Description and validation of a simple histological nerve tissue scoring system for nerve allografts.

BACKGROUND: In nerve allograft development, currently used subjective histological scoring systems to evaluate nerve structure in experimental studies are not uniform and have not been validated. The aim of this study was to describe and validate a simple, fast and inexpensive method to compare structural properties of nerve allografts on a histological level. MATERIALS AND METHODS: A total of 113 histological sections of rat (sciatic nerves) and human peripheral nerve segments (thoracodorsal and long thoracic nerve for motor and sural for sensory nerve) treated with various decellularization protocols were analyzed. Slides were stained with Hematoxylin & Eosin (H&E), Toluidine Blue and Laminin. A subjective scoring system using a 5-point scale was used to score each nerve section by four independent researchers. For validation of this score both inter- and intra-rater reliability were calculated using the Intraclass Correlation Coefficient (ICC). RESULTS: Overall, an excellent correlation between the different observers was found for the Toluidine Blue (ICC = 0.919, 95% CI: 0.891-0.941), H&E (ICC = 0.9, 95% CI: 0.865-0.927) and Laminin staining (ICC = 0.897, 95% CI: 0.860-0.926). Intra-rater reliability was good for all three staining techniques (ICC >0.8). CONCLUSION: This study demonstrates the validity of this simple scoring system to evaluate nerve structure after different processing protocols. All three evaluated staining techniques can reliably be used for both rat and human nerve tissue. We believe this method is suitable to compare different processing techniques to create processed nerve allografts, especially when the average scores of multiple raters are used.

Functional Outcome after Reconstruction of a Long Nerve Gap in Rabbits Using Optimized Decellularized Nerve Allografts.

BACKGROUND: Processed nerve allografts are a promising alternative to nerve autografts, providing an unlimited, readily available supply and avoiding donor-site morbidity and the need for immunosuppression. Currently, clinically available nerve allografts do not provide satisfactory results for motor reconstruction. This study evaluated motor recovery after reconstruction of a long nerve gap using a processed nerve allograft and the influence of storage techniques. METHODS: Nerve allografts were decellularized using elastase and detergents and stored at either 4° or -80°C. In 36 New Zealand White rabbits, a 3-cm peroneal nerve gap was repaired with either an autograft (group 1, control) or a cold-stored (group 2) or frozen-stored (group 3) processed nerve allograft. Nerve recovery was evaluated using longitudinal ultrasound measurements, electrophysiology (compound muscle action potentials), isometric tetanic force, wet muscle weight, and histomorphometry after 24 weeks. RESULTS: Longitudinal ultrasound measurements showed that the cold-stored allograft provided earlier regeneration than the frozen-stored allograft. Furthermore, ultrasound showed significantly inferior recovery in group 3 than in both other groups (p < 0.05). Muscle weight and isometric tetanic force showed similar outcomes in the autograft and cold-stored allograft groups [p = 0.096 (muscle weight) and p = 0.286 (isometric tetanic force)], and confirmed the inferiority of the frozen-stored allograft to the autograft [p < 0.01 (muscle weight) and p = 0.02 (isometric tetanic force)]. CONCLUSIONS: Frozen storage of the nerve allograft significantly impairs functional recovery and should be avoided. The cold-stored optimized nerve allograft yields functional recovery similar to the gold standard autograft in the reconstruction of a 3-cm motor nerve defect. Future studies should focus on further improvement of the nerve allograft.

Gene expression and growth factor analysis in early nerve regeneration following segmental nerve defect reconstruction with a mesenchymal stromal cell-enhanced decellularized nerve allograft.

The purpose of this study was to evaluate the molecular mechanisms underlying nerve repair by a decellularized nerve allograft seeded with adipose-derived mesenchymal stromal cells (MSCs) and compare it to the unseeded allograft and autograft nerve. METHODS: Undifferentiated MSCs were seeded onto decellularized nerve allografts and used to reconstruct a 10 mm gap in a rat sciatic nerve model. Gene expression profiles of genes essential for nerve regeneration and immunohistochemical staining (IHC) for PGP9.5, NGF, RECA-1, and S100 were obtained 2 weeks postoperatively. RESULTS: Semi-quantitative RT-PCR analysis showed that the angiogenic molecule VEGFA was significantly increased in seeded allografts, and transcription factor SOX2 was downregulated in seeded allografts. Seeded grafts showed a significant increase in immunohistochemical markers NGF and RECA-1, when compared with unseeded allografts. CONCLUSIONS: MSCs contributed to the secretion of trophic factors. A beneficial effect of the MSCs on angiogenesis was found when compared with the unseeded nerve allograft, but implanted MSCs did not show evidence of differentiation into Schwann cell-like cells.

Outcomes of Single versus Double Fascicular Nerve Transfers for Restoration of Elbow Flexion in Patients with Brachial Plexus Injuries: A Systematic Review and Meta-Analysis.

BACKGROUND: Elbow flexion after upper brachial plexus injury may be restored by a nerve transfer from the ulnar nerve to the biceps motor branch with an optional nerve transfer from the median nerve to the brachialis motor branch (single and double fascicular nerve transfer). This meta-analysis assesses the effectiveness of both techniques and the added value of additional reinnervation of the brachialis muscle. METHODS: Comprehensive searches were performed identifying studies concerning restoration of elbow flexion through single and double fascicular nerve transfers. Only C5 to C6 lesion patients were included in quantitative analysis to prevent confounding by indication. Primary outcome was the proportion of patients reaching British Medical Research Council elbow flexion grade 3 or greater. Meta-analysis was performed with random effects models. RESULTS: Thirty-five studies were included (n = 688). In quantitative analysis, 29 studies were included (n = 341). After single fascicular nerve transfer, 190 of 207 patients reached Medical Research Council grade 3 or higher (random effects model, 95.6 percent; 95 percent CI, 92.9 to 98.2 percent); and after double fascicular nerve transfer, 128 of 134 patients reached grade 3 or higher (random effects model, 97.5 percent; 95 percent CI, 95.0 to 100 percent; p = 0.301). Significantly more double nerve transfer patients reached grade 4 or greater if preoperative delay was 6 months or less (84 of 101 versus 49 of 51; p = 0.035). CONCLUSIONS: Additional reinnervation of the brachialis muscle did not result in significantly more patients reaching Medical Research Council grade 3 or higher for elbow flexion. Double fascicular nerve transfer may result in more patients reaching grade 4 or higher in patients with a preoperative delay less than 6 months. The median nerve may be preserved or used for another nerve transfer without substantially impairing elbow flexion restoration.

Seeding decellularized nerve allografts with adipose-derived mesenchymal stromal cells: An in vitro analysis of the gene expression and growth factors produced.

Mesenchymal stromal cells (MSCs) secrete many soluble growth factors and have previously been shown to stimulate nerve regeneration. MSC-seeded processed nerve allografts could potentially be a promising method for large segmental motor nerve injuries. Further progress in our understanding of how the functions of MSCs can be leveraged for peripheral nerve repair is required before making clinical translation. The present study, therefore, investigated whether interactions of adipose-derived MSCs with decellularized nerve allografts can improve gene and protein expression of growth factors that may support nerve regeneration. Human nerve allografts (n = 30) were decellularized and seeded with undifferentiated human adipose-derived MSCs. Subsequently, the MSCs and MSC-seeded grafts were isolated on days 3, 7, 14, and 21 in culture for RNA expression analysis by qRT-PCR. Evaluated genes included NGF, BDNF, PTN, GAP43, MBP, PMP22, VEGF, and CD31. Growth factor production was evaluated and quantified using enzyme-linked immunosorbent assay (ELISA). On day 21, semi-quantitative RT-PCR analysis showed that adherence of MSCs to nerve allografts significantly enhances mRNA expression of neurotrophic, angiogenic, endothelial, and myelination markers (e.g., BDNF, VEGF, CD31, and MBP). ELISA results revealed an upregulation of BDNF and reduction of both VEGF and NGF protein levels. This study demonstrates that seeding of undifferentiated adipose-derived MSCs onto processed nerve allografts permits the secretion of neurotrophic and angiogenic factors that can stimulate nerve regeneration. These favorable molecular changes suggest that MSC supplementation of nerve allografts may have potential in improving nerve regeneration.

Collagen Nerve Conduits and Processed Nerve Allografts for the Reconstruction of Digital Nerve Gaps: A Single-Institution Case Series and Review of the Literature.

OBJECTIVE: A single-institution case series is reported and a review of the literature on the outcomes of digital nerve gap reconstruction with the NeuraGen type 1 collagen nerve conduit (Integra Life Sciences, Plainsboro New Jersey, USA) and the Avance Nerve Graft (Axogen Inc., Alachua, Florida, USA) is presented. METHODS: Thirty-seven patients were included with a minimal follow-up of 12 months. Primary outcome was postoperative sensory recovery measured by static 2-point discrimination test or the Semmes-Weinstein monofilament test. Secondary outcome measurements were perioperative or postoperative complications. Final outcome data were stratified to grade results as excellent, good, or poor. RESULTS: The mean nerve gap length was 14 ± 4.9 mm for the collagen conduits versus 18.4 ± 9.3 for nerve allografts. After 12 months, outcomes were graded as excellent sensory recovery in 48% of the collagen conduit repairs and 39% of the nerve allografts (P = 0.608), good in 26% of the conduits and 55% of the allografts (P = 0.074), and poor in 26% of the conduits versus 6% of the allografts (P = 0.091). One neuroma and 1 infection were reported. Graft rejection or extrusion was not observed. CONCLUSIONS: Nerve conduits and processed nerve allografts offer convenient off-the-shelf options for digital nerve gap repair. Both techniques offer effective means of reconstructing a digital nerve gap <2.5 cm at a minimum of 12 months of follow-up. Future prospective randomized large sample size studies comparing nerve conduits with allografts are needed to perform subgroup analyses and to define their exact role in digital nerve injuries.

Bone vascularized composite allotransplantation model in swine tibial defect: Evaluation of surgical angiogenesis and transplant viability.

INTRODUCTION: In prior small animal studies, we maintained vascularized bone allotransplant viability without long-term immunotherapy. Instead, an autogenous neoangiogenic circulation is created from implanted vessels, sufficient to maintain bone viability with only 2 weeks immunosupression. Blood flow is maintained despite rejection of the allogeneic vascular pedicle thereafter. We have previously described a large animal (swine) pre-clinical model, reconstructing tibial defects with vascularized tibial allotransplants. In this manuscript, autologous angiogenesis is evaluated in this model and correlated with bone viability. MATERIALS AND METHODS: Allogeneic tibial segments were transplanted across a major swine leukocyte antigen mismatch. Microvascular repair of the bone VCA pedicle was combined with intraosseous implantation of an autogenous arteriovenous (AV) bundle. The bundle was ligated in group 1 (n = 4), and allowed to perfuse in group 2 (n = 4). Three-drug immunotherapy was given for 2 weeks. At 16 weeks micro-CT angiography quantified neoangiogenic vessel volume. Bone viability, rejection grade, and bone healing were analyzed. RESULTS: A substantial neoangiogenic circulation developed from the implanted AV-bundle in group 2, with vessel density superior to ligated AV-bundle controls (0.11 ± 0.05 vs. 0.01 ± 0.01, P = .029). Bone allotransplant viability was also significantly enhanced by neoangiogenesis (78.7 ± 4.4% vs. 27.7 ± 5.8%, P = .028) with higher bone healing scores (21.4 ± 2.9 vs. 12.5 ± 3.7, P = .029). Ligated control tibias demonstrated disorganized bone morphology and higher local inflammation (P = .143). CONCLUSION: Implantation of autogenous AV bundles into vascularized bone allotransplants resulted in the rapid formation of a neoangiogenic autogenous blood supply in a swine tibia model that maintained bone viability, improved bone healing, and minimized rejection.

In Vivo Survival of Mesenchymal Stromal Cell-Enhanced Decellularized Nerve Grafts for Segmental Peripheral Nerve Reconstruction.

PURPOSE: Adipose-derived mesenchymal stromal cells (MSCs) have emerged as promising tools for peripheral nerve reconstruction. There is a paucity of information regarding the ultimate survivorship of implanted MSCs or whether these cells remain where they are placed. The aim of the present study was to track the in vivo distribution and survival of MSCs seeded on a decellularized nerve allograft reconstruction of a peripheral nerve defect using luciferase-based bioluminescence imaging (BLI). METHODS: To determine the in vivo survivability of MSCs, autologous Lewis rat MSCs were stably labeled with luciferase by lentiviral particles. Labeled cells were dynamically seeded onto a Sprague Dawley decellularized rat nerve allograft and used to bridge a 10-mm sciatic nerve defect. The MSC survival was determined by performing in vivo BLI to detect living cells. Twelve animals were examined at 24 hours after implantation, 3, 7, 9, 11, and 14 days, and at daily intervals thereafter if signals were still present. RESULTS: Labeled MSCs could be detected for up to 29 days. Gradually diminishing BLI signals were observed within the first week following implantation. Implanted MSCs were not detected anywhere other than the site of surgery. CONCLUSIONS: The MSCs seeded on decellularized nerve allografts can survive in vivo but have finite survival after implantation. There was no evidence of migration of MSCs to surrounding tissues. CLINICAL RELEVANCE: The findings support a therapeutic approach that combines MSCs with a biological scaffold for peripheral nerve surgery. It provides understanding of the viability and distribution of implanted MSCs, which is a prerequisite before clinical translation can be considered.

Comparable functional motor outcomes after repair of peripheral nerve injury with an elastase-processed allograft in a rat sciatic nerve model.

BACKGROUND: A bridging nerve autograft is the gold standard for the repair of segmental nerve injury that cannot be repaired directly. However, limited availability and donor site morbidity remain major disadvantages of autografts. Here, a nerve allograft decellularized with elastase was compared with an autograft regarding functional motor outcome in a rat sciatic segmental nerve defect model. Furthermore, the effect of storage on this allograft was studied. METHODS: Sixty-six Lewis rats (250-300 g) underwent a 10-mm sciatic nerve reconstruction using either a cold- (n = 22) or frozen-stored (n = 22) decellularized nerve allograft or an autograft (n = 22). Sprague-Dawley rats (300-350 g) served as full major histocompatibility complex-mismatched donors. Functional motor outcome was evaluated after 12 and 16 weeks. Ankle angle, compound muscle action potential (CMAP), isometric tetanic force, wet muscle weight, and histomorphometry were tested bilaterally. RESULTS: For CMAP and isometric tetanic force, no significant differences were observed between groups. In contrast, for ankle angle, histomorphometry and muscle weight, the cold-stored allograft performed comparable to the autograft, while the frozen-stored allograft performed significantly inferior to the autograft. At week 16, ankle angle was 88.0 ± 3.1% in the cold-stored group, 77.4 ± 3.6% in the frozen-stored group, and 74.1 ± 3.1% in the autograft group (P < .001); At week 16, the muscle weight showed a recovery up to 71.1 ± 4.8% in the autograft group, 67.0 ± 6.6% in the cold-stored group, and 64.7 ± 3.7% in the frozen-stored group (P < .05). CONCLUSIONS: A nerve allograft decellularized with elastase, if stored under the right conditions, results in comparable functional motor outcomes as the gold standard, the autograft.

A Simple Dynamic Strategy to Deliver Stem Cells to Decellularized Nerve Allografts.

BACKGROUND: The addition of adipose-derived mesenchymal stromal cells to decellularized nerve allografts may improve outcomes of nerve reconstruction. Prior techniques used for cell seeding are traumatic to both the mesenchymal stromal cells and nerve graft. An adequate, reliable, and validated cell seeding technique is an essential step for evaluating the translational utility of mesenchymal stromal cell-enhanced decellularized nerve grafts. The purpose of this study was to develop a simple seeding strategy with an optimal seeding duration. METHODS: A dynamic bioreactor was used to seed rat and human mesenchymal stromal cells separately onto rat and human decellularized nerve allografts. Cell viability was evaluated by MTS assays and cellular topology after seeding was determined by scanning electron microscopy. Cell density and distribution were determined by Live/Dead assays and Hoechst staining at four different time points (6, 12, 24, and 72 hours). The validity and reliability of the seeding method were calculated. RESULTS: Cells remained viable at all time points, and mesenchymal stromal cells exhibited exponential growth in the first 12 hours of seeding. Seeding efficiency increased significantly from 79.5 percent at 6 hours to 89.2 percent after 12 hours of seeding (p = 0.004). Both intrarater reliability (r = 0.97) and interrater reliability (r = 0.92) of the technique were high. CONCLUSIONS: This study describes and validates a new method of effectively seeding decellularized nerve allografts with mesenchymal stromal cells. This method is reproducible, distributes cells homogenously over the graft, and does not traumatize the intraneural architecture of the allograft. Use of this validated seeding technique will permit critical comparison of graft outcomes.

Spinal accessory nerve to triceps muscle transfer using long autologous nerve grafts for recovery of elbow extension in traumatic brachial plexus injuries.

Despite continuous improvement and expansion of reconstructive options for traumatic brachial plexus injury, options to reinnervate the triceps muscle remain somewhat sparse. This study describes a novel option, using a spinal accessory nerve transfer to the long head of the triceps muscle with an intervening autologous nerve graft. The resulting quality of elbow extension and factors that influence outcome are discussed.

Evaluate muscle tension using intramuscular pressure device in rabbit tibialis anterior model for improved tendon transfer surgery.

Quantitative evaluation of passive tension in a muscle is important in tendon transfer surgeries, however, currently appropriate intraoperative measurement techniques are lacking. OBJECTIVE: Intramuscular pressure (IMP) is explored as an application to access force. APPROACH: The tibialis anterior (TA) in New Zealand white rabbits (n = 9) was used to test the hypothesis of a strong correlation between the IMP, muscle force, and length. This study also helped to develop intraoperative techniques for future human studies evaluating various insertion techniques (parallel versus perpendicular). MAIN RESULTS: The Pearson correlation between IMP and force for all trials was 0.74 ± 0.30. Separating out the parallel insertion from the perpendicular insertion revealed a significantly higher correlation for parallel, 0.91 ± 0.13 versus 0.56 ± 0.32. SIGNIFICANCE: These data indicate IMP sensors can be used to assess force in a single muscle and the parallel insertion method should be used. New findings • What is the central question of this study? Successful outcomes of tendon and muscle transfers depend on proper muscle tension. A near linear relationship has been seen between muscle force and intramuscular pressure. This study aims to develop an intraoperative technique for assessing passive muscle tension using intramuscular pressure. • What is the main finding and its importance? The findings from this study reveal a high correlation between pressure and passive tension in a single muscle. The techniques developed in this study will allow the translation to a human model. The work will help to improve surgical outcomes and aim to retain muscle strength in the patient following procedures such as tendon and muscle transfers.

Recipient-derived angiogenesis with short term immunosuppression increases bone remodeling in bone vascularized composite allotransplantation: A pilot study in a swine tibial defect model.

Current vascularized composite allotransplantation (VCA) transplantation protocols rely upon life-long immune modulation to maintain tissue perfusion. Alternatively, bone-only VCA viability may be maintained in small animal models using surgical angiogenesis from implanted autogenous vessels to develop a neoangiogenic bone circulation that will not be rejected. This study tests the method's efficacy in a large animal model as a bridge to clinical practice, quantifying the remodeling and mechanical properties of porcine tibial VCAs. A segmental tibial defect was reconstructed in Yucatan miniature swine by transplantation of a matched tibia segment from an immunologically mismatched donor. Microsurgical repair of nutrient vessels was performed in all pigs, with simultaneous intramedullary placement of an autogenous arteriovenous (AV) bundle in Group 2. Group 1 served as a no-angiogenesis control. All received 2 weeks of immunosuppression. After 16 weeks, micro-CT and histomorphometric analyses were used to evaluate healing and remodeling. Axial compression and nanoindentation studies evaluated bone mechanical properties. Micro-CT analysis demonstrated significantly more new bone formation and bone remodeling at the distal allotransplant/recipient junction and on the endosteal surfaces of Group 2 tibias (p = 0.03). Elastic modulus and hardness were not adversely affected by angiogenesis. The combination of 2 weeks of immunosuppression and autogenous AV-bundle implantation within a microsurgically transplanted tibial allotransplant permitted long-term allotransplant survival over the study period of 16 weeks in this large animal model. Angiogenesis increased bone formation and remodeling without adverse mechanical effects. The method may allow future composite-tissue allotransplantation of bone without the risks associated with long-term immunosuppression. © 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 35:1242-1249, 2017.

The learning rate in three dimensional high definition video assisted microvascular anastomosis in a rat model.

Three-dimensional (3D) high definition (HD) video systems are changing microsurgical practice by providing stereoscopic imaging not only for the surgeon and first assistant using the binocular microscope, but also for others involved in the surgery. The purpose of this study was to evaluate the potential to replace the binocular microscope for microarterial anastomoses and assess the rate of learning based on surgeons' experience. Two experienced and two novice microsurgeons performed a total of 88 rat femoral arterial anastomoses: 44 using a 3D HD video device ('Trenion', Carl Zeiss Meditech) and 44, a binocular microscope. We evaluated anastomosis time and modified OSATS scores as well as the subjects' preference for comfort, image adequacy and technical ease. Experienced microsurgeons showed a steep learning curve for anastomosis times with equivalent OSATS scores for both systems. However, prolonged anastomosis times were required when using the novel 3D-HD system rather than direct binocular vision. Comparable learning rates for anastomosis time were demonstrated for novice microsurgeons and modified OSATS scores did not differ between the different viewing technologies. All microsurgeons reported improved comfort for the 3D HD video system but found the image quality of the conventional microscope superior, facilitating technical ease. The present study demonstrates the potential of 3D HD video systems to replace current binocular microscopes, offering qualitatively-equivalent microvascular anastomosis with improved comfort for experienced microsurgeons. However, image quality was rated inferior with the 3D HD system resulting in prolonged anastomosis times. Microsurgical skill acquisition in novice microsurgeons was not influenced by the viewing system used.

Nerve Transfers to Restore Elbow Function.

The purpose of this article is to provide an overview of the various nerve transfer options for restoration of elbow function. This article describes nerve transfer strategies for elbow flexion and extension including the indications, limitations, and expected outcomes based on current literature.

Motor Nerve Recovery in a Rabbit Model: Description and Validation of a Noninvasive Ultrasound Technique.

PURPOSE: To develop and validate a noninvasive ultrasound technique for the longitudinal analysis of functional recovery after segmental peroneal nerve reconstruction in a rabbit model. METHODS: Twelve male New Zealand White rabbits underwent a 1-cm peroneal nerve autograft reconstruction. Ultrasound measurements were performed before surgery and at 1, 2, 4, 8, 12, and 16 weeks postoperatively. All rabbits were managed with manual restraint for the ultrasound procedure, avoiding the risks of anesthetics. At 12 and 16 weeks, we evaluated functional recovery using compound muscle action potential, isometric tetanic force measurements, wet muscle weight, and nerve histomorphometry. Data were compared with ultrasound measurements by calculating the Pearson correlation coefficient. We determined intra-rater and inter-rater reliability of the ultrasound measurements. RESULTS: Ultrasound demonstrated good correlation with isometric tetanic force measurements and wet muscle weight, good correlation with nerve histomorphometry, and moderate correlation with compound muscle action potential. Both intra-rater and inter-rater reliability of the ultrasound technique was excellent. CONCLUSIONS: Ultrasound analysis of the tibialis anterior muscle provided a reliable method for analysis of functional recovery in a rabbit peroneal nerve reconstruction model. The noninvasive nature allowed for longitudinal follow-up within the same animal and measurement of early recovery without the use of anesthesia. CLINICAL RELEVANCE: Application of this noninvasive technique can reduce the variability and sample size necessary in peripheral nerve reconstruction studies and may provide an ideal tool for comparative studies in larger animal models.

Noninvasive Ultrasound of the Tibial Muscle for Longitudinal Analysis of Nerve Regeneration in Rats.

BACKGROUND: Today's criterion standards for measuring functional recovery after nerve trauma in experimental studies are the muscle mass ratio and the isometric tetanic force; both tests are invasive and require a sacrificial procedure. The authors propose ultrasound as a noninvasive method to determine muscle atrophy, and evaluate its validity and reliability by comparing it to muscle mass ratio, isometric tetanic force, and histology. METHODS: Fifty rats sustained a 10-mm autograft sciatic nerve reconstruction. With a 2-week interval, five animals were tested with a total follow-up of 20 weeks. The functional recovery of the hind-limb muscles was measured with ultrasound, muscle mass ratio, and isometric tetanic force. In addition, neuromuscular junctions were analyzed histologically. The different evaluation techniques were compared and the reliability of the ultrasound was determined. RESULTS: Four weeks after denervation, extensive muscle atrophy resulted in a decrease of muscle mass up to 30 percent. Ultrasound showed good correlations with muscle mass ratio for both tibial (r = 0.85) and gastrocnemius muscles (r = 0.89). Both intrarater reliability (r = 0.97) and interrater reliability (r = 0.88) of the ultrasound were high. The correlation with force was lower (0.62) but still statistically significant. CONCLUSIONS: Ultrasound measurement of muscle atrophy was highly correlated with the criterion standard muscle mass ratio and was also significantly correlated with isometric tetanic force. Histologic evaluation confirmed the regeneration pattern observed with ultrasound. The authors propose that ultrasound can be used as a valid alternative to muscle mass ratio to study muscle atrophy after nerve injury in a less-invasive and more animal-friendly manner.