Abstracts

Fourteenth Annual Meeting, Fajardo, Puerto Rico

Jnauary 15-16, 2005

Outstanding Nerve Paper Session Restoration of Elbow Flexion after Avulsion Injuries of C5 and C6 Nerve Roots by Nerve Transfer. N. Zagoralos, Z. Kokkalis, George Sotirios Themistocleous, G.D. Chloros, G. Papadopoulis, and D.G. Efstathopoulos. KAT Hospital, Athens, Greece. Restoration of elbow flexion caused by brachial plexus injuries is extremely important. Bridging with neural grafts after avulsion of C5 and C6 nerve roots is not always feasible. Alternative options include tendon or nerve transfers or a combination of both. However, tendon transfer methods yield poor results and nerve transfers (principally intercostals and accessory nerves) do not always guarantee acceptable outcomes. Oberlin et al. (1994) reported very encouraging results in solving this problem through nerve transfer of ulnar nerve fascicles to the fascicle for the biceps brachii of the musculocutaneous nerve. From 1995 to 2003, 16 patients (14 males and two females, aged from 16 to 35 years, with a mean age of 24 years), presenting with an avulsion injury of C5 and C6 nerve roots, were operated on using the method described above. The mean postoperative follow-up was 2.5 years (range: 1 to 6 years). Following a short period of immobilization, a rehabilitation program was initiated, including passive mobilization and electrical stimulation. At 4 months, muscular contractions were noted in all patients; at 6 months, there was weak elbow flexion, which improved significantly at 1 year postoperatively. This method is simple, reliable, and effective, because the motor fascicle of the musculocutaneous nerve is thin and does not require an important sacrifice of a large ulnar nerve part, thus minimizing motor and sensory compromise. The ulnar nerve is situated near the branch to the biceps tendon, allowing end-to-end suturing. Biceps brachii re-innervation is more rapid than with any other method. (American Association for Hand Surgery) Operative versus Nonoperative Treatment of Global Brachial Plexus Birth Injuries. Patricia Di Taranto, John A.I. Grossman, Lorna E. Ramos, and Andrew E. Price. Miami Children's Hospital, Miami, Florida, USA. There are few data available on the outcome of surgical treatment for global brachial plexus birth injuries. This report evaluated the outcome following surgical treatment in comparison to a similar non-operated group. Over a 7-year period (1994-2000), 91 infants born at a single institution (38,589 live births) sustained a brachial plexus birth injury. All were followed by one of the authors. Twelve of these infants presented with a flail, insensate limb at birth and evidence of Homer's syndrome. These findings persisted past 6 months of age. At a second institution, over a 4-year period (1997-2000), 42 consecutive infants with similar clinical findings at birth and at 6 months of age, underwent surgical reconstruction of their brachial plexus by a single surgeon; 36 were available for follow-up. Both groups had a minimum follow-up of 2.5 years. Shoulder and hand function were evaluated with the Gilbert-Raimondi system. No child in the surgical group underwent a second operative procedure during the follow-up period. In the unoperated group (n = 12), shoulder function at final follow-up was rated as 100% poor, as compared to the surgical group (n = 36), in which shoulder function was 0% poor, 22.2% good, and 5.5% excellent. Final hand function in the unoperated group was 25% poor and 75% fair, and in the operated group, 19.4% fair, 58.3% satisfactory, 16.6% good, and 5.5% excellent. Children with global brachial plexus birth palsies showed an obviously better outcome with surgical treatment. The surgical strategy of nerve repair and transfer must focus on the recovery of shoulder stability and hand function. (American Association for Hand Surgery) One-Stage Reanimation of the Paralyzed Face Using the Rectus Abdominis Neurovascular Free Flap. Ali Sajadian, Angela Y. Song, Jessie Van Swearingen, Christopher Khorsandi, Todd C. Henkelmann, and Ernest K. Manders. University of Pittsburgh, Pittsburgh, Pennsylvania, USA. Surgical reanimation of long-standing facial paralysis using free muscle transfer is frequently hampered by long rehabilitation and outcome inconsistency. A one-step reanimation using a free rectus abdominis neurovascular flap has the potential for better function and shortened recovery time, compared to traditional two-stage reconstructions. The authors' aims were to 1) demonstrate a useful technique in free flap facial reanimation; 2) impart clear functional improvement in facial movement; and 3) illustrate objective, quantifiable improvement in muscle activity. Eight patients with long-standing unilateral facial paralysis underwent one-stage facial reanimation. Briefly, the deep inferior epigastric artery and vein were identified, and a 6- to 10-cm pedicle obtained. A 4- x 5-cm segment of muscle, the inferior epigastric vascular bundle, and a 12- to 15-cm length of the contributing intercostal nerve were harvested. The recipient buccal branch of the facial nerve was identified on the nonparalyzed side, and nerve anastomosis was performed. The recipient facial artery and vein were identified on the paralyzed side, and vascular anastomoses were performed. Patients underwent preoperative EMG testing, and were followed postoperatively with serial EMGs. Peak muscle activity was recorded for zygomaticus and levator muscle groups during voluntary movement. At each visit, the Facial Grading Scale (FGS) was used to assess impairment at rest, voluntary movement, and synkinesis. Of the eight patients, preoperative EMG data were available for five, and they comprised the study group. The baseline FGS score for the patients was an average of 16.2 (range: 6-29). At most recent follow-up all five patients in the series showed an improvement, with mean improvement of 186% and a final FGS of 41 (range: 26-55). The average initial EMG reading for the levators was 8.6 muV (range: 0-30). All five patients had increased EMG potential at their final postoperative visit, with an average of 27.6 muV (range: 13-50), and a mean increase of 219% (range: 67-430%). The authors demonstrated that the rectus abdominis neurovascular free flap is a safe and reliable alternative in the treatment of long-standing facial paralysis. The resulting EMG and FGS scores confirmed that the one-stage reanimation procedure can afford reliable clinical improvement. (American Society for Reconstructive Microsurgery) Management of Brachial Plexus Injuries Assisted by Intraoperative Nerve Recording. Mirsad Mujadzic, Tuna Ozyurekoglu, Amit Gupta, and Warren Breidenbach. Christine Kleinert Institute for Hand and Microsurgery, Louisville, Kentucky, USA. Accurate evaluation of the type, level, and extent of lesions involved in brachial plexus injury is crucial. It is known that imaging techniques showing anatomic continuity of roots may be misleading. Intraoperative nerve recording (IONR) has proved to evaluate functional continuity of roots with the central nervous system (CNS). The purpose of this study was to present results in management of brachial plexus injuries assisted by IONR. The authors analyzed 23 patients with brachial plexus injuries. With the exception of two patients with gunshot injury, all other injuries were caused by traction. The mean patient age was 36 years and the mean follow-up was 38 months. In the study, nerve action potentials (NAP) were used for localizing lesions distal to the dorsal root ganglion. The somatosensory evoked potential (SSEP) was recorded to clarify the functional continuity of the dorsal root sensory fibers with the spinal cord. Twenty-three roots were grafted and neurolysis was done on 15 roots. Neurotization with the intercostal nerves was used in three cases, C7 hemicontralateral root in four cases, spinal accessory nerve in seven cases, and phrenic nerve in two cases. In eight cases, muscle and tendon transfers were used for failed grafting and neurotization and to improve final functional results. From 85 explored and examined roots with IONR, 37 roots did not show continuity with the brain. Forty-eight roots were in continuity and amenable to repair. From 23 grafted roots, 14 (61%) grafted nerves showed clinically useful recovery. In 15 neurolysed roots, 10 (66%) showed useful recovery. With neurotization with other proximal nerves, clinically useful recovery was obtained in 8/16 cases (50%). In four patients having functional free gracilis muscle, two cases achieved clinically useful function. Muscle and tendon transfers improved the final results. 1) IONR was a helpful aid in the decision-making process in the management of brachial plexus injuries, by identifying the roots without a connection to CNS, and by separating the neuromas that should be resected and grafted from those which may be neurolysed. 2) An integrated concept is necessary for the treatment of traction lesions in brachial plexus injuries. In this group of patients, nerve repair is still possible, but the surgeon should be ready to combine other treatment options such as neurolysis, neurotization, free muscle transfers, and tendon transfers, in order to achieve better final functional results. (American Society for Reconstructive Microsurgery) Final Results of Grafting versus Neurolysis in Obstetrical Brachial Plexus Palsy. Ann Schwentker, Christine G. Curtis, and Howard M. Clarke. Hospital for Sick Children and University of Toronto, Toronto, Ontario, Canada. Neurolysis of conducting neuromata-in-continuity in obstetrical brachial plexus palsy is still undertaken in some centers on the premise that it is inappropriate to divide any axons connected to distal targets. This may limit the options available for reconstruction. The objective of this study was to examine the final outcome of interposition nerve grafting versus neurolysis alone in patients with obstetrical brachial plexus palsy. A retrospective analysis of prospectively collected data was undertaken. Patients undergoing primary operative reconstruction of obstetrical brachial plexus palsy in the authors' institution from 1988 to 1998 were eligible for inclusion, if they had been followed for at least 4 years from the date of surgery (range: 44-76 months). Available patients were classified as neurolysis only (16) or resection and grafting (92) on the basis of the operative records. Further stratification was made of Erb's palsy or total palsy. The active movement grading scale (AMS) was used to measure 15 joint movements preoperatively and at final follow-up. Comparison of AMS scores before and after intervention was undertaken, using the Wilcoxon two-sample. Fisher's exact test was used to compare the proportion of patients achieving a score of Grades 6 or 7 on AMS. The positive results of a previous study of neurolysis in Erb's palsy patients were not sustained at 4-year follow-up, in which significant improvements were seen in AMS for only supination, and where pronation decreased. In neurolysis for total palsy, AMS scores were improved across many joint movements, but only occasionally to Grades 6 or 7 level. A second analysis looked at the proportion of patients achieving a grade of 6 or 7 at final follow-up, to examine outcomes deemed functionally important. Neurolysis in Erb's palsy decreased the proportion achieving good function for pronation. Grafting in Erb's palsy produced significantly more useful function for shoulder function, elbow flexion, supination, and wrist and thumb extension. Neurolysis in total palsy showed significant increases in function for shoulder adduction and supination only. Grafting in total palsy, in contrast, produced significantly increased proportions of patients with useful function for shoulder function, elbow flexion, supination, wrist flexion, and extension of the wrist, finger, and thumb. The improvement in limb function produced by neurolysis in Erb's palsy was not sustained over time. Grafting for both Erb's palsy and total palsy produced significant improvements in AMS scores and significant improvements in the proportion of patients achieving functionally useful scores. (American Society for the Peripheral Nerve) Serial Imaging of Regeneration after Common Nerve Injury Paradigms. Terence M. Myckatyn, Susan E. Mackinnon, Daniel A. Hunter, Danielle Brakefield, and Alexander Parsadanian. Washington University, Plastic Surgery, St. Louis, Missouri, USA. Recent advances in molecular neurobiology include the development of transgenic mice that express genes encoding fluorescent proteins under neuron-specific promoters (XFP mice). These animals have been used in the field of developmental neurobiology, but their use has expanded to include the study of peripheral nerve axonal regeneration subsequent to crush or unrepaired transection injuries. These authors have developed a transgenic mouse that differs from previously reported and commercially available mice, in that enhanced yellow fluorescemt protein expression (EYFP) is driven by the human thy1 promoter (hThy1). Motor and sensory peripheral nerves in these mice appear as a bright yellow-green under fluorescent microscopy and can be tracked after nerve injury. In this study, nerve regeneration was serially tracked in the same living animals every 5 days for up to 25 days after initial injury. Serial nerve imaging was utilized to study regeneration after the clinically relevant nerve injury paradigms of crush, tibial nerve transection repaired with primary neurorrhaphy or graft, and end-to-side neurorrhaphy. In addition, rhodaminated bungarotoxin was utilized to label acetylcholine receptors in the living mice, to study denervation and reinnervation of the gastrocnemius muscle over time. A z-axis driver assisted with imaging of the three-dimensional surface of the gastrocnemius in live animals. A stacked series of images taken 0.2 to 5 um apart took into account depth of field, pulsations from the adjacent arteries, breathing, and other movement artifacts. A computer algorithm was used to align, and then focus, hundreds of stacked images into a single two-dimensional image of neuromuscular junctions. Live animal serial nerve imaging was compared with wet-mount fluorescent microscopy and histomorphometry in the same nerve specimens. The use of transgenic mice that strongly express EYFP in their peripheral neurons, coupled with serial nerve imaging, provided an important methodology for studying the heterogeneous nature of axonal elongation and muscle reinnervation following peripheral nerve injuries. (American Society for the Peripheral Nerve) Scientific Papers Neuronal Differentiation of Human Adipose Tissue-Derived Stromal Cells and Maintenace of Long-Term Neuronal Cultures. Eul Sik Yoon, Sanjay Dhar, and Gregory Evans. University of California, Irvine, Orange, California, USA. Human adipose tissue-derived stromal cells (hADSCs) can be greatly expanded in vitro, and induced to differentiate into multiple mesenchymal cell types, including osteogenic, chondrogenic, myogenic, and adipogenic cells. Recent studies reporting differentiation to early neural progenitors of hADSC cells have aroused interest among investigators for regenerative medicine. To date, differentiation to non-mesodermal fates of hADSC cells is still largely unknown. The aim of this reported study was to investigate the possibility of inducing hADSCs to differentiate into neuron-like cells, and the extension of life span of these differentiated hADSC cells under optimal differentiation conditions. Human adipose tissue was digested with collagenase, and adherent stromal cells were cultured. After primary culture in control medium and expansion in two to three passages, the cells were incubated in a new long-term neuronal induction medium (DE-1 medium). Neuronal differentiation was identified by immunocytochemistry, western blot, and semi-quantitative RT-PCR applied to detect mRNA expression of neurofilament 1 (NF1), nestin, and neuron-specific enolase (NSE). The optimal differentiation protocol induced the hADSC cells to express nestin and NeuN characteristic of early neuronal precursor stem cells, and they exhibited neuronal specific markers, neuron-specific enolase (NSE) and neurofilament (NF). Neuronal morphologic characteristics were recognized in about 50 to 60% of the cell populations maintained over 8 weeks, and 60 to 80% of the differentiated cells expressed neuronal specific markers. The data supported the hypothesis that adipose tissue contains stem cells capable of differentiating into neurons. Compared with cells harvested by bone-marrow aspiration, hADSC cells are easier to obtain, have lower donor-site morbidity, and are available in large numbers, which eliminate the need for costly and lengthy tissue-culture expansion. These hADSC cells can overcome their mesenchymal commitment, and may represent an alternative autologous stem-cell source for CNS cell transplantation. The authors reported a new medium which can hold cells in a differentiated state for more than 8 weeks. Motor Neuron Progenitor Cell Transplants Prevent Muscle Atrophy after Peripheral Nerve Injury. Melody N. Craff, Timothy Shane Johnson, Milan Prakash Ranka, Jose L. Zeballos, Robert Howard, Pejman M. Motarjem, Mark A. Randolph, and Jonathan M. Winograd. Massachusetts General Hospital, Boston, Massachusetts, USA. A major obstacle to functional recovery following peripheral nerve injury is the time-dependent loss of the ability of muscle fibers to successfully receive reinnervation and regain function. This study investigated the potential of motor neuron progenitor (MNP) cell transplants, derived from embryonic stem (ES) cells, to preserve muscle mass and architecture after denervation in an in vivo adult mammalian model of peripheral nerve injury. ES cells were differentiated using retinoic acid and sonic hedgehog peptide to induce cholinergic MNP cells. Fluorescent-labeled MNP cells were injected into the gastrocnemius muscle of Sprague-Dawley rats following denervation by ipsilateral sciatic nerve transection. Control rats were subjected to identical unilateral sciatic nerve injuries, but received injection of a phosphate buffered saline (PBS) carrier solution only. Gastrocnemius muscles were weighed and analyzed at 7 days using histology and immunostaining for MNP markers. Seven days after MNP transplant, little or no atrophy of the denervated muscle occurred, as evidenced by 100% conservation of muscle mass and 99% preservation of myocyte cross-sectional area on histology relative to the uninjured normal contralateral muscle. In contrast, control muscles had atrophied significantly after 7 days, with muscle mass reduced to 70% of normal and cross-sectional area reduced to 72% of normal. Fluorescence microscopy revealed the extension of long branching axons from the labeled MNP cells across the surface of the muscle and penetrating into the muscle tissue. Immunohistochemistry of labeled MNP cells stained positively for cholinergic and motor neuron specific markers, and demonstrated overlap between presynaptic terminals and alpha bungarotoxin-labeled acetylcholine receptors on the denervated muscle fibers. No neurons could be seen in the control muscles by fluorescence microscopy and immunostaining. However, 21 days after MNP transplant, muscle atrophy in the experimental muscles was in line with control muscles, and only a few MNP cells were detectable at this point. This study demonstrated that transplanted neuronal precursors derived from embryonic stem cells could prevent muscle atrophy after denervation. MNP transplants appear to form new neuromuscular junctions with denervated muscle fibers. Currently, the survival of the transplants is limited in duration and effect. If prolonged, this could represent a promising therapy for muscle preservation following denervation injury. Molecular Imaging and Adult Stem Cell Therapy for Applications in Neuronal Regeneration. Michael Charles Edwards, Juri Gelovani, and Saleh Shenaq. Baylor College of Medicine and M.D. Anderson Cancer Institute, Houston, Texas, USA. Non-invasive whole body imaging would significantly aid in the development and clinical implementation of progenitor cell-based therapies, tissue engineering, and biomimetic materials chemistry, for applications in brachial plexus and peripheral nerve injury repair and reconstruction, by providing the means for noninvasive monitoring of the fate of transplanted or endogenous neural progenitor cells over a long time period. Circumventing the major limitations of in vitro radiolabeling and the impracticality of using fluorescent imaging in humans, these authors genetically labeled neural progenitor cells with different PET-reporter genes and repetitively imaged their distribution, migration, differentiation, and persistence in the context of progenitor cell-based therapies for nerve root avulsion and traction injuries. The choice of the right promoter to drive the expression of a reporter gene is important in terms of long-term monitoring and imaging the fate of infused progenitor cells. To start, these authors have tested the ubiquitous expressing CMV promoter, using a self-inactivation (SIN) HIV-1 lentiviral (LV). They chose the SINLV vector because it allows for the construction of more stringent tissue-specific or regulatable vectors in later studies. By using SINLVs with a specific promoter, they expected to achieve targeted transgene expression in a specific lineage of transduced stem/progenitor cell-derived tissues. The progenitor cells were genetically labeled with a constitutively expressed CMV reporter gene driving the expression of the non-immunogenic human thymidine kinase type two gene (hTK2) fused to green fluorescent protein (GFP), and imaged with [11C]FEAU or [18F]FEAU - two novel hTK2-specific radiotracers. For neural specific expression, on the same lentiviral vector, they constructed neural specific enolase (NSE) driving expression of HSV1-tk-RFP (red fluorescent protein). In the studies, progenitor cells could be genetically labeled with a constitutively expressed CMV promoter driving the non-immunogenic human thymidine kinase type two gene (hTK2), and imaged with [11C]FEAU or [18F]FEAU - two novel hTK2-specific radiotracers. Progenitor cells guided to their neural fate can be simultaneously detected and imaged using the neural specific sensor (NSE-HSV1-tk-RFP). Non-invasive whole body imaging will significantly aid in the development and clinical implementation of stem cell-based therapies for nerve regeneration by providing the means for noninvasive monitoring of the fate of transplanted or endogenous progenitor cells. These initial studies demonstrated the feasibility of this approach using a novel lentiviral-based molecular imaging strategy. Effect of FK506 Preload Treatment on Nerve Regeneration. Alison K. Snyder, Ida K. Fox, Chris M. Nichols, Daniel A. Hunter, Susan R. Rickman, and Susan E. Mackinnon. Washington University School of Medicine, St. Louis, Missouri, USA. FK506 enhances nerve regeneration when administered at the time of nerve injury. However, these effects are diminished when FK506 administration or administration and nerve repair are delayed. The purpose of this study was to determine if administering FK506 prior to nerve injury improves nerve regeneration in a rat tibial nerve transection model. Male Lewis rats were randomized into three equal groups: placebo-treated control, FK506 therapy beginning immediately at the time of surgery, and FK506 preload therapy beginning 3 days prior to surgery. All animals underwent tibial nerve transection with repair and completed serial walking tracks for functional outcomes assessment. Nerve tissue was collected at postoperative day 21 for histologic and histomorphometric analysis. Preliminary analysis of the histomorphometric results showed superior nerve regeneration in the FK506 preload group. In particular, total nerve fiber counts were significantly increased in the FK506 preload group (1848 ± 490), compared to both the immediate FK506 (1288 ± 630) and placebo-treated (722 ± 449) groups. The FK506 preload group also demonstrated the lowest percentage of neural debris. This study showed that the timing of FK506 administration alters its subsequent neuroregenerative effect: improved peripheral nerve recovery can be achieved by preloading with FK506. Clinically, this might be applicable to patients undergoing iatrogenic peripheral nerve injury for tumor ablation surgery such as prostatectomy. Molecular Switch-off for Nerve Growth Factor Production In Vitro. Sanjay Dhar, Michael P. McConnell, and Gregory R.D. Evans. University of California, Irvine, Orange, California, USA. The authors' previous studies have established that Human Embryonic Kidney (EcR-293) cells can be genetically engineered to release NGF in vitro up to 9 days following induction with Ponasterone A (PonA). To further tightly regulate the NGF release, they incorporated herpes simplex virus thymidine kinase (HSV-TK) gene as a molecular "off switch" in these cells to tightly regulate the NGF release. The present study was performed to assess the NGF expression after suicide gene therapy. HSV-TK gene was closed in expression cassette of plasmid vector pcDNA6 (pcDNA-TK). NGF producing human embryonic kidney cells (hNGF-EcR-293) were transfected with pcDNA6-TK to produce a stable cell line named as hNGF-EcR-293-TK. These cells induced with PonA( ± ) and PonA(-) were left for 5 days without changing the medium. On day 5, gancyclovir (GCV) was added to the experimental wells. Supernatants were collected at 1, 2, 3, and 4 days post gancyclovir treatment and tested for NGF expression and bioactivity by ELISA and PC-12 cell assay, respectively. An hNGF-EcR-293-TK stable cell line was established by transfecting HSV-TK in NGF-producing EcR-293 cells, followed by cloning and expanding a single cell clone, which was used for molecular switch-off experiments. RT-PCR validated the incorporation of HSV-TK in the TK ± cells. There was an increase in NGF production until day 5 post induction with PonA. Standardization with GCV dose indicated a 10uM concentration was effective to kill all the transfected cells. Treatment with GCV at day 5 post induction resulted in the progressive decline in NGF production, reduced PC-12 bioactivity, and cell killing. Non-transfected hNGF-EcR-293 and non-induced hNGF-EcR-293-TK control cells neither had NGF expression nor exhibited any PC-12 bioactivity,, and did not exhibit any cell killing to GCV treatment. Further experiments are in progress. The current study successfully established a cell line which is capable of both producing and switching off NGF production whenever needed, enabling a tighter and more regulated control over NGF expression. Direct Muscular Neurotization of Denervated Skeletal Muscle by Segmental Epineurectomy and Burying: Side Muscular Neurotization. Ilker Yazici, Suhan Ayhan, Cigdem Elmas, Cagri Timucin, and Kenan Atabay. Gazi University Medical Faculty, Plastic Surgery, and Hacettepe University Medical Faculty, Ankara, Turkey. Trauma that destroys the hilus of a striated muscle or directed to the point of nerve entrance, eventually leads to denervation atrophy. As a distal nerve stump is not available for these types of injuries, traditional nerve reconstruction techniques are not possible. The only option available for reinnervation of these muscles is direct muscular neurotization. In this study, the authors' goal was to develop a new muscular neurotization technique, in which transection of the donor nerve is avoided. They investigated the outcome of insertion of a donor nerve segment within the denervated muscle (side neurotization), following segmentary epineurectomy. Thirty-five male Wistar rats were used in the study. Denervation of the right gastrocnemius muscles was performed for both bellies, and the tibial nerve used for neurotization in all groups. Neurotization was established by dividing the tibial nerve and direct implantation for the direct neurotizatiion group, and by segmentary epineurectomy and burying for the side neurotization group. Five groups were evaluated, including: sham controls (n = 5), denervation controls (n = 5), denervation and segmentary epineurectomy with no burying (n = 5), direct muscular neurotization (n = 5), and side neurotization (n = 5). Assessment was performed by electromyography, muscle weight measurement, and histologic evaluation at the postoperative second and third months. The denervation group was successful for denervation, compared to the sham control group for electromyographic (p = 0.014), weight (p = 0.012), and histologic (p = 0.008) parameters. The side neurotization group was successful in diminishing muscle atrophy and gaining reinnervation in electromyographic (p = 0.028), histologic (p = 0.008), and weight (p = 0.047) parameters, compared to the denervation group. The direct neurotization group was also found successful for histologic (p = 0.008) and weight (p = 0.028) parameters. No statistically significant difference was found between the epineurectomy and denervation groups for all parameters (p > 0.05), indicating that epineurectomy alone was not enough for side neurotization. The technique is promising for not interfering with fascicular integrity and is as successful in reinnervating as in the direct neurotization group. Since there is decreased donor nerve morbidity, it seems to be a good alternative for direct muscular neurotization. The technique may also be helpful for neurotization of free muscle transfers. Further studies are required to evaluate the force transmission of side neurotized muscles, to completely evaluate the efficacy of the technique. Effect of Injurious Exercise on Neurovascularized Muscle Transfers. Melanie G. Urbanchek, Mustafa Asim Aydin, Jack H. Van der Meulen, and William M. Kuzon. University of Michigan, Ann Arbor, Michigan, USA. Skeletal muscle contracts with lower maximal isometric force following neurovascular transfer. It is unknown whether transferred muscle is more fragile and susceptible to injury. The authors compared mechanical properties of transferred muscle both at 3 days and at 2 months after an in situ injury protocol. Their hypothesis stated that during a standardized lengthening contraction protocol, transferred neurovascularized muscle is more susceptible to injury which is observed in a larger relative:decrease in maximal whole muscle force, decrease in single muscle fiber force, and increase in denervated muscle fiber percentage post injury. The rat extensor digitorum longus (EDL) muscle was orthotopically transferred in the adult rat without vascular disruption. Proximal and distal tendons were transected and repaired in all groups. The peroneal nerve was either sham operated (CONTROL) or divided and repaired with all (TOTAL) or approximately 30% (REDUCED) of proximal axons included in the repair. After a 4-month recovery, whole EDL muscle force was measured in situ. The no-injury control group ended the protocol here, while the injury groups completed a muscle-lengthening contraction exercise. The injured muscle groups were either assessed again at 3 days or 60 days post injury. The EDL muscles were removed and portions were prepared for single muscle fiber force testing or histologic assessment of fiber denervation. The whole EDL muscle force was lower for both the TOTAL and REDUCED groups, compared to the CONTROL group and for the REDUCED group, compared to the TOTAL group. Both main effect variables (muscle transfer and injury state) were significant for the dependent variable absolute values of whole muscle force, single fiber muscle force, and percent of denervated fibers. The injury protocol had less relative negative effect on the transferred than on the CONTROL muscles at 3 days post injury. At 60 days post injury, the muscle transfers improved in whole muscle force, single fiber force, and percent fiber denervation relative to the no-injury control groups. The hypothesis was rejected. The authors found that transferred neurovascularized muscle is less susceptible than CONTROL muscle to injury following a standardized muscle-lengthening contraction protocol. The lengthening contraction exercise proved to have beneficial effects for neurovascular transfer muscle. Delayed Reinnervation of Denervated and Inappropriately Protected Peripheral Nerve Stumps. Raj Midha, Qing-Gui Xu, Caatherine A. Munro, Jason Belkas, and Tessa Gordon. University of Calgary, Calgary, Alberta, Canada. Delayed repair of peripheral nerve injuries often results in poor motor functional recovery, partially due to deconditioning of endoneurial pathways in the distal nerve before motor axons can regenerate. The authors' previous studies have shown that a 2-month protection of a nerve stump with a motor nerve improves reinnervation, compared to sensory protection or chronic denervation. In this study, the temporal sequence of reinnervation was examined in motor, sensory, and non-protected nerve stumps. Using the rat femoral nerve, they protected distal endoneurial pathways of the saphenous nerve with either cross suture of the quadriceps motor nerve (Group A) or re-suture of the saphenous nerve (Group B), to compare later motor regeneration into the "protected" saphenous nerve pathway to chronic denervation and "unprotected" saphenous nerve (Group C). After this protection (or lack thereof) for 8 weeks, the femoral nerve's motor branch was re-cut and sutured to the distal saphenous nerve to allow motor reinnervation. The quantitative assessment of reinnervation was carried out after 3 (n = 14), 6 (n = 15), and 9 (n = 14) weeks, 20 mm distal to the repair site, by nerve sampling for axon counts and retrogradely labeled motoneuron counts. Significantly more myelinated axons reinnervated the motor (A) than sensory (B) and no-protection (C) groups at 6 weeks. Axon numbers declined in all groups, but at 9 weeks, there were still significantly more myelinated axons innervating the distal endoneurial pathway in Group A than in Group C. At 3 and 6 weeks, there were approximately 50% more retrogradely labeled femoral motoneurons in Group A than in the B and C groups. Neuron counts in Group A remained stable despite considerable pruning of axons. Counts in Groups B and C improved over time and, by 9 weeks, were comparable in all three groups. The authors concluded that initial motor innervation biases distal endoneurial pathways to favor motor axonal regeneration potential, compared to inappropriately innervated or chronically denervated nerve stumps. Reinnervation of inappropriately innervated or denervated nerves is considerably delayed. Photochemical Tissue Bonding: A Novel Nerve Repair Technique. Jose Luis Zeballos, Timothy Shane Johnson, Pejman M. Motarjem, Mark A. Randolph, Irina E. Kochevar, Robert W. Redmond, and Jonathan M. Winograd. Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA. Photochemical tissue bonding (PTB) is a novel tissue repair technique that utilizes photoreactive dyes and visible light to produce non-thermal covalent bonding between two tissue surfaces. Bonding occurs immediately and forms a watertight seal. PTB is ideally suited for the repair of delicate tissues in which both an immediate bond and minimization of suture material would be beneficial. In this study, PTB was investigated as a technique for the microsurgical repair of peripheral nerves. Forty Sprague-Dawley rats underwent transection of their left sciatic nerve, followed by randomization into one of four repair groups (n = 10): primary repair with 10-0 suture, repair with PTB using an overlapped epineurial cuff, repair with an overlapped epineurial cuff alone, or no repair. Postoperatively, animals were monitored for functional recovery via walking track analysis. At 90 days, bilateral gastrocnemius muscles were harvested, weighed, and fixed for histology. Sciatic nerve tissue was harvested and analyzed with histomorphometry. Walking track analysis at 90 days postoperatively revealed mean sciatic function indices of -70.6 for primary repair, -76.9 for PTB cuff repair, and -80.7 for cuff repair alone. There was also evidence of retained gastrocnemius muscle mass in all three repair groups, with mean ratios of 0.586 for primary repair, 0.473 for PTB cuff repair, and 0.382 for cuff repair alone. Histologic evaluation of neural tissue revealed the presence of myelinated axons distal to the site of repair in each repair type; however, the PTB with cuff technique demonstrated the least amount of scarring at the repair site itself. Histomorphometric analysis indicated that the fiber diameter was greater in the PTB with cuff repair, followed by suture repair and, last, cuff alone. Additionally, myelin thickness and the g-value in the PTB with cuff and the primary repair were comparable in both proximal and distal nerve segments. This study demonstrated that PTB using an overlapped epineurial cuff technique can be used as a microsurgical technique for the repair of peripheral nerves and approaches the results seen with primary microsurgical repair. Further refinement of this technique is necessary and may provide the microsurgeon with a valuable tool for the repair of peripheral nerves. A Novel Rodent Model for Long Nerve Gap Reconstruction. Jonathan Cheng, J.G. Yan, L.L. Zhang, D.A. Hunter, and H.S. Matloub. Medical College of Wisconsin, Milwaukee, Wisconsin, USA. To date, the ubiquitous rodent model has had limited utility for study of long nerve gaps due to the small size of the animal. The entire length of the sciatic nerve typically measures 3.5 cm in these small animals. Prior attempts to develop a long gap rodent model have required more complex repair and distal dissection. The authors sought to overcome these limitations by developing a simple, effective rodent model for reconstruction of long nerve gap defects. This would allow researchers to capitalize on the well-known rodent advantages of hardy constitution, easy care, workable scale, simple anesthesia, and low cost. Laboratory evaluation of rodent sciatic nerve injury and recovery is widely described and reproducible. Of importance, functional assessment of similar quality and validity to the rodent walking track is not available in larger animal models. Nine 250-300 g male Lewis rats were assigned to two groups and received a sciatic nerve transection. Control rats (n = 3) received a 1-cm sciatic nerve syngeneic graft. Experimental rats (n = 6) received a 4-cm syngeneic graft. Syngeneic grafts were harvested from large (450-500 g) male Lewis rats. In order to eliminate potential redundancy and cross-over regeneration, 4-cm grafts were placed in a looped configuration around the anterior head of the biceps femoris muscle. Sciatic nerves were harvested at 6 weeks (1-cm grafts) and 12 weeks (4-cm grafts) for histology and histomorphometry. In the 1-cm syngeneic graft group, robust regeneration was noted in the distal nerve segment. In the 4-cm syngeneic graft group, robust regeneration also was noted in the distal nerve segment. Histomorphometric values of the distal nerve segment in all groups were as expected for the specified gap length and time point. This report described a simple and effective long nerve gap rodent model employing a looped configuration to prevent cross-over regeneration. This novel approach overcomes the size limitation of the small animal, yet provides all the advantages which make the rodent the model of choice for animal studies. Restoration of Exorotation Following Suprascapular Nerve Neurotization in Obstetric Brachial Plexus Lesions. Martijn J.A. Malessy, Willem Pondaag, Ralph de Boer, Sonja Hofstede-Buitenhuis, and Rietje S. van Wijlen-Hempel. Leiden University Medical Center, Leiden, The Netherlands. A total of 86 infants with an obstetric brachial plexus lesion (OBPL), including loss of suprascapular nerve (SSN) function due to rupture or avulsion from spinal nervee C5 or rupture of the superior trunk, were treated. Neurotization of the SSN was performed, aiming at reinnervation of the infraspinatus muscle to restore exorotation. Preferably, a graft was led out from C5 to the distal SSN target stump. The accessory nerve (XIN)-SSN transfer was performed when the C5 proximal nerve stump was avulsed or ruptured, but contained less than 50% myelinated fibers on frozen section examination. XIN-SSN transfer was also applied when the C5 stump had been used for nerve reconstruction aiming at reanimation of the hand or biceps muscle. Nerve grafting from C5 was performed in 65 patients, and XIN-SSN transfer in 21 patients. Restoration of true glenohumeral exorotation following the two SSN neurotization procedures was evaluated and compared. True glenohumeral exorotation, i.e., effected by reinnervated infraspinatus muscle, was defined as follows: the angle between the 90-degree (actively or passively) flexed elbow resting against the belly and the flexed elbow after exorotation with the upper arm in adduction, until the scapula begins rotating. In addition, two specific tasks which normally include glenohumeral exorotation were evaluated, namely, the ability to reach the mouth and the back of the head. The mean age of the infants at the time of surgery was 5.3 months and the mean follow-up was 3 years. Thirty-five patients (41%) were unable to perform true glenohumeral exorotation, and 20 patients (23%) could reach the sagittal plane. Only 31 patients (36%) were able to exorotate through the sagittal plane. However, functional scores showed that 88% could reach their mouths and 75% the backs of their heads, regardless of the presence of true glenohumeral exorotation. There was no statistically significant difference in outcome between nerve grafting from C5 and XIN transfer. The restoration of a fair range of true glenohumeral exorotation after neurotization of the SSN in OBPL infants, whether by grafting from C5 or XIN transfer, was disappointingly low. However, it seems that compensatory techniques, such as rotation of the scapula, contribute to effectuate a considerable range of movement. Neurotization (Nerve Transfer) of the Axillary (Circumflex) Nerve Using Elements of the Posterior Cord: Anatomic Considerations and Clinical Examples. Jose L. Borrero. Florida Hand Center, Altamonte Springs, Florida, USA. Transferring branches of the posterior cord to the axillary nerve allows for coaptation between a freshly cut functioning proximal nerve source and the recipient axillary nerve. This leads to a desirably reduced time for deltoid muscle reinnervation. The resulting improved shoulder motion is the combined product of deltoid and teres minor muscle action, providing abduction and external rotation. Neurotization with the lower subscapular nerves or the thoracodorsal nerve denervates strong internal rotators, an added benefit for patients with excessive internal rotation. A review of the literature on brachial plexus anatomy and 10 cadaver dissections supported the use of these transfers. Illustrations found in well-accepted textbooks of anatomy depict a "standard configuration" for the brachial plexus, with illustrations that are clear but do not represent the most common pattern. Subscapular and axillary nerves frequently have the same cervical root origin. Transfers using subscapular nerves make sense in situations in which the axillary nerve has been avulsed from its origin at the posterior cord and there is no injury to the cervical roots or the upper trunk. In cases of root avulsion, transfers using branches of the posterior cord make sense, only if the recipient axillary nerve and the supplier nerve do not have the same brachial plexus root origin. Since root origin and fiber quantity of a peripheral nerve are seldom known with exactitude or precision, the suitability of the nerve to be transferred is determined by direct stimulation during surgery. In three adult cases, deltoid reinnervation with subscapular nerves was quite successful. In one obstetric paralysis case, using the thoracodorsal nerve rapidly improved abduction and external rotation, and reduced the tendency toward excessive internal rotation. Our Experience with Ipsilateral C7. Julia K. Terzis and Epaminondas Kostopoulos. Eastern Virginia Medical School, Microsurgical Research Center, Norfolk, Virginia, USA. In devastating plexopathies with widespread avulsion injury, both ipsilateral and contralateral motor donors may be used to achieve useful function. In some instances, the use of ipsilateral C7 (IC7) could be the treatment of choice. Since November, 1991, 38 patients were treated with IC7; 21 of them had OBPP, while the other 17 had adult posttraumatic plexopathies. The IC7 was used for direct neurotization in nine cases in the OBPP group (5 AD, 4 PD) and indirectly, 31 times (15 PD, 16 AD), while in the adult group, direct neurotization took place five times and, indirectly 19 times (7 AD, 12 PD). The anterior division (AD) was the donor in four OBPP cases (19%), the PD in four (19%), and both of them in the remaining 13 (62%). In adult plexopathies, the PD was used in nine cases (53%), the AD in one (6%), and both of them in the remaining seven (41%). The overall muscle strength recovery was compared to the preoperative presentation. The scale used (British Medical Research Council Grading System expanded with ± and - grades) was as follows: poor, M0 to M2; fair, M2 ± to M3; good, M3 ± to M4-; and excellent, M4 to M5-. In the OBPP group, the results were excellent and good in most of the cases for: deltoid reconstruction (6/7 cases; always with PD), triceps (13/17, always with PD), and biceps reconstruction (11/12, 6 with AD and 5 with PD). In the posttraumatic adult group, the results for deltoid reconstruction (10 cases, always with PD) were excellent and good in six, fair in two, and poor in two. For biceps reconstruction (4 cases, all with AD), results were poor in one (1/4), good and fair in the others (3/4). In two patients, the IC7 was directed to free muscles for finger flexion and extension; in both cases, the results were fair. The use of IC7 is a useful therapeutic option for selective cases of brachial plexus paralysis. The best results were seen in young patients with short denervation times. Use of Botulinum Toxin Type A as a Surgical Adjunct in the Management of Brachial Plexus Birth Trauma. Patricia E. Di Taranto, John A. Grossman, Andrew Price, Ilkar Yaylali, Lorna Ramos, and Hebert Valencia. Miami Children's Hospital, Miami, Florida, USA. There is little information in the literature regarding the use of botulinum toxin type A in the management of the shoulder deformity associated with brachial plexus birth trauma. Over a 4-year period (1999-2002), botulinum toxin type A was used in 74 patients as an adjunct to surgical treatment. The surgical procedures included primary nerve reconstruction (44 patients) and subsequent muscle transfer for restoration of shoulder external rotation (30 patients). In all patients, a dosage of 10 units/kg was used. The recipient muscle(s) was either the pectoralis major only (10 patients) or the pectoralis major and latissimus dorsi (64 patients). Shoulder function was evaluated using the Gilbert and Miami Shoulder Scales. Outcome in surgical cases treated with botulinum toxin type A was compared to identical surgical cases performed by the same surgical team without botulinum toxin type A. Shoulder function grades were significantly higher at a minimum 2-year follow-up in cases in which botulinum toxin type A was used. Botulinum toxin type A injection into the shoulder internal rotators is a useful adjunct to the primary and secondary surgical treatment of brachial plexus birth injuries. A Randomized Prospective Multicenter Study of a Biodegradable NeurolacR Nerve Guide for Sensory Nerve Repair in the Hand. M.F. Meek, M.J.O.E. Bertleff, M.J.P.F. Ritt, B. Van der Lei, A. De Boer, P. Houpt, and J.-P. A. Nicolai. Department of Plastic Surgery, University Hospital Groningen, The Netherlands. Injury of peripheral nerves is a common problem, and is always associated with loss of nerve function. Current repair techniques include direct end-to-end suturing or interposition of autologous nerve grafts. The preferred technique to repair peripheral nerve gaps without tension is autologous nerve grafting. This technique has only minor problems, but the results are less than satisfactory and may lead to donor-site morbidity. An alternative for peripheral nerve repair is the use of nerve conduits, to guide regenerating nerve fibers toward the distal nerve stump, while neuroma formation and ingrowth of fibrous tissue into the nerve gap are prevented. Since 1989, the authors have had experience with the application of a biodegradable nerve guide composed of poly (DL lactide-ε-caprolactone). The purpose of this randomized prospective multicenter study was to compare the NeurolacR nerve guide (Polyganics BV, Groningen, The Netherlands) with standard end-to-end nerve repair for sensory nerve defects in the hand. From August, 2002 to May, 2003, 53 patients were randomized in a prospective multicenter study. Patients with a sensory nerve defect in the hand, meeting the selection criteria, were treated with either a Neurolac nerve guide or according to current practice (nerve grafting or end-to-end repair). Quantitative sensory testing using the Pressure-Specified Sensory DeviceTM (PSSD, Sensory Management Services, Inc., Baltimore, MD, USA) was performed in all patients at 3-month intervals for a 1-year follow-up period to evaluate sensory nerve recovery. Fifty-three patients have been included. At 6 months, sensory nerve recovery was observed in both arms of the study. The results were comparable in terms of sensory return. Twelve months of data on sensory nerve recovery were presented. Results indicated that the Neurolac nerve guide can be used for the repair of sensory nerve defects up to 2 cm in the hand. The return of sensory nerve function in both groups was comparable. Patients will be followed further in order to obtain long-term data on sensory nerve recovery in both experimental and control groups. Use of a Collagen Conduit for Primary Repair of Complex Digital Nerve Injuries. Randy Bindra, Barbara J. Sager, and Kimberley Betts. University of Arkansas for Medical Sciences, Little Rock, Arkansas, USA. Even the most meticulous microsurgical repair of a simple digital nerve laceration will not restore normal sensibility to the finger. In complex cases of digital nerve laceration, resection of the nerve ends results in a gap that does not allow tension-free end-to-end repair. Use of conventional nerve grafting techniques with donor morbidity is less desirable, and the use of a nerve conduit offers an attractive option. The authors presented their results with the use of a xenograft collagen nerve conduit (NeuraGen, Integra Life Sciences, Plainsboro, NJ, USA) for the repair of complex digital nerve injuries. This was a retrospective study of 10 consecutive digital nerve repairs (8 fingers, 2 thumbs) in 9 patients with complex digital nerve injuries, who were treated with the use of a collagen nerve conduit, as primary end-to-end repair was not achievable. The ages of patients ranged from 8 to 48 years, and the time to surgery varied from 3 weeks to 4 years. Nine cases were traumatic lacerations, 1 was a dog bite. After surgery, all patients underwent immediate protective mobilization under therapy supervision. Follow-up ranged from 4 months to 1 year. Sensory assessment was done by Semmes-Weinstein monofilament and static two-point discrimination testing by a certified hand therapist. In all cases, the wounds healed without complications. No patient developed painful scars or neuroma in the finger. At least protective sensation was recovered in all patients, and the mean static two-point evaluation was 9 mm. The xenograft collagen nerve conduit was easy to handle, provided a flexible and secure repair to allow early controlled mobilization, and yielded satisfactory clinical results without complications. The outcome was comparable to the reported results of using vein conduits or autogenous nerve grafts for similar injuries. In Vivo Three-Dimensional Imaging of the Human Median Nerve by Diffusion-Tensor MR. M.W. Stenekes, J.M. Hoogduin, J.-P.A. Nicolai, and M.F. Meek. Department of Plastic and Reconstructive Surgery, University Hospital, Groningen, The Netherlands. The in situ assessment of axonal projections of the peripheral nervous systems has been severely limited by the lack of noninvasive techniques to study this type of anatomy. These authors demonstrated that in vivo three-dimensional (3D) reconstruction of axonal projections can be achieved using a 3D diffusion-tensor imaging technique, combined with a designed fiber reconstruction algorithm. As a first example, the pathway of the median nerve was investigated. The images of a patient who sustained an isolated unilateral median nerve lesion were compared with images of healthy individuals. This was the first time in vivo images of peripheral nerves were obtained by the 3D diffusion-tensor imaging technique. High-Resolution MRI of the Human Median Nerve. Archie A. Heddings, Mehmet Bilgen, Randolph Nudo, E.B. Toby, Terence McIff, and William M. Brooks. Kansas University Medical Center, Kansas City, Kansas, USA. It is widely accepted that peripheral nerve repairs performed within 6 weeks of injury have much better outcomes than those performed at later dates. However, there is no diagnostic technique that can determine if a traumatic peripheral nerve injury (PNI) requires surgical intervention in the early post-injury phase. The objective of this study was to determine whether novel, noninvasive magnetic resonance imaging (MRI) techniques could demonstrate the microstructure of human peripheral nerves that is necessary for determining a prognosis, and determining if surgery is indicated following traumatic injury. Ex vivo MRI protocols were developed on a 9.4 Tesla research scanner using spin-echo proton density and gradient echo imaging sequences, and a specially designed inductively-coupled radio frequency coil. These imaging protocols were applied to in situ imaging of the human median nerve in four fresh-frozen cadaver arms. Noninvasive high-resolution images of the human median nerve were obtained. Structures in the nerve that were observed included fascicles, interfascicular epineurium, perineurium, and intrafascicular septations. Application of these imaging techniques to clinical scanners could provide physicians with a tool that is capable of grading the severity of nerve injuries and provide indications for surgery in the early post-injury phase. Nerve Regeneration in Hand Transplantation. Frederic Schuind, D. Abramowicz, Carlo Van Holder, D. Mouraux, Chantal Robert, and A. Meyer. Erasme University Hospital, Brussels, Belgium. The rate of nerve regeneration has been more rapid than expected in hand transplantation cases. On June 15, 2002, these authors transplanted the right distal third of a forearm of a 27-year-old brain-dead man to a 22-year-old male patient who had lost his right dominant hand in a work accident. The donor hand matched for size and skin tone and 3/6 HLA. Immunosuppression included antithymocyte globulins as induction therapy and tacrolimus, mycophenolic acid, and prednisolone as induction and maintenance therapy. The patient had immediate, although very poor, sensibility at two dorsal locations. This observation, confirmed by independent observers, was further demonstrated after 1 week by cortical activation on f-MRI. On a skin biopsy taken to rule out rejection at 2 months, normal myelinated axons were observed. The progression of Tinel's sign was quite rapid, as well as the return of sensibility. The patient perceived hot and cold, pain, proprioception, and could localize touch at the fingertips. There was motor return of the transplanted intrinsic muscles. Most activities of daily life are possible; functional tests showed the transplanted hand to be superior to a preoperative myoelectric prosthesis. The general appearance is very good with satisfactory nail growth. This young patient tolerates the immunosuppression well. Rejection of the transplanted hand has not occurred, and daily tasks are better performed than with his prosthesis. The quite satisfactory nerve regeneration could be related to the neuroregenerative and neuroprotective effects of FK506. The only explanation concerning the immediate sensibility and persistence of living axons is that some axonal fusion has occurred. The nerve sutures were performed without any tension, on two well-vascularized nerve extremities, using fibrin glue and a few stitches, a situation not very different from the peripheral nerve reconnection technique proposed by L. De Medinaceli. Common Risk Factors of Secondary Carpal Tunnel Surgery. Jeffrey Evan Schreiber, Mark Patrick Foran, David Jerome Schreiber, and E.F. Wilgis. Raymond M. Curtis Hand Center at the Union Memorial Hospital, Baltimore, Maryland, USA. Carpal tunnel syndrome (CTS) is a formidable disease that plagues the patient and incurs great costs to society as well. Currently, carpal tunnel release (CTR) is performed 400,000 to 500,000 times per year and usually results in a cure. Unfortunately, primary CTR can fail, due to reasons previously classified into persistent, recurrent, or new symptoms of CTS resulting from a variety of causes. Identification of specific risk factors for recurrent CTS leading to these causes would provide goals for avoiding such factors, to prevent the recurrence of CTS after CTR, in turn preventing the need for additional surgical procedures. The authors hypothesized that there are several common risk factors associated with most secondary CTRs, and chose to investigate these factors by analyzing the charts of those patients requiring a second CTR procedure. A retrospective chart review was performed to identify patients who underwent secondary carpal tunnel surgery from January, 2000 to March, 2004 at the authors' institution. Patients were identified by searching hospital medical record databases, using the Common Procedural Terminology, 2004, procedural codes for neuroplasty and/or transposition, median nerve at the carpal tunnel (64721), endoscopic carpal tunnel release (29848), and fat graft (15770). Forty-eight patients were found to have had secondary carpal tunnel surgery between January 1, 2000 and March 31, 2004. The mean interval between the first and second procedures was 6.6 ± 7.3 years (range: 1-28 years). Symptoms associated with the recurrence included numbness, pain, tingling, hypersensitivity, muscle spasm, and weakness. Nine patients had diabetes mellitus, 11 had hypertension, and six had gastrointestinal-related illnesses. A total of 603 patients had an open release during the data collection period, and 43 of them had a secondary procedure (7%). A total of 1754 patients had an endoscopic release during the data collection period, and four of them had a secondary procedure (0.2%). Primary release of the carpal tunnel with the open technique was one of the most commonly observed risk factors associated with secondary carpal tunnel surgery. Diabetes and hypertension also contributed to the need for secondary surgery. The endoscopic technique may reduce the incidence of secondary carpal tunnel surgery by reducing scar formation. Effect of Cable Size and Cable Numbers on the Functional Outcome of Nerve Allografts. Alper Sari, Sakir Unal, and Maria Siemionow. Department of Plastic Surgery, Cleveland Clinic Foundation, Cleveland, Ohio, USA. The aim of this study was to compare functional outcome of different nerve grafting techniques when different cable sizes and numbers were used for the reconstruction of nerve defects with nerve allografts. A total of 40 nerve allograft transplants were performed between Brown-Norway (BN RTIn) donors and Lewis (LEW RTI1 recipients. Sciatic nerve defects of 14 mm were created in the recipients, and reconstructed with monocable sural (Group 1, n = 5), monocable peroneal (Group 2, n = 5), tri-cable sural-peroneal-tibial (Group 3, n = 5), and whole sciatic trunk nerve allografts of 14 mm in length (Group 4, n = 5). A 7-day protocol of alpha beta T cell receptor monoclonal antibody (TCRmab) of 250 microgr/day combined with 16 mg/kg/day of cyclosporine A (CsA) was given to the recipients. The return of sensory and motor function was evaluated by clinical tests (pin-prick and toe-spread), and somatosensory evoked potential analysis (SSEP), conducted at 12 weeks. On day 100 post-transplant, nerve allografts and gastrocnemius muscles were sampled for histology. Tri-cable and trunk graft applications showed statistically significant better return of toe-spread function, compared to sural grafts (p < 0.01). This was confirmed by the higher muscle weights obtained in the tri-cable and trunk groups, compared to single cable sural allografts (p < 0.01). P1 and N2 wave latencies obtained by SSEP revealed better recovery of sensory function in tri-cable and trunk grafted allografts, compared to sural and peroneal monocable allografts. The differences in SSEP measurements were statistically significant between tri-cable and peroneal groups, and trunk and peroneal groups (p < 0.01). The regeneration of tri-cable and trunk nerve allografts under a short-term protocol of TCR-CsA immunosuppressive therapy was superior to the regeneration of mono-cable sural and peroneal nerve allografts. These findings differ from the authors' previous data when the same technique was applied in nerve isograft studies, where mono-cable grafts showed comparable results to whole trunk and tri-cable isografts. Antibody Dosing of Costimulation Blockade for Nerve Transplantation. Renata V. Weber, Susan R. Rickman, Daniel A. Hunter, and Thomas H. Tung. Washington University School of Medicine, St. Louis, Missouri, USA. Nerve allografts serve as conduits to guide nerve regeneration and, once the nerve has regenerated through the allograft, immunosuppression is no longer necessary. Anti-CD40 ligand costimulation-blocking antibody (MR1) induces an extended period of immune unresponsiveness. The authors hypothesized that a single dose of MR1 is sufficient for successful regeneration in the murine peripheral nerve allograft model. They wished to determine the minimal single MR1 dose necessary for successful regeneration through a peripheral nerve allograft in the murine model. The findings of this study were to evaluate the ability for successful clinical nerve transplantation using lower doses and shorter courses of antibody therapy than are required for organ transplantation. A tibial nerve allograft murine model was used. Male Balb/c nerve allografts were used in female C57BI/6 mice and divided into three groups. Thirty-two C57BI/6 mice were randomized to four groups. Groups 1 (isograft) and 2 (untreated allograft) served as control groups. Group 3 recipients received a single dose of MR1, 1 mg IP, on day 0 prior to transplantation, and Group 4 received a single dose of MR1 at 0.5 mg IP on day 0. A segment of host tibial nerve was resected to allow reconstruction with a 10-mm tibial nerve iso- or allograft, using standard microsurgical technique. Functional evaluation consisted of walking-track analysis performed at 3 weeks postoperative, prior to sacrifice. The nerve grafts were harvested and assessed by standard histology and quantitative histomorphometry to include total axon count, axon diameter, width and density. ELISPOT assay using host splenocytes quantified the level of immune responsiveness to donor antigen by measuring cellular proliferation and the production of IFN-γ and IL-4. Data were analyzed initially with a one-way repeated measures analysis of variance; if detected, statistically significant differences were further analyzed by a pairwise multiple comparison procedure using the Student-Newman-Keuls method. Immunosuppression was noted by in vitro assay after a single dose of MR1, preventing acute rejection of the nerve grafts. Nerve regeneration occurred through the allografts when a single MR1 dose of 1 mg was given preoperatively. There did not appear to be any significant regeneration, if the dose was lowered to 0.5 mg. Walking-track analysis is pending. A single dose of MR1 can prevent acute rejection of nerve allografts. Blockade of the CD40 costimulation pathway with MR1 allows nerve regeneration across murine nerve allografts, while maintaining immunocompetence. Effects of Modality Matched Nerve Grafts on Peripheral Nerve Regeneration. Chris M. Nichols, Michael J. Brenner, Ida K. Fox, Thomas H. Tung, Daniel A. Hunter, Susan R. Rickman, and Susan E. Mackinnon. Washington University College of Medicine, St. Louis, Missouri, USA. Autologous nerve grafting is the current standard of care for nerve injuries resulting in a nerve gap. This treatment requires the use of pure sensory grafts to reconstruct mixed and motor defects, but the consequences of mismatches between graft and native nerve are unknown. Motor pathways have been shown to preferentially support motoneuron regeneration. Functional outcome of motor nerve reconstruction depends on the magnitude, rate, and precision of end-organ reinnervation. The reported study examined the role of pathway type on regeneration across a mixed nerve defect. Thirty-six Lewis rats underwent tibial nerve transaction and received isogeneic motor, sensory, or mixed nerve grafts. Histomorphometry of the regenerating nerves at 3 weeks demonstrated robust nerve regeneration through both motor and mixed nerve grafts. In contrast, poor nerve regeneration was seen through sensory nerve grafts, with significantly decreased nerve fiber count, percent nerve, and nerve density, when compared with mixed and motor groups (p < 0.05). These data suggest that use of mixed or motor nerve grafts, rather than sensory nerve grafts, will optimize regeneration across mixed or motor nerve defects. Immunomodulation with Anti-CD40L Monoclonal Antibody Enhances Neuroregeneration in the Murine Nerve Allograft Model. Bryan Jacobs, Keith Bishop, Sherri Y. Chan, Melanie Urbanchek, and Paul S. Cederna. University of Michigan, Ann Arbor, Michigan, USA. Anti-CD40L monoclonal antibody (MR1) treatment blocks the CD40-CD40L costimulatory pathway and has been shown to increase survival of murine cardiac allotransplants. Previous work in the authors' laboratory has demonstrated the effectiveness of MR1 in dramatically reducing the Th1 and Th2 responses to peripheral nerve allografts. They have also demonstrated that MR1 results in improved functional muscle recovery following peripheral nerve allografting 60 days post-transplantatiion. This study sought to investigate the effect of MR1 inductive therapy on axonal sprouting and regeneration following peripheral nerve allografting. Three experimental groups were designed to investigate the effects of MR1 therapy versus vehicle on axonal sprouting and elongation. 1) The isograft control group included C57BL/6 isografts transplanted into C57BL/6 recipients with vehicle administration (n = 5). 2) The allograft control group included BALB/c allografts transplanted into C57BL/6 recipients with vehicle administration (n = 5). 3) The allograft experimental group included BALB/c allografts transplanted into C57Bl/6 recipients with MR1 administration (n = 5). Sciatic nerve grafts of 10 mm were harvested from the sciatic notch to the knee. Nerve grafts were utilized to reconstruct similar 10-mm defects in recipient mice. On days 0, 1, and 2, each recipient received an IP injection of either MR1 (1.0 mg in 0.2 ml saline) or vehicle (0.9% normal saline). All mice convalesced for 14 days following nerve grafting. The sciatic nerves were then harvested, fixed with gluteraldehyde, epon embedded, sectioned at 3 um, and evaluated using standard histomorphometric analysis. Fourteen days following sciatic nerve grafting, a significant increase in axonal sprouting, axonal elongation, and percent neural tissue was identified at the distal end of the sciatic nerve graft in the MR1-treated nerve allograft mice, compared to the untreated nerve allograft recipients. Analysis of variance showed a significant effect for the variable of treatment type (with or without MR1). Histomorphometric analysis of the proximal and mid portions of the nerve grafts will be completed to more accurately characterize the effect of treatment type on neuroregeneration. In this experiment, a significant increase in axonal sprouting and elongation was observed in nerve allograft transplant recipients that received MR1, compared to those allograft recipients receiving only vehicle. There were no differences in axonal sprouting and elongation identified between nerve allograft recipients receiving MR1 and isograft control mice, suggesting that the blunted Th1 and Th2 responses seen with CD40/CD40L costimulatory pathway blockade translate into improved functional recovery following nerve allografting. Schwann Cell Injected Cold Preserved Nerve Allografts for Reconstruction of the Long Nerve Gap: Preliminary Results. Ida K. Fox, Chris M. Nichols, Michael J. Brenner, Daniel A. Hunter, Andres Jaramillo, Patrick M. Wood, Thalachallour Mohanakumar, and Susan E. Mackinnon. Washington University, Saint Louis, Missouri, USA. Prolonged cold preservation of peripheral nerve allografts has been shown to produce a non-immunogenic scaffold that can be used to reconstruct short peripheral nerve gaps. The reintroduction of cultured Schwann cells to these cold preserved nerve allografts was not detrimental to, but did not significantly improve, regeneration in a short gap rodent model. The goal of this study was to determine the effect of autologous Schwann cell injection on augmenting regeneration across cold preserved allografts in a long nerve gap model. The ability of 8-week cold preserved, autologous Schwann cell injected nerve allografts to support regeneration across a long nerve gap was assessed in the inbred mini-swine and outbred non-human primate (NHP) models. Prior to transplantation, autologous Schwann cells were cultured from the harvested sural nerve of each recipient animal. At 8 weeks prior to transplantation, appropriate nerves were harvested from donor animals for cold preservation in University of Wisconsin Organ Preservation (UW) solution. On the day of transplantation, the appropriate graft was placed after injection with either cultured Schwann cells, placebo (phosphate buffered saline) or nothing (autografts). In mini-swine, a cold preserved allograft was directly compared to a contralateral cold preserved, Schwann cell-injected graft. In NHPs, cold preserved Schwann cell-injected allografts were compared to autografts. Outcomes assessed included histology, histomorphometry and immunologic testing (the mixed lymphocyte reaction in mini-swine and the ELISPOT assay in NHPs). Preliminary results showed that addition of cultured autologous Schwann cells did improve regeneration over that seen in cold preserved allografts alone. The correlation of immunologic assay results with measurement of transplanted organ success by histologic means was highly variable between individual experimental subjects. The combination of prolonged cold preservation with autologous Schwann cell injection may be useful for the reconstruction of long peripheral nerve gaps. Careful interpretation of ELISPOT assay results is necessary in the NHP model, as these may not correlate precisely with degree of nerve allotransplant rejection. Relatedness in the NHP model is difficult to assess, and use of immunologic assays to track these differences is challenging, due to a high degree of inter-individual assay response variability. Also, unlike humans, NHPs tend not to scar extensively, which may skew results. Although it may seem counterintuitive, the inbred mini-swine model may serve as a more appropriate pre-clinical animal model to assess regeneration across long nerve gaps. Successful Culture of Adult Lewis Rat Schwann Cells. Jonathan Cheng, Michael Agresti, J.G. Yan, and H.S. Matloub. Medical College of Wisconsin, Milwaukee, Wisconsin, USA. Schwann cells have been implicated in the establishment of a molecular milieu permissive toward regeneration of axons. Substantial tissue engineering efforts have consequently focused on development of a Schwann cell-based synthetic graft for nerve reconstruction. Although numerous techniques for Schwann cell culture have been described, no ideal method exists. Current techniques demand the use of clinically unacceptable mitogenic factors or donor nerve predegeneration. Furthermore, existing methods have failed in the culture of Lewis rat Schwann cells. Lewis rats are especially favorable for the study of nerve reconstruction, as they exhibit a negligible rate of self-mutilation following nerve injury, and possess a disposition which facilitates walking-track evaluation. The authors described a technique for the successful culture of adult Lewis rat Schwann cells. Adult Lewis rat sciatic nerves were microdissected to remove soft tissue, epineurium, and perineurium, and finely minced to create nerve explants. Explants were placed in DMEM + 10% FBS, and allowed to undergo in vitro degeneration. Explants were then dissociated with collagenase and dispase for 4 hr to form "microexplants." Microexplants were plated on poly-L-lysine (PLL)-coated flasks with Bottenstein-Sato (BS) media with a variable FBS regimen: Group 1 - 1 week in vitro degeneration, then BS + 10% FBS x 4 days, followed by BS + 2.5% FBS; Group 2 - 1 week in vitro degeneration, then BS + 2.5% FBS; Group 3 - 4 weeks in vitro degeneration, then BS + 10% FBS x 4 days, followed by BS + 2.5% FBS; Group 4 - 4 weeks in vitro degeneration, then BS + 2.5% FBS. When cells reached near-confluence, they were subcultured in BS + 2.5% FBS. If cells had not achieved near-confluence by 6 weeks, they were subcultured by default. At the time of subculture, cultures were examined for cell count by hemocytometer and for purity by immunofluorescence for S100. Groups 1 and 2 yielded > 1.9 × 107 cells/g nerve, while Groups 3 and 4 yielded > 8.4 × 107 cells/g nerve, with > 90% Schwann cell purity. Cultures of high purity and number were obtained by 7 weeks after initial nerve harvest. Effects of duration of in vitro degeneration and percentage of FBS were discussed. A technique for the successful culture of adult Lewis rat Schwann cells was described. Following a period of in vitro explant degeneration, the use of Bottenstein-Sato media with reduced levels of fetal bovine serum allowed the differential expansion of Schwann cells over fibroblasts. The authors defined the serial transfer of dissociated "microexplants" with accelerated cellular outgrowth. Nerve and Skin Regeneration - Shared and Interdependent Events Enhanced by GM284. David E. Weinstein, Khalid Parris, and Raymond Birge. GliaMed, Inc., New York, New York, USA. Integrity of the skin and intradermal structure is dependent on innervation. Four days following sensory denervation, keratinocyte proliferation is reduced by half, and epidermal thickness reduced by 70%. These deficits are reversed upon regeneration of the cutaneous nerves. The events in wound healing can be divided into two basic phases: re-epithelialization, in which wounds are sealed, and mesodermal regeneration, in which the follicular and glandular structures of the skin are regenerated. In the absence of repair of the mesoderm, re-epithelialization is unstable. The authors have recently described the ability of GM284, a novel non-immunosuppressive immunophilin ligand, to enhance nerve regeneration. In the reported study, they showed that topical administration of the same drug enhances wound healing, by merging the two phases of regeneration into a single, contemporaneous event, that occurs much more rapidly than in untreated or vehicle-treated wounds. In addition, they showed that GM284 recruits stem cells into the healing mesoderm, and thus hastens the generation of new glands and follicles following full-thickness injury, while it promotes the enhanced regeneration of cutaneous nerves. Finally, they showed that GM284 induces the up-regulation of a tissue-specific gene and the POU protein SCIP in healing skin, which is very similar to the events in regenerating nerve. These observations suggest that there are shared molecular characteristics in regenerating tissue, and thus this begins to define a "biology of regeneration." Serratus Anterior In Vivo Contractile Force Study. Scott D. Lifchez, James R. Sanger, Mario G. Gasparri, John A. LoGiudice, Hani S. Matloub, and William B. Tisol. Medical College of Wisconsin, Milwaukee, Wisconsin, USA. Late paralysis of the muscles innervated by the facial nerve and of the branches of the median and ulnar nerves to the hand, represents a serious surgical challenge. A major limitation of functional muscle transfers performed to reconstruct these nerve deficits is the inability to predict the force that will be generated by the transplanted muscle. No study has previously documented the in situ contractile force of a potential donor muscle. These authors studied the contractile force of the slips of the serratus anterior in situ in 11 patients undergoing thoracotomy. Force was measured via a sterile strain gauge directly sutured to the superficial muscle fascia. Each slip was then independently stimulated. Mean contractile force generated by a slip of the muscle was 0.193 pounds (range: 0.019 to 0.797 pounds). This result compares favorably with a previous study of the maximum force generated by smiling (0.307 pounds). In addition, the lowest slip of the serratus anterior showed a trend of less force generated than those above it (0.185 vs. 0.209 pounds); this difference did not reach statistical significance by Student's t-test. However, the contractile force of the lowest slip, when expressed as a percentage of the average force of the more superior slips, was significantly different (69.6% vs. 100%, Student's t-test) from that of the more superior slips for 9- and 10-slip muscles. Also demonstrated was a significant decrease in muscle strength with increasing age (r = -0.805, p = 0.005, linear regression). These data have significant implications for facial and hand reanimation. The ability of individual slips of the serratus to generate contractile force on par with that required for smiling suggests that each slip could be used to restore a separate facial motion. Further separation of the flap along pre-existing fascial planes may allow up to 10 independent force vectors to be generated by a single muscle flap. This far exceeds the number of force vectors obtainable by other currently used muscle donors, such as the gracilis, latissimus dorsi, and pectoralis minor. These features make the serratus anterior muscle flap an attractive option for reconstruction of deficits due to late facial and intrinsic hand nerve dysfunction. It is hoped that this report will stimulate further research on the strength and capabilities of muscles used to reconstruct permanent nerve deficits. Pilot Study of Postoperative Urinary Continence after Unilateral Cavernous Nerve Reconstruction During Prostatectomy. Catherine Curtin, Martin Sanda, John Wei, Rodney Dunn, James Montie, Kevin Chung, and Paul Cederna. University of Michigan, Ann Arbor, Michigan, USA. Prostate cancer is the most common form of non-skin cancer affecting American men. The surgical treatment is prostatectomy, which can result in urinary incontinence and sexual dysfunction. Erectile dysfunction after prostatectomy is caused, in part, by damage to the cavernous nerve plexus, and studies suggest that microsurgical reconstruction of the cavernous nerve(s) improves postoperative sexual function. There is also evidence that these same nerves may be important in maintaining urinary continence. The authors therefore hypothesized that reconstructing a cavernous nerve during prostatectomy would improve not only sexual function, but urinary continence as well. A validated instrument, the Expanded Prostate Cancer Index Composite (EPIC), was used to assess the urinary and sexual function of patients undergoing prostatectomy with cavernous nerve reconstruction. Scales on this instrument range from 1-100, with a higher score representing better function. Microsurgical reconstructions of the cavernous nerves were performed by two microsurgeons using sural nerve interposition grafts. Two comparison subject groups were included to provide context for the findings. These subjects had completed the EPIC questionnaire as part of a retrospective study to assess the quality of life after prostatectomy, and included age-matched men from the community, who did not have prostate cancer (n = 112), and men who underwent prostatectomy with resection of a cavernous nerve but no reconstruction (n = 50). During the study period, 24 unilateral sural nerve graft reconstructions of the cavernous nerves were performed. One year after surgery, 21 of these patients completed the EPIC questionnaire. The EPIC urinary incontinence score for the community men without prostate cancer was 93; those who underwent prostatectomy with unilateral cavernous nerve reconstruction had a mean continence score of 72; and control patients who had one nerve removed but no reconstruction had a score of 75. The sexual score for the men in the community was 61; those who had unilateral nerve reconstruction had a score of 49; and patients who did not have reconstruction had a score of 34. The data suggested that sexual function was improved in patients who underwent unilateral cavernous nerve reconstruction, a result consistent with other published series. However, the authors did not see improved urinary continence with unilateral cavernous nerve reconstruction. This pilot study was limited by small sample size; thus, the authors still recommend that surgeons assessing the outcomes of this procedure should follow both urinary and sexual outcomes. Using a Peripheral Nerve to Bridge a Lesion of the Spinal Cord: A Novel Surgical Concept and Operative Technique. David T.W. Chiu, GuoLi Hu, Jack H. Martin, and Richard T. Ambron. Columbia University/New York University, New York, New York, USA. The regenerative capacity of a peripheral axon that makes peripheral nerve repair successful continues to elude clarification. Attempts to graft a segment of peripheral nerve around the zone of injury, i.e., to bridge the gap between the proximal and distal stump of an injured spinal cord, has met with only limited success. A novel surgical approach, utilizing the regenerative capacity of a vascularized peripheral spinal nerve, was conceptualized, and an experimental procedure has been tested on a rodent model. The procedure involves: 1) dissection of the T13 spinal nerve along with its axial vessels and detaching it from its muscle; 2) a dorsal laminectomy, including the T13 foramen and the laminae of T13 and L1; 3) creation of an ipsilateral hemisection spinal lesion between T13 and L1 level; 4) transposition of the T13 spinal nerve to the spinal cord via the laminectomy; and 5) insertion of the cut end of the T13 nerve stump into the dorsal lateral quadrant of the spinal cord below the lesion. Four weeks postoperatively, anterograde and retrograde staining demonstrated outgrowth of motor axons from the T13 proximal stump into the motor nuclei at the insertion site and establishment of contact with the motoneurons. Most remarkably, the authors observed that: 1) there was minimal cavitation at the site of spinal nerve insertion and this was likely to be one explanation for the successful regeneration; 2) immunocytochemistry and confocal microscopy indicated that some of the proteins that govern the formation of the neuromuscular junction were also involved in the formation of the synapse between the motor axons and the lower motor neurons. Among these proteins is agrin, which is involved in clustering of acetylcholine receptors. This procedure has great promise, both for restoring connections between the brain and motor circuits below a spinal cord lesion and for patients with chronic spinal injuries. Since spinal circuits below a lesion retain function, a peripheral nerve bridge should interact with these circuit months or years after injury. Water-Immersion Skin Wrinkling of the Digit is due to Vasoconstriction. Ching-Hua Hsieh, Hu-Hong Tsai, and Seng-Feng Jeng. Chang Gung Memorial Hospital in Kaohsiung, Kaohsiung, Taiwan. Water-immersion skin wrinkling is a reliable and simple test for sympathetic innervation, which regulates digital blood flow. Skin wrinkling due to vasoconstriction has been suggested, because it occurs with significant reduction in blood flow velocity in the digital vessels. However, the coexistence of vasoconstriction and skin wrinkling has not been confirmed to be a cause-effect relationship or merely an epiphenomenon. In the clinical situation, O'Riain (1973) observed that no skin wrinkling occurred in the replanted finger after water immersion (O'Riain's sign). This reported study was designed to investigate the changes in blood perfusion after water immersion in both replanted and contralateral normal fingers, and to test the hypothesis that wrinkling is a function of digital vasoconstriction, the assumption being that the replanted finger will not demonstrate vasoconstriction after water immersion. The study was performed on 12 replanted fingers and 12 contralateral normal corresponding fingers of seven patients who had sustained complete amputation. Measurements were performed prior to and after immersion for 30 min in 0.5 mol/L NaCl solution at a constant 40 degrees C. Laser Doppler was used to measure the flow velocity and blood perfusion of the pulp. Mean laser Doppler perfusion measurements were given as relative flux units and were estimated for quantitative comparisons. All normal digits showed marked skin wrinkling with reduced blood flow and tissue perfusion (11%, p < 0.01). All corresponding replanted digits did not show skin wrinkling, but demonstrated improved blood flow and tissue perfusion (45%, p < 0.01) after water immersion. In addition, all the replanted fingers demonstrated remarkable sensory recovery. The study clearly demonstrated that skin wrinkling after water immersion was due to vasoconstriction of the digital pulp. The pulp has a very high concentration of arteriovenous vessels and is one of the few sites in the body containing large numbers of "glomus bodies" that are directly linked to the dermis and, through epidermal anchoring, to the epidermis. It has been suggested that the loss of volume of the glomus bodies through vasoconstriction generated a negative digital pulp pressure, pulling the epidermis, and forming skin wrinkling. The paradoxical vasoconstriction, but not vasodilatation, accompanying water-immersion wrinkling at increased temperatures may be due to thermoregulation of the body, reducing heat gain from the environment. Replanted fingers may lose this regulatory function related to sympathetic innervation. Intraoperative Nerve Recordings as a Useful Aid in the Management of Neuroma-in- continuity. Mirsad Mujadzic, Tuna Ozyurekoglu, Amit Gupta, Vasudeva G. Iyer, and Warren C. Breidenbach. Christine Kleinert Institute, Louisville, Kentucky, USA. The purpose of this study was to investigate the importance of intraoperative nerve recordings (IONR) in the surgical treatment of neuroma-in-continuity (NIC). The authors wanted to see if IONR changed the decision-making process significantly, regarding the excision and grafting of NIC vs. neurolysis. Thirty-eight patients with 47 NIC were managed using IONR and had an adequate follow-up. There were 19 neuromas at the brachial plexus level and 28 neuromas at individual nerve levels. The NICs with clinically poor or no function were classified as type A, and NICs with partially preserved sensory or motor function were classified as type B. There were 37 type A neuromas and 10 type B neuromas. Resection was anticipated for all type A neuromas. Nerve compound action potentials (NCAP) were recorded across each neuroma. In type A neuromas, neurolysis was done if the amplitude was over 30% of normal NCAP, and resection and grafting were performed if the amplitude was below 30% of normal NCAP. In type B neuromas, NCAPs were recorded to differentiate the functional nerve fascicles from nonfunctional ones. Overall, 26 neuromas were excised and grafted, and 21 were neurolyzed. Fourteen neuromas were analyzed histologically. Seventeen of 37 type A neuromas showed NCAP more than 30% of normal, and thus were only neurolyzed. The gain in information was 46%. Eleven of the brachial plexus level neuromas were excised and grafted, and six of these achieved > M3 motor recovery. Eight were neurolyzed, and five achieved > M3 motor recovery. Nine neuromas were excised and grafted at the individual nerve level, and five achieved > M3/S3. In nine cases, neurolysis was performed; > M3/S3 recovery was obtained in six cases. The percentage of myelinated nerve fibers in histologic sections highly correlated with the amplitude of NCAP. In type B neuromas, six were partially excised and grafted, and four were neurolyzed. Gain in information was 60%. All the patients achieved > M3 or > S3 recovery. IONR seemed to be a helpful aid in the decision-making process for surgical treatment of neuroma-in-continuity at the brachial plexus and individual nerve levels. NCAPs helped preserve the functional fascicles of both types of neuroma-in-continuity. How Do We Evaluate Functional Recovery after Nerve Repair? An Open Controlled Clinical Study of Complete Median Nerve Injuries. Maria Åberg, Christina Ljungberg, and M. Wiberg. Department of Surgical and Perioperative Science, Hand and Plastic Surgery, Omeå, Sweden. The current lack of reliable and objective methods for evaluating results of peripheral nerve repair is a problem when investigating novel treatments in clinical research and clinical practice. This study aimed at generating reference materials and evaluating the applicability of a battery of tests used for evaluation of peripheral nerve injuries. Fifteen patients with a history of complete median nerve transaction and 15 healthy volunteers were included in the study. The groups were comparable with regard to age, sex, and dominant hand. The patient group included six females and nine males, with a mean age of 32.2 years (range: 9-54 years). The healthy volunteer group included seven females and eight males, with a mean age of 34.5 years (range: 15-60 years). All subjects had their right hands as dominant hands, and the distribution of the previously injured hands included eight right-hand injuries and seven left-hand injuries. Each subject was investigated with a battery of tests, including evaluation of functional, sensory, and motor recovery, by means of questionnaires, clinical evaluations, and neurophysiologic and physiologic findings. The results were statistically analyzed, and comparisons were made within the patient group and between the patients and healthy volunteers using a "per protocol" and an "intention to treat" approach. Criteria for success, based on the specificity of each method and the ability to discriminate between patient/healthy volunteer and injured/non-injured hand within the patient group, were stipulated, in order to evaluate the usefulness of each method. Fourteen of 32 (44%) of the variables, representing five of the 15 included methods (33%), were not able to discriminate between injured/non-injured hand and/or between patients/healthy volunteers in this study, and are thus questionable for evaluating nerve repair. However, efficacy results implied that subjective questionnaires (Disabilities of the Arm, Shoulder and Hand score and the four-question form) and performance tests (Assessment of Motor and Process Skills and Sollermans subtests 4, 8, and 10) are useful for evaluating functional recovery after peripheral nerve injury. Sensory recovery could be evaluated by means of two-point discrimination, GAP detection test, recovery according to the non-modified British Medical Research Council, and sensory neurography. Motor recovery would preferably be evaluated by means of electromyography, motor neurography, and manual muscle testing according to Brandsma (1995). Short-Term Results for Peripheral Nerve Decompression in Diabetic and Non-Diabetic Patients. Jerome K. Steck. Orthopaedic Consultants and Alvine Foot and Ankle Center, Sioux Falls, South Dakota, USA. Neuropathy of any cause can be very difficult and frustrating to treat. Medical management is available, but often relief is fleeting. More recently, decompression of lower extremity peripheral nerves has been shown to restore or improve sensation and to decrease pain in neuropaths. Several recent articles demonstrate results for diabetic patients. This study reported on the short-term results of decompression for both diabetic and idiopathic neuropaths. Twenty-five diabetic and 25 idiopathic neuropaths were identified and underwent peripheral nerve decompression between November, 2003 and May, 2004. This consisted of decompression of the common peroneal nerve at the fibular head, tarsal tunnel, and deep peroneal nerve. Patients filled out a preoperative questionnaire at initial clinical examination, were evaluated by the PSSD, NCV and, in most cases, a neurologist preoperatively. Postoperatively, patients were evaluated by the PSSD, repeat questionnaire, and clinically. The mean duration of follow-up was 6 months for both groups; no patients were lost to follow-up. The average patient age was 62 years for the diabetics and 64.5 years for the idiopathics. Twenty-one (84%) of the diabetic patients and 22 (88%) of the idiopathic patients reported an 80% decrease in pain or greater and were considered to have a successful result. Eighteen (72%) of the diabetic patients and 17 (68%) of the idiopathic patients reported a 50% improvement in sensation or more at 3 months postoperatively. Sensation was further evaluated by the PSSD postoperatively, and the diabetic and idiopathic patients had a narrower spacing for two-point sensation and/or a lower pressure threshold in all patients who had improvement. A positive Tinel's sign at one or more of the three known compression sites and a positive PSSD neurosensory evaluation predicted a successful outcome 86% of the time. Peripheral nerve decompression for alleviating the symptoms of neuropathy seems to be a viable option for treatment. Furthermore, as earlier reports of decompression have been promising in the diabetic patient, this report seemed to be equally encouraging for idiopathic neuropathy as well in short-term follow-up. Relationship between Peripheral Nerve Decompression and Gain of Pedal Sensibility and Balance in Patients with Peripheral Neuropathy. Ivica Ducic, Nathan Taylor, and A. Lee Dellon. Georgetown University and Institute for Peripheral Nerve Surgery, Washington, D.C. and Baltimore, Maryland, USA. Previously, these authors reported a correlation between abnormal balance and foot sensibility in a population of patients with impaired lower extremity sensation. The purpose of the reported study was to determine if peripheral nerve decompression (tarsal tunnel release, neurolysis of medial and lateral plantar and calcaneal nerves) could lead to improved foot sensibility, increased proprioception and balance, and decreased falls in a population of patients with impaired lower extremity sensation. Fourteen patients with peripheral neuropathy were included in this study. Seventy-one percent of patients were females. The average patient age was 67 years. All patients were evaluated pre- and postoperatively, to assess their lower extremity sensibility, as well as their ability to stand still, maintaining their balance with their eyes open and then closed, which is defined as "sway." Lower extremity sensibility (one- and two-point discrimination) was measured at the pulp of the big toe and medial heel with the Pressure-Specified Sensory Device. The MatScan Measurement System measured each patient's sway. Neuropathy was the result of diabetes in 57% of patients, a combination of diabetes and hypothyroidism in 7%, and unknown etiology in 36%. Eight patients underwent peripheral nerve decompression in one leg, and six patients had it done bilaterally. Mean toe and heel sensibility improved 19% (p = 0.13, NS) and 18% (p = 0.4, NS), respectively, in the unilateral group, whereas the bilateral group experienced an improvement in mean toe and heel sensibility of 44% (p = 0.003, S) and 26% (p = 0.02, S), respectively. Overall, sensibility for the toe and heel improved 30% (p = 0.0028, S) and 26% (p = 0.02, S), respectively. Pre- and postoperative sway comparison in the unilateral group revealed a reduction in sway with eyes open and eyes closed by 11% (p = 0.65, NS) and 14% (p = 0.34, NS), respectively. Of interest, comparison of pre- and postoperative sway in the bilateral group showed a reduction with eyes open and eyes closed by 26% (p = 0.05, S) and 54% (p = 0.008, S), respectively. Results supported the notion that patients with peripheral neuropathy would benefit from bilateral lower extremity nerve decompression. Overall, postoperative sway with eyes open and eyes closed was reduced 13% (p = 0.2, NS) and 37% (p = 0.02, S), respectively. These results confirmed improvement in pedal sensibility, proprioception, and balance in patients with peripheral neuropathy, that was attainable with lower extremity peripheral nerve decompression. This procedure can help reduce the development of pressure ulcers and can decrease the frequency of falls and fractures in this patient population, leading to decreased morbidity and mortality. Neurolysis of the Lateral Femoral Cutaneous Nerve. Ivica Ducic, Nathan Taylor, and A. Lee Dellon. Georgetown University and Institute for Peripheral Nerve Surgery, Washington, D.C. and Baltimore, Maryland, USA. This study evaluated the efficacy of neurolysis in the treatment of meralgia paresthetica (MP). First described in 1878, MP is a painful mononeuropathy of the lateral femoral cutaneous (LFC) nerve with multiple etiologies, including pregnancy, obesity, external compression from tight clothing, belts, or gun holsters, diabetes, iliac bone graft harvesting, trauma, and entrapment of the nerve from surgical scarring. Occupational measures for treatment of MP involve removal of the source of compression (weight loss, avoidance of constricting garments or accessories), while neurolysis is reserved for those patients who respond poorly to occupational and/or medical management (NSAIDs, amitriptyline, gabapentin, carbamazepine, or steroids). The study retrospectively evaluated the outcomes of 29 patients who underwent neurolysis of the LFC nerve. Twenty-two patients underwent neurolysis on one side, whereas 7 patients underwent bilateral neurolysis, for a total of 36 procedures. Twenty-four (67%) of the 36 procedures involved neurolysis of the LFC nerve solely, whereas 12 (33%) of the 36 procedures involved of the LFC nerve and resection of one or more additional nerves (ilioinguinal, iliohypogastric, genitofemoral). Sixty-six percent of the patients were females, and the average age among all patients was 48 years. Preoperatively, all patients had either a positive Tinel's sign or tenderness in the distribution of the LFC nerve. All patients experienced pain and/or paresthesias in the distribution of the LFC nerve prior to surgery. Surgical outcomes were graded subjectively on a scale from 1 (least favorable) to 5 (most favorable). Overall, of the 36 procedures performed, 69% received a grade of 4 or 5, whereas 31% received a grade of 1 or 2. When comparing outcomes of LFC neurolysis alone vs. LFC neurolysis and additional nerve resection, 53% and 16%, respectively, received a grade of 4 or 5, whereas 14% and 17%, respectively, received a grade of 1 or 2. Following the procedure, it took approximately 2 to 3 weeks for relief of incisional pain and tenderness over the LFC nerve to occur. Relief of paresthesias from time of surgery ranged from 1 day to 4 weeks.. Measurable return of LFC nerve sensation occurred over several months after surgery. Individual differences in outcome regarding pain relief, paresthesia, and return of sensation, as well as the analysis of why certain patients did not respond well to surgery were discussed. The study confirmed that neurolysis of the LFC nerve is an effective treatment for meralgia paresthetica in properly selected patients. Long-Term Functional Results of Endoscopic Cubital Tunnel Release. Reimer Hoffmann and Maria Siemionow. Evangelisches Hospital, Oldenburg, Germany and The Cleveland Clinic Foundation, Cleveland, Ohio, USA. There is an interest in introducing minimally invasive techniques in peripheral nerve surgery. This is confirmed by the routine use of endoscopic carpal tunnel release by many surgeons. Overall success rates of surgical interventions for cubital tunnel syndrome are reported to be within 80-90%. Minimally invasive approaches are reported to give comparable results. However, the discussion about "simple" ulnar nerve decompression vs. anterior transposition is still open. This report presented the rationale, clinical and anatomic indications, a detailed technique, and the authors' results on an endoscopic approach to cubital tunnel release. Seventy-three patients and 75 elbows were treated with an endoscopic cubital tunnel release technique. All patients were evaluated clinically prior to surgery and classified according to Dellon's staging system for nerve compression. All patients were also assessed by electrophysiologic testing. The indications for surgical decompression of the cubital tunnel included symptomatic patients with evidence of muscle weakness and sensory deficits confirmed by neurosensory testing studies. Only patients with recurrent cubital tunnel syndrome were not considered candidates for an endoscopic procedure. A detailed description of the surgical technique, requiring a 15-mm incision and the use of a 4-mm facelift endoscope, was presented. Postoperative assessments included clinical evaluation according to the Bishop rating system, electrophysiologic studies, and a patient satisfaction questionnaire. The mean follow-up for the endoscopic cubital tunnel release was over 27 months. The average length of the endoscopic neurolysis of the ulnar nerve was 17 cm. The anatomic findings suggested that cubital tunnel syndrome may be a multifocal entrapment neuropathy. More than 90% of patients had full elbow motion within 48 hr after surgery, and early return of the function of the adductor digiti minimi and ulnar nerve innervated interosseous muscles confirmed by the cross-finger test. There were no major complications, except of tenderness at the incision in three patients. Based on these evaluations, the results revealed a more rapid functional recovery, lower morbidity, and an earlier return to manual labor, in contrast to the conventional open release. The authors concluded that endoscopic cubital tunnel release is a safe procedure in the hands of the experienced surgeon, and a reliable alternative to the conventional open approach. Surgical Management of Cubital Tunnel Syndrome: A Comparative Analysis of Outcome Using Four Different Techniques. Michel Saint-Cyr, Andrew Scott LaJoie, Soo-Heong Tan, and Tsu-Min Tsai. Christine M. Kleinert Institute for Hand and Microsurgery, Louisville, Kentucky, USA. Various options exist for the surgical management of cubital tunnel syndrome. This paper reported a comparison of the outcome of four different surgical techniques: simple decompression; endoscopic decompression; subcutaneous decompression; and submuscular decompression for the treatment of cubital tunnel syndrome. One hundred and seventeen patients with cubital tunnel syndrome were reviewed from 1986 to 2000. Inclusion criteria included all patients with cubital tunnel syndrome refractory to conservative treatment. Patients with previous cubital tunnel surgery were excluded. Parameters measured included: signs and symptoms, medical comorbidity, other nerve compressions, and anatomic pathology. Severity was evaluated using the Dellon classification and the symptom severity score SSS (pain, clawing, Froment, Wartenberg's). Bishop's rating was measured at final follow-up. Duration of symptoms, conservative treatment prior to surgery, and follow-up were all recorded. Statistical analysis included Student's and Kruskall-Wallis tests, and Spearman's rho for correlation. Parameters correlated included duration of symptoms vs. Dellon, SSS vs. Dellon, and Bishop vs. Dellon. Other outcomes measured included other nerve compression vs. Dellon, surgery type vs. Bishop, and surgery type vs. Dellon. Of the 117 cases treated (61 men, 56 women), nine were bilateral with right and left sides being equally affected (60 R vs. 57 L). The average patient age was 45.7 years ± 12.5. Diabetes and hypertension were present in 6.0% (7/117) and 12.8% (15/117), respectively. The average duration of symptoms prior to consultation was 11.5 months ± 14.5. The mean follow-up was 404 ± 568 days. Conservative treatment prior to surgery was 339 days. Dellon's classification at initial consultation was 9.9% mild, 32.5% moderate, and 37.6% severe. Correlation between severity of nerve compression and symptom duration was not statistically significant. A significant positive correlation existed between the SSS and Dellon score, 0.32. Carpal tunnel syndrome was the most commonly associated nerve compression. Bishop's rating was 46.5% excellent, 39.5% good, 7.9% fair, and 6.1% poor overall. A significant positive correlation existed between the Bishop rating and Dellon score, rho = 0.28 with p = 0.003. The average Bishop score was 1.74 ± 0.85, and no significant difference existed when comparing each surgical technique to one another. No significant association was found between the severity of compression (Dellon) and the surgery type performed. Increasing severity of ulnar nerve compression negatively influenced outcome (Bishop rating). No statistically significant differences were found between the type of surgery performed in regard to outcome and Dellon score. Endoscopic Release of the Cubital Tunnel: A Cadaveric Study. Enrique Pereira, Diego Miranda, and Tsu-Min Tsai. Christine M. Kleinert Institute for Hand and Microsurgery, Louisville, Kentucky, USA. The goal of this reported study was to assess the adequacy and complications of cubital tunnel release with a new technique using a current endoscopic carpal tunnel instrumentation set. Twelve cadaveric elbow specimens were used. A blinded study of endoscopic release of the cubital tunnel using endoscopic instrumentation was carried out. All procedures were performed by one surgeon using a single 3-cm incision. A second surgeon dissected the cubital tunnel after the procedure. The length of the release was measured, and the complications were assessed. The mean length of the release was 5.45 cm proximal (range: 4-8 cm) and 5.06 cm distal (3.5-8.5 cm), when measured from the medial epicondyle. The mean length of overall release was 10.64 cm (7.5-15 cm). There were no complications regarding the nerve, accompanying vessels, or ulnar collateral ligament injury. No subluxation of the ulnar nerve was observed after the procedure. This technique of cubital tunnel decompression through a 3-cm incision protects the ulnar nerve from the extensive dissection seen in other open techniques. Theoretically, it preserves nerve vascularity and prevents possible motor and cutaneous nerve injuries. The simplicity and low complication rate demonstrated in this cadaveric study would support the clinical application of this endoscopic technique using a current carpal tunnel instrumentation set. Minimally Clinically Important Difference in Carpal Tunnel Syndrome Symptom Severity Scale. Tuna Ozyurekoglu, Steven McCabe, and Andrew Scott LaJoie. Christine Kleinert Institute for Hand and Microsurgery, Louisville, Kentucky, USA. The authors aimed at identifying the change score in the Symptom Severity Scale (SSS) of the Carpal Tunnel Questionnaire that is associated with an important change in clinical care. Twenty-eight patients with carpal tunnel syndrome, treated with a steroid injection into the carpal tunnel, completed the Brigham and Woman's Hospital Carpal Tunnel Questionnaire and Katz-Stirrat Hand Diagram before steroid injections and at 3 weeks follow-up. Satisfaction was determined by the clinical behavior of the patients and the doctors' evaluation. Receiver operating characteristic (ROC) curves for overall SSS, pain domain, and sensation domain scores were created. The areas under the curves (AUC) were calculated to determine and compare the responsiveness of each. The minimal clinically important difference was determined by ROC methodology and Youden's index analysis. The SSS of the Carpal Tunnel Questionnaire, its pain domain, and sensibility domain can reveal a meaningful clinical improvement after carpal tunnel injection. The symptom severity scale (p = 0.004), its sensory domain (p = 0.031), and pain domain (p = 0.035) were statistically better than chance (AUC = 0.50). All three measures showed predictive ability. However, the total SSS was the most sensitive in detecting a change (AUC = 0.820). The MCID in the score of the SSS after carpal tunnel steroid injection was found to be 1.045. The SSS can be used to distinguish a clinically important change after carpal tunnel injection. A decrease of 1.045 or more in the score of the SSS may determine such a change from the patient's and physician's perspectives. Role of Free Radicals in Chronic Nerve Compression Injury: Experimental Study with a Rat Sciatic Nerve Model. Jason P. Rehm, Sheila G. Lindley, Michelle Tucci, Michael P. Angel, Paul Thacker, and Steven Longstreet. University of Mississippi Medical Center, Jackson, Mississippi, USA. Acute and chronic nerve compression is the etiologic factor resulting in a great number of visits to any physician caring for patients with hand complaints. Much is known about the gross pathology (i.e., compression of the median nerve under a tight transverse carpal ligament). However, exactly what happens at the molecular level is less well understood. Experimentally, it has been shown that free radicals and oxidants are present in nerves following an ischemic insult. An animal model, using compression provided by a Silastic tube placed circumferentially around a rat sciatic nerve, has previously been described, and free radical changes and fibrosis have been appreciated. This study looked at the effects of deferoxamine, an antioxidant, in a rat nerve compression model. Adult Sprague-Dawley rats were subjected to 6 to 8 weeks of sciatic nerve compression using a Silastic tube. The control rats were sacrificed and a section of the compressed nerve harvested for study. A treatment group received deferoxamine intravenously each week following nerve compression up to the time of their sacrifice and nerve harvest. Assays for malonyldialdehyde (MDA), nitric oxide, protein (internal control), and glutathione were then performed. Both MDA and nitric oxide levels were significantly lower (p = 0.032 and p = 0.004, respectively) in the treatment group, compared to control rats. There was no significant difference in protein levels or glutathione between groups. The iron chelation and antioxidant properties of deferoxamine may have ameliorated the deleterious effects of free radical damage induced during nerve ischemia. This may prove to be of benefit in patients, both preoperatively to minimize permanent fibrosis and to lessen the ischemia/reperfusion injury postoperatively following nerve decompression. Various Applications of Epineurial Sheath Grafts for Nerve Gap Bridging. Maria Siemionow, Yavuz Demir, and Abir Mukherjee. Cleveland Clinic, Plastic Surgery Department, Cleveland, Ohio, USA. Severe nerve injuries may result in significant nerve gaps, making surgical management challenging. Use of standard nerve autografts is associated with donor-site morbidity. In order to reduce morbidity, these authors investigated applications of various epineurial sheath grafts for bridging nerve gaps as an alternative method to nerve autografting. Nerve gaps were created by removing a 12-mm segment of sciatic nerve in 30 inbred Lewis rats. The entire epineurial sheath was next removed from the nerve segment. In Group 1 (n = 6), defects remained unrepaired. In Group 2 (n = 6), autograft repair using a conventional epineurial suture technique with 4 sutures was applied. In Group 3 (n = 6), the defect was repaired by one epineurial sheath strip with centrally aligned suture. In Group 4 (n = 6), two epineurial strips were used to repair the defect with two epineurial sutures placed 180 degrees apart. In Group 5 (n = 6), the whole rectangular epineurial sheath was used to repair the defect without longitudinal splitting, and was anchored with four sutures. Functional evaluation was performed by pin-prick and toe-spread tests at 3, 6, and 12 weeks and by electrophysiologic assessment of somatosensory evoked potentials (SSEP) at postoperative weeks 6 and 12. Denervation atrophy of the gastrocnemius muscles was evaluated by wet muscle weight and muscle morphometry. Toe-spread results were significantly better in rats repaired with full sheath grafts and conventional nerve grafts, compared to single strip graft (p < 0.01), while pin-prick test results did not show differences between study groups at 12 weeks. SSEP evaluation revealed no significant difference in latencies among groups. The wet weight of gastrocnemius muscles and mean muscle diameter were smaller in rats treated with one strip, compared to other groups (p < 0.05). The total number of myelinated fibers in the conventional autograft (4098 axons per specimen), two-strip (3868), and full-sheath groups (5596) indicated adequate regeneration, whereas nerve regeneration was inferior in single-sheath Group 3 (1795). When compared to Group 2, Group 3 animals showed a statistically significant reduction in the total number of myelinated fibers and axon diameter (p < 0.05). In this study, sciatic nerve gap repair with the epineurial sheath graft resulted in functional recovery and histomorphometric outcome comparable to the conventional autologous nerve graft repair technique. It can serve as an alternative method to conventional nerve autografting, as it provides an expanded source of graft material for bridging nerve defects, with significantly reduced donor-site morbidity. Effect of Bioabsorbable Fibers in Nerve Conduits for Peripheral Nerve Reconstruction. Hirokazu Tochigi, Yasushi Nakao, Masanori Takahashi, and Yoshiaki Toyama. Keio University School of Medicine, Tokyo, Japan. As a simple tubular conduit can facilitate repair of nerve gaps less than 10 mm in the rat sciatic nerve gap model, many types of nerve conduit were evaluated for repairing longer nerve gaps (e.g., Schwann call-seeded hybrid graft, drug delivery system from the nerve conduit, genetic modification using viral vector). But many problems, such as expansion and source of Schwann cells, drug side effects, and viral toxicity, should be resolved before using these nerve conduits in the clinical situation. The conduit with collagen fibers or laminin-coated fibers was shown useful to repair nerve gaps over 10 mm in a previous report, but there are no reports on the effect of different fiber numbers in nerve conduits composed of only non-biochemical substances on axonal density and total fiber number. In this reported study, the authors examined the different effects on axonal regeneration of fiber number in nerve conduits. Forty Wistar rats were used, and 15-mm sciatic nerve gaps were created and reconstructed. The tubular conduits and fibers were composed of polycaprolactone-polylactic acid copolymer. The study was divided into four groups. In Group 1 (n = 10), silicone tubes were used; in Group 2 (n = 10), tubular conduits; in Group 3 (n = 10), tubular conduits with 40 fibers; in Group 4 (n = 10), tubular conduits with 70 fibers. At 12 months postoperatively, all rats were reexplored, the sciatic nerve including the nerve conduit was harvested and fixed by glutaraldehyde. At the mid portion of the conduit, cross sections were taken and stained with toluidine blue, to evaluate the axonal contents. In axonal density (fiber number/mm2), statistically significant differences between the silicone and tubular conduit groups were detected (Group 1: 0 ± 0; Group 2: 9548 ± 4193; Group 3: 8536 ± 2059; Group 4: 8381 ± 2331; p < 0.05). In myelinated total fiber number, there were significant differences between the nerve conduit with 40 fibers and the other groups (Group 3: 2734 ± 471; Group 1: 0 ± 0; Group 2: 766 ± 563; Group 4: 124 ± 136; p < 0.05). The tubular nerve conduit with 40 fibers composed of polycaprolactone-polylactic acid copolymer had a good result in total fiber number, demonstrating potential use in nerve guidance in the clinical situation, without any predictable side effects. Characteristics of Neuropathologic Changes in Vibration Injury: Experimental Study. Yu-hui Yan, Hani S. Matloub, Ji-Geng Yan, Lin-Ling Zhang, James R. Sanger, and Danny A. Riley. Medical College of Wisconsin, Milwaukee, Wisconsin, USA. Vibration syndrome, a clinical condition arising from chronic use of vibrating tools, is associated with a spectrum of neurovascular symptoms. To date, only its vascular pathology has been extensively studied. These authors sought to determine what direct neurologic injury, if any, is caused by vibration. Hind limbs of anesthetized rats were affixed to a vibrating platform 4 hr per day for 7 days. Study animals were vibrated with set parameters for frequency, acceleration, velocity, and amplitude; control animals were not vibrated. On day 7, nerves were studied by light and electron microscopy. While light microscopy demonstrated minimal histologic differences between vibrated (n = 12) and control nerves (n = 12), electron microscopic changes were dramatic. Splitting of the myelin sheath and axonal damage (e.g., myelin balls and "finger ring") were consistently seen in both myelinated and nonmyelinated axons. Despite relatively short vibration, definite pathology was demonstrated, suggesting that vibration syndrome has a direct neurologic component. An Animal Model for the Study of Complex Regional Pain Syndrome. Jose Monsivais and Yang Sun. Hand and Microsurgery Center of El Paso, Texas, USA. Complex Regional Pain Syndrome (CRPS) usually develops after an initiating traumatic event. It is not limited to the distribution of a single peripheral nerve, and is apparently disproportionate to the inciting event. The development of an animal model is necessary for the study of this condition and development of treatment plans. Sixty Sprague-Dawlley rats, weighing between 200 and 230 gr, were randomly divided into six groups of 10 rats each. Group 1 - sham operation - nerve roots of C4, C5, C6, C7, and cervical plexus were exposed under the operating microscope with general anesthesia; Group 2 - same procedure as Group 1, but the roots of C5, C6, and C7 were exposed and ligated at the point where they emerged from the foramina; Group 3 - only C4 was ligated; Group 4 - only C5 was ligated; Group 5- same procedure as Group 1, but only C6 was ligated; Group 6 - the BP was exposed, but only C7 was ligated. The animals were observed during a 4-month period and sacrificed at 4 months. During this period, the rats were videotaped for walking-track analysis. Sudo-motor disturbances were recorded. X-rays were obtained to document osteopenia and longitudinal growth. Group 1 animals exhibited normal behavior with no pain or allodynia. Group 2 to 6 animals invariably showed signs of allodynia, which gradually diminished after 2 weeks. Groups 2 and 6 allodynia remained unchanged, and the animals refused to be touched, moved, to eat or drink. After 6 weeks of ligation, Groups 3, 4, and 5 improved, and the allodynia completely resolved. In Groups 2 and 6, improvement was noted, but animals never recovered to the level of the other groups. Allodynia and hypesthesia persisted throughout the study period. Walking-track analysis similarly showed improvement in all groups except Groups 2 and 6. X-rays revealed diffused osteopenia in all groups. All animals showed delayed growth at 2 months, and all were underweight by the fourth month. At 4 months, the forearms were 10 mm shorter than the normal contralateral sides. Ligation of cervical roots, independently or in combination, can produce a syndrome that resembles human CRPS I and II, but the changes are more pronounced when C7 is involved primarily or in combination with other roots. This model appears to be valid and practical for the study of neuropathic syndrome. Functional Recovery after Facial and Sciatic Nerve Crush in the Rat. Tessa A. Hadlock, James T. Heaton, Mack L. Cheney, and Susan E. Mackinnon. Masschusetts Eye and Ear Infirmary, Boston, Masachusetts, USA. The objective of this study was to systematically follow facial nerve recovery following crush to the main trunk, with respect to ocular and vibrissial function, and to compare the rate of facial recovery with the rate of recovery from a sciatic nerve crush in the same animals, as a way of validating the functional parameters of facial nerve recovery against the well-known measure of hind-limb function, the sciatic function index (SFI). Animals underwent exposure of the main trunk of the facial nerve and the proximal segment of the sciatic nerve. Both nerves underwent standardized crush injury, and subsequent daily functional testing. After a plateau of functional recovery was achieved, the animals were sacrificed, and measurements of the distance between the site of injury and the end musculature were made. These permitted determination and comparison of rates of recovery in both systems. All crushes resulted in loss of electrical conductivity, as validated by intraoperative proximal nerve stimulation. Recovery of ocular and vibrissial motor function occurred at a rate of 1.5 to 2.4 mm/day. Return of hind-limb function occurred more slowly, although when corrected for distance, the rate (2.26 mm/day) appeared to match that of facial nerve recovery. Recovery after facial nerve crush follows a predictable time course, and the rate of recovery is consistent with the rate of recovery from sciatic nerve injury. Return of the blink reflex, loss of vibrissial fibrillations, and return of vibrissial sweeping function appear to be internally consistent functional measures of facial recovery. They are quantifiable, and will be useful for future facial nerve manipulation studies.