2003 Summer Bioengineering Conference, June 25-29, Sonesta Beach Resort in Key Biscayne, Florida INTRODUCTION The glenohumeral joint is the most frequently dislocated diarthrodial joint in the body. Recurrent instability following a capsular shift procedure for posterior instability has been reported to be as high as 72%. (3) Additionally, capsular shift procedures used to treat anterior instability have resulted in an increase in posterior capsular strain (2.9%-6.6%) as the humerus was elevated. (4) While experimental and analytical models have focused on the uniaxial tensile properties of the glenohumeral capsule, (1, 7) surgical repair techniques shift the capsule in the medial-to-lateral and superior-to-inferior direction. This translates the posterior capsule perpendicular and parallel to the long axis of the posterior band of the inferior glenohumeral ligament (PB-IGHL). Moreover, the posterior capsule has been shown to transmit forces in the direction perpendicular to its longitudinal axis. (2) Therefore, the objective of this study is to determine the mechanical properties of the posterior capsule in the directions perpendicular (transverse) and parallel (longitudinal) to the longitudinal axis of the PB-IGHL. METHODS Four cadaver shoulder specimens (average age: 63 ± 7.4 years). Each shoulder was thawed at room temperature and dissected leaving the glenohumeral capsule intact. The PB- IGHL was identified and the posterior capsule was dissected free from the joint. One longitudinal and one transverse dog- bone (midsubstance: 12.5 mm x 2.5 mm) tissue sample was harvested from the posterior capsule of each specimen. Throughout the entire testing protocol, tissue samples were kept moist with physiological 0.9% saline solution. The cross- sectional area of the midsubstance of each sample was determined using a laser micrometer system while the specimens were fixed within customized clamps. A circular punch was used to obtain reflective plastic markers (1.6 mm diameter) for non-contact video strain analysis. Two markers were fixed to the midsubstance of each sample using cyanoacrylate, centered approximately 10 mm apart, and analyzed with a Motion Analysis TM system. The samples were subsequently mounted in a preheated saline bath (37°C) on a uniaxial material testing machine (Instron Model 4502). The load cell had a range of 0-44.8 N with an accuracy of ±0.07 N. A 0.1 N preload was applied to each sample and the tissue was preconditioned via cyclic elongation (0.0-0.3 mm) for 10 cycles. A load-to-failure test was then performed at a crosshead displacement rate of 10 mm/min. From the stress vs. strain curve, the tangent modulus and the ultimate stress were obtained. A paired t-test was used to compare the parameters from the longitudinal and transverse directions with statistical significance set at p<0.05. RESULTS The traditional “toe region” followed by a linear region prior to failure was found for four of the eight tissue samples examined ( Figure 1 ). Seven tissue samples failed in the midsubstance and one tissue sample failed at the tissue-clamp interface. A significant difference (p>0.05) was not detected between the cross-sectional areas of the transverse (3.7±1.3 mm 2 ) and longitudinal (4.2±1.2 mm 2 ) tissue samples. No significant differences could be demonstrated for the tangent modulus, stress at failure, strain at failure, and strain energy density of the transverse tissue samples compared to the BI-DIRECTIONAL MECHANICAL PROPERTIES OF THE POSTERIOR REGION OF THE GLENOHUMERAL CAPSULE Susan M. Moore, Patrick J. McMahon, Richard E. Debski Musculoskeletal Research Center Department of Orthopaedic Surgery University of Pittsburgh Pittsburgh, PA Starting page #: 0107