Harold K. Dunn

A Prospective Evaluation of Untreated Acute Grade III Acromioclavicular Separations

We prospectively studied the natural history of untreated acute grade III acromioclavicular separations. Twenty-five patients were treated nonoperatively with a sling for comfort through progressive early range of motion as tolerated. Ten additional uninjured subjects underwent strength testing to evaluate the difference between the dominant and nondominant sides so that patient data could be standardized. The patients were examined at intervals of 6, 12, 24, 36, and 52 weeks after injury, at which time they completed a subjective questionnaire and underwent isometric dynamometer testing as well as military press and bench press strength testing. One patient underwent a surgical procedure at 2 weeks after injury because of cosmetic concerns. Twenty of the 25 patients completed the 1-year evaluation and strength-testing protocol. Subjectively, 4 of the 20 patients (20%) thought that their long-term outcome was suboptimal, although for 3 of them it was not enough to warrant surgery. Objective examination and strength testing of the 20 patients revealed no limitation of shoulder motion in the injured extremity and no difference between sides in rotational shoulder muscle strength. The bench press was the only strength test that showed a significant short-term difference, with the injured extremity being an average of 17% weaker. This study documents the natural history of patients with an untreated acute grade III acromioclavicular separation and provides a reference with which to judge all other proposed methods of treatment.

Effects of flexural rigidity of plates on bone healing.

The effect of bone-plate flexural rigidity on the strength of healing bone was explored. Plates with similar shapes and a range of flexural rigidities (from 1.4 to 16.8 Nm2) were prepared from fiberpolymer composites or selected from among commercially available metal plates and applied to the anterolateral surface of osteotomized femora in dogs. At sixteen weeks, the plated femora and contralateral control femora were removed Bone specimens (composed of cortex and callus) were removed from the area beneath the plate and from the opposite side of the same femur as well as from the anterobateral surface of the contralateral control femur. The strengths of these specimens were determined in bending tests. Bone strength increased with decreasing rigidity of the plate, and the effects were more marked in the specimens taken from the side of the femur opposite that of the plate The results indicate that control of the bending * Read at the Annual Meeting of the Orthopaedic Research Society, Las Vegas, Nevada, February 3, 1977. t Division of Orthopedic Surgery, University of Utah Medical Center, Salt Lake City, Utah 84132. Please address reprint requests to Dr. Dunn. t National Bureau of Standards, Washington, D.C. 20234. § University of Utah Research Institute, Salt Lake City, Utah 84108. 1 Department of Materials Science and Engineering, University of Utah, Salt Lake City, Utah 841 12. loads to which bone is subjected during healing may be an important aspect of fracture treatment. CLINICAL RELEVANCE: Initial fracture healing appears to be enhanced by rigid internal fixation. However, the ultimate strength of the bone depends not only on this phase of fracture healing but also on the remodeling that occurs. Traditional metallic plates have been shown to have unfavorable effects on bone remodeling because they prevent the underlying bone from being subjected to the stresses that are normally present. Remodeling studies have demonstrated that when the normal forces to which living bone is exposed are removed, decreased cortical thickness and increased osteoporosis ensue. These changes have been shown to be a function of plate flexibility or stiffness. The present study, utilizing flexible plates made from fiber-reinforced composite materials, indicates that the presence of more normal stress may be beneficial in early fracture healing as well as remodeling. We suggest that so-cabled stress-shielding caused by rigid metallic plates can be diminished through the use of plates constructed from more flexible composite materials. Bone is a living tissue which responds to its mechanical environment. It has been shown that mechanical stresses are needed to minimize undesirable remodeling of FIG. I EFFECTS OF FLEXURAL RIGIDITY OF PLATES ON BONE HEALING 867 VOL. 61-A, NO 6. SEPTEMBER 1979 bone, but the effects of loading and movement during fracture healing have been explored less thoroughly’3. The optimum mechanical environment for fracture healing has not been delineated, a fact that is well demonstrated by present-day clinical use of such widely divergent treatments as cast-bracing and internal fixation. We studied the response of fracture healing to various mechanical environments provided by bone plates with a wide range of flexural rigidities, and in this report we describe our experimental approach and initial findings. In studies of bone remodeling, it has been noted that use of conventional metal fixation plates results in a bone plate-bone mechanical mismatch in which the relatively rigid plates ‘ ‘shield ‘ ‘ the bone from the stress to which it is ordinarily exposed. It has been demonstrated that this socabled stress-shielding results in bone remodeling causing cortical thinning or osteoporosis, or both2’9”#{176}”3. It has also been shown that the magnitudes of these changes are functions of flexural stiffness of the plate. For exampie, Woo and co-workers, using intact canine femora, compared the effects of a conventional metallic plate with the effects of a fiber-reinforced-polymer composite plate which was geometrically similar to the metallic plate but was one-tenth as stiff. Less cortical thinning was observed with the more compliant plate. Similarly, Tonino and associates noted less remodeling beneath an even more compliant non-reinforced-polymer plate than beneath a metal plate with which it was compared. The effect of flexurab stiffness of the plate on fracture healing has been explored to some extent”36. Fracture healing is a complex continuing process, but to facilitate understanding it can be considered as composed of phases. The first phase is characterized by proliferation of callus and the second, by bone remodeling. When considering the effects of plates in bight of these healing phases, one can hypothesize two conflicting requirements during the service life of a plate. First, it must have sufficient flexural rigidity to prevent excessive motion at the fracture site, to maintain alignment and allow initial fracture healing (calbus proliferation) to occur. Subsequently, the plate should have sufficient compliance to minimize stress-shielding and the undesirable aspects of remodeling such as cortical thinning and bone resorption. We performed this preliminary study to explore the effect on fracture healing of varying the flexural rigidity of bone plates from about 4 to 40 per cent of the rigidity of the bone to be plated. It is possible to measure many aspects of the biological response of bone to the mechanical environment present during fracture healing. We chose to determine bone strength because it is of direct clinical interest. Our findings indicate that within the limits explored, decreasing the flexurab rigidity of the plate increased the strength of the healing bone. Materials and Method One class of materials available for the fabrication of internal fixation devices is composed of polymers reinforced with continuous fibers 1,3.6.5,13 These materials can be used to produce devices such as bone plates by laminating layers of the materials in a heated die under pressure. A spectrum of mechanical properties can be achieved in devices of similar geometric configuration by using different materials and lamination patterns. For our project, geometrically similar plates with a range of flexural rigidities were obtained. Some were selected from clinically used stainless-steel plates, while others were fabricated from commercially available graphite-polysulfone and glass-epoxy laminates as well as from laboratoryprepared graphite-polypropylene composites . All composite materials were tested for biocompatibility in agaroverlay tissue-culture tests and were shown to produce responses similar to those produced by samples of the USP Negative Control Plastic Reference Standard . Examples of the fabricated plates and of a stainless-steel plate are shown in Figure 1. Examples of the plates used to stabilize the femoral osteotomies. Top graphite-polysulfone; middle glass-epoxy; bottom conventional 316L-type stainless steel. The plates were approximately I 15 millimeters long, and depending on the properties desired they were fabricated with widths between about eleven and thirteen millimeters and thicknesses between 2.5 and 6.3 millimeters. All plates were approximately 1 15 millimeters long, with six screw-holes located in the same positions as in the standard steel plates. The desired spectrum of flexural rigidity was achieved not only by varying the plate materiabs but also by varying the plate widths from eleven to thirteen millimeters and the thicknesses from 2.5 to 6.3 millimeters. The flexural rigidities ranged from I .45 to 16.8 Nm2, while the axial stiffnesses were more nearby the same. The eleven plates that were more flexible had axial stiffnesses that ranged from approximately I .4 x bO” to 2.8 x lO Nm2, and the six stiffer plates had axial stiffnesses that ranged as high as 10.8 x l0 Nm2. The flexural rigidities were determined by four-point bending tests, while the axial stiffnesses were determined in compression tests. The flexural rigidities were calculated using the

Analysis of particles in acetabular components from patients with osteolysis.

Acetabular polyethylene components were quantitatively analyzed for the presence of third body particles from 38 consecutively retrieved components. Backscattered electron imaging and correlated energy dispersive x-ray analysis were used for the assessments. Retrievals were divided into 4 groups based on methods of fixation and metal alloy types: 8 hydroxyapatite coated, 6 cobalt chrome porous coated, 17 titanium porous coated, and 7 cemented implants were evaluated. The backscattered electron imaging data showed that the components from the hydroxyapatite coated implants had larger particles than did the components from the cemented group. The hydroxyapatite group had 51 +/- 52 particles per mm2. The cobalt chrome alloy group had 10 +/- 9 particles per mm2, and the titanium alloy group had 9 +/- 16 particles per mm2. The cemented group had 5 +/- 4 particles per mm2. The difference between the cement group and the hydroxyapatite group was statistically significant. The elemental analysis showed that 70% of the particles in the hydroxyapatite group had calcium and phosphorus elements. Third body particles likely contribute to particulate generation. The results suggest that the hydroxyapatite coated components have the potential for producing greater amounts of particulate debris. Continued analysis of retrieved components for the presence of the third body particles is required.

Anterior Stabilization of Thoracolumbar Injuries

A new implant system has been designed to rigidly fix the spine anteriorly at a single level above and below the injury site for treatment of severe thoracolumbar fractures. The primary system consists of two rods rigidly linked by vertebral body bridges. These bridges are secured to each vertebra at two points: one posterolaterally on the body with a vertebral body screw and the other more anterolaterally with a vertebral body staple. This implant system has been used in 48 consecutive thoracolumbar injuries. There have been three late failures to attain spine fusion, but in no patient has there been loss of vertebral column alignment or spinal canal decompression. There have been no major catastrophes with the system. There have been one deep- and two superficial-wound infections, all of which have been successfully treated and resolved.

Simultaneous quantitation of knee ligament forces

An apparatus and experimental technique have been developed which allow the simultaneous quantitation of human cadaver knee ligament forces with the knee in various positions and subjected to various loads. The methodology permits all the major ligaments of the knee to be kept intact, thus preserving much of their normal relative load-bearing function. A novel measurement technique was employed for ligament force measurement. Strain gauges were placed at the ligamentous insertions and origins either on cortical bone or the cancellous/cortical transition. Gauge output was subsequently calibrated quantitatively by cutting the ligaments and subjecting them to known loads. The gauging technique was thermally stable, and the method as a whole yielded reproducible relative ligament force data when knee positions and loads were repeated for a given knee specimen. Three fresh human cadaver knee specimens have been subjected to testing thus far with gauges at six ligamentous sites. Initial evaluation of the data indicate that ligamentous loading patterns were quite different from those reported in the literature as determined by other methods.

Simultaneous Anterior-Posterior Approach Through a Costotransversectomy for the Treatment of Congenital Kyphosis and Acquired Kyphoscoliotic Deformities

BACKGROUND Congenital kyphosis and acquired kyphoscoliotic deformities are uncommon but are potentially serious because of the risk of progressive deformity and possible paraplegia with growth. Our current approach for the treatment of these deformities is to use a single posterior incision and costotransversectomy to provide access for simultaneous anterior and posterior resection of a hemivertebra or spinal osteotomy, followed by anterior and/or posterior instrumentation and arthrodesis. To our knowledge, this approach has not been reported previously. METHODS The medical records and radiographs for sixteen patients who had been managed at our institution for the treatment of congenital kyphosis and acquired kyphoscoliosis between 1988 and 2002 were analyzed. The mean age at the time of surgery was twelve years. The diagnosis was congenital kyphosis for fourteen patients and acquired kyphoscoliotic deformities following failed previous surgery for two. The mean preoperative kyphotic deformity was 65 degrees (range, 25 degrees to 160 degrees ), and the mean scoliotic deformity was 47 degrees (range, 7 degrees to 160 degrees ). Fifteen patients were managed with vertebral resection or osteotomy through a single posterior approach and costotransversectomy, anterior and posterior arthrodesis, and posterior segmental spinal instrumentation. The other patient was too small for spinal instrumentation at the time of vertebral resection. A simplified outcome score was created to evaluate the results. RESULTS The mean duration of follow-up was 60.1 months. The mean correction of the major kyphotic deformity was 31 degrees (range, 0 degrees to 82 degrees ), and the mean correction of the major scoliotic deformity was 25 degrees (range, 0 degrees to 68 degrees ). Complications occurred in four patients; the complications included failure of posterior fixation requiring revision (one patient), lower extremity dysesthesias (one patient), and late progressive pelvic obliquity caudad to the fusion (two patients). The outcome, which was determined with use of a simplified outcomes score on the basis of patient satisfaction, was rated as satisfactory for thirteen patients, fair for two patients, and poor for one patient. CONCLUSIONS A simultaneous anterior and posterior approach through a costotransversectomy is a challenging but safe, versatile, and effective approach for the treatment of complex kyphotic deformities of the thoracic spine, and it minimizes the risk of neurologic injury. LEVEL OF EVIDENCE Therapeutic Level IV.

Cementless Femoral Revision Arthroplasty 2- to 5-Year Results With a Modular Titanium Alloy Stem

Seventy-five consecutive cementless femoral revision arthroplasties were performed, using a modular titanium alloy stem (S-ROM, Joint Medical Products, Stamford, CT). Sixty-six hips were available for complete follow-up evaluation at 2 to 5 years (average, 3.4 years). Independent clinical analysis compared pre- with postoperative modified Harris hip scores and examined patient satisfaction. Independent radiographic analysis was also undertaken. There were two rerevisions, both for hematogenous sepsis. Of the rest, clinical scores improved from a preoperative value of 44 to a postoperative value of 83. Eighty-nine percent of patients were satisfied or very satisfied with the results of surgery. Fifty-two stems had solid bony ingrowth, seven were stabilized by fibrous tissue, and five were radiographically loose. Five-year survival, with rerevision as endpoint, was 96.4%.

Flat-top talus. A long-term report of twenty club feet.

Twelve adult patients who had been treated for talipes equinovarus were studied. A flat-top talus was found in all of the twenty ankles involved. The mean age of the patients was forty-two years. All patients had a measurably decreased range of motion in the ankle, but it caused them little or no pain. Only one patient had significant pain and limitation of activity.

Mechanical properties of some fibre reinforced polymer composites after implantation as fracture fixation plates

Continuous fibre reinforced composite bone plates made from graphite/polysulphone and glass/epoxy laminates were implanted for 16 weeks on osteotomized canine femurs and for 12 months on intact femurs. After sacrifice, the plates were removed and tested in four point bending for stiffness and strength. There were no significant differences in properties between control and implanted plates in the 16 week study. Both the glass/epoxy and the graphite/polysulphone systems showed deterioration after implantation for 12 months.

Distraction and compression loads enhance spine torsional stiffness

We have assessed the degree to which compression and distraction forces applied to the spine of the magnitude achievable through instrumentation systems increase torsional stiffness, thereby possibly enhancing the environment for arthrodesis. A functional relationship between spine torsional stiffness and axial compression and distraction forces is described. To establish this relationship whole human thoracolumbar spines as well as individual motion segments were tested in torsion under a variety of axial loading conditions consistent with forces applied clinically. These studies indicate that applying axial loads to the whole thoracolumbar spine through the action of compression or distraction increases the stiffness of the spine in torsion. Compression and distraction forces increased the torsional stiffness of thoracic segments, but only compression forces were found to significantly increase the stiffness of lumbar segments. The soft tissue structures of both the anterior and the posterior columns have the ability to increase torsional stiffness with axial loading.