Abstracts - Part I

When Not to Perform the Deep Inferior Epigastric Perforator (DIEP) Flap. Alex Keller. North Shore Long Island Jewish Health System, Great Neck, NY.

The DIEP flap was introduced in 1989 and first reported for use in breast reconstruction in 1994. The TRAM flap was introduced a decade before the DIEP flap. Those actively doing TRAM and free TRAM flaps have occasionally encountered situations in which the technical raising of a conservatively sized flap went well, yet there was a large area of flap necrosis. Recently, with the introduction of the DIEP flap, reports have questioned its reliability and the need for drainage through the superficial venous system. The superficial inferior system is often quite small, making microsurgical transfer difficult. The author reported an algorithm for preoperative planning of a reliable free autologous abdominal flap breast reconstruction.

All patients undergo Doppler examination for possible DIEP flap breast reconstruction. In addition, the superficial epigastric system is evaluated at the inferior border of the proposed flap. With experience, one will be able to discern a loud sound that can be followed as a large useful perforator. Similar conclusions can be drawn for the superficial system.

In situations in which the perforators were large and the superficial system was small, a DIEP flap without any augmentation was appropriate. In situations in which the perforators were few, low in volume and small, and the superficial system was loud and large, the superficial system was most appropriate for transfer. With the experience gained from having done over 200 DIEP flap breast reconstructions, a preoperative determination of the dominant system was possible over the last 9 months by Doppler examination, and was confirmed in intraoperative findings.

The clinical conclusion was that the DIEP system is the dominant blood supply to the lower abdominal skin and subcutaneous tissue most of the time but, in a few predictable situations, the superficial system is a better system for free flap transfer.

Outcomes Following 170 Consecutive Deep Inferior Epigastric Perforator (DIEP) Free Flaps for Breast Reconstruction. Thaddeus Sebastian Fabian and Bernard W. Chang. Mercy Medical Center, Baltimore, MD.

Breast reconstruction using autologous tissue and muscle-sparing technique has become increasingly popular. The deep inferior epigastric perforator (DIEP) free flap is one of the latest refinements in the use of abdominal soft tissue to reconstruct the breast, while preserving the rectus abdominis muscles. Free flap breast reconstruction has been shown to have many advantages over conventional pedicle TRAM flap surgery; however, controversy has arisen regarding the complication rates of DIEP flaps vs. free TRAM flaps. This study evaluated the outcomes of 170 consecutive DIEP flaps performed in 132 patients in a community hospital setting.

Thirty-eight patients had bilateral procedures. Immediate reconstruction was performed in 95 cases (56%), and 75 cases (44%) were delayed reconstructions. The mean age of patients was 50 years (range: 26 to 74 years). The mean hospital stay was 3.36 days. Risk factors were present: smokers (25%), previous radiotherapy (22%), and obesity (30%). The thoracodorsal vessels were the preferred inflow for the DIEP flaps which were based on one (41%), two (28%), three (25%), four (4%), or five (0.6%) perforators.

The mean postoperative follow-up period was 15.9 months (range: 3 to 34 months). Total flap necrosis occurred in 7 patients (4.1%), while partial flap loss occurred in 10 patients (5.9%). Fat necrosis occurred in 18 patients (11.1%). No abdominal hernias were noted, but 2 patients (1.1%) had slight unilateral abdominal wall bulges. All patients were able to resume their normal daily activities. One hundred sixty-one patients (95%) had a mean of 1.3 secondary procedures for breast-mound shaping, scar revision, and/or nipple reconstruction, including 36 patients (21%) who had mastopexy or reduction surgery performed on the contralateral breast for symmetry.

DIEP flap reconstruction offers the advantage of reducing abdominal-wall morbidity, allowing short hospital stays and a return to normal daily activity. These data indicate that while performing DIEP flaps may be more technically demanding, this procedure can be safely carried out in a community hospital setting, with complication rates no higher than those reported for free TRAM flaps.

Breast Reconstruction with the DIEP Flap and Global Venous Congestion. Nho V. Tran, Edward Buchel, and Phyllis A. Convery. Mayo Clinic, Rochester, MN.

The purpose of this report was to study the possible etiology of deep inferior epigastric perforator (DIEP) flap global venous congestion in breast reconstruction. Although venous congestion in zones I, II, and III of the DIEP flap occurs rarely, its cause is unknown. After no usual causes can be identified, a venous anastomosis between the superficial epigastric vein and any vein at the recipient site is the current treatment.

Ten fresh cadaver dissections showed that the superficial (SIEV) and deep inferior epigastric vein (DIEV) systems were well-connected, but that the valves within the SIEV could prevent retrograde drainage into the DIEV. Normally, the SIEV valves directed blood from the tissue superficial to Scarpa's fascia into the femoral vein. In the DIEP flap, the usual SIEV flow was interrupted. Hence, a retrograde flow within the SIEV had to occur to drain into the DIEV system. Larger SIEVs might have competent valves to prevent such necessary retrograde flow.

The authors proceeded to carry out 21 DIEP flaps for breast reconstruction in 12 consecutive patients. The average patient age was 34 years, and the reconstructed breast mound ranged from a B to a D cup. In each flap, the largest SIEV was exposed, measured, cannulated, and continuously monitored by a Hewlett Packard transducer during the entire flap dissection. Pressure within the SIEV in mmHg was recorded and correlated with SIEV diameter, flap capillary refill, and flap outcome. The number of perforators per flap ranged from 1 to 3, with a mean of 2.

Twenty-one flaps had a mean SIEV diameter of 1.7 mm, with a range of 0.08 to 2.5 mm. Mean initial, peak, and end pressure in the SIEVs was 19, 33, and 15 mmHg, respectively. Two blue flaps were found to have a capillary refill under 1 sec, elevated peak pressure of 46 mmHg in the 1.5-mm SIEV and 50 in the other 2-mm SIEV. With the dissection completed and the flap still perfused by the native deep inferior epigastric vessels, the venous congestion spontaneously resolved, as the SIEV pressure dropped to 15 and 21 mmHg. Microvascular anastomoses were performed as usual. No flap loss occurred due to no additional venous anastomosis.

The conclusion was that competent SIEV valves may play a role in the rare global venous congestion of DIEP flaps.

Bilateral Autogenous Breast Reconstruction Using Perforator Free Flaps. Moustapha Hamdi, Phillip Blondeel, Koenraad VanLanduyt, Thierry Tondu, and Stan Monstrey. Gent University, Gent, Belgium.

The authors presented a single center's experience in bilateral breast reconstruction using perforator free flaps. A series of 42 patients underwent this procedure between February, 1996 and September, 2001. The average patient age was 49 years (range: 26 to 71 years). Seventeen patients (41%) had one or more risk factors (smoking, radiotherapy, chemotherapy, etc.) A family history of breast cancer was reported in 21 patients (50%).

Primary and secondary bilateral breast reconstructions were done in 16 and 3 patients, respectively. Fifteen patients, who had had breast reconstruction with implant, had a tertiary breast reconstruction. Combined reconstruction (primary with secondary, primary with tertiary) was done in 8 patients. Seventy-six DIEP flaps and 8 superior gluteal artery perforator (S-GAP) flaps were used. Simultaneous bilateral breast reconstruction was performed in all patients with DIEP flaps (38 patients) and in one patients with an S-GAP flap. In three patients, a different breast reconstruction was done, using an S-GAP flap with a 6-month interval. The average operating time was 10.5 hr (range: 8 to 14.5 hr) for the simultaneous bilateral reconstruction.

Total necrosis occurred in one case of DIEP flap, and the reconstruction was salvaged with an S-GAP free flap 6 months later. Two pulmonary infections, one DVT, and one cardiac arrhythmia were reported as postoperative complications. The mean hospital stay was 9 days (range: 6 to 20 days). Abdominal bulging was reported in one patient. There were no recurrent disease or cancer manifestations, with an average follow-up of 3 years.

Decreasing donor-site morbidity and offering an excellent aesthetic and long-term outcome, the described procedure provides a high rate of patient satisfaction and is an adequate oncologic alternative.

Breast Reconstruction with the DIEP Flap after Segmental Resection for Breast Cancer. Anders G. Liss, Rafael Acosta, and Morten Kildal. University of Uppsala, Uppsala, Sweden.

Segmental resection of the breast in cancer cases has grown popular among general surgeons, the reasons being that the cosmetic results are arguably better than with mastectomy and, in selected cases, the prognosis is the same. However, patients are left not only with asymmetries in breast size but also, more difficult to correct, is the dislocation of the nipple/areola area in the direction of the resection scar.

The DIEP flap has become the standard method for reconstruction after mastectomy at the authors' clinic. Nine cases were presented, in which the DIEP flap was used to reconstruct breast size and the dislocation of the nipple/areola area.

The flap was proven safe and very useful in reconstruction of small defects, as in segmental resections of the breast in cancer cases. The cosmetic result is very good and, since there is an areola/nipple, there is usually no need for a second operative procedure.

Comparison of the Deep Inferior Epigastric Perforator (DIEP) and Free Rectus Abdominis Myocutaneous (TRAM) Flaps in Post-Mastectomy Reconstruction: A Cost-Effectiveness Analysis. Achilleas Thoma, Dana Khuthaila, Karen Veltri, and Eric Duku. McMaster University, Hamilton, Ontario, Canada.

The DIEP flap is becoming a popular technique for post-mastectomy breast reconstruction. It is presented as a superior technique to the free TRAM flap in reducing abdominal donor-site morbidity. The efficacy of both techniques has recently been documented. However, a literature review has failed to identify any article that deals specifically with the cost effectiveness of the DIEP flap, using proper health economic methodology. The purpose of this study was to compare the DIEP flap to the free TRAM flap, using cost-utility analysis, a variant of cost-effectiveness analysis.

A decision analytic model and a Ministry of Health perspective were used. Direct medical costs associated with the two techniques were estimated from the Ontario Ministry of Health schedule of benefits (April, 2002). Hospital costs were obtained from St. Joseph's Hospital, a university hospital in Hamilton, Ontario. Utilities were obtained from 33 plastic surgeons across Canada, who were knowledgeable about the techniques and associated complications. Utilities were converted in Quality Adjusted Life Years (QALYs). The probabilities of the various health states associated with DIEP and free TRAM flaps (e.g., total flap failure, partial flap failure, abdominal hernia, abdominal bulging, etc.) were obtained by reviewing the recent literature. An Incremental Cost-Utility Ratio (ICUR) was calculated on the assumption that the probabilities of health states and cost estimates reflected the two techniques. A sensitivity analysis was also performed by increasing the probabilities of key complications associated with the DIEP flap, to see if the results were stable.

The DIEP flap was more costly than the free TRAM flap, but it provided more QALYs. The ICUR was $15,523.76/QALY. The sensitivity analysis showed that the ICUR did not change substantially by increasing the probability of some of the key (plausible) complications of the DIEP. Using the grades of recommendation for adoption of new medical interventions (Laupacis et al.,1992), this study showed that there is strong evidence of adoption of the DIEP flap.

The study showed that the DIEP flap is cost effective in the Canadian health care system. However, due to the uncertainty and the biases surrounding the reported rate of complications, and the true estimates of costs associated with the two techniques, there is a need to collect sampled data (costs and utilities) directly from patients in a randomized controlled trial comparing the two techniques.

Immediate Free Flap Reconstruction of Locally Advanced Breast Cancer: Is It Safe? Babak J. Mehrara, Andrew Smith, Timothy Santoro, Eric Arcilla, Jeffery Sebastian, James Watson, and Andrew L. Da Lio. UCLA, Los Angeles, CA.

Numerous studies have demonstrated that immediate breast reconstruction (IBR) after mastectomy is associated with improvements in quality of life and body image. However, IBR of locally advanced breast cancer (LABC) remains controversial due to concerns about delays in postoperative adjuvant therapy, as well as a theoretical potential for increased incidence of local recurrence or delays in diagnosis of recurrence. The purpose of this study was to evaluate the safety of IBR in patients with LABC in the largest series to date.

The records of 766 consecutive patients treated over the last 10-year period were reviewed. Patients with the diagnosis of stage IIB or worse breast cancer, who were treated with mastectomy followed by IBR, were identified and analyzed. One hundred and seventy patients with LABC who underwent IBR were identified. There were 158 unilateral and 12 bilateral reconstructions (176 flaps). The TRAM flap was used in the vast majority (158). The average patient age was 48 and 24% of patients had comorbid conditions. The average hospital stay was 5.1 days, and 23 patients (13.5%) required banked blood transfusions.

There were 7 major complications (4.1%; 1 complete flap loss; 2 partial flap losses, 1 arterial and 1 venous thrombosis, both of which were salvaged, and 2 hematomas). There were 70 minor complications (41%; 9 infections, 33 patients with fat necrosis; 9 hernias or abdomifluid overload, and 13 donor-site complications). Adjuvant postoperative therapy was delayed in 8 patients (5%), and the maximum delay was 3 weeks in one patient. There were no delays in the diagnosis of local recurrences that were attributable to the reconstruction. Although flap shrinkage was noted in nearly one-third of patients treated with postoperative radiation therapy, only a handful of patients experienced severe breast distortion.

The authors showed that IBR in the setting of locally advanced breast cancer is safe, well-tolerated, and not associated with significant delays in adjuvant therapy. In addition, in a subset of patients, IBR with free tissue transfer enabled wide en-bloc resection and avoided skin grafting of the chest wall and irradiated tissues. These findings, together with the convenience, improved cosmetic outcome, and the known psychological and economic benefits of IBR, make a strong argument for immediate reconstruction in otherwise healthy patients, regardless of cancer stage. Patients who are likely to receive postoperative radiation therapy must be counseled regarding the risks of flap shrinkage and distortion.

To Feel or Not to Feel: Innervated Autologous Flaps in Post-Oncologic Breast Reconstruction. Jeffrey Scott Isenberg. University of Oklahoma Health Sciences Center, Oklahoma City, OK.

With advances in autologous breast reconstruction, remarkable aesthetic form can be provided following mastectomy. Parallel to these improvements in form attendant on the use of autologous tissue has been improvement in the texture of the reconstructed breast. However, little attention has been given to the sensory deficit present in all current modes of breast reconstruction. This is surprising, given the significant attention sensory restoration receives in other anatomic regions of the body. Due to the easy accessibility of appropriate sensory nerves in the operative field, it seemed natural to attempt to provide sensory innervation to autologous flaps that were utilized in post-oncologous breast reconstruction.

During a recent 38-month period, all women presenting for post-oncologic breast reconstruction, when indicated, were presented the option of autologous innervated breast reconstruction. Detailed discussion of the risks was carried out, and a separate consent form utilized for the neural part of the procedure. All microneurorrhaphies were performed under operative magnification. Beginning at the 8^th postoperative week and continuing bimonthly for 16 months, each patient was evaluated for several sensory modalities, including touch, two-point discrimination, sharp/dull sensation, and temperature. Data were collected from the native breast skin envelope and the skin island of the flap, with the contralateral unoperated breast serving as a control.

Thirteen women underwent innervated autologous breast reconstruction following mastectomy. The mean age of this group was 60.7 years (range: 51 to 70 years). There were 10 innervated TRAM flaps and three innervated latissimus flaps. All flaps survived. Surgical complications occurred in four patients and included a seroma, abdominal bulging, delayed healing of an abdominal incision, and delayed healing of infraumbilical abdominal skin. Sensory testing demonstrated early and sustained return of all modalities throughout the period of data collection. Sensory return approached, but never completely attained the levels of the contralateral unoperated breast. There was also a trend for the innervated TRAM flap to surpass innervated latissimus flaps in degree of sensory return.

Inntervated autologous breast reconstruction following oncologic surgery appears safe and efficacious in providing sensory return superior to the results reported with non-innervated autologous reconstructions.

A Reliable Patient-Based Evaluation Instrument for Assessing Autologous Breast Reconstruction Outcomes. Michael Cohen, Bradley Evanoff, and Keith E. Brandt. Washington University, St. Louis, MO.

The objective of this study was to create a reliable instrument that can be used to determine what aspects of autologous breast reconstruction outcomes matter most to patients.

A 9-item instrument was developed, with emphasis on those criteria that are specifically affected by the particular procedure performed. The criteria included vertical positioning, horizontal positioning, defects, projection, shape, inframammary fold, medial contour, overall appearance, and overall satisfaction. The instrument utilized a 4- to 7-point scale on each sub-item. Thirty-six patients who had previously undergone autologous breast reconstruction completed the instrument on two occasions, one month apart.

Test-retest reliability analysis demonstrated a “good-to-excellent” range for each criterion (weighted kappa of 0.53 - 0.87) and a lower reproducibility for the overall satisfaction item (weighted kappa of 0.46). The instrument's internal consistency was high when assessed both for the entire nine-item scale (Chronbach's alpha= 0.92). Most individual criteria correlated well with the overall appearance score (Spearman's rho of 0.61 - 0.80), with slightly lower correlations for the vertical and horizontal positioning and the inframammary fold items (rho of 0.55 - 0.58). Corrlations between individual criteria and patient satisfaction were weaker for horizontal positioning, defects, projection, and inframammary fold (rho of 0.47 - 0.59), and good for vertical positioning, shape, and medial contour (rho of 0.64 - 0.80). High correlation was found between overall appearance and overall satisfaction (rho=0.81).

These authors developed a reproducible and internally consistent scale that can be used by patients to assess the subtle and specific aspects of their own reconstruction appearance. The survey instrument shows good-to-excellent test-retest reliability, as well as correlation between the individual criteria and the patient's overall satisfaction and judgment of overall appearance. Application of this instrument to a large cohort of patients will allow prioritization of the criteria. By learning which criteria are most important in determining overall patient satisfaction, surgeons will then be able to choose the reconstructive procedures most likely to positively impact these criteria.

Breast Reconstruction with the Free Superficial Inferior Epigastric Artery Flap. Pierre Chevray. University of Texas M.D. Anderson Cancer Center, Houston, TX.

Breast reconstruction using the TRAM flap technique has evolved from the pedicled to the free TRAM, the muscle-sparing free TRAM, and then the DIEP flap. Each variation in this progression has involved harvest of decreasing amounts of rectus abdominis muscle and anterior fascia, with the aim of decreasing abdominal donor-site weakness, discomfort, complications, and hospital stay. Breast reconstruction using the lower abdominal free superficial inferior epigastric artery (SIEA) flap has the potential to virtually eliminate abdominal donor-site morbidity and complications, because the rectus abdominis fascia and muscle are neither excised nor incised. Despite its advantages, the free SIEA flap for breast reconstruction is rarely used.

The author reported his experience with 9 free SIEA flap breast reconstructions in 7 patients. The free SIEA flap was attempted in 20 consecutive free autologous tissue breast reconstructions between August, 2001 and April, 2002. A free SIEA flap was not used in 11 of these 20 cases (55%) because the superficial inferior epigastric artery was absent or deemed too small. An adequate vein was present in all 20 cases. The average patient age was 46 years, and the average BMI was 30.

One of the nine SIEA flaps was lost due to arterial thrombosis. The remaining 8 SIEA flaps healed without clinically significant fat necrosis. The superficial inferior epigastric vessel pedicle length averaged approximately 8 cm. The internal mammary vessels were used as recipients in all SIEA flap cases to allow positioning the flap sufficiently medially on the chest wall. Moderate and large sized breasts were reconstructed with hemi lower abdominal flaps, with aesthetic results equal to those obtained with TRAM and DIEP flaps. The average hospital stay was shorter for the SIEA flap patients, but the difference was not statistically significant in this initial small series.

The free SIEA flap is an attractive option for autologous tissue breast reconstruction. Harvest of this flap does not injure the anterior rectus fascia or underlying rectus abdominis muscle. This will potentially eliminate abdominal donor-site complications, such as bulge and hernia formation, and decreases weakness, discomfort, and hospital stay, compared to TRAM and DIEP flaps. Its disadvantages are the smaller and shorter pedicle, compared to TRAM and DIEP flaps, and the absence or inadequacy of an arterial pedicle in over half of the patients.

Autogenous Breast Reconstruction with the Superficial Inferior Epigastric Artery Flap. James E. Craigie, Robert J. Allen, and Andreas S. Heitland. Louisiana State University Health Science Center, Mt. Pleasant, SC and New Orleans, LA, and University of Aachen, Aachen, Germany.

The superficial inferior epigastric artery flap (SIEA) has been reliably used in microsurgical reconstruction of the breast. Since 1997, the authors have employed the flap in 133 reconstructions with only one failure. The advantages of the SIEA influde: reliable donor vessels, aesthetically placed donor scar, readily available donor tissue, and no risk for abdominal hernia or abdominal-wall weakness. The authors performed cadaver dissections in 100 specimens and found the SIEA vessels present in 72% of groins. The technique and surgical results of the SIEA flap for breast reconstruction were discussed.

Internal Thoracic Artery Perforators as Recipient Vessels for Perforator Flaps in Microsurgical Breast Reconstruction. Alexandre Mendonca Munhoz, Luis Henrique Ishida, Gustavo Gibbin Duarte, Gustavo Pinos Sturtz, Fabio Lopes Saito, and Marcus Castro Ferreira. University of Sao Paulo, Sao Paulo, Brazil.

The microanastomosis of a perforator flap pedicle to a perforator recipient vessel in the anterior thoracic wall, in order to avoid internal thoracic or thoracodorsal vessel dissection, is feasible in clinical practice. The authors presented their preliminary clinical experience with this technique.

Since June, 2001, a search for recipient perforator pedicles in the anterior thoracic wall was carried out in 30 patients who underwent microsurgical breast reconstruction. These vessels are perforators from the internal thoracic artery and vein, and vessel diameter and flow were evaluated, with recipient pedicles chosen based on the authors' experience. In 22 patients, immediate breast reconstruction was performed, and late reconstruction was done in 8 patients. The average patient age was 52 years (±3). In the late reconstruction group, all patients had radiotherapy after mastectomy, with a mean interval of 20 months between surgical procedures.

In the group which underwent late breast reconstruction, no perforator vessels clinically adequate as flap recipients were found. In the immediate breast reconstruction group, perforator vessels in the second intercostal space were found in 13 patients. In 8 of these patients, the vessels had a diameter and flow adequate for use as recipient pedicles. Seven DIEP flaps and one SGAP flap were transferred to these perforators in the anterior thoracic wall. The flaps demonstrated no perfusion complications. The 22 patients without recipient perforator pedicles underwent breast reconstruction with DIEP flaps transferred to the internal thoracic vessels.

Perforator-to-perforator breast reconstruction is clinically feasible in selected patients who undergo immediate breast reconstruction. This technique minimizes morbidity in internal mammary vessel dissection and spares the thoracodorsal vessels. The inconstant presence, variable diameter, and extra time spent in the delicate dissection of these perforator pedicles are the main disadvantages.

Internal Mammary Anterior Thoracic Perforators: New Option for Recipient Vessels in Free Flap Breast Reconstruction. Abhay Gupta and Joseph C. Camarata. University of Texas Health Science Center at San Antonio, TX.

The advent of microsurgery has revolutionized the art of autologous tissue breast reconstruction. Microsurgical free tissue transfer allows the reconstructive surgeon to restore a naturally shaped breast without the constraints of pedicled flaps. Although the free transverse rectus abdominis myocutaneous (TRAM) flap has become the procedure of choice in microsurgical breast reconstruction, other popular donor tissue options include the deep inferior epigastric perforator (DIEP) flap, the deep circumflex iliac (Rubens) flap, and the superior or inferior gluteal myocutaneous flaps. In general, free tissue transfer has several advantages over pedicled flap reconstruction, including better flap vascularity, easier flap shaping and inset, and decreased donor-site morbidity. The most popular recipient vessels used are the thoracodorsal vessels and internal mammary vessels. Other reported recipient vessels include the serratus branch of the thoracodorsal, the subscapular, and the thoracoacromial vessels. The thoracodorsal vessels are often used in immediate breast reconstruction because they are exposed during the mastectomy procedure, particularly following axillary lymph node dissection. However, in delayed reconstruction, many surgeons prefer to use the internal mammary vessels, to avoid repeat dissection through scar or fibrosis in the axilla. Many surgeons also prefer to use these vessels in immediate reconstruction.

Advantages of the internal mammary vessels include more optimal flap placement, with avoidance of lateral fullness and medial deficiency, and a shorter required pedicle length. However, access to the internal mammary vessels requires increased operating time, and excision of a medial segment of costal cartilage, which carries the potential complications of pneumothorax, chest-wall hernia or intercostal neuralgia.

The authors reported the use of a new, previously unreported, recipient vessel that has all the advantages of the internal mammary vessels, but avoids the necessity of costal cartilage resection and its associated morbidity. The anterior thoracic perforators of the internal mammary vessels are direct cutaneous perforators that emerge from the intercostal spaces, just lateral to the sternum, and form the vascular basis for the free or pedicled deltopectoral flap. These vessels are quickly and easily exposed, and are of adequate diameter for microsurgical anastomosis. The superficial location of these vessels avoids extensive dissection or rib removal, and their medial location allows for optimal positioning of the flap.

The authors presented a series of free flap breast reconstructions using these recipient vessels, and proposed that the anterior thoracic perforator is an excellent new recipient vessel option in free flap breast reconstruction, enabling a more efficient reconstruction with significantly decreased morbidity.

Choice of Recipient Vessels for Free Flap Breast Reconstruction. Moustapha Hamdi, Phillip Blondeel, Koenraad VanLanduyt, and Stan Monstrey. Gent University, Gent, Belgium.

The internal mammary (IM) vessels are the authors' first choice for free flap breast reconstruction, although their use may result in pain and thoracic deformity. The authors' efforts at reducing morbidity led to their adopting an algorithm in free flap breast reconstruction, and also to salvage compromised free flaps.

The IM perforators are sought out during the undermining of breast skin. If no suitable perforator is found, the pectoralis major (PM) muscle is split at the level of the 4^th rib. The PM muscle is dissected cranially to look for a good sized perforator. The internal mammary vessels will be considered for the microanastomosis if the perforator is small. The IM perforator can also be used as an additional recipient vein in case of a congested free flap or venous thrombosis. The thoracodorsal (TD) vessels are used as an additional blood supply or drainage in case a large flap is indicated. The thoracodorsal vessels are spared if a latissimus dorsi flap is needed as a secondary procedure. The thoracoacromial vessels are considered as a salvage technique for a flap with late venous thrombosis.

This protocol was used in patients between June, 1999 and December, 2001. Breast reconstruction with free flaps was done in 218 patients with perforator flaps. IM, TD, and thoracoacromial vessels were used in 180 patients (82%), in 10 patients (4.5%), and in one case (0.5%), respectively. IM perforators were successfully used in 27 cases (12.5%), but a revision of an arterial thrombosis occurred in one case, resulting in partial flap necrosis. Operative time was decreased by 60 min by limiting the dissection of the recipient vessels and of the main pedicle of the flap.

Using IM perforators as recipient vessels spares the IM vessels for eventual cardiac bypass surgery, avoiding recipient-site morbidity, and decreasing postoperative discomfort.

Recipient Vessels in Microsurgical Breast Reconstruction: A Qualitative and Histomorphometric Study. Luis Henrique Ishida, M.M. Gomes, H.A. Nakamoto, F.L. Siato, L. Rodrigues, G. Sturtz, A.D.Y. Kim, A. Munhoz, and M.C. Ferreira. University of Sao Paulo, Brazil.

The choice between the thoracodorsal and internal thoracic as recipient vessels for breast reconstruction has been discussed by several authors. Anatomical studies have revealed little difference between these pedicles; however, the presence of degenerative changes may lead to flap failure and might be a significant difference between them.

Fifty-four vessels of bilateral thoracodorsal, internal thoracic, and deep inferior epigastric pedicles were dissected in nine fresh cadavers. Samples of perforator vessels 1.5 cm long were harvested and fixed with formalin. Five-millimeter sections of paraffin-embedded tissue were stained with hematooxilin-eosin, Verhoeff, and Picrosirius. Histologic degenerative changes were evaluated and histomorphometric measurements were made. Data were analyzed with a computer-assisted device with Optimas^R software.

Major arterial degenerative changes were found in seven thoracodorsal (38.9%), three internal thoracic (16.6%), and two deep inferior epigastric (11.1%) vessels. The mean external diameters were 1.25, 1.82, and 1.67 mm, respectively. There was one case of chronic arterial thrombosis with signs of reperfusion. Although a more similar diameter between the deep inferior epigastric and internal thoracic arteries was found, in clinical practice the technical difficulties of vascular anastomosis with the thoracodorsal pedicle are also similar. The higher number of degenerative changes observed in the thoracodorsal vessels may be related to a higher number of thromboses and flap failures. The authors stated that careful patient selection is required to avoid complications. Comparative clinical studies must be done to elucidate the real importance of such findings.

Does Recipient Vessel Choice for Free TRAM Flaps in Irradiated Patients Affect Outcome? Claire L.F. Temple, Eric A. Strom, Adel A. Youssef,, and Howard N. Langstein. University of Texas M.D. Anderson Cancer Center, Houston, TX.

More breast cancer patients are undergoing post-mastectomy radiation based on reports of improved locoregional control and survival. Although immediate TRAM breast reconstruction yields favorable aesthetic results, delaying the reconstruction until completion of radiotherapy spares the flaps from unpredictable fibrosis. It remains unclear whether the thoracodorsal (TD) or the internal mammary (IM) vessels are the preferred recipient vessels, considering the effects of radiation and scarring from previous surgery.

The authors' goal was to compare the TD and IM vessels as recipient sites for usability rates and outcomes in a uniform group of patients undergoing TRAM flaps in a delayed fashion after post-mastectomy radiotherapy.

Between 1990 and 2001, 159 consecutive delayed, irradiated TRAM flap patients were identified from a prospective database. Charts were reviewed for recipient vessel choice, inspection and decline of any recipient vessel, and outcomes including flap failure, vascular complications, fat necrosis, and lymphedema. T-tests, chi-squared and multivariate analyses were performed to compare vessel choice and outcome.

During this period, 123 of 159 patients were planned to have a free flap, of which 106 free flaps were able to be performed. The remaining 17 were converted to a pedicle TRAM, due to unusable recipient vessels. Of the free flaps, 45 were anastomossed to the IM, 55 to the TD, and 6 to other vessels. The IM and TD group were well-matched for BMI, smoking history, number of scars in the flap, time between XRT and reconstruction, and ischemia time. Twenty-six percent of the TD and 20% of the IM vessels were unusable as recipients (p=0.42). Excluding the IM vessels deemed unusable by preoperative duplex Doppler examination, the number of IM vessels declined on intraoperative dissection and inspection, and decreased to 12% (p=0.056). Of the unusable TDs, 84% were too scarred, 11% were too small, and 5% were absent. Of the unusable IMs, 46% had unusable veins, 27% had unusable arteries, and 27% had both. The unusability rate for IM vessels was equivalent for left- and right-sided reconstructions (23% vs. 17%, p=0.55).

Outcomes were similar regardless of recipient vessel used (TD vs. IM): flap loss (3.6% vs. 0%, p=0.20), partial flap failure (5.5% vs. 8.9%, p=0.54), vascular complications (11% vs. 7%, p=0.46), fat necrosis (15% vs. 18%, p=0.69), lymphedema (9% vs. 4%, p=0.37). Logistic regression revealed a trend toward higher overall complication rates with the use of TD vessels (OR 2.8, p=0.19).

The authors' conclusion was that TD and IM recipient sites allowed for similar usability and outcome.

Small Intestinal Submucosa as a Carrier for Skin Flap Prefabrication. Can Zhu, Feng Zhang, Man-Ping Lei, Tanya M. Oswald, and William C. Lineaweaver. University of Mississippi Medical Center, Jackson, MS.

The vascular carrier is believed to a play a crucial role in revascularization of the prefabricated skin flap. In the clinical setting, vascular pedicles raised in continuity with a muscle cuff or fascia may result in a bulky pedicle and donor-site morbidity. Small intestinal submucosa (SIS) is a biomaterial that can provide a natural scaffold for cell migration and can induce neovascularization. In this study, the authors examined the feasibility of SIS as a carrier in skin flap prefabrication.

Thirty-eight adult Sprague-Dawley rats were randomly divided into five groups. The right saphenous vascular bundle was used as the vascular carrier. In Group 1, an arteriovenous fistula was made by anastomosis of saphenous artery and vein. An SIS patch (1.5 x 2 cm²) was placed underneath the vascular bundle. At 2 weeks postoperatively, the same size of medial thigh flap, including the SIS, was raised and replaced. In Group 2, the vascular bundle was isolated and laid over the SIS patch. The distal saphenous vessels were ligated when the flap was raised. In Group 3, an arteriovenous fistula was made without SIS implant. After 2 weeks maturation, the flap was raised with only the vascular bundle. In Group 4, the flap was raised with only the vascular bundle, and the distal end of the vascular bundle was ligated. In Group 5, an SIS was implanted, and the flap, including the SIS patch, was raied and replaced without vascular pedicles. The survival of the flaps and the histology were evaluated at 7 days after replacement.

The average survival area in Group 1 was 98.8%, and in Group 2 was 87.5%. there was no statistically significant difference in survival area between these two groups (p>0.05). The mean survival areas in Groups 3 and 4 were 60.0% and 25%, respectively, which were significantly lower than in Groups 1 and 2 (p<0.05, p<0.01). Histologic evaluation showed that the SIS patch was incorporated into the adjacent connective tissue, and a consistent pattern of exaggerated amounts of neovascularization was observed in the biomaterial and subcutaneous tissue of the prefabricated flaps with SIS implant.

The study demonstrated that small intestinal submucosa can incorporate into the adjacent tissue and can induce angiogenesis in flap prefabrication. This biomaterial can provide a scaffold for supporting and enhancing the survival of prefabricated skin flaps.

Transciatic Foramen Pedicled Vertical Rectus Myocutaneous Flap Coverage of Gluteal Defects. Nho V. Tran. Mayo Clinic, Rochester, MN.

Extensive low lumbosacral irradiated defects often require free tissue transfer to achieve closure. The well-known transpelvic path allows an inferiorly based vertical rectus abdominis myocutaneous flap (VRAM) closure of pelvic and sacrectomy defects. In three presented cases, the large defects were off the midline. Division of the sacral tuberous ligament enlarged the greater sciatic foramen to create a direct route for the ipsilateral, inferiorly based pedicled VRAM flap to close unilateral gluteal defects up to 300 cm². This maneuver allows wound closure without free tissue transfer.

Development of a Novel Spinal Cord Nerve Guide with an Injectable Trophic Factor Delivery System. Curtis L. Cetrulo, Jr., Kai Kai Wang, L. Nicholas Cetrulo, and Brooke R. Seckel. Plastic and Reconstructive Surgery Research Laboratory and Microsurgery Center, Burlington, MA.

The purpose of this research was to design an in vivo model for the study of the microenvironment of the transected spinal cord.

Eight spinal cord nerve guides were placed in a 5-mm gap in the cords of Sprague-Dawley rats. Gaps were created after microlaminectomy and cord exposure at the T8 level. The dura was then sutured closed around the guide's tubing for inflow and outflow ports. The injection ports were tunneled subcutaneously and sutured in position for postoperative injection. Rats were subjected to gentle injection of saline beginning 1 week postoperatively to assess guide and port patency. Five sham animals underwent microlaminectomy, cord exposure, removal of a 5-mm gap of spinal cord, and dural closure without guide placement.

The first three animals survived a mean of 68 days (range: 57-81 days), until sacrifice due to urinary tract infection. Postoperative care was amended to include antibiotic coverage, increased infection control, and a modified bladder-emptying technique. Of the next 5 animals, 4 survived until planned sacrifice at 90 days; 1 animal exhibited autophagia and was sacrificed at 75 days. All 8 guides and ports remained patent and all rats tolerated injections of saline without exhibiting spinal shock. The 5 sham animals survived for a mean of 11 days (range: 7-14 days) until sacrifice after exhibiting signs of spinal shock, exemplified by poor feeding, respiratory distress, and obtundation.

Post-mortem dissection revealed a regenerated cable that spanned the 5-mm gap in all 8 animals in which a nerve guide was placed. The diameter of this cable was significantly smaller than the native cord, yet histologic sections contained a significant number of axons. The mean axon count in the initial group of experimental animals (n=3) was 4256±154 (range: 1807- 6324). A marked fibrotic reaction was observed exterior to the guide's lumen, while the regenerated cord was sequestered within the lumen. No regenerated cables were seen in the sham animals.

The authors have designed a spinal cord nerve guide that spans a 5-mm gap in the spinal cord of the Sprague-Dawley rat. The nerve guide employs inflow and outflow ports, which should allow trophic factors to be injected into the lumen of the guide, while sequestering the proximal and distal stumps from the mechanical and possibly biochemical impediment of neurite outgrowth by the reactive gliosis and fibrotic components of the lesion scar. Studies are also underway to assess the functional capacity of these regenerated cables.

Dynamic Restoration of the Eyelid Sphincter with a Temporalis Sling as Part of the Reanimation in Facial Paralysis. Ashok Gupta. Bombay Hospital and Medical Research Centre, Bombay, India.

Complete reanimation of a paralyzed face is a reconstructive challenge, as well as a dream for patients so affected. Functioning muscle transfer is a dynamic method of facial reanimation, in which spontaneous recovery of muscle function either has not taken place or is not possible, as in cases of long-standing facial paralysis or following ablative surgery. Cross-face nerve grafts (CFNG) provide requisite motor control to these functioning muscle transfers, as well as controlling synchronized facial movements on both sides. End-to-end coaptation of a severed/damaged facial nerve or interposition nerve graft offer better results than all secondary reconstruction.

Despite all efforts for acceptable facial movements, restoration of the eyelid sphincter with near normal dynamic movements of the upper and lower eyelids has remained a challenge. This author uses a temporalis mini innervated muscle-fascia sling for eyelid sphincter control, along with cross-face nerve grafts, and a free functioning muscle transfer (FFMT) for elevation of the upper lip, angle of the mouth, and lower lip. The anterior belly of the digastric muscle is used to stabilize the lower lip and to replace the action of lower-lip depressors for a dynamic reanimation. The temporalis mini innervated muscle-fascia sling was initially described by Gillies and has been re-defined by the author with excellent results. There is no webbing of the medial canthus or eversion of the lower or upper eyelids.

The author described his experiences in using the temporalis mini innervated muscle-fascia sling transfer for restoration of the eyelid sphincter as part of total facial reanimation, with a 7- to 10-year follow-up of patients.

Free Anterolateral Thigh Flap for Craniofacial Reconstruction. Ayman Amin, Mohammed Rifaat, Francisco Civantos, Donald Weed, Mohammed Abu-Sedira, and Mahmoud Bassiouny. The National Cancer Institute, Cairo, Egypt.

Free tissue transfer has revolutionized skull-base surgery by expanding the ability to perform cranial-base resection, thus improving the quality of reconstruction. The rectus abdominis and latissimus dorsi are commonly used flaps for such reconstruction. In this report, the authors presented their experience using the anterolateral thigh flap for reconstruction of a variety of craniofacial defects.

The study involved a total of 15 patients over a 4-year period between 1998 and 2002. Fourteen patients had locally advanced head and neck cancer that required craniofacial resection, and one patient had a complicated gunshot wound of the forehead. The patients underwent extensive resection and debridement that resulted in large composite defects. Ten patients were treated at the National Cancer Institute (Cairo, Egypt) and five at the University of Miami, Florida. All patients were immediately reconstructed using the anterolateral thigh free flap. A two-team approach was applied for all patients. The success rate, recipient vessels used, donor-site morbidity, and complications were examined.

The anterolateral thigh flap was used to reconstruct a variety of extensive craniofacial defects, consisting of composite scalp and calvarium (3), anterior skull base (4), lateral skull base (3), and composite midface defects (5). The free flap was successful in 14 patients. One patient developed complete flap necrosis. Five patients received preoperative radiation therapy. One patient died 1 month postoperatively from pulmonary embolism. The most common recipient vessels were the facial vessels and the external jugular vein. The average length of the vascular pedicle was 12 cm. Skin perforators were present in a musculocutaneous pattern in most of the cases (13/15). There was no CSF leak. Three patients developed partial wound dehiscence. The donor site was closed primarily in 10 patients. Donor-site morbidity included partial loss of the skin graft in 2 patients, and wound infection and superficial sloughing in 1 patient.

The anterolateral thigh free flap is a versatile flap for head and neck reconstruction. It is a reasonable option for skull-base reconstruction. It has the advantages of allowing a comfortable two-team approach, providing a generous amount of tissue without jeopardizing the integrity of the abdominal wall, and less pain and atalectasis postoperatively.

Reconstruction of Complex Craniofacial Resection. Nho V. Tran, Uldis Bite, Paul M. Petty, Phyllis A. Convery, and N.B. Meland. Mayo Clinic, Rochester, MN.

The purpose of this study was to evaluate the reconstructive outcome of patients with extensive craniofacial tumor resection.

A retrospective study between 1988 and 2001 identified 69 patients with an average age of 55 years, sex ratio of 45 males to 24 females, and T4 malignancies that required extended craniofacial resections. Tumor types of 49% SCC, 23% sarcoma, 10% adenocarcinoma, 6% BCC originated from the face, sinus, ear, and parotid areas. Forty-two percent (29/69) of patients underwent 2.4 previous surgical procedures prior to the studied resections. Tumor recurrence after resection and radiation occurred in 23% of the patients. All were reconstructed with free rectus myocutaneous flaps. Seventy-five percent of the patients required skull-base reconstruction, and 52% received postoperative adjuvant irradiation.

The mean follow-up time was 31.4 months, with a median of 13 months. Early complications included 7 hematomas, 5 flap dehiscences, 4 arterial thromboses, 2 complete flap losses, 1 partial loss, 2 CSF leaks requiring a lumbar drain insertion during the early postoperative period, and 2 wound infections. Late complications included two ventral hernias and a palatal fistula in a patient with preoperative radiation. The flap loss rate was 3% (2/69). The overall complication rate was 26/69 (40%). No surgical mortality occurred.

Complex craniofacial resection of advanced tumor can be safely reconstructed with the free rectus myocutaneous flap.

Latissimus Dorsi Free Flap for Facial Reconstruction in the Pediatric Trauma Patient: Case Report and Literature Review. Jennifer L. Walden, Linda G. Phillips, Steven J. Blackwell, and John H. Miller. University of Texas Medical Branch, Galveston, TX.

The difficulty of pediatric facial reconstruction is accentuated by the unique and changing character of the face during childhood development, and the availability of local matching tissues. By necessity, distant free flaps have become a first line in the treatment of large, complicated, facial wounds in adults. Although free tissue transfer has been advocated for acquired defects after excision of congenital lesions and for hemifacial microsomia in children, microsurgical reconstruction of extensive traumatic facial defects in the pediatric population is described less frequently.

The authors presented the case of a 5-year-old boy who suffered a traumatic avulsion injury to the right hemi-face, scalp, and skull from a riding lawn mower. The patient sustained severe head trauma, with initial examination revealing exposed dura and frontal lobe, as well as calvarial, scalp, orbital, maxillary, and facial soft-tissue deficit. Initial and immediate operative intervention included reconstructive coverage of the facial defect with a free latissimus dorsi myocutaneous flap with thoracodorsal anastomosis to the facial artery and vein. Pericranial flaps were used to cover the large scalp defects, and a scalping flap based on the left supratrochlear vessels augmented on its undersurface with Alloderm^R was placed over a dural patch to provide brain coverage. Additional delayed procedures were performed for refinement of aesthetic results, including upper lip and oral commissure reconstruction with local transposition flaps, scalp reconstruction after tissue expansion, alloplastic augmentation of orbital and zygomatic deficiencies, and laser hair removal of previously rotated scalp flaps along the temporal region.

Microsurgical correction of facial defects with free tissue transfer has undergone rapid developments over the past decade, and its application in pediatric patients has emerged as well. The authors presented a case of a young patient who underwent microvascular resurfacing of a large area of his face with a free latissimus dorsi flap, along with staged ancillary procedures to optimize facial symmetry, contour, color, and texture. Long-term follow-up, as well as photographic analysis, was provided for this case that the authors consider a success in pediatric facial reconstruction after a devastating injury.

Role of Reconstructive Microsurgery in War Wounds. Ahmad Maghari, Kamal S. Forootan, and Sima Molavi. Hazrat Fatemeh Hospital, Tehran, Iran.

These authors were interested in describing the role of reconstructive microsurgery in treating combatant wounds during the war between Iran and Iraq, which began in 1981 and lasted for 8 years. They described experiences with extensive soft-tissue losses; most of the patients were treated primarily in the field or in small zone hospitals, and then referred again to the hospital in Tehran. The authors have utilized more than 1500 free flaps, and have achieved satisfactory results, with a success rate of about 85%. Most of the patients are now active in society, with suitable jobs and family connections. The details of the flaps were discussed and slides presented.

Flying Solo: A Single Year-Single Surgeon Community Hospital Replantation/Revascularization Experience. Jeffrey Scott Isenberg. University of Oklahoma Health Sciences Center, Oklahoma City, OK.

Once considered heroic and experimental, replantation and revascularization of acral parts have become accepted and routine at many medical centers worldwide. However, the reported results regarding survival and the function of replanted and revascularized parts have been predominantly from tertiary medical centers. These reports represent the results of specialized teams, including microsurgeons, fellows, residents, and ancillary staff. Such results, while laudable, may not reflect the actual outcome and experience of the community hospital surgeon. The purpose of this study was to evaluate the replantation and revascularization experience of a single microsurgeon in solo practice in a community hospital environment.

Prospective data were gathered on all patients presenting to the author over a recent 12-month period at a 200-bed community hospital for emergent replantation or revascularization of the forearm, hand, and digits. Data gathered included age, sex, etiology, concurrent medical conditions, tobacco use history, ischemic time, method of part preparation and transport, zone of injury, dominant handedness, procedure performed, use of vein grafts, outcome, and secondary procedures arising from failure of the initial intervention. The system of Merle was used to classify distal digital amputations.

Fifty-one patients presented for emergent replantation and/or revascularization of the digits, hand, and forearm. Since the author was the sole provider of these services at the hospital of contact, his series represents the entire institution's experience for the time period given. Among this cohort 94% were men whose mean age was 31.6 years. Concerning etiology, there were three almost equal causes: knife (30%), crush-avulsion (34%), and power saw (33%). The zone of injury was confined in the majority of cases from zones I (Merle zone IV-I) through III. A total of 67 digits and/or limbs were operated on. There were 41 replantations and 26 revascularizations. Ischemic time averaged 170 min. Fifty-eight digits and limbs (87%) were successfully salvaged via replantation or revascularization. There were 9 failed interventions (13%), all single digits with multiple levels of injury. No revascularization failed and no replantation of a sharply amputated part failed.

Over a 12-month period, 67 digits and/or parts were operated on emergently for replantation/revascularization by a solo microsurgeon in a community hospital. A success rate of 87% was noted. All failed interventions were in crush-avulsion injuries. These results compare well with those reported by major tertiary centers.

Microsurgery in Private Practice: Review of a Single Surgeon's First 100 Cases. Frederick J. Duffy. UT Southwestern, Dallas, TX.

Microsurgery has evolved into an integral part of reconstructive surgery, particularly regarding reconstruction of the breast, head and neck, and extremities. High success rates have been documented by several large centers. Microsurgery is also practiced in the private practice setting, but success rates in private practice are poorly documented. A review of a single surgeon's first, consecutive 100 free flaps was presented.

Recipient sites included the lower extremity (57%), breast (15%), femoral head (9%), chest wall (1%), hand (5%), and head and neck (13%). Ten flaps were re-explored within the first 36 hr; nine of these flaps were salvaged (90%). Re-exploration revealed venous occlusion in 7 patients, arterial occlusion in 1, and hematoma under the flap in 2 patients. In addition to the one flap that failed despite re-exploration, two additional flaps failed. One was discovered to be obviously nonviable on postoperative day 3, and another patient returned after discharge with a nonviable flap. Of the three flaps that failed, two patients had subsequent successful reconstructions with a second free flap, and one patient underwent below-knee amputation. There were two patient deaths within the first month following surgery, one from stroke and one from an apparent myocardial infarction. The review also demonstrated an increasing interest in and use of perforator flaps: two of the three flaps lost were perforator flaps, but there was no statistically significant difference in complications between the perforator and traditional free flap groups.

The review suggested that high microsurgical success rates are possible in private practice and that perforator flaps can be successfully learned and performed in the private practice setting.

Breast Reconstruction with Free Tissue Transfer at a City Hospital: A 12-Year Experience. Hooman Soltanian, Alexes Hazen, Jennifer M. Capla, Raymond R. Chang, Jamie P. Levine, and Geoffrey C. Gurtner. NYU Medical Center, New York, NY.

Free tissue transfer has become a routine part of reconstructive surgery. To study the evolution of free flap procedures for breast reconstruction in a city hospital setting (Bellevue Hospital Center in New York City), the authors reviewed the records of all patients who underwent breast reconstruction with free tissue transfer from 1/1/90 to 12/31/01. In this hospital, residents provide the majority of patient care under direct attending supervision and participation.

The following parameters were examined: type of flap, length of stay, operative time, delayed vs. immediate reconstruction, need for blood transfusion, use of anticoagulants, comorbid conditions, and complication rate. Procedures performed during the last 2 years were compared to those of the first 10 years of the study. Forty-six patients underwent breast reconstruction with free flaps during the study period - 33 during the first 10 years and 13 during the last 2 years. There was a decrease in the length of hospital stay and number of blood transfusions over the recent years. There was an increase in the number of patients requesting immediate breast reconstruction. The anesthesia records documented a trend toward shorter operative times. The length of the procedures for immediate and delayed reconstructions was similar. Aspirin was used in nearly all cases. Free flaps have become the procedure of choice when autologous breast reconstruction is desired. TRAM flaps remain the most often used tissue for this purpose.

The authors concluded that microvascular free tissue transfer is a reliable and safe option for post-mastectomy reconstruction at a city hospital. The establishment of a comprehensive and multi-disciplinary breast cancer team is an important factor for better patient education and higher degree of satisfaction. Preoperative planning and coordination between the ablative and reconstructive teams have resulted in shorter operative times, improved aesthetic results utilizing skin-sparing mastectomy techniques, and better patient outcomes. Based on the findings of the study (shorter operative times and decreased need for blood transfusion), the authors do not consider routine preoperative autologous or donor-directed blood donation necessary or cost effective.

Update on Limb-Sparing Surgery for Limb Sarcomas. E. Gene Deune, Maurice Nahabedian, Anthony P. Tufaro, Deborah Frassica, Steve Lietman, and Frank J. Frassica. Johns Hopkins School of Medicine, Baltimore, MD.

There are 6000 new cases of sarcomas annually, with 50% located in the extremities. Wide tumor resection, immediate soft-tissue reconstruction, and adjuvant therapy with limb preservation have become the treatment of choice whenever possible.

Between February, 1998 and May, 2002 (51 months), 74 patients (M=52, F=22; mean age: 50.8± 17.1 years) underwent salvage of 75 limbs (54 lower extremity, 21 upper extremity) at the authors' medical institution by a multidisciplinary team of orthopedic oncologist, radiation oncologist, and reconstructive surgeons. Sixty-nine percent of the tumors were high-grade (n=51); 41% of the tumors were MFH (n=30), and the thigh was the most common location (n=30). Seventy-five percent of the coverage was done by the rectus abdominis (n=28), latissimus dorsi (n=19), and the gastrocnemius (n=13). Thirty-three (45%) patients had additional reconstructive surgeries, the most common being tendon transfers/grafting and allograft bone or alloplast joint placement.

Follow-up has been 13.1±10.1 months (range: 1-43 months) with four lost to follow-up. Surgery took 8.2±2.9 hr, and patients stayed 7.8±3.8 days. Of the 84 flaps, 33 were free, with one acute loss (3%) and three late losses (9.1%). All pedicle flaps survived (total flap survival: 95.1%). Sixty-three patients had adjuvant therapy, 35 preoperatively and 45 postoperatively. Eighteen had brachytherapy radiation starting on 6.5±1.7 days, with no flap loss.

Thirty-nine (52%) patients had wound complications. Delayed wound healing (n=16) and wound seromas (n=15) were the most common. There were 10 non-wound-related complications (18.4%), the most common being nerve palsy (n=4). Twenty patients (27.0%) required operative intervention with wound debridement (n=9) and seroma drainage (n=6). Three developed local recurrence. Five had metastasis preoperatively, with 12 developing metastasis postoperatively; the most common location was the lung. Five patients died, with mean survival at 11.3±7.4 months after surgery. Perioperative mortality was zero. Limb function was graded as excellent (return to preoperative level), moderate (some limitation in range or strength), or poor (minimal use due to pain or poor motion). Forty-one regained full limb use, 24 moderate use, and 5 limited use. Overall, 65 (94%) had full-to-moderate limb function. No limbs were amputated.

This ongoing analysis of the authors' experience with limb salvage for sarcomas indicates favorable results in function of limbs that historically would have been amputated. Although the local wound complication rate was high, limb salvage should be performed whenever possible to avoid the morbidities associated with amputation.

Free Fibula Growth Plate Transfer for Proximal Humerus Reconstruction in Children. Neal S. Topham, Geoffrey L. Robb, and David W. Chang. M.D. Anderson Cancer Center, Houston, TX.

Resection of the proximal humerus in children can result in loss of the humoral growth plate. Limb salvage procedures in these instances require replacement of the growth plate to insure continued growth of the upper extremity, as well as restoration of limb function. Reconstruction can be achieved using a vascularized fibula flap that has been extended to include the epiphyseal growth plate located in the head of the fibula. Successful transfer of the epiphysis requires modification of the harvested fibula flap pedicle to include the anterior tibial artery, with or without the peroneal artery.

The authors discussed the vascular anatomy of the upper fibular epiphysis and presented a patient who underwent proximal humerus reconstruction utilizing the free fibula flap with inclusion of the epiphysial growth plate. The patient was a 9-year-old female with osteosarcoma of the left proximal humerus. Resection of her tumor resulted in an 11.5-cm proximal humerus defect. The 14.5-cm harvested fibula was procured with both the peroneal artery and anterior tibial vessels as pedicles to preserve growth plate viability. The distal fibula was placed within the medullary canal of the cut end of the humerus with 2.5 cm of override. The fibular head was positioned in the glenoid fossa. The knee was stabilized with the remaining lateral collateral ligament and biceps tendon. The anastomosis of the anterior tibial artery to branches of the brachial artery was performed in a retrograde fashion to provide pedicle length.

Four months after reconstruction, there was no knee instability, compared to the uninvolved side. The reconstructed arm had full range of motion of the elbow and the shoulder demonstrated 80 degrees of forward flexion and abduction and 50 degrees of extension. She had no significant internal or external rotation. Radiographic examination showed no sign of subluxation near the glenoid and good alignment of the fibula within the humerus distally. The growth plate was open radiographically, but no measurable growth has yet been demonstrated.

Growth plate transfer for reconstruction of the proximal humerus can be accomplished using the fibula and fibular head, but a detailed understanding of growth plate anatomy and vascular supply is required. Further observation is also required to determine the actual growth of the reconstructed arm.

Devascularizing Complications of Free Fibula Harvest: The Peronea Arteria Magna. Gedge D. Rosson and Navin Singh. Johns Hopkins Hospital, University of Maryland, Baltimore, MD.

A retrospective review of 93 consecutively imaged limbs over 32 months demonstrated a peronea arteria magna (PAM) prevalence of 5.3% in an urban population. Free fibula transfer is now a commonplace technique for long bone and mandibular reconstruction, but the utility of preoperative angiography continues to be debated. The authors presented a case report and performed a cost-effectiveness analysis for preoperative vascular imaging of the donor limb using MRA and traditional angiography (TA).

A 42-year-old man with a shotgun blast to his forearm had a segmental bony defect of the radius of 10 cm reconstructed with a free fibula osteocutaneous flap. Preoperatively, the patient had a normal vascular examination of the donor leg, including palpable DP and PT pulses. Intraoperatively, DP and PT signals were evaluated by Doppler probe; however, in the PACU, the foot appeared mottled with no arterial signal. Angiogram revealed absent PT, spasm of the AT, and a surgically absent peroneal. Vascular bypass was required for salvage.

Donor-site complications of fibula harvest range from 20 to 30%, but are rarely limb-threatening. Limb loss is a dreaded complication of congenital PAM, which can be present with a normal vascular examination. The reported incidence of PAM from the plastic surgery and radiology literature ranges from 0.2% to 8.3%. In the reported case of a patient with unrecognized PAM, the excess direct medical cost of the complication was $140,000. The authors performed a cost-effectiveness analysis using the cost of arterial imaging at their institution, which suggested that TA is expected to either cost $1020 excess or save $6100 per patient, depending on the actual prevalence. Similarly, MRA is expected to either cost $144 excess or save $6900 per patient. This represents the direct medical costs and is an underestimate of the actual costs, which might include time lost from work, disability, pain, and medico-legal action.

Some microsurgery groups advocate using no preoperative imaging of the donor limb; they rely on intraoperative assessment of the vascular anatomy. An aborted harvest due to aberrant anatomy leads to both direct and indirect added costs. The authors' group believes that MRA imaging of the donor limb, being minimally invasive, is cost effective and is indicated for all free fibula transfers. For equivocal results, conversion to more invasive and costly TA may be necessary.

Vein Grafts as a Useful Tool for Achieving Optimal Results in Digit Replantation. M.H. Cheng, Chang-Gung Memorial Hospital, Taipei, Taiwan.

Since the first reported case of a successful digit replantation, new instrumentation and surgical techniques have been developed, which have significantly improved survival as well as functional outcome of these replantations. Reported success rates in the recent literature have ranged between 60 and 80 percent survival. Many methods are employed to achieve good results. The purpose of this reported study was to evaluate the methods used by a single surgeon, including the use of vein grafts in the majority of cases, to achieve a success rate of over 95% survival.

Between April, 1994 and April 1998, 64 consecutive digits were replanted in 53 patients by a single surgeon. A retrospective chart review was performed on all patients. Survival, complications, and functional outcomes were evaluated.

The mechanism of injury was crush in 47 patients, clean-cut amputation in 12, and avulsion in 5 patients. The level of amputation was zone 1 in 38 patients and zone 2 in 26 patients. Vein grafts were used in 68 vessels; 30 vein grafts were for arterial anastomoses, and 38 for venous anastomoses. In no patient was the bone shortened to decrease the distance between vessel ends. Two patients had shortening of the phalangeal bone due to the severity of the crush injury. There were two complete failures out of 64 digits. Two digits underwent reexploration with salvage of one, and another digit failed and was not reexplored. A success rate of 96.9% was achieved.

As in any surgical procedure, constant attempts were made to improve outcome. In this study, multiple methods including the use of 11-0 suture, limited use of bone shortening, the use of anticoagulant therapy, and meticulous surgical technique, had a significant influence on achieving a low failure rate. However, the use of vein grafts and the choice of the appropriate vessels for vein grafts had a significant role in improving the survival rate, as well as the functional outcome, in the digital replantations. The use of vein grafts as tools to improve results and to achieve a survival of over 95% was emphasized, and indications and methods of use were discussed.

AlloDerm Combined with Free Fascia Lata Graft for the Reconstruction of a Complicated Abdominal-Wall Defect. Gedge D. Rosson and Ronald P. Silverman. Johns Hopkins Hospital, University of Maryland, Baltimore, MD.

These authors presented a novel use of AlloDerm and free fascia lata graft for abdominal-wall reconstruction in a case of complicated recurrent ventral hernia repair. Ventral hernias are a common surgical problem, with high recurrence rates. The usual method of repair involves a synthetic prosthesis, which is suboptimal for patients who are immunocompromised, have an enterocutaneous fistula, have had a recent intraabdominal infection, or have intraoperative enterotomies. Patients can develop wound infections, necessitating removal of the synthetic material. Autologous fascial grafts become vascularized and resist infection better than synthetic mesh. However, there is a limited supply of fascia lata and its use produces donor-site morbidity in the thigh.

The case report presented involved a 57-year-old man with a history of multiple enterocutaneous fistulae and hernia repairs. On January 26, 2002, he underwent repair of recurrent ventral hernia. During laparotomy, enterotomies were carried out. The defect measured 20 × 30 cm. A component separation was first performed by incising the external oblique fascia lateral to the rectus muscles, and medially advancing the rectus muscles bilaterally. The fascia lata was harvested from the thigh and sutured to the abdominal fascial edges. This did not close the entire length of the wound; the remaining defect was 5 × 7 cm. A sheet of AlloDerm was sutured directly to the edge of the fascia lata graft and to the remaining abdominal fascial edges.

On February 19, 2002, the patient was readmitted with purulent drainage from his umbilicus. A CT scan showed an intraabdominal abscess. The AlloDerm was not removed, and he was treated with percutaneous drainage and I.V. antibiotics. He is now healed without signs of infection or hernia recurrence.

AlloDerm is an acellular dermal matrix processed from human cadaver skin. It is efficacious in the treatment of burns, for oral resurfacing, as soft-tissue filler, and for dural replacement. In a rabbit ventral hernia model, AlloDerm revascularized and maintained functional integrity for 30 days. AlloDerm produces no donor- site morbidity and is readily available. The authors believe that this was the first report of free fascia lata graft sutured to AlloDerm for abdominal-wall reconstruction.

Bowel Preparation before Free Colon Transfer: Is It Necessary? Samir Mardini, Christopher J. Salgado, and Hung-Chi Chen. Chang-Gung Memorial Hospital, Taipei, Taiwan.

Basic surgical principles emphasize the need for bowel preparation before elective colorectal surgery. In emergent situations, a two-stage procedure is performed in which bowel anastomosis is carried out in the second stage. This is done to avoid complications associated with anastomotic leakage causing bacterial contamination of the intraabdominal viscera. This basic surgical principle has been questioned by several investigators in relation to elective colorectal surgery. The goal of this study was to evaluate the safety and effectiveness of free-tissue transfer of a segment of colon without preoperative bowel preparation.

Patients who underwent upper esophageal reconstruction with a segment of colon between September, 2001 and April, 2002 were evaluated. A retrospective chart review of all patients who did not have preoperative bowel preparation was done. All patients included in this study were maintained on a regular diet until the midnight before surgery. There was no selection bias of these patients, as they were a group who did not understand or did not follow instructions given to them by the hospital's department of preoperative testing.

Nine patients underwent reconstruction using a segment of colon with ileum for the reconstruction of the hypopharynx and cervical esophagus, with concomitant reconstruction of the voice tube. All the patients healed and were able to tolerate a soft diet. One patient had a recurrence of tumor. No deaths were noted in the study. One patient had a leak at the upper colon anastomosis, which was repaired using a pectoralis major muscle flap. One patient had an abdominal-wall dehiscence with no evidence of infection.

This was the first report in the literature of free microvascular transfer of a segment of colon without preoperative preparation of the bowel. The outcomes of this study allow the plastic surgeon the option of performing the transfer of colon even in the absence of preoperative preparation of the bowel. However, this option should not be considered as a first choice but, when necessary, it can be done safely and effectively.

Strategies for Successful Reexploration of a Free Jejunum Flap. Hung-Chi Chen and Yueh-Bih Tang. Chang-Gung Memorial Hospital and National Taiwan University Hospital, Taipei, Taiwan.

Free jejunal flap transfer has become the first choice for reconstruction of the pharynx and cervical esophagus. The ischemic time should be as short as possible, since the tissue material has a high metabolic rate and contains intestinal bacterial flora that predisposes to a rapid development of suppuration in the presence of ischemia. Therefore, strategies have evolved for the reexploration of the free jejunal flap.

Two hundred and twenty-three cases of free jejunal flap transfer were reviewed between 1983 and 2002. Among them, 16 cases had been reexplored. The time before revascularization, thrombosis of vessels (artery, vein, or both), other findings on reexploration, and complications were analyzed.

Seven of the 16 reexploration cases had been successfully salvaged. However, two of them developed later stricture in the reconstructed esophagus. Five of them had thrombosis of the artery, and 11 had thrombosis of the vein. The time before revascularization was from 2.5 to 7 hr (mean: 230 min). Among the 5 cases of arterial thrombosis, 4 were salvaged. Among the 11 cases of venous thrombosis, 3 were salvaged, but 2 of these subsequently developed stricture despite the survival of the muscle component of the jejunal flap. Five of the 7 salvaged cases had leakage that required secondary repair.

Early detection of vascular compromise is mandatory for salvage of a free jejunal flap, especially for those with venous thrombosis. The intestine is more susceptible to venous congestion than to arterial occlusion because of hemorrhagic infarct. (This is different from other free flaps.) When congestion is found, release of tension by removing sutures should be done immediately, before sending the patient to the operating room for reexploration. For prevention of additional congestion, the selection of a large recipient vein and the liberal use of skin graft or local flaps should be considered, to avoid compression from tight closure of the wound and a subsequent vicious circle.

Free Flap Salvage with the Vacuum-Assisted Closure Device. Salvatore C. Lettieri and L. Scott Levin. Mayo Clinic, Phoenix, AZ and Duke University Medical Center, Durham, NC.

Since its introduction, the vacuum-assisted closure device (V.A.C.) has been used for a multitude of wound problems. It has been used primarily for wound care and closure. It has also been used for marginally ischemic wounds, in order to promote a demarcation between viable and non-viable tissue. This was a case report of salvage of a free flap using the device.

The case involved a 28-year-old male patient who had a stage IIIb tibia/fibula fracture with extensive degloving of the pre-tibial region. A rectus abdominis free tissue transfer was performed with vein interposition grafts for the arterial and venous anastomoses. The patient suffered repeated arterial anastomotic leaks and repairs over an 8-day period. The anastomosis leaked at the vein interposition graft to the superficial femoral artery (SFA). Repair performed included a vein patch graft to the SFA, segmental resection of the SFA with reversed interposition vein graft, and finally an extra-anatomic bypass vein graft. Each of the first repairs leaked, and the extra-anatomic bypass thrombosed.

The SFA was ligated, as was the arterial inflow to the free flap, leaving the flap with just the wound bed as its only blood inflow from soft-tissue inosculation. The outflow via the interposition vein graft to the superficial femoral vein was intact. The flap was viable during all the revascularization procedures. After the ligation of the inflow, there was a change in the appearance of the flap consistent with ischemia. The wound V.A.C. was applied, and there was pronounced recovery of the flap. The V.A.C. was changed every other day for 10 days, and a split-thickness skin graft was then applied.

The patient had an external fixator removed 10 months later, after there was satisfactory healing of the fracture. No bone grafting was required, the patient did not require any revision of the flap, and there was a symmetric appearance.

The vacuum-assisted closure device has been used on wounds to help with closure and to prepare wound beds. It has also been used to help ischemic tissue secondary to trauma or burns. In this reported case, the algorithm for patient care followed accepted treatment for the severe complication of infected anastomoses. The final treatment, though, incorporated the V.A.C. and saved the patient from a number of more complicated limb-salvage measures, and possible amputation.

Trans-Endothelial Migration of Lymphocytes is Synergistically Enhanced by the Nitric Oxide Precursor, L-Arginine. Jeffrey Scott Isenberg. University of Oklahoma Health Sciences Center, Oklahoma City, OK.

Tremendous advances have occurred in the technical and anatomic areas of microsurgical transplantation. However, elucidation of the biomolecular events of such reconstructive units remains to be made. Particularly lacking is an understanding of the events that occur at the endothelial cell interface during periods of ischemia and secondary inflammation. Such information would be especially useful in the development of pharmacologic and/or molecular therapies for low flow and no-flow situations in microsurgical transplantation. The purpose of this study was to evaluate the migration of peripheral blood monocytes across cytokine-stimulated vascular endothelium, both with and without nitric oxide donor.

Primary cultures of human umbilical vein endothelial cells (HUVEC) were established from cords taken following birth of healthy-term newborns. Peripheral blood lymphocytes (PBLC) were obtained via standard Histopaque centrifugation. A single healthy donor was used for each assay. Migration of peripheral blood lymphocytes across a confluent layer of HUVEC was assayed in chemotaxis Transwell cell culture chambers (Costar). Monolayers of HUVEC were stimulated with IL-1B (10 ng per well) or BSA-anti-BSA immune complex (100-200 ug/ml/well), in the presence or absence of the nitric oxide donor, L-arginine (100-200 uM), and then seeded with PBMCs (1 x 10⁶ cells), and incubated for 48 hr. Cell counts of the supernatant were then made and levels of TNF-a and INF-g determined, using standard ELISA kits (Pharmigen). Differences between control and experimental samples were analyzed by a two-tailed Student's t-test and a commercial software package (GraphPad Prism, San Diego, CA).

Transendothelial migration (TEM) of PBLCs was not increased in the absence of IL-1B or BSA-anti-BSA immune complex, with or without L-arginine. Stimulation of HUVEC with cytokines IL-1B or BSA-anti-BSA immune complex did not result in increased TNF-a production, again regardless of the presence of L-arginine or not. However, a significant increase in TEM was noted in HUVEC stimulated with Il-1B or BSA-anti-BSA immune complex. Also, levels of INF-g were increased in this group. This response (TEM and INF-g production) was augmented in an additive manner in the presence of L-arginine.

Lymphocyte transendothelial migration was increased following cytokine and immune complex stimulation. This effect was enhanced by pretreatment with the NO precursor, L-arginine. In addition, an increase in INF-g production was noted in HUVEC stimulated with Il-1B or BSA-anti-BSA immune complex. These findings suggest a possible role of NO in the transendothelial migration of inflammatory cells and in the secondary production of proinflammatory molecules.

Evaluation of Fructose Diphosphate for Salvage of Ischemic Gracilis Muscle Flaps in Rats. Vipul Sud, Sheila G Lindley, W. Dorsett-Martin, William C. Lineaweaver, Feng Zhang, Olga McDaniel, and Steve Bigler. University of Mississippi Medical Center, Jackson, MS.

Metabolic intermediate fructose 1,6 diphosphate (FDP) has been observed to bypass the effect of ischemia-reperfusion injury in affected tissue. Prior administration of FDP to ischemic-islanded pedicled skin flaps increases the area of survival significantly. The purpose of this study was to evaluate whether intravascular administration of FDP would help in salvaging ischemic microvascular rat muscle tissue transfer.

Forty-eight adult male Sprague-Dawley rats (400- 450 g) had their right gracilis muscle flap harvested and subsequently intraarterially irrigated with FDP (0.125 g/ kg b.w.). The flap was then anastomosed to the left femoral vessels after varying periods of ischemia. The ischemia time was sset at 1, 1.5, 2., 2.5, 3, and 3.5 hr (n=4 in each group). An equal number of rats were not given FDP treatment to serve as controls. The muscle flaps were reexamined at 72 hr for viability by visual confirmation and histologic examination. Frozen sections were taken at the same time for semi-quantitative estimation of surviving muscle, using electrophoretic estimation of DNA fragmentation of tissues, as well as oxidating mitochondria by succinate dehydrogenase NitroBlueTetrazolium (NBT) stain.

Although visualization and histologic examination did not show any statistically significant difference in survival of the flaps related to the period of ischemia, the FDP-treated flaps showed decreased inflammatory damage in the surviving flaps at the 1- to 1.5-hr ischemia period. This was further confirmed by better quantification of DNA fragmentation electrophoretically in FDP-treated flaps, as well as better staining of surviving oxidative mitochondria by succinate NBT staining.

Administration of FDP after flap harvest did not affect flap survival relative to ischemia time. However, FDP did decrease inflammatory damage in the surviving flap after ischemia time of 1 to 1.5 hr. Such an effect could be useful in salvage of complicated flaps or replanted tissue requiring reexploration or multiple anastomotic revision. Further methods of delivery are being investigated to obtain better survival rates.

Expanded Parascapular Cross Jump Free Flap. Sima Molavi, Kamal S. Forootan, and Ahmad Maghari. Hazrat Fatemeh Hospital, Tehran, Iran.

There are rare instances when only one huge, suitable, free flap should be transferred, and rarer cases when this huge flap should be applied for both legs and knees. These authors have applied such a huge, expanded, scapular free flap for both legs and knees in a patient whose case was reported.

A 13-year-old, crippled, short stature, Afghan girl was referred for treatment of bilateral hypertrophic scars and post burn flexion deformity of both knees in 90° flexion, with chronic recurrent ulcers. An expanded scapular free flap was planned. An 1100-cc rectangular expander was applied to the right scapular area, after providing a subcutaneous pocket; simultaneously, transverse incisions were made in both popliteal areas, releasing scar tissue and allowing for gradual straightening of the knees through skeletal traction. A skin graft was harvested and applied to the popliteal zone after 2 weeks. Controlled expansion was performed to 4450 cc over 6 months. A Corlett loop was attempted 48 hr prior to microvascular free flap transfer, but the loop clotted several times because of the extensive chronic damage and inflammation of the vein graft.

The right radial vessels were chosen as recipients. The expanded scapular flap was raised based on the circumflex scapular vessels. The flap donor site was closed primarily, and the flap was transferred after application of a skeletal fixator to mobilize the knees and forearm. The scars were excised, the middle part of the flap was tubed, covering the right knee, crossing to the left knee to cover it and part of the expanded popliteal area. The circumflex vessels were anastomosed to the right radial vessels under X8 magnification with end-to-end anastomoses. The operation was uneventful, and postoperative management was usual. The patient was seated in a semi seated position for 4 weeks.

Three weeks post procedure, fluorcein and tourniquet tests revealed incomplete perfusion of the flap; by 4 weeks, there was complete perfusion, so the flap was divided from the right forearm, and it survived well.

A similar procedure was carried out in May, 2001, involving a 12-year-old Iranian girl who had sustained a complete degloving injury in a motor vehicle accident. In this case, the vascular pedicle could be divided 5 weeks after tourniquet tests.

Prevention of Venous Compromise in Reversed Fasciocutaneous Flaps. Navin Singh. Johns Hopkins University School of Medicine, Baltimore, MD.

Several reversed fasciocutaneous flaps are currently utilized in plastic surgery, including radial forearm, sural artery, and lateral arm flaps. Venous outflow is contingent on a favorable circuitous flow through communicating veins. Venous congestion is a recognized complication, and inclusion of secondary draining veins is often recommended. This author advocated placing a draining angiocatheter into the distal end of the vein - at the opposite pole from the venous anastomosis - which can be aspirated to alleviate any venous congestion. The technique was demonstrated in four case reports.

The first patient undergoing limb salvage with a reversed sural artery flap developed on-table venous congestion. Every 30 min, aspiration of 5 cc led to complete resolution of the congestion, and 100% survival with uneventful limb salvage. In a second patient, venous congestion was alleviated with drainage; however, only 50% of the flap was salvaged. The remainder of the wound was successfully skin grafted after 1 month of VAC application. For the third patient, an angiocatheter was placed prophylactically but was not utilized, as the flap clinically remained viable. Similarly, in a radial forearm reversed flap for dorsum of the hand coverage, drainage was initiated on the table for 1 hr until the flap equilibrated. Subsequently, the angiocatheter was removed at 24 hr without any untoward effect. In the fourth patient, flap congestion was transient and improved without any drainage.

This simple technique was previously reported by others in the literature for drainage of TRAM flaps by placement of an angiocatheter in an adequately sized subcutaneous vein, but is usefully extended in reversed flaps that are by design predisposed to congestion. Conceptually, a draining angiocatheter in the distal vein of a reversed flap seems sound. It may be used to mechanically “leech” a flap until it equilibrates. Conversely, decreased venous pressure may alter the physiology of the flap, preventing the cross-communicating veins from reversing flow. Further clinical and bench work would serve to elucidate the applicability of this technique in both pedicled and free reversed flaps.

Versatility in Breast Reconstruction with the Deep Inferior Epigastric Perforator (DIEP) Free Flap. Yoav Barnea, Aharon Amir, Arik Zaretski, David Leshem, Jerry Weiss, Raphael Shafir, and Eyal Gur. Sackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel.

Breast reconstruction using abdominal-wall tissue has become the gold standard in autologous reconstruction for the past four decades and has proven to be a reliable and safe technique. Deep inferior epigastric perforator (DIEP) flap breast reconstruction is an improved method of autologous tissue breast reconstruction, with minimal insult to the abdominal wall. The DIEP flap, including skin and fat only, spares the rectus abdominis muscle, and therefore preserves adequate abdominal-wall competence. The flap, with its long pedicle, is highly versatile, with minimal inset limitations, and can be molded into virtually any breast shape desired. The lower abdomen donor site can provide sufficient material for partial, total, and even bilateral breast reconstruction. The flap is compatible with immediate or delayed reconstruction, matching various breast sizes, and is natural looking and cosmetically acceptable.

The authors presented their series of 18 patients who underwent breast reconstruction with the DIEP flap. Oncological breast surgery included radical mastectomy (3 patients), modified radical mastectomy (10 patients), bilateral mastectomy (1 patient), skin-sparing mastectomy (1 patient), and lumpectomy/quadrantectomy (3 patients). Adjuvant radiation therapy was given to 12 patients. In 2 patients (11%), the reconstruction was done immediately, and in 16 patients (89%), was delayed. In 3 patients, the reconstruction with the DIEP flap took place after the failure of implant reconstruction.

The overall success rate was 100%, with three complicated cases, including one distal flap fat necrosis, one partial skin loss at the distal flap edge, and one small dehiscence of the donor-site wound. All abdominal donor sites were intact, with no signs of herniation or bulging. All patients, while supine, were able to elevate both feet to 10 degrees or more of bed level at postoperative day 7.

The authors concluded that breast reconstruction with the DIEP flap is a highly versatile reconstructive method, applicable for many breast ablative surgery options.

Late Venous Thrombosis in TRAM Flap Breast Reconstruction. Christopher J.M. Brooks and Joseph M. Serletti. University of Rochester, Rochester, NY.

Since its introduction in 1982, the TRAM flap has become the mainstay of autologous breast reconstruction. The free TRAM flap, with its low incidence of both total and partial flap loss, is the method of choice in many institutions. Flap failure, while uncommon, is usually secondary to vascular thrombosis. While most series report venous thrombosis occurring within the first 48 hr following surgery, these authors have observed an unusual occurrence of late venous thrombosis in their free TRAM flap patients. This paper reported on their experience of venous thrombosis in free TRAM flap breast reconstruction, highlighting the significant incidence of late venous thrombosis.

A retrospective review of all free TRAM flaps for breast reconstruction at the authors' participating institutions was performed, identifying patients who had postoperative venous thrombosis. Over 700 free TRAM were performed since 1992. Ten patients developed venous thrombosis. Of these 10, five were discovered in the first 48 hr. The remaining five were not identified until between 4 and 11 days postoperatively. Four of the patients had already been discharged from the hospital and returned because of changes in their breast reconstruction, while the fifth was discovered on the evening prior to her planned discharge. All five of the flaps with early thrombosis were salvaged by repeat anastomosis. Of the five flaps with late venous thrombosis, two resulted in complete flap loss, while the remaining three were salvaged by repeat anastomosis and urokinase infusion.

The majority of postoperative venous thromboses have previously been reported to occur in the early postoperative period (within 48 hr). These authors' experience in free TRAM flap breast reconstruction differs significantly, in that half of the thromboses did not occur until the patients were discharged from the hospital or, in one case, on the evening prior to discharge. While the mechanism or prevention of delayed thrombosis are unclear, at the minimum, patients should be educated on the signs of venous thrombosis and should have a low threshold for contacting their surgeons in the event of any flap changes.

A Subjective Rating Scale for Evaluating the Appearance Outcome of Autologous Breast Reconstruction by Physicians and Patients. Michael Cohen, Bradley Evanoff, and Keith E. Brandt. Washington University School of Medicine, St. Louis, MO.

These authors attempted to create and assess the reliability of a subjective evaluation instrument to be used by both physicians and patients to judge the results of post-mastectomy autologous breast reconstruction, and to examine correlations between patient and medical opinions.

An instrument was created based on a literature review of appearance criteria used to assess autologous breast reconstructions. Criteria included in the rating scale were positioning, defects, projection, shape, inframammary fold, medial contour, and overall appearance. Postoperative photographic images of 32 patients who had previously undergone TRAM flap breast reconstruction were evaluated by four medical judges (2 microsurgery fellows and 2 breast center nurses at Barnes-Jewish Hospital,) and by the patients themselves. In addition, the patients rated their overall satisfaction. The evaluations were performed twice, one month apart, by the judges and the patients.

Test-retest reliability for each aesthetic item on the scale was better for patients than for judges (weighted kappas, 0.53 - 0.87 vs. 0.46 - 0.63). The instrument demonstrated slightly better overall internal consistency when used by patients (Chronbach's alpha value of 0.92) than by judges (alpha, 0.85 - 0.91). In addition, for patient evaluations, all individual aesthetic criterion scores correlated with the overall appearance self-score (Spearman's rho, 0.55 - 0.89). For judge responses, most, but not all, individual aesthetic criteria correlated with the overall appearance scores (rho, 0.50 - 0.73); positioning, inframmary fold, and medial contour scores correlated less with the overall judge scores (rho, 0.28 - 0.38). Patient ratings of overall appearance and of overall satisfaction correlated highly with each other (rho, 0.81). Overall inter-rater agreement on each criterion was poor among judges, as well as between judges and patients (weighted kappa, -0.07 - 0.35). Finally, no correlation was found between the judges' perceptions of any individual criterion and patient overall appearance or satisfaction scores (rho, -0.19 - 0.56).

The instrument demonstrates a slightly better internal consistency and reproducibility when used by patients than by medical professions. Expert evaluations did not correlate well with patient evaluations, and the poor inter-observer agreement among the judges creates a potential problem for interpreting independent judge scores. The data suggest that patient input regarding item-specific criteria should definitely be included in evaluations of breast reconstructions. Ratings of the same set of patient images are being performed by a panel of reconstructive surgeon judges, and were included in the presentation.

Skin-Sparing Mastectomy and Autogenous Microsurgical Breast Reconstruction in an Asian Population: The Influence of Biopsy Scar Location and Skin Preservation on Aesthetic Outcomes. Karen W.E. Sng, Bien-Keem Tan, Soo-Wan Hong, Siew Bok Wee, Bee Kiat Ang, Winston Chew, and Kok-Chai Tan. Singapore General Hospital.

Results of skin-sparing mastectomy (SSM) breast reconstruction are superior because of the preservation of the breast skin envelope. The breast biopsy scar location can hinder preservation of breast skin.

Forty-one cases of SSM and reconstruction over 2.5 years were assessed. The authors identified the location of biopsy scars and correlated these with incisions used for mastectomy, and hence the amount of skin preserved. Reconstruction was by autogenous tissue; 88% involved free tissue transfer. Aesthetic outcomes were analyzed regarding breast shape and symmetry, areola positioning, and visibility of scars.

Patients were divided into three groups: Group 1 - complete SSM - only nipple-areola complex (NAC, n=20); Group 2 - partial SSM - NAC + surrounding skin containing scar, n=14); Group 3A - near complete SSM - NAC and scar excised through separate incisions, preserving intervening bridge of skin; 2^nd defect reconstruction with 2^nd skin paddle; Group 3B - near complete SSM - NAC and radially positioned scar excised through teardrop incision; the triangular tail can be closed during nipple reconstruction to give a linear scar (n=2).

There was one partial flap loss because of fat necrosis. Donor-site problems were minimal: 1 hematoma and 1 abdominal hernia. Group 1 (complete SSM) had superior aesthetic results; the skin preservation provided a good shape and symmetry and, after nipple-areola reconstruction, the breast was virtually scarless. In Group 2 (partial SSM), the neo-NAC was surrounded by a circular rim of donor skin, creating a “circle-within-a-circle” appearance. The modified approaches in Group 3 helped to maximize skin preservation, minimize scars, and retain the anatomic position of the areola.

From an aesthetic viewpoint, the breast can be divided into three zones: zone 1 - within the NAC; zone 2->3.5 cm from edge of NAC; zone 3-<3.5 cm from edge of NAC. Biopsy scars should be confined to zone 1. Mastectomy can then be performed through a circumareolar incision, allowing maximal preservation of breast skin. This is especially so in the Asian breast, which is smaller and more amenable to complete SSM. Scars situated in zone 2 can be excised using a separate elliptical incision to conserve skin. Scars in zone 3 should be avoided, whenever possible. If biopsy is performed on a suspicious lesion in this zone, a radially positioned scar is recommended.

Long-Term Results of Transmetacarpal Complete Amputation. N. Gerostathopoulos, D. Efstathopoulos, S. Spiridonos, I.K. Triantafillopoulos, L. Kolintzas, and N. Daoutis. Hand Surgery and Microsurgery Clinic, KAT Hospital, Athens, Greece.

An evaluation of the functional results obtained following replantation of transmetacarpal amputations was undertaken. Over the last 14 years (1987-2001), 22 patients with complete mid-palm amputations were admitted to the authors' department. All of them underwent replantation, while five other patients were rejected on the grounds of age, degree of injury, time under ischemia, or other associated problems. Twenty-one patients were men, and one patient was a young, pregnant woman. The average age of the patients was 31 years. The type of associated soft-tissue trauma was graded at the time of admission as guillotine in 12 patients, crush in six patients, and avulsion in four patients. All patients were available for evaluation and the follow-up averaged 41 months after injury. Evaluations of hand function were based on the degree of sensibility on both sides of each finger, range of motion, and flexion. Power grasp and lateral pinch were measured using a Jamar dynamometer.

Twenty patients were successfully replanted. All guillotine type of amputations were successful. One crush type and one avulsion injury failed due to non-reflow and infection, respectively. The best functional outcome and return of digital sensibility occurred in a 28-year-old patient whose nerves were immediately repaired without tension. Intrinsic muscle function was absent or weak in all patients. Secondary reconstructive operative procedures were performed in all successful cases, including tenolysis, bone-grafting procedures, nerve grafting, and tendon transfers.

Replantation following transmetacarpal amputation led to successful rates of functional restoration and gave cosmetic appearance superior to any prosthesis. However, the return of hand intrinsic muscle function was poor. The functional outcome of the replantation depends on surgical skill, the type of injury, and the age of the patient.

Revascularization/Replantation of Wrist and Distal Forearm Amputations: Experience of Function Outcome from Management of 16 Cases. D. Efstathopoulos, S. Spiridonos, P. Aretaiou, P. Mistidis, S. Seitaridis, and N. Daoutis. Hand Surgery and Microsurgery Clinic, KAT Hospital, Athens, Greece.

Since 1963, when Chen-Zhong Wei first replanted a distal forearm amputation in China, the number of revascularized extremities has increased tremendously. Management of proximal to wrist amputation is critically dependent on the time from injury to revascularization. Cool ischemia must not exceed 6 to 8 hr. Factors such as the level and type of amputation, patient age, as well as concomitant injuries, must also be considered in the decision-making.

From 1988 to 2001, 16 patients who sustained amputations (10 with wrist amputation, 6 with distal forearm amputation) have successfully been treated at the authors' institution. Thirteen patients were males and three were females. The mean patient age was 38 years (range: 17 to 55 years). Fourteen patients had sustained complete amputation, while two patients had incomplete, non-viable amputation. The vast majority of amputations were due to injury by disk saw, two from rotating belts, one from a propeller, and one from a heavy metal sheet. All patients underwent an emergency procedure within less than 6 hr. Bone shortening was performed in a few cases, using plates with K-wires as internal fixators.

The primary goal was primary reconstruction of all anatomic parts, in order to undergo an intense rehabilitation program as soon as possible. All patients but one achieved satisfactory function. According to Chen, criteria used for evaluation were: 3 points in class I, 10 points in class II, 2 points in class III, and 1 point in class IV. Using subjective patient evaluation, all the patients were satisfied, including the one with a poor functional outcome. No patient was willing to substitute a prosthetic device, regardless of outcome.

Technical Particularities in Replantation/Revascularization. Alexandru V. Georgescu, Ivan Ovidiu, and Calin Melincovici. Cluj Napoca University of Medicine, and Rehabilitation Hospital, Cluj Napoca, Romania.

The amputations of various segments of the limbs are often accompanied by complex defects of the soft tissue, bone, or vessels, making an intervention much more complicated than a simple vascular anastomosis. Microsurgical techniques have allowed for the combination of various procedures to restore the vascular axes and for repairing several defects at the same intervention, as well as obtaining superior functional results.

The authors have had experience with 256 cases of replantation/revascularization. In 34 cases, special techniques were used: branched vein grafts for multiple digital amputations (8), flow-through flaps for extensive cutaneous defects (20), composite flow-through flaps with bone or tendon (4), and either simple or composite regional pedicle flaps (2).

In all 34 cases, the distal segment survived. All the associated free flaps survived and were well-integrated. Bone consolidation was faster and limb shortening was avoided in all cases. There were minor reinterventions for tenolysis or flap readjustment.

Skillfully using all the possibilities of micro and plastic surgery, there are only a few cases in which a replantation cannot be performed. The presence of a well-vascularized bone graft in composite flaps stimulates a more rapid consolidation, and avoiding limb shortening allows for a better and faster rehabilitation. The composite flow-through flap is an excellent choice for many complex traumas with concurrent vascular, tendinous, or bone defects.

Neglected Scaphoid Non-Unions: Reconstruction with Vascularized Bone Grafts. K.N. Malizos, Z.H. Dailiana, V. Zachos, and S. Varitimid. HSRM, Greece.

A neglected non-union of the scaphoid of the wrist causes secondary changes that limit joint function, and imposes difficulties in establishing bone union. A vascularized bone graft from the distal radius (dorsal or palmar) has been prospectively utilized in 40 patients with a 3.7-year (range: 1 to 17 years) non-united scaphoid fracture, and in two patients with enchondroma.

Forty patients were males and two were females, with ages ranging from 19 to 52 years. The injury affected the dominant hand in 34 cases. Six of the patients had had an unsuccessful previous surgical procedure, and scapho-styloid arthritis was present at the time of the procedure in 11 patients. Preoperative evaluation included x-ray examination of both wrists and additional imaging evaluation (tomograms-CT-MRI) in selected cases, as well as assessment of pain, range of motion, and grip strength.

A dorsal radial vascularized bone graft was used in 38 patients, and a palmar graft in four. The procedure was carried out under axillary block anesthesia in 34 cases. Through a dorsal or palmar approach, the respective branch of the radial artery (recurrent intercompartmental or pronator quadratus branch) was identified, and a 12 × 10-mm rectangular cortico-cancellous graft of at least 10 mm in depth was elevated, with the vascular bundle protected by a 3-mm wide strip of fascia and periosteum.

The pathologic tissue from the scaphoid non-union was excised and, after curettage and tourniquet release, punctuate bleeding was assessed; this was absent in the proximal segment in 13 cases. The length of the (proximal and distal) scaphoid segments was subtracted from the length of the normal scaphoid and determined the dimensions of the graft, which was interposed to the post-curettage defect (intercalary graft) and secured with K-wires. A long arm cast was applied for 6 weeks and, after removal of the wires, a short arm cast was applied for 1 to 5 more weeks. In 7 of the 11 cases with scapho-styloid arthritis, a closed wedge osteotomy of the distal radius was carried out at the same operation.

Follow-up time ranged from 12 to 86 months. Union and graft incorporation were assessed on x-rays in the majority of patients and on CT scans in 5 patients. Union was accomplished in all cases over a time period of 7 to 11 weeks. The length of the scaphoid was reconstructed in a range from 76 to 93% (average: 85%) of the normal side. Wrist flexion-extension improved from 63 degrees preoperatively to 86 degrees postoperatively, and radial-ulnar deviation from 17 degrees to 29 degrees. Grip strength improved from 67% to 88% of the contralateral site. Complete absence of pain in any kind of activity was noted in 31 patients. Despite fracture healing, in the patients with arthritic changes, pain with strenuous use of the hand was not completely eliminated postoperatively. All patients returned to their previous occupations in a period of 4 to 6 months, and were satisfied with the general improvement of wrist function.

The single surgical approach, the possibility of completing the procedure under regional anesthesia, the ease of dissection of the graft with a constant vascular pattern, and avoidance of anastomosis, in addition to the very high rate of successful healing, are certain advantages of the technique used in this series. Findings demonstrated the efficacy of the vascularized pedicle graft from the adjacent radius in establishing union and correcting major alignment defects in neglected scaphoid fractures. In cases of degenerative changes of the joint architecture, the final functional outcome was compromised.

Elbow Flexion Reanimation Via Musculocutaneous Neurotization in Brachial Plexus Palsy: Preliminary Results. M.D. Vekris, A.E. Beris, A.V. Korobilias, and P.N. Soucacos. HSSRM, Greece.

> In brachial plexus palsy patients, the neurotization of the musculocutaneous nerve is one of the primary goals in the reconstruction of the injured plexus, since the return of elbow flexion is of paramount importance.

During the last 4 years (March, 1998 to March, 2002), 51 adult patients with post-traumatic brachial plexus palsy were operated on in the authors' clinic. The patient mean age was 24.5 years and the most common cause of injury was motor vehicle accident. Exploration of the brachial plexus was performed in 39 patients, while 12 were late cases (2 years) and only secondary procedures were done (e.g., muscle transfer). Seven patients had an extended infraclavicular lesion, and from the 32 supraclavicular lesions, 21 had an element of avulsion (4 global, 10 with four root avulsions, and 7 with three roots avulsed). Neurotization of the musculocutaneous nerve was performed in the majority of cases via nerve grafts from the intraplexus (C5, C7, C8). Extraplexus donors were utilized in 14 patients; in 7 patients, the phrenic was used alone or with intraplexus donors (5). In three cases, the accessory nerve was used, in one patient, the accessory and cervical plexus motor branches, and in three patients, three intercostals were used.

All patients with intraplexus neurotization of the musculocutaneous nerve except two, regained useful function of the biceps (M3+ to M4++). In extraplexus neurotization, the phrenic nerve as a conjunctant donor gave some functional result; when used alone, the results were M3 and M3-. Accessory neurotization yielded M3+ in combination with the cervical motors and M3- when used alone. Intercostal neurotizations provided M2+ and M3.

In brachial plexus paralysis, when the element of avulsion is present, neurotization is often based on extraplexus donors. The return of biceps function is greater and more rapid when intraplexus donors are utilized. Extraplexus neurotization can give satisfactory results when used in combinations.

Shoulder Fusion and Free Functioning Gracilis Transplantation in Patients with Elbow and Shoulder Paralysis Caused by Poliomyelitis. Tarek Abdalla El-Gammal, Amr El-Sayed Ali, and Mohammed Mostafa Kotb. Assiut University, Assiut, Egypt.

Six children between 7 and 16 years of age presented with flail shoulder and elbow caused by poliomyelitis. Shoulder fusion was followed by free functioning gracilis transplantation to replace the atrophied biceps muscle. The transplanted muscle was reinnervated by either the spinal accessory or the phrenic nerve. The follow-up averaged 44 months (range: 56 to 23 months). All patients developed at least a grade 3 power of elbow flexion and were able to place their hands to their mouths. Five of six cases were able to flex the elbow against resistance. One patient required tension readjustment, and elbow flexion contracture of 45 degrees developed in another patient. On average, the transplanted gracilis began to contract 3 months after transplantation, muscle power reached grade 2 at 5 to 6 months, and grade 3 at 9 to 12 months. Muscles supplied by the spinal accessory nerve contracted earlier and ultimately attained more power than those supplied by the phrenic nerve, probably because of easier rehabilitation. Shoulder fusion and free functioning gracilis transplantation for biceps replacement provide a solution for restoration of function in children with flail shoulder and elbow, caused by poliomyelitis. The procedure can be used in other neuromuscular conditions, such as late presenting Erb's palsy, especially when no other muscles are available for local transfer.

Gracilis Fasciocutaneous Perforator Flap: A Clinico-Anatomic Study. E. Lykoudis, C. Vlastou, and D. Kintos. Hygeia Hospital, Athens, Greece.

It is well-documented in the literature that only the proximal third of the skin overlying the gracilis muscle is reliable, when harvested with the muscle as a musculocutaneous flap. This is attributed to the constant presence of musculocutaneous perforators, branches of the vascular pedicle of the gracilis muscle, which nourish the overlying fascia and skin. In this study, the possibility of raising a fasciocutaneous flap based on these perforators was investigated.

The study was performed on a clinical and an anatomic basis. The clinical portion comprised two groups (A,B). In Group 1, both legs of 30 patients were examined by Doppler ultrasound for the presence and location of a sizable perforating vessel. Group B consisted of 12 patients who underwent unilateral gracilis muscle harvesting for various reasons. During muscle harvesting, the presence of sizable perforating vessels was recorded. The anatomic portion comprised one group (C) of 15 cadavers. In this group, the presence of a sizable perforating vessel was also recorded.

In Group A, Doppler examination located bilaterally a sizable perforator at a distance of 5 to 8 cm from the origin of the gracilis muscle, in 29 of 30 cases. In Group B, operative dissection revealed the existence of a musculocutaneous perforator of adequate diameter in 11 of 12 patients, who underwent gracilis muscle harvesting. Regarding the anatomic study (Group C), a significant perforating musculocutaneous vessel on both muscles was recorded bilaterally in 14 of the 15 cadavers dissected. The perforators were located at a mean distance of 6.2 cm from the origin of the muscle. The perforators were then dissected through the muscle to their origin from the vascular pedicle of the gracilis.

According to the authors' findings, in more than 95% of limbs, there exists a sizable musculocutaneous perforator coming through the muscle proximal to the main vascular pedicle, to the skin and fascia overlying the proximal third of the gracilis muscle, making the harvesting of a fasciocutaneous perforator flap quite feasible.

Thoracic Outlet Syndrome: Is There Any Treatment? G. Bouhlis, C.H. Eliades, C.H. Triantafillou, and Al Koskinas. Hospital Henry Dunant, Mesogion, Athens, Greece.

A total of 26 patients (20 women and 6 men) suffered from neurogenic thoracic outlet syndrome (TOS) and were reviewed. The patient mean age was 31.5 years (range: 20 to 52 years). The follow-up was 33.53 months (range: 1 to 76 months). All patients presented with pain and half of them with neurologic symptoms as well. Of 132 patients referred to the authors' unit with TOS during the last 7 years, these 26 patients underwent surgical decompression. The decision for a surgical procedure was made after the failure of conservative treatment for more than 6 months. In the majority of cases, the procedure was supraclavicular decompression, while decompression at more than one level was necessary in three cases. Nineteen of the 26 patients had perfect or very good results regarding pain and the recovery of muscle strength. Five patients had fair results, and two of them had full recurrence of symptoms during the first 2 months postoperatively; these latter patients refused re-operation.

Thoracic Outlet Syndrome: Reappraisal of the Value of Surgical Treatment by Improvement of Technique. David Chwei-Chin Chuang. Chang Gung Hospital, Taipei, Taiwan.

Thoracic outlet syndrome (TOS) is a complex entrapment syndrome of the brachial plexus in the thoracic outlet, consisting of various vague symptoms and signs. Surgical treatment for TOS is controversial due to the demands of the surgical techniques and the high rate of complications.

From 1993 to 2001, after an 8-year experience in brachial plexus exploration, 35 highly selected patients (positive in clinical examinations, reconfirmed by a neurologist, and failures in conservative treatment), with 38 TOSs, were treated by surgery at the author's institution by the same surgeon. Three were operated on bilaterally at different times. Five TOS patients with fractured clavicles were excluded. All were subjected to a supraclavicular and/or subclavicular approach by a 10-cm C-curved neck incision with the apex toward the sternoclavicular junction. The upper, middle, and lower trunks of the brachial plexus were identified and protected. Nearly total removal of the scalene anterior muscle and resection of the first rib and cervical rib (if present) with help of Kellison punch forceps was the surgical procedure.

Thirty-five TOSs (92%) had excellent to good results. Only three TOSs in three patients had fair results: two had further cervical spine laminectomy for abnormal vein compression in the cord, one due to recurrent lung apex tumor (thoracic neurofibroma). Complications were minimal: two pleural effusions, one lymph leakage, and four temporal motor weaknesses or numbness in the operated limb. There was no case with recurrence of symptoms and signs, except the case of recurrent lung apex neurofibroma.

Surgical decompression with this improved technique was re-appraised. The modified and improved surgical technique is well supported and will be kept and utilized for the treatment of intractable TOS.

In Vivo Effect of NGF-7S on Motor Neuronal Regeneration through Nerve and Vein Grafts. A. Gravvanis, E. Lykoudie, G. Tagaris, A. Papalois, P. Panayotou, and I. Ioannovich. Microsurgery and Burn Center, General Hospital of Athens, G. Gennimatas, Athens, Greece.

The aim of this study was to investigate the effect of local administration of nerve growth factor-7S (NGF-7S) on the axonal regrowth of mixed peripheral nerves through nerve or inside-out vein grafts.

One hundred and twenty male Wistar rats were randomized into four groups of 30 animals each. A defect, 12 mm long in the right sciatic nerve, was created and repaired either with a nerve autograft (Groups A, B) or an inside-out vein graft (Groups C,D). NGF-7S (in Groups A and C) or normal saline (Groups B and D, controls) was locally administered daily during the first 3 weeks. For this purpose, a subcutaneously implanted reservoir connected to a silicone tube was used. Electrophysiologic and histologic-morphometric studies were carried out at 4, 6, 8, 10, and 12 weeks postoperatively (subgroups a, b, c, d, and e, respectively, n=6 each).

The histologic appearance of the cross sections was better in the NGF groups, with better axonal orientation and distribution. Analysis of variance and linear regression analysis showed that NGF consistently provided a significant increase in the average number, myelin thickness, and the diameter of myelinated axons throughout the course of time (Group A vs. B, and Group C vs. D). Most important, electrophysiologic findings verified the beneficial effect of NGF on motoneuronal regeneration, since the recovery of the compound muscle action potential was faster in the NGF-treated groups and reached normal value at 8 (Group A) and at 6 (Group C) weeks.

According to these findings, there is strong evidence of the beneficial effect of NGF-7S on peripheral nerve regeneration through nerve and inside-out vein grafts.

Neuronal Lesion with OX7-Saporin Intraneural Injection: Experimental Study in Rats. Zijie Zhang, Allan J. Belzberg, Richard A. Meyer, and John W. Griffin. Johns Hopkins University, Baltimore, MD.

Painful neuroma can develop after axonal injury and is often resistant to surgical interventions. In this reported study, the authors investigated whether axonal transport of a neuron-specific toxin, OX7-saporin, led to selective loss of neurons in the dorsal root ganglia (DRG) and ventral horn of the spinal cord.

Forty Sprague-Dawley rats were randomly assigned to five groups. In Group A1, fluorogold (FG) was used to label the left posterior tibial nerve, while fluororuby (FR) was used to label the left common peroneal nerve. In Group A2, a reverse labeling method to A1 was used. In Group B (GB), the same method was used as in A1, plus OX7-saporin injection. In Group C (GC), the same method was used as in A1, plus OX7-saporin injection. In Group D (GD), the same method was used as in A2, plus Dulbecco's PBS injection instead of OX7-saporin. At 3 or 6 weeks post injection, the rats were sacrificed by transcardial perfusion. The DRG and spinal cord were harvested and 60-um serial sections were cut and mounted using a sliding freezing microtome. Fluorescent labeling of the neurons in the DRG and spinal cord was observed with a fluorescence microscope using appropriate filters.

The results showed that there was a significant reduction of FG (GB) or FR (GC)-labeled neurons in both the DRG and spinal cord of the OX7-saporin-treated rats. Of interest, the fluorescence-labeled neurons were almost completely destroyed by the OX7-saporin in the spinal cord, while non-targeted neurons in the DRG and spinal cord were kept intact.

These results indicated that OX7-saporin could effectively be transported to the dorsal root ganglia and spinal cord by injection into a selected peripheral nerve, where it could selectively ablate the targeted parent neurons. The suicide transport of neuron-specific toxins offers a potential molecular neurosurgical approach for the treatment of painful neuroma.

Clinical Experience with Distally-Based Soleus Flaps. David A. Jansen, Kathy Langston, Esteban Marten, Ross Hansen, and Ralph Edward Newsome. Tulane University Health Sciences Center, New Orleans, LA.

Distal third traumatic and tumor defects traditionally have been reconstructed with microvascular free flap reconstruction. In many non-academic centers, private practitioners find this onerous and difficult to achieve without a standing microvascular team. Thus, local flaps could potentially be possible alternatives for reconstruction of these difficult areas. Historically, distally based soleus flaps have received minimal attention due to variable survival rates and questions about overall blood supply. The authors presented their clinical series of eight distally based soleus flaps for reconstruction of ankle and lower third traumatic and tumor defects.

A series of eight patients over a one-year period were deemed excellent candidates for distally based soleus flaps; they were presented with variable follow-ups. The data showed that distally based soleus flaps are an excellent alternative for reconstruction in the difficult distal third of the lower extremity. The authors discussed the techniques, pearls, and pitfalls, so that solo practitioners could be comfortable with these reconstructive challenges.

Experience with Reconstruction of Ankle and Foot Defects. C. Vlastou and E. Lykoudis. Hygeia Hospital, Athens, Greece.

The difficulties encountered in reconstructing the ankle and foot are: 1) lack of available large bellies for transfer; and 2) the need for reconstruction of a durable weight-bearing surface. A systematic approach to these problems was presented.

Over the last 10 years, 72 flaps were performed on 64 patients, aged 2 to 75 years (mean: 37 years). The choice of flaps was based on the size, nature, and location of the defect, and the vascular status of the limb. Regional flaps were avoided in severely traumatized extremities, in diabetic limbs, or in patients with vascular impairment. If vascular impairment was present, a bypass graft to restore limb circulation preceded the flap transfer, or an A-V loop was used to provide circulation to the flap. Twenty regional and 52 free flaps were used. Regional flaps included the extensor digitorum brevis (4), dorsalis pedis (2), lateral calcaneal (9), medial plantar (4), and flexor digitorum brevis (1). Free flaps included a) fasciocutaneous (8), deltoid (4), scapular (1), radial forearm (1), lateral arm (1), groin (1); b) myocutaneous (3); and c) skin grafted muscle (41): latissimus dorsi (26), rectus abdominis (4), and gracilis (11).

Fifteen regional and six free flaps were sensory; protective sensation was achieved in all cases. Nine regional and 20 free flaps were used on plantar defects. Nine limbs with vascular impairment, four of which had arterial bypass grafts, were successfully reconstructed with free flaps.

Complications included a partial loss of a regional flap and one free flap failure. One free flap was aborted, due to recipient vessel fibrosis, because of an extended zone of injury. In long-term follow-up, ulceration developed on two skin-grafted muscle free flaps covering heel defects. These were resolved by refashioning the flap and using special footwear. In addition, one extremity was amputated despite successful free flap coverage, because of persistent osteomyelitis and destruction of the ankle joint.

An algorithm for coverage of the ankle and foot was suggested. Sensory flaps are preferable for plantar surface reconstruction, although skin-grafted muscle may be adequate. Custom-made footwear is frequently necessary in cases of plantar defects.

Limb Reconstruction for Sarcomas Located in the Distal Third of the Lower Extremities. E. Gene Deune, Anthony Tufaro, Deborah Frassica, and Frank J. Frassica. Johns Hopkins School of Medicine, Baltimore, MD.

Fifty percent of the 6000 new annual cases of soft-tissue sarcomas occur in the extremities, with a 2:1 ratio between the lower and upper extremities. Those occurring in the distal third of the lower extremity are more challenging to reconstruct because free flaps are required and functional restoration may be necessary.

Between February, 1998 and May, 2002 (51 months), 74 patients underwent limb-sparing surgery and immediate soft-tissue coverage for sarcomas in the authors' institution. Fifty-five of the sarcomas were in the lower extremity, with nine in the distal third (leg and foot). Nine patients were male and three were female, with a mean patient age of 45.9±21.9 years. The most common pathology was malignant fibrous histiocytoma (n=6, 67%) with three high grade, three intermediate grade, and one low grade; two could not be graded. All coverage was done with free flaps (gracilis, 7; rectus abdominis, 2); four patients underwent simultaneous functional restoration surgery. Three had autologous tendon grafts to restore ankle/toe extension, and one underwent placement of an Achilles tendon allograft to restore plantar flexion.

The length of follow-up was 11.8±8.7 months (range: 1-29 months). Of the nine free flaps, one failed. Four patients developed wound-healing complications (donor-site hematoma, delayed skin-graft healing, and pressure ulceration of the insensate flap). Seven patients had adjuvant therapy (two preoperatively, four postoperatively, and one both pre- and postoperatively). No patients developed either metastasis or local recurrence, and no limbs had to be amputated. Postoperative function was rated excellent, good, or poor: five (71%) had excellent function, returning to preoperative activities, and two (29%) reported having moderate function with restrictions of preoperative activities. None reported poor function. Two patients were retired and one was disabled preoperatively. Of the five gainfully employed preoperatively, all have gone back to their previous employment without limitation.

Limb salvage has now become the standard of care in the surgical management of limb sarcomas. The distal third provides an additional challenge because of the need to restore function as well as coverage. Similar to the experience with traumatic wound coverage, defects in the distal third of the lower extremity require free flaps for coverage. The results showed that aggressive management of sarcomas followed by soft-tissue coverage, along with functional restoration surgery, when required, provide excellent outcomes.

Ten-Year Experience in Perforator Flaps for Reconstruction of the Lower Third of the Tibia, Heel and Malleoli. A. Kepenekidis. Burn Hospital Unit, Kifisia, Athens, Greece.

The aim of this study was to show alternatives in the use of perforator flaps to cover soft tissue defects of the lower third of the leg, and to clarify some technical points in flap dissection. The advent of perforator flaps was a breakthrough in the surgical treatment of soft tissue defects in difficult areas, such as the distal tibia, malleoli, and posterior heel. The initial use of fasciocutaneous flaps by Ponten and the description and refinement of several flaps proximally or distally based, enriched the armamentarium of the reconstructive plastic surgeon facing such difficult problems.

During the last 10 years, 1992-2001, perforator flaps have been used in 87 patients presenting soft tissue defects of the lower leg. Fifty-three patients were males (60.9%) and 34 were females (39.1%), with ages varying from 18 to 84 years. In 45 cases, a distally based perforator flap was used as a fasciocutaneous (island or not) or fasciosubcutaneous flap.

In a majority of cases, 62 (71.26%), the postoperative period was uneventful, although some complications were experienced, such as partial necrosis of the flap (especially in the distally based flaps) and infection.

Perforator flaps offered a solution in covering areas in the distal leg, the malleoli, and the heel. One- or two-stage reconstructions with proximally or distally based flaps reduced the indications for microsurgical transfers, and there were several advantages, such as conservation of the vascular axes of the leg, flexible pedicles, no risk from microsurgical anastomoses, reliability and easy dissection. A venous congestion for the first 3-4 postoperative days can be anticipated, with inclusion in the flap base of as many perforators as possible or the occasional use of leeches. In cases of infected bone, the indication is for a muscular, rather than a perforator, flap.

Flaps based on perforator branches of the vascular axes of the leg can solve problems of soft tissue defects in the distal leg, malleolar region, and posterior heel, having several advantages over other reconstructive procedures.

Ten-Year Experience in Foot and Ankle Reconstruction: Pedicled Muscle Flaps vs. Free Flaps. Ivica Ducic, Christopher Attinger, Paul Cooper, Jamie Schwartz, and Ryan Gocke. Georgetown University, Washington, D.C.

Local muscle flaps, pioneered by Ger in the late 1960's, were extensively utilized for foot and ankle reconstruction until the late 1970's when, with the evolution of microsurgery, microsurgical free flaps became the reconstruction of choice. To reassess the role of pedicled vs. free flaps in foot and ankle reconstruction, the authors examined the Georgetown Wound Registry and identified all patients who underwent flap reconstruction from 1990 through 2000, with a 2-10-year follow-up.

By protocol, flap coverage was the reconstructive choice for defects with exposed tendons, joints, or bone. Local muscle flaps were always selected over free flaps, if the defect was small (3×6 cm or less) and within the reach of the local muscle flap. Otherwise, free flaps were utilized. The role of diabetes was also evaluated.

Eighty consecutive patients were identified: 32 patients (19 diabetic and 13 non-diabetic) received local muscle flap reconstruction, while 48 received 52 free flaps (21 diabetic and 30 non-diabetic) for 51 limbs. All wounds after debridement had exposed bone, joint, or tendon at their base. Revascularization prior to reconstruction was required in 34% of diabetics. Subsequently, 240 procedures were required to heal the wounds, of which 42% were limited to debridements only. Thirty-four pedicled muscle flaps were utilized (19 abductor digiti minimi, 9 abductor hallucis brevis, 3 extensor digitorum brevis, 2 flexor digitorum brevis, and 1 flexor digiti minimi). Fifty-two free flaps were done (18 gracilis, 13 rectus abdominis, and 15 fasciocutaneous flaps).

When comparing the success of pedicled vs. free flaps, there were no significant differences in limb salvage (94 vs. 88%), survival (78 vs. 75%), healing rate (87 vs. 94%), days to heal (99 vs. 109 days), flap success rate (97 vs. 94%). The complication rate was higher in pedicled than free flaps (33 vs. 18%), and almost reached statistical significance (p=0.051). Diabetes did not affect any of the above parameters except long-term patient survival.

Local muscle flaps are as effective as free flaps for coverage of foot and ankle defects. Proper selection based on the size and location of the defect provides the reconstructive surgeon with two excellent options for soft tissue coverage. Diabetes does not appear to adversely affect their effectiveness.

Optimizing Graft Positioning in Avascular Necrosis with the Aid of CAD-CAM Technology. A.E. Beris, B.K. Kostopoulos, A.H. Payatakes, A.E. Korobilias, M.D. Vekris, and P.N. Soucacos. University of Ioannina Medical School, Ioannina, Greece.

The extracapsular placement of a vascularized fibular graft in the subchondral region of the femoral head is a widely accepted method for treating patients with avascular necrosis (AVN) of the femoral head. The “Ioannina method” is a technique that aims to improve the precision of targeting the lesion of the femoral head, thus increasing the survival of the femoral head in patients with osteonecrosis.

Twenty-seven patients (30 hips) with femoral head osteonecrosis were treated with this technique in the orthopaedic department during the last 4 years. With the use of CAD-CAM, the lesion is located, and a custom-made metallic aiming device is manufactured. This device is then used to place the graft in an optimal position in the center of the lesion. This group was compared with 20 patients with conventional targeting.

Results were evaluated clinically and radiologically. The short-term results showed precision of targeting in 89% of the patients, compared to 55% using the conventional method. X-ray exposure and operating time were also significantly reduced.

This technique is the result of more than 12 years of experience in the treatment of osteonecrosis with transfer of free vascularized fibular grafts. The technique was designed and developed by the senior author and emphasizes the contemporary trend toward precision in the location and description of the lesion. The method decreases x-ray exposure, minimizes operating time, and optimizes the placement of the graft.

Twelve-Year Experience with the Free Vascularized Fibular Graft for Osteonecrosis of the Femoral Head. A.V. Korompilias, K.N. Malizos, A.E. Beris, and P.N. Soucacos. University of Ioannina Medical School, Ioannina, Greece.

The purpose of this study was to present a 12-year experience with the free vascularized fibular graft (FVFG) for osteonecrosis (ON) of the femoral head.

The results of 184 hips in 152 patients, followed-up for a minimum of 3 years and up to 12 years (average: 6 years) were presented. All patients were evaluated yearly regarding function, as assessed with the Harris Hip Score (HHS), radiographic progression of the disease, and conversion to a total hip replacement (THR). The disease was bilateral in 101 patients (59%) and 34 of them (34%) had bilateral FVFG. Associated etiologic factors included use of steroids (85 hips), consumption of alcohol (22 hips), trauma (27 hips), hemoglobinopathies (9 hips), and autoimmune diseases (14 hips). Idiopathic osteonecrosis was present in the remaining 45 hips. The stage of ON, according to the criteria of Steinberg, was: stage II for 47 hips (23%), stage II for 50 hips (25%), stage IV for 83 hips (40%), and stage V for 24 hips (12%).

Excluding hips that were eventually treated with THR, the mean HHS improved for all groups of hips: from 85 to 96 points for stage II hips, from 74 to 91 points for stage III hips, from 69 to 85 points for stage IV hips, and from 61 to 76 points for stage V hips. The increase was significant for all stages (from p<0.05 to p<0.001, paired two-tailed t-test). Progressive collapse of the femoral head, acetabular changes, or loss of the joint space was noted in 5% of stage II hips, in 36% of stage III hips, in 48% of stage IV hips, and in 61% of stage V hips. In this series, 14 hips (8%) were eventually treated with THR; none that had been in stage II (0%), 1 of 45 that had been in stage III (2%), 9 of 77 that had been in stage IV (12%), and 4 of 23 that had been in stage V (17%). One case of septic arthritis and two cases of deep wound infections all led to THR.

It is generally agreed that ON of the hip can be arrested in most patients, if the procedure is performed prior to development of a subchondral fracture (stages I and II). In stage III, fibular grafts may arrest or delay the progression of the disease. Even in those patients with more advanced ON, including those with collapse of the articular surface, the FVFG may be of benefit, prolonging reduction of symptoms and postponement of THR.

The FVFG is the authors' favored technique in the management of ON of the femoral head. This technique does not violate the joint capsule, as do other bone grafting procedures (pedicled iliac crest, quadratus femoris graft) and, in comparison with conventional grafts, supports the subchondral surface with a viable cortical strut (bone healing process), and enhances the revascularization process.

Physicians with Avascular Necrosis of the Femoral Head: Management with Free Vascularized Fibular Grafting. James R. Urbaniak, Marco Rizzo, Philip Clifford, and Eunice Gunneson. Duke University Medical Center, Durham, NC.

Osteonecrosis of the femoral head is a very challenging medical condition to treat. Since the condition is irreversible and often progressive, surgical management is a mainstay of therapy in symptomatic cases. In young patients, preservation of the femoral head for as long as possible is a goal, as endoprosthesis placement in this group will most likely require at least one revision within their lifetime. The purpose of this study was to compare the results at one institution of free vascularized fibular grafting for the treatment of osteonecrosis between 1) physicians and 2) a control group of the population at large. A second purpose was to determine how compliant physicians are as patients, when asked to follow another physician's instructions.

Physicians treated for stages II, III, and IV avascular necrosis of the femoral head included a treatment period from 1986 to 1999. The control group consisted of the entire population of patients treated with free vascularized fibular grafting from the onset of the senior author's experience, which ranged from 1979 to 1999. A total of 21 physicians (32 hips) were treated: 10 stage II, seven stage III, and 15 stage IV. The control group had a total of 1257 hips: 221 with stage II disease, 220 stage III, and 816 stage IV. All patients were followed for a minimum of 2 years.

A total of 21 physicians (32 hips) were treated for avascular necrosis of the femoral head. The control group consisted of 1257 hips. When compared with the senior author's entire experience treating stages II, III, and IV avascular necrosis, the physician group displayed an excessively high degree of compliance during the immediate postoperative period (p>0.001). In addition, the overall rate of recommended or conversion to total hip arthroplasty was significantly lower in the physician group (6.25), compared to the control group (20.6%). Exercise, weight control, and diet were all greatly stressed in the physician survey.

The better outcome in the physicians over the control group may be related to one or more factors, including extensive non- and partial weight bearing, diet, weight control, and exercise.

Non-Union of Femoral Neck Fractures with Avascular Necrosis of the Femoral Head: Treatment with Free Vascularized Fibular Graft Combined with Subtrochanteric Valgus Osteotomy. A.E. Beris, B.K. Kostopoulos, A.H. Payatakes, A.V. Korobilias, and P.N. Soucacos. University of Ioannina Medical School, Ioannina, Greece.

Non-union and avascular necrosis of the femoral head are the main complications of fractures of the femoral neck. Treatment in young adults is very difficult, as arthroplasty is not indicated. A technique was described, using a free vascularized fibular graft to promote healing and to prevent osteonecrosis, combined with valgus subtrochanteric osteotomy to convert shearing forces into compression and to correct leg-length discrepancy.

Seven patients with non-united femoral neck fracture were treated by valgus osteotomy and free vascularized fibular graft. There were four males and three females, with a mean age of 27.2 years (range: 16 to 39 years). Six of them sustained a femoral neck fracture in a motor vehicle accident and initially underwent fixation with Steinmann pins or cannulated screws. One sustained a pathologic fracture due to an aneurysmal bone cyst of the femoral head and neck. The mean delay between injury and reoperation was 8 months (range: 2 to 16 months). All patients presented some degree of absorption of the femoral neck and avascular necrosis of the femoral head at various stages. To secure the osteotomy in its altered position, a 120 to 150-degree blade plate was used. The amount of valgus angulation varied from 20 to 30 degrees. A derotation screw was placed proximal to the vascularized graft.

There were no intraoperative complications. One patient underwent a second procedure (removal of the implants) 26 months after the osteotomy. Ossseous union was achieved, and there was clinical and radiologic improvement of the avascular necrosis in all patients.

These authors recommended this procedure for the treatment of young adults with difficult intracapsular fractures of the femoral neck complicated with non-union, avascular necrosis, resorption of the neck, and leg shortening.

Role of Vascularized Fibular Grafts for Difficult Cervical Spinal Arthrodesis. Michael J. Lee, Kent V. Hasen, Neil A. Fine, Stephen Ondra, and Gregory A. Dumanian. Northwestern University, Chicago, IL.

Anterior cervical spine fusion remains a mainstay of treatment for spinal instability and deformity; rates of fusion vary from 5 to 40%. Corticocancellous bone grafts and long iliac bone grafts aid in intervertebral fusion; however, in difficult cervical spine arthrodesis cases, such as with infection, radiation, previous or multilevel repair, the likelihood of fusion decreases with these nonvascularized bone grafts. Vascularized bone grafts such as the fibula have increased compressive, as well as torsional, strength, compared to iliac crest or rib, and provide more rapid biomechanical stability and faster rates of incorporation from 3 to 5 months.

Six patients underwent anterior cervical spine arthrodesis with vascularized fibula graft fusion. Five were males and one was female, with a mean patient age of 59.3 years. Etiology included osteomyelitis, failed arthrodesis, cancer resection, and radiation. The fibula graft was harvested in standard fashion and plated by the neurosurgeon. The external carotid and internal jugular vein were most often the target vessels.

Results revealed one intraoperative death secondary to blood loss resulting from the spinal surgery, and one fibular graft thrombosis. The patient with the thrombosed graft received an immediate reconstruction with a second vascularized fibula graft without postoperative complications. Follow-up ranged from 6 to 12 months. Bone scans revealed successful fusion at 3 months in five of the patients. No long-term complications were found at the donor site or the cervical spine.

Vascularized fibular grafts can be an effective and reliable method of fusion for difficult spinal arthrodesis. Because of the high morbidity associated with these spinal surgeries, vascularized fibular grafts should be reserved for failed spinal arthrodesis, radiated or infected spines, or multilevel fusions.

Vascularized Fibular Grafts in the Management of Congenital Pseudarthrosis of the Tibia. A.V. Korompilias, A.E. Beris, K.N. Malizos, T.H. Xenakis, G. Aphendras, M.D. Vekris, C.H. Zalavras, and P.N. Soucacos. University of Ioannina Medical School, Ioannina, Greece.

Congenital pseudarthrosis of the tibia remains one of the most challenging problems confronting the orthopaedic surgeon. The operative results are frequently less than successful, many cases requiring several surgical procedures, and a significant number of these ultimately end in amputation. The purpose of this study was to assess the surgical results, complications, secondary procedures, and long-term results of the vascularized fibula in the treatment of congenital pseudarthrosis of the tibia.

Seven patients who had congenital pseudarthrosis of the tibia were treated consecutively between 1992 and 2000 with free vascularized fibular grafts. There were four females and three males, with a mean patient age at the time of operation averaging 6.5 years (range: 1 to 12 years). Four left tibias and three right tibias were involved. Two patients had been operated on before, but no union was obtained. All had been treated previously with cast and braces. After wide resection of the pseudarthrosis with surrounding fibrotic and thick scar tissue, reconstruction was performed with free vascularized fibular graft from the contralateral leg. Stability was maintained with internal fixation in four patients, external fixation in two patients, and an intramedullary pin in one patient.

The average follow-up was 2.6 years (range: 6 months to 8 years). In five patients, both ends of the graft healed primarily within 2.7 months (range: 1.5 to 3 months), and hypertrophy of the fibular graft occurred rapidly with a well-formed medullary canal. In one patient, the distal junction did not unite and, although requiring three subsequent operations, still did not heal. Stress fracture occurred in the middle of the grafted fibula in one patient, who underwent four additional procedures before union was achieved.

The free vascularized fibula transplant for congenital pseudarthrosis of the tibia is a valid method for treating this complex problem. Despite continuing problems and the relatively high complication rate, the ultimate results are generally good, especially as compared with the published series that report the failure of conventional grafting techniques. However, even achieving union in pseudarthrosis is not sufficient for the resolution of this disease and is only half the problem; the other half is maintenance. Long-term follow-up beyond skeletal maturity, if possible, is necessary to evaluate surgical results.

Upper Extremity Reconstruction with the Free Vascularized Fibular Graft. C. Zalavras, A.E. Beris, K.N. Malizos, A.V. Korompilias, M.D. Vekris, I. Kostas, and P.N. Soucacos. LAC+USC Medical Center, Los Angeles, CA; University of Ioannina Medical School, Ioannina, Greece; University of Thessaly, Larissa, Greece; and Microsurgical Research Center, Norfolk, VA.

Reconstruction of the upper extremity is a challenging problem when extended bone loss has occurred, either primarily during trauma or secondarily after pathologic tissue excision. The authors presented their experience in the management of large skeletal defects of the upper extremity with the free vascularized fibular graft (FVFG).

Eleven patients, aged 7 to 43 years (mean: 27 years) were treated with the FVFG for upper extremity defects ranging from 6 to 18 cm (mean: 10 cm). Follow-up has been from 1 to 10 years (mean: 4 years). Nine defects involved the forearm and two involved the humerus. The etiology of the defects was infected non-union in four cases, tumor in four, osteomyelitis in one, aseptic non-union in one, and congenital pseudarthrosis in one case. The microvascular flap was osseous in 10 and composite in one patient, with transfer of the soleus muscle. Plate and screw fixation was used in all cases.

Graft healing was evident in 20 of 22 junction sites (91%) at an average of 3 months (range: 2.5 to 4 months). No perioperative complications and no stress fractures occurred. Donor-site morbidity was negligible. Functional outcome was good or excellent in 10 of 11 patients (91%), and no amputations were necessary.

Microsurgical reconstruction of the upper extremity with the FVFG constitutes a viable treatment option. The FVFG allows for radical excision of pathologic tissue and improves the biology at the recipient site. The shape and size of the FVFG matches those of the forearm bones and the humerus. Therefore, no hypertrophy is necessary, and early resumption of function is possible on graft incorporation, regardless of the defect dimensions. In the authors' opinion, the FVFG is a treatment of choice for large skeletal defects of the upper extremity.

Free Osteofasciocutaneous Fibula Flap: Nine-Year Experience in 77 Patients. Nikolaos Papadopulos. Technical University of Munich, Germany.

During the last 9 years, a total of 77 free osteofasciocutaneous fibula flaps have been used in several clinical cases, including 52 neophallus constructions in female-to-male transsexuals, 21 composite mandible reconstructions (after tumor surgery or osteoradionecrosis), as well as in three lower leg and one forearm reconstructions for oncologic surgical defects.

In order to evaluate the utility of these flaps, their security was analyzed in relation to the magnitude of the skin paddle used and the fibular length, as well as pedicle length, position and number of the osteofasciocutaneous branches of the peroneal artery, and the modifications performed during flap harvesting and flap fixation at the recipient site. Additionally, donor-site morbidity was examined.

Great security of the blood supply by large skin paddles was observed, although some of these flaps undergo massive modeling of soft tissue and bone according to their clinical application. Donor-site morbidity was moderate.

The utilization of the free osteofasciocutaneous fibula flap is an excellent therapy for difficult composite reconstructions. It provides a long, broad, and strong bicortical bone of good quality that supports physical stress and offers good intrinsic rigidity for neophallus construction, a periosteal blood supply that allows multiple osteotomies, a long vascular pedicle with large vessels, and almost constant anatomy. In addition, the skin paddle has a uniform medium thickness, and is sufficiently large and pliable, and ideal for the reconstruction of associated soft-tissue defects. Donor-site morbidity is relatively insignificant. For these reasons, the widespread utilization of this flap is supported for the reconstruction of composite defects.

Free Vascularized Corticoperiosteal Bone Graft for the Treatment of Radiation Necrosis of the Clavicle. Allen T. Bishop, Bruno Fuchs, and Scott P. Steinmann. Mayo Clinic, Rochester, MN.

Radiation-induced necrosis and resultant pathologic fracture of the clavicle may follow radiation treatment of malignant bone and soft-tissue tumors. Partial excision of the clavicle, although frequently recommended, may result in persistant pain, weakness, instability, and brachial plexopathy. Clavicle reconstruction is a reasonable alternative. Previously reported methods have been problematic, including use of vascularized fibula or rib, or reconstruction with Dacron mesh with or without bone graft. The authors described the use of a free vascularized corticoperiosteal flap as an alternative in these difficult cases.

Two patients with radiation-induced pathologic clavicle fractures were treated with free vascularized corticoperiosteal bone flaps. These included a 36-year-old female treated for dermatofibrosarcoma protuberans overlying the left anterior shoulder and clavicle, and a 55-year-old male with a metastatic squamous cell carcinoma to the left neck. Both were treated by surgical excision of the tumor and subsequent radiation therapy. Spontaneous pathologic fractures occurred 30 and 9 months afterward, respectively, and remained non-united for 12 and 9 months, until the flap procedure. Preoperatively, both patients complained of limited shoulder motion, and rest and activity-related pain. Imaging studies and surgical inspection both demonstrated atrophic non-union and avascular changes.

At surgery, the clavicle was reduced and stabilized with compression plating. A free vascularized corticoperiosteal bone graft was harvested from the medial femoral condyle, based on the descending genicular arteriovenous pedicle. The flap was wrapped around the clavicle, and microvascular anastomoses performed to the thoracoacromial trunk. Vessel patency checks and a bleeding flap surface confirmed successful transfer. A shoulder immobilizer was used for 6 to 8 weeks. Passive ROM exercises were then begun.

Healing was confirmed by tomography at 5 and 7 months, respectively. No donor-site morbidity other than transient medial knee pain was seen. The function of the upper extremity was remarkably improved at final follow-up 2 years later.

Vascularized free corticoperiosteal bone flaps are ideal in the clavicle. The flap is thin and able to be wrapped around the clavicle without visible bulk. This enables clavicular healing at one bone contact site, and avoids the difficulties of interposing, fixing, and healing a bridging vascularized structural graft. The technique of periosteal grafting is ideally suited to chronic non-unions without substantial bone loss, but with poor chances of healing on their own. Rapid subperiosteal new bone and improved local blood flow serve to correct many of those changes attributed to radiation necrosis of bone.

Clinical Correlates for Functional Outcome and Complications in Pharyngoesophageal Reconstruction. Bruce H. Haughey, Blake Anderson, Mina Charepoo, and Yash Patil. Washington University School of Medicine, St. Louis, MO.

Studies of paryngoesophageal (PE) free flap reconstruction seldom report the statistical correlation between clinical variables and functional outcomes.

Fifty-five PE reconstructions were performed following resection or stenosis, with or without laryngectomy. Fifteen variables relevant to functional outcomes and postoperative morbidity were recorded. The Functional Outcome Swallowing Score (FOSS), a 0-5 system, measured swallowing (0-normal swallowing). Uni- and multivariate statistical models were used for outcome analysis. Complication rates were recorded.

Seventy-four percent of resections included total laryngectomy and 81% of reconstructions were performed primarily. Sixty-four percent of patients received prior treatment and 80% underwent neck dissection. The mean best swallowing score was 1.8, achieved at a median interval of 2 (0.5 to 9.5) months. The ultimate swallowing score was 2.4, reached at an interval of 7 (1-84) months from reconstruction. Voice was by larynx in 22% of cases, tracheoesophageal puncture in 35%, and electrolarynx in 43%. Fistulas occurred in 15%. Significant associations existed between poorer best swallow scores (2-5) and fasciocutaneous flaps (odds ratio, O.R., =5.7), comorbidity (O.R=5.4), and use of saliva bypass tubes (BT) (O.R.=10.5) in univariate analyses. In multivariate models, poorer ultimate swallowing scores correlated significantly with the presence of comorbidity (O.R.=5.6) and use of BT (O.R.=4.1). No variables predicted voicing method. Complications were significantly associated with prior treatment (O.R.=8.4), neck dissections (O.R.= 9.6), and primary reconstruction (O.R.=7.8) with multivariate analysis.

Surgeons may wish to limit the use of BT, be cautious in predicting ultimate swallowing results for patients with comorbidity, and expect more neck wound complications/fistulas with salvage surgery, primary reconstruction, and neck dissections.

Critical Oropharyngocutaneous Fistulas after Microsurgical Head and Neck Reconstruction: Indications for Management Using the “Tissue-Plug” Technique. Christopher J. Salgado, Samir Mardini, and Hung-Chi Chen. Darnall U.S. Army Hospital, Ft. Hood, TX, and Chang-Gung Memorial Hospital, Taipei, Taiwan.

Despite advances in head and neck reconstruction with free tissue transfer techniques, oropharyngocutaneous fistulas continue to present challenging and potentially lethal complications. The authors presented a classification system for these fistulas and the management of nine patients who developed critical oropharyngocutaneous fistulas after microsurgical head and neck reconstructions following tumor extirpation or injury from corrosive agents.

The indications for management of the fistulas were primarily dependent on their specific location. Three peristomal (type I) and six midneck (type II) fistulas were considered critical due to their risk of aspiration pneumonia and carotid artery blow-out, respectively. Using the concept of a “tissue plug” for fistula repair, a dermal component of either a deltopectoral (5) or pectoralis major (4) pedicled flap is guided through the fistula with an appropriately-sized gallstone dilator and used to “plug” the tract both internally and externally. Traction is placed on the internal dermal component by way of a non-absorbable suture and tied over an external bolster, with no sutures placed directly in the surrounding friable tissue.

There were no partial or total flap losses. There were two fistula recurrences in patients who had received postoperative radiation therapy. One of the fistula recurrences was believed to be due to tumor recurrence within the previous fistula. All patients resumed oral feeding except for the patient with suspected tumor recurrence.

The authors concluded that using the tissue-plug technique in the management of critical oropharyngocutaneous fistulas serves not only to effectively obliterate the tract, but also to augment vascularity and volume in already damaged, ischemic, and deficient tissues.

Incidence and Etiology of Perioperative Complications in Microvascular Free Flap Reconstruction of the Head and Neck. Jeffrey D. Suh, Joel A. Sercarz, Jeffrey D. Rawnsley, Elliot Abemayor, Thomas C. Calcaterra, Daniel Alam, and Keith E. Blackwell. UCLA, Los Angeles, CA.

The objective of this reported study was to determine the incidence and etiology of perioperative complications in patients who undergo microvascular free flaps for reconstruction of the head and neck. The setting was an academic tertiary care medical center.

Four hundred consecutive microvascular free flaps were performed for reconstruction of the head and neck, with 95.5% of the defects arising after the treatment of malignancies. Donors included radial forearm flaps (n= 183), fibula flaps (n=146), rectus abdominis flaps (n=38), subscapular system flaps (n=27), iliac crest flaps (n=5), and a jejunal flap. Patient-related characteristics (age, sex, diagnosis, comorbidity level, tumor stage, defect site, primary vs. secondary reconstruction, and history of prior surgery, radiation therapy, or chemotherapy) and the incidence of perioperative complications were recorded prospectively over a 7-year period.

The perioperative mortality was 1.3%. Overall, perioperative complications occurred in 37.3% of all cases. Free flaps proved to be extremely reliable, with a 0.8% incidence of free flap failure and a 3% incidence of partial flap necrosis. Perioperative medical complications occurred in 20.5% of cases, with pulmonary (8.8%) and cardiac (6.5%) complications predominating.

This study confirmed that free flaps are extremely reliable for achieving successful reconstruction of the head and neck. The American Society of Anesthesiology comorbidity index, which is relatively simple and widely utilized in clinical practice, compared to previously advocated comorbidity scales, was shown to be a statistically significant (p=0.003) predictor of the incidence of perioperative medical complications.

Free Transfer of Ileo-Colon for Total Reconstruction of Cervical Esophagus and Voice Tube: Experience with Nine Patients with Preliminary Evaluation of Speech. Samir Mardini, Christopher J. Salgado, and Hung-Chi Chen. Chang-Gung Memorial Hospital, Taipei, Taiwan.

Reconstruction of the larynx after total laryngeal resection has classically been treated with prosthetic devices, either placed surgically or used as external devices. Surgical reconstruction of the voice tube has not been common, due to the complexity of the surgery, as well as the inadequacy of the results. In this study, the results of experience with nine patients who underwent simultaneous reconstruction of the trachea and esophagus, using a segment of ileo-colon, were evaluated. Details of the procedures, as well as technical refinements and preliminary results, were presented.

Between September, 2001 and April, 2002, nine patients underwent simultaneous reconstruction of a segment of upper esophagus, as well as the voice tube, using a segment of ileo-colon. Evaluation of outcomes of voice was performed using a three-point scale. A score of 3 was given to a patient who was deemed to have a good result, indicating speech that was intelligible. A score of 2 was given to a patient who had non-intelligible speech, and 1 for a patient who had no speech, either due to the lack of a connection with the trachea at the time or due to a problem with the reconstructed voice tube.

An average of two procedures were performed per patient to complete the reconstruction. Complications included one patient with leakage of the upper cervical anastomosis, one donor-site complication, and one patient with recurrence of tumor. No patient received a score of 3 at this point; seven patients received a score of 2; and two patients received a score of 1. Both of the patients who have received a score of 1 have not completed the final stage of voice tube reconstruction (connection of ileum with trachea). Results of these patients will be evaluated subsequently.

The impact of achieving a successful reconstruction of the voice tube is great, with the rewards being potentially very satisfying. External, as well as surgically placed devices, have achieved adequate intelligible speech; however, the patients are usually unsatisfied by the quality of the voice and the impracticality of using the device. Technical refinements, as well as outcome of speech using this relatively new technique, i.e., a segment of ileo-colon, with or without the appendix, for total reconstruction of the voice tube, were outlined, and the results in speech were demonstrated.

Temporoparietal Free Flap for Reconstruction of the Larynx in the Radiation Failure Setting. Ralph Gilbert, Peter C. Neligan, Patrick J. Gullane, Jonathon Irish, and Dale H. Brown. Princess Margaret Hospital, University Health Network, Toronto, Ontario, Canada.

Current approaches to the management of early stage glottic carcinoma include radiation as one of the primary modalities of therapy. While local control rates are high, when radiation fails, the options for partial laryngectomy and preservation of the voice are limited, as radiation injury to the local tissues compromises the usual reconstructive options. This paper described a new technique for reconstruction of the larynx using the free temporoparietal flap as a vascular carrier for grafts of mucosa and cartilage.

The paper reported experience with 18 consecutive patients undergoing laryngeal reconstruction for T1 and T2 glottic carcinomas, who failed radiotherapy. Outcome measures of vocal quality, and objective assessment of patients' perceptions of vocal disability, flap success, and time to decannulation were presented. All patients were decannulated with an average of 12 days to decannulation. Voice results varied from fair to excellent, depending on the extent of laryngeal resection. The objective voice index suggested that most patients felt their voices were improved from the pre-surgical state. There were no flap failures or partial flap losses in this series.

The technique of the free temporoparietal flap as a vascular carrier for reconstruction of the larynx following radiation failure appears to show promise in this pilot series of patients.

Tracheal Reconstruction Using Free Flaps. Hung-Chi Chen, David Chuang, and Yueh-bih Tang. Chang-Gung Memorial Hospital, and National Taiwan University Hospital, Taipei, Taiwan.

Tracheal reconstruction is challenging to surgeons. The difficulty is due to the unique location and structure. Following trauma or ablation of tumor, the defect can involve only the partial wall or can be totally circumferential; the latter can involve short or long segments. Reconstruction is impossible with a single autologous tissue. Various methods have been tried with modest success, such as resection and primary anastomosis, periosteal flaps, composite costal cartilage-perichondrium grafts, vascularized bone carrying a skin flap, forearm flaps, and others.

Three patients with tracheal defects have been reconstructed with free flaps, two with free forearm flaps and one with a jejunum flap. Structural stability was maintained with costal cartilage or periosteum followed by prolonged stenting. In one patient, the tracheal defect was due to trauma, and in the other two, to cancer.

One patient reconstructed with jejunum had recurrence of cancer at 3 months after surgery; however, the stent could be removed at 2 months, because the tracheal defect was located only in the anterior wall. The other two patients with forearm flaps had the stent removed at 6 months. One of these patients had a segmental defect of 3 cm, and the other had a partial circumferential defect.

The tracheal defect of partial circumference or short segment can be reconstructed with free jejunum or a forearm flap supported by cartilage or cortico-periosteum followed by long stenting. However, tracheotomy may need to be maintained in cases of longer segmental defects.

Innervated Anterolateral Thigh Flap for Tongue Reconstruction. Peirong Yu. M.D. Anderson Cancer Center, Houston, TX.

Glossectomy defects are commonly reconstructed with rectus abdominis myocutaneous flaps or radial forearm flaps, depending on the size of the defect. To minimize donor-site morbidity, 16 consecutive anterolateral thigh (ALT) free flaps were performed in a period of 7 months by a single surgeon for head and neck reconstruction: 11 for the tongue, three for temporal-orbital-maxillary, and two for pharyngoesophageal reconstruction. All flaps survived without partial necrosis. The average patient age was 63±11 years, and their body mass index was 25.5±4.5.

A perforator near the midpoint between the anterior superior iliac spine and superolateral patella (perforator B) was present in all cases; only two of them were septocutaneous. A more distal perforator (perforator C) was present in 11 casess, which was musculocutaneous in each case, while a more proximal perforator (perforator A) was present in eight cases, with five being septocutaneous. The average pedicle length to perforator A was 10.9±2.5 cm, while that to perforator B was 14±2.8 cm. The average thickness of the flap at the level of perforator B was 15.5±5.9 mm (range: 5 to 30 mm). In the 11 glossectomy cases, eight were total or subtotal, and three were hemi-glossectomy defects. Three patients also underwent segmental mandibulectomy and free fibular flap reconstruction. In each case, a portion of the flap was deepithelialized to reconstruct the floor of the mouth; the larynx or hyoid bone was re-suspended to the mandible. Extensive thinning of the flap was performed in an obese patient; flap thickness was reduced from 30 mm to 5 mm without compromise of the blood supply to the flap. All patients were decannulated.

Modified barium swallow studies were performed postoperatively in all patients to assess swallow function and leakage. A small contained leak was found in one patient, which healed spontaneously. Prolonged aspiration was present in two patients whose resections included the anterior mandible and extended to the epiglottis. Consequently, they were dependent on tube feeding. Excellent speech and deglutition occurred in the remaining patients. The lateral femoral cutaneous nerve was included for a sensate flap in six patients, with the lingual nerve as the recipient in each case. All patients regained sensation to a certain degree.

It was concluded that the ALT flap is well-suited for tongue reconstruction with minimal donor-site morbidity. Flap thinning may be necessary for hemi-glossectomy defects.

Dual Purpose for Single Free Flaps in Reconstruction of the Intraoral Cavity: Augmentation of the Neck with a Deepithelialized Segment of the Anterolateral Thigh Flap. Tewodros Gedebou and Fu-Chan Wei. Chang-Gung Memorial Hospital, Linkou, Taoyuan, Taiwan.

The purpose of this study was to evaluate the protective and aesthetic benefits of augmenting the neck with a deepithelialized portion of the anterolateral thigh (ALT) free flap designed to resurface intraoral defects.

In this prospective study, two groups of 30 patients each with intraoral carcinoma were selected and matched for age and extent of ablative surgery. Group 1 patients underwent augmentation of the neck with a deepithelialized portion of the ALT flap, in addition to restoring intraoral lining. Patients in Group 2 underwent reconstruction of the intraoral defect only. Early postoperative results and long-term appearance were evaluated and compared.

In Group 1, the average size of the ALT flap harvested was 7.5x21 cm, of which 47% was deepithelialized and used to cover skeletonized vessels of the neck. Postoperative recovery was unremarkable for 28 patients. Two re-explorations were undertaken, resulting overall in one partial flap failure and one (3.3%) orocutaneous fistula. In Group 2, the average ALT flap size was shorter; however, there were three re-explorations, resulting in one partial failure and four (13%) orocutaneous fistulas. In both groups, there were no carotid arterial blow-outs, and during an average follow-up period of 10 months, no cancer recurred. On final analysis, the aesthetic quality of the augmented necks of patients in Group 1 was superior to that in Group 2. There were no donor-site complications recorded, given that primary closure was possible in all.

The ablative effort in management of intraoral carcinoma presents a multi-faceted challenge to the reconstructive surgeon. In addition to restoration of intraoral lining, protection of neck vessels and aesthetic restoration of the neck should be considered. Prevention of potentially disastrous complications, such as orocutaenous fistulas or carotid artery blow-out, should be a prime consideration when selecting a reconstructive method. As one of the most versatile soft-tissue flaps, the ALT allows for provision of enough length with minimal expense. By providing a well-vascularized buttress to the suture line, this method reduced the rate of fistulas in the reported series. Furthermore, the sunken-in appearance that follows neck dissection and irradiation was not apparent in the augmented group.

When the extirpation of intraoral carcinoma involves neck dissection, a single ALT free flap can reconstruct both the intraoral defect, as well as augment the neck, which may reduce the rate of complications associated with neck dissection.

Bilobed Fibular Free Flaps for Reconstruction of Large Through-and-Through Defects in the Head and Neck. Daniel Alam, Elliot Abemayor, Thomas C. Calcaterra, Joel A. Sercarz, Jeffrey D. Rawnsley, and Keith E. Blackwell. UCLA Medical Center, Los Angeles, CA.

For extensive defects with loss of oral cavity mucosa, mandibular continuity, as well as skin, a number of reconstructive options have been used, including single flap reconstructions based on “mega” flaps like the scapular vessel system. While the fibular free flap provides excellent bone for mandibular reconstruction, concerns have been raised regarding the viability of the skin paddle and its reliability. Most authors argue against the harvesting of skin paddles large enough to allow for simultaneous management of intraoral as well as cutaneous defects. These authors reported a series of 18 patients with large, complicated, through-and-through defects of the oral cavity and facial skin, who were successfully reconstructed with a single fibular free flap utilizing a large skin paddle.

Deepithelialization of the center of the paddle and creation of a “bilobed” flap allowed for closure of mucosal and skin defects concurrently. Surgical defects in the study were all composed of full-thickness defects encompassing a mucosal, bone, and skin component. Five patients had isolated anterior bony defects with concurrent lip defects. The remainder of the patients had extension of soft tissue defects into the lateral cheek/buccal mucosal regions. Flaps were harvested with skin paddles ranging in size from 12 × 9 cm to 29×19 cm. The skin paddles were deepithelialized in their central region to allow for accommodation and folding required for appropriate insetting.

All patients were successfully reconstructed. Follow-up ranged from 1 to 41 months. No deaths occurred in the immediate perioperative period. There were no flap failures, and skin paddle survival in all patients. Seven of 18 patients were able to progress to oral diets. One major flap-related complication occurred with non-take of the split-thickness skin graft to the donor site, requiring regrafting. In 6/18 patients, minor flap complications occurred. Three patients had small areas of non-take of the split-thickness skin graft, which were treated conservatively with local debridement. The remaining three had minimal skin loss (<10%). Three patients underwent secondary lip commisuroplasties, and one had minor soft tissue debulking.

These authors believe that, in contrast to previously reported cases in the literature, the fibula osteomyocutaneous free flap can provide a safe and reliable large skin paddle, allowing bilobed reconstruction of complicated through-and-through head and neck defects.

Mandibular Reconstruction: Two Flaps Better Than One? Essam M. Gabr, Mark R. Kobayashi, Arthur H. Salibian, William B. Armstrong, Michael Sundine, Jay W. Calvert, and Gregory R. Evans. University of California, Irvine Medical Center, Orange, CA.

In this study, the authors compared the combined iliac and ulnar forearm free flaps with myocutaneous fibular free flaps for mandibular reconstruction.

A retrospective study was conducted of 40 patients who had oromandibular reconstruction, of whom 23 patients had a combined iliac crest without skin and ulnar forearm free flap. Seventeen patients had a myocutaneous free fibula flap. There were 10 females and 30 males, with a mean age of 57.47 years in the study population. Ninety percent of the cases were squamous cell carcinoma (55% T4), of which 11% were recurrent tumors. Anterolateral mandibular defects constituted 52.9% of the fibular reconstructions, and 60.9% for the iliac/ulnar reconstructions. The mean bone gaps were 8.79 cm and 8.95 cm, respectively. Functional evaluation was based on the University of Washington Questionnaire (UWQ) through phone calls and personal communication with patients.

The mean hospital stay was 15.43 and 10.09 days for the fibulas and iliac/ulnar flaps, respectively. The facial artery (64.7%) and facial vein (60%) were the main recipient vessels for the fibular reconstructions, while the external carotid (95.6%) and internal jugular (66.7%) were the main recipient vessels for the iliac/ulnar. The overall flap survival was 96.8% (100% of fibulas and 95.65% of iliac/ulnars). Two flaps were completely lost in the iliac series because of unsalvageable venous thrombosis. Local complications for the iliacs were 30.4%, while they were 5.9% for fibular reconstructions. Functions such as speech, swallowing, and chewing were significantly better in the fibulas than the iliac/ulnars. Cosmetic acceptance of 77.8% of the fibulas was excellent and good, while 71.4% of the iliac/ulnars was rated good.

It appeared that within this study population, the free myocutaneous fibular flap had fewer local complications and higher flap survival than did the combined iliac and ulnar forearm flaps. Overall functional outcome also was improved. The use of the double flap may be appropriate in massive oromandibular defects, but may be less appropriate in the more modest functional reconstruction of mandibular defects.

2-Octylcynanoacrylate-Assisted Microvascular Anastomosis in a Rat Model: Histologic and Biomechanical Features. Erik Ang, Kok-Chai Tan, Robert Ng, I.C. Song, and Irene Kee. Singapore General Hospital, Singapore.

The aim of this study was to establish the safety and superiority of 2-Octylcyanoacrylate-assisted microvascular anastomosis over suture-only anastomosis in the laboratory rat model.

One hundred forty male Sprague-Dawley rats were used. They were sacrificed at 1 week, 1 month, 3 months, and 6 months, and the anastomoses were subjected to evaluation. The femoral arteries were then tested with a micro-caliper instrument for stiffness and breaking force. Longitudinal and transverse sections were sent for histologic evaluation.

There was no significant difference in maximum break force or stiffness at 1- week, 1-month, 3-month, and 6-month intervals for the glue vs. suture. At 1 week and 1 month, suture had a higher maximum break force, but at 3 and 6 months, glue had a higher maximum break force. Glue vs. suture were both stiffer than control (but not statistically significant) at 1 week and became significantly less stiff than controls at 3 months. There was acceptable inflammatory reaction and minimum distortion of the intimal layer in the glued vessels.

This study showed that the maximum break force and stiffness of the glue-assisted microvascular anastomoses were similar to those of the suture-only anastomoses up to 3 months, and superior after 3 months. It demonstrated comparable histologic appearance, and an absence of tissue toxicity or adverse reaction to the glue.

The biomechanical properties of 2-Octylcyanoacrylate-assisted anastomoses are superior to those achieved by suture-only anastomoses, approaching normal physiologic parameters. In addition, the glue reduces the need to insert more sutures, thereby reducing the risk of vessel narrowing, foreign body reaction, and intravascular thrombosis that may occur with standard suture technique, leading ultimately to shorter times required for flap preparation, shorter digital ischemic time, and reduced intravascular thrombosis, and resulting in improved flap or digital survival.

Based on the histologic, biomechanical, and visual observations in this rat model, the authors intend to develop a new technique of microvascular anastomosis that is clinically relevant, and one that can result in decreased ischemic time and better flap or digit survival. They also intend to perform further studies on the anastomosis of structures other than blood vessels, like intestine and bile duct.