________________________________________________________

ASSOCIATION OF BONE AND JOINT SURGEONS

Fifty-Seventh Annual Meeting

June 8-12, 2005

Carmel Valley Ranch

Carmel, California ________________________________________________________________

Paper #1

DISLOCATION RATE OF HEMIARTHROPLASTY IN TUMOR CONDITIONS

Sean P. Scully, MD, PhD1 (n), Michaela M. Schneiderbauer, MD2, Rafael J. Sierra, MD3, Cathy Schleck, BS3, William S. Harmsen, MS3

1University of Miami, Miami, FL, 2Orthopaedische Universitaetsklinik, Basel, Switzerland, 3Mayo Clinic, Rochester, MN

Background: Hemiarthroplasty is frequently used for treatment of femoral neck insufficiency resulting from neoplastic disease in the proximal femur. A recent study analyzed the dislocation rates of hemiarthroplasty but excluded patients with tumor involvement at the site of surgery since the authors hypothesized that the dislocation rates in patients with tumor involvement would be markedly higher. The current study compares the dislocation rates of hemiarthroplasty performed without tumor involvement to those performed with tumor involvement at the surgical site. Materials and Methods: Patients undergoing hemiarthroplasty for tumor involvement of the proximal femur were identified within a total joint registry and the patient charts were reviewed retrospectively for dislocation rates, presurgical conditions, and postsurgical outcomes and treatments. Between 1974 and 2001, 1812 patients were treated with hemiarthroplasty for reasons other than tumor involvement, and 320 hemiarthroplasties were performed for tumor related conditions. Patients in the tumor related group differed in that they were younger and included a higher proportion of male patients. Results: Hemiarthroplasty performed for tumor conditions at the surgical site dislocated more frequently than hemiarthroplasty performed for non-tumor conditions with ten-year dislocation rates of 10.9 percent (tumor) and 2.1 percent (non-tumor). The median time to dislocation in patients with tumor-related conditions is markedly shorter than in the patients without tumor involvement, 24 days (tumor) versus 37 days (non-tumor). The preservation of the greater trochanter in patients with tumor involvement showed a statistical trend towards influencing the dislocation rate favorably, Hazard ratio = 3.5 (p=0.06). Conclusions: Hemiarthroplasty performed for tumor related conditions at the site of surgery shows a significantly higher short-term and long-term dislocation rate compared to hemiarthroplasty performed for other reasons than tumor involvement. The time to dislocation is shorter in patients with tumor involvement. The preservation of the greater trochanter seems to be a dominant factor in determining the likelihood of dislocation of hemiarthroplasties; more so than the level of resection and the extent of soft tissue compromise.

Paper #2

EFFECT OF TGF-β ON METASTATIC RENAL CELL CARCINOMA GROWTH AND OSTEOLYSIS

Kristy L. Weber, MD (n) Johns Hopkins Outpatient Center, Baltimore, MD

Bone metastases develop in approximately 40% of renal cell carcinoma (RCC) patients and, despite diagnostic and therapeutic advances, treatment remains palliative. RCC bone metastases are characterized as highly vascular and predominantly osteolytic, however the interactions between RCC cells and the bone environment contributing to this phenotype are largely unexplored. The potent and multifunctional cytokine TGF-β is present at high levels in the bone environment and has been implicated in the development of bone metastasis from breast and prostate cancer. In this study, we investigated the role of TGF-β in the growth of RCC cells in the bone and subsequent bone destruction. TGF-β and its receptors (TGF-βRI and TGF-βRII) were expressed in 18/18 cases of human RCC bone metastasis immunohistochemical analysis (IHC). Correspondingly, human bone-derived RCC cell lines (RBM-IT4, RMB23, RBM21A, RBM17C) also express TGF-βRI and TGF-βII by Western analysis. Treatment of human bone-derived RCC cell lines with TGF-β (2.5 ng/ml) resulted in phosphorylation of SMAD2 and SMAD3, suggesting that the classical TGF-β signaling pathway is intact in these cell lines. Transgenic RBM1-IT4 cells stably expressing a dominant-negative (dn) TGF-βRII were generated. Two clones (DN1 and DN2) with a 46% reduction in TGF-β signaling were selected for further study. RBM1-IT4, DN1, DN2, and empty vector control cells were injected into the tibiae of athymic nude mice to determine whether TGF-β stimulation of bone-derived RCC cells played a role in their ability to grow in the bone microenvironment and induce osteolysis. After ten weeks, digital radiography showed that mice injected with DN2 cells had a significantly decreased incidence of bone lesions (p<0.05) and osteolysis (p<0.001) relative to parental cells, while those injected with DN1 cells had no detectable lesions (p<0.001) or osteolysis (p<0.001). Tumor growth, determined by tumor weight and H&E staining, was significantly decreased in mice injected with DN1 and DN2 cells compared to parental cells (p<0.05), correlating with tumor-induced osteolysis. In light of these results, we determined the ability of TGF-β to stimulate the production of cytokines and matrix metalloproteinases (MMPs) involved in cell growth, angiogenesis, and bone resorption. TGF-β treatment of bone-derived RCC cell lines stimulated increased secretion of IL-6 (⎯x = 1.5-fold) and VEGF (⎯x = 1.3-fold) as determined by ELISA. TGF-β also induced MMP-2 and MMP-13 expression in a time-dependent manner as observed by RT-PCR. Correspondingly, IHC analysis showed that IL-6, VEGF, MMP-2, and MMP-13 were expressed in 8/9, 18/18, 8/10, and 12/13 cases of human RCC bone metastasis. Other growth-promoting and bone resorptive factors, such as IL-8, TNF-α, and IL-1β were not induced by TGF-β in bone-derived RCC cell lines as tested by ELISA. Taken together, these data suggest that TGF-β may play a substantial role in promoting the growth of RCC bone metastases and concomitant bone destruction, possibly involving the induction of cytokines (IL-6, VEGF) and MMPs (MMP-2, MMP-13).

Paper #4

MISDIAGNOSIS IN OSTEOSARCOMA: EFFECT OF ERRONEOUS SURGICAL PROCEDURE ON RECURRENCE AND SURVIVAL

Miguel A. Ayerza, MD (n), D. Luis Muscolo, MD, Luis Aponte-Tinao, MD

Italian Hospital of Buenos, Buenos Aires, Argentina

Introduction: The purpose of this study was to determine whether an erroneous surgical procedure based on an initial misdiagnosis in osteosarcoma has prognostic relevance in relation with recurrence rate and survival, and if limb salvage can be safely performed in such patients. Material and methods: We reviewed the surgical treatment and oncologic results in 117 high grade osteosarcomas that had been treated in our institution between 1990 and 2000. Neoadjuvant therapy was given to all patients. In 9 patients previous intralesional curettage was performed based on an erroneous initial diagnosis. Final surgical treatment after multiagent therapy included 3 amputations and 6 limb salvage procedures. In the remaining 108 patients with no misdiagnosis, final surgical treatment included 6 amputations and 102 limb salvage procedures. Local recurrence and survival rates were recorded. Results: Fifteen out of the 117 patients developed a local recurrence. In the misdiagnosed group the local recurrence rate was 55%, while for those without initial erroneous diagnosis was 8%. Differences were statistically significant (p< 0.05). There were five local recurrences in six limb salvage procedures in the group with a previous erroneous diagnosis. The 5-year estimate of post recurrence survival for the 15 patients was 33%, while for those without local recurrence the 5–year overall survival rate was 65%. Discussion: We conclude that osteosarcoma patients, treated with an erroneous surgical procedure based on an initial misdiagnosis have poor prognosis with a higher recurrence rate and lower life expectancy, and may not be good candidates for limb salvage even after being treated with multiagent chemotherapy.

The purpose of this study was to analyze the effect of previous erroneous treatment on recurrence and survival in osteosarcoma.

Paper #5

MOLECULAR MECHANISMS OF PTH MEDIATED ENHANCEMENT OF FRACTURE REPAIR

Sanjeev Kakar, MD, MRCS (n), Geol Barnes, PhD, Michael Song, MSc, Kevin Wang, MSc, Lee J. Silkman, BSc, Louis C. Gerstenfeld, PhD, Thomas A. Einhorn, MD Boston University School of Medicine, Boston, MA

Introduction: Fracture healing is a unique and well orchestrated biological process resulting in optimal skeletal repair. Despite this, it is estimated that between five and ten percent of the fractures that occur annually in the United States exhibit some degree of impaired healing. Several compounds have been approved for the management of these conditions including BMP-2 and OP-1 (BMP-7). While these compounds show promising results, their applicability is limited due to the need to implant the materials in the fracture site usually in the context of an operative procedure. The ideal modality for the treatment of fractures would be a compound that could be administered systemically at the early stages of fracture repair to enhance the healing process and reduce the incidence of nonunion. Several studies have reported on the enhancement of fracture repair by systemically administered PTH (1-4). Recently Alkhiary et al. (2004) demonstrated in a rat femoral fracture model that intermittent systemically administered PTH resulted in enhanced bone healing as indicated by an increase in callus volume, bone mineral density and biomechanical properties compared to controls. The mechanisms by which this enhanced repair is achieved, however, remains unknown. We hypothesize that one way PTH mediates its effects on fracture healing is by amplifying and accelerating the chondrogenic phase of the endochondral repair process, the predominant mechanism by which callus develops during fracture repair. Materials and Methods: Standardized closed femoral fractures were produced in C57Bl/6 mice and treated with PTH (40μg/Kg) or vehicle alone (sterile saline) via daily subcutaneous injections over the first 14 days of the 28 day standard repair time course. Calluses were harvested at days 5, 7, 10, 14, 21 and 28 post fracture for faxitron radiographic analyses, histologic analyses and isolation of RNA for molecular analyses using ribonuclease protection assay (RPA). These times points represent early chondrogenic recruitment (day 5), chondrogenic maturation (days 7 and 10), replacement of cartilaginous callus (day 14) and bone remodeling (days 21 and 28). Results: Radiographic analyses demonstrate that PTH-treated fractures calluses display enhanced mineralization at earlier time points post fracture than controls. Fracture calluses from PTH-treated animals exhibit an increased mineralization by day 14 combined with a moderate increase in callus size as compared to controls. Molecular analysis of genes associated with chondrogenic and osteogenic cellular differentiation demonstrate that, in PTH-treated animals, chondrocyte and osteoblast maturation is enhanced. Expression of collagen type X, a marker of hypertrophic chondrocytes, peaks at day 7 in PTH-treated fractures as compared to day 10 in control fractures demonstrating an increased rate of cartilage maturation in the presence of PTH. The chondrogenic genes (collagens type II and X) are also expressed over a broader time frame in PTH-treated fractures suggesting an overall increase in chondrocyte differentiation in addition to earlier maturation. Finally, in PTH-treated fractures, osteocalcin expression, a marker of mature osteoblasts, is delayed but enhanced. In control animals increased osteocalcin expression is detectable by day 10. In PTH-treated fractures osteocalcin expression is not significantly induced until day 14 but increases to maximal levels by day 21 reaching a higher level of expression than observed in control fractures. Conclusion: These data show that daily systemic administration of PTH accelerates and augments fracture healing by enhancing both the chondrogenic phase and osteogenic phases of endochondral bone repair. Clinical Relevance: A systemically administered drug that enhances bone repair could have widespread clinical applications in the use of treating fractures, enhancing osteointegration of porous implants and promoting joint and spinal arthrodeses. PTH is the first bone formation agent that has been shown to be effective for systemic treatment of bone disease and may represent an attractive new modality in the management of bone healing. References

1. Alkhiary YM, Gerstenfeld LC, Krall E, et al. Enhancement of experimental fracture healing by systemic administration of recombinant human parathyroid hormone (PTH 1-34). J Bone J Surg Am; In print.

2. Andreassen TT, Ejersted C and Oxlund O (1999). Intermittent PTH (1-34) treatment increases callus formation and mechanical strength of healing rat fractures. JBMR 14(6); 960-968.

3. Holzer G, Majeska RJ Lundy MW, Hartke JR and Einhorn TA (1999). PTH enhances fracture healing: A preliminary report. Clin. Orthop. 366:258-263

4. Nakajima A, Shimoji N, Shimizu S, Moriya H, Einhorn TA, and Yamazaki M (2002). Mechanisms for the enhancement of fracture healing in rats treated with intermittent low-dose human PTH (1-34). JBMR 17(11); 2038-2047.

Paper #6

PTH(1-34) ENHANCES SPINE FUSION RATE AND MASS VOLUME

Joseph M. Lane, MD (a-Lilly), Susan V. Bukata, MD, Emre Tomin, BS, Matthew Cunningham, MD Hospital for Special Surgery, New York, NY

Introduction: Low back pain affects a substantial proportion of the population and despite current techniques, up to 40% of lumbar fusions fail to heal in this group leading to ongoing pain and additional surgery. In both humans and rats, PTH administered in low pulstatile doses leads to increased bone formation in fracture and nonfracture conditions. The effect of PTH on spine fusion is unknown. We hypothesize that intermittent PTH induces the bone formation stage early, with reduction in bone resorptive stage, leading to improved fusion rate and quality. Materials and Methods: 44 male New Zealand white rabbits underwent harvest of bilateral iliac crest autograft and bilateral spine fusion at the L5-L6 level following the Boden spine fusion model. The animals were divided into two equal groups and beginning on post-operative day 4, half of the animals received daily subcutaneous injections of PTH(1-34) at 10 micrograms/kg/dose. All animals were sacrificed 6 weeks from the date of surgery. All animals had an x-ray of the bone-grafting site immediately post-operatively and at the time of sacrifice. The L5-L6 vertebral segments were removed and analyzed by manual bending for fusion in the coronal and sagital planes, as well as in torsion. All specimens were analyzed by faxitron radiograph and scored by three independent investigators for radiographic fusion and quantity of bone formation. All segments underwent fine cut (0.6mm) CT and volume analysis of the fusion mass. Results: Manual-bending testing identified fusion in 30% of the control animals while 81% of the animals treated with PTH fused (Chi-squared test =10.8 p<.001, Fisher exact test p<.002). Faxitron radiographic analysis of individual fusion sites showed 20% of control animal sites fused and 69% of PTH animal sites fused (p<.001). Using a radiographic fusion score (0-5) the controls had a score of 3.36 and the PTH animals 4.51 (p<.001, 83% inter-observer agreement). Fine cut CT analysis and volume measurement demonstrated a 75% increase in bone formation in the PTH treated animals with an average mass of 3.5cc in the control animals and 6.1cc with PTH treatment (p<.001). Conclusions: Intermittent PTH administration as used in osteoporosis increased the rate of spine fusion by 170% and increased the mass of bone formation by 75% in the Boden rabbit spine fusion model. PTH has potential to not only treat osteoporosis but also improve the rate of fusion and improve the quantity of bone mass formed in the fusion. Osteoporotic patients undergoing spine fusion would have the fusion enhanced with the anabolic agent PTH(1-34) while they are addressing their underlying osteoporosis.

Paper #7

THE USE OF APLIGRAF IN CHRONIC WOUND CARE THERAPY

Scott T. McMullen, MD (n), Stephen M. Hansen, MD, Tracy M. Campbell, BA, Timothy C. Fitzgibbons, MD, David J. Inda, MD, Dennis F. Hayes, MD, Laura Landon, RN

Orthopaedic Office, Omaha, NE

Introduction: Chronic lower extremity wounds present a challenging problem. Such wounds may contribute to the development of local soft tissue infection, deep seated bone and joint infection, and systemic sepsis. Along with this comes the risk of lower extremity amputation as a final treatment option. Risk factors for the development of chronic lower extremity ulceration include diabetes with the associated peripheral neuropathy and/or peripheral vascular disease and also chronic venous insufficiency. Traditional wound care treatments focus on the maintenance of a clean, moist wound environment, off-loading to minimize ongoing mechanical tissue injury, and anti-edema therapy to minimize cutaneous hydrostatic forces. In recent years treatment adjuvants have been utilized to promote soft tissue healing. Our initial wound care center experience was tailored around the use of Procuran, a concentrate of growth factors from a platelet releasate. Regranex, a recombinant platelet growth factor gel, represents a newer method of delivering a cutaneus growth factor. An extension of this is cell based wound therapy in which active cells are utilized to create an interactive growth factor delivery system. One such product of tissue engineered skin is Apligraf by Organogenesis. The purpose of this study is to review the use of Apligraf in lower extremity wounds in our wound care clinic setting. Material and Methods: 52 patients with 72 wounds were retrospectively reviewed. There were 32 men with 47 wounds and 20 women with 25 wounds. The average age of the patient population was 64 years with a range of 33 to 88 years. The predominant diagnosis was diabetes with peripheral neuropathy followed in frequency by venous insufficiency disease. All patients’ wounds were treated according to a wound care clinic protocol including maximization of off-loading measures, edema control, evaluation and treatment of peripheral arterial occlusive disease, treatment of infection, and serial wound debridements in the outpatient wound care clinic setting. Apligraf was considered as an adjuvant to wound treatment if the wounds were not progressively healing. Diagnoses required for the use of Apligraf are diabetic neuropathic wounds of the foot and ankle and/or venous insufficiency wounds of the lower extremities. The average wound treatment time prior to presentation to the wound care clinic averaged 3.8 months with a range of 1-36 months. Average wound care clinic treatment time prior to Apligraf application was 4 months with a range of 0-24 months. Wound volume averaged 1,275 cubic millimeters prior to presentation to the wound care clinic and wound volumes at the time of Apligraf application averaged 1,291 cubic millimeters. Results: Overall 48 of the 72 wounds have progressed to healing, a healing rate of 67%. The average time to heal was 2.3 months with a range of 2 weeks to 9 months. Neuropathic wounds healed at a rate of 70%, venous insufficiency wounds at 53%, and post-surgical wounds healed at a rate of 78%. There appear to be two healing patterns, rapid and slow, with rapid healing occurring within 2 to 4 weeks and slow healing occurring over several months with a gradual decrease in wound size. Discussion: The use of Apligraf bilayered skin substitute offers an active growth factor delivery system. It contains two cell lines, fibroblast in the dermal layer and keratinocytes in an epidermal layer. The patients presenting in the wound care clinic environment are a very heterogenous and challenging group. When basic wound care has not allowed for progression onto wound healing, the use of Apligraf as an active growth factor delivery system has been a helpful adjunct to assist with healing in the majority of these wounds.

Paper #8

NEUROTIZATION VS. NERVE REPAIR: A COMPARISON OF SURGICAL TECHNIQUE IN AN ANIMAL MODEL OF CHRONICALLY DENERVATED MUSCLE

Scott W. Wolfe, MD (n), Andrew Swanson, MD, Steve Doty, MD, Joseph Feinberg, MD, Mike Khazzam, BA, John

Ehteshami, MD, Charlotte Shum, MD Hospital for Special Surgery, New York, NY

Introduction: Reinnervation of chronically denervated muscle is clinically unpredictable and poorly understood on a biological basis. Current operative strategies include either direct nerve repair, nerve grafting, nerve transfer, or neurotization of the affected muscle with a proximal donor nerve. The goal of the present study is to compare the quality of muscle recovery using microneural repair versus direct nerve-to-muscle implantation (neurotization) in a rat model of chronic denervation. Methods: Forty-two Sprague-Dawley rats underwent surgical denervation of the tibialis anterior muscle by transecting the common peroneal nerve. After 0, 8 or 12 weeks of denervation, animals were assigned to either a direct repair cohort (a freshly axotomized common tibial nerve was coapted to the distal common peroneal nerve stump) or a neurotization cohort (a freshly harvested peripheral nerve graft was coapted to a freshly axotomized common tibial nerve and the distal end directly implanted into the tibialis anterior muscle). An additional five animals were used for a “sham” cohort (no surgical denervation) and six animals used for a control group (no surgical repair). After a twelve week recovery period, animals underwent functional testing of the tibialis anterior muscle using electrical stimulation to generate a tetanic contraction. Peak force and electromyographic characteristics (latency and amplitude) were compared to the un-operated, contralateral limb. Tibialis anterior muscles were then harvested for mass and histologic evaluation. Results: Forty-five animals completed histologic and physiologic testing. Denervated control animals demonstrated a significant decrease in muscle mass, contractile strength and peak EMG amplitude compared to “sham” animals. Repaired animals tended to perform better than neurotized animals with respect to muscle mass, contractile strength and peak EMG amplitude after 0, 8 and 12 weeks denervation (Figures 1,2). Differences in contractile force were only significant in the 0 week denervation group (94 ± 30% vs. 50 ± 20%, Repair vs. Neurotization; p=0.01). In all groups, duration of denervation adversely affected mass, contractile strength and EMG characteristics. Neurotized muscles processed for histologic analysis demonstrated acetyl-cholinesterase activity at the nerve-muscle interface, confirming the formation of motor end-plates de novo. Discussion: Although animals in the neurotization group demonstrated the ability to reinnervate a chronically denervated muscle via a peripheral nerve graft bridge, animals in the direct repair group consistently performed better with respect to contractile strength, muscle mass and electromyographic characteristics. Recent work has focused on the inability of Schwann cells within the chronically denervated distal stump to support axonal reinnervation [1,2]. To date, evidence from this study does not support the hypothesis that muscle neurotization is superior to standard nerve repair for functional restoration of chronically denervated muscle. As the duration of denervation increases, however, neurotization may be more efficacious than direct nerve repair for functional muscle recovery. References: 1. Sulaiman OA and Gordon T. Effects of short- and long-term Schwann cell denervation on peripheral nerve regeneration, myelination, and size. Glia. 2000 Dec;32(3):234-46 2. Emery E, Rhrich-Haddout F, Kassar-Duchossoy L, Lyoussi B, Tadie M, Horvat JC Motoneurons of the adult marmoset can grow axons and reform motor endplates through a peripheral nerve bridge joining the locally injured cervical spinal cord to the denervated biceps brachii muscle. J Neurosci Res. 2000 Dec 15;62(6):821-9. **Oakland Medical Center, Department of Orthopaedics, Oakland, CA

Figure 1. Tibialis anterior strength ± SD of animals tested after denervation for 0, 8 or 12 weeks and subsequent repair or neurotization. Contractile strength is expressed as a percentage of the contralateral limb. Red numbers indicate p values comparing repair and neurotized groups at each time point. N for each group = 3-6 individuals.

Figure 2. Anterior compartment mass ± SD of animals tested after denervation for 0, 8 or 12 weeks and subsequent repair or neurotization. Mass is expressed as a percentage of the contralateral limb. Red numbers indicate p values comparing repair and neurotized groups at each time point. N for each group = 3-6 individuals. ACKNOWLEDGEMENTS: Aided by a grant from the Orthopaedic Research and Education Foundation.

0

10

20

30

40

50

60

70

80

90

100

Control 0 WeeksDenervation

8 WeeksDenervation

12 WeeksDenervation

Con

tract

ile s

treng

th (%

of

cont

rala

tera

l lim

b) ShamNo RepairRepairNeurotization

0

10

20

30

40

50

60

70

80

90

100

Control 0 WeeksDenervation

8 WeeksDenervation

12 WeeksDenervation

Con

tract

ile s

treng

th (%

of

cont

rala

tera

l lim

b) ShamNo RepairRepairNeurotization

TTiibbiiaalliiss AAnntteerriioorr SSttrreennggtthh

0.14

0.10

*0.01*0.001

AAnntteerriioorr CCoommppaarrttmmeenntt MMaassss

20

30

40

50

60

70

80

90

100

110

Control 0 WeeksDenervation

8 WeeksDenervation

12 WeeksDenervation

Mus

cle

mas

s (%

of c

ontr

alat

eral

lim

b)

ShamNo RepairRepairNeurotization

20

30

40

50

60

70

80

90

100

110

Control 0 WeeksDenervation

8 WeeksDenervation

12 WeeksDenervation

Mus

cle

mas

s (%

of c

ontr

alat

eral

lim

b)

ShamNo RepairRepairNeurotization0.07

*0.02 0.06

*0.002

Paper #9

PRIMARY AND REVISION THA IN PATIENTS WITH SMALL FEMORAL MORPHOMETRY

William L. Lanzer, MD1(a-Portland Orthopaedics), R. Sekel, MD2, R. Eberle2, M. Richardson3, D. Gibson4, L. Kwong5, B. Mallin6, A. Infante7

1University of Washington, Seattle, WA, 2St. George Hospital Campus, Sydney, Australia, 3Epworth Hospital, Richmond, Australia, 4Yale University, New Haven, CT, 5UCLA Medical Center, Los Angeles, CA, 6Institute for Bone and Joint

Disorders, Phoenix, AZ, 7Tampa General Hospital Tampa, FL Introduction and Aims: Currently, multiple femoral component types and sizes exist for primary total hip arthroplasty. However, component sizes for small femoral geometry are generally not available. The purpose of this study is to present the short-term use of a femoral component with sizes that extend into small femoral morphometry applications. Method: Between November of 2001 and December of 2003, 20 primary THA cases and three revision THA cases were performed utilizing a non-cemented, dual threaded, cone shaped (DTCS) modular femoral component manufactured in off-the-shelf sizes which include those sizes for small femora. The components are made of CoCr and include a size “Z” (19mm proximal, 9mm distal) and a size “Y” (17mm proximal, 8mm distal). Both components have hydroxyapatite coating for stimulating increased bone on-growth and a modular neck allowing intra-operative adjustments of leg length, version, offset and neck length. Results: The average patient follow-up was 10 months (range: 64-27 months). There were 19 (83%) hips in which the “Z” component was used, and four (17%) hips with the “Y” component. Radiographic evaluation revealed well fixed and positioned components with evidence of bone densing in areas in intimate contact with the DTCS component. Radiographic evidence of minor stress shielding was observed in the greater trochanter (Gruen Zone 1) and the proximal calcar/neck cut region (Gruen Zone 7). Two revision cases (8%) required the additional use of a 6cm modular extension component (MEC) to bridge a proximal femoral deficiency. Two cases (8%) required adjunctive strut allografting at the time of surgery to protect a thin or deficient femoral cortex. There were no reported postoperative compilations related to the femoral component. There was no disassociation of the modular neck from the femoral stem and there was no incidence of femoral component fracture. Conclusion: While expanding component profile offerings into larger sizes is common, developing similar component designs for abnormally small femora is uncommon, beyond the scope of the materials used and only done as a “custom” order. The DTCS modular femoral component used affords a versatile option when presented with cases involving small femoral morphometry. We conclude that the DTCS component in small sizes is promising and warranted for continued use.

Paper #10

"MINI-RELEASE" FOR LATERAL EPICONDYLITIS

Wadih S. Macksoud, MD1 (n), Kurt A. Gasner, MD2, Jeffrey A. Deren, MD1 1Orlando, FL, 2St. Paul, MN

Retrospective chart review of 74 patients with refractory lateral epicondylitis of the elbow undergoing a “mini-open” release procedure was conducted. Details of the procedure are described. Demographic data, duration of symptoms, pre-operative treatment, failure rate and complications were identified. Four patients were lost to follow-up. All patients underwent the procedure with no post-operative immobilization prescribed. In fact, use of the extremity/elbow was encouraged. Results showed 93% success rate (65/70) with return to full activity at an average of 44 days post-operatively. Complications included short duration (minor) RSD symptoms in one patient and spontaneously resolving post-op subcutaneous hematoma in another. This “mini-open” release technique is a safe, easy, and effective treatment for refractory lateral epicondylitis of the elbow with limited negative post-operative impact on the lives of our patients.

Paper #11

PATIENT PERCEIVED FUNCTION AND LONG TERM FUNCTIONAL RESULTS AFTER ALLOGRAFT ACL RECONSTRUCTION

David N. Caborn, MD (a,b-CryoLife), John Nyland, EdD, PT, SCs, ATC

Shea, Tillett, Malkani & Caborn, Louisville, KY

Although allograft tissue use has become increasingly popular, few studies have reported patient perceived function or long-term outcome results. This two-phase study evaluated patient self-reports and objective knee joint function at > 2 years following primary ACL reconstruction with allograft tissues. Survey packets including the IKDC 2000 Knee Forms and the Knee Outcome Survey (ADL and Sports Activity Scales) were mailed to 335 patients. Survey packets were completed by 188 subjects (95 male, 93 female), 56% return. Patients were 5 + 2.8 years post-surgery (range = 2-12 years). Patients scored normal or nearly normal for the IKDC Subjective Knee Evaluation (66.4%), Knee Outcome Survey ADL Scale (79.8%), Knee Outcome Survey Subjective ADL Rating (83.5%), Knee Outcome Survey Sports Activity Scale (72.3%) and the Knee Outcome Survey Subjective Sports Activity Rating (65.9%). IKDC Current Health Assessment Subscores were comparable to national age group and gender matched normative values. Patients self-reported sports activity decreased from strenuous to moderate, sports activity frequency decreased from 4-7 times/week to 1-3 times/week, and overall knee function decreased 10% compared to pre-injury values. Clinical evaluation of objective knee joint function was performed on 65 patients with comparable survey results to the larger subject pool. Concomitant meniscus repair or debridement was performed on 66.2% of patients. Mild effusion (8%) and medial joint line tenderness (10.3) was experienced by some patients. Mean active involved side knee flexion was decreased 2.2% (140° vs. 143°). Mean involved knee anterior laxity was increased by < 2 mm compared to the uninvolved knee. Single leg broad jump scores were normal or nearly normal for 83.1% of patients. IKDC Clinical Examination scores were 78.5% normal or nearly normal. Instrumented single leg vertical hop revealed similar values for hop height, concentric phase force, and power production (P > .05), however eccentric phase force and power production were decreased compared to the uninvolved lower extremity (P = .04). Both patient self-reports and objective knee joint function examination revealed similar findings to previous reports of patients following primary ACL reconstruction using autograft tissues. Long-term lower extremity extensor eccentric function deficits at the side of knee injury and surgery may influence long-term function.

Paper #12

POSTERIOR SHOULDER INSTABILITY: DIAGNOSIS ACCURACY OF PHYSICAL TESTS

Edward G. McFarland, MD (n), George El Rassi, MD, Harpreet S. Gill, MD, Hyung B. Park, MD, Harpal S. Selhi, MD

Lutherville, MD

Introduction: The goal of this study was to evaluate the diagnostic value of common clinical examinations for posterior shoulder instability which has not been previously reported. Materials and Methods: 308 patients who subsequently underwent diagnostic arthroscopy and shoulder surgery were prospectively examined for signs of posterior instability. All patients underwent a preoperative physical examination which included: (a) posterior apprehension sign for the presence of pain, subluxation, or apprehension; (b) posterior laxity testing using a posterior drawer sign with the patient supine; (c) evaluation of the voluntary component of instability. Of this cohort, 12 underwent a posterior stabilization for symptomatic instability and 296 were utilized as the control group for statistical analysis. A diagnostic score was developed using standard scale development techniques, including multivariate logistic regression and ROC (receiver operator characteristic) curves. Results: Overall sensitivity of the tests was low (less then 50 percent for all tests), while specificity was high (better than 80 percent for all tests). A diagnostic score was developed combining the five best independent predictors of posterior shoulder instability: voluntary instability (p less than 0.001), the subluxation (p less than 0.001) and pain (p less than 0.05) components of the posterior apprehension sign, and the pain (p less than 0.2) and symptom reproduction (p less than 0.001) components of the posterior drawer sign. This combined score ranged between 0 and 6, and a cutoff value of 2 resulted in sensitivity and specificity of 83 and 89 percent, respectively. ( Table 1). Conclusion: While most signs for posterior instability are not sensitive, a newly devised scoring system can make the diagnosis in over 83 percent of cases. Table 1. The accuracy of various tests for posterior shoulder instability.

Sensitivity Specificity PPV NPV AC LR

Voluntary Instability 0.42 1.00 1.00 0.98 0.98 >

99.99 PA - Subluxation 0.25 0.98 0.38 0.97 0.95 14.50 PA - Apprehension 0.08 0.99 0.25 0.96 0.95 8.06 PD-Apprehenssion 0.42 0.92 0.19 0.97 0.90 5.30 PA - Pain 0.50 0.86 0.13 0.98 0.84 3.45 PD- Pain 0.50 0.82 0.11 0.97 0.80 2.72 Algorithm A (cutoff = 1) 0.92 0.79 0.15 1.00 0.80 4.45 Algorithm A (cutoff = 2) 0.83 0.89 0.23 0.99 0.88 7.35 Algorithm A (cutoff = 3) 0.75 0.92 0.27 0.99 0.91 9.09 Algorithm A (cutoff = 4) 0.67 0.97 0.44 0.99 0.95 19.40

PA= Posterior apprehension. PD= Posterior drawer. PPV=Positive predictive value. NPV= Negative predictive value. AC= Accuracy. LR= Likelihood ratio.

Paper #13

COMPARISON OF A NOVEL FIBERWIRE-BUTTON CONSTRUCT VERSUS METALLIC SCREW FIXATION IN A SYNDESMOTIC INJURY MODEL

Michael D. Stover, MD (a-Arthrex), Kevin Forsythe, MD, Kevin Freedman, MD, Avinash Patwarden, PhD

Loyola University Medical Center, Maywood, IL

Background: There is minimal experience with a syndesmotic fixation device that is less rigid in order to approximate the normal mechanics of the distal tibio-fibular relationship during healing. This study evaluates the ability of a novel Fiberwire-button implant (Arthrex, Naples, FL) to maintain syndesmotic reduction as compared with a metallic screw. Hypothesis: A novel Fiberwire-button implant will be as effective as a metallic screw for maintenance of ankle syndesmotic reduction in an external rotation model. Study Design: Controlled laboratory study. Methods: Twelve matched fresh-frozen pairs of cadaveric lower extremities were used. In phase one, the ankles were tested (12.5 N-m external rotation force) with all ligaments intact to establish physiologic syndesmotic diastasis. In phase two, the same force was applied to the ankles after sectioning of the syndesmotic and deltoid ligaments. In phase three, within the pairs, each limb was randomly assigned to receive either the Fiberwire-button implant or a metallic screw (Synthes, Paoli, PA); the ankles were then tested for syndesmotic diastasis with a progressive external rotation force, from 2.5N-m to 25N-m (or failure). Results: There was no significant difference in diastasis amongst pairs with either an intact or sectioned syndesmosis (p=0.64, p=0.80 respectively). There was significantly greater diastasis in the Fiberwire-button group at all external rotation loads (p<0.0001). Nine of the ten pairs failed, all through fracture of the distal fibula. There were no hardware failures. The metallic screw group failed, on average, at 15N-m, whereas the Fiberwire-button group failed at approximately 18N-m (significant difference, p=0.0004). The metallic screw group maintained syndesmotic reduction up to, and including, 5N-m of force. Conclusions: The Fiberwire-button construct was unable to maintain syndesmotic reduction of the ankles at any of the external rotation forces applied. The ankles fixed with the Fiberwire-button implant demonstrated significantly greater widening of the syndesmosis in comparison to the metallic screw, at all applied loads. Clinical Relevance: The Fiberwire-button implant may not maintain adequate ankle syndesmotic reduction in the immediate post-operative period relative to a metallic screw.

Paper #14 CLINICAL OUTCOME OF THE TIBIAL INLAY PCL RECONSTRUCTION:

MID TERM RESULTS AND REVIEW OF THE LITERATURE

Richard D. Parker, MD (n), William Wind, MD, John A. Bergfeld Cleveland Clinic Foundation, Cleveland, OH

Treatment of posterior cruciate ligament (PCL) injuries continues to be a significant area of debate among orthopaedic surgeons. Most would agree that current surgical indications include symptomatic grade III laxity, combined injury patterns and bony avulsion fractures. Determining the optimal method for PCL reconstruction has been difficult. In this presentation we present our experience using the single-bundle tibial inlay technique for PCL reconstruction. Methods: We evaluated 25 patients at mid term follow-up following both isolated and combined reconstructions using the single-bundle PCL inlay technique. Our patient population consisted of 25 patients who were treated using the PCL inlay technique between 1993 and 2001. All patients who underwent PCL reconstruction using the PCL inlay technique and were available for at least 18 months of follow-up were included in the study. Those patients requiring additional ligament surgery at the time of PCL inlay reconstruction were also included. Exclusion criteria involved those patients requiring revision PCL surgery, contralateral PCL injury and those with severe ipsilateral lower extremity trauma. Results: All patients enrolled in the study completed the subjective portion of the MOON (Multicenter Orthopaedic Outcomes Network) form. There were no statistically significant differences in the subjective scores between the patients with isolated versus combined injury. However, there was a trend toward inferior results in patients with the combined injury patterns. At a final follow-up averaging 49.2 months, there was an average flexion loss of approximately 7 degrees by goniometer testing. Final ROM averaged 0 degrees of extension (range, -3 to +10) and 131 degrees of flexion (range, 110-145). Resting posterior sag was measured with the knee at 70 degrees. In 68% (17/25) of the knees, the tibial plateau was anterior to the femoral condyles. The plateau was flush with the condyles in the remaining 32% (8/25) of the knees. Approximately half of the patients (13/25) had 3-5mm of total AP laxity with the knee at 70 degrees of flexion. Twenty-eight percent (7/25) had 0-2mm of laxity and 20% (5/25) had 6-10mm of laxity. When a straight posterior drawer was applied with the knee in a resting position at 70 degrees of flexion, 72% (18/25) had 2mm or less of posterior translation. Twenty-four percent (6/25) had 3-5mm and 4% (1/25) had 6-10mm of posterior translation with clinical testing. Almost all patients 96% (24/25) had a firm posterior drawer endpoint. Twenty-eight percent (7/25) had a grade I (glide) reverse pivot shift with the remainder having a negative reverse pivot shift. All patients with a residual grade I reverse pivot shift had an associated PLC injury. Arthrometric KT 1000 testing at 20lb of force and the knee at the quadriceps neutral angle resulted in a average SSD of 1.6mm of posterior translation between the involved and uninvolved knees. There were no statistically significant differences when comparing the involved and uninvolved knees with respect to functional and Biodex testing. Eighteen of the 25 knees had >90% single-leg hop for distance compared to the unaffected extremity. Instrumented stress radiography was evaluated in all patients with measurements made by an independent radiologist. Average SSD between involved and uninvolved knees was 6.3mm (range, 4.8mm - 11.5mm). Patients with isolated PCL reconstruction had less posterior translation (average, 5.0mm) on instrumented testing compared to patients with combined injury (average 6.5mm). There were no significant differences between the isolated and combined reconstruction in terms of the development of arthritic changes. MRI revealed all PCL grafts to be intact with an average thickness of 11.2mm (range, 9-14mm). Final objective IKDC scores showed 6 normal (A), 12 nearly normal (B), 6 abnormal (C) and 1 severely abnormal (D). The severely abnormal patient suffered a knee dislocation along with a peroneal nerve palsy. Mean Lysohlm score was 84. There were two complications of arthrofibrosis requiring post operative manipulation and arthroscopic lysis of adhesions. Further surgery was necessary in an additional five patients. Three had removal of retained femoral hardware. The patient with the preoperative peroneal nerve palsy required a later neurolysis and one patient deteriorated to a total knee arthroplasty 16 months following the index procedure.

Paper #15 ARTHROSCOPIC FINDINGS IN ACETABULAR DYSPLASIA: A PARADIGM TO EVALUATE CHONDRAL

INJURIES

Joseph C. McCarthy, MD (b-Innomed, Arthrex), Jo-ann Lee, MS Boston, MA

Introduction: In hip dysplasia conventional radiographs, including high-contrast gadolinium-enhanced MRI, are not always sensitive enough to diagnose a labral tear or chondral pathology. This study examines the relationship between mild acetabular dysplasia, labral and acetabular cartilage injury and development of early osteoarthritis. Materials and Methods: Between 1989 and 2000, 170 hips in 163 patients with mild acetabular dysplasia underwent arthroscopic evaluation of their hip. Acetabular dysplasia is defined as a patient with a lateral Central Edge (CE) Angle of less than 27 degrees (range 19–26). Surgical findings were classified by location and by severity of the chondral lesions of the femoral head, acetabulum and labrum. Results: Of the 170 hips with dysplasia, 122 of these had labral tears (72%) at the free-margin articular surface and 113 were anterior (66%). One hundred hips (59%) had anterior acetabular chondral lesions. Among the 113 patients who had anterior labral tears, 78 hips (69%) had anterior acetabular chondral defects, and 44 hips (39%) had anterior femoral head chondral lesions. Conclusion: Even in mild dysplasia uncovering of the anterior femoral head subjects the labrum to increased load and potential susceptibility to tearing. Labral tears may contribute to or can occur in association with articular cartilage lesions of the contiguous femoral head or acetabulum. Findings in this study support the concept that labral disruption is frequently part of the continuum of degenerative joint disease.

Paper #16

CARTILAGE DEGENERATION IN NORTH AMERICAN AND SAUDI ARABIAN OSTEOARTHRITIC KNEES

W. Andrew Hodge, MD (n), Melinda K. Harman, MSc, Scott A. Banks, PhD

The BioMotion Foundation, West Palm Beach, FL Introduction: The anterior cruciate ligament (ACL) affects the tibiofemoral contact location and cartilage degeneration in osteoarthritic knees. ACL deficient knees have more posterior tibial degeneration than ACL intact knees. Considering that preoperative knee deformity can predict polyethylene wear after total knee replacement (TKR), comparisons of ACL integrity and cartilage degeneration in different patient populations provides valuable insights about knee function and strategies for successful TKR. This study compares cartilage wear patterns in Saudi Arabian (SA) and North American (NA) patients’ knees. Methods: 32 SA knees and 42 NA knees underwent TKR for osteoarthritis and varus malalignment. The ACL was intact in all knees prior to tibial resection. The area and location of cartilage wear were measured on the resected tibial articular surfaces. Results: SA tibias had larger, more anterior cartilage wear. Wear occupied 46%+19% and 37%+16% of the medial plateau on SA and NA tibias, respectively. Wear patterns were located >5 mm anterior to the medial plateau center on 34% and 29% of the SA and NA tibias, respectively. Despite varus malalignment, 22% of the SA tibias and 74% of the NA tibias had wear on both the medial and lateral plateaus. Conclusions: SA knees have elongated cruciate ligaments because of chronic stretching during deep knee flexion activities. Extreme anterior wear patterns correspond to anterior tibiofemoral contact and may be characteristic of substantial (30º) tibial internal rotation occurring with deep flexion. Restoring knee alignment and accommodating the altered knee kinematics that occur with ACL deficiency after TKR could prove beneficial for deep flexion activities. Summary: Comparisons of ACL integrity and cartilage degeneration in different patient populations provides insights about knee function and successful TKR strategies.

Paper #17

THE OSTEOARTHRITIS MATRIX

Scott F. Dye, MD1 (n), Mailine Chew, MD2, Christopher Dye, Rebecca M. Larsen 1University of California, San Francisco, CA, 2California Pacific Medical Center, San Francisco, CA

To the present, the conceptualization of the development and progression of osteoarthritis (OA) of knee and other joints has been determined solely by the structural changes, as seen on radiographs (e.g., Fairbank changes and the Kellgren-Lawrence scale.) Consequently, the very lexicon of OA is replete with structural terminology (osteophytes, osteosclerosis, osteopenia, joint space narrowing, cysts, etc.) that restricts and limits one’s concept of the condition. Over the past decade, observing the dynamic osseous tissue homeostasis characteristics of the adult knee by the use of sequential technetium 99m-MDP scintigraphy has led us to a profoundly different conceptualization of the natural history of osteoarthritis. We have found that the persistent loss of osseous homeostasis (as manifested scintigraphically) identified knees at risk of developing OA at a time when radiographs are still normal. Therapeutic intervention of this “pre-arthritis stage” that results in restoration of osseous homeostasis aborts the development of OA of the knee. Also, the natural history of OA of the knee is often a process of punctuated equalibrium with periods of metabolic activation and progression, interspersed with periods of relative metabolic dormancy and non-progression. We have developed a simple method of manifesting the possible combinations of structural and metabolic factors that may pertain in a given joint at a given time that allows visualization of the actual dynamics of OA. We term this the Osteoarthritis Matrix. This Matrix presents the radiographic data in order of severity (0,1,2,3) on the horizontal axis - abscissa , and the severity of a concurrently obtained bone scan (0,1,2,3) on the vertical axis - the ordinand. (See attached.) We have documented all possible examples of structural and metabolic characteristics in the human knee, with the exception of bone scan 0, x-ray 3. The use of this matrix allows a simple method to change one’s conceptualization of the dynamic character of OA of the knee (and other joints) by the addition of osseous metabolic data to the already familiar structural data. The use of this graphic representation of metabolic and structural information can provide a new way in which the effectiveness of current and future therapeutic methods of treating OA may be more accurately evaluated.

Paper #18

THE HISTORY OF ORTHOPAEDIC SURGERY AT THE PRESIDIO OF SAN FRANCISCO 1776–1995

Eugene G. Galvin, MD (n)

San Francisco, CA

The Presidio of San Francisco was the longest continuously serving military base in the United States, 1776 to 1995. It was under Spanish control until 1821, then under Mexican Control until 1846 when the Americans took over. All these transfers were done without firing a shot. The first permanent hospital was built in 1857, Wright Army Hospital, which today is a museum. In 1898 at a cost $103,333.50, the first general hospital in the Army was built to serve those injured in the Spanish American War in the Philippines. Being undamaged by the Great San Francisco Earthquake of 1906, it cared for the majority of the injured. In 1911 it was re-named Letterman Army Hospital for Maj. Jonathan Letterman, the Medical Director of the Army of the Potomac during the Civil War and later the Medical Director (coroner) of San Francisco. During World War I it expanded greatly and served as the Amputation Center for the entire Army. In the 1920’s the first internships and residency training programs were instituted. The orthopaedic training program started in 1947. During the Great Depression it provided medical care for the western states for the Conservation Corps. During World War II, Letterman was the main treatment facility for the war in the Pacific. On one day alone in 1945, over 5000 wounded soldiers were admitted. To transfer patients to facilities closer to their homes, a network of hospital trains was created headquartered at the Presidio. Early in the war, the Army Surgeon General, Gen. Norman Kirk, an orthopaedist and former Commander of Letterman, recruited Dr. Sterling Bunnell to create a series of hand surgery centers throughout the country headquartered at Letterman. A group of young general and orthopaedic surgeons were “volunteered” to become hand surgeons to include Ray Curtis, William Littler, Paul Brand, etc., and so started hand surgery in this country. The Korean War and Vietnam brought huge influxes of the wounded. By the time of the Vietnam War, with the advent of helicopters and the medivac system, casualties reached the hospital within days or even hours of being wounded on the battlefield. In 1972, the Army’s first total joint replacement center and fellowship were started by then LTC John Feigan. On October 17, 1989, the Loma Prieta Earthquake struck San Francisco. Being the only hospital with a functioning heliport, Letterman received the majority of the casualties. After serving in the First Gulf War, the Presidio and Letterman Army Medical Center were stood down as the first casualties of the military cutbacks at the end of the Cold War and the BRAC (Base Realignment and Closure) Act of Congress. It has a long, proud and dedicated history of service to the military, San Francisco and the United States.

Paper #19

ORTHOPAEDIC DISORDERS IN FICTIONAL LITERATURE

Marlene De Maio, MD (n) Bone & Joint/Sports Medicine Institute, Portsmouth, VA

Purpose: The goal of this talk is to review some interesting orthopaedic disorders described in fictional literature. This review will show that orthopaedic disorders have been clearly recognized by respected authors, that orthopaedic surgeons are familiar with these authors and their fictional creations, and that certain of these conditions are more prevalent in fictional literature. Materials and Methods: A medline search and literary search were conducted to find fictional characters with orthopaedic disorders. Only five citations were obtained by this method. Psychiatric, medical, and substance abuse conditions are much more frequent. Results: The characters with orthopaedic disorders may be categorized by true syndromes, developmental disorders, infectious disease, and post-traumatic conditions. Tiny Tim (A Christmas Carol, C. Dickens) had vitamin D resistant rickets. Ichabod Crane (The Legend of Sleepy Hollow, W. Irving) most likely had Marfan’s syndrome. Mr. Johnson (Lady Susan, J. Austen) suffered a gout attack and gout is described 500 years earlier in the Canterbury Tales. The Munchkins of Oz comprise the gamut of dwarfism types (L. Frank Baum). The Seven Dwarfs were most likely achondroplasts. Who can forget the kyphoscoliosis of The Hunchback of Notre Dame (V. Hugo) and Igor in Frankenstein (M. Shelley)? With respect to Forest Gump, his polio was not lethal, but Little Nell (The Old Curiosity Shop, C. Dickens) and Mimi’s (La Boheme, G. Puccini) tuberculosis was. Some of the most vivid characters are those who sustained amputations. These include upper extremity amputations (Beren, Lord of the Rings; Captain Hook, Peter Pan), lower extremity amputations (Captain Ahab, Moby Dick; Long John Silver, Treasure Island), and upper and lower extremity amputations (Joe Bonham, Johnny Got His Gun). Conclusion: While psychiatric disease, substance abuse, and medical disease account for most of afflictions described in literary fiction, there are noteworthy examples of orthopaedic disorders. Amputees and characters with kyphoscoliosis appear with higher frequency and have very colorful descriptions of their afflictions and personalities. Amputees are often portrayed as villains. Dwarfs and sick children are often described tenderly. There is a lack of research in this area. No peer reviewed papers were found addressing this topic. The orthopaedic community could raise itself to the literary level of psychiatrists and internists if we addressed this topic aggressively.

President’s Presentation

CHARLES O. TOWNLEY: DESIGN BY REASON

James W. Pritchett, MD Seattle, WA

The career of Charles O. Townley tells a nearly complete story of the development of joint replacement surgery. The basic concept for all his procedures was to replicate normal anatomy. His designs were calculated to transmit stress to supporting bone in a physiologic fashion. He validated this approach both in the laboratory and through clinical research. Townley envisioned the ideal replaced joint as “pseudo biological” with porous implanted material serving only as an articular barrier to allow joint movement and pain relief while fibrous tissue in growth restored a soft joint surface. This goal has not yet been realized. Townley did not wait for the development of either polyethylene or methylmethacrylate to perform joint replacement surgery. He used alternative cements and polymers. Since 2001 he has held a patent on a new polymer, which he is testing. The Townley Anatomic Knee, which retained both cruciate ligaments, was widely used worldwide in the 1970’s and 1980’s. It was one of the first total condylar knee prostheses. The TARA (Total Articular Replacement Arthroplasty) with its characteristic curved stem has been one of the most enduring of the surface replacement hip prostheses in both its metal and ceramic versions. Starting in the 1940’s, Townley designed implants for the toe, hand, shoulder, ankle, and elbow in addition to hip and knee implants. While maintaining a busy private practice in Port Huron, Michigan for 53 years, he was a tireless innovator and investigator. He developed his concepts, implants and instruments primarily without industry support and took a finished and tested product to implant manufacturers. Prior to the enactment of Medicare, Townley provided his implants at no or nominal cost to patients. He spent years defining the healing of fibrous tissue in anterior cruciate injury and shoulder dislocations. He performed human and animal dissections and published techniques for repair. In the 1950’s, he was interested in compression fixation for fractures, osteotomies, and arthrodeses. Townley lectured widely around the world and published a number of original papers from 1947 to 2004. He was a founding member of the Knee Society and its president in 1988. He has been a member of the Association of Bone and Joint Surgeons since 1962, presenting eleven scientific papers.

Paper #20

TWO-INCISION MINIMALLY INVASIVE SURGERY: ONE SURGEON’S INITIAL EXPERIENCE

Bradley K. Vaughn, MD (c-Zimmer), Liz Fuller, PA, Huong Nguyen Raleigh, NC

Interest in less invasive surgical approaches has recently increased with minimally invasive surgery. We compared acute complications and early outcomes of the two separate incision technique to a standard antero-lateral approach. The first 50 consecutive Minimally Invasive Surgery (MIS) 2-incision cases were reviewed. Operative times, fluoroscopy times, estimated blood loss, and incision measurements were recorded at surgery. Operative and post-operative complications were documented. Height and body weight were obtained to determine Body Mass Index. Post-operative Harris Hip Scores and radiographs with component position were reported on all patients. The patients were compared to demographically matching standard approach patients. Average BMI for the 16 females and 33 males was 27. Mean fluoro time was 59 seconds. Operative time averaged 100 minutes. Estimated blood loss averaged 410.5 cc. Skin incision lengths averaged 5.7 cm. anterior and 3.19 cm. posterior. A trochanteric fracture and Stage 2 femoral neck fracture were reported operative complications and two avulsed trochanters were reported post-operatively. Mean Harris Hip Score was 94. No infections or dislocations occurred, and radiographic review revealed increased variability of acetabular component position. With increased demand for total hip arthroplasty comes the search for newer techniques to satisfy patient and surgeon pursuit of decreased morbidity and better outcomes. These new techniques cannot be arbitrarily applied to all patients. Although patients have higher recovery scores and shorter hospital stays, the 2-incision procedure is technically demanding and complications occur. Long term results will be examined to determine the efficacy of this method compared to the standard approach.

Paper #21

COMPLICATIONS AFTER TOTAL HIP ARTHROPLASTY: RESULTS OF THE ASSOCIATION OF HIP AND KNEE SURGEONS (AAHKS) SURVEY

Richard D. Heekin, MD (n)

Jacksonville, FL The Research Committee of the American Association of Hip and Knee Surgeons (AAHKS) initiated a survey among AAHKS members to determine the incidence of some of the more serious complications after total hip arthroplasty. For primary total hip arthroplasty, responding surgeons reported a 0.119% rate of sciatic nerve palsy, 0.009% rate of femoral neurovascular injuries, 0.08% fatal pulmonary emboli, 0.66% deep venous thrombosis, 0.45% recurrent late (> 6 weeks post-op) dislocation requiring re-operation, 0.08% acute (< 3 weeks post-op) sepsis, 0.15% chronic sepsis, 0.33% mechanical implant failure requiring re-operation, 0.66% polyethylene failure requiring re-operation, and a 2.1% revision rate among primary total hips for all causes. Conversion and revision hips had a higher incidence of thromboembolic and neurovascular complications; fewer patients had complete resolution of neurologic deficits. Revisions were four times more likely and conversions three times more likely than primary hips to experience recurrent late dislocation requiring re-operation. Revision hips had a fourfold greater risk of both acute and chronic sepsis compared to primary hips. While revisions were twice as likely to experience non-polyethylene mechanical failure, primaries had a 50% greater risk of polyethylene-related failure than revisions. Key words: total hip arthroplasty, complications, AAHKS, survey.

Paper #22

FAILURE MECHANISM OF POLISHED STEMS WITH FOUR DIFFERENT GEOMETRIES (SINGLE SURGEON SERIES)

Dennis K. Collis, MD1 (a-Zimmer, Inc.), Brian A. Jewett, MD1, Craig G. Mohler, MD1, Daniel C. Fitzpatrick, MD2

1Eugene, OR, 2Springfield, OR Introduction: Experience with loosening and lysis with rougher surface cemented stems in the 1980s convinced the senior surgeon to use only a polished surface on stems cemented after 1992. Having seen significant lysis of the bone with loosening within the first five years with some rougher surface stems which made revision difficult, the senior surgeon was interested in the early results of polished stems. Materials and Methods: Between 1993 and January 2004, 1031 consecutive cemented stems with a polished surface of 0.4 microinches were used. These stems were of four somewhat different geometries. Beginning in 1993, 120 polished Iowa stems with an oval stem geometry were inserted. Beginning in 1994, 148 Triumph stems with a more rectangular geometry and larger size were inserted. The Heritage stem, which had a proximal geometry similar to the polished original Charnley prosthesis of several decades earlier, was introduced in 1996, and 283 of these stems were cemented. The Versys Heritage cemented stem, which duplicated the proximal Charnley geometry even more closely and added significant taper distally for centralization, was introduced in 1999. 480 of these stems were inserted. Throughout this 11-year period, for the most part in patients under 60 years of age, an uncemented stem was used. However, 106 of these 1031 stems were in patients under the age of 60. 189 patients weighed more than 200 pounds. This group was 36 percent male, 64 percent female. Intensive recall follow-up on the first 100 stems inserted in each of the four geometries was done to determine if there were any radiographic or clinical differences, or any unknown revisions. Results: As of March 2005, 6 of the 1031 stems have required revision: 1 for infection; 1 Iowa polished stem had lysis at five years which progressed to full bone-cement interface loosening requiring revision at 10 years; 1 Triumph stem developed lysis at four years which resulted in fracture requiring revision at 10 years; 2 Heritage stems loosened at the bone-cement interface requiring revision at four and eight years; and 1 Versys Heritage stem loosened at four years at the bone-cement interface. (This stem had been painful throughout the postoperative period and was not thought to have good bone-cement interdigitation.) The effort to update the follow-up of the first 100 stems in each of the four geometries found no additional failures and only three hips were noted to have some minor bone-cement lysis on current radiographs. 91 of the 400 patients were deceased. Average follow-up of the 309 living patients in this group was 5.6 years and 234 had 5 or more years of follow-up. Conclusions: Polished cemented stems are quite successful in the midterm, with a revision rate at 5 years postoperative that is one-fifth that of the previously-used cemented rougher surface stems. All revisions started with loosening at the bone-cement interface and there was much less bone loss than that seen with rougher surface stems. The polished stems were also slower to require revision when they became loose than the rougher surface stems.

Paper #23

CLINICAL FOLLOW-UP COMPLIANCE IN TOTAL JOINT ARTHROPLASTY PATIENTS

John C. Clohisy, MD (n), Gregory D. Byrd, Madelyn C. Curry, BSN, Rick W. Wright, MD St. Louis, MO

Introduction: Routine radiographic analysis of total joint arthroplasty (TJA) has been recommended to ensure early detection of polyethylene wear, osteolysis and to prevent catastrophic implant failure. The purpose of this study was to evaluate early clinical follow-up compliance in patients undergoing total hip arthroplasty (THA) or total knee arthroplasty (TKA). Methods: Retrospective review of clinical follow-up compliance for THA (372 primary, 145 revisions) and TKA (201 primary, 78 revisions) was performed. All 796 patients were treated by one surgeon and were instructed to return for 1 and 2 year follow-up visits. Follow-up compliance was assessed from our hip and knee registry. A logistical regression analysis was used to model likelihood of returning for follow-up. Results: Patient follow-up with primary THA was 60% at 1 year and 31% at 2 years while revision THA was 54% at 1 year and 34% at 2 years. Primary TKA follow-up was 66% at 1 year and 36% at 2 years. The revision TKA patients had 57% return for 1 year and 28% for 2 years. Higher pre-operative Harris Hip Scores in THA patients were associated with an increased frequency of follow-up (p<0.001). Caucasian THA patients trended toward being more likely than non-Caucasian patients to return for 2 year follow-up (p=0.051). Conclusion: Clinical follow-up compliance of TJA patients was poor in this series with an average of 32% returning for their 2 year follow-up visit. These data raise concern that routine clinical and radiographic surveillance of TJA patients is limited by poor patient follow-up compliance.

Paper #24

MUSCLE STRAIN DURING SIMULATED TOTAL HIP ARTHROPLASTY

Michael J. Grecula, MD (n), William L. Buford, PhD, Jason P. Norcross, MD University of Texas Medical Branch, Galveston, TX

Introduction: The total hip arthroplasty (THA) procedure involves extreme positioning of the limb to accomplish the tasks of hip dislocation, acetabular preparation, femoral preparation, and component insertion. With the increasing interest in minimally invasive “muscle sparing” THA, knowledge of the strain and elongation of the individual muscles around the hip during these maneuvers is important. Methods: A whole-body kinematic simulation developed in our lab was used to interactively move the lower limb into the positions needed to perform a total hip arthroplasty using a minimally invasive posterior approach. The following muscles were enabled in the simulation: Piriformis, Obturator Internus, Obturator Externus, Quadratus Femoris, Gluteus Maximus, Medius, and Minimus. During the simulation, joint angles, muscle origin to insertion path lengths, and moment arms were monitored. Comparisons were made at four static positions corresponding to the leg position with hip dislocation, acetabular preparation, femoral preparation, and component insertion. Resting muscle fiber length was measured during a fresh cadaver dissection and also calculated using the whole body kinematic structure. Muscle strain was then calculated as change in fiber length relative to resting length. Results: Over all of the muscles studied, the measured strain ranged from 89% to 277%. The average strain was 157%, with a standard deviation of 57%. The lowest strain was seen in the anterior fibers of the gluteus medius. The quadratus femoris underwent the greatest strain, ranging from 207% in the hip dislocation position to 277% in the femoral component insertion position. Discussion/Conclusion: Much of the posterior muscle architecture was placed at or beyond the ultimate tensile strength of fresh cadaver or living muscle as reported in the literature. This highlights the importance of considering the potential for indirect trauma to the peri-articular muscles during a minimally invasive approach. This 3D kinematic simulation can be a valuable tool in developing and evaluating different MIS approaches which can minimize direct and indirect tissue trauma.

Paper #25

UNICOMPARTMENTAL OSTEOARTICULAR ALLOGRAFTS OF THE KNEE: SURVIVAL ANALYSIS AND COMPLICATIONS

Miguel A. Ayerza, MD (n), D. Luis Muscolo, MD, Luis Aponte-Tinao, MD, Eduardo Abalo, MD

Hospital Italiano, Buenos Aires, Argentina Introduction: Unicompartmental osteoarticular defects of the knee are challenging due to demands of stability and function of this weight-bearing joint. Prosthetic reconstructions often require sacrificing the uninvolved compartment. Osteoarticular allograft reconstructions can restore the anatomy, and allows reattaching soft tissue structures such as meniscus and ligaments from the host. The purpose of this study was to perform a survival analysis of unicompartimental osteoarticular allografts of the knee and evaluate their complications. Material and Methods: Forty unicompartmental osteoarticular allografts of the knee performed in 38 patients during the period 1962-2001 were followed for a mean of 10 years. Twenty-nine transplants were located at the femur (11 medial and 18 lateral) and 11 at the tibia (4 medial and 7 lateral). Allografts survival was determined with the use of the Kaplan-Meier method. Complications as local recurrence, fracture, articular collapse and infection were analyzed. Results: The global rate of allograft survival was 85% at five years. There were 8 complications in 6 patients: 2 local recurrences, 2 infections, 1 fracture, 1 massive resorption and 2 articular collapses. Six allografts were removed and they were considered failures. All were reconstructed with a second allograft that included 2 unicompartmental and 4 bicompartmental. There were two articular collapses that were converted to allograft prosthesis. Discussion: Although the incidence of reoperations due to allograft complications is high, the allograft survivor rate at five years was 85%. Unicompartmental allografts appear to be an alternative in those situations in which the massive osteoarticular bone loss to be reconstructed is limited to one knee compartment. The purpose of this study was to analyze survival and complications of unicompartimental osteoarticular allografts of the knee.

Paper #26

A KINETIC AND KINEMATIC ANALYSIS OF A NEW TRABECULAR METAL AUGMENTATION PATELLA

Bassam A. Masri, MD, FRCSC1 (a-Zimmer), John Mountey, MB, BCh, FRCS(Orth) 2, Michael Paice, BSc3, Nelson

Greidanus, MD, FRCSC3, David Wilson, PhD3 1Vancouver, BC, Canada, 2Southampton, United Kingdom, 3Vancouver General Hospital, Vancouver, BC, Canada

Purpose: To evaluate the kinetic and kinematic function of a new trabecular metal augmentation patella. Methods: Eight frozen human cadaveric knees (mean age 68 years) were prepared for an “Oxford type” knee rig. Alignment and offset were addressed and the soft tissue envelope was kept as intact as possible. A load cell was introduced into the extensor mechanism. Femoral, patellar and tibial motion were recorded using the Optotrak system. Extensor mechanism forces were measured for the native knee, knee replacement with a polyethylene patellar button, the same TKR after excision of the patellar button (patelloplasty simulating patellar button excision after a revision TKR), and for the same knee replacement after the patelloplasty was reconstructed using a trabecular metal augmentation patella (Zimmer, Warsaw, IN). The quadriceps tendon force was measured as a function of degree of flexion as the knee was going from flexion to extension (extension force) and from full extension to flexion (flexion force). Results: The mean and maximum extensor mechanism forces in extension and flexion were lowest for the native knee (p<0.05). For the TKA group, the only statistically significant difference was noted when the patelloplasty was performed, with higher forces in extension, and higher forces in flexion at the initiation of the flexion arc, and lower forces after 45 degrees of flexion (p=0.024 flexion force; p=0.00015 extension force). This difference was normalized with the augmentation patella (p=0.198). The mean and maximum patellar mediolateral maltracking in flexion and extension were higher for the patelloplasty group compared with the other TKA groups (p=0.024), but this was normalized with the augmentation patella (p>0.05). Significance: This study demonstrates that the augmentation patella restores the abnormal tracking and higher extensor mechanism forces seen after patelloplasty in revision TKA to those normal after a TKA, and could potentially improve the results of TKA when the patella is otherwise not resurfaceable.

PAPER #27

3D GAIT ANALYSIS BEFORE TOTAL KNEE ARTHROPLASTY FOR PRIMARY OSTEOARTHRITIS. A PREOPERATIVE COMPARISON OF THE INVOLVED KNEE TO THE UNINVOLVED

James M. Ray, MD (n), William J. Kuminka, MS

National Training Center Sports Medicine Institute, Clermont, FL Introduction: Treatment of osteoarthritis by total knee arthroplasty corrects mechanical deformities in order to improve function. Gait characteristics in patients diagnosed with primary osteoarthritis were examined before unilateral total knee arthroplasty. To objectively assess improvement of function after total knee arthroplasty certain gait characteristics were measured preoperatively using force plate analysis and 3D kinematic and kinetic analysis. Methods: 18 patients (10 males and 8 females) with the diagnosis of primary osteoarthritis of the knee were pre-operatively evaluated for kinetic and kinematic measurements using the Peak Motus motion analysis system and 2 AMTI force plates. Overall cadence, average gait velocity and 15 gait parameters comparing right and left differences (involved and uninvolved knees) were examined. In addition, 3D knee joint angles, angular velocities, angular accelerations, knee joint moments, joint forces, gait power and gait reaction forces were measured. All patients underwent total knee arthroplasty using the same implant system, by the same surgeon, and all followed the same post-operative rehabilitation protocol.

Results: Analysis of the 15 gait parameters for the entire group showed a statistically significant comparison of the involved knee to the uninvolved knee in regards to the stance phase (uninvolved) and the swing phase (involved) p > 0.05. 3D kinematic and kinetic analysis demonstrated changes in preoperative data in regards to ground reaction forces in the uninvolved knee versus an increase in angular velocity of the involved knee. In the 8 females there was noted a higher gait velocity in the involved knee as compared to the uninvolved knee at a p value >0.05. Discussion: Monitoring patient’s gait prior to total knee arthroplasty for primary osteoarthritis determines the baseline deterioration of ambulatory function. In this small study group, documentation of gait parameters were noted in order to explain altered function of an arthritic knee compared to the contralateral normal knee. Long term monitoring of ambulatory function is needed after total knee arthroplasty which would provide needed information in regards to total knee arthroplasty outcome studies.

Paper #28

USE OF REINFUSION DRAINS AFTER TOTAL KNEE ARTHROPLASTY (TKA) IN PATIENTS GIVEN LOW MOLECULAR WEIGHT HEPARIN (LMWH) FOR THROMBOEMBOLIC DISEASE (TED)

PROPHYLAXIS

Richard H. Walker, MD (n), Alexander C. Gordon, MD, Pamela Pulido, RN, Clifford W. Colwell, MD, Steven N. Copp, MD, Kace A. Ezzet, MD

Scripps Clinic, La Jolla, CA

Introduction: Use of a reinfusion drain, in combination with LMWH for TED prophylaxis, after TKA has not been previously reported. Use of LMWH has caused concern regarding an increased prevalence of postoperative hematoma, hemarthrosis, and blood transfusion after TKA. The purpose of this study was to determine the effect of a reinfusion drain on prevalence of wound complications and blood transfusion after primary TKA in patients using LMWH prophylaxis. Methods: A retrospective review of patients who underwent unilateral TKA from 2001 to 2003 at one institution was performed regarding wound complications and TKA outcome. No drain was placed in 180 TKA (NoDrn Group), with wound closure prior to tourniquet deflation. A reinfusion drain was placed in another 188 TKA (Drn Group), with tourniquet deflation and hemostasis prior to closure. All patients received LMWH 20 mg 6 hours postoperatively and 40 mg QPM for 7-10 days. Minor wound complications were defined as those treated with local wound care or oral antibiotics. Major wound complications were defined as those needing reoperation. A subcohort of 95 patients from the NoDrn Group, who each donated one unit of autologous blood preoperatively (NoDrn-Auto Group), was compared to 96 patients from the Drn Group, who did not donate autologous blood (Drn-NoAuto Group), regarding preoperative and discharge hemoglobin (Hgb) and overall and allogeneic transfusion rate. Results: The NoDrn and Drn Groups showed no difference regarding age, diagnosis, length of stay, symptomatic TED rate, manipulation rate, or flexion and arc of motion at one and three months postoperatively. However, the Drn Group showed a lower wound complication rate which approached significance (No Drn 12%, 4 major/12 minor; Drn 4%, 2 major/5 minor, p=0.08). The Drn-NoAuto Group had a higher mean preoperative and discharge Hgb and a lower overall transfusion rate (NoDrn-Auto 13.1 g/dL, 9.9 g/dL, 40%, respectively; Drn-NoAuto 13.5 g/dL, 10.5 g/dL, 16%; all p<0.05), but a higher allogeneic transfusion rate (NoDrn-Auto 5%; Drn-NoAuto 16%; p=0.0003). For both NoDrn-Auto and Drn-NoAuto Groups. In the NoDrn-Auto Group, the autologous blood wastage rate was 60%. In the Drn-NoAuto Group, mean reinfusion was 300 cc. Transfused patients had a significantly lower preoperative Hgb than patients who were not transfused in both the NoDrn-Auto Group (12.6 g/dL & 13.5 g/dL, respectively; p=0.0001) and Drn-NoAuto Group (12.7 g/dL & 13.7 g/dL, respectively; p=0.007). For both groups, a pre-operative Hgb of <13.0 g/dL was a significant risk factor for transfusion. For the NoDrn-Auto Group, preoperative Hgb did not predict allogeneic transfusion need. Conclusion: Use of a reinfusion drain for TKA patients receiving LMWH TED prophylaxis had no detrimental effect on TKA flexion, arc of motion, or need for manipulation, but resulted in a trend toward a lower wound complication rate (p=0.08). This study did not demonstrate efficacy of a reinfusion drain, regarding allogeneic transfusion rate, as a substitute for preoperative autologous blood donation. If a surgeon employs LMWH TED prophylaxis and endeavors to substitute use of autologous blood predonation with use of a reinfusion drain and tourniquet deflation/hemostasis prior to TKA closure, this study suggests that methods to assure a preoperative Hgb >13.0 g/dL should be employed in order to minimize allogeneic transfusion.

Paper #29

MOLECULAR MECHANISMS OF ARTHROFIBROSIS FOLLOWING TOTAL KNEE ARTHROPLASTY

Michael D. Ries, MD1 (n), Jay Parvizi, MD2, Joshua Sleeper, BS2, Teresa Freeman, PhD2, Marla Steinbeck, PhD2

1University of California, San Francisco, CA, 2Rothman Institute, Philadelphia, PA Introduction: Idiopathic arthrofibrosis, when no discernible causes for stiffness can be identified, affects between 4 to 8% of patients with TKA. The exact etiology of this condition is unknown. This study evaluates the molecular mechanism responsible for mediating aggressive periarticular fibrosis. Methods: Periarticular tissues from the affected knees of 12 patients with idiopathic arthrofibrosis were retrieved. 2x2 cm tissue samples were taken from various regions around the knee. Control tissues were taken from the same regions of the knee in patients undergoing primary knee arthroplasty, revision arthroplasty for periprosthetic infection, and revision surgery for aseptic loosening of the components. The surgical samples were processed either for immunohistochemistry or for isolation of cells for cell culture. Results: Histology revealed that the retrieved tissue from the stiff knees consisted of extensive lattice work of fibrotic tissue with abundant myofibroblasts and fibrocartilage. There was no evidence of mineralization. Large numbers of immune cells (mast and T-cells) were observed in the arthrofibrotic tissues. By immunohistochemistry we were able to confirm the abundant presence of various byproducts and markers of Reactive Oxygen Species (ROS) such as superoxide anion, H2O2, hypochlorous acid/chlorine gas (HOCl/Cl2) and proteolytic enzyme myeloperoxidase (MPO). Discussion: The exact etiology of idiopathic stiffness remains unknown. This study has confirmed that aggressive fibroblast proliferation occurs in these patients as shown by PCNA analysis. Furthermore, the fibroblasts escape the regulatory apoptosis process. This may be a consequence of the increased expression of COX-2 observed in these cells and the subsequent induction of Bcl-2 by COX-2 products. Immune cells were found in abundance in periarticular tissues of patients with stiffness. The release of ROS and various growth factors from these cells was confirmed to be the driving factor in inducing metaplasia and maintaining the aggressive fibroblast proliferation.

Paper #30

TREATMENT OF BONE DEFECTS AND LIMB LENGTH DISCREPANCIES IN CHILDREN BY LENGTHENING FOLLOWED BY A FREE VASCULARIZED FIBULAR GRAFT

Kenneth L. B. Brown, MD, MSc, FRCSC (n) BC Children's Hospital, Vancouver, BC, Canada

Introduction and Aims: The treatment of bone defects secondary to congenital pseudoarthrosis of the tibia, infections and tumors is problematic. The vascularized fibular graft has been used for many years as a way to improve blood supply and successfully achieve union. Lengthening the limb prior to grafting can improve outcomes. Methods: Forty-one patients with major bone defects secondary to tumor resections, infections and congenital pseudoarthroses had reconstruction with a vascularized fibular graft. Of these, 10 patients had limb length discrepancies which were treated by application of an external fixater for lengthening through the bone gap. Following restoration of length with an external fixater, a vascularized fibular graft was inserted to bridge the bone defect. The external fixater was not removed until union of the graft to the host bone and initial hypertrophy occurred. Results: The ten patients (5 males and 5 females) were aged 2.5 to 14.5 years (mean 7.6 years). The affected bones included eight tibias, one humerus and one ulna. The limb length discrepancies ranged from three cm to 20 cm (mean 6.44 cm). The duration of lengthening prior to definitive vascularized fibular graft ranged between one to 15 weeks (mean seven weeks) in nine patients. In the patient with a discrepancy of 20cm, lengthening spanned 52 weeks. At the time of the definitive vascularized fibular graft procedure, the fixater was partially disassembled to facilitate surgery and microvascular anastomosis. The frame was then reassembled and used as the fixation device to protect the graft. The fixater was removed from seven to 24 weeks (mean 16 weeks) after definitive surgery. There were no complications during the lengthening process. However, two patients experienced non-unions which were successfully treated by autologous bone grafts. One patient had a fracture of the vascularized fibular graft which healed uneventfully. Conclusion: A staged approach to reconstruction of major long bone deficiencies leads to a better outcome than insertion of vascularized fibular grafts without addressing the limb length discrepancy. In this type of procedure, the patients not only bridged their bone defect, but achieved limb length equality as well.

Paper #31

COMPLIANCE WITH ACCP GUIDELINES FOR PREVENTION OF VENOUS THROMBOEMBOLISM AFTER TOTAL HIP AND KNEE ARTHROPLASTY

Richard J. Friedman, MD1 (a-Center for Outcomes Research, University of Massachusetts, b-Sanafi-Aventis),

Alexander Gallus, MD2, Fred Cushner, MD3, Gordon FitzGerald, MD4, Fred Anderson, PhD4 1Charleston, SC, 2Flinders Medical Centre, Australia, 3ISK Institute for Orthopaedics, New York, NY, 4University of

Massachusetts Medical School, Worcester, MA Background: The Global Orthopaedic Registry (GLORY) collects data about in-hospital management and 12-month clinical outcomes of elective total hip or knee arthroplasty (THA/TKA) patients in a range of practice environments across North and South America, Europe, Japan, and Australia. The American College of Chest Physicians (ACCP) has published evidence-based recommendations for preventing venous thromboembolism (VTE) after THA or TKA. Clinical practice for VTE prophylaxis varies greatly among orthopaedic surgeons, with previous studies suggesting that there is a wide discrepancy between clinical practice and evidence-based medicine. The purpose of this study is to compare VTE prophylaxis patterns with the evidence-based ACCP guidelines. Methods: Compliance with ACCP guidelines was assessed as “partial” if ≥1 recommended type of prophylaxis was used, and as “full” if the recommended timing, duration and target range were achieved. Recommended types of prophylaxis included warfarin or low-molecular-weight heparin (LMWH) for THA and TKA. Recommended timing for the start of prophylaxis was within 12 hours before, or 24 hours after surgery. Duration was ≥7 days, and therapeutic range for warfarin was INR 2–3. Results: Data on VTE prophylaxis use was available for 3,259 THA and 4,247 TKA patients. Data from THA/TKA patients are pooled in this analysis because compliance was similar in both groups. Warfarin use was restricted almost entirely to the USA: 58% of US THA/TKA patients received warfarin in hospital compared with 0.2% THA/TKA patients elsewhere. The median in-hospital duration of warfarin prophylaxis was 3 days, and this was continued after discharge in only 44% of patients. Among 2,823 patients receiving warfarin, 23% received their first dose before surgery, 74% within 24 hours after surgery, and 3% on a later day. The locally nominated target INR was 1.5–1.9 in 62.2% and ≥2.0 in 37.4% patients. Only one-third of all patients on warfarin achieved an INR 2–3 during hospitalization. Low-molecular-weight heparin (LMWH) use was less common in the USA (41%), compared to other countries (92%). Initiation of LMWH prophylaxis was correct according to guidelines in 93% and 74% of US and non-US patients, respectively. The median duration of LMWH was 8.5 days (in hospital 3 days; post-discharge 6 days) in the USA and 25 days (in hospital 8 days; post-discharge, 17 days) in other countries. Conclusions: Only 38% of US patients and 54% of non-US patients in GLORY received VTE prophylaxis according to ACCP recommended timing, duration, and therapeutic range. Failures to comply with ACCP recommendations were primarily inadequate dosing and low target INR with warfarin, although insufficient duration of LMWH also contributed to lower compliance in US patients. There is a wide gap between guideline recommendations and practice, with the majority of US surgeons using warfarin with a low target INR of 1.5–1.9 rather than the recommended INR 2–3. There is currently no evidence to support this low-dose warfarin regimen in orthopaedic surgery.

Paper #32

Not Available

Paper #33

THE EFFECT OF OXYGEN TENSION ON THE COLONY FORMATION AND EXPANSION OF HUMAN CONNECTIVE TISSUE PROGENITOR CELLS

George F. Muschler III, MD (n), Sandra Villarruel, MS, Cynthia Boehm, BS, Mark Pennington, BS, Kimberly Powell,

PhD The Cleveland Clinic Foundation, Cleveland, OH

Bone marrow contains connective tissue progenitor cells (CTPs) that can be activated to form new tissues including; bone, cartilage, fibrous tissues, muscle, and fat. The heterogeneous CTP population in various tissues can be characterized by analysis of the progenitor cell colonies that are derived from these cells in vitro. Using a colony-forming unit (CFU) assay clonal formation, expansion, and differentiation of CTP progeny can be effectively characterized. The number of CFUs found in a population reflects both the number of CTPs in the population and the activation of these cells under the culture conditions. Proliferation can be assayed in each colony, based on the number of cells in each colony at a preset time following activation. Optimization of the use of CTPs in therapeutic cellular grafts requires an improved understanding of the bone healing environment and the response of CTPs to those conditions. One of the most characteristic features of the fracture site or site of cell transplantation is a lower than normal oxygen tension, due to an imbalance between the metabolic demand of local cells and the initial lack of vascularization within the graft site. The response of CTPs to a low oxygen tension environment is therefore critical to the biological performance of these cells following transplantation. This study assessed the effect of local oxygen tension on CTPs using quantitative analysis of CFUs under four oxygen tension conditions.

Using an approved IRB consent protocol (#1309) bone marrow was harvested from the iliac crest of fifteen patient donors. Bone marrow derived cells were plated at 1.0x106 cells/2x2cm2

and cultured under conditions that promote an osteoblastic phenotype for six days at four different oxygen tension levels: 20%, 10%, 5%, and 1% (corresponding to PO2=152 mmHg, PO2=76 mmHg, PO2=38 mmHg, and PO2=7.6 mmHg, respectively). The cultures were then fixed using 1:1 acetone/methanol and stained with DAPI, to fluorescently label nuclei. CTP cultures were imaged using a 2048x3072 Quantix K6303E 12 bit digital camera (Roper Scientific) attached to a Leica DMXRA motorized microscope and controlled by imaging software Metamorph (v4.6). Individual images were collected at 512 x 768, 8-bit gray level, using a 10x objective (pixel size = 1.78 μm) and then montaged together (20 columns x 24 rows) to create one image of the entire cell culture well (2cm x 2cm square). The image analysis software used for quantitative characterization of CTPs was developed using algorithms written in the C/C++ programming language and Motif X-Windows TM environment. We assessed the effect of oxygen concentration on the data using a mixed effects model analysis while controlling for patient. We estimated the mean outcome differences for all possible pairs of oxygen concentrations, and analyzed them using Tukey’s correction for multiple comparisons. In fifteen patient donors, a total of 9,174 colonies containing a total of 367,418 cells were assayed. CFU formation increased at lower oxygen tensions (1%, 5%, 10%) compared to CTPs cultured at 20%, reflecting an increased activation rate, even at 1% O2. Figure 1 shows the colony forming efficiency (CFE), which was determined relative to the standard 20% oxygen. CFE was increased by 60% at 5% O2 compared to 20% O2 indicating greater activation frequency of CTPs at lower oxygen tension conditions.

Figure 1: Median Colony Forming Efficiency

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

1 5 10 20

Oxygen Tension (%)

CFE

* p<0.001**p<0.002

***

*

A count of the number of cells per colony was also performed. The median number of cells per colony was increased at lower oxygen tension, reflecting an increase in clonal expansion among the progeny of CTPs at lower oxygen tensions. In this setting, colonies containing 8, 16, 32 … cells can be viewed as having undergone 3, 4, 5 … cycles of proliferation (Effective Proliferation Rate or EPR). Figure 2 shows the EPR data. EPR was significantly greater at lower oxygen tension (1%, 5%, 10%) than at 20% O2 (p-value < 0.0001) despite large variation between individual CFUs at all oxygen tensions. A cell per colony distribution for each oxygen tension can be determined by evaluating each individual colony. Figure 3 clearly demonstrates the out performance of oxygen tensions 5% and 10% compared to the extreme cases of 1% and 20% oxygen.

Figure 2: Effective Proliferation Rate Figure 3: Cells per Colony Distribution

0

1

2

3

4

5

6

7

8

9

10

8 16 32 64

128 256

512

1024

2048

4096

8192

16384

Cells in Colony

Num

ber o

f Col

onie

s

1% 5% 10% 20%

Low oxygen tension conditions, even down to 1% O2, appear to increase both CTP clonal activation and expansion among the progeny of CTPs derived from bone marrow. The fracture healing environment, in particular the characteristic low oxygen tension, will have a great impact on composite cellular grafts used in the clinical setting. These findings have important implications relative to the biological response and survival of CTPs following local trauma or transplantation and may also influence the design and development of delivery systems for CTPs in tissue engineering applications.

Paper #34 A PROSPECTIVE CLINICAL AND ELECTROMYOGRAPHIC COMPARISON OF VASTUS SPLITTING

AND MEDIAN PARAPATELLAR APPROACHES TO TOTAL KNEE ARTHROPLASTY

Vincent D. Pellegrini, Jr., MD1 (n), Matthew Kelly, MD2, Michael Parentis, MD2, Mustasim Rumi, MD2, William Parrish, MD2, Katrina Bailey, RPT2, Milind Kotari, MD2

1University of Maryland Orthopaedics, Baltimore, MD, 2Pennsylvania State University, Hershey, PA Purpose: A prospective randomized clinical trial was conducted comparing the vastus splitting and median parapatellar approaches for primary total knee arthroplasty (TKA). Materials and Methods: Informed consent and IRB approval were obtained; 42 consecutive patients (51 knees) were enrolled. Functional evaluation, operative data, and preoperative and postoperative electromyography (EMG) were assessed. Results: Early postoperative results at 5.8 months showed no differences in range of motion, strength, knee scores, tourniquet time, proprioception, kinesthesia, or frequency of patellar resurfacing. There were significantly more lateral releases and greater blood loss in the median parapatellar group. Nine of twenty-one (43%) knees in the vastus splitting group had abnormal post-operative EMG of the vastus medialis; all 25 in the median parapatellar group were normal. Long-term follow-up in 31 of 42 patients (39/51 knees) at mean 5.27 years (range 4.32-5.97 years) reaffirmed comparable functional evaluation and knee scores in the two groups. Electromyograms were repeated in those nine patients with prior abnormal studies; seven had returned to normal and each had the muscle split developed by blunt digital dissection. One patient had chronic changes indicative of reinnervation and one demonstrated evidence of ongoing denervation. No significant functional differences existed in these nine patients. Discussion: The vastus splitting approach for primary TKA offers a reasonable alternative to the median parapatellar approach and significantly reduces the need for lateral release. Early EMG abnormalities represent reversible neurapraxic injury that can be avoided by blunt muscle dissection. Knowledge of the functional innervation of the extensor mechanism will be important in less invasive approaches to knee arthroplasty.

Paper #35

ELBOW RANGE OF MOTION IN PROFESSIONAL BASEBALL PITCHERS

Rick W. Wright, MD (n), Mark O'Neal, ATC, Barry Weinberg, ATC, Brett L. Wasserlauf, MD, George A. Paletta, MD Barnes-Jewish Hospital, St. Louis, MO

Background: Physicians involved with the care of baseball players have long noted elbow range of motion changes in pitchers. Objective data regarding the extent of these changes has rarely been documented. In addition, any factors related to the extent of range of motion differences have not been demonstrated. These factors would include player age, length of service, innings pitched, or history of elbow surgery. We hypothesized that elbow range of motion changes are common in baseball pitchers and that these changes are related to player age, amount and length of time professionally pitched and any history of surgical procedures on the dominant extremity. Methods: Thirty-three professional pitchers were evaluated for elbow range of motion during spring training preseason physical examination for the 2003 season. Dominant and nondominant elbow range of motion including flexion, extension, supination, and pronation was measured with a goniometer using techniques established by the American Academy of Orthopedic Surgeon’s guidelines for joint motion measurement. All players had been participating in their usual winter preseason preparation, but no player had begun preseason drills or competition. Range of motion data was evaluated for dominant vs. nondominant elbow extension and extension loss, elbow flexion and flexion loss, and elbow flexion extension total arc and arc loss. Elbow supination and pronation and total supination pronation arc was compared for dominant and nondominant elbows. Dominant and nondominant range of motion differences were evaluated using a t-test assuming unequal variances. Baseball records were reviewed for arm dominance, age, years of professional pitching, professional innings pitched and history of elbow surgery. These factors were then statistically evaluated for their possible effect on the range of motion changes noted during physical examination. Age, # years and # innings pitched are inherently continuous measures. To analyze their association with ROM, Pearson correlations (r) were used followed by an unpaired t-test comparing two groups formed by a median split. Previous surgery and dominant side are categorical measures. To analyze their association with ROM unpaired t-tests were used. Results: Table 1. Sample descriptives (N = 33). Cell values represent mean ± standard deviation or n (%). Mean ± SD or

n (%) Min, max

Age (yrs) 29.6 ± 3.9 23.6, 40.1 Years pitching 8.6 ± 3.7 3.0, 19.0 Innings 943 ± 565 180, 2568 Dominant side: Left 8 (24%) Right 25 (76%) Surgery on dominant side: No 19 (58%) Yes 14 (42%) Table 2. ROM measures by side (N = 33).

Data presented are mean ± standard deviation and (min, max). P-value based on paired t-tests comparing sides.

Dominant Nondominant Difference p-value Extension˚ 5.5 ± 8.0 (-6.0, 28.0) -2.4 ± 3.2 (-9.0, 3.0)* -7.9 ± 7.4 (-29.0, 3.0)* <0.0001 Flexion˚ 130 ± 7.1 (110, 142) 135 ± 4.7 (126, 145) -5.5 ± 7.8 (-30.0, 8.0)** 0.0003 Extension/flexion total arc˚ 124 ± 13.6 (88.0, 146) 137 ± 6.0 (126, 149) 13.3 ± 13.7 (-5.0, 46.0)* <0.0001 Supination˚ 82.6 ± 5.2 (73.0, 93.0) 83.3 ± 4.6 (73.0, 97.0) -0.73 ± 4.2 (-14.0, 7.0)** 0.33 Pronation˚ 83.8 ± 5.0 (68.0, 91.0) 83.7 ± 4.5 (71.0, 92.0) 0.15 ± 5.8 (-17.0, 16.0)** 0.88 Supination/pronation total arc˚

166 ± 8.1 (149, 181) 167 ± 6.2 (155, 182) -0.58 ± 6.5 (-14.0, 14.0)** 0.62

1. *Difference calculated by subtracting the dominant side from the nondominant side. 2. **Difference calculated by subtracting the nondominant side from the dominant side.

3. *Negative value represent hyperextension Tables 1 and 2 describe the sample and indicate that there are significant differences in ROM between the dominant and nondominant side for extension/flexion measures. No significant difference between sides was found for the supination/pronation measures. No associations with ROM were found (all correlation p-values > 0.10) for age, years pitched, innings pitched, dominant arm or history of surgery. A logical next step in the analysis would be to develop a multivariate model predicting ROM difference between sides. Since no significant predictor of between-side differences was noted, there is no reason to proceed to a multivariate model. Conclusions: Our study demonstrated a significant loss of range of motion of a pitchers’ dominant elbow. A direct correlation of age, experience or history of elbow surgery on the loss of range of motion was not shown. Based upon the range of motion necessary for overhand throwing these findings may be of minimal consequence. Physicians and athletic trainers involved in the care of these athletes should understand this represents a common finding that may represent no impact upon their ability to throw a baseball and no impact upon their length of career.

Paper #36

DEPRESSION IN ORTHOPAEDIC TRAUMA PATIENTS: PREVALENCE AND SEVERITY

Renn J. Crichlow, MD1, Patricia L. Andres, MS, PT2, Suzanne M. Morrison, MPH3, Stephen M. Haley, PhD, PT2, Mark S. Vrahas, MD3 (n)

1Shock and Trauma Center, Baltimore, MD, 2Sargent College of Health and Rehabilitation, Boston, MA, 3Massachusetts General Hospital, Boston, MA

Background: There has been a longstanding connection between physical injury, disability, and emotional distress. Several investigators have shown a relationship between trauma, depression and poor outcomes. The body of literature on trauma and depression, however, is limited in its ability to clearly evaluate the following: the relationship between the degree of injury and depression; the degree to which physical injuries confound the evaluation of depression; and the relationship between physical function in patients with less severe injuries and depression. Methods: One hundred and sixty-one patients who presented to the Orthopaedic Trauma Service’s Clinic were enrolled and interviewed. We obtained information about patients’ demographics and administered the following self-report outcome measures: the Beck Depression Index (BDI); the Short Musculoskeletal Function Assessment (SMFA); and the Physical Function-10 (PF-10). Outside of this method, we documented the nature and severity of the injury(s), and calculated correlations between the outcome measures and the BDI. Injury-specific factors such as the AO classification, Abbreviated Injury Scale (AIS), Injury Severity Score (ISS), Gustillo and Anderson Type were also examined. Results: Patients’ depression severity scores as measured by BDI were as follows: minimal depression - 55%; moderate depression - 28%; moderate to severe depression - 13%; and severe depression - 4%. When the somatic elements of the BDI were removed, the prevalence of moderate to severe depression was 26%. The SMFA scores had a strong negative correlation with the BDI: – 0.75 with a p value < 0.001. Of the injury-specific factors, only open fractures were found to have an impact on the presence of depression, with an odds ratio of 4.58 (95%CI 1.57-12.35). Conclusions: The prevalence of clinically significant depression approaches 45% in a diverse cohort of postoperative orthopaedic trauma patients. Global disability is strongly correlated with the presence of depression. The lack of vitality and physical function alone does not explain the extent of the depression, as there is a strong emotional component. The presence of an open fracture substantially increases the risk for depression. Level of Evidence: Level III retrospective cohort study

Paper #37

♦SYNOVIAL FLUID ANTIBIOTIC LEVELS IN TWO STAGE PROSTALAC REVISIONS FOR INFECTED TOTAL JOINT ARTHROPLASTIES

Christopher P. Beauchamp, MD (b-Depuy), Sean C.L. Frost, MD, Richard E.

Donnelly, PA-C, Mark J. Spangehl, MD Mayo Clinic, Scottsdale, AZ

The gold standard for treatment of infected total hip or knee arthroplasties remains a 2-stage reconstruction with implant removal and insertion of antibiotic impregnated spacers, iv antibiotics, and reimplantion with a new joint replacement once the infection has been eradicated. Thirty such two-stage reconstructions were performed by the senior author over the period covering January 2001 – September 2004. At the time of reimplantation, synovial fluid was analyzed for drug levels of vancomycin, gentamycin and/or tobramycin. The mean time between stages was 87.7 +/- 28.9 days. The mean ESR and CRP at the time of Stage I prostalac insertion was 60 +/- 25 and 45.9 +/- 8.7 respectively. The ESR and CRP just before Stage II revision dropped significantly to 21.9 +/- 7 (p<0.01) and 12.8 +/- 0.66 (p<0.001) respectively. The synovial fluid at Stage II revision contained a mean vancomycin level of 8.45 +/- 5.2 (n=30), gentamycin level 18.8 +/- 8.3 (n=20), and tobramycin levels 15.6 +/- 5.9 (n=10). Of the 30 patients, there were 2 with persistent infection, one resulting in revision Stage I, and one resulting in amputation. There were 2 patients who developed recurrent instability requiring revision surgery, and one patient developed acute renal failure requiring revision of the palacos cement to avoid gentamycin. Twenty-nine of the 30 patients ultimately had infection free joint replacements.

Paper #38

THE INCIDENCE OF PELVIC DEEP VENOUS THROMBOSIS AFTER ANTERIOR SPINAL FUSION SURGERY

Serena S. Hu, MD (n), Anubhav Sinha, MD, Sigurd H. Berven, MD, David S. Bradford, MD, Vedat Deviren, MD

University of California, San Francisco, CA Background: Deep venous thrombosis is noted to be a consequence of anterior spine surgery due to retraction of the iliac vessels and the contraindication of prophylactic anticoagulation. Previous studies1 have reported an incidence of up to 6% in patients post-operatively, however those studies used clinical signs and Doppler ultrasound to diagnose lower extremity thrombi, which is known to have a poor detection rate for pelvic thrombi. Since anterior spinal approaches involve significant retraction of the great vessels, detection of pelvic thrombi may be more important in this patient population. Our study used pelvic magnetic resonance venography (MRV) to look for thrombi in the post-op period, a method that has been shown to have greater than 95% sensitivity and specificity for detecting thrombi. Materials and Methods: For this study, we asked all patients who were operated on by our spine service between 2001-2004 for anterior spine surgery and were over 40 years of age to participate. Thirty-six patients have been enrolled to date and in addition, there were 4 patients who we were not able to consent prior to surgery, (either because they were not asked or because they declined) who required an MRV due to clinical suspicion of DVT (either leg swelling or shortness of breath in the post-operative period). Three to five days after their anterior procedure, the patients underwent a pelvic MRV. If the MRV was positive for a pelvic thrombus, the patients received treatment as recommended by the consulting hematology service. All patients who were studied were called three months after their procedure to see if they had been diagnosed with or had symptoms of a DVT in the interim. Results: Of the 40 total patients who underwent MRV, 2 had definite pelvic thrombi and one other patient had a probable thrombi. All three of these patients had a tear of the iliac vessel during surgery (3 tears with DVT/7 total tears = 43%), and all three had allograft placed anteriorly (3/22=13.6%). Patients who had a DVT also tended to have on average, more anterior levels fused (7.0 vs. 6.1) and longer operating times though these were not statistically significant. Of the two patients with definite thrombi, one had a history of PE and had an IVC filter placed and is currently on long-term anticoagulation without complications. The other was placed on short-term anticoagulation in the hospital without complications. The patient with probable thrombi was monitored clinically without anticoagulation and developed no further sequelae. Of the four patients who did not initially consent to the procedure but received MRV due to clinical suspicion, 1 patient had a definite positive MRV and is included in the above statistics. One additional patient who did not receive an MRV and did not have clinical symptoms of a DVT while in house, developed a DVT on a plane ride back home, one week after surgery. Discussion: We determined an incidence of 7.5% (3/40) of DVT among patients after anterior spine surgery in the immediate post op period, this is similar to previous studies. We observed an incidence of 25% in patients who had clinical suspicion of a DVT, which is significantly higher than previously reported. In patients who have a positive MRV, no patients received complications from either the IVC filter with long-term anticoagulation or short-term anticoagulation to date. In the patient who developed DVT while on a plane ride, it is difficult to know whether or not the DVT was caused by the surgery or by venous stasis caused by prolonged immobilization while seated in an airplane. From this data we conclude that DVT is a fairly common entity in this patient population and should be treated aggressively when clinically suspected or if a tear in the iliac veins occurred during surgery. Pelvic MRV is a more sensitive study for detecting DVT in this patient population than Doppler ultrasound and should be used if available. Patients should also be advised to avoid immobilization in the postoperative period. Further studies should proceed to determine what other risk factors determine which patients are at greatest risk for developing DVT. 1 Dearborn et al, 1999.

Paper #39

SALTER-HARRIS I AND II FRACTURES OF THE DISTAL TIBIA: DOES MECHANISM OF INJURY RELATE TO PREMATURE PHYSEAL CLOSURE?

Scott J. Mubarak, MD (n), Michael T. Rohmiller, MD, Tracey P. Gaynor,

MA, Jeff Pawelek, BS Children’s Hospital, San Diego, CA

Introduction: Salter-Harris (S-H) type I/II physeal fractures of the distal tibia are commonly treated by closed methods due to the belief that these injuries have a low risk of physeal arrest. However, some authors report complications and believe that anatomic reduction improves outcome. We attempt to answer if operative treatment can reduce the rate of premature physeal closure (PPC) in S-H type I/II fractures of the distal tibia. Methods: At our institution, 147 distal tibial S-H type I/II fractures were treated from 1994-2002. Complete reviews were performed on all patients and 101 fractures met our inclusion criteria (largest series reported). Patients were categorized according to treatment performed. Results: Average patient age at the time of injury was 12 years. Patient follow-up averaged 12 months. Two major fracture patterns were defined radiographically. We found a difference in PPC based on injury mechanism. Type of treatment may prevent PPC in some fractures. The most important determinant of PPC is the fracture displacement following reduction. Conclusion: Premature physeal closure occurs more commonly in S-H type I/II fractures of the distal tibia than previously believed. Operative treatment may decrease the frequency of PPC in some fractures. Significance: The risk of physeal arrest in Salter-Harris type I/II fractures of the distal tibia is higher than previously reported, and may be decreased by operative management.

Paper #40

THE ROLE AND VALUE OF THE ORTHOPAEDIC REPRESENTATIVE

Robert R. Karpman, MD, MBA (n) Sun City West, AZ

Introduction: As hospitals continue to strive to improve their bottom line, more emphasis has been placed on the cost of medical supplies including orthopaedic implants. Little data has been collected regarding the role and value of the orthopaedic representative to the hospital and what contribution that may have to the overall price of an orthopaedic implant. The purpose of this study was to evaluate the time that orthopaedic representatives spend providing service to several hospitals in the Phoenix metropolitan area. Methods: Five orthopeadic representatives. were asked to keep daily time logs of all of their activities for a consecutive 2 week period. Major categories included service in the O.R., cleaning instruments, physician consulting, inventory, marketing, etc. The results were then tabulated and the mean times calculated for each activity. Results: The mean number of hours worked each day by the orthopaedic rep was 11.6 hours. In addition, the hospital expected 24 hour service call for urgent or emergent cases. 63% of the daily time was utilized in the operating room assisting scrub techs with instruments, preparing cases, etc. 17% of the time was spent consulting with surgeons regarding difficult cases, templating, providing procedure texts, etc. Less than 10% or 1 hour per day was spent on marketing or other company activities. Discussion: Unlike other medical supplies sold to a hospital, orthopaedic implants require a significant amount of technical support that is provided by the local orthopaedic representative. According to this present study nearly 60% of a representative’s time is spent in the operating room assisting hospital staff to assure that all of the equipment is available for each case. In some of our community hospitals, instruments for an orthopaedic case are no longer pulled by hospital staff but by the orthopaedic representative. If hospitals want to lower implant costs, serious attempts need to be made to supply adequate personnel to assist the orthopaedic surgeon to assure efficiency and patient safety or they must bear the burden of higher implant costs so that adequate technical support can be available.

Paper #41

THE ORIGIN OF MODERN CLINICAL RESEARCH

Stuart A. Green, MD (n) Los Alamitos, CA

The single-blind or double-blind, placebo-controlled randomized clinical trial is considered the “gold standard” for evaluating the alleged value of newly developed pharmaceuticals and devices. Indeed, it is difficult, if not impossible, to market an innovative drug or device in the modern world without first conducting such an investigation. Where did this method of evaluating the claimed benefit of a particular therapy arise? The history of blind assessment with sham intervention can be traced back to a French royal commission charged with investigating the mesmerism craze of the 1780s. Among the commissioners were Benjamin Franklin, the oldest member; Antoine Lavoisier, the father of modern chemistry; Jean Sylvain Bailly, a distinguished royal astronomer and later president of the National Assembly and mayor of Paris after the revolution; and Joseph-Ignace Guillotin, a well-known physician who argued successfully for “humane” executions with the device that later took his name. Franz Anton Mesmer, a Swiss-born Viennese physician who arrived in Paris in February, 1778, believed that an ultrafine magnetic fluid, emanating from the stars and planets, passed through all objects in the universe, both animate and inanimate. He attributed illness to blocking of the magnetic fluid’s natural flow and thought that he had the power to correct such stoppage. Without fully realizing how it happened, Mesmer hypnotized his subjects after pre-treatment preparation that utilized soft lighting, eerie music and other measures now recognized as helping evoke a hypnotic state. Speaking softly, Mesmer generated an altered type of consciousness in the patient. Part of Mesmer’s therapeutic program involved inducing a “crisis” whereby a patient would either faint or convulse uncontrollably. (This particular aspect of the treatment alarmed observers.) The patient would then be carried, if necessary, to a padded “crisis room” to avoid injury during the fits. As far as is known, the strategy Franklin and the other commission members employed to debunk Franz Mesmer’s assertions about illness and treatment appear in the first published report of a single-blind placebo-controlled clinical trial. Subsequent evaluators, looking into the legitimacy of certain therapeutic devices and medications, explicitly duplicated the commission’s method of using intentional subject ignorance (combined with real and sham treatment) to arrive at their conclusions. The royal commission’s report castigated Mesmer. The commissioners concluded that there was no such thing as an invisible magnetic fluid. They attributed the effects of mesmerism to the subjects’ “imagination.” After the publication of the royal commission’s report, there followed numerous placebo-controlled inquiries—both in Europe and America—into the professed benefits of Hahnemann’s homeopathic remedies. Other blind assessments followed, scrutinizing everything from rheumatic fever and psychological illnesses to testicular extract injections and cocaine. The final step in creating a thoroughly modern method of verifying the benefit of a particular treatment followed mathematician R.A. Fisher’s insistence that randomly assigning subjects to either a treatment group or a control (placebo) group permits valid statistical comparisons between the two groups to some definable level of confidence (the “p” value).