Dennis L. Powers

Efficacy of hyaluronic acid/nonsteroidal anti-inflammatory drug systems in preventing postsurgical tendon adhesions

Tendon adhesion is acknowledged to be a function of both an overwhelming inflammatory response at the surgical site and the loss of physical separation that is normally present between the tendons and the synovial sheath. Adhesions bind the flexor tendons to each other and to surrounding structures, interfering with their normal gliding function. The clinical result of adhesion formation following flexor tendon surgery is poor digital function. This study investigated the effect of intraoperative treatments of high viscosity absorbable gels made of various combinations of hyaluronic acid and nonsteroidal anti-inflammatory drugs, on adhesion formation in a leghorn chicken flexor tendon model. Forty-eight mature, white leghorn chickens were used to verify the surgical model and to test five different gel treatments. The gels were formed from: 2% sodium hyaluronate in phosphate buffered saline alone or combined with 1 mg/mL tolmetin sodium; 1 mg/mL naproxen sodium; 0.216 g/mL calcium acetate; or 0.216 g/mL calcium acetate plus 1 mg/mL naproxen sodium. The gels were applied by injecting 0.2 mL of the specified composition into the intrasheath space near the conclusion of the surgical procedure. Gross and histological evaluations were conducted to analyze the efficacy. All of the treatments significant reduced the extent and severity of postsurgical tendon adhesion in this animal model as compared with the control (no gel treatment) (p < 0.05). The combination of naproxen sodium and calcium acetate in a high viscosity sodium hyaluronate carrier was the most effective composition. The combination of a high viscosity gel and nonsteroidal anti-inflammatory drugs appears to maintain the natural separation between the tendons and their sheaths and decrease the tissue inflammatory response through mediating two of the major stimuli in adhesion formation.

THE PREVENTION OF PROSTHETIC INFECTION USING A CROSS-LINKED ALBUMIN COATING IN A RABBIT MODEL

We evaluated the effects of a serum protein coating on prosthetic infection in 29 adult male rabbits divided into three groups: control, albumin-coated and uncoated. We used 34 grit-blasted, commercially pure titanium implants. Eleven were coated with cross-linked albumin. All the implants were exposed to a suspension of Staphylococcus epidermidis before implantation. Our findings showed that albumin-coated implants had a much lower infection rate (27%) than the uncoated implants (62%). This may be a useful method of reducing the infection of prostheses.

Fixation of osteotomies using bioabsorbable screws in the canine femur.

The purpose of this study was to compare the healing properties of femoral osteotomies fixed by bioabsorbable screws (20:80 polyglycolic copolylactic acid copolymer) to standard stainless steel screws of a similar design in a dog femoral model. Two osteotomies were used, the trephine osteotomy (10 mm diameter) in the metaphyseal lateral femoral condyle and in the femoral diaphysis, and a unilateral osteotomy in the lateral femoral condyle. Two months after the trephine osteotomies, the femurs that contained the polymer screws were not significantly different in mechanical strength from the femurs treated with the stainless steel screws, either in the diaphyseal or metaphyseal model. There was no histological difference in bone healing between the metallic and polymer screws for all periods (2, 9, and 17 months). There was no adverse inflammatory response to the polymeric or metallic screws. By month 17, the polymer screws were resorbed completely. All the diaphyseal screw tracks had healed with bone and areas of remodeling were evident in two specimens. For the femoral condyle osteotomy model at 2 months, the polymer screws were present and intact, and all osteotomies healed with no evidence of inflammation. By 9 months, only one specimen had polymeric material left in the screw track. At 15 months, the screw tracks still were present but no evidence of any polymer remained. The tracks were filled with fibrous and adipose connective tissue. All osteotomies stabilized with either bioabsorbable polymer screws or stainless steel screws did heal satisfactorily without any complications, inflammation, or osteolysis. The polyglycolic polylactic acid copolymer may have a clinical role as a bioabsorbable material without the concerns for the osteolysis, foreign body reaction, and sterile abscess formation that have occurred with bioabsorbable fixation methods in the past.

Hydroxyapatite composites designed for antibiotic drug delivery and bone reconstruction: a caprine model.

The purpose of this study was to investigate the feasibility of fabricating a drug delivery system that serves a dual function, to eradicate infection as well as to provide a scaffold for osseous integration. Two porous composite systems were fabricated using hydroxyapatite (HA) as the carrier for gentamicin sulfate (GS), an aminoglycoside antibiotic. Structural and mechanical properties of porous HA-GS composites were characterized and the in vitro release behavior of GS from fabricated composites was monitored and compared with the well-known polymethylmethacrylate (PMMA)-GS delivery system. Scanning electron microscopy revealed a macropore range of 150 to 200 microm and 100 to 190 microm for the sintered and unsintered HA-GS composites, respectively. The effect of GS inclusion on bone apposition and ingrowth was assessed using a caprine model. Plugs 10 mm x 6 mm of cylindrical tricalcium phosphate, sintered HA, and sintered HA-GS were implanted in the femoral diaphysis for a period of 6 weeks. Data collected during the in vitro study showed that GS can successfully be incorporated into HA and used as a drug delivery system to eradicate Staphylococcus aureus. In vivo data confirmed that the inclusion of GS within a ceramic matrix did not stimulate or inhibit osteointegration or bone apposition. In conclusion, the fabricated sintered HA-GS composite may be beneficial in the treatment of infected osseous sites as a drug delivery system.

The effect of site of implantation and animal age on properties of polydioxanone pins

Absorbable polymeric orthopaedic pins (Orthosorb) of 2.0 mm diameter were implanted at different sites in mature (3.5 kg, > 5 months) and immature (5 weeks old) rabbits (total 36) for 2, 4, and 5 weeks. The sites of implantation were the medullary canal of the femur, muscles of the thigh and subcutaneous tissue of the dorsum. In mature rabbits, 1.3 mm diameter pins were also implanted in the medullary canal of the femur. The shear strength of the pins harvested from the rabbits, was measured at each time period using a fixture that shears the pins into three parts symmetrically about the load axis. In both mature and immature rabbits the rate of degradation in mechanical properties was higher in the medullary canal of bone than in the muscle and in the subcutaneous tissue (p < 0.05). The strength retention was lower in immature than in mature rabbits after 4 and 5 weeks. The 1.3-mm pins had higher initial strength (174.7 +/- 7 MPa), higher strength retention and slower degradation within the medullary canal of femur of mature rabbits as compared to the 2.0-mm pins (157.5 +/- 4.8). DSC and X-ray diffraction results of control and implanted pins showed higher initial crystallinity and a wider range of crystallite size in the 1.3-mm pins. After 5 weeks in vivo, the crystallinity increased indicating degradation within the amorphous phase. The smaller crystallites underwent recrystallization to form larger crystallites. The results indicate that site of implantation and age of recipient influence the degradation and associated effects on mechanical properties of absorbable implants. The size of the implant, though important in determining its properties, should be considered in association with its microstructure, which also plays an important role in determining strength and strength retention of absorbable polymeric systems.

The Immature Goat as an Animal Model for Legg-Calvé-Perthes Disease

Legg-Calvé-Perthes disease (LCPD) results from avascular necrosis of the capital femoral epiphysis in growing children. This disease often yields a significant deformity of the proximal femur, which may result in osteoarthritis. Its cause is unknown, although extensive radiographic, clinical, and histologic evaluations have been performed. Attempts at developing an animal model for LCPD have been unsuccessful. Previous models have been based predominantly on determining the vascular etiology of the disease. There is a need for an animal model that mimics the growth pattern of the proximal femur seen in LCPD. Such a model would allow for the development and testing of new treatments. Thus far, no treatment strategy has been completely successful. A study involving graphic analysis of radiographs found that arrested anterolateral physeal growth with continued or accelerated perichondrial ring and posteromedial epiphyseal growth would account for the most severe morphologic changes observed in the femoral heads of patients with LCPD. A surgical procedure was performed to ablate the capital femoral physis in goat kids in an attempt to mimic the changes noted in this study. The procedure was evaluated with radiographs, gross specimens, and histopathologic slides. Graphic analysis of the radiographs revealed changes in the shape of the operated femoral head compared with the unoperated femoral head. While bone, fibrous, and fibrocartilaginous bridges were histologically observed across the physis, the resultant deformities did not mimic the changes identified in the graphic analysis study, perhaps because of inconsistencies in the surgical ablative techniques, which will require further modification. This study provides the basis for further research to develop a successful model.

Glutaraldehyde-Cross-linked Meniscal Allografts: Clinical, Gross, and Histological Results

Osteoarthritic changes in the knee are often a late result of total meniscectomy. In cases of total resection, availability of a prosthetic meniscus might limit development of these changes. The objective of this research was to evaluate a glutaraldehyde-cross-linked medial meniscus as a morphologically and biologically compatible prosthesis in a canine model. Medial and lateral menisci were harvested from donor dogs, frozen in saline, and cross-linked with glutaraldehyde. Five host animals were selected and matched with donors. Glutaraldehyde-cross-linked medial menisci were implanted bilaterally in the stifle joints and one glutaraldehyde cross-linked lateral meniscus was implanted subcutaneously. Clinical results showed asymptomatic limb and joint usage during the 12 postoperative weeks. Gross and histological evaluations indicated acceptable biocompatibility. The subcutaneous implants were encapsulated with a thin fibrous tissue capsule that was only mildly inflamed. Within the joints, the anterior attachment and periphery were maintained in position by their sutures; however, there was dehiscence of the posterior suture in all cases. The articulating surfaces of the implants were intact. There was an initial loss in the quantity of proteoglycans following glutaraldehyde treatment, with significant recovery after implantation into the joints. There were significant degenerative changes (loss of proteoglycans and fibrillation) in the articular cartilage on the femoral condyle and tibial plateau most likely a result of the posterior attachment failure. It was concluded that glutaraldehyde-cross-linked meniscal allografts showed an acceptable degree of histocompatibility. However, failure of the posterior attachment interfered with testing the efficacy of the prosthesis.(ABSTRACT TRUNCATED AT 250 WORDS)

Glutaraldehyde-cross-linked meniscal allografts: mechanical properties.

Removal of a severely damaged medial meniscus has been shown to lead to degradation of the articular cartilage and formation of degenerative arthritis. To counter this degenerative effect, meniscal prostheses, including glutaraldehyde-cross-linked allografts, have been evaluated in dogs. The purpose of this research was to quantify the mechanical properties of both fresh and glutaraldehyde-cross-linked canine medial menisci. Mechanical properties quantified were tensile strength, tensile modulus, and compressive stiffness. In addition, water content of compressive test samples was measured. Analysis of variance showed significantly lower tensile strength and tensile modulus and significantly higher compressive stiffness for the glutaraldehyde-cross-linked menisci, as compared to fresh specimens. Measurement of the weight percentage of water in fresh and cross-linked samples revealed no significant differences in water content. When implanted into a joint, the increased compressive stiffness could increase the peripheral tensile load. Due to the decreased tensile strength in this region, the prosthetic meniscus could be susceptible to peripheral tears.

Growth Profiles and Articular Cartilage Characterization in a Goat Model of Legg-Calvé-Perthes Disease

Numerous animal species, including the goat, have been evaluated as potential models for human Legg-Calve-Perthes disease (LCPD). These models disrupt the vasculature of the femoral head, causing it to collapse, and therefore do not mimic all the clinical patterns of the human disease. Baseline data regarding the weight and femoral length in the growing goat are not available. This study characterized the goats normal growth for comparison with that of humans. The growth aberrations in the proximal femur created by surgically ablating the capital physis were described and compared with the aberrations observed in human LCPD cases. Age, weight, and femoral length (test and control) data were obtained for goats approximately 1 to 14 months of age. At 4 months of age, a craniolateral surgical approach was used to expose the cranial lateral capital physis so that it could be cauterized. Postoperative radiographs were evaluated by graphic analysis to assess the resultant changes in the morphology of the proximal femur. The articular cartilage of the femoral head and acetabulum was evaluated mechanically, using indentation testing, to determine the apparent modulus of elasticity, and histopathologically regarding its thickness and proteoglycan content. The proximal femurs of goats and humans exhibit similar morphology and growth patterns. There was a positive correlation between age, weight, and femoral lengths in the goat. The surgical procedure was effective in ablating the capital femoral physis as indicated by shorter femoral lengths and fragmented, flattened, and mushroomed femoral heads. The histopathological data revealed that the articular cartilage was significantly thicker in the operated hip joints at the ventrocaudal and cranial acetabula and the dorsal and ventral femoral heads. The test cartilage exhibited significantly less positive staining for proteoglycans in the dorsocaudal and the cranial acetabula as well as the ventral femoral head. The apparent modulus of elasticity, of the test cartilage was significantly lower than the control value at the dorsocaudal acetabulum. These data show that the surgical procedure produced morphological changes that mimic those in human LCPD. The increased thickness of the articular cartilage of the LCPD femoral head may account for the articular degeneration observed in older patients with LCPD, as increased cartilage thickness is associated with decreased tissue quality.

Anatomical Reconstruction of the Anterior Cruciate Ligament in Goats

A surgical procedure was developed for the implantation of an anatomical, two-banded anterior cruciate ligament (ACL) prosthesis. Prostheses were fabricated of braided long-chain polyethylene fibers. The left ACL of adult male goats was surgically excised and replaced with either an anatomical reconstruction (5 goats) or a conventional reconstruction (5 goats). The anatomical reconstruction required drilling four bone tunnels, two each in the femur and tibia. Each band of the prosthesis was placed through one tunnel in the femur and the corresponding tunnel in the tibia, recreating the anteromedial and posterolateral bands. The two bands were tensioned independently and stapled in place. In the conventional procedure, the prosthesis was doubled and placed through two larger tunnels, one in the femur and one in the tibia, tensioned and stapled together. All animals were terminated 3 months after surgery. Clinical evaluation of passive range of motion, antero-posterior laxity and the appearance of the joint space showed little or no difference between the reconstruction methods. The ultimate failure load for the natural (unoperated) ACL was 1691 +/- 210 N, while the anatomical and conventional reconstruction groups had mean ultimate failure loads of 1233 +/- 732 and 1012 +/- 220 N, respectively. The elongation to failure of all groups was similar: the natural ACL group = 7.1 +/- 2.8 mm, the anatomical group = 7.2 +/- 2.9 mm, and the conventional group = 7.7 +/- 3.9 mm. The slope of the load-deformation curve, or stiffness, was significantly higher for the natural ACL (4.53 +/- 1.24 x 10(5) N/m) than for either of the reconstruction methods (2.75 +/- 1.59 x 10(5) N/m for the anatomical and 2.34 +/- 0.60 x 10(5) N/m for the conventional). The energy to failure, or area under the load-deformation curve, showed no significant difference between groups. In conclusion, both types of reconstructions were less strong, stiff, and tough than the natural ACL. There was no significant difference observed between the anatomical and conventional reconstruction methods over the 3-month implantation time in either clinical evaluation or mechanical testing. Therefore, at 3 months postsurgery, the anatomical reconstruction technique was considered no better and no worse than the conventional reconstruction technique.