Abstracts

January 12-15, 2002

Cancun, Mexico

Computer-Assisted Sensorimotor Testing in Staging Cubical Tunnel Syndrome. Oscar C. Aszmann, Manfred Frey, Isler Can, and Veith Moser.

Cubital tunnel syndrome is the second most common chronic nerve entrapment of the upper extremity; yet, both diagnosis and staging of the severity of the progression of the disease rely mostly on the keen observation and interpretation of clinical signs and symptoms. Nerve conduction studies offer little help, since in more than 30% of cases, they will indicate false positive or false negative test results; they do not provide any basis for staging chronic peripheral nerve compression. To be valid, a staging system must correlate well with the known pathophysiologic mechanisms of chronic nerve compression, and objective parameters must be available to quantify differing degrees of sensory and motor dysfunction, and must allow for different therapeutic consequences.

In this reported study, the authors evaluated 20 normal individuals and 20 patients who presented with the clinical diagnosis of cubital tunnel syndrome. Classic two-point discrimination (2PD), one-point pressure threshold, two-point pressure threshold, pinch and grip strength, were measured with the use of a computer-assisted testing system. All individuals were then stratified according to age (above and below 45 years), absence or presence of disease, and the predictive value of each testing modality evaluated according to 99% confidence intervals. Progression of the disease was staged according to loss of sensory and motor function. While abnormal two-point discrimination was indicative of severe nerve compression, early stages of nerve compression were delineated along the 2 PD pressure threshold continuum. Similarly, the gradual loss of pinch and grip strength to wasting and atrophy of intrinsic musculature was used to differentiate between mild, moderate, and severe degrees of peripheral nerve dysfunction.

While the prognostic value of the proposed staging remains to be validated, the results of this study underline the high diagnostic value of sensorimotor testing in chronic nerve compression.

Accuracy of Radial Tunnel Injections Using a Cadaver Model. Robert E. Meehan, Stephen P. DeSilva, John L. Andary, and Gregory R. Hill.

The injection of a local anesthetic and cortisone into the radial tunnel can be helpful in the treatment of radial tunnel syndrome and in discriminating it from lateral epicondylitis. A description or analysis of the technique has not been reported, to the authors' knowledge. They described and compared two methods of injecting a dyed tracer in a cadaver model.

Eighteen paired unembalmed forearms were injected posterolateral with 1 ml of 0.5% methylene blue, at 3 cm (9 specimens) or 4 cm (9 specimens) from the radiocapitellar joint. A successful injection was defined as a stained posterior interosseus nerve (PIN) within the radial tunnel boundaries. A digital picture (Olympus 3030) was taken of the resected radial nerve on a white background. The color intensity (percent density) of the stained and unstained portions was measured using the Image Pro Plus graphics program (Media Cybernetics). The color intensity of the stained nerve reflected its concentration.

Methylene blue was accurately delivered to the radial tunnel in 17 of 18 (94%) specimens. Color intensity measurements between the stained (mean: 74.2) and unstained (mean: 25.8) nerve portions were statistically significant (p<0.001). Comparison between the two injection levels demonstrated the 3-cm site (mean: 78.2) to be statistically more intense (p<0.03) than the 4-cm site (mean: 69.7). No dye migration to the lateral epicondyle was seen in any specimen.

The authors' technique successfully injected dye into the radial tunnel and might be useful as a diagnostic and therapeutic tool in the workup of lateral elbow pain.

Outcomes from 101 Microreconstructions in 99 Patients with Obstetrical Brachial Plexus Paralysis. Julia K. Terzis and Konstantinos Papakonstantinou.

The majority of patients diagnosed with obstetrical brachial plexus paralysis (OBPP) recover spontaneously. However, approximately 20% require microsurgery to obtain acceptable function in their paralyzed upper extremities and to prevent debilitating contractures. In this reported study, the authors presented the outcomes of primary nerve reconstruction and the results of secondary procedures performed in their center between 1978 and 2000 for restoring or enhancing function in the upper extremitiy.

Ninety-nine patients were operated on between 1978 and 2000. Results were analyzed in 84 patients with adequate follow-up. Seventy-five patients underwent 77 primary brachial plexus reconstructions, and 24 underwent only secondary procedures. The mean denervation time for this patient population was 41.1 months. Nerve reconstruction included microneurolysis (n&equals;132), direct end-to-end (n&equals;85) and direct end-to-side neurotizations (n&equals;14), indirect neurotizations with interposition nerve grafting (n&equals;252), and direct nerve-to-muscle neurotizations (n&equals;4). Muscle (n&equals;135 pedicled and 48 free) and tendon (n&equals;80) transfers were used to enhance function.

Results were analyzed in relation to the type of injury (Erb's vs. global paralysis), the severity of the injury, and the denervation time. The diagnostic value of CT/myelography and electromyography was also analyzed. Results of reconstruction showed improvement in all muscles tested at a statistically significant level (p<0.001): good and excellent for 84.8% of biceps, 73.55% of supraspinatus, 71% of deltoid, and 67.8% of triceps restorations. The Mallet scores and the Gilbert-Raimondi scores improved after reconstruction in all patients at a statistically significant level.

In this study, outcomes were better if the number of avulsed roots was fewer. The denervation time (DT) primarily affected the outcome of hand function. Patients with DT less than 3 months underwent fewer surgeries (1.3 surgeries per patient) to complete the reconstruction, compared to patients with DT between 3 and 6 months (3.1 surgeries per patient). This is the largest outcome series in North America operated on by a single surgeon.

Electrical Stimulation of Denervated EDL Muscles of Rats for 15 Weeks Does Not Enhance Recovery of Maximum Force Following Reinnervation. Douglas E. Dow, Paul S. Cederna, Cheryl A. Hassett, Robert G. Dennis, and John A. Faulkner.

Following peripheral nerve injury and repair, residual muscle weakness hinders effective rehabilitation. Recovery is especially poor if reinnervation is delayed for many weeks. Prolonged periods of denervation result in greater muscle atrophy, which may inhibit functional recovery. Electrical stimulation of denervated muscles reduces atrophy. EDL muscles of rats denervated for 17 weeks maintain only 27% of muscle mass and 2% of maximum force of control muscles. In contrast, electrical stimulation of denervated muscles maintains 99% of the mass and 75% of the force. The authors' hypothesis was that electrical stimulation of denervated EDL muscles of rats during 15 weeks of peroneal nerve division, enhances recovery of muscle mass and maximum force following nerve repair.

The EDL muscles of 16 rats received one of four experimental treaments: control-EDL muscle remained innervated (n&equals;6); immediately repaired-peroneal nerve that innervates EDL muscle was divided and immediately repaired (n&equals;10); denervated/repaired-peroneal nerve remained divided for 15 weeks, then repaired (n&equals;7); stimulated/denervated/repaired-peroneal nerve remained divided for 15 weeks while EDL muscle was electrically stimulated (100 Hz, 20 pulses/contraction, 200 contractions/day) by an implanted stimulator; then nerve was repaired (n&equals;9).

Functional evaluation occurred 6 months following nerve repair. Maximal force was measured by supramaximal electrical stimulation of the peroneal nerve, to generate isometric tetanic contractions measured by a force transducer tied to the exposed and divided distal tendons of the EDL muscle. Subsequently, the muscle was harvested and weighed. Reinnervation failed in two muscles, one each from the denervated-repaired and stimulated-denervated-repaired groups. Muscle mass recovered to a higher level in the stimulated-denervated-repaired group than in the denervated-repaired group, but maximum force recovered no better.

The reduced level of atrophy in the denervated EDL muscles that were electrically stimulated during the 15 weeks of nerve division did not enhance recovery of maximum force following nerve repair. The critical issue for recovery following long-term denervation and nerve repair appears to be not just the maintenance of muscle mass and maximum force. Other factors not adequately affected by this stimulation protocol limit recovery.

Predicting Outcome in Obstetrical Brachial Plexus Palsy. Nancy de Kleer, Christine G. Curtis, Derek Stevens, and Howard M. Clarke.

The purpose of this study was to identify factors at an early age which reliably predicted outcome in cases of obstetrical brachial plexus palsy. Currently, lack of spontaneous recovery of elbow flexion at 3 months of age is used as an indication for surgery in centers around the world. This has not been scientifically validated and, currently, no gold standard exists to predict which children will improve with conservative management, and which children will benefit from primary surgery.

Six hundred and four children assessed at the authors' brachial plexus clinic over the last 10 years, were evaluated in a retrospective review of data which had been collected prospectively. Seventy-nine children, all of whom underwent a standardized brachial plexus physical examination at 3 months of age and had a minimum follow-up of 2 years, were analyzed in this study.

Four test scores, based on various combinations of validated physical examination maneuvers, were developed and calculated for children at their 3-month physical examination. One of the test scores was elbow flexion alone, while others incorporated additional movements, such as wrist extension. The 3-month physical examination test scores were analyzed with respect to final functional outcome, regardless of whether or not surgery was undertaken. The ability of the test scores to predict outcome was analyzed with logistic regression, receiver operating characteristic (ROC) curves, and tree-based analyses.

Two of the four test scores were shown with logistic regression to predict benefit, with statistical significance (p&equals;0.01 and p&equals;0.013). Elbow flexion alone did not predict outcome. ROC and tree-based analysis supported these findings.

Outcome in obstetrical brachial plexus palsy can be predicted using two recently developed test scores. Elbow flexion at 3 months of age did not predict outcome. Future studies will evaluate the test scores prospectively, and should yield a formula to calculate which child will benefit from conservative vs. surgical management.

Preservation of Sexual Function after Radical Prostatectomy Using Cavernous Nerve Reconstruction. Joseph J. Disa, Peter G. Cordeiro, James McKiernan, and Peter T. Scardino.

Radical retropubic prostatectomy (RRP) for prostate cancer can result in the loss of sexual function. Bilateral nerve-sparing prostatectomy is the gold standard for the preservation of erectile function. Resection of one cavernous neurovascular bundle results in a decreased chance for the return of erectile function, while resection of both bundles essentially eliminates return of function. Additionally, pretreatment of the disease with chemo-, hormone, or radiation therapy is known to significantly diminish sexual recovery. This study evaluated the role of interposition nerve grafts in the restoration of sexual function after RRP with cavernous nerve resection.

Three hundred ninety-three consecutive patients underwent RRP over a 33-month period. Cavernous nerve reconstruction was performed in patients who reported preoperative potency and underwent resection of one or both cavernous nerves. Nerve grafting was performed at the time of prostatectomy. Patients were evaluated postoperatively using the International Index of Erectile Function (IIEF), a psychometrically validated assessment tool for sexual function.

Eighty-two patients underwent nerve grafting (61 unilateral and 21 bilateral). The mean patient age was 58 years in the nerve grafted patients and 60 years in the patients not receiving nerve grafts. Blood levels of prostate specific antigen, the incidence of pretreatment (chemo-, hormone, or radiation therapy), Gleason pathologic score, and clinical stage were all significantly higher in patients undergoing nerve resection and reconstruction. Nerve graft donor sites were the sural (55), genitofemoral (18), and ilioinguinal (9). The mean graft length was 6.0 cm. IIEF surveys were sent to patients who were at least 6 months postoperative from the procedure. One hundred sixty-nine (45%) responded, with a 56% response rate in patients receiving nerve grafts. Potency (IIEF score >15) was observed in 53/92 (58%) with BNSP; 8/24 (33%) with unilateral NSP; and 0% with both neurovascular bundles resected. Potency was observed in 17/39 (44%) with unilateral nerve graft reconstruction and 2/8 (25%) with bilateral reconstruction (median follow-up: 13 months). Of note, both patients recovering function after bilateral grafting received pretreatment. Evaluation of patients with unilateral reconstruction without pretreatment demonstrated 52% (11/21) potency by IIEF.

The results of this study supported the efficacy of cavernous nerve reconstruction with nerve grafts. A prospective randomized study (particularly with respect to unilateral reconstruction), with the objective measurement of erectile function, is warranted.

Comparison of Microsurgical and Conventional Reconstruction of Transverse and Longitudinal Congenital Hand Deficiencies. Neil F. Jones, James Chang, Prosper Benhaim, and John F. Lawrence.

Children born with congenital absence of the digits due to transverse or severe longitudinal deficiencies or constriction band syndrome, have conventionally been reconstructed using non-vascularized toe phalangeal bone grafts. More recently, microsurgical toe-to-hand transfers have been advocated. The purpose of this study was to compare the conventional technique of non-vascularized toe phalangeal bone grafting with reconstruction by microsurgical techniques by toe-to-hand transfers, analyzing the growth of the transferred bone and hand function.

Seventeen patients who had undergone non-vascularized toe phalangeal bone grafting were compared with a similar group of 17 children who had undergone microsurgical toe-to-hand transfers. Radiographic analysis of growth was evaluated by the appearance of open epiphyseal plates, phalangeal growth over time, and comparison of the transferred toe phalanges with the contralateral foot. The parents evaluated improvement of their children's hand function.

Seventeen patients underwent 19 surgical procedures to transfer 29 non-vascularized toe phalangeal bone grafts harvested extra-periosteally from the second, third, or fourth toes. Mean age at operation was 1.8 years, and mean radiographic follow-up was 4.5 years. Thirty-six percent of growth plates remained open, but no growth of the transferred proximal phalanges was observed over time. Only three parents believed that hand function had improved after toe phalangeal bone grafting (18%). Twenty-one microsurgical toe transfers were performed in 17 children at a mean age of 4.6 years, with a mean follow-up of 2.5 years. Ninety-six percent of epiphyseal plates remained open, and progressive growth over time was seen in 90%. In those patients in whom it was possible to compare radiographs of the contralateral foot, there was equal growth in the transferred toe phalanges. All parents thought that microsurgical toe transfers improved both fine pinch and grasp of large objects.

Despite meticulous extra-periosteal dissection, conventional reconstruction by non-vascularized toe phalangeal bone grafts did not provide any bone growth over time. Conversely, with microsurgical toe-to-hand transfers, growth potential was preserved and was comparable to the corresponding toe on the contralateral foot. Little improvement in hand function was seen after conventional reconstruction, except in those children in whom the toe phalangeal bone grafts were placed distal to the PIP joints. Microsurgical toe transfers into the thumb or ulnar border digit position improved both pinch and grasp function, while double second-toe transfers provided improved three-point pinch and grasp.

Palmar Arch Revascularization for Arterial Occlusion to the Distal Upper Arm. Renata V. Weber, William Suggs, Teresa Benacquista, and Berish Strauch.

During a period of 5 years, seven patients (ages: 42 to 62 years, 52±7, mean ±SD) underwent upper extremity bypass for ischemic changes to the hand not responsive to conservative management. Patients were referred from the vascular department at the authors' institution.

Preoperative angiograms were performed and attempts at angioplasty or intravenous tPA were done when possible. Patients with persistent upper-extremity ischemia and an obvious occlusion with reconstitution in the hand, were candidates for upper-extremity bypass to the palmar arch. All patients had upper-extremity bypasses performed with reverse saphenous vein grafts. The proximal anastomoses (end-to-side) were performed by either the vascular or plastic team, while all distal anastomoses (end-to-side) were performed by one of two plastic surgeons under a microscope, to the deep or superficial palmar arch. Postoperative follow-up ranged from 3 months to 2 years.

The resultant healing period and resolution of ischemic changes were similar to previous reports of upper-extremity revascularization with end-to-end anastomosis to the palmar arch. The bypass graft to the upper arm improved pain and tissue ischemia in all cases. Patients with preoperative ulcers were completely healed by 3 months.

These authors' results agreed with those of previous studies which validated that upper-extremity bypass is effective in healing ulcers present on the hand, associated with decreased blood flow, and in improving symptoms of ischemic changes. However, they also believe that end-to-side anastomosis to the palmar arch offers significant advantages, in that the continuity of the arch is maintained with all possible outflow vessels, eliminating size discrepancy in the vessels, and allowing for greater patency.

Musculocutaneous Radial Artery Forearm/Flexor Carpi Radialis Flap-A Clinical and Cadaveric Study. Alexandrina S. Saulis, Samir S. Sukkar, and Gregory A. Dumanian.

The authors described a modification of an old flap and the creation of a new one, in order to solve some common clinical problems. These flaps utilize the flexor carpi radialis muscle belly and the adjacent radial artery (FCR/RA), either harvested in conjunction with a radial forearm free flap (four patients), or else as a reversed flap without skin based on the distal radial artery (three patients). The FCR/RA free flaps are useful for coverage of microanastomoses when a radial forearm free flap skin paddle is inset at some distance from the anastomotic site. The reversed distally-based FCR/RA pedicle flap is quite useful for coverage of a dorsal hand wound, without microsurgery and without the troublesome reversed radial forearm skin paddle scars.

A retrospective chart review was performed of the 7 cases utilizing the FCR/RA flap. Donor-site morbidity was evaluated by grip strength measurement. The neurovascular status of the donor extremity was compared to the normal unoperated contralateral upper extremity. Six fresh cadaver arms were dissected to better categorize the FCR/RA flap. Muscle dimensions, vascular anatomy, and neural pedicle lengths were recorded.

In this series of 7 cases, all FCR/RA flaps survived. No significant functional or neurovascular compromise of the donor limb was encountered as a result of FCR removal. Grip strengths were not different from the contralateral limb. The cadaveric dissections demonstrated that the body of the muscle was supplied consistently by 3 to 4 segmental perforating branches from the radial artery. The muscle was, on average, 3×15 cm in size. The nervous supply was primarily off the median nerve, but large variability was noted in branch length (range: 1 to 5 cm) and branching patterns to the pronator teres and flexor digitorum superficialis.

Like the radial forearm flap, the FCR/RA is a reliable flap. Donor-site morbidity was negligible in the 7 cases reviewed. The muscle belly was consistently supplied by 3 to 4 segmental vessels arising from the medial side of the radial artery. The flap is unique in its ability to cover microanastomoses when the radial forearm skin is inset at a distance, and can be used locally to cover small dorsal hand wounds without microsurgery. However, this study did not support the use of the flap as a neuromuscular flap, due to the variability of the nervous supply to the muscle.

Composite Free Serratus Anterior Osteomuscular Flap in Treating Bone Defects of the Upper Limb. Alexandru V. Georgescu, Ivan Ovidiu, Calin Melincovici, and Simona Serbu.

The relatively easy harvesting, the possibility of using different numbers of muscle components, and the relatively long pedicle, have made the free serratus anterior flap an important competitor to the latissimus dorsi flap in the coverage of complex osteocutaneous defects. The flap can be harvested with a bony component, including one or two segments of ribs. It can also be part of a double simultaneous or consecutive free transfer associated with the latissimus dorsi or Chinese flap.

Between 1998 and 2000, the authors used this flap in 12 consecutive cases, 7 males and 5 females, aged between 4 and 54 years. The etiology of the defects was post-traumatic, and the arm was involved in 3 cases, the forearm in 3, and the hand in 6 cases. The flap was used predominantly for complex defects, infected or with a high risk of infection. The flaps were harvested with a bony component, including one or two segments of ribs. In three cases, the flap was part of a double simultaneous (2) or consecutive free transfer associated with a latissimus dorsi or Chinese flap.

A single flap was lost due to an irreversible arterial thrombosis. Minor superficial or distal necroses were observed in 3 cases, usually resolved spontaneously. In all cases, the bony fragment consolidated, and 12-month radiographs demonstrated very good integration. The flaps were monitored for a period of 6 to 24 months, with only minor volume readjustment being necessary in 2 cases. No infections developed in any of the cases.

Compared with other free muscle flaps, and especially with the latissimus dorsi flap, the serratus anterior flap has additional qualities that make it extremely useful in limb trauma: a large surface for reduced volume; a flat belly muscle useful in hand reconstruction; the possibility of partial harvesting of 1 to 3 digitations, essential for small bone defects in which the flap is only a bone carrier; ease of harvesting with bone segments; a long pedicle; and reduced functional deficit in the donor area.

Reconstruction of Large Post-Traumatic Skeletal Defects of the Forearm by Vascularized Free Fibular Graft. Roberto Adani, Luca Delcroix, Marco Innocenti, Luigi Tarallo, Giordano Pancaldi, and Massimo Ceruso.

The most common application of vascularized bone graft has been for reconstruction of the lower extremity. However, the indications for vascularized bone grafts to the upper extremity have been expanded, as this technique became more widely appreciated. Between 1993 and 2000, 12 patients who had segmental bone defects following trauma of the forearm, were managed with vascularized fibular grafts. There were 7 men and 5 women, with an average age of 37 years (range: 19 to 66 years). Two patients had traumatic bone defects; six patients had post-traumatic non-unions; and four had osteomyelitis. The reconstructed sites were the radius in eight patients and the ulna in four. The length of the bone defect ranged from 6 to 13.5 cm. In four cases, the fibular graft was raised as a vascularized osteoseptocutaneous fibular graft. Of these four flaps, one was taken for monitoring purposes only. For fixation of the grafted fibula, plates were used in 10 cases, and screws and Kirschner wires in two cases; in these latter two cases, an external skeletal fixator was used for immobilization of the extremity.

Follow-up ranged from 89 to 7 months. Eleven grafts were successful. Two patients required additional bone grafts. There were no cases with fractures of the grafted bone, but malunion occurred in one case and non-union at the proximal site in one case. No patient had evidence of resorption of the graft or symptoms referable to the donor leg. No recurrence of local infection was encountered in the cases with previous osteomyelitis. The mean period to obtain radiographic bone union was 4.5 months (range: 3 to 10 months). One fibular artery and one fibular vein were anastomosed with the recipient vessels; in two cases, the peroneal artery was used as an interposition graft between the radial artery. According to Tang's classification, the 11 successful cases were graded as excellent in four, good in three, fair in three, and poor in one case.

Use of fibula grafts allows transfer of a segment of diaphyseal bone that is structurally similar to the radius and ulna, and that is of sufficient length for the reconstruction of most skeletal defects in the forearm. The vascularized fibular graft is indicated in patients with intractable non-unions, in whom conventional bone grafting has failed, or for large bone defects (in excess of 6 cm) in the radius or ulna.

Challenge of Axillary Reconstruction Following Radical Lymphadenectomy. John D. Potochny, Charles E. Butler, and Merrick Ross.

Radical axillary lymphadenectomy for metastatic melanoma can be a morbid procedure, often performed for palliation. Cutaneous and soft-tissue defects can be extensive. Recent evidence suggests that, compared with nonsurgical treatment, surgical resection of a single, isolated, nodal metastatic melanoma is associated with a higher patient survival rate. Axillary reconstruction can be accomplished using a wide variety of reconstructive options; however, excision of adjacent local tissue in these cases can limit local flap options and can increase the procedural complexity.

All patients who underwent axillary reconstruction following resection of metastatic melanoma over a 10-year period (1990 to 2000) at M.D. Anderson Cancer Center were reviewed retrospectively. In addition, two separate cases were reviewed in which patients had previously undergone radical axillary lymphadenectomy and subsequently required radical re-excision for recurrent axillary metastatic melanoma followed by flap reconstruction. Ten patients underwent axillary radical lymphadenectomy and reconstruction for metastatic axillary melanoma. Defects were reconstructed with regional (n&equals;5), free (n&equals;4), or supercharged (n&equals;1) flaps. Free and supercharged flaps were either fillet of upper extremity (n&equals;2) or rectus abdominis flaps (n&equals;3). There were no flap losses. Complications occurred in 5 cases (50%) and included hematoma, seroma, partial flap necrosis, and wound dehiscence.

Reconstruction for the two patients who underwent radical axillary re-excision was performed with local fasciocutaneous flaps and a pectoralis major muscle flap with a skin graft in one patient, and a free vertical rectus abdominis flap (VRAM) in the other. Both patients received postoperative radiotherapy. Delayed wound healing and late axillary abscess occurred in the patient who underwent reconstruction with local and pectoralis major flaps. There were no postoperative complications following the VRAM reconstruction.

The challenges encountered with reconstruction of these axillary defects included acute preoperative infection with sepsis and the need for resection of surrounding structures, such as latissimus dorsi and pectoralis major muscle, chest-wall skin, and axillary, thoracodorsal, and subscapular vessels. Specific reconstructive options and techniques were required for each unique circumstance. Axillary reconstruction following radical lymphadenectomy poses a special set of challenges, but can be accomplished with acceptable results through selection of the appropriate reconstructive technique for each individual case.

Emergency ``All-in-One'' Reconstruction in the Treatment of Destructive Traumas of the Upper Limb. Alexandru Georgescu, Ivan Ovidiu, Radu Onoe, Calin Melincovici, and Calin Morar.

The incidence of upper-limb trauma has stimulated the imagination of surgeons, regarding surgical strategy and suitable surgical techniques. When reconstruction through local and/or regional methods is not suitable, microsurgical techniques have allowed for free flaps to resolve these problems. The importance of ``per primam intentionem'' repair has been outlined for all anatomic lesions. When local resources are not suitable for reconstruction, microsurgical techniques have been used to compensate for limited resources.

Between January, 1996 and December, 200, these authors have treated 231 patients with destructive traumas of the upper limb. The etiology was crushing (84), complete and incomplete amputations (125-electrical saw in 94 cases), explosions (10), gun shot wounds (2), and burns (10-electric burns in 7 cases). In 56 cases, the final procedure was delayed for up to 72 hr because of extended lesions; all other cases were treated with emergency repair of the affected tissues. The use of local resources or replacement of the affected tissue with equivalent functional and anatomic tissue was paramount, using microsurgical methods-revascularization (27), replantation (85), flow-through flaps (30), free flaps (154 cases presenting extensive defects). Free flaps were used for coverage (fasciocutaneous-52; muscular-22; musculocutaneous-41; osteomuscular-5; digital transfer-27; two or three simultaneous flaps-15). Free flaps were also used both as surface coverage and for functional rehabilitation.

For interpreting the results, the authors considered local morphologic reconstruction considered acceptable by both patient and surgeon, and functional recovery, which varied from 30% (42 cases) to 90% (7 cases). Socioprofessional reintegration was possible for all of the 216 patients with favorable results (104 returned to work at the same place and job). In 109 cases, one to two reinterventions were necessory for tenolysis, tenoplasties, arthrodeses, or other palliative procedures.

Management of severed upper limb trauma currently is aided using microsurgical techniques, especially replantation and free flaps. Early repair emphasizing per primam intentionem represents the key for good functional results, as it prevents local fibrosis, joint stiffness, muscle denervation, and nerve degeneration. For tissue loss, local regional resources were first considered and, when these were not available, free flaps were adapted to the local anatomy and function.

Management of the Mangled Hand and Upper Extremity. Batia Yaffe and Joel Engel.

The treatment of mangled hand and upper extremity requires microsurgical techniques and more than one surgical procedure. Restoration of function is not always possible, and sound clinical judgement, as well as skill and ingenuity, are necessary in treating these patients.

Between 1985 and 1999, 40 patients were treated with mangled hand or upper extremity, with patient ages ranging between 18 months and 61 years. In two patients, the restored hand had to be amputated and, in one patient, partial amputation was required. Twenty-eight patients were available for final evaluation and were followed for 2 to 11 years post end of treatment. The mean rehabilitation period of these patients was 2.5 years. Most of them had more than four additional surgical procedures during this period, and almost every patient needed some kind of free flap.

The functional result was excellent in 6 patients, good in 9 patients, fair in 9, and poor (no function) in 4 patients. The most significant predictive factor in the final outcome was the patient's age. All patients in the excellent or good result categories were younger than 24 years, and all in the poor result category were older than 40 years of age.

Salvage of a mutilated hand or upper extremity is worthwhile in younger patients, but the morbidity involved in multiple additional surgical procedures and years of rehabilitation, in view of poorer results, might be too much of a price for older patients.

Functional Upper Extremity Reconstruction Using Microvascular Tensor Fasciae Latae. Mark R. Kobayashi, Rangan Gupta, Patrick McMahon, and Gregory R. Evans.

While microvascular muscle transfers to the forearm have been both reliable and successful, functional restoration of the shoulder girdle and upper arm has been disappointing, with suboptimal return of both strength and motion of the shoulder and elbow. The tensor fasciae latae (TFL) myofascial flap is an excellent alternative to traditional shoulder and/or biceps reconstruction. This study focussed on the functional outcomes of the TFL flap.

The TFL flap was used for traumatic reconstruction of the shoulder (n&equals;2) and biceps (n&equals;1). Both casese of shoulder reconstruction were secondary to brachial plexus stretch injuries in association with shoulder dislocations. The biceps reconstruction was the result of biceps muscle transection with distal muscle necrosis and musculocutaneous nerve transection. In all cases of shoulder reconstruction, EMG/NCV revealed complete denervation of the deltoid muscle, mild-to-moderate AC separation, complete inability to abduct, flex, and extend the shoulder. The TFL was inset between the middle and anterior deltoid heads. Reinnervation required nerve graft for the case with axillary nerve transection, and thoracodorsal nerve pedicle transfer for the upper trunk brachial plexus injury. Vein grafts were necessary for vascular reconstruction. In the biceps reconstruction, EMG/NCV demonstrated some reinnervation of the proximal biceps, and limited elbow flexion in neutral position with lack of supination. The TFL was placed in an anatomic position, and reinnervation carried out to the previously transected musculocutaneous nerve. All flaps were performed 12 to 18 months following injury.

Shoulder reconstruction of the anterior and middle deltoids resulted in full range of motion for abduction and flexion (strength 5/5). Pain was markedly reduced, with correction of the shoulder separation. The biceps reconstruction completely restored biceps function for flexion and markedly improved supination (strength 5/5).

Traditional shoulder reconstruction following brachial plexus injuries involving the deltoid and shoulder girdle musculature, has been suboptimal. Shoulder fusion eliminates the pain secondary to chronic subluxation or dislocation, and eliminates scaphohumeral motion. The use of the TFL myofascial flap not only reduces pain by strengthening the shoulder girdle, but also can allow restoration of both flexion and abduction. Careful selection of the recipient nerve is crucial for success and requires preoperative EMG/NCV studies. Return of both adequate strength and motion allows return to near normal premorbid activities.

Hand Resurfacing with the Superthin Latissimus Dorsi Perforator-Based Free Flap. Jeong Tae Kim.

Perforator-based free flaps have been introduced for various kind of reconstruction and resurfacing. The author presented experience with the ``superthin'' latissimus dorsi perforator-based free flap for hand resurfacing, by preserving important donor structures such as motor nerve and muscle. The superthin flap with 5 to 7 mm of thickness can be elevated, including skin and thin superficial adipose layer, based on a single perforator of the back. These thin flaps can be used for resurfacing moderate and severe defects after electrical burn, scar contracture release, and crushing injuries.

Six clinical cases were managed with this superthin flap, without flap necrosis or other serious postoperative complications. All the flaps were thin enough for defects on the palm and wrist after electrical burn, scar contracture release, and crushing injury. Palmar or circumferential resurfacing of the thumb and palm resurfacing in children were available with the superthin flap, and there was no need for secondary debulking procedures. The dimensions of the flap could be safely harvested up to 16x6 cm, based on a single perforator, and the donor site was closed primarily without any functional deficit. An additional pedicle length of 3 to 5 cm was obtained through tortuous intramuscular dissection, and a more proximal pedicle was taken for a T anastomosis of the pedicle at the recipient artery.

This superthin flap is adequate for hand resurfacing and can be elevated from the back by dissection based on a single perforator. This early clinical experience can validate the real possibility for the superthin perforator-based free flap, and can promise widening of its utility for the purpose of resurfacing.

Endoscopic Harvesting of the Temporal-Parietal Fascial Free Flap. Paul S. Cederna and Kevin C. Chung.

The temporal-parietal fascial (TPF) flap is a thin and pliable flap that is ideally suited for soft-tissue coverage of hand and foot wounds. However, the open harvesting technique often creates substantial donor-site deficits, including long scars and alopecia of the temporal scalp. In order to reduce these deficits, these authors have utilized an endoscopic approach to harvesting the TPF free flap through a single 3-cm preauricular incision.

Five patients underwent endoscopic harvesting of the TPF free flap for reconstruction of hand and foot wounds. Three patients had open wounds on their hands following trauma, tumor resection, and radical resection of recurrent Dupuytren's contracture. Two patients required Achilles tendon coverage following trauma. The TPF flap was harvested through a 3-cm preauricular facelift-type incision. Pedicle dissection was performed under direct visualization, and TPF flap harvest was accomplished using endoscopic visualization (5 mm, 30-degree endoscope).

All flaps survived completely. One patient developed a hematoma under the flap from a fall 3 weeks after reconstruction. The hematoma was evacuated, and the wound healed uneventfully. Total flap harvest time was less than 2 hr in all cases. No patient developed alopecia in the temporal scalp. No frontal nerve palsy occurred in this series.

The TPF free flap is a versatile flap that can be safely harvested using endoscopic techniques. Reduction in donor-site deficits, using a minimally invasive approach, should encourage wider applications of this technique.

Functional Adrenergic Responses of the Human Digital Artery. Jason A. Castle, Tom L. Smith, and L. Andrew Koman.

Raynaud's phenomenon affects nearly 10 percent of the general population and up to 25 percent of premenopausal women. Symptoms are thought to arise from an intense vasoconstriction of the digital arteries secondary to hyperactivity of the adrenergic receptors in the vessel wall. To date, the microvascular control mechanisms of the human digital artery are largely extrapolated from animal models, human vessels from non-thermoregulatory beds, and indirect measurements of total digit flow such as venous plethysmography. This study utilized vessel ring wire myography to demonstrate the differences between accepted models of thermoregulatory vascular reactivity and actual functional responses of the human digital arteries.

A 3-5-cm section of the common and ulnar-sided proper digital artery of the index finger was recovered from organ donors. Vessels were stored in UW cryopreservative at 40 degrees F. The artery was separated from surrounding connective tissue and isolated into 3-mm ring segments. Lumen diameters were approximately 1 mm. Ring segments were allowed to equilibrate in physiologic Kreb's buffer at 34 degrees C, and 2 g of tension for 1 hr. Vessels were then treated with propranolol (5×10-7 M, beta adrenergic antagonist), and allowed to equilibrate for 10 min prior to introduction of increasing doses of NE (10⁹-10⁴ M, n&equals;10). Following 10 serial washings with Kreb's buffer, vessels were treated with propranolol and terazosin (1×10⁵ M, alpha-1 adrenergic antagonist) and allowed to equilibrate for 10 min prior to repeating the NE dose response curve.

Tension plots demonstrated a biphasic response to each dose of norepinephrine. This response was unaffected by beta blockade with propranolol. The shape of the tension plots varied significantly from those seen in animal models under identical experimental conditions. Even under direct stimulation with NE, the arteries could not maintain constant tension, suggesting that the digital artery is a conductance, rather than a resistance, vessel, and therefore does not play a significant role in moment-to-moment regulation of total digit blood flow.

The addition of terazosin completely inhibited vessel responses to NE up to 10⁴ M, confirming previous assumptions that the alpha-1 adrenergic receptor is primarily responsible for vasoconstriction of the digital arteries. Vessel reactivity to NE could not be reestablished, even after repeated washings with physiologic buffer. These results provided a valuable insight into the microvascular control mechanisms of human digital arteries, and are relevant for all clinicians treating patients with vasospastic disorders of the hand.

Research supported by Carolina Donor Services and NIH Physician Scientist Fellowship Grant.

Redistribution of Plantar Pressure after Big Toe-to-Thumb Transplantation. Darrell Brooks, Michele Lee, Peter Siko, Harry J. Buncke, Rudolf Buntic, Gabriel Kindd, and Gregory M. Buncke.

The aim of this reported study was to compare peak pressure and contour curves for the sole of the foot in patients who have undergone unilateral big toe-to-thumb transplantation, with the contralateral intact foot.

This was a retrospective study. Gait patterns in 14 patients who had big toe-to-thumb transplantation between 1988 and 1991 were analyzed. The patients were all males, with an average age of 31 years (range: 15 to 55 years). The EMED system was used to measure the plantar pressure distribution in all patients. Patients would contact a pressure-sensitive plate with their bare feet during normal gait, one foot at a time. Each foot was evaluated 10 times. Peak pressures and pressure contours were averaged from these recordings. All patients underwent big toe harvest with preservation of the first metacarpal head and reconstruction of the trans-metatarsal ligament.

The mean peak pressures were higher under the first and second metatarsal heads and lower under the heels of feet with the big toe removed, compared with the contralateral intact foot. After big toe-to-thumb transplantation, the patient's gait changed, as weight was redistributed to the metatarsal heads and away from the heel.

Head and Neck Microsurgical Reconstruction: A Three-Year, Three-Center Collaborative Experience. Aharon Amir, Tommy Shpitzer, Dean Ad-El, Yehuda Ben-Asher, Dan Fliss, Benjamin Shlomi, Jacob Rapapport, Arie Shtayer, Raphael Shafir, Raphael Feinmesser, and Eyal Gur.

Pathologies of the head and neck direct the reconstructive surgeon to respond both to function and aesthetic aspects. Microsurgery makes possible the successful reconstruction of composite defects and has improved patients' quality of life. This paper presented a collalborative experience of three medical centers in Israel.

Over a 3-year period, 91 patients underwent 95 microsurgical procedures. Most of the patients had undergone malignant or benign tumor excisions (82 and 2, respectively), and a small group of patients had reconstructions for traumatic and congenital defects (2 and 5 patients, respectively). Defects of the oral cavity or tongue were the most common (n&equals;27), followed by mandibular defects (n&equals;21). The trauma group involved scalp replantation and neck contracture release. The congenital deformity group was comprised of facial paralysis pediatric cases (6) and 1 complex hypospadias boy. Flaps used included the free radial forearm flap (n&equals;28), free fibula (n&equals;22), free latissimus dorsi muscle (n&equals;8), sural nerve graft (n&equals;7), free jejunum (n&equals;6), sural nerve and free gracilis muscle (n&equals;4), free scapular/ parascapular (n&equals;4), free rectus abdominis muscle (n&equals;3), free lateral thigh (n&equals;1), and free scapular-latissimus dorsi muscle (n&equals;1).

Six patients had failed reconstructions, with a success rate of 93.6%. Most failed flaps (5) were in a patient group that had recurrent malignancy of the head and neck region, had extensive neck dissections, and were previously irradiated. There were 4 recipient and 4 donor-site infectious complications, one patient with pulmonary embolism, and one with DVT. No major donor-site morbidity was recorded, but one patient demonstrated fibular donor-site infection, necrosis of muscle, and residual limping. Seven patients in this series succumbed to their disease. All the rest are free of disease, with agreeable functional and cosmetic results. All facial paralysis patients had a good functional smile, and all nerve-graft patients demonstrated fair muscle recovery.

The microsurgical reconstruction options that were chosen provided both soft-tissue and bony reconstitution, regaining function and cosmesis for most patients. The variety and versatility of options for such head and neck pathologies are remarkable, and sometimes the only remedy for restoring the patient's quality of life.

Salvage of the Pedicled Colon Flap with Microvascular Anastomosis. Hung-chi Chen, Ming-huei Cheng, and Fu-chan Wei.

A long defect of the esophagus is often reconstructed with a pedicled colon segment as a pedicled flap. However, there are situations that cause circulatory problems and eventual loss of the flap, either partially or totally: elderly patients who have arteriosclerosis affecting the vessels supplying the transposed colon segment; anatomic variations, either artery or vein; or inadequate space in the retrosternal tunnel which causes compression to the vascular supply of the colon segment.

Among the 186 cases of colon segment interposition for esophageal reconstruction, three patients were found to have impaired circulation during surgery. Salvage was performed with microvascular anastomosis of the terminal vessels to the neck vessels. Two were due to arteriosclerosis and one was due to anatomic variation. All three pedicled flaps were salvaged successfully. The wound healed well, with good swallowing function in subsequent follow-up. The microsurgical procedure averaged 1 hr.

The authors concluded that microvascular techniques can be applied to improve the results of colon segment interposition whenever necessary. Many cases of leakage can actually be prevented by providing good circulation to the terminal portion of the pedicled flap.

Trans-Oral Reconstruction of the Mobile Tongue Using the Radial Forearm Free Flap. Thomas Shpitzer, Dean Ad-El, Aharon Amir, Raphael Shafir, Raphael Feinmesser, Jakob Cohen, Dan Fliss, and Eyal Gur.

The complex arrangement of muscles in the tongue makes it unlikely that replacement of these muscles will be possible in the near future. It is therefore essential to preserve the remaining mobility in the residual tongue after performing a partial glossectomy. Hence, the authors' approach to reconstruction of the anterior tongue is to allow the residual tongue to maintain its maximum mobility by introducing thin, pliable cutaneous free flaps to restore shape and volume. The aim of this reported study was to share experience of partial anterior tongue resection through trans-oral reconstruction using radial forearm free flaps (RFFF).

Between January, 1999 and December, 2000, eight patients underwent partial glossectomy, followed by trans-oral reconstruction and use of the radial forearm free flap. Patients were reviewed to determine their aesthetic and functional outcomes, relating to speech, deglutition, and aspiration. All patients included in the study underwent resection of 40 to 50 percent of their mobile tongue, as well as reconstruction with radial forearm free flaps trans-orally without lip or mandibular split-thickness grafts.

Mobile tongue reconstruction with RFFFs was performed in eight patients. One flap was explored for ischemia and was not salvaged. Following flap failure, the patient underwent a successful contralateral RFFF reconstruction. There were no other general recipient or donor-site complications. Decannulation of the tracheostomy was performed on postoperative day 8, and the nasogastric tube was removed after 10 days. All patients were followed postoperatively for a minimum of 3 months. All returned to a normal diet within 4 weeks. After 1 month of a normal diet, all patients reported the ability to sense food that was trapped in the floor of the mouth on the side of the reconstruction. Another interesting finding was the diminution in biting the neo-tongue, which frequently occurs during the first weeks postoperatively. Patients stated that they ceased favoring the non-operated side of the oral cavity for chewing. Those who underwent resection and reconstruction of a full 50% of their mobile tongue had excellent cosmetic results. However, those who underwent resection of less than 50% of the mobile tongue had less satisfactory cosmetic results, due to some differences in the height of the dorsal surface of the tongue.

Reconstructing a large portion of the mobile tongue, using a free radial forearm flap through the mouth opening only is a procedure that is simple to perform, and that produces very good functional and aesthetic results.

Microvascular Reconstruction of Complex Orbital Defects Following Resection of Massive Sinonasal and Cutaneous Malignancies. Emad A. Magdy and Brian B. Burkey.

Surgical ablation of extensive sinonasal or skin tumors involving the orbit creates complex three-dimensional defects caused by a combined loss of bone and soft tissue. These defects often occur in previously irradiated fields or in areas that will be irradiated. Management of such defects presents many challenges, including reconstructing a barrier between the intracranial and sinonasal spaces; reestablishing palatal competence; cosmetically rehabilitating the exenterated orbit; and recreating near-normal facial aesthetic contour after composite tissue loss.

Microvascular free-tissue reconstruction was used in 19 patients following resection of extensive malignancies involving the orbit between September, 1992 and March, 2001. Fourteen patients had received previous therapy; nine were recurrent tumors following surgical resection, and five were radiation therapy failures. Ten patients received postoperative radiotherapy. Fifteen of the 19 tumors arose primarily from the sinonasal region; 11 extended to the anterior skull base; and five needed palatal resections. The remaining four were cutaneous tumors with massive facial skin loss following surgical extirpation.

Flaps were chosen based on the reconstructive needs of each patient, as noted. The algorithm for choosing the flap type was discussed. Musculocutaneous free flaps were used for cranio-orbito-facial reconstruction in 15 patients, with the rectus abdominis flap used in 11, and the latissimus dorsi in four patients. Four patients were reconstructed using fasciocutaneous free flaps, three with the radial forearm flap (to allow for future placement of a prosthetic eye), and one with a scapular flap.

All free flaps survived. Only five patients had postoperative complications: postoperative pneumonia in one patient, two minor donor-site related complications, and two patients with wound breakdown following radiotherapy. Patients were followed-up for a period of up to 81 months (mean: 23.7 months). Seven of the 13 patients (53.8%) who continued their regular follow-up visits remain alive with no evidence of disease. Based on these findings, the authors believe that the aggressive use of microvascular free-tissue transfer in reconstruction of complex orbital defects is both safe and effective.

Combined Anterolateral Thigh Flap and Vascularized Fibula Osteoseptocutaneous Flap in Reconstruction of Extensive Composite Mandibular Defects. Naci Celik, Fu-chan Wei, Hung-chi Chen, Ming-huei Cheng, and Wei-chao Huang.

Reconstruction of extensive composite oromandibular defects involves not only mandibular bone, oral lining, and external face, but also cheek bulkiness. This report presented the authors' preferred reconstructive method, using two free flaps.

Between October, 1998 and June, 2000, 22 extensive composite mandibular defect reconstructions, using a combined free anterolateral thigh flap and vascularized free fibula osteoseptocutaneous flap, were performed. All of the patients were males. The tumor type was squamous cell carcinoma and the tumor stage was T3 or T4 in all cases. The inner lining defects were between 4x10 cm and 20x22 cm. The bone defects were between 7 cm and 14 cm. The outer lining defects were between 6x6 cm and 14x16 cm.

Complete flap survival was 90.90% (40 of 44 flaps). A complete loss was seen in an anterolateral thigh flap (2.27%). Skin islands of two osteoseptocutaneous free fibula flaps and one anterolateral thigh flap developed partial necrosis (6.81%). Thirteen patients underwent revision procedures after the first operation, including debulking of the flap or revision of the mouth angle, or both.

An ideal reconstruction of extensive composite oromandibular defects could be achieved by simultaneous use of the osteoseptocutaneous free fibula flap with another free soft-tissue flap. The two most commonly employed flaps, forearm flap and rectus abdominis myocutaneous flap, used for the external face and soft-tissue reconstruction, have been replaced by the anterolateral thigh flap in recent years at the authors' center. The forearm flap is usually too thin to cover the fibular bone, and the reconstruction plate and rectus abdominis myocutaneous flap can cause a subclinical reduction in abdominal strength. Both flaps are difficult to harvest during tumor excision. The anterolateral thigh flap has a large cutaneous area and the vastus lateralis muscle can be included, in order to gain more soft-tissue volume. It can be elevated as a sensate flap by including the lateral femoral cutaneous nerve. The donor site can usually be closed primarily, if the width is less than 6 to 8 cm. There is no need for positional changes, and simultaneous flap elevation with tumor resection is possible.

The free anterolateral thigh flap, combined with the vascularized fibula osteoseptocutaneous flap, is a good choice in the reconstruction of extensive composite defects of the oromandibular region. This procedure gives a better functional and aesthetic result than that achieved with other reconstructive options.

Should the Donor Radius be Plated Prophylactically after Harvesting of a Radial Osteocutaneous Free Flap? - A Cost-Utility Analysis. Gloria M. Rockwell and Achilleas Thoma.

Despite significant reductions in the incidence of donor radius fracture with improved techniques, fractures continue to occur after harvesting of radial osteocutaneous free flaps. In many centers, this has led to prophylactic plating of the donor radius, instead of treatment of fractures after their occurrence. The objectives of this reported study were to assess each of these two treatment options, in terms of a cost-utility analysis, based on secondary data from the literature using a decision analytic model. Costs were calculated from a ministry of health perspective, and effectiveness was calculated in terms of quality-adjusted life years (QALYs).

A literature search identified 22 studies reporting complications of the radial osteocutaneous free flap, and 10 studies reporting complications of radius plating. The data collected were used to create a decision tree with probabilities for 25 different outcomes. Direct costs, utilities, and QALYs were determined for each outcome state.

A decision analytic model was used to determine the total expected costs, utilities, and QALYs for each of the treatment options. Total costs/QALY prophylactic plating: $2071/8.55; total costs/QALY treatment as fracture: $140/9.92; baseline incremental costs/ QALY: $1931/-1.36 years. The results were plotted on a cost-effectiveness plane falling within the ``loose-loose'' quadrant of less effective and more costly. Single and multi-variant sensitivity analysis of the probabilities and costs of various outcomes were performed to determine the effect on the incremental results.

Prophylactic plating of the donor radius after harvesting of an osteocutaneous free flap was not cost-effective, when compared to the treatment of donor radius fractures as they occur. The conclusions were stable, based on single and multi-variant sensitivity analysis.

Reconstruction of the Combined Supraglottic Laryngectomy/Base of Tongue Defect Using the Lateral Arm Fasciocutaneous Free Flap. Brian B. Burkey, Emad A. Magdy, and James L. Netterville.

Resection of an extensive portion of the base of the tongue (beyond the circumvallate papillae) in cases requiring a supraglottic laryngectomy, imposes an increased risk of life-threatening aspiration and deglutition problems, if the resulting defect is not properly reconstructed. A method of reconstruction with a sensate free-tissue transfer was proposed.

The lateral arm fasciocutaneous free flap was used in the primary reconstruction of complex defects resulting from combined supraglottic laryngectomy/ extensive base of tongue resections in five cases between July, 1996 and September, 1998. Three of these patients had undergone prior irradiation, and two patients had resections that would not allow primary closure. Four of the five flaps were innervated by neural anastomoses with local recipient nerves, including the superior laryngeal and lingual nerves. All flaps survived. All five patients were successfully decannulated in periods ranging from 15 days to 3 months. Three of five patients achieved complete oral nutrition within 4 months. One patient remained partially dependent on enteral feedings 4 years postoperatively, and one patient died of disease at 9 months without achieving oral nutrition. Four of five patients (80%) remained without disease at a minimum of 3 years follow-up.

The lateral arm fasciocutaneous free flap was found to be a good option in reconstructing complex defects resulting from combined supraglottic laryngectomy/base of tongue resection, allowing the resumption of an oral diet and airway in the majority of patients, even if irradiated.

Free DCIA Groin Flaps for Mandibular Reconstruction. Batia Yaffe, Ran Yahalom, Haim Kaplan, Nahum Givol, and Schlomo Teicher.

Composite mandibular defects following oncologic resection or trauma cause significant functional and aesthetic deformity, and their reconstruction is a surgical challenge usually requiring a free flap.

Between 1998 and 1999, 30 patients underwent microsurgical reconstruction of a composite mandibular defect in the authors' unit. In 22 patients, 23 DCIA-based free groin flaps were performed, and 13 other flaps were used in eight patients. This presentation dealt with the indications, pitfalls, technique, complications, and immediate and late outcome in the DCIA-flap-treated patients.

There were 13 osteocutaneous and 10 myoosseous flaps. Three flaps were lost, and 70% of the patients needed additional minor surgery (mostly vestibuloplasty and dental implants) to reach optimal results. Seventy-two percent of the patients achieved normal or near-normal appearance and function.

The authors prefer the use of the DCIA groin flap over other osseous or osteocutaneous flaps due to the quality, shape, and height of the available bone and the minimal donor-site morbidity. The flap is especially suited for lateral defects with loss of a large amount of soft tissue.

Distraction Mesenchymogenesis and Double Fibula Free Flaps in Children with Micrognathia. Brian Boyd, Eric Stelnicki, Joav Barnavon, and Craig Ueker.

These authors presented a new technique which combines the concepts of distraction osteogenesis and tissue expansion, and utilizes the advantages of free vascularized bone grafting in the management of severe micrognathia. Two cases were described: a 3-year-old child with an extreme case of Goldenhar's syndrome, and a 3-year-old child with severe bilateral Treacher Collins. Neither had TM joints. Both had tracheostomies and were tube-dependent for feeding. In each, the mandible was distracted away from the skull base, using distraction wires attached to a special halo device. Over- correction was deliberately produced, giving both children a class III malocclusion. At this point, the posterior mandibles were constructed in one stage, using bilateral free fibular osteocutaneous flaps.

The first child was 9 months postoperative at the time of this report. He eats a normal diet by mouth and his tracheotomy is decannulated 12 hr per day. The tube is reinserted at night because the child has chondromalacia as a result of prolonged cannulation. This will be surgically corrected in the near future. The second child is now 2 months postoperative and still in intermaxillary fixation, but her parents have noticed that she can make normal sounds from her larynx-something not possible before. The authors believe that this is due to a vastly increased AP pharyngeal distance, strikingly demonstrated on CT scan.

Comparison between Radial Forearm and Anterolateral Thigh Free Flaps. Naci Celik, Fu-chan Wei, Huan-tang Chen, David C.C. Chuang, Seng-feng Jeng, and Yu-te Lin.

The free radial forearm flap has been the flap of choice for many microsurgical reconstructions for many years. However, the anterolateral thigh (ALT) flap has surpassed it in recent years at the authors' center. In this study, they shared their experience with these flaps in a large number series, and compared the two flaps.

Between August, 1989 and August, 2000, 1028 radial forearm flaps were harvested, mostly for tumor and trauma reconstruction. Of these, 923 were used for intraoral reconstruction, and 72 flaps were used for face, scalp, and neck reconstruction. Of the remaining flaps, 17 were used for the upper extremity, 9 for the lower extremity, and 7 for the trunk and genitalia. After January, 1995, in all cases, the radial forearm flaps were raised with a suprafascial dissection technique, to reduce donor-site problems. Between June, 1996 and August, 2000, 672 ALT flaps were used in 660 patients. In head and neck reconstructions, the authors used 484 ALT flaps, 58 for the upper extremity, 121 for the lower extremity, and 9 for trunk reconstruction.

Complete survivals were 98.83% and 95.68% for radial forearm and ALT flaps, respectively. Reexploration for vascular insufficiency was done in 76 cases (7.39%) for the radial forearm flaps, and in 56 cases (8.3%) for the ALT flaps. The number of reexplorations and partial and total failures was significantly higher in the ALT flap group (chi-square test, p&equals;0.000500).

The ALT flap has a long and large pedicle, a large area of available skin, and little donor-site morbidity. The thickness of the flap can be adjusted, and it can be harvested as a musculocutaneous or chimeric flap. On the other hand, the radial forearm flap is easy to harvest and has a lower flap failure rate. To eliminate donor-site morbidity to a degree, suprafascial dissection of the radial forearm flap has been proposed, but sacrificing a major artery is still a disadvantage; in addition, the thickness cannot be adjusted, and the available skin is not sufficient enough for large defects in extensive head and neck cancers.

With increasing experience, the major concern of microsurgeons has concentrated on donor-site morbidity. The radial forearm flap still remains useful, especially for intraoral reconstruction, but it has been largely replaced by the versatile ALT flap in the reconstruction of soft tissue by the authors in recent years.

Single Artery Successful Replantation of the Totally Avulsed Face and Scalp. Brad Wilhelmi, Robert Kang, Kuimars Movassaghi, and W.P. Andrew Lee.

Successful replantation of the scalp with microanastomosis of a single artery has been reported to produce reliable results. However, only one face and scalp replantation has been reported in the literature. This involved the repair of more than one artery. These authors performed the first successful replantation of a face and scalp with repair of only a single artery.

A 21-year-old man presented after completely severing his face and scalp. The patient had caught his long hair in a conveyor belt at work. The face and scalp were replanted, repairing the right superficial temporal artery with an interposition vein graft. Initially, only the right superficial temporal vein was repaired. One week post-replantation, the patient returned for treatment of venous congestion, with repair of the left superficial temporal vein as well. This saved the replanted portion and it survived entirely.

The face and scalp can be replanted with only single artery repair.

Quixil Fibrin Sealant: A Real Alternative for Nerve Repair. Lorraine Ornelas, Luis Scheker, Luis Padilla, Mauricio Di Silvio, and Manuel Lopez Iglesias.

These authors have previously demonstrated that Quixil human fibrin sealant, used to repair peripheral nerve lesions, produces less inflammatory reaction, less all-around fibrosis, and more axonal regeneration with good fiber alignment, than other fibrin glues and microsuture. The purpose of the reported investigation was to show that the sealant also results in better function.

Thirty Sprague-Dawley rats had their right sciatic nerves transected. Two groups were formed: Group 1 animals were repaired with microsuture, epineurial technique; Group 2 animals were repaired with Quixil fibrin sealant. Tracking and EMG studies were performed 3 and 6 months post-repair. The left sciatic nerves were used for controls.

Results demonstrated that Quixil is a good alternative for peripheral nerve reconstruction, and that it enhances nerve regeneration and recuperation.

Modification of the Oberlin Procedure for Elbow Flexion in Brachial Plexus Injuries. Thomas H.H. Tung and Susan E. Mackinnon.

The Oberlin ulnar nerve fascicle transfer has achieved good-to-excellent recovery of elbow flexion, but with the occasional need for additional procedures such as a Steindler transfer to increase strength. The purpose of this study was to report a modification of the Oberlin procedure in which transfer from the ulnar nerve to the biceps branch of the musculocutaneous (MC) nerve is augmented by the transfer of medial pectoral nerve branches to the brachialis branch.

Brachial plexus reconstruction, in which the Oberlin procedure was augmented by nerve transfers to the brachialis muscle, was performed in four patients, aged 16 to 46 years. Each patient had a severe post-traumatic brachial plexopathy and underwent reconstruction 3 to 5 months after injury. In all cases, a redundant fascicle of the ulnar nerve innervating the flexor carpi ulnaris was transferred directly to the biceps branch of the MC nerve. Reconstruction of elbow flexion was further augmented by transferring medial pectoral nerve branches to the brachialis branch of the MC nerve with a nerve graft. In two cases, the lateral antebrachial cutaneous nerve wass also reinnervated and transposed proximally to neurotize the biceps muscle. In one case, intercostal nerves were used with a medial pectoral branch to reinnervate the brachialis muscle. In three cases, shoulder reconstruction was also performed with nerve transfers to the suprascapular and/or axillary nerves.

Follow-up varied from 14 months to 3 years. At their last visit, two patients were noted to have 4+/5 elbow flexion and two had 4/5 elbow flexion. All are very pleased with their results. One patient was a high school student, and three were employed at the time of injury. Two patients have been able to return to their previous employment (cement mixing and mail carrier).

Oberlin's ulnar nerve fascicle transfer has achieved good recovery of elbow flexion in the reconstruction of brachial plexus injuries, with results varying from 3-4/5 flexion strength. The biceps is a primary supinator and a secondary flexor of the forearm, while the brachialis muscle is the primary forearm flexor. To obtain more consistent and powerful elbow flexion, the authors have augmented the Oberlin procedure by also reinnervating the brachialis muscle. By reinnervating both muscles, they have been able to achieve 4-4.5/5 elbow flexion and excellent functional recovery, as evidenced by the return to previous employment in two of three patients.

Technique and Strategy in Anterolateral Thigh Perforator Flap Surgery-Based on Analyses of 15 Complete and Partial Failures in 439 Cases. Naci Celik, Fu-chan Wei, Hung-chi Chen, Chih-hung Lin, Ming-huei Cheng, and Yi-chieh Chen.

These authors analyzed the causes of flap failures, and presented their technique and strategy concerning anterolateral thigh (ALT) perforator flap surgery.

Between June, 1996 and August, 2000, 672 ALT flaps were used in 660 patients. Their ages at the time of surgery ranged from 3 to 94 years (mean: 49.7 years). To analyze only the perforator flap failures, flaps based on the septocutaneous vessels, musculocutaneous ALT flaps, and chimeric ALT flaps were excluded from the study. There were a total of 439 flaps which were based on musculocutaneous perforators.

Overall, of 439 flaps, 424 flaps survived completely, providing a success rate of 96.58%. Of the 15 failed cases, 8 were complete failures and 7 were partial (10% to 60%) failures. Thirty-four flaps were reexplored after surgery. Nineteen of these 34 flaps (55.8%) were totally salvaged.

In this study, some of the reasons for flap failure, unique to ALT perforator flaps, were identified. These included relative inexperience in perforator flap elevation, fragility of the musculocutaneous perforators, and easy twisting of the pedicle. Important technical concerns in the harvesting of ALT perforator flaps are mapping of the skin vessels with a hand-held Doppler probe preoperatively, meticulous harvesting of the flap under the surgical loupe or even the microscope with lower magnification for inexperienced surgeons, and intermittent topical use of 2% xylocaine during the intramuscular dissection of the perforators, in order to prevent vasospasm. During the reexploration of an ALT perforator flap, one must search for twisting of the pedicle or skin vessels and small bleeders from the intramuscular perforators.

The ALT flap is a reliable flap with a constant anatomy, a long and large pedicle, a large area of available skin, and little donor-site morbidity. Knowledge of the common causes of failure unique to ALT perforator flaps will guide the surgeon in preventing failures and will permit salvage of flaps in jeopardy.

Nerve Allografting without Immunosuppression in a Long Gap Model. D.M. Godat, J.G. Yan, L.L. Zhang, I.E. Shehadi, P.P. Narini, and H.S. Matloub.

The successful transplantation of allograft nerves without the use of immunosuppression will provide readily available material for nerve reconstruction without donor-site morbidity. Success with various materials has been demonstrated in short gap models; however, processed allografts have not been tested in long gap models. This study investigated the feasibility of transplanting processed allograft nerves without immnosuppression in a long gap (6 cm) model using physiologic, as well as histomorphometric, data.

Sciatic nerves were harvested from outbred rabbits to yield an acellular, freeze-dried nerve graft. The nerves were hydrated in saline just prior to transplantation. Twenty rabbits, 2.5 to 3.0 kg, were divided into four groups. A 6.0-cm gap was created in the right peroneal nerve prior to transplantation. Group 1 (autograft) animals received the contralateral saphenous nerve; Group 2 (fresh allograft) received an unprocessed allograft from an outbred rabbit; Group 3 (untreated gap) received no graft; and Group 4 (processed allograft) received the processed allograft. Multiple outcome measures were performed after 6 months, and included moist muscle weight of the extensor digital longus muscle, tetanic muscle test, electromyelogram (EMG), and histomorphometric analysis.

Group 4, with processed allograft, showed no significant difference in maximal muscle force generated, compared to Group 2, the fresh allograft. The mean muscle force for Group 4 was 78 g, compared to 47 g for Group 2 and 1.1 g for Group 3. These were all significantly less than the autograft, Group 1, 285 g. EMG data were consistent with the muscle force data. Muscle weight was less in Groups 2-4, compared to Group 1. Histomorphometric data were also discussed.

The rabbit long nerve gap is an important model for assessing nerve regeneration. Animals receiving processed nerve allografts in a 6-cm gap did not demonstrate a significant physiologic improvement. Processed allograft alone appeared not to be sufficient to stimulate functional regeneration across longer gaps.

Analysis of Limb Salvage Surgery for Soft-Tissue Sarcoma. E. Gene Deune, Maurice Nahabedian, Anthony Tufaro, Deborah Frassica, Steve Lietman, and Frank Frassica.

Extremity sarcoma is a life- and limb-altering diagnosis. Whenever possible, limb preservation is performed by wide tumor resection, immediate soft-tissue reconstruction, and adjuvant therapy. This was a retrospective analysis of the limb-salvage surgery performed at the Johns Hopkins Hospital by reconstructive surgeons, radiation oncologists, and orthopedic surgeons.

Between 1998 and 2001 (38 months), 49 patients (m&equals;35, f&equals;14) underwent 50 limb-salvage procedures (33 lower extremity, 17 upper extremity, 31 left, 20 right). The mean follow-up was 10.2±7.8 months (range 1 to 32 months). The most common diagnosis was malignant fibrous histiocytoma (n&equals;22). Thirty-seven (75%) tumors were high grade. The thigh was the most common location (n&equals;16). The rectus abdominis (n&equals;20), latissimus dorsi (n&equals;13), and gracilis (n&equals;5) were the most common flaps. Twenty-four were free flaps, with one acute loss (4.2%) and two late losses (8.3%). All pedicle flaps survived (total flap survival: 94%). Nineteen patients had concurrent surgery, the most common being tendon reconstruction. The mean time of surgery was 8.3±3.2 hr. The mean hospital stay was 9.0±4.1 days.

Preoperatively, 21 had radiation, chemotherapy, or both. Postoperatively, 15 patients (four of whom had preoperative radiation) were treated with brachytherapy starting at 5.5±2.1 days. No flaps were lost. Sixteen others had postoperative external radiation, chemotherapy, or both. Wound complications were seen in 25 (51%), most commonly, seroma (n&equals;11). There were 10 non-wound related complications (18.4%), the most common being nerve palsy (n&equals;4). Sixteen patients (32.7%) had surgery for their complications: seroma drainage (n&equals;4), wound debridement (n&equals;4). Perioperative mortality was zero. Four had metastasis at the time of limb salvage surgery. Seven later developed metastasis. Three had local recurrence. Two patients died, surviving 5.5±3.5 months after surgery.

Limb function was graded as excellent (defined by return to preoperative level), moderate (functional but limited in range or strength), or poor (severe limitation due to pain or poor motion). Twenty-eight have regained full limb use; 19 have moderate use; and four have limited use. Overall, 47 (92%) have full-to-moderate limb function. No limbs have been amputated.

This was a continuing analysis of experience with limb salvage for sarcomas. The data indicate that limb-salvage surgery results in functional limbs which, historically, would have been amputated. Although the complication rate was high, particularly for wound healing, limb salvage should be performed whenever possible, to avoid the morbidity associated with amputation.

Experience with the Capana Method for Limb Salvage: Using the Free Fibula Flap and a Cadaveric Allograft in 20 Bone Sarcoma Patients. Eyal Gur, Aharon Amir, Yehuda Kolander, Yoav Barnea, David Leshem, Arik Zaretski, Jerry Weiss, Raphael Shafir, and Issak Meler.

The fibula free flap has become the gold standard in orthopedic oncology for limb salvage after long-bone tumor resection. Large intercalary resection is possible when the fibula can be used in exchange. The fibula is still a narrow and weaker bone than the original resected bones, and for functional replacement, a long period of immobilization is needed. During that period, osteotomy sites will unite, and a longer period will elapse for the fibula to hypertrophy through a process of pressure transport, creation of micro-fractures, and callus formation. To achieve strength for the reconstructed limb in the early period (until the fibula will support all of the physical load), Capana advised the use of a combination of the free fibula flap, an allograft, and hardware fixation.

Between 1997 and 2001, these authors treated 20 patients who had sarcoma limb-salvage resections. All of them were reconstructed using the free fibula flap, according to the Capana technique. There were 11 males and 9 females, with a mean age of 21 years. The major diagnosis was Ewing sarcoma (12 patients), who were all reconstructed in a delayed fashion (3 to 5 years post-resection). The rest had osteogenic sarcoma (8 patients), and were all reconstructed immediately. The femur was involved in 12 patients, the tibia in 4, the radius in 3, and the humerus in only 1 patient. In all patients, the allograft used was taken from the national bone bank, and the hardware varied from plate to intramedullary fixation, or their combination.

All flaps were postoperatively immobilized in a plaster cast and monitored using an implantable Doppler device. All flaps clinically survived (100%) and, at day 10 postoperatively, bone scans were performed to monitor their vitality. In all 20 patients, a positive scan was obtained. All fibulas demonstrated good union and callus formation in x-rays performed 3 to 5 months postoperatively. Patients were kept immobilized for 6 to 9 months (transition period in which they used a brace and gradually learned to bear weight on the reconstructed limb until they were fully weight-bearing). In this series, there was no major recipient or donor-site complication. In two patients, a failure of the hardware fixation system was observed and surgically repaired.

The authors found free fibula flap reconstruction, using the Capana method, an optimal reconstruction for intercalary bony resection, with low morbidity and a good functional outcome.

Role of Free-Tissue Transfer and Sural Neurocutaneous Flaps for Reconstruction of Leg Wounds. Mark A. Grevious, Loren S. Schechter, Risal Djohan, David H. Song, and Robert F. Lohman.

Neurocutaneous flaps are a relatively recent addition to the armamentarium of techniques for lower extremity reconstruction. The role of sural neurocutaneous flaps for soft-tissue reconstruction of the leg, compared to free flaps, has not yet been fully clarified.

Forty-seven patients, operated on consecutively between January, 1999 and May, 2001, were analyzed. They required either sural flaps or free flaps for reconstruction of leg wounds. Sural flaps were attempted in 22 patients and successfully completed in 21 (in one patient with collagen-vascular disease, the skin paddle did not appear to be perfused and the flap was replaced prior to transfer). In three patients, the sural nerve was divided distally, and the skin paddle was transferred to the proximal third of the leg. Reversed sural flaps were used in the remainder of patients. Free flaps were used for 25 patients (18 rectus abdominis flaps and 7 radial forearm flaps). The two groups of patients were similar with respect to age, sex, American Society of Anesthesiology class, tobacco use, and wound duration. There was a trend (p&equals;0.03) toward greater utilization of radiotherapy among patients with free flaps (20%), compared to patients undergoing sural flaps (0%). The free flaps were larger than the sural flaps: 200 cm² vs. 22 cm², p<0.0001).

Patients with free flaps also required more ICU days than did patients with sural flaps: 4 days vs. 0.5 days, p<0.0001. In the free flap group, there were two major complications (total flap necrosis) and 7 other complications treated non-operatively, including donor-site complications (3), flap congestion treated with leeches (1), partial skin graft loss (2), and minor wound separation (1). In the sural flap group, there were no episodes of total flap necrosis. However, there were 2 major complications, including congestion treated with leeches, leading to partial flap necrosis (1), and inability to transfer the flap (1).

Sural flaps can be useful for treatment of small wounds of the leg, without resorting to microsurgery or muscle transfer. Free flaps are the method of choice for treating large wounds, but are associated with a greater need for ICU care. There is minimal risk of total flap necrosis with sural flaps, but minor wound-healing problems are common.

Decreased Activation of Neutrophils in Young Adolescent vs. Adult Rats Following Ischemia/Reperfusion Injury to the Gracilis Muscle. Arian Mowlavi, D. Garth Meldrum, Bradon J. Wilhelmi, and Michael W. Neumeister.

These authors observed that procedures, such as replantation of amputated extremities and free flap transfers, are better tolerated by young adolescents, compared to older adults. This may be due, in part, to decreased ischemia/reperfusion (IR) injury in younger patients following these procedures. Neutrophils have been identified as main culprits in IR injury. In this reported study, the role of neutrophil activation in IR injury was assessed in young adolescent vs. mature adult rats.

The expression level of activated neutrophil surface markers and neutrophil associated oxidant production were analyzed following 4 hr of ischemia and 1, 4, and 16 hr of reperfusion, using a gracilis muscle flap model. These outcome measures were determined by assessing for the mean fluorescence intensity (MFI) of CD-11b and CD-45 markers and dihydrorhodamine (DHR) 123 oxidation to rhodamine, respectively, using flow cytometry assay. Additionally, muscle edema, as determined by wet-to-dry muscle weight ratio, and muscle viability, as determined by nitro blue tetrazolium staining following 24 hr of reperfusion, were completed for each of the groups. The Student t-test was utilized to analyze the statistical significance of the outcome measures above.

Neutrophil activation was significantly decreased in the young adolescent rats, compared to the mature adult rats at 4 hr (54.878 MFI vs. 71.102 MFI, p&equals;0.025). In contrast, the MFI of positive CD-11b and CD-45 events was significantly decreased in the mature adult rats, compared to the young adolescent rats at 16 hr of reperfusion (81.958 MFI vs. 86.873, p&equals;0.007). Neutrophil oxidative was significantly decreased in the young adolescent rats, compared to the mature adult rats, after both 1 hr (51.775 MFI vs. 78.100 MFI, p&equals;0.035) and 4 hr of reperfusion (46.548 MFI vs. 83.690 MFI, p&equals;0.005). Decreased neutrophil activation after 1 and 4 hr of reperfusion correlated with significantly lower edema formation in the young adolescent rats, compared to the mature adult rats (1.123 vs. 1.248, p&equals;0.0011). Decreased neutrophil activation also correlated with a trend toward improved muscle viability in the young adolescent rats, compared to the mature adult rats (32.3% viability vs. 23.7% viability, p&equals;0.189).

The authors presented evidence for differences in IR injury outcome measures throughout the reperfusion period that were dependent on age-related differences.

Minimally Invasive Harvesting of the Gracilis Muscle without Endoscope Assistance. Nai-siong Kueh, Seng-feng Jeng, and Yur-ren Kuo.

Minimally invasive surgery with endoscopic assistance for muscle harvesting has recently become popular. However, the procedure entails the use of special instruments and also requires special training for an extensive learning period. The purpose of this study was to develop a new minimally invasive procedure for harvesting the gracilis muscle flap without endoscopic assistance.

In this series (Group1, 12 patients), gracilis muscle flaps were harvested using a minimally invasive technique, with the aid of a blunt dissector (MacCollum-Dingman breast dissector), without endoscopic assistance. Another 12 patients (Group 2) had flaps harvested using the conventional open technique.

Group 1 patients had an average incision length of 6.5 cm, compared to 17.5 cm in Group 2. The average harvesting time in Group 1 was 36 min, and that in Group 2 was 41 min. The total amount of fluid drainage postoperative day 4 was 99.87±30.45 and 139.79± 49.92, respectively, in Group 1 and Group 2. All the free muscle transfers were successful without major complications in both groups. All the patients in Group 1 were satisfied with the painless procedures, as well as the cosmetic results, compared with those in Group 2.

Compared with a conventional technique, this procedure is advantageous in its easier performance, shorter incision, lower morbidity, and a better appearance at the donor site.

Experience in the Management of Distal Tibial Bone Loss and Nonunion with Free Vascularized Fibular Transfer. Michael E. Pannunzio, Lorenzo L. Pacelli, Marvin R. Brown, and William C. Pederson.

Fractures of the distal tibia present difficult management problems. They often involve the joint surface of the ankle and are frequently open, leading to a relatively high rate of infection and nonunion. While transfer of a free fibula is an accepted procedure for tibial-shaft nonunions, there is little in the literature regarding free fibula transfer in the management of tibial plafond fractures.

Over a 6-year period, 13 patients underwent free vascularized fibular transfer for bone loss and/or nonunion after tibial plafond fracture. All patients had been considered for below-knee amputation prior to fibular transfer. The average age at the time of fibular transfer was 39 years, with an average time from injury to the index procedure of 8 months. Three of the initial fractures were closed, with the remaining 10 open. Five of 13 patients developed infected nonunions; however, all were culture negative at the time of fibula transfer. The average number of prior operations was 3.8. All patients had an external fixator placed, and all fibulas held with minimal internal hardware. Seven patients had tibiotalar arthrodesis, while the remaining had the fibula placed into the tibial plafond without crossing the joint.

Three patients required early return to the operating room for vascular compromise, but all had vascular salvage. One patient required free muscle transfer due to venous compromise of the skin paddle (with patent vessels to the bone). Ten of 13 (77%) patients went on to primary union of the fibula and tibia, with three patients developing a nonunion of one end. One patient opted for secondary cancellous bone graft of the distal site with eventual healing, while two other patients (15%) opted for below-knee amputation. One patient developed a draining sinus at 13 months (despite healing), which resolved after debridement and hardware removal. Of the 11 patients with eventual union, partial weight-bearing began at an average of 4 months, and weight-bearing without assistive devices by 8 months. The presence of prior infection did not significantly increase the time to full weight-bearing. Donor-site morbidity was low, in 2 of 13 patients (15%).

Management of nonunion and infected nonunion of distal tibial fractures remains a challenging problem. This small series demonstrated that free fibula transfer offers a relatively high union rate (85%) in these complex cases, with an acceptable rate of complications. In properly selected patients, this procedure can offer a functional alternative to below-knee amputation.

Functional Restoration Surgery in Limbs Treated for Sarcoma. E. Gene Deune, Anthony Tufaro, Deborah Frassica, Steve Lietman, and Frank Frassica.

Between 1998 and 2001, 49 patients underwent limb-salvage surgery for sarcomas at the Johns Hopkins Hospital. Eight patients required restoration of limb function with tendon transfer, tendon grafts, or innervated muscles, when tumor resection eliminated crucial limb functions. A retrospective review of these cases was undertaken.

There were eight patients (males&equals;4, females&equals;4; mean age 51.3±25.3; range: 23-84 years). Seven had undergone immediate functional restoration. Four had immediate tendon transfer; two required tendon grafts to restore tendon continuity; and two required innervated muscle transfer (pedicled latissimus, free gracilis) to restore elbow flexion. Six of the tumors were located in the upper extremity, and two were in the lower extremity. MFH was the common diagnosis (n&equals;4). Seven were free flaps, and one was a pedicled flap: rectus abdominis, n&equals;3; latissimus dorsi, n&equals;3; gracilis, n&equals;2. No flaps were lost.

Three patients underwent additional concurrent surgery: DRUJ arthrodesis, vein graft repair of the brachial artery, and peroneal nerve grafting. The mean operative time was 11.4±2.1 hr, compared to 7.8±3.1 hr for limb salvage cases, in which no functional restoration surgery was required. Five patients had postoperative complications; three were wound-related; three required surgery for complications; four had preoperative adjuvant therapy; and four had postoperative therapy.

The mean follow-up was 13.3±6.2 months (range: 5 to 22 months). Two patients developed local recurrences treated with further excision; one had a metastasis at the time of the surgery; and one developed a metastastic lesion to the shoulder, which was excised. Six remain free of metastasis; seven remain alive; and one died of an unrelated cause.

Limb function was graded as excellent (defined by return to preoperative level), moderate (functional but limitation in range or strength), or poor (severe limitation due to pain or poor motion). Excellent to good return of function was reported in 87.5% of cases (1 excellent, 6 moderate). Both patients who had innervated muscle transfers are now able to flex their elbows at 135° against gravity. One patient had significant pain that severely limited motion. No limbs have been amputated.

Technically feasible functional limb restoration can be performed at the time of resection and soft-tissue coverage, unless contraindicated. Although the numbers and follow-up periods are limited, the results are encouraging. The authors believe that functional restoration is a crucial element in attaining a successful outcome in limb-salvage surgery.

Transcutaneous Tissue pH Monitoring via IPG. Raymond M Dunn, Jocelyn Songer, and Stevan Kun.

Advances in flap monitoring over the past 5 years have resulted primarily in incremental improvements of existing technologies. Tissue pH has been shown to reliably predict flap compromise in experimental and clinical studies, but current pH measurement technology requires the placement of an indwelling tissue pH probe (invasive). The authors were interested in designing a pH-based flap monitor utilizing non-invasive technology (impedance plethysmography, IPG) and in establishing its ability to correlate with direct tissue pH measurements in experimental and clinical ischemia in arterial and venous compromise.

Utilizing impedance, a monitor was designed to transcutaneously measure alteration in blood flow, and to correlate simultaneously with tissue pH measurement via traditional implanted glass electrodes. On confirmation of the technology in a rabbit hindlimb model, the addition of electronic filter and noise reduction algorithms was applied, to improve clinical interpretability. Clinical studies were performed in tourniquet extremity cases of >30 min duration (n&equals;20), to establish and confirm clinical applicability and effacacy.

The experimental results and signal analysis revealed a close correltation of non-invasive IPG and direct pH measurement. Experimental and clinical trials revealed an excellent correlation of direct and IPG tissue PH measurement, and confirmed existing reports of rapid response of pH to tissue ischemia and reperfusion.

Previous reports have documented the efficacy of free flap tissue pH monitoring. These authors presented a technologic advance in pH monitoring that allowed this technique to be performed non-invasively. Further refinements of this technology should provide other steps toward the development of an ideal free-tissue transfer monitor.

Engineering a Blood Supply to Target Tissue: What Conduit Configuration Works Best? Robert L. Walton, Elisabeth K. Beahm, and Kevin Maguire.

In the fabrication of tissue engineered constructs, vascular supply can be provided by the placement of a pedicled flap or vascular conduit into or adjacent to the construction site. The efficacy of the various methods to achieve vascular integration has not been clearly characterized or quantified. The purpose of this reported study was to investigate the effectiveness of various autologous blood vessel conduits in promoting neovascular integration of a target entity.

Thirty-two New Zealand White rabbits (4.5 g b.w.) were utilized. The femoral artery and vein of each groin were dissected over a distance of 5 cm. The femoral vessels were then subjected to the following randomized protocol: Group 1-control (no further manipulation; Group 2-segmental 3-cm resection and anastomosis of the femoral artery only; Group 3-test of femoral artery only; Group 4-segmental (3 cm) resection and anastomosis of femoral artery only; Group 5-test of femoral vein only; Group 6-segmental (3 cm) resection and anastomosis of femoral vein only; Group 7-distal division and anastomosis of femoral artery and vein to create an A-V loop; Group 8-segmental (3 cm) resection and proximal repair of femoral artery and vein, and distal anastomosis of femoral artery to femoral vein to create an A-V graft loop.

Disks of expanded polytetrafluoroethylene measuring 2.0x2.0x0.2 cm were placed adjacent to the manipulated vessels of all preparations, and then sealed between leaves of silicone sheeting. At 3 and 6 weeks, the disks were examined and the distance of neovascularization was measured from the central axis of the vascular conduit.

Controls demonstrated the most prolific neovascular response, with complete integration of the scaffolds (4/4). Group 7 (A-V loop) demonstrated vascular integration of 3 of 4 scaffolds at 6 weeks. The femoral artery or vein alone failed to yield a neovascular response. Anastomosis of the artery alone, vein alone, or both, yielded variable neovascular responses. The A-V loop of arterial and vein segments yielded very minimal vascular growth.

The optimal engineering of a ``vascular capillary network'' to a target entity requires the presence of an intact functioning artery and vein. Segmental grafts of both arteries and veins demonstrate significant delay in neovascularization, when compared to controls. Isolated artery and vein preparations did not demonstrate neovascularization during the 6-week period of observation. The findings have important implications in the development of in vivo tissue engineering strategies.

Reperfusion Injury is Attenuated in iNOS Gene Deficient Mice. Long-En Chen, Anthony V. Seaber, James R. Urbaniak, and Li Zhang.

The authors' preliminary data have shown that inhibition of inducible nitric oxide synthase (iNOS) diminishes ischemia/reperfusion (I/R) injury in rat skeletal muscle. To confirm the role if iNOS in I/R injury, this study evaluated I/R injury in skeletal muscle in iNOS gene deficient [NOS (-/-)] mice.

The left cremaster muscles of male iNOS (-/-, n&equals;11) and C57BL/6J (n&equals;11) mice, weighing 20 to 25 g, were isolated. Three hours of ischemia were achieved by clamping the vascular pedicle of the isolated muscle, and denervating by resecting a 3-mm segment of the genitofemoral nerve. Vessel diameter and overall blood flow of the muscle were measured at 10-min intervals over a 90-min reperfusion period. Weight ratio (% of normal) and pathologic features of reperfused muscles were also determined.

At 10 min of reperfusion, the muscle blood flow was 32±21% (mean±SD) of baseline in controls, and 80±24% in the iNOS (-/-) group. While it gradually increased in controls to a maximum of 60±57% at 90 min, blood flow in the NOS2 group reached baseline level at 20 min, and remained at that level throughout the experiment. Blood flow was significantly (p<0.001) greater in the iNOS (-/-) group than in controls at all time points. Also, the average vessel diameters were between 58% and 66% of baseline in controls at 10 min of reperfusion, and increased to a maximum of 74±5% in 10-20-mm, 78±6% in 21-40 mm, 68±10% in 41-70 mm vessels at 90 min. In contrast, the diameter in the iNOS (-/-) group sharply increased to over 85% of baseline in each vessel category at 10 min, and reached a maximum of 106±8%, 96±8%, 88±5% at 90 min, with a significant (p<0.001) difference at each time point. The muscle weight ratio was greater (p<0.001) in controls (196±64% of normal) than in the iNOS (-/-) group (116±8%). Inflammation and neutrophil extravasation was less severe in the iNOS group than in controls.

Using iNOS (-/-) mice, the authors have confirmed the role of iNOS in the mechanism of I/R injury of skeletal muscle. Inducible NOS activity of leukocytes outside the vasculature during reperfusion leads to excess NO production that can affect many cellular functions, such as changing ion currents; can alter the plasma membrane potential, inhibiting cellular respiration and enzyme activities; and can mediate DNA damage. These effects lead to cellular damage or necrosis. The data suggest a potential pathway to attenuate or prevent I/R injury via inhibition if iNOS.

Mechanism of Microsurgical Thrombosis Inhibition by the Platelet Glycoprotein IIb/IIIa Antagonist SR121566A. Shim Ching, Achilleas Thoma, and John G. Kelton.

Glycoprotein (GP) IIb/IIIa is a platelet surface receptor that is crucial in platelet aggregation. Prior studies have evaluated a specific and potent inhibitor of GP IIb/IIIa, SR121566A, in a rabbit ear model of arterial microsurgical thrombosis. In these studies, the authors noted a significant four-fold increase in vessel patency following administration of SR121566A over saline control (80% vs. 20% patency, respectively, p<.01). This was correlated with marked inhibition of platelet aggregation and statistically insignificant differences in coagulation assays and bleeding times. Scanning electron microscopy of occlusive thrombus suggested decreased platelet deposition in SR121566A-treated vessels.

The purpose of the reported study was to further delineate the mechanism by which SR1211566A inhibits microsurgical thrombosis. This was approached by quantification of radio-labelled platelets and fibrinogen deposition, following a standardized arterial injury, that reliably results in occlusive microvascular thrombus formation.

Experimental animals were divided into two groups: saline control (n&equals;18) and an SR121566A-treated group (2 mg/kg, n&equals;18). Both ear arteries were then injured in a standardized fashion. 51CR platelets and 1251 fibrinogen were administered 15 min prior to microvascular clamp release. At 10 min prior to release, the saline vehicle or drug was administered in a blinded fashion. Assessments were carried out at 10, 20, and 30 min after clamp release (n&equals;6 at each time interval per treatment group). At these intervals, vessels were tested for patency by the Acland refill test prior to sacrifice. Two-centimeter segments of arteries centered over the area of damage were measured for radioactivity content. Results were compared by ANOVA. As the study was still blinded at the time of report and was still ongoing, current results were available at the time of presentation.

GP IIb/IIIa antagonists represent a new class of anti-platelet agents that may be ideally suited for inhibiting microsurgical thrombosis. The results reported will quantify the effect of SR121566A on platelet and fibrinogen deposition in the authors' model. In addition, the study provides additional insights into the relative contribution of platelets and fibrinogen to arterial microvascular thrombosis.

PKC Contributes to Microvascular Protection through Delayed Ischemic Preconditioning in Skeletal Muscle. Wei Z. Wang, Linda L. Stephenson, Kayvan T. Khiabani, Chandra Nataraj, and William A. Zamboni.

The authors' hypothesis was that microvascular protection induced by delayed ischemic preconditioning (IPC) in skeletal muscle is protein kinase C (PKC)-dependent. Vascular isolated cremaster muscle of SD rats underwent 4 hr of warm ( 30 to 32 degrees C) ischemia, and then 60 min of reperfusion. IPC consisted of 45 min of ischemia, but was done 24 hr before prolonged ischemia. To mimic the effects of IPC in the late phase, 4-phorbol 12 myristate 13-acetate (PMA, a PKC activator) 10-7M was given 24 hr before prolonged ischemia via local intra-arterial infusion. To block the effects of IPC in the late phase, Chelerythrine (CHE, 1 PKC inhibitor) 10-5 M was given 30 min before prolonged ischemia via local intra-arterial infusion. Arterial diameters (including feeding and terminal arterioles) and capillary perfusion were measured, using intravital microscopy. Four groups were compared: 1) IPC; 2) sham IPC; 3) CHE + IPC. 4) PMA + sham IPC.

IPC on day 1 produced significant microvascular protection against prolonged ischemia on day 2 (p<0.05, compared to sham IPC). Adminstration of PMA on day 1 without IPC produced a similar protection on day 2 to that induced by IPC alone. In contrast, blocking PKC enzymes by a PKC inhibitor, CHE eliminated the IPC-induced microvascular protection seen on day 2 (p<0.05, compared to IPC).

The results indicated the IPC-induced microvascular protection in the late phase was PKC-dependent.

(This study was part of an ongoing research project supported by the NIH, RO1 HL 65384-01).

BioGlue^R as a Tissue Adhesive for Seroma Prevention. Nathan Gopi Menon, Stephen Downing, Nelson Goldberg, and Ronald Paul Silverman.

Donor-site seroma formation is a common free-tissue transfer complication due to the creation of a large dead space. Previous studies have shown that fibrin glue is successful in reducing the rate of seroma formation in a rat model. However, commercially produced fibrin glue is made from pooled human blood products, and therefore may be unacceptable to some patients because of the risk of viral transmission. Autologous fibrin glue made from cryoprecipitate is not as strong an adhesive owing to lower fibrinogen concentrations, and requires preoperative blood donation by the patient. The use of a synthetic tissue adhesive would eliminate concerns about viral transmission and preoperative blood donation.

In this study, the authors investigated the efficacy of a synthetic tissue adhesive, BioGlue^R, in the prevention of seromas in a rat model. This is a tissue adhesive that has been approved by the FDA for use in the United States for the treatment of acute thoracic aortic dissections. The tissue adhesive is composed of purified bovine serum albumin (45%) and glutaraldehyde (10%). The glue is dispensed through a glue gun and is easy to use in this application.

Twenty-two Sprague Dawley rats were randomly assigned to either a control group (n&equals;11) or an experimental group (n&equals;11). All the rats underwent resection of the pectoralis major along with axillary lymphadenectomy and disruption of subcutaneous lymphatics, consistent with the established model. The control group received 2.5 cc of saline before closure; 2.5 cc of the BioGlue was injected into the experimental animals. On POD 7, the resultant seromas were quantified. Wound tissues were harvested for histologic examination on POD 2 and 7.

The rats in the control group had a mean seroma volume of 5.19 cc (SD±3.65 cc, n&equals;11), while the rats treated with BioGlue had a mean seroma volume of 0.25 cc (SD±0.43 cc, n&equals;11). An unpaired Student's t-test was performed showing a statistically significant difference bween the control and experimental groups (p<0.0003). Histologic analysis of the wound tissues performed on POD 2 showed an inflammatory response consistent with postoperative changes, but no foreign-body reaction, while the tissues harvested on POD 7 showed both an inflammatory response and a foreign-body reaction.

The study demonstrated that BioGlue can be used as a tissue adhesive that significantly reduces seroma formation in this rat model. Longer term studies should be performed to investigate the long-term effects on surrounding tissues.

Brief Extremity Ischemia Prior to Flap Ischemia has a Preconditioning Effect on Flaps in a Rat Model. Markus V. Kuentscher, Eva U. Shirmbeck, Marta M. Gebhard, Cristoph Heitmann, and Guenter Germann.

Ischemic preconditioning (IP) is a protective endogenous mechanism to reduce ischemia/reperfusion injury, and is defined as a brief period of ischemia that is termed ``preclamping.'' This is followed by tissue reperfusion, and is thought to increase ischemic tolerance. Murray was the first to describe this phenomenon for the heart in 1986. The objective of this reported study was to determine if ischemic preconditioning and enhancement of flap survival can be achieved not only by preclamping of the flap pedicle, but also by induction of an ischemia/reperfusion phenomenon in a body area distant from the flap, prior to flap harvest.

Forty male Wistar rats were divided into four experimental groups. An extended epigastric adipocutaneous flap (6x10 cm) was raised. In the control group, a 3-hr flap ischemia was induced. In the preclamping group, a brief ischemia of 10 min was induced by clamping the flap pedicle, followed by 30 min of reperfusion. Ischemia of the right hindlimb was induced in Group 3 by clamping the femoral artery and vein for 10 min after flap elevation. The limb was then reperfused for 30 min. Thereafter, flap ischemia was induced, as in the control group. A similar protocol was used in Group 4. A tourniquet was used to induce hindlimb ischemia, and further experimentation was then performed, as in Group 3.

The mean flap necrosis area was assessed for all groups on POD 5 using planimetry software. The average flap necrosis area was 68.2% (±18.1%) in the control group, 11±8.38% in the preclamping group, 12.5±5.83% in the femoral ischemia group, and 24±11.75% in the tourniquet group. All preconditioned animals demonstrated a significantly lower area of flap necrosis than the control group (p<0.0001, Student's t-test).

The data demonstrated that ischemic preconditioning and enhancement of flap survival can be achieved not only by preclamping of the flap pedicle, but also by induction of an ischemia/reperfusion event in a body area distant from the flap, prior to harvest. These findings indicated that IP is a systemic phenomenon, and not just a local reaction in the flap. The data suggested that remote IP could be performed simultaneously with flap harvest in the clinical setting, for improved flap survival without prolongation of the operative procedure. This may decrease the rate of partial flap loss or fat necrosis, especially in high risk groups such as smokers, those with irradiated tissues, and obese patients.

Endothelial Cell-Coated Small-Diameter PTFE Grafts in an Animal Model. Rolf Buettemeyer, Julian W. Mall, Axel Rademacher, Andreas W. Philipp, and Manrico Paulitschke.

In the past 15 years, various research groups have focused on the endothelialization of PTFE grafts, to minimize thrombogenicity in small-diameter grafts. Endothelial cells grown under static conditions on PTFE grafts have not been sustained on the graft surface with normal in vivo blood flow. Recent publications have emphasized that initial shear stress on endothelial cells is able to modulate the functional and structural status of cells. These authors have recently described a novel perfusion system which enables culturing endothelial cells on PTFE grafts under a wide range of shear stresses up to a confluent cell layer. The aim of the reported study was to validate in an animal model that endothelial cells in seeded grafts under shear stress will remain as a confluent layer with a normal systemic blood flow in an animal.

The study was performed with ``Deutsche Landrasse'' female pigs, weighing 25 to 35 kg (n&equals;13). Endothelial cells were taken from the external jugular vein, isolated, and seeded until a sufficient number of cells could be harvested for graft coating (∼800,000 cells). The PTFE grafts, 4 mm in diameter with a length of 5 cm, were soaked with fibrin glue, coated with the cell suspension, and then immediately hooked up to a perfusion circuit with a steady and slowly increasing flow to a maximum pulsed shear stress rate of 15 dyn/cm². Two weeks after cell harvesting, the PTFE grafts demonstrated a confluent cell layer in vitro, and were implanted end-to-end to an identical animal utilizing the femoral artery and standard vascular surgical techniques. Six weeks later, the graft was harvested for histologic examination. The blood perfusion rate through the graft was measured immediately after implantation and before explantation. Histologic evaluation was done by an independent pathologist. Three uncoated grafts were used as controls.

All of the uncoated grafts were clotted. All of the coated grafts showed normal blood perfusion after 6 weeks. The perfusion rates after implantation and before explantation were similar. Histologic examination showed a confluent layer of endothelial cells adherent to the fibrin glue surface of the PTFE graft.

The authors demonstrated that endothelial cells seeded under flow on the surface of small-diameter PTFE grafts grow to confluent cells in vitro. These cells, grown under shear stress, remain as a confluent layer on the surface of the PTFE graft after exposure to physiologic blood flow, providing firm cell adherence.

Early Capillary No-Reflow during Low-Flow Reperfusion after Hindlimb Ischemia in the Rat. Florian Fitzal, Frank A. DeLano, Corey Young, and Geert W. Schmid-Schoenbein.

Increasing evidence has indicated that pathological events in the microcirculation, rather than at the arterial or venular segment, are responsible for flap failure. Reduction of arterial inflow after ischemia (low-flow reperfusion) is associated with capillary no-reflow and an increase in flap necrosis. However, the underlying mechanisms remain unresolved. The development of these complications may be strongly flow-dependent. Therefore, the authors examined the difference between normal-flow and low-flow reperfusion by assessing functional capillary density, leukocyte and platelet accumulation in capillaries, and capillary perfusion and leukocyte behavior in post-capillary venules after 2 hr of hindlimb ischemia. In addition, since proteases may play a role in microvascular complications, they investigated the role of a serine protease inhibitor (gabexate mesilate&equals;FOY) during low-flow reperfusion.

In a rat model, after 2 hr of hindlimb ischemia, normal- or low-flow reperfusion was produced, and the microcirculation in the gracilis muscle was monitored by intravital microscopy. In a third group of animals, gabexate mesilate (FOY), a serine protease inhibitor which interferes with platelet aggregation and leukocyte adhesion, was administered intravascularly before and during low-flow reperfusion. Low-flow reperfusion (confirmed by measurement of the femoral arterial red blood cell velocity), resulted in capillary no-reflow at an earlier stage, compared to normal-flow reperfusion (functional capillary density 10 min after reperfusion: 101±7 vs. 210±18 um/capillary). The capillary lumen was not visible during ischemia and did not open on reperfusion. They observed an increase in leukocyte adhesion forces to the post-capillary venules at a later stage of low-flow reperfusion, compared with normal-flow reperfusion (leukocyte adhesion index 60 min after reperfusion: 0.19±0.18 vs. 0.75±0.12). However, neither a significant number of adherent leukocytes to the post-capillary venules nor obstruction of capillaries by platelet aggregates could be detected during low-flow reperfusion. FOY did not attenuate capillary no-reflow, but reduced leukocyte adhesion forces to the post-capillary venules (leukocyte adhesion index 60 min after reperfusion: 0.49±0.06).

Low-flow reperfusion led to early capillary no-reflow which may be at least partially, responsible for further reperfusion damage, and thus flap failure. The mechanism seems to be independent of leukocyte adhesion to the post-capillary venules or platelet aggregation. Instead, endothelial cell and/or tissue swelling, in combination with luminal obstruction and leukocyte plugging, may be responsible for the early capillary no-reflow phenomenon.

Clinical Utility of Cyclosporin A and Thymoglobulin against Ischemia/Reperfusion Injury in Rat Skeletal Muscle. D. Garth Meldrum, Arian Mowlavi, Ryan Naffziger, Ashkan Ghavami, Bradon Wilhelmi, Michael W. Neumeister, and Timothy O'Connor.

Replantation and transplantation of tissues may suffer from the effects of ischemia/reperfusion (IR) injury. Cyclosporin A (CsA) and thymoglobulin provide protection against graft rejection through lymphocyte immunosuppression. Evidence for an independent protective effect of CsA against IR injury during organ transfer has prompted studies showing benefits of CsA in various ischemia-exposed visceral organs. Thymoglobulin has demonstrated favorable results in several types of organ transplants, and has also been hypothesized to have a potential role in prevention of IR injury. These authors evaluated CsA and thymoglobulin administration in rat IR injury models, to investigate the potential benefits of immunosuppression in skeletal muscle IR injury.

To determine the effects of CsA on IR injury, 4 hr of ischemia wass induced in the gracilis muscle in a rat model. CsA (15 mg/kg PO) was administered in two experimental groups: a) pre-ischemic (n&equals;6), 48, 24, and 3 hr prior to ischemia and b) post-ischemic (n&equals;6), a half-hour after induction of ischemia. The effects of CsA on IR muscle injury were observed in each of the experimental groups, as well as a control group (n&equals;6) exposed to similar ischemia and administered a saline vehicle. The effect of thymoglobulin in IR was assessed with 4 hr of ischemia in a rectus femoris rat model (n&equals;5) vs. a 4-hr ischemic control (n&equals;5). Thymoglobulin (5 mg/kg) was administered I.V. 2 hr after the induction of ischemia. Muscle viability (NBT staining) and muscle edema (weight to dry weight ratio) were assessed 24 hr following reperfusion.

The pre-ischemic CsA-treated gracilis muscle group demonstrated improved viability (39.1±4.8%), compared to the ischemic control group (23.8±7.1%, p&equals;0.039). Furthermore, the pre-ischemic CsA-treated group demonstrated decreased edema (1.137±0.095), compared to the control ischemic group (1.248±0.045, p&equals;0.011). Only a trend toward improved muscle viability (32.1±4.2%) and decreased edema formation (1.200± 0.062) was observed in the pre-ischemic CsA-treated group, compared to the control ischemic group. Likewise, the thymoglobulin-treated rectus femoris muscle group demonstrated only a trend toward improved viability vs. control (52.2±8.18% vs. 39.4±5.2).

The observations confirmed the beneficial effects of pre-ischemic CsA therapy observed in organ transplantation research, and suggested a clinical utility for CsA therapy for patients undergoing upper-extremity transplantation. Potential beneficial effects of pre-ischemic treatment with thymoglobulin treatment are currently underway.

Improvement of Muscle-Flap Hemodynamics by Angiopoetin-1. Raffi Gurunluoglu, Przemyslaw Lubiatowski, Corey K. Goldman, Blazenka Skugor, and Maria Siemionow.

Angiopoetin-1 (Ang-1) constitutes a novel family of endothelial growth factors that are ligands for the endothelium-specific receptor, tyrosine kinase. These authors designed an experimental study to investigate specifically the angiogenic potential of Ang-1 and its hemodynamic effects, by using a cremaster muscle flap model.

Forty-five male Sprague-Dawley rats (200 to 250 gr) were divided into three groups of 15 animals each. A cremaster tube flap was prepared on its pudo-epigastric pedicle. An intraarterial adenoviral gene therapy was chosen to deliver Ang-1. Injections were given into the common iliac artery after clamping the pedicle. Control flaps received saline. In Group 2, the flaps were treated with adenovirus vector encoding Ang-1 (Ad-Ang-1, 108 pfu). Another group served as treatment control, Ad-GFP (control gene encoding green fluorescent protein). The tube flap was inserted into a subcutaneous tunnel in the lower limb. The flaps were withdrawn from the leg at days 3, 7, and 14, and prepared for in vivo microcirculatory measurements, such as capillary count, RBC velocity, vessel diameter, leukocyte-endothelial interaction, and permeability index. Sections from the flaps were stained with H&E and factor VIII staining.

At day 7, capillary density was significantly higher in the Ad-Ang-1-treated group, compared to control and Ad-GFP groups (p<0.01). Similarly, the permeability index was significantly lower, compared to that of the Ad-GFP group (p<0.01). At 14 days, the number of capillaries was significantly higher in the Ad-Ang-1 group, compared to control and Ad-GFP groups (p<0.01). The permeability index was significantly lower in the Ad-Ang-1-treated group, compared to the Ad-GFP group (p<0.01). Capillaries in the Ad-Ang-1-treated flaps were more tortuous, with respect to control flaps.

In the Ad-Ang-1-treated cremaster flaps, functional capillary density increased by 46% and 98% over time at 7 and 14 days, respectively (p<0.01). However, the permeability indices were not significantly different from control cremaster flaps at any time point. Increased capillary perfusion, coupled with the formation of more stable and mature vessels resistant to leakage, provided in vivo evidence that Ang-1 promoted muscle-flap hemodyanamics. Ang-1 alone or in combination with other angiogenic factors may have promise as an alternative therapeutic agent to improve perfusion in muscle flaps, ischemic tissues, and chronic wounds.

Vascular Endothelial Growth Factor Gene Transfection to Improve Skin-Flap Survival. Nicole Zook Sommer and Michael W. Neumeister.

The purpose of this reported study was to investigate the protective effects of vascular endothelial growth factor (VEGF) gene transfection in a skin-flap model, using a liposomal carrier.

Eighteen Wistar rats, divided into three groups, were injected subcutaneously along an area corresponding to the mid-portion of a planned extended fasciocutaneous McFarlane flap on the back of the animal. One group was injected with commercially available VEGF-165 gene mixed with a liposomal carrier, lipofectamine. Two control groups were injected with either lipofectamine or normal saline. At 1 week, flaps of 3×10 cm were raised, then stapled back into place over a 0.5-mm-thick silicone sheet. Surface planimetry was used to calculate the area of flap necrosis at 1 and 2 weeks. At 2 weeks, tissue samples from the flap and uninvolved back skin were harvested and analyzed for capillary counts. Statistical comparison was performed with independent t-tests.

The difference in the area of necrosis between the normal saline group and the VEGF+lipo group was statistically significant (p&equals;0.03) at 2 weeks, but not significant at 1 week (p&equals;0.05). The difference in the area of necrosis was found to be more statistically significant between the VEGF+lipo group and the lipofectamine group at 1 and 2 weeks (p&equals;0.0078 and p&equals;0.0025). The lipofectamine group had a larger mean area of necrosis (55 mm and 66 mm) at both weeks, compared to the normal saline group (51 mm and 59 mm). The VEGF+lipo group and the lipofectamine group demonstrated a highly significant difference in capillary count (p&equals;0.0005).

The authors hypothesized that the transfection of the VEGF gene into the tissue may further improve flap survival with continuous local production of the VEGF protein. The lipofectamine group displayed more necrosis than the normal saline group, suggesting that lipofectamine, used only as a carrier in this study, may have a deleterious effect. This may explain why the VEGF+lipo mixture group did not demonstrate as significant a decrease in necrosis, compared to the normal saline group. The lipofectamine may have negated the positive effects of VEGF. This is also supported by the capillary counts. The lack of statistically significant differences in capillary counts between the normal saline group and the VEGF+lipo group, may be secondary to the negative effects of the lipofectamine. The carrier of the VEGF gene may be inhibitory and, by eliminating this factor with the use of a different carrier, VEGF may have a more profound effect on flap survival.

Myogenin Expression of Human Skeletal Muscle after Denervation. Jian-guang Xu, Han-wei Huang, and Yu-dong Gu.

A degenerative process occurs in human skeletal muscles following peripheral-nerve injury. As muscle atrophy progresses, it gradually becomes impossible for skeletal muscle to regain innervation. Myogenin is a transcriptor, playing a key role in myogenic differentiation. It regulates the expression of embryonic AchR and other muscle-specific genes. The presented research focused on myogenin expression in human skeletal muscles after denervation.

In the first part of the experiment, 11 skeletal muscle samples were obtained from different body regions of patients with different denervation periods. Myogenin mRNA levels were detected with quantitative RT-PCR, with Aldolase-A mRNA as the internal standard. Then, nine further skeletal muscle samples were obtained from patients with different denervation periods. Immunohistochemistry was used to detect myogenin expression (ABC). Several sections were selected for H&E staining.

In the first set of samples, myogenin mRNA levels in skeletal muscle elevated after denervation, culminated by 7 months 37.5 times greater than normal. By 3, 12, and 26 months, the myogenin mRNA expression was greater by 25 times, 21.3 times, and 11 times. In the second set of samples, with myogenin protein located at the nuclei of myofibers, a few positive nuclei could be observed in samples with denervation periods of less than a half-year, and much fewer in 9-month and 10-month sections. No positively stained specimens were investigated in normal and 26-month samples.

Human skeletal muscles lost their normal differentiation gradually after denervation. The elevation of myogenin expression after peripheral-nerve injury, reflected that myofibers ``re-entered'' the regeneration process within 1 year, but no regenerative reaction of the muscle satellite cells could be demonstrated. The myogenin mRNA level decreased at 1 year after denervation, but the morphologic changes were not as severe as at 2 years. In correlation with clinical experience, the skeletal muscles could recover some contractile function after 1 year of denervation, but this was not as good as with a short denervation period. The authors suggested that if the myogenin expression level in human skeletal muscle were elevated in some way between 6 months and 1 year (such as with gene therapy), this would be beneficial for the reinnervation of myofibers.

Ischemia/Reperfusion Regulates NOS Expression in Peripheral Nerve. Wen-Ning Qi, Peter G. Whang, Zuo-Qin Yan, Long-En Chen, Anthony V. Seaber, and James R. Urbaniak.

Nitric oxide (NO) is believed to be an important messenger molecule in the nervous system. The purpose of this reported study was to characterize the distribution of NOS isoenzymes, and to observe any changes in their transcriptional and transnational levels in peripheral nerves subjected to ischemia/reperfusion (I/R) injury.

The sciatic nerves of 66 male adult rats were divided into two groups. Nerves in the ischemia group underwent 2 hr of ischemia. In nerves in the I/R group, a 5-mm segment of each nerve was subjected to a 100-g crush load for 2 hr to simulate ischemia, followed by 3 hr of reperfusion, after which the nerves were harvested. In each group, the opposite sciatic nerve of each rat was used as a normal control.

Following 2 hr of ischemia, mRNA expressions measured by RT-PCR remained at normal level, but i-NOS mRNA showed a 6-fold increase from the normal. Following 2 hr/3 hr (i/R), n- and e-NOS mRNA expression diminished to 91±30% and 62±25% of normal, respectively. In contrast, i-NOS mRNA further up-regulated to 14-fold of normal. Western blots showed that n-NOS protein level decreased slightly from normal in the nerve following 2 hr of ischemia, but further decreased to 63.9% following 2 hr/3 hr I/R. e-NOS protein decreased in response to ischemia (80.9%) and I/R (81.0%). i-NOS protein expression was not identified in any group. Immunohistochemical evaluation showed n-NOS positive staining in Schwann cells and axons of nerves, and e-NOS in vascular endothelia, but no i-NOS positive staining in any group.

In the authors' opinion, this is the first study to examine altered expressions of NOS genes and proteins in the somatic nerve following I/R. The results reveal the dynamic expression of individual NOS isoforms during the course of I/R injury. The findings that decreased constitutive NOS expression and increased i-NOS expression during I/R, indicate that NO plays an important role in the physiologic or pathologic processes involved in reperfusion injury of peripheral nerves. NO may be beneficial or detrimental, depending on the stage of evolution of the event (such as ischemia and I/R) and on the cellular compartment producing NO. It is conceivable that an understanding of this modulation on a cellular and molecular level may lead to an understanding of reperfusion injury observed in vivo, and to improvement of clinical outcomes in peripheral-nerve injury.

Ex Vivo Transfection of Free-Flap Microvascular Beds: New Method of Efficient, Targeted Gene Therapy. Jamie Levine, Tamara Elias, Seum Chung, Robert Galiano, Wendy Olivier, Anil Aluri, Oren Lerman, Alexes Hazen, and Geoffrey Gurtner.

Gene therapy using viral vectors holds the promise of treating a variety of disorders. These authors have developed a unique method of gene delivery, which may avoid many of the problems of viral transfection, including systemic toxicity. An explanted vascular bed is transfected ex vivo, and then reattached to the native circulation using microvascular techniques.

Fifty Sprague-Dawley rats had groin free flaps raised, based on the superficial epigastric vessels. These flaps were transfected with a replication-deficient adenovirus containing a b-galactosidase (b-gal) reporter gene driven by the CMV promoter. The virus was infused via the artery at concentrations ranging from 1×106 to 4×1011 PFU. Following transfection, the flaps were anastomosed to the native vasculature. The flaps were analyzed histologically at various time points for b-gal staining. The protein was quantified using an ELISA assay, and PCR analysis was used to systemically evaluate gene activity.

High regional transfection of the b-gal gene was noted in the flap and was not restricted to the endothelium. Transfection efficiency was as high as 100% for endothelial cells, 40% for connective tissue, 30% for adipocytes, and 15% for keratinocytes. No b-gal staining was noted outside of the transfected flap. These observations were confirmed by PCR. These studies revealed that the transgene was localized to and expressed throughout the transfected flap.

Gene therapy has had limited clinical success because of difficulties in developing targeted, sustained, high-level gene expression, with acceptable host toxicity. The authors' method represents a technological advance over currently utilized techniques of gene delivery, and appears to be an ideal method of targeted gene transfer. This method of free-flap transfection has the potential of replacing missing gene function, while circumventing many of the systemic problems of gene therapy. It may prove useful for treating a host of congenital, acquired, and oncologic disorders.

Microcirculation is Improved in the L-Selectin and CD-18-Knockout Mouse During Reperfusion. Colin Cooney, Long-En Chen, Anthony V. Seaber, and James R. Urbaniak.

Neutrophils play an important role in ischemia/reperfusion injury (IRI). Netrophils first attach to the endothelium in a weak interaction (rolling) mediated by selectins: P-, E-, and L-selectin. In addition, leukocytes firmly adhere to endothelium in a step mediated by CD11/CD18. Although the authors' previous work demonstrated that blocking L-selectin reduced IRI, there are few data to compare the function of these leukocyte adhesion molecules. This reported study compared the roles of L-selectin, P-selectin, and CD18 during IRI in skeletal muscle, using a murine knockout model.

The left cremaster muscle from mice deficient in the gene for L-selectin, P-selectin, or CD18 (n&equals;16 per group) underwent 4 hr of ischemia, followed by 90 min of reperfusion. C57BL/6J mice were used as controls. During reperfusion, blood flow or vessel diameter were assessed at 10-min intervals. Following reperfusion, the muscles were weighed to evaluate edema, and were sectioned for histological evaluation. Vessels were divided into three categories (<20, 21-35, and 36-60 mm), according to their baseline diameter. The results were expressed as a percentage of the baseline measurement, and muscle wet weight was expressed as percentage weight of the contralateral muscle.

The L-selectin (-/-) and CD18 (-/-) groups demonstrated significantly increased vessel diameters vs. controls in all three vessel categories throughout the 90 min of reperfusion and vs. the P-selectin (-/-) group in the <20 and 21-35-mm arterioles. Vessel diameters in the P-selectin (-/-) group were not statistically significantly different from controls. Blood flow in the CD18 (-/-), L-selectin (-/-), and P-selectin (-/-) groups reached a maximum flow of 119±8%, 118±12%, and 99±12%, respectively, with a significant over 2.3-fold difference vs. controls. The wet weight ratio in the three knockout groups was significantly reduced vs. controls. Histology results indicated severe edema, widespread vacuolization, and extensive necrosis in the controls. Similar changes were observed in the P-selectin (-/-) group, but were markedly reduced in the L-selectin (-/-) and CD18 (-/-) groups.

Versus controls, the microperfusion was significantly improved in the L-selectin (-/-)and CD18 (-/-) groups. Although some improvement was also found in the P-selectin (-/-) group, the magnitude was less than that in the L-selectin (-/-) and CD18 (-/-) groups. The difference could be explained by the time frame in which these molecules function. L-selectin is constitutively expressed, while P-selectin is expressed early on the endothelium (within minutes). The results suggest that the adhesion molecules, L-selectin and CD18, are attractive targets for potential therapy against IRI.

An Outcome Analysis Comparing the Thoracodorsal and Internal Mammary Vessels as Recipient Sites for Microvascular Breast Reconstruction: A Prospective Study of 100 Patients. Steven L. Moran, Guillermina Nava, Amir B. Behnam, and Joseph M. Serletti.

The thoracodorsal (TD) vessels have been the standard recipient vessels for the majority of surgeons performing free TRAM reconstructions. Recently, the internal mammary (IM) vessels have been recommended as the first choice of recipient vessels for microvascular breast reconstruction. These authors performed a prospective randomized trial examining the differences in operative and aesthetic outcomes between each recipient site.

A prospective trial of 100 patients presenting for microvascular breast reconstruction was conducted. The first 60 patients were randomized so that 30 flaps were anastomosed to the IM vessels, and 30 were anastomosed to the TD vessels. The recipient vessels for the remaining 40 flaps were left to the discretion of the surgeon. To evaluate aesthetic outcome, a group of five blinded, non-medical observers and three blinded plastic surgeons graded each result for symmetry and overall aesthetic result on a scale of 1 to 5.

Forty women underwent planned reconstruction using the IM vessels, while 60 underwent proposed reconstruction with the TD vessels. Average operating time was 6 hr in each group. Average hospital stay was 5.8 days in each group. Conversion rate from initial recipient vessel to a secondary recipient site occurred in 12.5% of IM reconstructions and 7% of TD reconstructions. All converted IM cases involved left-sided reconstructions, and were due to venous inadequacy. Twenty-five percent of left-sided IM reconstructions were found to have inadequate recipient veins. Unusable TD vessels were found only in delayed reconstructions, at a rate of 15% in the delayed setting. All flaps from converted procedures survived without complications. The average follow-up was 15 months, during which time there was one flap loss in the TD group. There were no significant differences in complication rates between groups. The average aesthetic grade was 3.5 in each group. Postoperative recovery time averaged 8 weeks. Overall patient satisfaction was given an average grade of 4 in each group.

There were no significant differences in operating time, length of stay, return to work, or final aesthetic results in either group. The left IM site was acceptable for microvascular reconstruction in only 75% of the patients, possibly making the TD site a more reliable option for left-sided reconstructions. The TD vessels were acceptable in 85% of delayed reconstructions, possibly making the IM site a better option for delayed reconstruction. Surgeons should be aware of conversion rates, and should plan appropriately, if recipient vessels appear unusable for free-tissue transfer.

Comparison of Free and Pedicled Transverse Rectus Abdominis Myocutaneous Flaps in Postmastectomy Reconstruction: A Cost-Utility Analysis. Achilleas Thoma, Dana Khuthaila, and Gloria Rockwell.

Since its description by Hartrampf et al. (1982), the transverse rectus abdominis myocutaneous (TRAM) flap has become the preferred method of post-mastectomy reconstruction. However, in the last decade, there has been an ongoing controversy about whether the TRAM flap should be used as a pedicled or free flap. The efficacy of both procedures has been documented in the literature. But a literature review has failed to identify a single article that deals specifically with the cost-effectiveness of these procedures, using proper health economic methodology. The purpose of this reported study was to compare the free TRAM to the pedicled TRAM flap in post-mastectomy reconstruction, using a cost-utility analysis.

A decision analytic model and a Ministry of Health perspective were used for this study. Medical costs associated with the two techniques were estimated from the Ontario, Canada, Ministry of Health Schedule of Benefits (1998). Hospital costs were obtained from St. Joseph's Hospital, a university hospital in Hamilton, Ontario. Utilities were obtained from ten ``experts,'' who were knowledgeable about the two techniques and their complications. They were presented with scenarios outlining clinically important health states associated with each technique. Utilities were converted into quality adjusted life years (QALYs). The probabilities of the various health states (e.g., total or partial flap loss, hernia of the abdomen, etc.), associated with the pedicled and free TRAM flaps, were obtained by reviewing 15 key articles.

An incremental cost-utility ratio (ICUR) was calculated, based on the assumption that the probabilities of health states and cost estimates reflected the two techniques. A sensitivity analysis was also performed by altering the probabilities associated with the two techniques. The free TRAM flap was found to be more costly than the pedicled flap, but it provided more QALYs. The ICUR was $12098.94/QALY. Using the grades of recommendations for adoption of new medical interventions (Laupacis et al., 1992), this study showed that there is strong evidence for adoption of the free TRAM flap.

The study demonstrated that the free TRAM flap is more cost-effective than the pedicled TRAM flap in the Canadian health care system. Due to the uncertainty surrounding the true rate of complications and the true estimates of the costs associated with the two techniques, there is a need to collect sampled data (costs and utilities) in a randomized controlled trial comparing the two techniques.

Free Superficial Inferior Epigastric Flap for Breast Reconstruction. Ming-Huei Cheng, Fu-Chan Wei, and Hung-Chi Chen.

The use of lower abdominal skin and fat is the gold standard for autologous tissue transfer in breast reconstruction. The main problem in the majority of lower abdominal fat transfers is that they may be associated with significant donor-site morbidity due to the harvest of some or all of the rectus abdominis muscle. The goal of this reported study was to investigate the minimal donor-site morbidity of lower abdominal fat transfer by using the superficial inferior epigastric flap for breast reconstruction.

From September, 2000 to April, 2001, there were seven consecutive cases that underwent breast reconstruction using the free superficial inferior epigastric flap. Two flaps were performed as an immediate breast reconstruction. Five of seven patients underwent delayed breast reconstruction. The mean patient age was 42 years. The flap design was the usual usage for the transverse rectus abdominis myocutaneous (TRAM) flap. The lower abdominal incision was made to explore the superficial inferior epigastric vessels. The pedicles were identified and dissected. The flap was then harvested similarly to the TRAM flap without opening the rectus fascia.

All seven flaps survived well, giving a 100% success rate. One of the flaps had temporary venous congestion, with subsequent infection. This flap survived after partial debridement. All patients could get out of bed within 1 day postoperatively. The mean length of the superficial vessels was 10 cm, and the mean diameter was 1.8 cm in the artery and 2.0 cm in the vein.

Abdominal weakness and hernia can theoretically be completely avoided by harvesting neither fascia nor rectus abdominis muscle, when elevating the lower abdominal fat flap. The superficial inferior epigastric flap is the better option than the free TRAM flap for breast reconstruction, in terms of donor-site morbidity. The only drawback of this technique is the relatively shorter and smaller pedicle. However, this can be adjusted by using the internal mammary vessels as the recipient.

Robotic Endoscopic Harvest of Internal Mammary Vessels for Free-Flap Breast Reconstruction. Brian Boyd, Kenneth Stahl, Michel Samson, and Raul Rosenthal.

Internal mammary artery harvesting via minimally invasive techniques is becoming more common in the modern era of cardiac surgery, for use as a bypass conduit in coronary bypass surgery. Over the last 24 months, port access endoscopic LIMA and RIMA harvesting, aided by robotic devices, has been advanced in several centers. The authors' current method of choice is the use of the ``Aesop'' voice-activated robotic arm (Computer Motion, Inc., Santa Barbara, CA). With small modifications of the standard technique used in cardiac surgery, it has been possible to harvest the internal mammary artery, and pass the pedicle containing the artery and vein through the chest wall for use in TRAM flap reconstruction after mastectomy. The principal modification involves port placement through the mastectomy incision into the thoracic cavity, allowing for a more appealing final cosmetic result. The remainder of the pedicle mobilization is as has been reported in the cardiac surgery literature. The entire internal mammary artery is mobilized, and a 3- to 4-inch-long pedicle is brought out through the second interspace, without the need of resecting ribs or even costal cartilage. This pedicle has abundant length to accommodate any free flap with ease. Six illustrative cases were shown.

Use of Lidocaine-Norepinephrine Significantly Reduces Blood Loss and Operating Time in Free TRAM Breast Reconstruction. Asko M. Salmi, Esa Muuronen, and Camilla Hellevuo.

Free TRAM breast reconstruction should be a safe, routine operation, considering the vast need for breast reconstructions. Lidocaine-epinephrine injection is in common use with breast reductions, because it has been proved to decrease blood loss during surgery. The aim of this study was to evaluate if lidocaine-epinephrine injection could be safely used in free TRAM breast reconstruction, to reduce blood loss and operating time.

Fifteen consecutive free TRAM procedures (control group), operated on previously, were compared to 14 consecutive free TRAM procedures operated on by the same surgeons, with a new technique using lidocaine-epinephrine injection (experimental group). In the experimental group, the authors used 0.5% lidocaine-Adrenalina which was infiltrated into the lower abdomen (60 ml) before elevating the TRAM flap, 20 ml in the chest wall before excising the scar, and 20 ml in the other breast, if reduction mammoplasty or mastopexy was required.

The mean range of blood loss during the operation (ml): 990 250-2500 380 140-610, p<0.001. Blood loss after the procedure (ml): 750 170-1680 660 230-1370. The need for blood (units of red cells): 2.2 0-6 1.5 0-4. Operating time (min): 188 137-270 142 100-185, p<0.001.

The study demonstrated that lidocaine-epinephrine reduces blood loss during a free TRAM procedure by 60%. In addition, the operating time is reduced by 25%, mainly because of less bleeding. Lidocaine-epinephrine did not have any effect on patient hemodyanmics during the operation. It also appeared that patients in the control group had less postoperative bleeding and needed fewer transfusions.

Vasoconstrictive agents have been thought to be contraindicated in microsurgery; however, this study showed that the combination of lidopcaine-Adrenalina can be safely used in free TRAM breast reconstruction. Since the study commenced, the authors have performed a total of 60 free TRAM reconstructions with the described technique, without postoperative thrombosis or flap loss.

Critical Coverage of Large Chest and Breast Defects. Mussab A. Al Shammari and William W. Shaw.

Large chest-wall defects due to tumor resection, recurrence, irradiation, or infection, present life-threatening or disabling problems requiring large-flap coverage. Questions remain regarding the extent of resection feasible, skeletal support, flap choice, and outcome. The UCLA plastic surgery flap records (1990-2000) were reviewed to identify patients with large chest and breast lesions requiring major flap coverage. The diagnosis, defect size, resection, flap used, and outcome were determined.

Twenty-three patients were identified with the following diagnoses: large primary tumor (5), recurrent carcinoma after radiation (11), large or multiple recurrent breast cancer (5), and radiation necrosis (15). Most resections involved more than half the chest, and the largest flap used measured 23x50 cm. Flaps used included the TRAM (19), TFL (2), and LTD (2). Two cases required skeletal support with Marlex mesh, and none required bone graft. There were no operative mortality, flap loss, or major wound complications. In all cases, the wounds healed and functional results were satisfactory. No significant disabling paradoxical movement was noted. Four patients subsequently died of metastasis. The reconstruction was beneficial to the patient's quality of life in all cases.

Extensive chest and breast defects can be salvaged reliably with large flaps. Common flap choices are the TRAM, tensor fasciae latae, and latissimus dorsi flaps. Once satisfactorily reconstructed, the functional/ aesthetic outcomes are excellent. Patients with advanced cancer can have effective palliation, with wide resection and reconstruction to improve the quality of life.

Irradiated Thoracodorsal Vessels as Recipients for Free-Flap Breast Reconstruction. Yvonne L. Karanas, David N. Sayah, and William W. Shaw.

Radiation to the axilla frequently results in fibrosis and damage to the thoracodorsal vessels. These authors evaluated the ability to utilize the irradiated thoracodorsal vessels as recipients for autologous free-tissue breast reconstruction.

A retrospective review was performed of all free TRAM-flap reconstructions at the authors' institution from 1991 to the present. Patients who had received preoperative radiation were identified and served as the experimental group. The non-irradiated patients served as the control group. The ability to utilize the thoracodorsal vessels was determined for each group. This decision was made by the reconstructive surgeon, and was based on the diameter of the vessels and the amount of flow within the vessels. Use of alternative vessels indicated that the thoracodorsal vessels were not available. The incidence of vascular complications was determined for each group. The groups were then compared, using the Student's t-test.

Those who underwent preoperative radiation prior to free TRAM breast reconstruction comprised 106 patients; 583 underwent free TRAM-flap breast reconstruction without radiation. In 74 (70%) of the irradiated patients, the thoracodorsal vessels were suitable, and were used as the recipient vessels. In the non-irradiated group, 483 (83%) of the patients had suitable thoracodorsal vessels that were used as the recipients for free-tissue transfer. The difference between these two groups was statistically significant (p&equals;0.002). The incidence of vascular complications for the irradiated thoracodorsal group was 1.4%, and 1.9% for the non-irradiated thoracodorsal group. There was no statistically significant difference in vascular complications between the two groups.

Preoperative radiation therapy to the axilla results in a slight decrease in the usability of the thoracodorsal vessels. The surgeon should therefore be prepared in these cases to select alternative vessels, such as the internal mammary or circumflex scapular, for anastomosis. If the thoracodorsal vessels are usable, preoperative axillary radiation does not result in an increased incidence of vascular thrombosis.

Post-Mastectomy Free-Flap Reconstruction of the Previously Augmented Breast. Yvonne L. Karanas, Andrew L. Da Lio, Darren Leong, James P. Watson, Kathleen Waldron, and William W. Shaw.

Post-mastectomy breast reconstruction in the previously augmented patient has become an increasingly common problem, as the augmented patient reaches the peak age of cancer incidence. These patients raise several issues. On average, they are thin, with a paucity of donor sites for autologous tissue reconstruction. They may wish to maintain their large breast size. They place an increased emphasis on the appearance of their breasts, and demand superb cosmesis. These authors evaluated their management of these patients, to determine the solutions to the above issues.

They performed a retrospective review of all previously augmented patients who were treated for breast cancer at the UCLA/Revlon breast cancer center between 1991 and the present. Patients who underwent a mastectomy were identified. Their charts were reviewed to determine their height, weight, implant size, cancer diagnosis, type and timing of reconstruction, additional treatment, and outcome. Each patient's body mass index (BMI) at the time of surgical treatment was calculated to characterize the patient's body habitus at the time of the initial surgical intervention.

Forty-one previously augmented patients underwent a modified radical mastectomy to treat their breast cancer during the study period. Thirty-eight (93%) underwent immediate reconstruction. Initially, 24 patients had autologous tissue reconstruction, and 14 had expander reconstruction. Six expander patients later underwent autologous tissue reconstruction for contracture, rupture, and poor cosmesis. In total, 30 of 38 patients (79%) had autologous breast reconstruction, and 28 (74%) had free-tissue transfer. Twenty patients (53%) had free TRAM flaps; five (13%) had free gluteal flaps; three (8%) had free TFL flaps; and two (5%) had pedicled latissimus dorsi flap reconstruction. The use of non-TRAM free flaps in this series was 21%, which is higher than the authors' average rate of 15%. The average BMI in this study population was 23.2, less than the national average of 30 for women aged 35 to 70 years. In summary, their patients were taller and lighter than the national average.

In the authors' experience, autologous free-tissue transfer provided the largest volume of tissue, with superior aesthetic results that satisfied many of the requirements of this patient population. Alternative flaps were used more frequently because of the unique body habitus of these patients.

Free TRAM Breast Reconstruction in Private Practice. David Frederick Ruebeck, Michael Cloud Fasching, and Michael Zhang.

Free TRAM breast reconstruction is a safe and effective method for correcting the deformities of mastectomy. The literature supporting this finding generally comes from teaching programs and university hospitals. These studies report the use of multiple surgical teams, simultaneous procedures in separate surgical fields, and a generous support staff to complete the procedure in a timely fashion. These authors studied their experience with free TRAM flaps in a private-practice community hospital setting. The procedure was performed by single surgeons utilizing a minimal support staff.

All free TRAM flaps performed by the senior authors from October, 1998 to November, 2000, were studied. A total of 111 free TRAM flaps were performed in 72 patients whose charts were available for review. Thirty-three cases were unilateral and 39 were bilateral. Eighty-one were immediate reconstructions and 30 were delayed. There was one total flap failure and one sub-total loss, for a flap success rate of 98%. The mean reconstructive operative time for the unilateral free TRAM was 3 hr 40 min, and for the bilateral free TRAM was 5 hr 17 min. The thoracodorsal vessels were used for 69 flaps, and the internal mammary vessels for 41 flaps. The average blood loss was 184 cc for unilateral cases, and 275 cc for bilateral cases. Five percent of the patients received a blood transfusion. The average hospital stay was 4.6 days. The major complication rate (outcomes which required further surgery) for donor and recipient sites was 15%. The average follow-up was 11.5 months.

Free TRAM breast reconstructions can be performed in a safe and effective manner by single surgeons in a community hospital setting.

Effect of Smoking on Flap and Donor-Site Complications in Pedicled TRAM Breast Reconstruction. Ivica Ducic, Scott Spear, and Frank Couco.

The effect of smoking on complications in patients undergoing pedicled transverse rectus abdominis myocutaneous (TRAM) flap breast reconstruction was studied in a retrospective review of 224 pedicled TRAMs over a 10-year period. Patients were divided into active smokers (15.5%), former smokers (17.5%; stopped smoking at least 4 weeks prior to reconstruction), and non-smokers (67%; no smoking history). There were no statistically significant differences in age, weight, radiation/chemotherapy history, flap pedicle types, timing of reconstruction, or percentage of delay procedures performed.

Overall flap complications (erythema, infection, ecchymosis, delayed wound healing, hematoma, seroma, fat necrosis, partial and complete flap necrosis) were seen in 50% of total flaps. Overall donor-site (infection, delayed wound healing, hematoma, seroma, hernia) and other (hypertrophic scars, partial nipple loss, pyogenic granuloma, DVT, pulmonary embolism, lipoma) complications occurred in 35.5% and 7.5% of patients, respectively. Approximately 20% of TRAM reconstructions were complicated by two or more flap complications, while only 2.5% of patients in the study had more than one donor-site complication associated with surgery. No patients experienced more than one ``other'' complication associated with the procedure. There was no statistically significant difference in overall (one or more) flap complication rates between subgroups; however, when compared to nonsmokers, both active and former smokers had a higher incidence of multiple flap complications (32.3% vs. 12.7% and 34.3% vs. 12.7%, p&equals;0.0138 and 0.0049,respectively).

Compared to nonsmokers, active smoking carried a relatively risk of 3.3 for having more than one flap complication, while former smokers had a 3.6 times greater risk for multiple flap complications. There was no statistically significant difference found between active and former smokers. Active smokers also had a higher rate of TRAM infection, compared to nonsmokers (14.3% vs. 3.4%, p&equals;0.0243, odds ratio&equals;4.7). For TRAM delayed wound healing, former smokers had a relative risk of 4.7, compared to nonsmokers (14.3% vs. 3.4%, p&equals;0.0485); however, the odds ratio was not significant. There were no statistically significant differences found in the incidence of donor-site or other complications among smoking subgroups.

In summary, logistic regression identified active smoking as a statistically significant risk factor for developing multiple (two or more) flap complications (p&equals;0.0061) and TRAM infection (p&equals;0.0255), while former smoking was a risk factor for multiple flap complications (p&equals;0.01) and TRAM delayed wound healing (p&equals;0.0433).

Breast Reconstruction in Older Women: Advantages of Free TRAM over Pedicled TRAM. Joan E. Lipa, Adel A. Youssef, Roy L.H. Ng, Steven J. Kronowitz, and David W. Chang.

Studies evaluating the efficacy of breast reconstruction in older patients have focused primarily on implant reconstructions. Although operative time is a consideration, these patients should not be denied the benefits of autologous reconstruction. The interest of these authors was in evaluating the morbidity of performing autogenous breast reconstruction in this older population. They compared the complication rates associated with pedicled transverse rectus abdominis myocutaneous (TRAM) flaps vs. microvascular TRAM flaps in this population.

Retrospective chart reviews were conducted for women >65 years of age who underwent post-mastectomy reconstruction with a TRAM flap, pedicled (pTRAM) or free (fTRAM), between April, 1987 and December, 2000. Data were analyzed using Stata 6.0.

Of 32 women (ages 68.4±2.8 years) undergoing TRAM flap reconstructions, 13 (40.6%) were pTRAM and 19 (59.4%) were fTRAM flaps. Nine patients (28.1%) underwent bilateral reconstruction, and 23 (71.8%) reconstructions were immediate. The mean hospital stay was greater for pTRAM (6.5±2.1 days) than fTRAM (6.1±1.2, p<0.05). The incidence of having a breast-site complication was 30.8% in pTRAM and 36.8% in fTRAM (NS). There was one (7.69%) pTRAM partial flap loss and no fTRAM losses. One medical complication (atrial fibrillation without hemodynamic instability) occurred in a patient undergoing an fTRAM, but this incidence was not statistically significant, and there were no long-term sequelae. However, there was a significant difference in the incidence of having a donor-site complication: pTRAM 7 (53.8%) and fTRAM 1 (5.3%). By multivariate logistic regression, the odds for a donor-site complication were 99.99% less in fTRAM, compared to pTRAM, being adjusted for bilateral harvest, body mass index, current smoking, ASA rank, and use of Marlex mesh. The number of additional surgical procedures to complete the reconstruction or to correct complications was higher in pTRAM (1.2±1.2) than in fTRAM (0.7±0.7). There were no perioperative deaths. The mean follow-up time was 4.2±3.6 years (0.3 to 13.4 years). 79.5% of patients undergoing pTRAMs were alive with no evidence of disease at 10 years, and 91.7% of fTRAMs were alive at 5 years.

Autogenous breast reconstruction can be achieved in the elderly population with acceptable morbidity. Microvascular TRAM flap breast reconstruction does not confer additional risk, and it may be preferred in this patient population over pedicled TRAM flap reconstruction, because it may reduce hospital stay and decrease donor-site complications.

Component Reconstruction of Difficult Poland's Syndrome: Role of Tissue Expansion and Free Flaps. David N. Sayah, William W. Shaw, and Jeff D. Hoefflin.

Congenital aplasia (hypoplasia) of the breast, pectoralis muscles, and ribs in Poland's syndrome has traditionally been reconstructed using implants with/without ped-icled latissimus dorsi flaps. Problems remain with incomplete soft-tissue and contour correction, mal-positioned nipple, skeletal defects, absence of the latissimus muscle, and multiple implant corrections in young patients. These authors' experience with staged correction of various components of the deformities was reviewed.

A retrospective review was carried out of 11 patients with Poland's syndrome who underwent staged microvascular reconstruction. Patients' medical records and pre- and postoperative photos were evaluated. Patient ages at the time of reconstruction ranged from 18 to 54 years (average: 33 years). There were 10 females and 1 male. There was an average of 2.3 implants (range: 0 to 5) prior to presentation. With these implants, 83% complained of capsular contracture, 33% of implant rupture, and 17% of implant erosion. After explantation in six patients, the skin envelope was sufficient to allow for proper positioning of the nipple-areola complex prior to free tissue transfer. Two patients with anterior chest wall depressions received Marlex mesh and simultaneous tissue expansion to lower the nipple-areola complex, while making room for a free flap, followed by a gluteal free flap.One patients with a previous rib graft had tissue expansion to the lower nipple, followed by a gluteal free flap and small custom implant for breast and chest-wall reconstruction. Another patient with minimal chest-wall depression received tissue expansion, followed by a gluteal free flap. Sixty-seven percent of patients received an ipsilateral gluteal free flap, 22% a TRAM, and 11% a deep circumflex iliac artery flap. The thoracodorsal vessels were found to be hypoplastic or atretic in 44% of patients, mandating the use of the circumflex scapular or internal mammary as recipient vessels.

Complications included one hematoma and one pneumothorax. Second-stage revision of the breast mound, donor-site liposuction, or scar revision was performed in all patients. Staged breast reconstruction, using tissue expansion and free flaps, allowed for good aesthetic outcomes in all patients. Patient satisfaction remains high, with the average follow-up being 3.2 years.

A variety of anatomic components observed in Poland's syndrome can be addressed satisfactorily to achieve an overall good result. The skin envelope and nipple position can be improved through tissue expansion. Rib defects can be stabilized with Marlex mesh or rib grafts. Contour and breast volume deficit can be corrected with autologous flaps, supplemented with occasional custom implants. Flap choices may include the latissimus, gluteal, TRAM, and DCIA flaps. The sequence of the various components of reconstruction must be coordinated from the beginning.

The purpose of this reported study was to evaluate the role of microvascular reconstructive surgery in the management of pediatric cancer patients at a major cancer center. All pediatric cancer patients who had free-flap reconstruction at M.D. Anderson Cancer Center between 1989 and 2000 were retrospectively reviewed for types of flaps used, associated morbidity, and overall outcome, including long-term functional and aesthetic results, disease-free interval, and patient survival.

A total of 43 free-flap reconstructions were performed in 39 patients (15 M, 24 F), with a mean age of 13 years (range: 4 to 18 years). Twenty-nine patients had malignant disease, mostly of a sarcoma type, and 10 patients had a benign disease. Tumor location involved the head and neck in 27 patients and the extremity in 12 patients. Free flaps used included the rectus (13), fibula (12), latissimus (6), scapula (5), and others (7). Eleven patients had received preoperative radiation therapy.

Major complications occurred in seven patients (18%), including flap loss (2), orocutaneous fistula (2), vessel thrombosis (1), infected plate (1), and meningitis (1). Minor complications (9) included minor wound-healing problems. Outcome results, including long-term functional and aesthetic outcome, disease-free survival, and overall patient survival were addressed.

Free-flap reconstruction can be performed safely, with acceptable associated morbidity, in the management of pediatric cancer patients.

The Bipedicled TRAM in Unilateral Breast Reconstruction: Microsurgical Variations on a Theme. Roy L.H. Ng, Steven J. Kronowitz, Joan E. Lipa, Adel Youssef, Gregory P. Reece, and John Potochny.

In TRAM flaps, poorly perfused zones lie furthest from the vascular pedicle. An infra-umbilical midline scar further restricts reliable perfusion to the ipsilateral hemi-TRAM. These ischemic zones are excised in most unilateral breast reconstructions. However, inclusion of these zones to achieve adequate volume/projection for a large breast requires additional vascular input. The authors examined their indications and techniques over a 10-year period, ranging from the conventional bipedicled TRAM through various permutations to the bipedicled DIEP.

They retrospectively reviewed all unilateral breast reconstructions with TRAM and DIEP flaps performed at the M.D. Anderson Cancer Center from January, 1991 through March, 2001. Reconstructions that required more than one vascular input were identified. Details concerning patient age, body mass index (BMI), brassiere cup size, operative indications, risk factors, complications, and follow-up were obtained.

Of the 863 unilateral breast reconstructions performed, there were 34 DIEP (DP), 107 unipedicled TRAM (UT), 43 bipedicled TRAM (BT), and 678 free tram (FT) flaps. Seventy-four reconstructions (8.6%) required an additional vascular pedicle (mean age&equals;48, range: 27-73; mean BMI&equals;24.9, range: 18.9-34.9; median cup range A-DDD, comprising 38 BT, 18 UT supercharged to ipsilateral deep inferior epigastric vessels (Utsc), 6 UT with contralateral FT/DP component [(U+FT)], 3 bipedicled FT/DP (BFT), 5 stacked hemi- UT (UT+UT), and 2 stacked hemi-FT/DP (FT+FT) flaps. Indications for Utsc were intraoperative flap congestion, while size/projection of reconstruction (22), infra-umbilical midline scar (27), obesity (5), smoking (7), and not stated (5) were indications for an additional contralateral pedicle.

Partial flap loss occurred in 5.3% BT (2/38) and 11.1% UTsc (2/18); fat necrosis occurred in 23.7% BT (9/38), 16.7% UTsc (3/18), and 50% UT+FT (1/2); and abdominal bulge/hernia occurred in 10.5% BT (4/38) and 5.6% UTsc (1/18), but these differences were not statistically significant. The incidence of these complications was 0% in the remaining reconstructions. The mean hospital stay was 6 days (range: 3-18 days). The mean follow-up was 24.3 months (range: 1-87 months).

A TRAM flap with two pedicles is infrequently indicated, hence, the low numbers in this series. The use of microvascular pedicles over conventional pedicles appears to benefit flap perfusion and donor-site morbidity. Any differences between TRAM or DIEP microvascular pedicles could not be ascertained.

Free Forearm Flap and its Adipofascial Tissue for Reconstruction of Oral Cancer Defects. Yur-Ren Kuo, Po-Chung An, and Seng-Feng Jeng.

The radial forearm flap has been one of the most popular flaps to reconstruct defects after oral cancer ablation. However, it sometimes may not provide sufficient soft tissue to obliterate the dead space after tumor excision and lymphy-node dissection, which results in deep wound infection of the neck or even orocervical fistula. The authors modified the radial forearm flap along with its adipofascial tissue, to prevent such postoperative complications.

From January, 1997 to December, 2000, 50 patients following ablative oral cancer surgery were studied. Twenty-nine patients (Group 1) were reconstructed with the traditional radial forearm flap, the other 21 (Group 2) were reconstructed with the radial forearm flap along with a sheet of its adipofascial tissue. The radial forearm flap was designed on the axis of the radial artery, 8x4 cm² to 12x10 cm² in size, and sufficient to resurface the intraoral defect. In Group 2, the radial forearm skin flap along with a sheet of adipofascial tissue 8x8 cm² to 12x10 cm², was used to obliterate the dead space of the oral floor and neck. The donor site of both groups was resurfaced with split-thickness skin graft. In Group 2, the original skin flap and adipofascial tissue were resutured at the original site.

The total success rate was 96%. Two flaps in Group 1 failed due to arterial occlusion, and required secondary skin flaps for reconstruction. Five patients in Group 1 developed postoperative hematoma, and needed surgical treatment for drainage. None in Group 2 had hematoma formation. Nine patients in Group 1 had neck wound infection, compared with only two patients in Group 2; this difference was statistically significant. Both groups had one orocervical fistula each. The average volume of drainage and hospitalization days were similar in both groups. Morbidity at the donor site in both groups was not significant.

The advantage of the described modification includes the following. Suitable soft tissue is available for dead space obliteration, to decrease the chances of postoperative hematoma. The important vessels in the neck can be protected. There is a decrease of neck wound infection. Donor-site morbidity is similar to that in traditionally treated patients.

Refinements in the Use of the Fibula Osteocutaneous Free Flap for Extensive Maxillary Defects. Howard N. Langstein, Ramon Llull, Rhona Jacob, David W. Chang, and Geoffrey Robb.

Reconstruction of the maxilla must adequately replace the critical bony support of the midface, but little is known about the best way to achieve this. This reported study was designed to determine the ideal geometric flap configuration for maxillary reconstruction.

A retrospective review of patients with extensive maxillectomy defects was performed, in whom the fibula free flap was used to reconstruct the lower midface. Postoperatively, 3-D images were used to critically analyze the actual clinical results. Computer-assisted design technology was then used to test potential flap geometric improvements, and to develop parameters for ideal maxillary reconstruction.

The series consisted of four cases, one with a traumatic defect and the remainder with post-oncologic deformities. Successful fibular osteocutaneous flap transfer was achieved in all patients. Two of the four cases involved primary reconstruction (immediate transoral bilateral maxillectomy and reconstruction), and two were delayed (replacement of a failed fibula flap following maxillectomy, and a traumatic maxillary defect sustained from a self-inflicted GSW to the face). In the GSW defect, the fibula flap was used as a straight strut in the upper midface between zygomas. In the remaining cases, closing osteotomies were used to fashion the fibula into an arch, which was oriented anteriorly and inferiorly in order to support the cheek, lip, and nose. The exact position for bone inset was guided by: a) a sterilizable nasal-based external prosthesis which fixed the preoperative upper lip position in space; and b) information obtained from 3-D analysis of previous results, evolving during the series. The skin paddle of the fibula closed the oronasal communication.

Results varied by placement of the fibula and whether the reconstruction was immediate or delayed. Immediately reconstructed patients had better oral competence, upper lip and nasal projection, and better overall outcomes. The GSW patient had a stable platform for subsequent nasal reconstruction, but the higher bone position did not support the upper lip and was too cranial to support a prosthesis. The other patient with delayed reconstruction had adequate bony alignment, lip and nasal support, and improved facial dimensions, but has persistent hypernasality, presumably because of soft-palate dysfunction. All patients can tolerate soft diets without difficulty, although the patients reconstructed immediately have progressed to more solid foods.

In this series, immediate reconstruction and lower angled arched placement of the fibular flap were associated with technical and clinical success. Computer-assisted design has allowed for a better undestanding of ideal flap geometry.

Role of Computed Tomography for the Preoperative Assessment of Fibular Dimensions and Implant Reliability: Clinical Perspectives. Alberto Bedogni, Giorgio De Santis, Stefano Valsecchi, Luisa Ganassi, and Pier Francesco Nocini.

The use of osseointegrated implants in reconstructed jaws enhances oral function and improves the quality of life. The fibula is one of the most suitable sources of vascularized bone for the reconstruction of bony defects of the jaws; it has been recently introduced as an elective method for the treatment of extreme alveolar ridge atrophies, in order to gain an ideal bone support for implant prosthetic rehabilitation. Nevertheless, different opinions are reported in the literature about the risks of putting implants into fibular bone, due to its size. There is not yet a reliable method which allows preoperative assessment of fibular dimensions. The aim of this reported study was to determine preoperative fibular bone dimensions and morphology by means of sequential CT scan images, in order to predict 8 mm, 10 mm, and 13 mm implant placement.

Sixty healthy adults were studied. They were recruited so as to have three age groups, consisting each of 10 males and 10 females. Subjects underwent CT scans of both legs, with 120 fibulas analyzed. The left and right fibular dimensions of each segment were compared, in order to evaluate any differences in size and implantability. If both right and left sides were considered, patients could undergo a successful implantation of 8-mm long segments at all fibular sites. A 10-mm long segment could be implanted in 28 of 30 females at the proximal fibular third, and in 100% of the males and females at the remaining sites. A 13-mm long segment could be implanted in 22, 27, and 29 of 30 males at the proximal, middle, and distal third, respectively, while the corresponding figures for females were 13, 23, and 11.

Males had a substantially greater probability of having implantable proximal and distal sites with 13-mm segments, compared to females. Gender had no effect on the implantability of the middle fibular third, and age had no influence on the implantability of any segment. Not surprisingly, weight influenced the probability of having an implantable distal, more than a middle or proximal, segment, and the same was true for BMI, which was the best predictor of implantability in the pooled sample. A significant interaction of weight and BMI with gender was found only at the distal site.

The results of this study were discussed and the clinical implications were reported.

Free-Tissue Transfer after Resection of Peri-Orbital Neoplasms. David S. Warsaw, Craig H. Johnson, Stephen J. Finical, and Uldis Bite.

Peri-orbital neoplasms range from small superficial skin lesions to extensive tumors than can even extend to the brain. Unfortunately, in many cases primary repair with local tissue is not possible after the tumor is resected. Also, exposure of bone, dura, or brain requires definitive coverage. In these cases, reconstruction with free-tissue transfer is preferred.

In a retrospective study, charts were reviewed from January, 1988 to December, 2000, of patients who underwent free-tissue transfer after resection of peri-orbital neoplasms. Thirty-seven patients fit these criteria and were included in the study. Patient age at the time of surgery, gender, diagnosis, recurrence at time of presentation, surgical complications, date of last follow-up, and cause of death were recorded.

Mean follow-up after surgery was 3.2 years. Patients ranged in age from 12 to 82 years (mean: 52.9 years). Twenty-three (62.2%) were male. The majority of patients had one of four diagnosis: sarcomas (22%), squamous-cell cancer (19%), basal-cell cancer (14%), or adenoid cystic cancer (11%). Fifteen patients (40.5%) had recurrent disease on presentation. Seven (49%) received preoperative radiation therapy. Thirty patients (81%) underwent reconstruction with a rectus flap. Ten (27%) had postoperative complications, ranging from small seromas to complete flap loss secondary to thrombosis.

One-and 3-year survival rates were 73.2% and 53.6%, respectively. All of the patients who died did so from complications of their disease. No variables were found to be statistically significant, in terms of survival.

Peri-orbital neoplasms can be a devastating disease, with slightly more than half the patients surviving 3 years. Nonetheless, reconstruction with a free-tissue transfer is an acceptable method to provide coverage after resection of peri-orbital tumors. Age, preoperative radiation, or recurrent disease should not be a deterrent to performing this type of reconstruction, which can range from simple filling of the defect and closure of the wound, to complex reconstruction of a socket capable of retaining an orbital prosthesis. The rectus abdominis muscle with overlying subcutaneous fat and/or skin was the most frequently used free-tissue transfer in this series.

Experience with Eye Reanimation Microsurgery. Julia K. Terzis and William Bruno.

This clinical study reviewed the surgical techniques utilized over the past century to correct lagophthalmos. The various procedures were analyzed and compared with the outcome of a single surgeon's experience with microsurgical reanimation of the paralyzed eye sphincter.

The study was a retrospective chart review of 133 patients who underwent eye reanimation procedures by the same surgeon from 1978 to 2000. The criteria for eligibility in the study were dependent on the length of follow-up for each type of procedure, and the availability of proper video/photographic documentation of each patient. Of the 133 patients reviewed, 117 met the inclusion criteria for the study. Of the 474 procedures performed, 389 met the inclusion criteria. Sixty-eight percent of the patients were females, and 32% were males. The mean age of the population was 26 years, and the majority of their facial paralysis resulted from neoplastic (39%) and developmental (33%) etiolgies. The mean denervation time (duration of the paralysis prior to reinnervation) of each patient was slightly greater than 10 years. Nearly half (45%) of all patients underwent previous surgical procedures at other institutions.

The restoration of eye sphincter function involved various static, as well as dynamic (muscle and/or nerve transfers), microsurgical procedures. These included prosthetic implants (gold weights and eye springs), mini-tendon transfers, neurotizations, pedicle muscle transfer, and free microneurovascular muscle transfers. An independent panel of four reviewers evaluated pre- and postoperative videos documenting eye closure and involuntary blinking. These reviewers were asked to judge the functional and aesthetic outcome of both eye closure and blink, and to grade each on a 5-point scoring system. The senior surgeon was not involved in the video assessment process.

Of all patients rviewed, 97% had a positive improvement in their postoperative eye closure scores, with 94% achieving a moderate or better postoperative result. Regarding blink scores, 97% of patients had improved postoperatively, and 87% had a moderate or better outcome. The mean improvement of all patients for eye closure and blink score was statistically significant (p<0.001).

The patients who underwent dynamic procedures more often yielded better results than those who did not. These procedures included innovative surgical procedures introduced by the senior surgeon, namely, free platysma transfers and frontalis pedicle muscle transfers.

Free Vascularized Flap and Flap-Plate Options: Comparative Results of Reconstruction of Lateral Mandibular Defects. Eyal Gur, Thomas Shpitzer, Patrick J. Gullane, Peter C. Neligan, Johnathan C. Irsh, Jeremy E. Freeman, and Michael Van den Brekel.

Reconstruction of the mandible and oral cavity after segmental resection is a challenging surgical problem. Although osteocutaneous free flaps are generally accepted to be optimal for reconstruction of anterior defects, the need for bony reconstruction of a purely lateral mandibular defect remains controversial.

A retrospective comparative study of short and long-term outcomes of three different reconstruction techniques for lateral defects was carried out. In total, 62 patients were included, of whom 30 had a plate and pedicled pectoralis major myocutaneous flap (PMMF), 17 had a plate and free radial forearm flap (FRFF), and 15 had an osteocutaneous free flap. Functionality, flap failure, and complications were scored.

Plates had to be removed in 8 of the 30 pedicled pectoralis major myocutaneous flaps, 2 of 17 radial forearm flaps, and none of the 15 osteocutaneous free flaps failed. The difference was borderline statistically significant (p&equals;0.055). Long-term functional outcome revealed no statistically significant difference in oral deglutition (p&equals;0.76) or in facial contour (p&equals;0.36). Oral continence was significantly better in patients treated with free radial forearm flaps (88.2%), compared to pedicled pectoralis major myocutaneous flaps (53.3%) or osteocutaneous free flaps (46.7%, p&equals;0.02). On the other hand, the results for speech favored the osteocutaneous free flap group: 14 of 15 (93.3%) scored normal, compared to 13 of 17 (76.5%) for the plate and radial free flap group, and 18 of 30 (60%) for the plate and pedicled pectoralis major myocutaneous flap group. However, this represented a borderline statistically significant result (p&equals;0.06).

For lateral mandibular defects, the osteocutaneous free flap is reliable and durable at long-term follow-up. However, in a selected group of patients, the flap-plate option is a viable reconstructive option.

Reconstruction of Defects at the Scalp, Forehead, and Temporal Area after Tumor Therapy. Barbara S. Lutz, Per-Erik Janson, and Allen Rezai.

The aim of this reported study was a retrospective evaluation of reconstructive options for covering scalp and forehead defects after tumor therapy, regarding functional and aesthetic outcomes. Midface, intraoral, and neck defects were not included in the study.

Ten consecutive patients, aged 65.8±9.2 years, were treated with free microvascular tissue transplantation for reconstruction of scalp and forehead defects after tumor therapy. Histology of the tumors included squamous cell cancer (8), angiosarcoma (1), and radionecrosis following irradiation of an angiosarcoma (1). Four patients presented with tumor recurrences after previous surgery, irrradiation, and selective chemotherapy. The defects involved the scalp with bone exposure in all patients, and dura defects in two patients. Exenteration of the orbit (1) and ablation of the ear (2) were performed in three patients. The average extension of defects was 171 cm (range: 40 to 600 cm). Free flaps employed for reconstruction included the anterolateral thigh flap (7), suprafascial radial forearm flap (1), lateral arm flap (1), and latissimus dorsi muscle flap (1). Two patients required dura repair with fascia lata at the skull base.

There was no flap failure. Donor-site morbidity was negligible. Secondary procedures such as tissue expansion are planned in one patient after radiation ulcer coverage. Two patients could not undergo a radical surgical procedure because of tumor ingrowth into the brain stem. One of these died 4 months postoperatively. Hospitalization stay averaged 9.3±0.6 days.

Free skin flaps, such as the anterolateral thigh perforator flap or the suprafascial radial forearm flap, are recommended for reconstruction of tumor-related defects in the non-hair-bearing areas in the forehead or temporal region, and for large scalp defects. The customized harvested myocutaneous anterolateral thigh flap is proposed as an excellent option for covering defects which consist of both deep and superficial areas, and in vascularly compromised areas such as irradiation ulcers.

Ten-Year Experience in Jaw Reconstruction and Dental Rehabilitation with the Fibular Flap: Pitfalls and Complications, How to Treat Them. Giorgio De Santis, Alberto Bedogni, Massimo Pinelli, and Pier Francesco Nocini.

These authors discussed their experience with the use of the fibula free flap for mandibular or maxillary reconstruction and implant-prosthetic rehabilitation. The report focused on the immediate and delayed complications associated with complex dentofacial rehabilitation.

From December, 1989 to April, 2001, 55 patients were treated for mandibular or maxillary reconstruction with vascularized fibular flaps. Fifty-two of 55 fibular flaps healed primarily, with a success rate of 94.5%. Three failures occurred. One hundred fifty fixtures were inserted into fibular bone, and 30 implant-supported prostheses were manufactured. Long-term follow-up ranged between 6 months and 10 years (mean follow-up: 46.5 months) of functional loading. Seven implants were lost. One was removed because of instability. An additional two implants were removed because of infection of the mini-plate used for fixation of bony segments. Three implants were lost because of an uncorrected masticatory loading and one implant was lost because of loading associated with a malposition. The implant success rate was 95.5%.

A retrospective analysis of the series was attempted, to detect the main complications related to bone and implant surgery, and to establish how correction was possible without compromising the results. Immediate complications were always vascular, such as bleeding and/or arteriovenous thrombosis. Thrombosis may be sustained by compression of the pedicle, due to hematoma formation, or by masticatory movements. Infections and free-flap necrosis always occurred without a second surgical revision of the anastomoses. Fixation plays an important role in bone healing and callus formation, since unstable bone segments are more prone to non-union. Incorrect fixation is responsible for rupture of plates. In 2 of 52 patients, plate fracture caused delayed healing at the osteotomic site.

Delayed complications are due to a number of conditions that may alter long-term aesthetic and functional results. They are due to incorrect surgery as regards bone orientation, stabilization, and final volume; incorrect implant surgery as regards the type, number, and spatial positioning of fixtures, along with the lack of primary stability and/or ossseointegration; or the incorrect management of peri-implant soft tissue.

Technical Problems in the Reconstruction of the Voice Tube after Total Laryngopharyngectomy. Hung-chi Chen.

The patient with cancer in the pharynx and larynx may refuse surgical intervention, due to the fear of losing the voice. There are several methods of voice reconstruction; however, all are far from perfect. Since the jejunal flap has been introduced, it has been applied for reconstruction of both the cervical esophagus and the voice tube after total laryngopharyngectomy. The author has investigated four types of voice tubes created by a second loop of the transferred jejunal flap, comparing wound-healing problems, secretions, rigidity of the wall of the voice tube, loudness of the voice, voice pitch, and problems in eating or aspiration.

The features of an ideal voice tube were found after comparative study: proximity of its lower end to the permanent tracheosteomy; soft and redundant skin around the tracheostomy; small lumen of the tube; small junction between the tube and the pharynx; an antiperistaltic limb; a limb with antigravity; and minimal secretion but sufficient for self-cleansing functions, to maintain the patency of the voice tube.

It was concluded that a voice tube constructed of jejunum can be improved to obtain better vocal function and food intake, with a minimal chance of aspiration. This solution is more physiologic than a voice prosthesis or external device.

Superchaarged Pedicled Jejunum for Reconstruction of Total Esophagectomy Defects. Ayman Nabawi, Roy L.H. Ng, and Michael J. Miller.

Restoration of gastroesophageal continuity following total esophagectomy may be accomplished by gastric pull-up or pedicled transposition of the colon or jejunum. Potential ischemia in the transposed segment of stomach or bowel may be preempted by the technique of supercharging, that is, microanastomosis of the mesenteric blood vessels in the distal transposed segment to vessels in the neck or chest. These authors reported their early experience with supercharged ped-icled jejunal transpositions for primary and secondary total esophageal reconstruction.

Between May, 2000 and April, 2001, six patients (mean age: 55.4 years; range: 41 to 71 years, M:F&equals;5:1) underwent total esophageal reconstruction with supercharged jejunum at the M.D. Anderson Cancer Center. Indications for esophagectomy were carcinoma (5) and Barrett's esophagus (1). There were three primary and three secondary reconstructions (two failed gastric pull-ups, 1 re-operation for positive margins following successful gastric pull-up). The path of the jejunal segment was retrocardiac in three patients and retrosternal in three patients. All proximal bowel anastomoses were performed end-to-end with a single layer of interrupted 3-0 Vicryl sutures. Distal bowel anastomoses were performed in a Roux-en-Y fashion. Four were stapled and two were sutured in two layers. Feeding jejunostomy tubes were inserted in all patients. The recipient vessels were the internal mammary (4), and the superior thyroid artery and internal jugular vein (2). Postoperatively, all patients were ventilated and monitored in the intensive care unit.

One patient developed a proximal anastomotic leak on POD 10, and another developed a mid-jejunal leak unrelated to either anastomosis on POD 7. Both leaks resolved spontaneously, but the latter patient subsequently required dilatation of a stricture 6 months postoperatively. A third patient developed an esophagocutaneous fistula in the neck, following successful resuscitation from cardiac arrest on POD 3, and remains on tube feeding pending further surgery. The mean hospital stay was 19 days (range: 11 to 43 days). By the time of discharge, four patients had re-commenced oral intake. At follow-up (mean: 7 months, range: 3 to 12 months), three patients had resumed a normal diet, while two still required feeding-tube supplemantation of their oral intake.

The authors' early experience with the supercharged jejunum in total esophageal reconstruction was favorable. However, a longer follow-up is required to establish the final functional outcome in terms of dietary intake.

Vascularized Bone Transfers for the Management of Allograft Non-Union in Cancer Patients. Roy L.H. Ng, Sanjay K. Sharma, Steven J. Kronowitz, Kristin L. Weber, Alan Yasko, and Michael J. Miller.

Cadaver bone allografts afford effective limb salvage in treatment of extremity sarcomas, but are associated with non-union at the allograft-host bone interface or at sites of late allograft fractures in up to 42% of cases. Failure to successfully achieve osseous healing can lead to extremity amputation. The authors reported the use of vascularized bone transfers to salvage non-healing intercalary allografts used in extremity reconstruction following ablative cancer surgery.

Between February, 1996 and November, 2000, four patients were treated for non-uinion of cadaver allografts by vascularized fibula transfer. Patients included one male and three females, with an average age of 30 years (range: 13 to 40 years). Time to diagnosis of non-union was 7 months (range: 5 to 8 months), radiographically and confirmed intraoperatively. Non-unions involved the allograft-host bone interface in the femur (n&equals;2) or humerus (n&equals;1). One patient experienced two fracture non-unions in the same allograft segment of the tibia at different locations and times. Both were treated with vascularized fibula: the first was transferred as an ipsilateral pedicled flap, and the second as a contralateral free flap.

Microvascular transfers were revascularized using end-to-side sutured anastomosis or a microvascular coupler device to nearby vessels. No vein grafts were required. Bone was inset to the region of non-union in an intramedullary position (i.e., inlay) or in contact with the cortical surface (i.e., onlay) and rigidly fixed. The allograft was supported with intramedullary rods and/or anti-rotation hardware. Evaluation of osseous healing was by serial radiographic examination, and functional recovery was noted for each patient. Surgical success was 100%, but there was one case with delayed donor-site healing. The patients were reported as to operative technique, function, radiographic findings, healing, and final result.

Vascularized bone transfer can promote osseous healing in cases of allograft non-union in cancer patients. Bone may be transferred as pedicled flaps or free-tissue transfers, and used to bridge the site of non-union. Improved allograft healing is likely due to the native osteogenic properties of bone flaps. These techniques provide an important option to achieve limb salvage in patients treated for extremity sarcomas.

Intermedullary Rod fixation vs. External Fixation for Treatment of Grade IIIB Tibial Fractures Requiring Microvascular Free Flaps. Geoffrey Furtner, Wendy-Ann M. Olivier, Jamie Levine, and Oren Lerman.

In recent years, there has been an increased use of intermedullary rods (IM rod) for treatment of open tibial fractures. This paper compared IM rods to conventional external fixation (ex-fix) in patients with open tibial fractures requiring microvascular free flaps (MVFF) for soft-tissue closure.

A retrospective chart review was performed on all patients at the authors' institution with Grade IIIB tibial fractures requiring MVFF between 1995 and 2000. There were a total of 25 patients, with an average age of 36 years. Mechanisms of injury were MVA (80%), GSW (8%), crush (8%), and fall (4%). Free-flap donor sites were the rectus (65.4%), latissimus (30.8%), and gracilis (3.8%). There was one flap failure.

Fourteen patients underwent ex-fix, and 11 patients underwent IM rods. Thirteen limbs in the ex-fix group were salvaged (92.9%), and 10 limbs in the IM rod group were salvaged (91.0%). The average time to MVFF was 13.5 days for the ex-fix group (57% within 7 days) and 8.2 days for the IM rod group (90% within 7 days). The average number of I&Ds was two for both groups. Bone gaps were observed in six patients in the ex-fix group (average gap: 5.3 cm) and six patients in the IM rod group (average gap: 4.7 cm). Primary bone reconstructions were accomplished using bone grafts in eight patients, distraction osteogenesis in two patients, and free fibula grafts in two patients. Non-union was observed in five patients in the IM rod group (50%) and in two patients in the ex-fix group (15.4%).

The IM rod group had a higher incidence of wound infection, infected bone graft, and osteomyelitis (30%, 20%, 30%, respectively) than the ex-fix group (0%, 0%, 15.4%, respectively). There was one pin-site infection in the ex-fix group, which was treated with local pin care and antibiotics.

Many studies report excellent results using the IM rod for the treatment of open tibial fractures. However, very few patients in these studies require an MVFF for closure of their open wounds. This reported study showed inferior results for IM rod stabilization of Grade IIIB tibial fractures in the subgroup of patients who required MVFF for soft-tissue coverage. There was an increased incidence of non-union, osteomyelitis, and infection in those patients treated with the IM rod, compared to those treated with ex-fix.

``Tap Flap'': Anatomic Basis and Results of Tibial Artery Perforator Flaps. Raymond Ray Dunn, Adam Vernadakis, and Michael Diaz.

Advances in understanding the anatomic basis of blood supply to fasciocutaneous flaps based on single transmuscular or transsseptal perforators have extended the applications of DIEA-based free flaps and gluteal artery free flaps for breast reconstruction. The concept of perforator flaps in local/regional reconstruction in general, and from the tibial artery system in particular, has not been fully exploited. Distal middle third and lower third tibial wound defects remain a challenge for reconstructive surgery, with limited options using local/regional flaps. The posterior tibial artery vascular territory supplies a large lower territory, and can be isolated on two or three major septal perforators. Localization and exploitation of the vascular territory associated with these perforators may allow for extended applications of local/ regional fasciocutaneous flap reconstruction in the distal lower extremity.

Preoperative duplex and intraoperative Doppler examination was carried out in patients with tibial wound defects undergoing reconstruction, to isolate two to three major perforators in the mid distal third of the leg. Flaps were designed as island, superior-based fasciocutaneous flaps, isolated on single perforators down to the tibial axis, with mobilization allowing flap advancement and rotation to reconstruct defects which might ordinarily require free-flap reconstruction. Fresh cadaveric dissections (n&equals;4) using methylene blue to outline the vascular territories were presented.

In four patients undergoing island tibial artery perforator flaps, there was 100% flap viability, primary wound healing, and superior donor-site and operative outcomes. The cadaveric dissections demonstrated four (±1) major cutaneous perforators from the posterior tibial artery to the overlying skin territory. The largest perforators (>5 mm) were located at approximately 6 and 10 cm above the tip of the medial malleolus.

A conceptual understanding of the cutaneous blood supply based on perforators of the tibial artery in the lower extremity should allow for greater flexibility in local/regional lower-leg fasciocutaneous flap design, allowing wounds of the distal tibia, which might otherwise require free-flap reconstruction, to be covered with well-vascularized local/regional fasciocutaneous flaps, with a shorter operating time and equivalent outcomes. The ``tap flap'' represents an additional reconstructive option for consideration in lower extremity defects.

Assessing the Effectiveness of a Reconstructive Algorithm for Dorsal Foot Wounds. Samir Mardini, Karen Kim, and Christopher Attinger.

Reconstruction of dorsal foot wounds often poses a challenge to the reconstructive surgeon in choosing the option that yields the optimal functional and aesthetic result. Available options range from allowing the wound to heal by secondary intention to more complex measures such as free-flap reconstruction. These authors sought to assess the validity of a reconstructive algorithm based on the size and depth of the wound in dorsal foot wound reconstruction, utilized at their institution.

All consecutive dorsal foot wounds treated by the senior author at Georgetown University Medical Center between February, 1991 and March, 2000, were reviewed. Forty-six patients were treated with 85 procedures. Five patients were lost to follow-up. All patients were serially debrided and infection-free prior to planning the reconstruction. The etiologies of the wound were infection (34%), trauma (24%), post surgical (26%), ischemia (7%), and unknown (9%). The size of the wound ranged from 0.15 cm² to 180 cm², with an average size of 40 cm². Co-morbidities included diabetes (54%), peripheral vascular disease (30%), coronary artery disease (26%), and renal failure (9%). Twenty-four percent of patients were smokers, and 24% underwent revascularization. Modes of reconstruction included secondary intention (15%), delayed primary closure (4%), skin graft (41%), local flap (18%), and free tissue transfer (22%). The complication rate was 13%: skin graft loss (7%), partial flap loss (4%), and hematoma (2%). Four percent of patients underwent below-knee amputation. A successful reconstruction was achieved in 95% of the patients.

The algorithm used over the last 10 years has been successful in healing most dorsal foot wounds seen at the authors' institution. For superficial wounds (without exposed tendon or bone) that are less than 4 cm² in size, secondary intention or delayed primary closure was optimal. In superficial wounds greater than 4 cm², a skin graft was more appropriate. For deeper wounds with exposed tendon or bone, local, pedicled, or microsurgical free flaps were used. For defects that were small and were located in the proximal part of the foot, local and pedicled flaps were preferentially used. For large wounds and for more distal foot defects, free-tissue transfer was used.

Role of Microsurgery in Orthopedic Oncology: Three Years of Experience with 39 Cases of Limb-Sparing Surgery. Eyal Gur, Aharon Amir, David Leshem, Tommy Shpitzer, Arik Zaretski, Dean Ad-El, Yoav Barnea, Raphael Shafir, Yehuda Kolander, and Issak Meler.

Orthopedic oncology deals with the preservation of functional limbs after resection of bone and soft-tissue malignant tumors. In order to avoid amputation, but without compromising cancer surgery principles, large bone, joint, and soft-tissue defects are created. Microsurgery plays an over-growing role in the reconstruction of these defects. This is done primarily (during ablation of the tumor) or secondarily (delayed procedures after ``spacers'' placement), or for complications such as infections of implants.

From January, 1998 to April, 2001, 39 patients were treated. There were 25 males and 14 females, with a mean age of 20 years (range: 8 to 42 years). The different diagnoses were osteosarcoma (11), Ewing sarcoma (20), and high-grade soft-tissue sarcoma (8). The anatomic locations of the tumors were femur (25), heel (3), tibia (4), fibula (1), radius (4), forearm (1), and humerus (1). In 11 patients, the reconstruction was done primarily and, in 28, a delayed reconstruction was done. Twenty-six patients had their fibula transferrred to provide bony reconstruction. In 15 patients, the fibula was used for the reconstruction of an intercalary resection of the femur, in four for intercalary resection of the tibia, in two for knee arthrodesis, in four for distal radius, and in the last for humerus reconstruction. Eight patients had soft-tissue reconstructions. In six, the latissimus dorsi flap was used as a muscle-only or myocutaneous flap. The scapular flap was used once for the reconstruction of a complex forearm defect in a young girl, and the gracilis muscle was used once for shin coverage.

In this series, the flap transfer success rate was 95%, where bone viability was assessed by radioisotope scanning at day 10 after surgery. Bone viability was further assessed clinically and radiologically at follow-up. In a 3-year to several month follow-up interval, no major complications could be reported, except for one patient with failure of his fibula plating system that was reinforced. All the fibula patients show, so far, an established union of their osteotomy sites, with no case of infection or major flap-related functional deficit. Several patients were reported with various reconstructive needs, to detail the problems and microsurgical solutions for orthopedic oncology.

Tensor Fasciae Latae Perforator Flap for Trochanteric Defect Coverage: Anatomy and Surgical Technique. Luis Henrique Ishida, Alexandre Mendonca Munhoz, Eduardo Montag, Gustavo Strutz, Luis Carlos Ishida, Hugo Alberto Nakamoto, Fabio Lopes Saito, and Marcus Castro Ferreira.

Considering techniques with less morbidity for trochanteric defect coverage, these authors carried out an anatomic study, followed by a clinical series using this flap. Forty-eight thighs on 24 fresh cadavers were dissected. The parameters measured included location, distribution, diameter, and length of the perforating vessels. In addition, an initial clinical study was performed in six patients with trochanteric pressure sores.

The study revealed a reliable presence of perforator pedicles anterior to the greater trochanter, which provides an adequate rotation arc for harvesting a flap, without sacrifice of the underlying muscles. This flap is nourished by perforators from the asscending branch of the lateral circumflex femoral artery, which arises from the deep femoral artery and runs through the intermuscular septum, tensor fasciae latae, or rectus femoralis muscles. The flaps were raised based on perforators located preoperatively, using a unidirectional Doppler. Good results were obtained with primary closure of the donor area without surgical complications.

This flap is an alternative for myocutaneous flaps, as it preserves local musculature, without functional sequelae in patients who deambulate. It also preserves the local anatomy for musculature use in cases of recurrence, as usually seen in plegic patients with pressure sores.

Classification and Prospective Evaluation of Complications in Free-Flap Perfusion. Riccardo E. Giunta, Andreas Geisweid, Edgar Biemer, and Axel-Mario Feller.

The aim of the reported study was to present a new clinical classification of disturbances of the circulation in free flaps, and to assess the results of treating complications at each of the stages described. Within a period of 29 months, 194 free microvascular flaps were prepared and evaluated in a prospective study. Postoperative monitoring was carried out from a purely clinical point of view. A distinction was made between arterial and venous circulatory disturbances and, in both cases, four stages were characterized according to the time required for recapillarization, the color of the transplant, and the bleeding on puncture.

Complications occurred in 69 cases (36%), including intraoperative revisions during the first operation. The ratio of arterial to venous insufficiency was 33:35. In the presence of arterial disturbances of the circulation, the proportion of the losses increased in progressive stages to 41%. With venous disturbances, the losses which occurred in stages I and II were predominantly partial. In 164 cases (85%), the flaps remained undamaged. In 15 cases (8%), there was a partial loss of the transplant, and in a further 13 cases (7%), complete loss had finally to be accepted.

The authors' results confirm that the presented clinical classification is indeed a measurement of the severity of circulatory impairment. Despite the relatively high complication rate, the use of various methods of treatment finally leads, in the overwhelming majority of cases, to a positive result for the patient.

Microgurgery in the New Millenium: Endoscopic-Assisted Microsurgery and its Realistic Expectations. Mohamed El-Shazly, Assem Kamel, Mostafa El-Sonbaty, Mohamed Zaki, and Rudiger Baumeister.

The long hours spent in microsurgery looking through the operating microscope, have always been accepted as the price paid for working in this surgical subspecialty. With the introduction of endoscopy, it has become apparent that magnification similar to that provided by the surgical microscope can be achieved, with the advantage of operating at a distance.

An experimental study was carried out with 120 SD rats classified into six groups. In the first two (Groups 1,2), the femoral vessels and nerves were sectioned and anastomosed in 20 animals, using the operating microscope and, by the application of an endoscopic unit, in another 20 animals. In the second two groups (Groups 3,4), groin flaps based on the femoral pedicles wer elevated as free flaps to be anastomosed in the neck with the common carotid and internal jugular vessels, using the microscope in 20 animals and the endoscope in the other 20. The thoracic ducts of the third two groups (Groups 5,6) were divided and anastomosed, using the microscope in 20 animals and the endoscope in the other 20 animals.

All data from the 240 vascular, neural, and lymphatic anastomoses of the 120 animals were collected and assessed. There were no technical differences between use of the microscope and the endoscope, regarding the depth of focus, diameter of the field, or image quality. Also, there was no clinically significant difference regarding the patency rate and free flap viability. In this series, the authors recorded a mean time of 27, 19.6, and 27.4 min for the vascular, neural, and lymphatic endoscopic anastomoses, respectively. These results were statistically significantly shorter than those of the microscopic anastomoses (p<0.0005), which recorded means of 36.5, 25.7, and 31.9 min for the vascular, neural, and lymphatic anastomoses, respectively. In addition, the authors recorded in the endoscopically-assisted groups a significant reduction in the total operating time by more than 50 min, and in vessel preparation time by more than 20 min (p<0.0005).

The endoscope seems to be a promising new tool in performing microsurgery with better technical aspects, regarding the magnification and visualization, significantly shorter anastomotic time, better handling, and more physical comfort for the surgeon. It is reliable, time-efficient, and cost-effective. As to the feasibilites and difficulties of this project, the authors have established some technical modifications, making the endoscope more compatible for being a clinically accepted operating microsurgical instrument in the future.

Incorporating Perforator Flaps into an Established Microsurgical Practice: A Single Surgeons' Experience. Frederick J. Duffy, Jr.

Perforator flaps represent a significant advance in microsurgery. Despite the expanding literature on this novel approach to free-flap dissection, most plastic surgery residents in the United States have no exposure to this type of surgery during their training. Can the principles and techniques of perforator flaps be learned and safely incorporated into an established microsurgical practice? One surgeon's initial 15-month experience with perforator flaps was presented.

Forty peforator flaps were performed during the 15-month period following attendance at the Munich perforator course in November, 1999. Twenty-eight were free flaps and 12 were pedicle flaps. Fourteen of the free flaps were used for lower extremity reconstruction, 11 for breast, two for head and neck, and one for upper extremity. The pedicle flaps were used for groin, sacral, and ischial wounds and, in five cases, for a novel approach to coverage of large myelomeningocele skin defects. One pedicle flap and two free flaps failed. The mechanism of failures was discussed in terms of whether or not they could be attributed to use of the perforator flap technique.

Proponents of perforator flaps believe that they represent a major advance in reconstructive microsurgery. However, wide acceptance of these flap techniques will require a greater number of surgeons using this approach. This small series suggests that perforator flaps can be safely incorporated into an established microsurgical practice. Some suggested guidelines for this process were presented.

Prelaminated Free Radial Forearm Skin Flap with Buccal Mucosa Tube for Urethral Reconstruction in a Hypospadias Compromised Child. Yoav Barnea, Aharon Amir, Yaakov Ben-Chaim, David Leshem, Arik Zaretski, Jerry Weiss, Raphael Shafir, and Eyal Gur.

Surgical reconstruction for complicated, recurrent failed hypospadias repair can present a great challenge to the surgeon, because there may not be any local tissue for further reconstruction attempts. The authors presented a 13-year-old case of hypospadias. In another country, the patient underwent no less than 10 previous failed hypospadias repairs. His penis was contracted and embedded partly into his scrotum. He had severely scarred ventral, dorsal, and scrotal skin, penile deviation, and a long urethral defect with a scrotal meatus and remnants of glanular urethra.

The patient underwent reconstruction in two stages. In the first stage, the radial forearm skin was prelaminated with a 7-cm-long, tubularized buccal mucosa graft from both oral walls. In the second stage 3 weeks later, the penile shaft scar tissue was totally removed from both corpora cavernosa, the deformed meatal orifice was refreshed, and the prelaminated radial forearm skin flap was transferred to the penile area. The neo-urethra was anastomosed to the proximal meatus and the distal remnant of the glandular urethra, and was covered by the radial skin flap. The radial artery, venae comitantes, and cephalic veins were anastomosed to the femoral vessels with no length defect. Postoperatively, the patient developed urethral fistula and stricture at the distal urethral anastomosis, which was successfully repaired 3 months later.

At 1 year, functional and cosmetic results are satisfactory. The patient returned to normal urination and reported normal erectile function.

CO₂ Laser Soldering of Arteriotomy Incisions in Rats Controlled by a Temperature Regulation Fiberoptic System. David Leshem, Avi Ravid, Tamar Vasilyev, Aharon Amir, Arik Zaretski, Yoav Barnea, Andrea Gat, Noam Kariv, Avraham Kazir, and Eyal Gur.

Laser soldering of tissue is a method of biologic tissue adherence, facilitated by a solder material and energy. This is a technique that firmly binds tissue, is wateright, introduces no foreign body, is faster than conventional suturing, and requires less skill to master. The authors assumed that the quality of the weld depends on the ability to monitor and control the surface temperature of the welded zone during the procedure. Their objective was to use a ``smart'' fiberoptic welding system, with temperature control, for sealing arteriotomy incisions in rat blood vessels, compared to the standard suture technique.

A CO₂ laser delivered light to the weld site through a silver halide fiber. A second fiber monitored the infrared light emitted from the tissue. This fiber was connected to a radiometer and measured surface temperature that was controlled by a computer. Albumin solder was applied to the arteriotomy. Laser soldering was carried out on arteriotomy sites in 24 rat femoral veins, with a surface temperature of 55 to 65°C, for a period of 5 sec. Another 24 rats served as a control group and their arteriotomy sites were sutured under the operating microscope using 10-0 nylon sutures. The study and control groups were divided into two subgroups (12 rats each): one was reoperated on at day 7 and the other at day 21. Patency was assessed clinically and histologically.

The patency rate for both study and control groups was 85% at days 7 and 21. Histopathologic studies showed no thermal damage and less inflammatory and foreign-body reaction than those caused by standard suturing.

The success rate in these preliminary experiments demonstrated that temperature-controlled laser soldering can produce effective and rapid sealing of blood vessels. This ``smart'' system will be used for further trials in microvascular anastomosis.

Effect of Anti-Lymphocyte Serum and Cyclosporine A Treatment on Induction of Donor-Specific Tolerance in Rat Hindlimb Allografts. Maria Siemionow, Kagan Ozer, Ramadan Oke, and Rita Prajapati.

Induction of tolerance to composite tissue allografts is the ultimate goal in reconstructive surgery. In this study, the authors reported the long-term results of rat hindlimb allograft transplantation under an immunomodulation protocol of Cyclosporine A (CsA) and anti-lymphocyte serum (ALS).

Hindlimbs were transplanted between Lewis-Brown-Norway (LBN) and Lewis (LEW) rats in six groups of six animals each. Groups 1 and 2 served as isograft and allograft controls without treatment. Groups 3 and 4 received single therapy with CsA and ALS for 3 weeks, respectively. In Groups 5 and 6, animals had combined treatment of CsA and ALS. The therapy in Group 5 was for 3 weeks and in Group 6 for only 1 week. CsA was started a night before surgery with a dose of 16 mg/kg, and tapered to a half every week. ALS was also started a night before transplantation, and 0.4 ml/day was administered every day within the first week, every other day within the second week, and every 3 days during the third week. Levels of CD3, CD4, CD8, CD90, NK, and TCR were determined by flow cytometry, and mixed lymphocytic reaction (MLR) assay was performed on long-term survivals. At 8 weeks, long-term allograft survivals were challenged with skin grafts from the donor (LBN), recipient (Lew), and third party (ACI) to evaluate tolerance.

In Group 1, isografts survived indefinitely. In Group 2, all limb allgorafts rejected between 5 and 9 days. In Group 3, extended survival wass achieved up to 20 days. In Group 4, rejection was observed within 3 to 5 days after the end of treatment. In Group 5, allografts were accepted without a need for chronic immunosuppression for more than 210 days. In Group 6, a significant increase in allograft survival up to 24 days was achieved in five animals, and up to 100 days in transplants.

Long-term survivals were found to be tolerant, as confirmed by acceptance of donor-specific skin grafts, and rejection of third-party skin grafts. T cells were uniformly depleted between 10 and 21 days and recovered at 64 days. In MLR analysis, lymphocytes were functionally tolerant to host and donor-strain alloantigens, yet competent to respond to third-party alloantigens in each case.

The 21-day protocol validated that the combination of CsA and ALS allowed for induction of tolerance across the MHC barrier without the need for chronic immunosuppression. This protocol may have an important clinical application in the treatment of composite tissue allograft transplants.

Breast Reconstruction with the Free TRAM or DIEP Flap: Patient Selection, Choice of Flap, and Outcome. Maurice Y. Nahabedian and Paul Manson.

Recent reports of increased fat necrosis and venous congestion within the DIEP flap have created uncertainty regarding the benefit of this flap. The purpose of this study was to establish selection criteria for the free TRAM and DIEP flaps, in order to minimize flap morbidity.

A total of 163 free TRAM or DIEP flap breast reconstructions (107 unilateral and 28 bilateral) were performed between 1997 and 2000. Of the 143 free TRAM flaps, 38 included the full width of the rectus abdominis (MS-0), 31 spared the lateral muscle segment (MS-1), and 74 spared the medial and lateral muscle segments (MS-2). Patient mean age was 49.4 years, and mean patient weight was 176 pounds. Tobacco use was documented in 23 of 118 women (19.5%). The mean follow-up was 19.5 months. Of the 20 DIEP flaps (MS-3), 14 were unilateral and three were bilateral. The mean patient age was 46.4 years, and the mean patient weight was 147 pounds. Tobacco use was documented in two of 17 women (11.8%). The number of perforators per flap was one (17 flaps), or two (three flaps). The mean follow-up was 8.2 months ( 4 to 15 months).

Morbidities related to the 143 free TRAM flaps included return to the operating room for 11 flaps (7.7%), flap loss in five (3.5%), fat necrosis in 14 flaps (9.8%), and mild venous congestion in two flaps (1.4%). Morbidities related to the 20 DIEP flaps included return to the operating room in two flaps (10%), one flap loss (5%), fat necrosis in two flaps (10%), and mild venous congestion in no flap. Flaps with fat necrosis occurred in women with a mean age of 51 years, active tobacco use in 12.5%, and a mean body weight of 182 pounds. Fat necrosis occurred in three of 38 MS-0 flaps (7.9%), four of 31 MS-1 flaps (12.9%), seven of 74 MS-2 flaps (9.5%), and two of 20 MS-3 flaps (10%).

The occurrence of fat necrosis appears to be independent of patient age, degree of muscle sparing (MS), and tobacco use. There is a trend toward increased fat necrosis in heavy women who require a large-volume breast reconstruction with a greater zone 2 component. Selection of the free TRAM or DIEP flap is based on patient weight and breast volume requirements, as well as on the number, caliber, and location of the perforating vessels.

Reconstructive Microsurgery in the Lymphatics. Corradino Campisi and Francesco M. Boccardo.

The authors reported the modern surgical management of peripheral lymphedema. For an appropriate selection of candidate patients for lymphatic microsurgery, an adequate diagnostic route is essential. It must include, in addition to patient history and clinical examination, lymphangioscintigraphy, an accurate study of venous circulation and, if required (in case of angiodysplasia), the arterial circulation. Based on the authors' 30-year clinical experience (more than 1000 patients), the role of lymphatic microsurgery was particularly emphasized. Derivative lymphatic microvascular procedures have currently their most exemplary application in multiple lymphaticovenous anastomoses (LVA). For those cases in which a venous disorder is associated with a more or less latent or manifest lymphostatic pathology, of such severity to contraindicate a lymphatic-venous shunt, reconstructive lymphatic microsurgery techniques have recently been developed (autologous venous grafts or lymphatic-venous-lymphatic-plasty, LVLA). With a follow-up from 5 to 15 years after surgery, positive results have been achieved in more than 80% of cases, with the best evidence among patients at stages II or III who have undergone operative procedures at an early stage. Long-term clinical outcomes lead to the conviction that currently, reconstructive microsurgery can be advantageously used to treat primary or acquired lymphodemas, even in pediatric patients.

Ring Avulsion Injuries: Treatment Outcome with Microsurgical Techniques. Darrell Brooks, Karin Schott, Rudolf Buntic, Gabriel Kind, Gregory M. Buncke, and Harry J. Buncke.

Microsurgical techniques have drastically improved the outcome in salvaging ring finger avulsion injuries. These authors presented the prognosis and treatment outcome of 84 ring avulsion injuries.

A retrospective review of a database revealed 84 patients with ring avulsion injuries. These digits were graded according to the Urbaniak classification, along with subsequent authors' modifications of this classification, including Kay and Adani. Twenty-three digits were deemed non-replantable (no graftable target), and underwent immediate amputation. Sixty-one digits underwent replantation. Eleven patients required vein grafting. Five patients had venous flow-through flaps for simultaneous soft-tissue coverage and revascularization. Eleven patients required secondary surgery. Outcome included survival and function; functional assessment included active range of motion, sensation, and grip strength. The average follow-up was 10 months (range: 3 to 25 months). The data were presented, with the Urbaniak system used to classify ring avulsions.

The study demonstrated excellent survival and good functional outcome after microsurgical treatment of seleted ring avulsion injuries. Survival appeared to be the norm for digits, regardless of the classification. Modifications of the Urbaniak system were based on anatomic injury and were more predictive of functional outcome. Surgeons should be aware that worsening class is not necessarily predictive of survival, but may be helpful as a descriptive tool and possibly in predicting eventual function.

Association of Vascularized Fibula and Massive Allograft in Skeletal Reconstruction of the Extremities. Marco Innocenti, et al.

In the late eighties, an original technique was developed for skeletal reconstruction after long intercalary resection of malignant bone tumors in the extremities. The procedure is based on the association of a massive allograft with a vascularized fibula autograft. The rationale of the ``combined graft'' is to associate the advantages offered by the mechanical resistance of a massive allograft with the biologic properties of a vascularized fibula graft.

Between 1988 and 2000, 59 patients were treated using this procedure. Reconstruction was carried out on the tibia in 40 cases, on the femur in 16, and on the humerus in three. In tibial reconstruction, a concentric assembling of the two components was preferred in all cases. A longitudinal slot is made in the allograft, and the medullary canal is curetted until the fibula fits. A stable osteosynthesis is then achieved by means of plates distally and compression screws proximally. In reconstruction of the femur, the fibula can either be inserted in the reamed medullary canal of the allograft, or placed medially in a parallel dual assembling.

The biologic properties of a vascularized fibula graft allow for early union at the level of the osteotomy. When associated with an allograft, this also enhances the allograft union. Weight-bearing induces fibular hypertrophy. Thus, the fibula can support the allograft when it is weakened by the resorption process. Using this procedure, even in young growing patients, intercalary reconstruction is possible in those cases in which intra-epiphyseal resection allows for saving the epiphysis, and in other conditions in which a minimal fixation of the graft is mandatory.

In two recent cases, the same concepts have been applied in the reconstruction of the proximal humerus in young adults. This type of reconstruction is more challenging because of the articular surface of the allograft. In order to improve the possibilities for revascularization of the humeral head, the epiphyseal cancellous bone is completely removed and replaced with autologous cancellous bone augmented with a buffer coat containing growth factors and stem cells. The vascularized fibula is inserted into the allogenic humeral head and in the medullary canal of the recipient humerus, obtaining a sort of ``biologic prosthesis'' that provides good functional results.

Experimental Model of Facial Transplant. Luis Eduardo Bermudez, Alida Santamaria, Tomas Romero, and Diego Fernando Calderon.

Treatment of severe facial soft-tissue deformities is a challenge. Even using state-of-the-art surgical techniques in the treatment of these extensive deformities, like the one described by Angriniani (1997), the final results are not ideal, because facial skin is unique and there is no other skin like it in the human body. Composite tissue allotransplantation has recently been used to restore limb function in humans and could be used for facial reconstruction. Hand-transplant procedures have had some success, and might serve as a background, but the only facial replantation that has been described was that of Thomas and Obed in 1998.

These authors were interested in developing an experimental model in animals to study the functional recovery of the transplanted face and to obtain expertise in various techniques. They also wished to determine if the complete hemi-face could be transplanted, based only on the facial vessels. The study is not an immunologic transplant study, although there have been several relating to immunology in composite tissue allotransplantation in animals and humans.

A hemifacial transplant was performed between two dogs. The data collected from such a procedure would have the same value as those collected from a complete facial transplant, but with less morbidity. To transfer the skin necessary for flap inclusion, at least a 3-cm muscle strip was required along the course of the facial artery, the orbicularis oculi muscle, and the forehead muscles. The recipient dog receeived immunosuppressive therapy from 10 days before surgery (cyclosporin 4 mg/kg and prednisone 1 mg/kg/day), to avoid hyper-acute rejection of the transplant.

The hemifacial transplant was successful. The recipient dog was kept alive during 7 days, and the flap was perfusing normally over 5 days. The flap was acutely rejected (skin biopsies) by days 6 and 7 postoperatively. The authors' hypothesis was validated: facial transplantation can be safely accomplished by using only the facial artery and external jugular vein as pedicles.

Effect of Tirofiban on Microvascular Thrombosis (Crush Model). York J. Yates, Carlos L. Farias, Jr., Charles L. Puckett, and Matthew J. Concannon.

Microvascular anastomotic thrombosis is a significant clinical problem, particular in crush and avulsion injuries. Platelet deposition plays a particularly significant role in the initiation and propagation of microvascular thrombosis; thrombin has little effect in the acute phase of thrombus formation. Nevertheless, heparin, a specific thrombin inhibitor, remains the most widely used microvascular irrigant. The purpose of this reported study was to evaluate Tirofiban HCI (Aggrastat), a glycoprotein IIb-IIIa inhibitor, and its role in preventing postoperative thrombosis in a crush anastomosis model. This class of drug blocks the initial step of vascular thrombosis-the attachment of platelet glycoproteins to their vessel-wall ligands.

A crush injury animal model using the rat femoral artery was utilized (Reichel, 1987). Three ``clicks'' of a Webster clamp were applied directly on the vessel for 60 sec prior to dividing the artery. An end-to-end microvascular anastomotic repair was then performed using 10-0 nylon suture. Either 1 cc of Tirofiban in saline (50 ug/ml) or 1 cc of saline only was randomly used to irrigate the vessel lumen prior to placement of the last suture. The same irrigant was used within both femoral arteries in each rat, to avoid contralateral interaction from systemic absorption. The vessels were re-examined 24 hr postoperatively, and the patency assessed. Any surgical complications, such as hematoma formation or excess bleeding, were also noted and recorded.

A total of 35 vessels in 20 rats were utilized in the study. The patency rate of the experimental group using Tirofiban was 11/16 (68.8%), while that of the control group was 2/19 (10.5%, p<0.05). No hematomas or other hemorrhagic complications were noted in either group.

This study demonstrated a significant improvement in patency with Tirofiban irrigation in a crush anastomosis rat model. There appeared to be no increased incidence of hemorrhagic complications associated with its use. Clinically, this may have usefulness as a potent anticoagulant, particularly in anastomoses that have been subjected to crush/avulsion injury.

Microvascular Reconstruction of a Massive Thoraco-Abdominal Wall Defect Allows Aggressive Resection for Cure. Joan E. Lipa, Roy L.H. Ng, W. Roy Smythe, and Pierre M. Chevray.

These authors reported what they believed to be the largest described defect of the thoracic and abdominal walls. A multidisciplinary approach achieved a structural and functional reconstruction.

A 36-year-old albino male presented with a 15×20-cm recurrent baso-squamous-cell carcinoma of the epigastric skin which had eroded through the sternum and was abutting the pericardium and liver. A metastatic workup was negative. Curative resection was undertaken with frozen section control.

Due to adjacent microscopic tumor rests, the resulting excision was larger than anticipated, measuring 26×38 cm (988 cm²), with the skin edges pulled inward. The full-tickness defect spanned from the suprasternal notch to 1 cm above the umbilicus, and from the midclavicular line on the right to the anterior axillary line on the left. The resection included the entire sternum, right anterior ribs 1-10, left anterior ribs 3-10, the right clavicle, anterior pericardium, central diaphragm, and upper abdominal wall musculature and peritoneum.

Functional restoration allowing independent ventilation was the primary goal, followed by coverage of exposed vital structures. The 10×10-cm diaphragmatic defect was repaired with a 2-mm-thick Gortex soft-tissue patch. A neosternum with emanating rib-like structures made of methylmethacrylate sandwiched between layers of Marlex mesh, was used for rigid chest-wall reconstruction. To prevent abrasion of exposed epicardial veins, Preclude pericardial membrane was sutured to the mesh undersurface. Due to the central truncal nature of the defect, no local pedicled flaps were adequate for coverage. A combined right free latissimus dorsi, serratus anterior, and 221×9-cm scapular flap was elevated for chest-wall coverage. Removal of the right medial clavicle and first rib allowed access for end-to-side anastomoses of the subscapular to the innominate vessels. Next, the abdominal wall was reconstructed with a 33×13-cm right free tensor fasciae latae (TFL) myofasciocutaneous flap, anastomosed to the right gastroepiploic vessels. Meshed split-thickness skin grafts closed the TFL donor site and covered the serratus anterior and latissimus dorsi flaps. Mechanical ventilation was continued until competance for spontaneous ventilation was assured. The patient was easily weaned off ventilator support on POD 10.

Three months following surgery, the patient was ambulatory, ventilating without difficulty, tolerating a regular diet, and living independently. This report demonstrated that expertise in free-tissue transfer, coupled with accessibility to a multitude of alloplastic materials, extended the limits of what is technically resectable. This allows aggressive resections with intent to cure.

Free Inferior Gluteal Myocutaneous Flap Harvest with Sparing of the Posterior Femoral Cutaneous Nerve. John A. Millard and Michael R. Zenn.

Loss of posterior thigh and popliteal sensibility is a frequent occurrence during isolation of the inferior gluteal artery (vascular pedicle) while harvesting the free inferior gluteal muscle-musculocutaneous flap. The posterior femoral cutaneous nerve of the thigh lies directly on the deep surface of the gluteus maximus muscle, running medial to the sciatic nerve in the gluteal area, and having a very close anatomic relationship with the inferior gluteal artery. The purpose of this study was to gain a better understanding of the anatomy of the posterior femoral cutaneous nerve (PFCN) and its relationship with the inferior gluteal artery (IGA). Eight fresh human pelvic halves were dissected, in order to examine the relationship of the PFCN to the IGA during harvesting of the inferior gluteal muscle as a free muscle flap.

Seven of eight pelvic halves had at least a portion of the PFCN and/or its branches intact after isolation of the IGA pedicle. One of eight of the pelvic halves had the entire PFCN and its branches intact after isolation of the IGA pedicle. One of eight pelvic halves required complete transection of the PFCN and its branches in order to isolate the IGA pedicle.

The authors demonstrated that, in a majority of cadaveric dissections, it is possible to maintain at least a portion of the PFCN and/or its branches intact after isolation of the inferior gluteal artery pedicle, while harvesting the free inferior gluteal muscle-musculocutaneous flap.

Routine Use of CT Angiography in Microsurgical Planning. Matthew B. Klein, Geoffrey Rubin, James Chang, and Larry Chow.

Preoperative angiography is a standard component of microsurgical reconstruction. The goals of preoperative angiography are to identify vascular anomalies and peripheral vascular disease which would alter the surgical plan. However, there are several drawbacks to traditional angiography, including lengthy acquisition time, expense of study, and the risks of arterial thrombosis and pseudoaneurysm.

Ten consecutive patients undergoing free flap transfer for reconstruction of lower extremity (n&equals;4) and head and neck (n&equals;6) defects, received preoperative CT angiograms, to assist with the planning of their reconstructive procedures. All patients underwent standard history and physical examination, including peripheral pulse examination. The average patient age was 44.2 years. All patients had normal creatinine and no history of allergy to intravenous contrast. All images were obtained using a Siemens Somatom Plus 4 Volume Zoom. All studies were read by staff radiologists and evaluated for arterial anatomy, presence of vascular disease and, in the case of lower-extremity reconstruction, relationship of leg vessels to the zone of injury.

Ten patients underwent preoperative CT angiography prior to microvascular reconstruction. There were no complications from the radiographic procedure or surgical procedure. All studies provided detailed three-dimensional images of arterial, as well as soft-tissue, anatomy. The surgical plan of one patient was changed, based on the study results.

The importance of preoperative angiography in microsurgical reconstruction has been well-established. The principal drawback to routine angiography is the risk of complications associated with arterial puncture, including vessel occlusion, vessel dissection, pseudoaneurysm, and hematoma. With the improvement in the quality of computer tomographic imaging, CT angiography is capable of providing high-resolution vascular imaging, without the need for arterial puncture. Furthermore, CT angiograms can allow for easy three-dimensional image reconstruction, as well as the ability to image soft tissue and bone, which can be particularly helpful in determining the relationship of potential recipient vessels with zone of injury. When compared with traditional angiograms, the acquisition time and examination cost are significantly lower.

Flap Delay: Optical Scanning of Flap Microcirculation. Danniel Reichner, Socorro Gutierrez, Thomas Scholz, Victoria M. VanderKam, Gregory R.D. Evans, and Earl Steward.

The purpose of this study was to test a new method of determining blood flow in skin flaps. Traditionally, flap blood flow has been measured by various methods, including clinical assessments of color, temperature, capillary refill, transcutaneous oxygen saturation, and fluorescein dye perfusion. This study examined a novel way to measure flap perfusion after surgical delay, examining skin vascularity by optical scanning.

Twenty rats were grouped into controls (Group 1) and surgical delay (Group 2). Surgical delay procedures were performed on day 0. Flap elevation was performed in both groups on day 7. Measurement of flap survival and vascularity was performed in both groups on day 14. Determination of flap survival was based on percentage of flap loss, histology, evaluation of microcirculation flow, vessel size, and thrombosis, by orthogonal polarization spectral (OPS) imaging scanning (Cytometrics, Inc.).

Data demonstrated that the OPS scanning device effectively measured areas of compromised vascularity in control and delay flaps. The control group had 33% flap survival and the delay group had 56% flap survival. These measured areas of skin viability correlated with the immediate post-elevated OPS scanned control and delay flaps.

The authors found that OPS scanning was an effective, noninvasive method to measure changes in microcirculation. Data also suggested that optical scanning is a noninvasive means to measure differences in blood flow and tissue perfusion in delayed skin flaps. Correlation between observed and measured changes compared with histology were discussed. Clinical applications include real-time monitoring of free-flap perfusion and intraoperative design of local flaps.

Review of the Implantable Doppler Probe for Intraoperative and Postoperative Microvascular Surveillance. Jorge I. de la Torre, William Hedden, Joel S. Solomon, Jobe Fix, John H. Grant, Paul M. Gardner, and Luis O. Vasconez.

Reliable monitoring of free-tissue transfer flaps allows for early identification of microvascular failure and increases the opportunity for flap salvage. Currently, there are several techniques to monitor free-tissue transfer in the postoperative period. The development of the internal Doppler allows for reliable monitoring, both intraoperatively and postoperatively.

The charts of 118 patients who underwent free-tissue transfer were reviewed. All patients underwent reconstruction using free-tissue transfer by multiple surgeons at a single institution. Microvascular anastomoses were monitored using the 20Hz internal Doppler probe; this included both arterial and venous anastomoses. Patients were monitored by the nursing staff either in the ICU for 1 day, or went directly to the floor. Monitoring was continued on the floor for approximately 1 week, using a predetermined protocol.

Over a 2-year period, 260 anastomoses were monitored with the internal Doppler (118 arterial and 142 venous). False positive results were noted in six cases (4 intraoperative and 2 postoperative). In eight cases, true positive signals were observed (2 intraoperative and 6 postoperative), which resulted in one flap loss. The free-flap success rate was 99% and the re-exploration rate was 8%, resulting in a salvage rate of 83%.

The internal Doppler offers an easy and reliable way to monitor microvascular free-tissue transfer both intraoperatively and postoperatively. The improved reliability of intraoperative monitoring allows for revision of the anastomosis at the initial surgery, and increasing flap success rates. In addition, early detection of flap failure permits an improved rate of flap salvage. The internal Doppler following free-tissue transfer is of particular value for surgeons who are early in the course of a microsurgical career.

Novel Technique for Additional Length and Improved Exposure of the Gracilis Pedicle. Kent Victor Hasen, Richard E. Tepper, Miguel L. Gallegos, and Gregory A. Dumanian.

The elevation of the gracilis muscle for use as a free flap is not difficult, but one must dissect beneath the adductor longus muscle to obtain a suitable pedicle length for the microvascular anastomosis. These authors described a technical modification that allows a single surgeon to dissect a long pedicle, without the need for specialized lighting or additional retractors.

After making the standard incision, the gracilis and the adductor longus muscles are identified and separated from each other and the surrounding muscles. The gracilis is easily freed on its superficial surface, and both proximal and distal portions of the muscle are divided, based on length of muscle needed. The dominant pedicle is exposed on the deep surface of the gracilis muscle. Intramuscular branches to the adductor longus are carefully divided. A key to the modified technique is wide elevation of the adductor longus from its fibrous connections to the surrounding adductor muscles on both sides of the vascular pedicle. The innovation involves early division of the origin and insertion of the gracilis muscle, allowing its passage under the adductor and into the previously developed anterior space. With the gracilis in this new position, the final dissection of the pedicle is more easily accomplished.

This technique permits a direct approach and visualization of the proximal aspect of the dominant pedicle, without the need for a headlight and without ``working in a dark hole'' beneath the adductor muscle. Beside the obvious advantage of greater pedicle length, one can frequently obtain a larger single vein from the merging of the venae comitantes, which aids in the microvascular anastomosis of the gracilis free flap.

Breast Reconstruction using the Free Deep Inferior Epigastric Perforator Flap: Study of the Perforator Vessels Intramuscular Anatomy and Clinical Application. Alexandre Mendonca Munhoz, Luis Henrique Ishida, Eduardo Gustavo Arruda, Gustavo Sturtz, Eduardo Montag, Fabio Lopes Saito, Fabio Busnardo, and Marcus Castro Ferreira.

Through anatomic study and initial clinical experience, the relationship between the lateral and medial perforators in the intramuscular trajectory was comparatively studied, establishing the main anatomic parameters in the dissection of the deep inferior epigastric (DIEP) flap. A total of 30 flaps from 15 fresh cadavers were used in the study. The following parameters were measured: location, diameter, intramuscular course of the perforating vessels, and length of the perforating vessels and pedicle. In addition, an initial clinical study was performed in 12 patients, using the DIEP flap in breast reconstruction.

One hundred ninety-one perforating vessels were detected (6.3 per flap). Sixty-six percent were located at the medial row, 14.2% of large diameter. In the lateral row, a rectolinear course was observed in 79.17% of the perforators, while only 18.1% of the perforating vessels in the medial row presented this configuration (p&equals;0.001). In the clinical study, all flaps were based on two perforating vessels, without the presence of muscle.

The majority of the perforators of the lateral chain presented a rectolinear intramuscular trajectory. This anatomic characteristic favors their dissection with reduced operating time and vascular lesion morbidity, constituting an important anatomic parameter in the harvesting of the DIEP flap.

Effect of Pre- and Post-TRAM Radiation on Pedicled TRAM Complications. Ivica Ducic, Scott Spear, and Frank Cuoco.

The effect of pre- and post-reconstruction radiation therapy on complications in patients undergoing pedicled transverse rectus abdominis myocutaneous (TRAM) flap breast reconstruction was studied in a retrospective review of 224 pedicled TRAMs over a 10-year period. Patients were divided into those who received pre-TRAM radiation (38 patients, 42 TRAMs) post-TRAM radiation (34 patients, 38 TRAMs), and a control group who never received radiation or chemotherapy (TRAM only, 78 patients, 91 TRAMs). There were no statistically significant differences in age, weight, radiation/chemotherapy history, flap pedicle types, timing of reconstruction, or percentage of delay procedures performed.

Compared to patients who received no neoadjuvant or radiation therapy, pre-TRAM radiation patients had a statistically significant difference in timing of reconstruction (70.3% vs. 35.7% unilateral, respectively; p&equals;0.0003). There was also a significant difference in timing of reconstruction for TRAM only vs. radiation post-TRAM patients (70.3% vs. 97.4%, respectively; p&equals;0.003). Overall flap complications (erythema, infection, ecchymosis, delayed wound healing, hematoma, seroma, fat necrosis, partial and complete flap necrosis) were seen in 50% of total flaps. Overall donor-site complications (infection, delayed wound healing, hematoma, seroma, hernia) and other complications (hypertrophic scars, partial nipple loss, pyogenic granuloma, DVT, pulmonary embolism, lipoma) occurred in 35.5% and 7.5% of patients, respectively. Approximately 20% of TRAM reconstructions were complicated by two or more flap complications, while only 2.5% of patients in the study had more than one donor-site complication associated with their surgery. No patients experienced more than one ``other'' complication associated with the procedure.

There were no statistically significant differences found in the incidence of donor-site, overall, and multiple flap, or other complications among the radiation and neoadjuvant therapy subgroups. The only difference that reached statistical significance was for TRAM ecchymosis: patients receiving radiation pre-TRAM had a higher incidence of ecchymosis than the patient control group (19% vs. 6.6%; p&equals;0.0374, relative risk &equals;2.6). Logistic regression analysis did not identify pre- or post-TRAM radiation as a statistically significant risk factor for any flap or donor-site complication.

Thoracoacromial Trunk for Recipient Vessels: A Lifeboat in Head and Neck Reconstruction. Lawrence J. Gottlieb, Alex Kaplan, Kerstin Stenson, and Loren Schechter.

Obtaining adequate recipient vessels for microvascular anastomosis in free-tissue transfer may be difficult in the re-operative head and neck cancer patient. Recipient neck vessels may have been sacrificed during cancer extirpation or previous failed reconstructions. In addition, scarring due to prior surgery and/or radiation may preclude safe dissection of recipient vessels. The thoracoacromial artery and its venae comitantes are alternative vessels that can be safely used in such situations.

The authors presented a series of seven patients in whom eight sets of thoracoacromial vessels were used as recipient vessels for free-tissue transfers. Previous transfer of the pectoralis major muscle as a pedicle flap for head and neck coverage did not preclude the use of the thoracoacromial vessels. Use of the thoracoacromial trunk for recipient vessels for free-tissue transfer is a reliable option in re-operative head and neck reconstructive surgery.

Complex Lower-Extremity Reconstruction in Young Children-When Free-Tissue Transfer is not an Option. Jeffrey Friedman, Randy Sherman, and Larry Hollier.

Free-tissue transfer remains a reliable and often preferred method for lower-extremity reconstruction, even in the young pediatric age group. However, there are situations when free-flap coverage is contraindicated, due either to recipient vessel unavailability or the potential for unacceptable donor-site morbidity. The authors presented their experience with these cases and the outcome of the reconstructions.

A retrospective review of two patients treated since 1977 was performed. Both patients were less than 3 years of age and presented with complex ankle deformities which prevented unassisted ambulation. Patient 1 had congenital arthrogyposis of both lower extremities, requiring surgical release of the Achilles tendons and ligamentous structures of the ankles. Latissimus dorsi free flaps were thought to be contraindicated in light of the potential for crutch use in the future. Therefore, bilateral posterior leg fasciocutaneous flaps were serially elevated and rotated to cover the associated ankle defects. Patient 2 developed full-thickness skin loss in the posterior ankle region secondary to meningicoccemia. Healing was by secondary intention and resulted in a fixed equines deformity, which prohibited normal ambulation. Following soft-tissue release and repositioning of the ankle to a neutral position, flow in the posterior tibial and peroneal arteries could not be demonstrated. As a result, a proximally-based posterior thigh flap was designed and used to cover the defect. Three weeks later, the flap was divided and inset. Both patients (three wounds) healed uneventfully and now ambulate without assistance.

Circumstances do exist in which free-tissue tranfer to the lower extremity in children is either ill-advised or technically impossible. The authors proposed two novel approaches for the use of fasciocutaneous flaps to treat these difficult wounds with good functional results.

Urethral Substitution Using an Intestinal Free Flap: Report of a Novel Approach. Lawrence J. Gottlieb, Dimitri Kuznetsov, Gregory T. Bales, and Kaveh Alizadeh.

Urethral reconstruction has undergone significant evolution since Russell described the first modern approach to surgery of the urethra in 1914. Today, a broader understanding of penile blood supply and the application of tissue-transfer techniques has improved the overall success in treating urethral strictures. Despite these advances, there remains a select group of patients with extensive urethral disease or urethral loss, in whom traditionally available reconstructive options are inadequate. Many of these patients have previously failed or are otherwise poor candidates for skin or mucosal grafts, and are also poorly suited for local or regional vascularized tissue transfers.

The authors reported a case which introduced a new dimension in urethral reconstruction heretofore not available. They have successfully used a tailored jejunal free-tissue transfer to reconstruct the proximal 16 cm of a severely diseased urethra in a 32-year-old male. The report detailed the technical considerations utilized in performing this procedure. The technique represents an entirely unique approach to urethral reconstruction, and should find a ready application for urethral stricture cripples and other men with extensive urethral loss.

Reconstruction of Upper and Lower Lips and Mandible with Multiple Flaps Following Resection of Extensive Squamous-Cell Carcinoma. Roy Lip Hin Ng, John Potochny, Jeffrey Myers, and David W. Chang.

A 56-year-old white male sought treatment for an extensive squamous-cell carcinoma involving upper and lower lips. Examination revealed an ulcerating exophytic lesion involving the entire lower lip, right two-thirds of the upper lip, floor of the mouth, buccal mucosa, and skin of the chin and right cheek. The tongue was mobile and unrestricted. There was numbness in the mental area, and bilateral cervical lymphadenopathy at levels I and II. CT scan revealed destructive changes in the right mandible back to the angle, and enlarged lymph nodes to level III. There was no evidence of distant metastases. The patient was staged as T4N2cM0.

Composite resection with a 2-cm margin was performed of both lips and the mandible from left mid-body to right angle, with right level I-IV and left level I-III selective neck dissections. Two centimeters of left upper lip were preserved. A free fibula osteocutaneous flap with 14 cm of bone and a 20×8-cm skin paddle was used to reconstruct the mandible, intra-oral lining, and lower lip externally. Microvascular anastomosis was performed to the facial vessels. A palmaris longus tendon sling was used to suspend the lower-lip reconstruction. The right upper lip was reconstructed with a pedicled scalp flap to provide hairy skin externally, and a pedicled tongue flap to provide vermilion and mucosal lining. Finally, a residual skin defect over the chin was reconstructed with a pectoralis major myocutaneous flap. Three weeks later, pedicle division and insetting of the scalp and tongue flaps were performed.

At 6 weeks postoperatively, the patient had intelligible speech, was tolerating a soft diet, and was looking forward to resuming work as a laborer, after completion of his adjuvant radiotherapy. Pre- and postoperative videos of function were presented.

Reconstruction of Pectus Excavatum Using Microvascular Free-Tissue Transfer. Jerome Donald Chao, Jeffrey R. Marcus, and Geoffrey C. Fenner.

Reconstruction of chest-wall deformities is a challenging problem, and best illustrated by the reconstruction of Poland's syndrome. However, the most common congenital deformity of the chest wall-pectus excavatum-is a more complicated deformity. It can have significant functional consequences related to compressive effects on the lungs and displacement of the heart. The traditional approach to such deformities includes cartilage resection, sternal osteotomy, and internal fixation. Mild cases include those in which the patient is asymptomatic. However, in these cases, the contour deformity still can be quite severe, and many surgeons have turned to custom implants to correct the deformity. But the use of autologous tissue has distinct and obvious advantages. These authors presented a case report of a female patient with significant deformity who had chest-wall and breast asymmetry corrected using a buried free TRAM flap.

An 18-year-old female presented with an untreated pectus excavatum. Her complaints centered on the contour deformity, as well as malposition of her breasts. After being presented with a number of options for contour correction, the patient and her family favored autologous reconstruction. The use of a buried TRAM flap was planned. Due to an insufficient superior epigastric artery, a de-epithelialized free TRAM was performed. There were no complications. Pre- and postoperative photographs were presented.

As illustrated in this case, pectus excavatum can result in significant soft-tissue deformity, when uncorrected. Manifestations may be more problematic for females due to the potential for significant breast asymmetry and malposition. The case presented, in which autologous tissue was utilized, yielded excellent results as well as high patient satisfaction. The use of autologous tisssue avoids the need for secondary procedures due to displacement of prostheses, seroma, and infection. The primary disadvantage relates to the donor site. Furthermore the technique removes one reconstructive option, should the patient require breast reconstruction in the future. Ultimately, any decision for re-contouring in this patient group should be made only after considerable patient counseling. The patient's lifestyle, expectations, and aesthetic concerns should guide in the decision-making process.

Segmental Intrathoracic Esophageal Reconstruction Using a Free Jejunal Tissue Transfer. Lawrence J. Gottlieb, Carl-Christian A. Jackson, Walter J. Chwals, and Kaveh Alizadeh.

In 1877, Czerny described the first esophageal replacement for reconstruction following cervical esophagectomy. Since that time, various conduits have been tried with varying degrees of success. These include gastric pull-up, mobilized jejunal segments, gastric tubes, colonic interposition, pedicled and free skin tubes, and free jejunal transfers. Although free jejunal tissue transfers have become standard for cervical esophageal replacement, current practices tend to focus on colonic interposition or gastric tube esophageal reconstruction, when the intrathoracic esophagus is involved in children. Intrathoracic free jejunal transfers are generally not performed, due to the difficulty in monitoring vascularity.

These authors presented a unique solution to the problem of severely scarred mid-esophageal segment in an 8-year-old autistic male. He developed a severe, persistent 7-cm mid-esophageal stricture following lye paste ingestion. Gastrostomy tube feedings were required for adequate nutrition, and Collis gastroplasty with Nissen fundoplication were performed for gastroesophageal reflux secondary to esophageal foreshortening. Frequent esophagoscopy and dilations did not provide lasting patency. Previous surgery precluded the use of a gastric pull-up, and colonic interposition was not chosen as a reconstructive option due to the required specialized feeding regimens that this autistic child could not follow.

Therefore, an anatomic reconstruction of this child's mid-esophagus was performed using an interpositional free jejunal tissue transfer, modified to allow for easy vascular monitoring. The transferred jejunal vessels were anastomosed to the right 8^th intercostal artery and azygous vein, and a temporary remote monitoring segment, attached to an extended mesenteric arcade, exited the thoracic cavity below the right scapula. Normal feedings were initiated within 7 days, and an 11-month postoperative endoscopic evaluation demonstrated a widely patent anatomically normal-appearing esophagus, without scarring. This case represented the first reported pediatric mid-esophageal reconstruction using an intrathoracic free jejunal tissue transfer.

Free-Flap Reconstruction as Long-Distance Visiting Microsurgeons. Barbara S. Lutz and Jan B. Wieslander.

During a period of 7 months, 17 free-flap reconstructions were performed by either one of the authors as long-distance (320 and 450 English miles, respectively) visiting microsurgeons.

Free-flap transplantations were performed for head and neck reconstructions in 16 patients, and for soft-tissue reconstruction in the lower leg in one patient. There were 13 males and four females, with an average age of 62 years. Free flaps included fibula with (6) and without skin paddle (1), radial forearm flap (4), anterolateral thigh flap (4), crista iliaca (1), and myocutaneous latissimus dorsi flap (1). Nine patients underwent resection of a primary head and neck tumor. Four patients were operated on for tumor recurrences, one for radionecrosis, two for fistulas after previous mandible reconstruction with free fibula, and one for coverage of a traumatic soft-tissue defect in the lower leg. Seven patients were irradiated before free-flap reconstruction, and three patients received a second free flap after previous mandible reconstruction with a free fibula 1 to 3 years preoperatively.

No acute reexploration was necessary. There was no complete flap loss. One partial skin loss after free osteomyocutaneous fibula transplantation healed spontaneously in one case without further sequelae. In a second patient, a pedicled trapezius flap and hyperbaric oxygen therapy healed a persistent orocutaneous fistula in a severely irradiated case. A small oronasal fistula after maxilla reconstruction healed uneventfully.

The microsurgeon arrived 1 day before reconstructions and stayed 1 to 3 days postoperatively. Preoperative investigations and postoperative care were done by the local plastic surgeons (who had no microsurgical experience), the ENT surgeons, and the nurses, following the regimen given by the respective microsurgeon. Although preoperative investigations and postoperative treatment varied between the two microsurgeons, individual guidelines were followed meticulously. Contact via e-mail or telephone was always possible.

Microvascular reconstructions performed by a long-distance visiting microsurgeon, have proved to be a reliable and safe option. This may be of advantage in remote areas with no local microsurgeon available, thus avoiding long transports of patients and the accompanying high costs.