Search

Objectives: The purpose of the current study was to demonstrate the efficacy of in situ tissue engineering of the cricoid and trachea in a canine model. Methods: Marlex mesh tube reinforced with polypropylene threads and covered by collagen sponge was used as a tissue scaffold for airway regeneration in 9 beagle dogs. The anterior half of the cricoid cartilage was resected in 5 dogs, whereas the cricoid cartilage and cervical trachea were simultaneously resected in 4 dogs. The tissue scaffold was implanted into the resultant defect. Results: Endoscopic examination showed no airway obstruction for a postoperative period of 3 to 40 months in all dogs. Granulation tissue was observed in 2 dogs, and slight mesh exposure in 1 dog, although all were asymptomatic. Light microscopy and electron microscopy showed the endolaryngeal and endotracheal lumen to be covered by ciliated epithelium. According to strain-force measurement, the framework was firmly supported by regenerated tissue, as well as the normal cricoid and trachea. Conclusions: Our current tissue scaffold provides a rigid framework for the airway, and the collagen coating invites tissue regrowth around the tube. This study presents the possibility of successful reconstruction of the cricoid and trachea with epithelial regeneration by means of in situ tissue engineering.

Bronchoscopic lung volume reduction (BLVR) for severe emphysema is less invasive than lung volume reduction surgery. Fibroblast growth factor-2 (FGF-2) has been reported to enhance fibrogenesis and angiogenesis. The aim of this study was to investigate the feasibility of BLVR with the FGF-2, and ability to reduce lung volume and promote recovery of lung function. The BLVR based on FGF-2 is less invasive than surgical procedures, and can be performed repeatedly if the effectiveness of volume reduction is inadequate. This simple bronchoscopic approach allows selective reduction in the volume of the emphysematous parenchyma, and intratracheal administration of FGF-2 induces an increase in pulmonary blood flow, thus allowing recovery of pulmonary function.

Study objectives Fibroblast growth factor (FGF)-2 is one of the most powerful angiogenic growth factors to be evaluated as an agent for the promotion of angiogenesis. The aim of this study is to investigate whether intratracheal administration of controlled-release FGF-2 microspheres restores pulmonary function in beagle dogs with emphysema Design Randomized, controlled, experimental animal study Subjects Eighteen Wister rats and 15 adult beagle dogs Methods In the rat study, we compared the time profiles of the radioactivity remaining after intratracheal injection of 125 I-labeled FGF-2, either incorporated with the controlled-release microspheres or as an aqueous solution. In the dog study, elastase-induced emphysema models were developed in 10 animals, classified into the following three groups: control group (n = 5), emphysema model with empty microspheres-treated group (FGF − group, n = 5), and emphysema model with FGF-2 containing microspheres-treated group (FGF + group, n = 5) Results In the rat study, controlled-release microspheres maintained higher whole-lung FGF-2 concentrations after intratracheal administration. In the dog study, Pao 2 in the FGF + group was significantly higher than in the FGF − group after treatment. Pulmonary perfusion dynamic MRI revealed significant improvement in the signal intensity of damaged lung with the FGF + group. Linear intercept of the FGF + group was significantly reduced than the FGF − group Conclusion Results indicate that intratracheal administration of FGF-2 induced an increase in pulmonary blood flow in the damaged lung and led to recovery of pulmonary function. The controlled-release microsphere system increased the effectiveness of FGF-2

PURPOSE: To create an experimental model of aortic dissection with a long-lasting patent false lumen as a proper animal model for development of less-invasive treatment for aortic dissection. MATERIALS AND METHODS: Fifteen adult beagle dogs (weight, 10–12 kg) were used. The descending aorta was exposed by a left thoracotomy at the sixth intercostal space. The entry for the aortic dissection was created surgically just distal to the origin of the left innominate artery and the reentry was 5 cm distal to the entry point. Normal saline solution was injected into the aortic wall (ie, media) between these two points to create the dissection. The dogs were followed up at 1 day, 3 months, 1 year, and 2 years. RESULTS: All 12 surviving dogs had completely patent true and false lumina without any thrombi. Microscopic examination showed that the dissection was created in the tunica media layer, making it identical to aortic dissection in humans. Color Doppler imaging confirmed the patency of the true and false lumina and the relatively narrowed true lumen. CONCLUSION: In this canine model of aortic dissection, the false lumen has excellent long-term patency and the dissection plane is histologically similar to that in human aortic dissection. This model may contribute to the development of new treatments for Stanford type B aortic dissection.

Primary pulmonary hypertension (PPH) is still a refractory disease, and patients deteriorate despite any treatment. We hypothesized that neovascularization in the lung could increase the volume of the vascular bed in the pulmonary circulation and thus reduce the development of pulmonary hypertension (PH). Endothelial progenitor cells (EPCs) might be a potential cell source for neovascularization. We examined the effects of EPC transplantation into the lungs of dogs with dehydromonocrotaline-induced PH. The lung parenchyma of PH model dogs was injected with ex vivo-expanded, autologous EPCs originated from peripheral blood (experiments, n=4) or culture medium (control, n=3), using a bronchoscope. EPC transplantation gave significant improvements in mean pulmonary artery pressure, cardiac output, and pulmonary vascular resistance. Histological evaluation revealed both improvement in the medial thickness of the small pulmonary artery and neovascularization of the lung tissue. These results indicate that EPC transplantation into the lung is effective at preventing the progression of dehydromonocrotaline-induced PH in dogs, and suggest a new therapeutic option for PPH.

The feasibility of an in situ tissue-engineering method employing cell-based therapy with autologous periodontal ligament-derived cells was investigated. Periodontal ligament cells were obtained from six beagle dogs. Periodontal fenestration defects (6 x 4 mm) were created bilaterally at a location 6 mm apical to the marginal alveolar crest in the maxillary canines. Alkaline phosphatase-positive periodontal ligament cells (3 x 10(5) cells) were seeded onto a collagen sponge scaffold just before implantation. One defect was filled with the cell-scaffold construct, and another was left empty as the control. All animals were killed 4 weeks after surgery, and specimens were evaluated histomorphometrically. All the histomorphometrical data were analyzed by three-way analysis of variance with the Bonferroni multiple comparisons test. Regeneration of apical tissue was faster than that of coronal and isolated tissues on the control side (apical > coronal > isolated; p < 0.0001). On the other hand, on the cell-seeded side, regeneration of the cementum was observed uniformly on the root surface. Our data suggest that the seeded cells induced cementum regeneration on the root surface, indicating the potential of in situ tissue engineering using autologous cells for the regeneration of periodontal tissues.

The aim of this study was to regenerate the injured vocal fold by means of selective cultured autologous mesenchymal stem cells (MSCs). Eight adult beagle dogs were used for this experiment. Selective incubation of MSCs from bone marrow was done. These MSCs were submitted to 3-dimensional incubation in 1% hydrochloric acid atelocollagen. Three-dimensional incubated MSCs were injected into the left vocal fold, and atelocollagen only was injected into the right vocal fold of the same dog as a control. Four days after injection, the posterior parts of the vocal folds were incised. The regeneration of the vocal fold was estimated by morphological and histologic evaluations. Our results showed that 3-dimensional incubated MSCs were useful in the regeneration of the injured vocal fold. This study shows that damaged tissues such as an injured vocal fold would be able to be regenerated by tissue engineering.

We studied capillary endothelial injury, as demonstrated by fibrin deposition and changes in thrombomodulin (Tm) expression, in rats receiving continuous or bolus iv lipopolysaccharide (LPS). Rats were continuously infused with iv LPS (0.1, 0.2, 0.5, or 1.0 mg/kg/hr) for up to 6 hours. Others were given a bolus iv dose of LPS (20 mg/kg), and then the same dose of saline as a continuous infusion was administered for up to 3 hours. Harvested lungs, livers, and kidneys were examined immunohistochemically for thrombomodulin expression and fibrin deposition. Tm expression began to diminish dose- and time-dependently in lung, liver, and renal peritubular capillaries within 2 to 4 hours of the start of continuous LPS administration (1.0 mg/kg/hr) and had completely disappeared by 3 hours, although Tm remained in the glomerulus. The amount of fibrin deposition observed varied with the organ, dose, and duration of treatment in rats that received continuous LPS administration, but little was deposited in the lung. After bolus LPS administration, Tm in the endothelia of lung, liver, and peritubular capillaries diminished 20 to 40 minutes after treatment and then recovered 120 to 180 minutes after treatment, but the Tm activity of the glomerulus did not change. Fibrin deposition in the capillaries was observed in the liver, glomerulus, and peritubular capillaries, but not in the lung. Endothelial injury by LPS administration is dependent on the dose of LPS and the duration of treatment. The amount of fibrin deposition differs among organs and with the duration of contact between the endothelium and the endotoxin.

BACKGROUND A previous report has noted that tracheal allotransplantation is possible using allografts without immunosuppressants. In the present study, we investigated whether blood typing is necessary in immunosuppressant-free tracheal allotransplantation in dogs. METHODS In group 1, the blood types of paired dogs were mismatched (n=6, three pairs), while the blood types of paired dogs were matched in group 2 (n=6, three pairs). In paired dogs, an intrathoracic 5-ring tracheal segment was exchanged. No immunosuppressants were used. RESULTS All animals survived for more than 8 weeks. In each group, different degrees of airway stenosis were observed in four dogs. However, no stenosis was observed in the other two dogs. Histologically, no vascular rejection was detected. Ciliated columnar epithelium covered the inner surface of the implanted graft in the long-term survival cases. CONCLUSIONS In immunosuppressant-free tracheal allotransplantation, no hyperacute rejection occurred in any recipient dog whose blood type was mismatched with the donor. Blood typing was not the key factor to successful immunosuppressant-free tracheal allotransplantation. Immunosuppressant-free tracheal allotransplantation may be possible using mismatched blood type tracheal grafts.

The recurrent laryngeal nerve (RLN) does not regenerate well after it has been cut, and no current surgical methods achieve functional regeneration. Here, we evaluate the functional regeneration of the RLN after reconstruction using a biodegradable nerve conduit or an autologous nerve graft. The nerve conduit was made of a polyglycolic acid (PGA) tube coated with collagen. A 10-mm gap in the resected nerve was bridged by a PGA tube in 6 adult beagle dogs (group 1) and by an autologous nerve graft in 3 dogs (group 2). Fiberscopic observation revealed functional regeneration of the RLN in 4 of the 6 dogs in group 1. No regeneration of the RLN was observed in any dog in group 2. We also tested for axonal transport, and measured the compound muscle action potential. The RLN can be functionally regenerated with a PGA tube, which may act as a scaffold for the growth of regenerating axons.

We investigated whether saber sheath-type tracheomalacia could be treated by the slow release of bone morphogenetic protein (BMP)-2 from a gelatin sponge. A 1 cm gap was made in the middle portion of each of 10 consecutive tracheal cartilage rings in the canine cervix (control group, n = 3), then a gelatin sponge containing 12 microg of BMP-2 solution was implanted in the gap (12 microg group, n = 3). In another group (120 microg + P group, n = 3), the implanted gelatin sponge contained 120 microg of BMP-2 solution, and the gap was covered with periosteum. All of the control dogs developed saber sheath-type tracheomalacia, whereas tracheomalacia was not observed in the 12 microg and 120 microg + P groups. In the 12 microg group, fibrous cartilage was observed at the ends of the cartilage stumps. In the 120 microg + P group, newly formed bone and cartilage were observed to form a bridge between the cartilage stumps. The regeneration of cartilage or bone induced by the slow release of BMP-2 from a gelatin sponge might be useful for treatment of tracheomalacia.

Background/Purpose: Circular myotomy often has been used to elongate the esophagus in end-to-end anastomosis for esophageal atresia with a long gap. Defects of the muscular layer often have led to the ballooning of the esophagus during the postoperative course. Recently, successful regeneration of the neo-esophagus using collagen sponge scaffold (CSS) for self-tissue regeneration has been reported. The authors applied this CSS to the muscular defect after circular myotomy in animal experiments. Methods: Myectomy or myotomy of the cervical or thoracic esophagus was performed in 11 piglets. CSS was applied to the amuscular site, which was 20 to 30 mm in length, to examine the regeneration of the muscular layer in 8 piglets (CSS group). The amuscular site was not wrapped by CSS in 3 piglets as a control (control group). Two months after operation, the animals were killed, and their operated esophagi were examined macroscopically and microscopically. Results: All 11 piglets increased in weight normally without any sign of dysphagia or vomiting. The amuscular site onto which CSS was implanted was covered with loose connective tissue, but muscle regeneration was not found microscopically. The amuscular site, which had been wrapped by CSS, was not distended in any of the 8 piglets when inflated, and ballooning was observed in all 3 of the control group. Conclusion: The results suggested that application of CSS to the site of circular myotomy for esophageal atresia with a long gap might not induce muscle regeneration but could prevent postoperative ballooning by inducing connective tissue growth. J Pediatr Surg 37:1409-1413. Copyright 2002, Elsevier Science (USA). All rights reserved.

The present investigation examined the key factors in the preparation of low-antigenic tracheal allografts by detergent treatment in dogs. In group 1 (n = 5), the grafts were treated by detergent at room temperature for 48 h and rinsed with running water until use. In group 2 (n = 4), detergent treatment was performed at 4 degrees C for 48 h, but the rinsing step was omitted. In group 3 (n = 6), the grafts were treated at 4 degrees C for 48 h, rinsed thoroughly with physiologic saline, and stored at 4 degrees C. The grafts were then used for tracheal replacement without immunosuppressant in dogs. The epithelium and mixed glands had been removed completely from the grafts in groups 1 and 2, and in five of the six grafts of group 3. In groups 1 and 2, cartilage viability appeared to have been lost and all animals died of airway stenosis within 38 days. In group 3, the chondrocytes were viable and all animals survived uneventfully. These results suggest that maintaining cartilage viability in a tracheal graft is necessary for successful immunosuppressant-free allotransplantation and, consequently, to maintain an open airway. The treatment temperature and sufficient rinsing are key factors for maintaining cartilage viability of grafts.

Recently, we have demonstrated canine peroneal nerve regeneration with functional recovery across an 80 mm gap using a polyglycolic acid (PGA) -collagen tube filled with laminin coated collagen fibers. In that study, the laminin coating was applied before a dehydrothermal (DHT) treatment designed to extend preservation of laminin in situ. To address concerns that the biological activity of laminin might consequently be reduced, the present investigation examined the influences of DHT crosslinking on the activity of laminin in terms of neural cell growth in vitro. DHT crosslinking was performed on collagen (type I or IV) spread on glass in three groups: (1) before coating, (2) after coating, and (3) both before and after coating. PC-12 cells were disseminated in each of the three groups. All three groups were cultured, and the number of cells were compared statistically. Cell growth achieved through application of laminin coating after DHT crosslinking was statistically greater than that achieved when laminin coating was performed before crosslinking. A reduction in laminin activity induced by DHT crosslinking was demonstrated. The optimal timing for the crosslinking of biornaterials treated with trophic factors such as-laminin should be examined in terms of the effects of crosslinking on the activity of the trophic factors.

Bone density is inherited as a complex polygenic trait. Previously, we identified two quantitative trait loci (QTLs) specifying the peak relative bone mass (bone mass corrected by bone size) on chromosomes (Chrs) 11 and 13 by interval mapping in two mouse strains: SAMP2 and SAMP6. The latter strain is an established murine model of senile osteoporosis and exhibits a significantly lower peak relative bone mass than SAMP2 mice. In this study, we report the effects of the Chr 13 QTL on peak bone density (Pbd2). First, we constructed a congenic strain P6.P2-Pbd2b, which carried a single genomic interval from the Chr 13 of SAMP2 on an SAMP6-derived osteoporotic background, to dissect this polygenic trait into single gene factors. This congenic strain had a higher bone density than the background strain using three measurement methods with different principles for bone density. Next, we measured the peak relative bone mass of the AKR/J strain and the 13 senescence-accelerated mouse (SAM) strains, which are considered to be a series of recombinant-like inbred (RI) strains derived from the AKR/J strain and other unspecified strains. We then determined the microsatellite marker haplotypes of these strains around the Pbd2 locus, in which three strains with a high relative bone mass shared the same haplotype over the 26-centimorgan (cM) region. In the Pbd2 locus, a high relative bone mass was associated with alleles of the unknown strain, whereas a low relative bone mass was associated with the alleles from the AKR/J strain. These results confirmed the existence of a Pbd2 locus regulating bone density in the SAM strains.

A novel artificial nerve conduit was developed and its efficiency was evaluated on the basis of promotion of peripheral nerve regeneration across an 80-mm gap in dogs. The nerve conduit was made of a polyglycolic acid-collagen tube filled with laminin-soaked collagen sponge. Conduits filled with either sponge- or fiber-form collagen were implanted into an 80-mm gap of the peroneal nerve (five dogs for each form). Twelve months postoperatively nerve regeneration was superior in the sponge group both morphometrically (percentage of neural tissue: fiber: 39.7 +/- 5.2, sponge: 43.0 +/- 4.5, n=3) and electrophysiologically (fiber: CMAP 1.06 +/- 0.077, SEP 1.32 +/- 0.127 sponge: CMAP 1.04 +/- 0.106, SEP 1.24 +/- 0.197, n=5), although these differences were not statistically significant. The observed regeneration was complementary to successful results reported previously in the same model, in which collagen fibers exclusively were used. The results indicate a possible superiority of collagen sponge over collagen fibers as filling materials. In addition, the mass-producibility, superior scaffolding potential, and capacity for gradual release of soluble factors of the sponge provide make it an attractive alternative to fine fibers, which are both technologically difficult and costly to produce. This newly developed nerve conduit has the potential to enhance peripheral nerve regeneration across longer gaps commonly encountered in clinical settings.

Two types of flat plates made from a polyvinyl alcohol (PVA) hydrogel with a water content of 80 and 20 (PVA-H80, PVA-H20), 20 × 10 × 1 mm in size, were subcutaneously implanted into each of 50 young, male Wistar rats. As a control, a sham operation was done on another set of 50 rats (Sham Op group). The shape and transparency of the PVA hydrogel were unchanged for up to 24 months. Tumors arose in 14 rats from the PVA-H80 group. In the PVA-H20 group, tumors appeared in 15 rats. The average tumor latency was 598 ± 109 days in the PVA-H80 and 637 ± 94 days in the PVA-H20. There was no difference in tumor incidence between the PVA-H20 and PVA-H80 groups (p < 0.05). In the Sham Op group, no malignant tumors appeared. Histopathologically, the tumors induced by hydrogel plates were malignant tumors resembling fibrosarcoma or malignant fibrous histiocytoma. This indicates that PVA hydrogel implants also induce solid state carcinogenesis at a similarly high rate to medical grade hydrophobic material reported in a previous study. © 2001 John Wiley & Sons, Inc. J Biomed Mater Res 56: 289–296, 2001

Background Poor healing of the sternum often limits the use of bilateral internal thoracic arteries (BITAs) after coronary bypass surgery in diabetic patients. We have reported that a gelatin sheet that incorporates basic fibroblast growth factor (bFGF) accelerates sternal healing after BITA removal in normal rats. This study evaluated the effects of the above method for sternal healing in diabetic animals. Methods and Results Diabetic Wistar rats with blood glucose levels >400 mg/dL and body-weight loss >20 g were established by a single intravenous injection of streptozotocin (55 mg/kg). After median sternotomy and BITA removal, 16 diabetic rats received either a gelatin sheet that incorporated bFGF (100 μg/sheet) on the posterior table of the sternum (FGF group, n=9) or no gelatin sheet (control, n=7). Peristernal blood flow, as measured by a noncontact laser Doppler 4 weeks after surgery in the FGF group, recovered to the preoperative level (106±10% versus 82±9%, P P P P Conclusions A gelatin sheet that incorporates bFGF may offset sternal ischemia and accelerate sternal bone regeneration and healing, even in diabetic patients.

Background . We have shown that a gelatin sheet incorporating basic fibroblast growth factor enhanced bone regeneration of the devascularized sternum. The purpose of this study was to determine if topical use of the gelatin sheet accelerated normal sternal regeneration and bone remodeling. Methods . Thirty Wistar rats had median sternotomy and were divided into 3 groups: 10 had the bilateral internal thoracic arteries removed and basic fibroblast growth factor sheet applied on the sternum (group A), 10 had just the bilateral internal thoracic arteries removed (group B), and 10 had intact bilateral internal thoracic arteries (group C). Results . Four weeks later the peristernal blood flow significantly increased and marked angiogenesis was seen around the sternum in group A. Histologically, the sternum was almost completely healed only in group A. In group A the bone mineral content was highest, but the bone mineral density was similar to that in other groups. The osteoclast index in group A was highest at the border zone of bone formation and remained high in regenerated bone. Conclusions . The basic fibroblast growth factor sheet offset sternal ischemia and accelerated normal sternal bone regeneration and remodeling, not only by callus formation but also by callus resorption.

There is now considerable evidence that peripheral nerves have the potential to regenerate if an appropriate microenvironment is provided. However, there are only a few reports of the successful use of artificial nerve conduits to repair major nerve defects more than 30 mm in length. In this study, we examined nerve regeneration across a long gap in the dog peroneal nerve using a novel artificial nerve conduit developed by our group. The conduit consists of a polyglycolic acid (PGA) collagen tube filled with laminin coated collagen fibers. In 12 dogs, the nerve conduit was implanted across an 80 mm gap in the left peroneal nerve. Three months after surgery, compound muscle action potentials (CMAPs) and somatosensory evoked potentials (SEPs) were detected. Evaluation of locomotor function revealed obvious limping for up to 3 months, but no marked difficulty in walking by 6 months. Microscopic observation of the regenerated nerve segment at 12 months showed numerous myelinated nerve fibers, which were smaller in diameter and enclosed in a thinner myelin sheath than normal axons. These results suggest that our artificial nerve conduit has potential usefulness in enhancing peripheral nerve regeneration, even across large gaps.

Object. Facial nerve paralysis due to a surgical procedure or trauma is a frequently observed complication. The authors evaluated facial nerve repair achieved by the interposition of a collagen nerve guide.Methods. Ten cats were divided into three groups. Group 1 consisted of six animals in which a 5-mm facial nerve segment on one side was resected and replaced by a collagen tube that was sutured to bridge both nerve stumps. On the opposite side a 5-mm segment of facial nerve was resected, reversed 180°, and sutured to the stumps as an autograft nerve. Group 2 consisted of two cats in which the collagen nerve guide was interposed on one side and the nerve on the other side was left intact. Group 3 consisted of two cats in which a reversed autograft nerve was placed on one side and the nerve on the other side was left intact. Histological, electrophysiological, and horseradish peroxidase labeling examinations were performed starting 3 weeks after surgery.Light and electron microscopic examinations of collagen tube—implanted specimens revealed a well-vascularized regenerated nerve. The electrophysiological study confirmed the recovery of electrical activity in regenerated axons. Horseradish peroxidase labeling also confirmed restoration of the whole facial nerve tract.Conclusions. The collagen nerve guide shows great promise as a nerve conduit.

Carinal reconstruction has the highest operative mortality rate in pulmonary surgery, and high tension at the anastomoses is one of the reasons. We have reported previously that a new Y-shaped artificial trachea gives acceptable short-term results. Here, we report our long-term observations and histologic findings. A Y-shaped Marlex mesh tube reinforced by a polypropylene spiral was coated with collagen, which was extracted from porcine skin, to increase the biocompatibility and airtightness of the prosthesis. Replacement of the carina was performed in 20 beagle dogs. The prosthesis was covered by the omental pedicle in all dogs. Six of the 20 dogs survived past 6 months, and 4 were in good condition when they were killed. The other 13 dogs died within 11 days of the operation. The main cause of death was air leakage from the prosthesis. Bronchoscopic examination of the surviving dogs 12 months after surgery revealed that the prosthesis was covered with regenerated tissue, and neither stenosis nor dehiscence was recognized; 15 months after implantation microscopic examination showed the growth of ciliated epithelial cells over the lumen of the prosthesis. These results suggest that our Y-shaped artificial trachea might be useful for reconstruction of the carina with tissue regrowth, although some improvement of the prosthesis is necessary.

We have developed a new adhesive for surgical use. The new adhesive is made of three components: porcine collagen, poly(L-glutamic acid) and water-soluble carbodiimides (WSC). The optimum concentration of each component was determined by measuring the time required for gel formation in experiments in vitro. Using these optimum concentrations, we applied the adhesive to wounds made on rats. A conventional fibrin glue was used as a control. Measurement of tensile strength and histological examination were performed 5, 7, 10, and 14 days after the operation. The tensile strength of wounds treated with 2.5 mg/mL collagen glue was not significantly different from that of wounds treated with fibrin glue except at 7 days after the operation (p < 0.05 by Student's t-test). Histological examination revealed that the speed of cell infiltration into, and absorption of 2.5 mg/mL collagen glue was slower than for fibrin glue, but faster than for 5.0 mg/mL collagen glue. One of the important advantages of our collagen glue is that the absorption rate of it can be controlled by the collagen concentration. Therefore, it seems to be adequate for sealing air leakage from the lung, which takes a relatively long period for recovery. Moreover it does not contain human serum, and, hence, it requires no blood donation and can be obtained with low cost.

Background . Human amnion has useful biomedical applications because it contains a large amount of human collagen fibers. We prepared purified human collagen membrane (HCM) from human amnion and used it to develop a new sheet by combining it with synthetic bioabsorbable polyglycolic acid (PGA) mesh. We evaluated its efficacy in preventing air leakage from the lungs of dogs. Methods . In 20 dogs, HCM-PGA sheet (n = 5), sheets using fibrin glue with a separate application method (n = 5), a mixed application method (n = 5), and fibrin glue alone (n = 5), were used as dressing materials after partial lung resection. Results . The HCM-PGA sheet using fibrin glue with a separate application method was shown to be significantly more effective by an air leakage pressure test than the other three methods. These results indicate that the HCM-PGA sheet is useful for preventing air leakage from the lung. Conclusions . The HCM-PGA sheet is more effective than conventional fibrin glue for controlling postoperative air leakage.

Objectives: Intrathoracic esophageal replacement with an artificial esophagus is considered difficult. We attempted to replace the intrathoracic esophagus with an artificial esophagus composed of a collagen sponge with a double-layered silicone tube and examined the state of host tissue regeneration. Methods: A 5-cm long gap was created in the intrathoracic esophagus in 9 dogs and repaired by interposition of our prosthesis. The dogs were fed only by intravenous hyperalimentation for 28 days. The silicone tube was removed at 29 days after the operation, and oral feeding was reintroduced. Results: One dog was put to death at each of the following times: 1, 2, 3, 3, 6, 12, and 24 months after the operation. One dog is still surviving without problems after more than 26 months. One dog died of malnutrition at 10 months. In all dogs, the host regenerated tissue had replaced the resulting gap at the time of silicone tube removal. The mucosa had fully regenerated within 3 months and the glands within 12 months. The process of stenosis and shrinkage was complete within 3 months and did not advance thereafter. The lamina muscularis mucosae were observed as islets of smooth muscle within 12 months. Although the skeletal muscle regenerated close to the anastomoses, it did not extend to the middle of the regenerated esophagus even after 24 months. Conclusions: Use of a collagen sponge with a double-layered silicone tube was shown to be feasible even in the thorax and to allow the regenerated host tissue, consisting of the mucosa, glands, and lamina muscularis mucosae, to replace the esophageal gap. (J Thorac Cardiovasc Surg 1999;118:276-86)

We manufactured an esophageal prosthesis made of a collagen sponge supported by an inner silicone tube. The collagen was derived from porcine skin and was crosslinked by dehydrothermal treatment. We implanted our prosthesis in nine dogs after resecting a 5 cm length of the thoracic esophagus. Intravenous hyperalimentation was performed for 4 weeks after the surgery. After 29 days, the inner silicone tube was removed endoscopically and oral feeding was restarted immediately. All nine dogs survived more than 1 month, and three of the nine dogs survived for more than 1 year. Endoscopically, the luminal surface of the regenerated esophagus was covered with a lustrous, smooth mucosa, but stenosis was observed in the midregion of the regenerated esophagus. Microscopically, no foreign material remained at the reconstructed site 1 month after surgery; the collagen sponge was absorbed and replaced by host tissue. Complete epithelization was observed within 3 months. Regeneration of the lamina muscularis mucosae, esophageal glands, and skeletal muscle were observed near the anastomoses. Our artificial esophagus holds promise for esophageal reconstruction in a clinical setting, even in the thorax.

We cloned a full-length cDNA encoding a novel mouse protein, A-C2, by differential display method using mouse embryonic fibroblast C3H10T1/2 cells and mouse chondroprogenitor-like EC cells, ATDC5. The deduced amino acid sequence of A-C2 consisted of 106 amino acids with no significant homology to the sequences previously reported. Northern blot analysis showed two major bands of 2.1 and 1.8 kb sizes. Expression of A-C2 mRNA was exclusive to ATDC5 cells at their undifferentiated stage. None of ATDC5 cells at their differentiated stage and adult mice tissues examined expressed A-C2 gene.

An ideal drug delivery system (DDS) releases an appropriate drug at specific locations and times. We tried to create a new antibiotic delivery system that releases gentamicin only when wounds are infected by Pseudomonas aeruginosa (P.A.). Exudate from the dorsal pouch of rats infected with P.A. showed significantly higher hydrolytic activity-thrombin-like activity-toward Boc-Val-Pro-Arg-MCA than exudate from noninfected wounds. We therefore constructed a device for controlled release of an antimicrobial drug triggered by thrombin-like activity. Briefly, gentamicin was bound to a polyvinyl alcohol derivative (PVA) hydrogel through a newly developed peptide linker cleavable by the proteinase, PVA-(linker)-gentamicin. In vitro experiments showed that proteinases from wounds infected with P.A. cleaved the linker and gentamicin was released while the exudate from noninfected wounds had no hydrolytic activity toward the linker. This device shows potential as an occlusive dressing with an effective antibiotic delivery system for treating infected wounds.

Objective: Attempts have been made to replace esophageal defects with a variety of artificial materials. However, because of the artificial nature of the materials, problems such as infection, leakage, stricture, or dislocation could not be avoided. Therefore we have designed a new type of artificial esophagus that is gradually replaced by host tissue. Methods: Our artificial esophagus was a two-layered tube consisting of a collagen sponge matrix and an inner silicone stent. We used it to replace 5 cm esophageal segmental defects in 43 dogs, and the inner silicone stent was removed endoscopically at weekly intervals from 2 to 4 weeks. Results: In the 27 dogs from which the silicone stent was removed at 2 or 3 weeks, constriction of the regenerated esophagus progressed and the dogs became unable to swallow within 6 months. In the 16 dogs from which the silicone stent was removed at 4 weeks, highly regenerated esophageal tissue successfully replaced the defect, leaving no foreign body in the host. Moreover, the regenerated esophagi had stratified flattened epithelia, striated muscle tissue composed of an inner circular and an outer longitudinal muscle layer, and esophageal glands. Conclusions: Even in mature adult higher mammals, esophageal high-order structures can be regenerated provided that an adequate three-dimensional extracellular structure is put in place for a sufficient period. (J Thorac Cardiovasc Surg 1998;116:98-106)

Reconstructions of the intrathoracic trachea in 24 dogs were done with the use of 50 mm long collagen-conjugated tracheal prostheses. Omental wrapping was also done in 14 of the dogs (omentopexy group) to evaluate the efficacy of this option in comparison with results in the other 10 dogs (control group). All 24 dogs had uneventful postoperative courses and were killed at 4 weeks or 3, 6, or 12 months after the operation. Better epithelialization and fewer complications, such as mesh exposure and luminal stenosis, were observed in the omentopexy group than in the control group. Angiography and analysis of regenerated blood vessels revealed that vessel ingrowth had started within 4 weeks and that vessel formation reached its maximal point within 6 to 12 months in the omentopexy group. In contrast, revascularization of the subepithelial region in the control group was poor even after 3 months, and vessel formation continued for as long as 12 months. The differences between the two groups were considered to be mainly a result of the speed of blood vessel ingrowth into the regenerated mucosa. We conclude that our prosthesis can be used safely for intrathoracic tracheal reconstruction and that omental wrapping is a useful supplementary method that reduces the occurrence of complications. (J Thorac Cardiovac Surg 1997;113:691-700)

Artificial esophagi designed thus far can be classified into three types in terms of the materials used: natural, artificial, and composite. In conventional models, even when artificial esophagi were made of ideal materials with high tissue affinity, they remained in the tissue as a foreign body, and therefore were not free of the complications caused by implanted material. The authors have designed a new type of artificial esophagus composed of a Silicone tube covered with nonantigenic collagen. The novel feature of this artificial esophagus is that the prosthesis does not remain in the implanted site, but is replaced by regenerated host tissue. Using this artificial esophagus, the authors have already succeeded in replacing a 5 cm gap in the esophagus. In this study, replacement of longer portions of the esophagus was assessed in seven dogs using a 10 cm long artificial esophagus. Stenosis did not occur in five of the seven dogs and, consequently, these dogs survived by oral feeding alone for more than 6 months without dry weight loss. The other two animals died of anesthetic accidents at the time of stent removal 6 weeks after surgery. In both cases, the internal surface of the neoesophagus was covered with a polylayer of squamous epithelium. Regenerated esophagi had normal esophageal glands and immature muscle tissue. It is therefore concluded that this new artificial esophagus is also applicable for replacement of long segments of esophagus.

We have designed a new tracheal prosthesis to overcome problems with an earlier device, which included stenosis and exposure of its constituent mesh. A polypropylene mesh cylinder, reinforced with a polypropylene spiral, is sealed with collagen sponge made from porcine dermal collagen. Using this prosthesis, we performed cervical tracheal reconstructions on 11 dogs. Three dogs died within 3 months of reconstruction. Their causes of death were anesthetic accident, diarrhea, and suffocation, respectively. Bronchoscopically, the inner surface of the prosthesis was almost covered with host tissue by 2 months. However, in one dog, a relatively large area of the mesh was exposed in the tracheal lumen at 6 months; a smaller area was exposed in two other dogs. The appearance of the inner surface rapidly became lustrous, and central stenosis was not significant, even after 12 months. Histologically, an incomplete epithelial lining at 6 months was seen on the reconstructed surface and included ciliated columnar, cuboidal, and squamous epithelium. These observations have revealed that this prosthesis has high biocompatibility and the potential to overcome problems of stenosis of the prosthesis lumen. However, because ulceration persists with this prosthesis, additional improvement is needed to reduce the incidence of mesh exposure.

Poly-L-lactide (PLLA) plates, 20 × 10 × 1 mm in size, with flat surfaces, were implanted into 50 young male Wistar rats subcutaneously, and the incidence of tumors for 2 years under specific pathogen-free (SPF) conditions was evaluated. Tumors arose in 22 rats (at the implant sites in 20, and distant from the implantation site in 2). As a control, plates of medical-grade polyethylene of the same shape were implanted into 50 rats. Tumors appeared in 23 of these rats (21 at the sites of the plates, and 2 ectopically). In 30 rats given sham operations, no tumors appeared in the 2 year observation period. Histologically, all the tumors were mesenchymal malignant tumors resembling fibrosarcoma or malignant fibrous histocytoma (MFH) in humans. In 6 of the PLLA tumors and 3 of the PE tumors, bone formation was observed. There was no significant difference between the incidence of tumors caused by PLLA plates and that of tumors caused by medical-grade polyethylene plates. Macroscopically, PLLA plates retained their initial shape for 2 years. Scanning electron microscopy revealed small holes on the surfaces of the plates, and their molecular weight was reduced to 8% of the initial value. © 1994 John Wiley & Sons, Inc.

The authors studied the efficacy of a new tracheal prosthesis made from mesh. The prosthesis, 50 mm long and 18-22 mm in diameter, is made from a Marlex mesh cylinder reinforced with a continuous polypropylene spiral that is grafted and coated with porcine collagen to increase its biocompatibility and provide an airtight seal during the initial implantation stage. Circumferential surgical resection and replacement of a seven to nine ring segment of the cervical trachea was performed in 20 adult mongrel dogs. At the time of surgery, a silicone tube was inserted into the tracheal prosthetic lumen to promote secretory transportation until the prosthesis was covered with host tissue. The silicone tube was removed during fiber bronchoscopy 1 month after surgery. With the exception of the prostheses in 3 dogs that died of unrelated causes, all were infiltrated by connective tissue and incorporated completely by the host. One of these 17 dogs died of suffocation caused by luminal stenosis 2.5 months after surgery, but the others survived until they were killed at more than 6 months. The luminal surfaces of the reconstructed tracheae were covered with respiratory epithelium to varying degrees, and in one dog killed at 22 months after surgery, confluent epithelization throughout the length of the prosthesis was confirmed histologically. In eight dogs, prosthetic luminal stenosis occurred because of overgrowth of granulation tissue, which generally was mild in all but three dogs. The authors conclude that this tracheal prosthesis is highly biocompatible and shows promise for clinical application.

This study investigated ways of preventing late stenosis after replacing the esophagus with a prosthetic device and increasing survival time in an animal model. In a previous study, the authors induced neoesophageal formation and re-epithelialization at an early stage, although stenosis in the chronic stage remained a problem, and long-term survival was prevented. The authors developed an artificial esophagus made from a collagen-silicone composite tube. The silicone tube (25 mm in diameter, 50 mm long, and 1 mm thick) was covered with collagen sponge (5 mm thick). The collagen sponge was intended to be replaced by host tissue, leading to neoesophageal regeneration. The authors thought stenosis was caused primarily by poor regeneration of submucosal tissue, rather than reepithelialization. Thus, they examined the grade of stenosis relative to the time the stent became dislodged. They concluded that the occurrence of stenosis after anastomosis depends upon the duration of stenting and that stenosis did not develop when the portion replaced by the artificial esophagus was stented for at least 4 weeks.

A new artificial esophagus with a bilayered structure made of porous collagen sponge and silicone was studied. The concept of this study was not to replace an esophageal defect permanently with prosthesis but to promote tissue regeneration by collagen. Five centimeters of cervical esophagus were replaced by this artificial esophagus in 19 adult mongrel dogs. Two weeks after implantation, the collagen sponge was replaced by autologous tissue and regeneration of the "neoesophagus" was observed in all animals. The inner surface of the neoesophagus was covered with mature mucosal epithelium similar to the intact esophagus 4-5 weeks after implantation. The replacement site was not complicated by infection, anastomotic leakage, or exuberant granulation tissue development on the luminal surface. In a long-term survival study, animals showed moderate-grade stenosis but could take normal feed orally and with no clinical problems observed.

• A new method of dynamic reconstruction by suturing the cricothyroid (anticus) and the lateral cricoarytenoid (lateralis) muscles was devised for the treatment of unilateral recurrent laryngeal nerve paralysis. Eighteen adult dogs underwent this surgical procedure after sectioning a recurrent laryngeal nerve. Immediate postoperative observations revealed a marked increase in tension of the paralyzed vocal cords and remobilization to the median line of 17 of 18 dogs. More than 4 months later, fiberscopic examination revealed a satisfactory adduction of the cords. Electrophysiologic and histologic examination showed reinnervation of the affected lateralis muscle by the external branch of the superior laryngeal nerve. The clinical applications of this surgery may serve as a dynamic reconstruction for unilateral recurrent laryngeal nerve paralysis. (Arch Otolaryngol Head Neck Surg.1993;119:498-503

The effects on gas exchange of superimposition of high-frequency oscillation (HFO) (40 Hz) on conventional mechanical ventilation were investigated in mongrel dogs with eucapnic gas exchange on conventional mechanical ventilation (CMV). The dogs were anesthetized, paralyzed, and ventilated with CMV until stable. Oscillation was then superimposed for 15 min, followed by CMV alone for a further 30 min. During HFO superimposed on CMV (CMV-HFO), the arterial carbon dioxide tension (PaCO2) increased from 43.6 +/- 1.2 mm Hg to 47.2 +/- 1.4 mm Hg (p less than 0.02), whereas the arterial oxygen tension (PaO2) did not change at all. The change was inhibited completely by administration of intravenous cromolyn sodium (CS) (6 mg/kg/min). The mean pulmonary arterial pressure (mPAP), cardiac output (CO), pulmonary capillary wedge pressure (PCWP), and pulmonary vascular resistance (PVR) did not change during the experiment. These results demonstrate that CMV-HFO appears to cause CO2 accumulation and eliminates the impaired O2 transfer, and that these effects are inhibited completely by CS administration.

An attempt was made to develop an airway stent for nonsurgical transluminal implantation using a bronchofiberscope. The stent is composed of a single wire with a diameter of 0.5 mm. The wire is made of thermal shape memory titanium nickel alloy, with a transition temperature of 25-30 degrees C. To avoid direct contact between the metal and the tissue, the wire is covered with a 300 microns thick coating of silicone. The stent is horseshoe shaped in cross-section. It is 15 mm in outer transverse diameter and 40 mm long. Ten stents were implanted in 10 dogs whose tracheal cartilages had been previously broken. The stents were first cooled in ice water to reduce their diameter and then inserted into the target site. The wire was warmed to body temperature and recovered its initial shape within 10 sec. The dogs were killed for examination between 1 week and 6 months after implantation. All the stents were located at the implantation sites and were patent. Tissue reactions, such as ulceration and granulation formation, were less severe than in those with previously implanted non-silicone covered stents. Microscopic observation showed that the wires became gradually covered with epithelium within 2 months. This transluminal technique for preserving airway patency shows promise for clinical application.

Synthetic biomaterials have been developed and used for bone grafting. Here, we developed a biodegradable sponge composite for bone tissue engineering by combining beta-tricalcium phosphate (beta-TCP) and collagen. In addition, we sought to determine the optimal beta-TCP granules/collagen ratio by evaluating and bone formation in vivo. Porous beta-TCP granules were mixed with atelocollagen hydrochloride solution at various ratios--0.02, 0.05, 0.1, and 0.2 g/mL. The resultant mixtures were freeze-dried and subjected to dehydrothermal treatment in vacuo. The final composites obtained were designated beta-TCP/collagen sponge composites (beta-TCP/CS). Through compression testing, it was found that the stress values for beta-TCP/CS (0.2 g/mL) were higher than those of the other three composites over the whole strain range. Histological evaluation at four weeks after implantation revealed that the collagen sponge had degraded and newly formed bone was present on the surface of the beta-TCP granules. At 12 weeks, the beta-TCP granules were completely degraded and remodeling of the lamellar bone was observed.

In a previously reported attempt to regenerate small intestine with autologous tissues, collagen scaffolds were used without cell seeding or with autologous mesenchymal stem cell seeding. However the regenerated intestine lacked a smooth muscle layer. To accomplish regeneration of a smooth muscle layer, this present study used collagen scaffolds seeded with the smooth muscle cells (SMC) in a canine model. Autologous SMC were isolated from stomach wall and cultured. Two types of scaffolds were fabricated: in SMC (+), cultured SMCs were mixed with collagen solution and poured into a collagen sponge; and in SMC (-), SMCs were omitted. Both scaffolds were implanted into defects of isolated ileum as a patch graft. Animals were euthanized at 4, 8, and 12 weeks; for the last time point, the ileal loop had been reanastomosed at 8 weeks. At 12 weeks, the SMC (-) group showed a luminal surface covered by a regenerated epithelial cell layer with very short villi; however only a thin smooth muscle layer was observed, representing the muscularis mucosae. In the SMC (+) group, the luminal surface was covered completely by a relatively well-developed epithelial layer with numerous villi. Implanted SMCs were seen in the lamina propria and formed a smooth muscle layer. Thus, we concluded that collagen sponge scaffolds seeded with autologous SMCs have a potential for small intestine regeneration.

The development of a novel biocomposite of apatite (Ap) and collagen incorporating low-level additions of silicon (Si) as an osseopromotive agent is detailed. Designed to mimic the structural and compositional characteristics of developing bone, this composite is produced via a coprecipitation method, through which the weight percentage of Ap (i.e., the Ap/collagen ratio) can be varied. Coprecipitates produced at Ap contents of 80 wt % (Ap/collagen = 4:1), 60 wt % (Ap/collagen = 3:2), and 40 wt % (Ap/collagen = 2:3) Ap showed markedly different morphologies, ranging from ceramic-like particulates to rope-like macro-fibrils; at all three Ap contents, however, the nanostructural features of the composites remained qualitatively indistinguishable, with equiaxed Ap nanocrystals distributed randomly throughout a matrix of amorphous collagen. Si incorporation was observed to occur preferentially in the collagenous phase—a result with potential impact on local controlled release of Si. © 2005 Wiley Periodicals, Inc. J Biomed Mater Res, 2005

The present report details the successful development of a model for spinal cord injury (SCI). This model is simple, reproducible, and requires no laminectomy. Development of the model was carried out using fourteen dogs. A balloon catheter was inserted into the extradural space via the intervertebral foramen of each dog, then the balloon was inflated at the L1 level by injection of saline. Six dogs underwent compression with a balloon volume of 1.5 ml, three dogs with a volume of 1.0 ml, and the remaining five dogs were used as uninjured controls. We applied the Basso, Beattie, and Bresnahan (BBB) locomotor rating scale to the dogs. Compression of the spinal cord for 10 min at 1.5 ml produced severe paraplegia (BBB remained zero or one for 6 months following surgery), while compression for the same time interval at 1.0 ml produced moderate paraplegia. Electrophysiological tests showed no hindlimb movement upon stimulation cranial to the site of injury in the 1.5-ml group. The volume of abnormal-intensity lesions in the 1.0-ml group calculated using MR imaging showed no marked changes in either high- or low-intensity lesions after 3 months, whereas in the 1.5-ml group, the low-intensity lesions alone showed a marked increase. Pathological examination of the damaged spinal cord showed the formation of cavities surrounded by scar tissue containing high levels of collagen. These findings closely resembled those of clinical cases. It was concluded that 10 min of balloon compression with a volume of 1.5 ml caused irreversible paraplegia in dogs.

The purpose of the present study was to evaluate the efficacy of cricoid regeneration via in situ tissue engineering in a canine larynx for the treatment of subglottic stenosis. As the tissue scaffold, a Marlex mesh tube coated by collagen sponge was used for a rigid airway framework and for tissue regrowth around the tube. On 5 dogs, the larynx was exposed and the anterior third of the cricoid cartilage was resected. The tube was anastomosed to the lower edge of the thyroid cartilage and to the first tracheal cartilage. By postoperative endoscopic examination at 3 to 7 months, no airway obstruction was observed in any of the dogs. There was granulation tissue in 2 dogs and slight mesh exposure in 1 dog, but they were asymptomatic. Confluent regeneration of the epithelium over the scaffold and good incorporation of the scaffold mesh into the host tissue were observed after surgery.

We have developed a bioabsorbable polyglycolic acid (PGA) tube filled with collagen sponge (PGA-collagen tube) as a nerve connective guide, and compared its effectiveness with that of autograft in terms of nerve regeneration across a gap. The PGA-collagen tube was implanted into 24 beagle dogs across a 15-mm gap in the left peroneal nerve. The right peroneal nerve was reconstructed with the autograft harvested from the left side, as a control. After the surgery, the connective tissue extended from both cut ends in the PGA-collagen tube and connected again at the center. Pathologically, the collagen sponge in the tube provided adequate scaffolding for nerve tissue extension, and the nerve tissue reconnected within 3 weeks. Electrophysiologically, muscle-evoked potentials (MEPs) and compound nerve action potentials (CNAPs) were detected 18 days after the surgery. For up to 6 months postsurgery, CNAPs and somatosensory-evoked potentials (SEPs) on the PGA-collagen side had a shorter latency and larger peak voltage than those on the autograft side. The myelinated axons on the PGA side were larger in diameter than those on the autograft side. It is suggested that the PGA-collagen tube has the potential to be an effective alternative to conventional autografting for the repair of some peripheral nerve defects.

Mastoid is a pneumatic bone, composed of small interconnecting chambers covered by a mono-layer of mucosa with an abundant blood supply. One of its main functions is gas exchange according to the concentration/pressure gradient. The final goal of our research project is to regenerate mastoid air cells and their unique physiologic functions. The aim of the present study is to determine appropriate cultivating conditions for the cells cultured on the surface of artificial hydroxyapatite. In our in vitro experiment, to imitate the skeleton of mastoid bone, we used two types of three-dimensional hydroxyapatite (3D-HA), i.e. with a high (90%) and low (60%) percentage of micropores. The former type was divided into two groups: collagen-coated and non-coated. Canine mucosal- and bone marrow-derived stromal cells (BSCs), from the oral floor and femur respectively, were harvested and cultured on the 3D-HA under different conditions. To estimate the proliferation/distribution of the cultured cells over the surface of the 3D-HA, these cells were stained with the dye DiI and hematoxylin-eosin. There were no significant differences in the proliferation of cultured cells on the 3D-HA with high and low percentages of micropores. Collagen-coated HA was a better material for the cultured cells compared with the non-coated HA. Co-cultured mucosal and BSCs proliferated better than those cultured separately. In conclusion, this tissue engineering technique may be applied for the regeneration of mastoid air cells.

Objectives We investigated whether bone morphogenetic protein 2, released slowly from a gelatin sponge, could induce cartilage regeneration in a canine model of tracheomalacia and evaluated the long-term results. Methods A 1 × 5-cm gap was made in the anterior cervical trachea by removing 5-cm long strips of 10 sequential cartilagines. In the control group (n = 5), the gaps were left untreated. In the gelatin sponge group (n = 5), a gelatin sponge soaked in a buffer solution was implanted in each defect. In the bone morphogenetic protein group (n = 5), a gelatin sponge soaked in a buffer solution containing 12 μg bone morphogenetic protein 2 was implanted in each defect. Results Tracheomalacia was observed in the control and gelatin sponge groups but not in the bone morphogenetic protein group. No regenerated cartilage was detected in the control or gelatin sponge groups, even 6 months after surgery. In contrast, regenerated cartilage, which had developed from the host perichondrium, was observed around the stumps of the resected cartilagines in the bone morphogenetic protein group. This regenerated cartilage maintained the integrity of the internal lumen for longer than 6 months. A compressive fracture test revealed that the tracheal cartilage in the bone morphogenetic protein group was significantly more stable than that in the gelatin sponge and control groups ( P = .0015 and P = .0001, respectively). Conclusions In this canine model of tracheomalacia, cartilage regeneration was induced around the stumps of tracheal cartilagines by bone morphogenetic protein 2 released slowly from a gelatin sponge. This regenerated cartilage was not reabsorbed for longer than 6 months and was strong enough to maintain the integrity of the internal lumen of the trachea.

Purpose To examine the feasibility of using sterilized, freeze-dried amniotic membrane (FD-AM) as a substrate for cultivating autologous corneal epithelial cells for ocular surface reconstruction. Methods Human AM deprived of amniotic epithelial cells by incubation with EDTA was freeze dried, vacuum packed, and sterilized with gamma-irradiation. The resultant FD-AM was characterized for its physical, biological, and morphologic properties by stretch stress tests, immunohistochemistry, electron microscopy, and cell culture. In addition, 3 weeks after an ocular surface injury, the conjunctivalized corneal surfaces of eyes in eight rabbits were surgically reconstructed by transplantation of autologous cultivated corneal epithelial cells on FD-AM. Results A stretch stress test revealed no significant differences between sterilized FD-AM and cryopreserved AM. Immunohistochemistry for several extracellular matrix molecules and electron microscopic analysis of FD-AM revealed that the process of drying and irradiation did not affect its biological and morphologic properties. The corneal epithelial cells cultivated on FD-AM had four to five stratified, well-differentiated cell layers. Corneas that were grafted with the cultivated corneal epithelial cells on FD-AM were clear and were all epithelialized at 10 days after surgery. Conclusions The sterilized, freeze-dried AM retained most of the physical, biological, and morphologic characteristics of cryopreserved AM; consequently, it is a useful biomaterial for ocular surface reconstruction.

The rate of biodegradation of new types of polyglycolic acid (PGA)-collagen composite tubes for nerve regeneration was evaluated in the peritoneal cavity. PGA mesh tubes with a diameter of 2 or 4 mm were coated with collagen solution and dried at room temperature. The tubes were then subjected to dehydrothermal treatment (composite tube). A 2 mm PGA-collagen composite tube filled with collagen sponge was also investigated in this study (sponge tube). Tubes with a length of 15 mm were fixed at the parietal peritoneum of BALB/c mice and excised 2 weeks and 1, 2, and 3 months after the operation. The inner areas of the excised tubes were measured microscopically. Statistical analysis was performed by one way ANOVA and Fisher's PLSD test. Although the inner areas of the 2 and 4 mm composite tubes were not maintained 1 month after the operation (62 +/- 6.8% and 21 +/- 3.8%, respectively), they were well maintained in the sponge tubes (83 +/- 6.4%). The inner areas of the sponge tubes were significantly larger than those of the composite tubes until 2 months after surgery. These results suggest that sponge tubes are more suitable than composite tubes for nerve regeneration in the peritoneal cavity.