.5 to 50 years). Eight patients had had at least one previous recurrence. The mean radiation dose administered prior to reconstructive surgery was 58 Gy. Five patients had skull-base resections. A total of 19 flaps were used in these 13 patients, including 11 free rectus abdominis myocutaneous flaps, two free radial forearm flaps, one free latissimus dorsi myocutaneous flap, one free fibula osseous flap, two pedicled temporalis muscle flaps, and two pedicled temporoparietal fascia flaps. Two flaps failed secondary to venous thrombosis, with an overall flap survival rate of 89%. Three patients developed at least one post-reconstructive local recurrence. Six patients were alive at a mean of 3.4 years following reconstruction (range: 0.1 to 10.3 years).

Four of the 17 patients were evaluated for consequences of prior radiotherapy, including radiation-associated sarcoma (1), mandibular osteoradionecrosis (1), and facial deformity (2). One patient had a skull-base resection. A total of two flaps were used in these four patients, including one free rectus abdominis myocutaneous flap and one free scapular flap. Two patients were awaiting surgery for functional and cosmetic correction.

The microsurgeon's role in the treatment of patients with head and neck RMS is challenging, as it commonly involves skull-base reconstruction in previously irradiated patients. Furthermore, these patients are typically young, are troubled with cosmetic and functional sequelae of childhood irradiation, and are prone to recurrences requiring future resection and multiple flap reconstructions.

Bleeding after Microsurgical Free Flap Reconstruction in the Head and Neck. Cheng-Hui Lin and Fu-Chan Wei. Chang Gung Memorial Hospital, Taoyuan, Taiwan.

Microsurgical free tissue transfer has become a standard reconstruction method after head and neck cancer resection. Without proper management, postoperative bleeding and hematoma formation may cause wound infection, flap necrosis, and even massive bleeding from ruptured major vessels.

From January, 1997 to December, 2001, 2015 patients with head and neck cancers underwent microsurgical free tissue transfer after tumor resection. Retrospective chart review identified 41 consecutive cases complicated by postoperative bleeding or hematoma accumulation requiring neck reexploration (2.0%).

Among these 41 cases, there was only one female. The median patient age was 56 years (range: 36 to 79 years). The cancer resection involved the tongue (8), buccal (15), lip (4), mouth floor (2), gum (4), retromolar (3), and pharynx (3). A higher incidence of bleeding and hematoma formation was noted in the more advanced tongue and buccal cancer patients. In the 8 patients with tongue cancer, 5 were T3 cases, and 3 were T2 cases. In the 15 patients with buccal cancer, 8 were T4 cases, 4 were T3 cases, and 3 were T2 cases. The bleeding was found from the flap itself (12), raw surface of reconstruction beds (8), branches of pedicle (8), and vessels other than the pedicle (13). Possible etiologies were a bleeding tendency due to abnormal liver function (5) or massive intraoperative transfusion (2), vessel clip dislodgement (13), inadequate anastomosis (1), infection over the deep neck space (7), and suction erosion of the vessel (1).

Four cases required microsurgical reanastomosis. Four cases developed marginal and partial necrosis of the flaps, and two of them received local flap reconstruction. Wound dehiscence was noted in two cases. Three cases were complicated by infection. Repeated bleeding was noted in two cases.

Bleeding in the head and neck region after microsurgical free tissue reconstruction is not uncommon. Good hemostasis and careful application of the clips are most important in prevention of postoperative bleeding. Severe coughing or pain should be avoided postoperatively. Whenever significant bleeding or hematoma in the head and neck region occurs, prompt action should be taken to avoid subsequent major complications.

Microvascular Salvage of Failed Cranioplasty in a Radiated Field. Hugo St. Hilaire, Jin Chun, Michael Christy, and Lester Silver. Mount Sinai Medical Center, New York, NY.

The goals of cranioplasty include protection of intracranial content and dural repair, as well as reestablishment of preoperative cosmesis. Alloplastic materials are often used to correct calvarial defect, but have a lower success rate when used in a radiated field. Failure may present as a chronic fistula draining cerebrospinal fluid, osteomyelitis of the calvarium, brain abscess, meningitis, or seizures. The hypocellular, hypovascular, and hypoxic environment produced by radiation therapy offers a challenge to the reconstructive surgeon. Microvascular tissue transfer is the ideal therapeutic measure in this patient population. Furthermore, in large areas where the dura is exposed, a free flap is a durable reconstruction and can also aid in enhancing dural repairs.

The authors presented five cases in which six free flaps were used to reconstruct failed alloplastic cranioplasties following radiation therapy. In two of these cases, the flaps were used in conjunction with a titanium mesh cranioplasty. Three radial forearm free flaps were used, two of which were in the same patient who had a recurrence of malignancy. The latissimus dorsi flap was used on two occasions, while the last patient was treated with a serratus anterior flap. Complete coverage of the wound was obtained in all patients with no long-term complication.

The latissimus dorsi is an expendable muscle that provides a free flap with a very long dominant vascular pedicle. This allows for microvascular anastomosis to be performed outside the radiated field. If needed, the anastomosis can be carried out at a lower level without the need for an interpositional vein graft. In addition, the latissimus dorsi free flap provides a good volume of muscle, able to cover large defects and protect a dural repair. The radial forearm flap provides an excellent choice for small defects in which there is no need for osseous reconstruction. The thickness of the forearm flap is comparable to normal scalp. Therefore, there is no excessive bulkiness. The serratus anterior flap can provide vascularized muscle and bone, if needed, on a single long vascular pedicle.

Results indicated that microsurgical free flap transfer allows for immediate, single-stage coverage of wounds previously irradiated, with well-vascularized tissue. The authors discussed in detail their strategy in flap selection, as well as a treatment planning algorithm used in this specific patient population.

Large Volume Cranioplasty with Free Tissue Transfer and Alloplastic Combination. Salvatore C. Lettieri, Maurice Nahabedian, Paul N. Manson, E. Gene Deune, and Bradley C. Robertson. Mayo Clinic, Phoenix, AZ, Johns Hopkins University, Baltimore, MD., and University of Maryland Shock Trauma Center, Baltimore, MD.

Acquired deformities of the calvarium are generally treated with various forms of cranioplasty. The techniques of cranioplasty can address the surface defect, but cannot be reliable for large volume defects. In this reported series, free tissue transfers were utilized for the volume fill with a second stage cranioplasty over the healed flap. The cases were reviewed and a treatment algorithm was presented.

In a retrospective review, four patients presented with large calvarial defects from trauma. Each of them had absent underlying brain, in addition to the missing segment of bone. All patients were well-healed from the original injury and were neurologically stable, although all of them had a neurologic deficit. The defects were studied by three-dimensional computed tomography (CT) and the volume deficit measured. All patients underwent first stage free tissue transfer and, at a second hospitalization, cranioplasty was performed. The volume was filled largely by the free tissue transfer, and an alloplastic material was used to re-create the rigid surface contour.

Pre- and postoperative CT scans were performed on all patients. There were no healing problems, and the flaps were viable at the time of cranioplasty. The volume deficit was measured preoperatively, and then reviewed on the postoperative CT scan. The resultant volume fill by free tissue transfer and the surface area of the alloplast were measured. All patients left the hospital without neurologic change and remained in their preoperative, post-morbid state.

Large volume calvarial defects can be very challenging to repair. Cranioplasties can be performed using alloplastic, cadaveric, or autologous materials. The extent of the volume can be only partially addressed with the materials used for the cranioplasty. In this small series of patients, free tissue transfers were safely used to fill the large volume defects, followed by a secondary cranioplasty. With this experience, the authors presented an algorithm for the treatment of large volume calvarial defects.

Dynamic Reconstruction of Eye Closure by Free Functioning Muscle Transplantation in Comparison to Muscle Transposition in Facial Palsy. Pietro Giovanoli, Chieh-Han John Tzou, Susanne Friedl, and Manfred Frey. Division of Plastic and Reconstructive Surgery, Vienna, Austria.

For patients suffering from facial palsy, lagophthalmus is often a more serious problem than the inability to smile. The vulnerability of the cornea and the continuous need for artificial tears or ointment lead to a restricted quality of everyday living. Reconstruction of eye closure by muscle transposition and free functioning muscle transplantation, respectively, offer good solutions to regain near normal eye protection without the need for implants.

In the last 3 years, 44 patients have been treated for facial palsy, including reconstruction of eye closure, using either a temporal muscle transfer in 16 cases, and a regionally differentiated part of a free gracilis muscle transplant in 10 cases, respectively. The patients were documented by a 3-D video analysis system. The mean patient age was 46.4±17.5 years.

In the 12 patients with temporalis muscle transfer for eye closure, the distance of the upper and lower lid points could be reduced from preoperative 10.2±2.4 mm to 4.1±0.5 mm postoperatively on the paralyzed side, compared to 0.0±0.0 mm preoperatively and 0.0±0.0 mm postoperatively on the non-paralyzed side during closing the eyes as in sleep. In the (territorially differentiated) muscle transplant group, consisting mainly of congenital facial palsies, eye-lid closure could be improved from a lagophthalmus of 6.6±1.5 mm preoperatively to 0.7±0.4 mm postoperatively. In three cases, the resting tonus of the part of the gracilis muscle transplant around the eye increased so much, that muscle weakening by direct excision was necessary in the upper lid.

Temporalis muscle transposition, as well as using a free functional gracilis muscle transplant for reanimation of the eye and mouth at the same time, are reliable methods to reconstruct eye closure with clinically adequate results. Temporalis muscle transposition guarantees an innervation completely independent of the facial nerve system. Therefore, synkinesia will not develop. This seems more important than the possibility of emotional coupling, which can be achieved by reinnervating part of the gracilis muscle transplant by an additional cross-face nerve graft. Therefore, temporalis transposition is the first choice in adult patients. In children, the authors prefer to use the free muscle transplant for eye closure, because they do not want to interfere with the growth of the face by transposing a local muscle, and because they believe in a higher degree of central plasticity in children.

Maintenance of Secure Fixation of Free Muscle Flaps for Facial Palsy. R. Jobe Fix. University of Alabama at Birmingham, Birmingham, AL.

Free innervated muscle transfer is currently the most effective technique to restore motion of the lower face and the lip, to create the dynamics of a smile. One of the problems associated with free muscle transfer is dehiscence of the muscle from its points of fixation, with resultant compromise of transferred muscle function.

There are three techniques that have been developed to improve muscle fixation: use of the GIA stapler across the ends of the gracilis muscle; use of the Mitek suture anchor into the zygoma; and placement of an internal splint that is removable at 6 weeks postoperatively. The use of the GIA stapler provides a double row of staples across each end of the muscle. This allows for a stable end of the muscle into which sutures can be passed, and can prevent “cheese wiring.” This method also allows for radiologic marking of the positions of the muscle ends, and can allow verification of intact fixation of the muscle near the lip. The use of the Mitek suture and its placement into the areas of the zygoma was reviewed. The placement of a pull-out Prolene suture from the commissure to the parietal skin is used for a period of 6 weeks to protect the muscle while it heals into the points of fixation at the zygoma and commissure level.

These techniques were used in 6 patients with facial palsy, one patient with Moebius syndrome, and one patient with Volkmann's ischemic contracture. A description of the techniques was presented, as well as pre- and postoperative videos displaying successful function.

Long-Term Follow-Up of Hypoglossal-Facial Crossover in Facial Nerve Palsy: Pure End-to-Side Anastomotic Technique. Kyung Suck Koh and San Yoon Kim. Asan Medical Center, University of Ulsan, Seoul, South Korea.

Hypoglossal-facial crossover is the most popular method of reconstructing facial palsy resulting from proximal facial nerve injury near the brain stem. Conventional hypoglossal-facial crossover involves performing a partial hypoglossal nerve section or incision and using an interpositional nerve graft to bridge the gap between the two nerves, sometimes resulting in hemiglossal atrophy and its sequelae. In an attempt to solve these problems, the authors performed “pure” end-to-side anastomosis between the hypoglossal and facial nerves in seven patients with facial palsy. Facial reanimation was achieved after 8 months. The longest follow-up was 5 years. With time, patient facial animation has been further improved. Despite the small number of cases, the authors believe that this technique of hypoglossal-facial crossover with “pure” end-to-side anastomosis, and mobilization of the intratemporal facial nerve, can decrease donor nerve morbidity in facial nerve rehabilitation.

Aberrant Reinnervation of Facial Paralysis and Its Surgical Treatment. David Chwei-Chan Chuang. Chang Gung Hospital, Taipei, Taiwan.

Aberrant reinnervation of the facial nerve is a complication of facial recovery after paralysis, causing postparetic facial synkinesis. This is the presence of unintentional movements in another area of the face, when intentionally performing movements in one area of the face, and includes dominant oculo-to-oral synkinesis (most common) or its reverse, oro-to-ocular synkinesis, and less common oculo-to-neck, oro-to-neck, and oro-to-forehead synkinesis. Most patients are most concerned with oro-ocular synkinesis.

Between 1986 and 2000, 116 patients with facial paralysis were treated with functioning free muscle transplantations. Seven cases were due to aberrant reinnervation of the facial nerve. Five cases involved sequelae to Bell's palsy, and two were traumatic sequelae. The strategy of reconstruction for this type of facial paresis is a three-stage procedure. In stage 1, a cross-face nerve graft is used. In a stage 2 procedure, 1 year after stage 1, the procedure removes the zygomatic and buccal branches of the injured facial nerve and its innervated facial muscles, related to upper lip movement, as extensively as possible. This surgery may cause mouth movement in the intact side, but does not cause oro-ocular synkinesis. In stage 3, 1 week after stage 2, restoration of facial smiling is performed by functioning free gracilis muscle transplantation, innervated by cross-facial nerve grafting.

The seven patients have had at least 1 year follow-up, and results are encouraging. The procedures involve careful patient explanation preoperatively, a good knowledge of the anatomy of the facial nerve branches and muscles, and experience and confidence in free muscle transplantation for facial reanimation in these complex reconstructions.

Expression and Distribution of NOS Gene and Protein in Reperfused Skeletal Muscle. Wen-Ning Qi, Long-En Chen, Anthony V. Seaber, and James R. Urbaniak. Duke University Medical Center, Durham, NC.

The authors' previous studies suggest an ischemia/reperfusion (I/R)-induced NOS change in skeletal muscle. However, no documentation has been provided for which of the three NO synthesis pathways (neuronal, inducible, and endothelial, i.e., nNOS, iNOS, and eNOS) is involved and at what level changes occur. This study was designed to identify gene and protein expression of individual NOS isoenzymes and to characterize their distribution in rat skeletal muscle subjected to I/R injury.

One side of the extensor digitorum longus (EDL) muscle from 20 adult rats underwent 3hr/3hr I/R. The opposite EDL was used as a control. Quantitative RT-PCR and Western blot technique were used to determine expression of NOS mRNA and protein. The relative expression of NOS mRNA and protein in the reperfused EDL muscle was determined by comparison with the normal side (as 100%). Statistical significance was determined by ANOVA.

RT-PCR showed that neuronal (n-), inducible (I-), and endothelial (e-) NOS mRNAs were each expressed in the normal EDL. I/R up-regulated nNOS to 116±22% (mean±SD) of normal and iNOS to 5682±180%, while eNOS remained at normal level (100±22%) in the EDL. There was significant upregulation in iNOS mRNA. Western blot detected greater eNOS (136±28%) and iNOS (384±137%), but less nNOS (63±15%) in the I/R EDL, with a significant difference compared to normal. Immunolocalization showed nNOS antigens were prominent in the muscle sarcolemma in both normal and reperfused EDL and also in the cytoplasm in reperfused EDL. eNOS expressed in vascular endothelial cells in normal and reperfused EDL, and also expressed in the cytoplasm of some muscle fibers in reperfused EDL. iNOS antigens were not found in normal EDL, but were present in the muscle cytoplasm in reperfused EDL.

The results demonstrated that I/R mediated not only expression changes at transcriptional and translational levels, but also the distribution changes of proteins of all three NOS isoforms in rat skeletal muscle. Because each NOS has distinct functions, these changes may imply the mechanisms of I/R injury. Hugely increased iNOS mRNA and protein during reperfusion could affect many cellular functions, to lead to cellular damage or necrosis. It is conceivable that an understanding of this modulation on a cellular and molecular level under different conditions may lead to an understanding of I/R injury observed in vivo, and to improvement of the clinical outcomes in reperfusion injury.

Mesenteric Revascularization Using a Superiorly-Based Transperitoneal Deep Inferior Epigastric Artery. Kenneth J. Moquin, Vigen B. Darian, Mary Grzybowski, and L. Scott Levin. Duke University Medical Center, Durham, NC, Henry Ford Hospital, and Wayne State University, Detroit, MI.

These authors confirmed the anatomic feasibility of using a pedicled deep inferior epigastric artery, with its attached rectus abdominis muscle, to revascularize the mesenteric circulation, by studying three human cadavers (fresh and formalin-preserved) and two dog cadavers (fresh). They then evaluated the physiologic efficacy of the mesenteric revascularization in dogs.

Using three large (50 lb) female dogs as an animal model for acute mesenteric ischemia, they performed laparotomies and divided the deep inferior epigastric artery and rectus abdominis muscle caudally in each dog. The muscle was then transposed into the peritoneal cavity on its superiorly-based vascular pedicle. Under loupe magnification (×4.5), the deep inferior epigastric artery was then anastomosed to the middle colic artery in an end-to-side fashion.

Intraoperative ultrasonic blood-flow measurements demonstrated flows of 365-370 mL/min in the superior mesenteric artery (SMA), 65-70 mL/min in the ileocolic artery, 25-30 mL/min in the middle colic artery, and 5-10 mL/min in the deep inferior epigastric artery bypass. Approximately 2 weeks later, a repeat laparotomy was performed, and the SMA was ligated at its origin. Acute SMA ligation in dogs produces acute mesenteric ischemia and is normally lethal within 24 hr. However, the dogs survived between 3 days and 5 months. Angiography confirmed complete ligation of the SMA but did not demonstrate the epigastric artery bypass. This was likely due to equipment and technique limitations. No abdominal-wall hernia or intestinal volvulus was present at autopsy.

This pilot study demonstrated the anatomic feasibility in humans and physiologic efficacy in dogs, of a superiorly-based transperitoneal deep inferior epigastric artery bypass for mesenteric revascularization. The results suggested that this bypass may prove to be a viable, less invasive treatment option for chronic mesenteric ischemia, where minor supplemental blood flow may alleviate the clinical symptoms associated with enteral nutrition, and prevent the need for extensive aorto-visceral bypass procedures.

Implantable Doppler Probe for Continuous Monitoring of Hepatic Artery and Portal Vein Blood Flow in Pediatric Liver Transplantation. Liza C. Wu, Loren S. Schechter, Robert F. Lohman, Somchai Limsrichamrern, Charles G. Winans, J. Michael Millis, and David C. Cronin. University of Chicago Hospitals, Chicago, IL.

Acute vascular thrombosis is a potentially fatal complication in solid organ transplantation. Pediatric liver transplantation is associated with a high incidence of hepatic artery thrombosis. Current methods of surveillance for vascular thrombosis are indirect and do not provide early detection. These authors have adopted the use of the 20-MHz implantable Doppler probe to assess hepatic artery and portal vein blood flow in the immediate postoperative period after pediatric liver transplantation.

Four consecutive pediatric liver transplants were performed between January and July, 2001. All four patients had hepatic artery anastomoses constructed with the aid of an operating microscope. The implantable Doppler probe was attached to the hepatic artery and portal vein distal to the anastomoses, using topical fibrin sealant. During the postoperative period, the presence, quality, and character of the Doppler signal from the hepatic artery and portal vein were assessed for 5 to 7 consecutive days. In all four patients, the implantable Doppler probe was successfully placed. Monitoring was easily learned and done by staff.

One patient developed hepatic artery thrombosis, which was accurately detected by the implantable Doppler probe. The patient was reexplored, the problem addressed, and the graft salvaged. All patients had their probes removed without complication, and were discharged from the hospital.

The implantable Doppler probe provided direct, real-time monitoring of hepatic artery and portal vein blood flow. Signal quality and character were easily assessed by physician and staff, and reliably reflected intravascular flow. This allows for early detection and intervention in microvascular complications of pediatric liver transplantation.

Abdominal-Wall Reconstruction with the Free Tensor Fasciae Latae Myofasciocutaneous Flap Using Intraperitoneal Gastroepiploic Recipient Vessels. Navin Singh and Pierre Chevray. Johns Hopkins University School of Medicine, Baltimore, MD, and University of Texas M.D. Anderson Cancer Center, Houston, TX.

The authors reported their experience with a method for definitive abdominal-wall reconstruction using the free tensor fasciae latae myofasciocutaneous flap anastomosed to the intraperitoneal gastroepiploic vessels.

This is a single-stage reconstruction capable of reliably reconstructing a full-thickness defect involving any region of the abdominal wall. The fascial component of the flap reconstructs the abdominal wall with like tissue, including epimysium, to minimize postoperative bowel adhesions, and the cutaneous portion of the free tensor fasciae latae provides durable and aesthetically acceptable external cover. Very large flaps can be harvested for defects up to one-third of the abdominal wall, and have the potential to be neurotized for function. The intraperitoneal gastroepiploic recipient vessels allow for an uninterrupted tight fascial closure, restoring structural integrity to the abdominal wall that is not topologically possible with extra-peritoneal recipient vessels. These recipient vessels, up to 10 cm in mobile length, once dissected and 2 to 3 mm in diameter, are conceptually superior because they are located immediately adjacent to the deep surface of the defect, and should be preferred even when the extra-peritoneal vessels are adequate.

The authors presented a series of three cases of full-thickness upper and lower abdominal-wall reconstruction utilizing this method for oncologic reconstruction. Prior abdominal surgery was not found to be an impediment. The technique is extendable to previously unreconstructable hernias.