the IL-1 functions are inhibited with IL-1 receptor an-tagonist (IL-1ra) in the stress-shielding model. We havehypothesized that a local administration of IL-1-ra mayinhibit the deterioration of the stress-shielded patellartendon. On the other hand, the tendon tissues have ahierarchical structure composed of bundles, fascicles,fibrils and fibers [7]. Yamamoto et al [8] found that theeffect of stress shielding on the mechanical properties ofpatellar tendon fascicles is much smaller than that ofbulk tendons. They suggested that the stress shieldingpredominantly affects mechanical interaction among col-lagen fascicles or among collagen fascicles and otherminor matrix proteins in the patellar tendon rather thanaffects the mechanical properties of collagen fasciclesthemselves. Therefore, we have made the second hypoth-esis that the effect of the IL-1-ra application on mechan-ical properties of the fascicles harvested from the stress-shielded patellar tendon may not be significant. Thepurpose of this study is to test these two hypotheses.Materials and Methods: A total of 30 mature femaleJapanese White rabbits with a body weight of 3.5 � 0.2 kgwere used in this study. In all animals, the right patellartendon was completely released from stress by stretching aflexible steel wire installed between the patella and the tibialtubercle [2]. After the stress shielding procedure, the ani-mals were divided into the following two groups of 15rabbits each. In Group I, 5-�g recombinant human IL-1ra(Peprotech EC) with 0.2-ml phosphate buffered saline(PBS) solution was injected between the patellar tendon andthe infra-patellar fat pad. In Group II, 0.2-ml PBS alone wasinjected in the same manner as Group I. All rabbits weresacrificed 3 weeks after the surgery [2,3]. We consideredthat the 3-week period was suitable for detecting the effectof a local application of IL-1-ra on the mechanical proper-ties of the stress-shielded patellar tendon. In each group,eight animals were used for biomechanical evaluation of thepatellar tendon in the bundle level having a width of 2.5 mmand five were used for biomechanical evaluation of thepatellar tendon in the fascicle level having a diameter of 300micrometers. The remaining two were used for histologicalobservation. Statistical analyses were performed using un-paired t-tests to compare the tissue dimension and themechanical properties of the patellar tendons betweenGroups I and II. A significance level was set at p� 0.05.Results: The averaged stress-strain relation curves areshown. The tangent modulus of Group I was significantlygreater (p�0.0163) than that of Group II, while thetensile strength of Group I was significantly greater(p�0.0063) than that of Group II. There were no signif-icant differences in strain at failure between Groups I andII. All fascicle specimens failed between the two gauge-length markers drawn on a specimen. We could not

detect any significant differences in the tangent modulus(p�0.558), the tensile strength (p�0.764), and the strainat failure (p�0.235) between Groups I and II. Histolog-ically, the number of fibroblasts in the patellar tendons ofGroups I and II was significantly higher than that of thenormal patellar tendon. In the patellar tendons of Group I,fibroblasts with spindle-shaped nuclei were predominantlyobserved, while cells with round or oval nuclei were scat-tered in the patellar tendons of Group II. In addition, split-ting tendon fibers in the patellar tendon appeared to be lessremarkable in Group I than in Group II.Discussion: This study demonstrated that a local adminis-tration of IL-1ra significantly inhibits the reduction of themechanical properties of the PT caused by stress-shieldingat 3 weeks. This result implied that the over-expressedIL-1� [4] plays an important role in the deterioration mech-anism of the mechanical properties of the PT. This studyalso showed that the effect of the IL-1ra application ontendon fascicles in the stress-shielded PT was not signifi-cant at 3 weeks. Yamamoto et al [8] reported that stressdeprivation reduces mechanical interactions among the col-lagen fascicles or among other minor matrix proteins suchas proteoglycan and fascicles in the PT. The results in thepresent study suggested a strong possibility that the IL-1�over-expressed in the stress-shielded PT [4] does not affectthe fascicles themselves but the mechanical interactionsamong the collagen fascicles or among other minor matrixand fascicles of the PT. As to clinical relevance, an anti-IL-1 strategy has recently attracted notice as a clinicaltherapy for inflammatory diseases such as a rheumatoidarthritis [9]. The present study highlights a possible appli-cation of anti-IL-1 strategies for reducing the mechanicaldeterioration of tendons and ligaments in response to stressdeprivation.References:

1. Amiel et al, Connect Tissue Res 1980.2. Yamamoto et al, JBE 1993.3. Majima et al, JOR 1996.4. Uchida et al, JB 2005.5. Tsuzaki et al, JOR 2003.6. Bhatnagar et al, Biochem Int. 1986.7. Kastelic et al, Connect Tissue Res 1978.8. Yamamoto et al, Clin Biomech 1999.9. Afeltra et al, Curr Drug Targets Immune Endocr

Metabol Disord 2001.

Paper 3: Prospective Randomized Study to CompareSingle Bundle Versus Double Bundle ACL Recon-struction in Restoring Rotational 3D Kinematics ofthe Knee WILLEM MARE VAN DER MERWE, MBCHB FCS SAORTHO, SOUTH AFRICA, PRESENTING AUTHOR

e334 ABSTRACTS

ABSTRACTBiomechanical studies have shown increased rota-

tional stability with double bundle compared to singlebundle ACL reconstruction. The aim of this study was toevaluate the difference in rotational kinematics after sin-gle and double bundle ACL reconstruction.

40 Subjects undergoing ACL reconstruction wasprospectively randomized to DB (n�20) and SB(n�20) groups. Exclusion criteria included meniscalpathology, collateral ligament instability, previous in-jury and previous surgery. All surgery was performedby one surgeon using Semitendanosis and gracilis asgraft and Endobutton and BioRCI screw as fixation.Demographics data were comparable between the 2groups. Passive laxity was measured by determiningthe 6-degree-of-freedom position and orientation ofthe femur and tibia under a known torsional load usingMRI. Each subject underwent 3 studies one on theuninjured knee that served as the control one preoperative and one post operative.Results: Rotational stability increased after ACL injuryand improved after surgery but not to normal. There wasno statistical difference between single and double bun-dle Reconstruction in restoring rotational stability. APtranslation in the medial compartment was also measuredduring rotational loading and there was a statistical dif-ference between DB and SB reconstruction with DBbeing superior. The axis of rotation was different for DBand SB reconstruction although this was not statisticallysignificant.

The subjective IKDC score was the same for bothgroups but the objective IKDC score was significantlybetter for the DB groupConclusions: Our study shows that although there wasno difference in the overall rotational stability betweensingle and double bundle ACL reconstruction, medial APtranslation and rotational axis differences might explainthe better objective score. Further studies are necessaryto determine if abnormal translation in the medial com-partment lead to increase risk for OA.

Paper 4: Prospective randomized study on DoubleBundles ACL Reconstruction (Out-In technique) ver-sus Single Bundle (STG) ACL Reconstruction (TransAntero-Medial Portal) LUIGI A. PEDERZINI, MD, ITALY,PRESENTING AUTHOR

MASSIMO TOSI, ITALY

MAURO PRANDINI, ITALY

LUIGI MILANDRI, ITALY

FABIO NICOLETTA, ORTHOPAEDIC SURGEON, ITALY

ANDREA DAVIDE COSSIO, ORTHOPAEDIC SURGEON, ITALY

ABSTRACTPurpose: The purpose of this study is to compare in aprospective randomized study, the clinical outcome andradiographic results between Double Bundle Out-Intechnique and Trans Antero-Medial portal Single Bundle(STG) technique in a 2 years follow-up.Materials and Methods: From March 2005 to March2006, 70 patients were operated and involved in a pro-spective randomized clinical study, 35 of these had ananatomical ACL reconstruction using Out-In technique(group A) and 35 had Trans Antero-Medial portal SingleBundle technique (STG 4 strands) (group B).

In Both the group anatomical insertional areas wereevaluated and measured arthroscopically, notch-plastywas not performed and patients with associated meniscal,peripheral lesions and condral problems were excluded.

In the group A, the tunnels were drilled over theanatomical areas in a convergent Out-In manner, and thenew bundles fixed by 4 absorbable screws at femoral andtibial levels. In the Group B the femoral tunnel was donedrilling through medial portal in order to obtain a newinsertion at 10/2 o’clock. All the cases were evaluated byan external observer: ROM evaluation was done at 1, 3,6 and 12 months, IKDC score and a standard X-Ray wasperformed at 24 months.Results: The average ROM at 1 month was 0°-125° inthe Group A, and 0°-100° in the Group B and respec-tively 0°-135 and 0°-125 at 3 months, and not significallydifferent at 12 and 6 months.

The IKDC score at 24 months follow-up evidenced82% grade A, 10% grade B and 8% grade C for group A,while group B showed 76% grade A, 14% grade B, 10%grade C. The residual pivot shift (glide) was 4% in groupA and 8% in group B. Standard X rays at 24 monthspost-op showed a mild enlargement for both the groupsA and B obviously in respect of the initial tunnel diam-eters.Conclusions: Out–in technique in anatomical doublebundle ACL reconstruction gave us better results com-pared to Trans Antero-Medial portal Single Bundle ACLreconstruction in early ROM, A-P stabilization and inresidual pivot shift test.

Paper 5: Prospective Clinical Comparisons of Ana-tomic Double Bundle Versus Single Bundle AnteriorCruciate Ligament Reconstruction Procedures in 328Consecutive Patients EIJI KONDO, MD, PHD, JAPAN, PRE-SENTING AUTHOR

NOBUTO KITAMURA, MD, JAPAN

SHIN MIYATAKE, MD, JAPAN

KAZUNOBU ARAKAKI, MD, JAPAN

YOSHIE TANABE, RPT, PHD, JAPAN

e335ABSTRACTS