A finite element analysis of augmented glenoid components

e328

the SST score and VAS pain for patients treated with a total shoulder arthroplasty. Methods: 319 patients had a primary anatomic total shoulder arthroplasty (TSA) or primary reverse total shoulder arthroplasty (RSA) by one of 5 shoulder and elbow surgeons at one of two institutions. Each patient completed the Simple Shoulder Test (SST), American Shoulder and Elbow Surgeons (ASES) score and a visual analog pain scale (VAS pain) preoperatively and at a minimum of 2 years postoperative (average 3.5 years). At final follow-up, each patient also answered the question “Since your total shoulder arthroplasty, please rate your response to the surgery. A: none—no good at all, ineffective treatment, b: poor—some effect but unsatisfactory, c: good—satisfactory effect with occasional episodes of pain or stiffness, d: excellent—ideal response, virtually pain-free”. Patients who answered “none” or “poor” were classified as having no change in the status of their shoulder. Patients who answered “good” were classified as having experience a small change equivalent to the “minimal clinically important difference”. The MCIDs were calculated for the ASES score, SST and VAS pain by subtracting the mean change score of all patients classified as having no change from the mean change score of all patients who were classified as experience a minimal important difference. T-tests were performed to compare means between unchanged and minimal important difference groups. A secondary analysis was performed to determine the effect of age, sex, duration of follow-up and type of arthroplasty (TSA or RSA) on the MCID. Results: 112 patients reported their results as “good” whereas 20 patients reported their results as “none” or “poor”. The MCID for the ASES score, SST and VAS pain was 20.9 (P < .001)), 2.4 (P < .0001) and 1.4 (P = .158), respectively. Duration of follow-up and type of arthroplasty did not have a significant affect on the MCIDs (P > .1) except shorter follow-up correlated with a larger MCID for the ASES score (P = .0081). Younger age correlated with larger MCIDs for all scores (P < .024). Female sex correlated with larger MCIDs for the VAS pain (P = .123) and ASES score (P = .05). Conclusions: Patients treated with a shoulder arthroplasty require a 1.4 point improvement in VAS pain scores, a 2.4 point improvement in the SST and a 21 point improvement in the ASES score in order to achieve a minimal clinical improvement from the procedure. Females and younger patients typically require a larger change to report this improvement. Duration of follow-up and implant type (RSA vs TSA) had no effect on achieving this change.

Paper #22 REVERSE SHOULDER ARTHROPLASTY IN PATIENTS WITH IMPAIRED LOWER EXTREMITIES

Adam L. Kemp, BS, Thomas W. Wright, MD, Department of Orthopaedics and Rehabilitation, University of Florida College of Medicine, Gainesville, Florida, USA Introduction: Patients with impaired lower extremities place an increased demand on their upper extremities due to mobility requirements. The rotator cuff is frequently the weak link in the upper extremity as it is not designed for long term weight bearing activities. Because of the increased loads these individuals may see catastrophic cuff failure at a relatively early age. This greatly increases their degree of impairment and further reduces mobility. The goal of this study is to report on this patient population with symptomatic irreparable rotator cuff defects or arthritis treated with a reverse shoulder arthroplasty (RTSA). It is our hypothesis that these patients will greatly benefit from a RTSA, that it will be a durable solution with only a moderately increased rate of complications. Material and Methods: From 1/2006 and 5/2013, 19 shoulders in 16 patients at one institution fulfilled the criteria for undergoing a RTSA with impaired lower extremities (primary mode of transportation wheelchair). Outcome variables included: Shoulder Pain and Disability Index (SPADI), Simple Shoulder Test (SST-12), American Shoulder and Elbow Surgeons Questionnaire (ASES), UCLA Shoulder Score (UCLA), Short Form Health Survey (SF-12), Constant Shoulder Score, and active range of motion. Complications were noted. Results: 3 of the 19 patients had a significant adverse event occurring early (<3 months). One was incurred

J Shoulder Elbow Surg 2016

in a postop fall in a Parkinson’s disease patient that dislodged the structurally bone grafted baseplate (6 weeks post op) and 2 patients dislocated, one with a proximal humerus nonunion and one with severe RA. One patient died at 9 months postop unrelated to the shoulder arthroplasty surgery. 12 out of the 15 remaining shoulders had at least 2 year follow-up. Follow-up averaged 40 months with a range of (24-66). All patients improved dramatically with the preop and postop scores reported in that order: SPADI 100/45, SST 2/7, ASES 31/73, UCLA 10/25, Constant 27/65 Active elevation 75/112 and Active external rotation 0/29. 2 patients had late complications one with a periprothethic fx at 3 years postop and one scapular fracture due to an assault. The periprosthetic fracture was revised. Discussion: RTSA for treatment of the symptomatic arthritic or irreparable rotator cuff tears in patients with impaired lower extremities can result in marked improvement in pain and function. However the surgeon must accept an increased major adverse event rate 26%. The big challenge is managing these patients immediately post op as they are very depend on their limbs. Often this means placement in a total care nursing facility for 6 to 12 weeks. However once past this postoperative period the patients greatly benefit from the increased mobility and function afforded them by the RTSA. Conclusion: RTSA for the proper indications in patients where the wheelchair is their major mode of mobility can be very rewarding but the surgeon must be prepared to manage these patients in the immediate post-operative period when they are limited to one functional extremity. Most complications occur during this post-operative period.

Paper #23 A FINITE ELEMENT ANALYSIS OF AUGMENTED GLENOID COMPONENTS

Nikolas K. Knowles, MESc, G. Daniel G. Langohr, MASc, George S. Athwal, MD, FRCSC, Louis M. Ferreira, PhD, University of Western Ontario and Roth|McFarlane Hand and Upper Limb Centre, London, Ontario, Canada Background: Acquired glenoid bone loss due to osteoarthritis can be classified by two common morphological erosion patterns: symmetric and asymmetric. In symmetrically eroded glenoids requiring joint replacement, the articular surface is typically reamed to match the curvature of the implant for full backside contact. This method, when used on bi-concave asymmetrically eroded glenoids to correct the version angle, is referred to as eccentric reaming. Alternatively, posterior augmented glenoid components provide another surgical treatment option. The purported advantage of augmented implants is their inherent bone preserving design, which corrects version without removal of excessive anterior bone by eccentric reaming. For these components, there is a lack of data pertaining to the biomechanical characteristics, specifically in the remaining supporting glenoid bone. The purpose of this study was to characterize three augmented glenoid component designs under varying simulated net joint loading directions and magnitudes. Materials: Three augmented glenoid component designs were modeled to assess varying backside geometries (Fig. 1). The implants were constructed as CAD models with identical peg sizes/locations, peripheral geometrical shape, and articular and backside curvatures. Three different augment designs were chosen to fully correct acquired retroversion while minimizing paleoglenoid reaming. The selected augments were: 7 mm posteriorstep, 35° posterior-wedge, and 16° full-wedge. A 1 mm cement mantle surrounded each peg for complete peg fixation. The scapular model was created with advanced asymmetric glenoid erosion (Walch type B2). Bone was discretized using tetrahedral elements and a heterogeneous distribution of cancellous bone properties were assigned based on CT intensity. Cortical elements were assigned homogenous properties. Net load vectors were applied to the articular surface of the augmented component by a simulated humeral component with a 4 mm mismatch in curvature. Loads of 500, 750 and 1000 N were applied to the center of the glenoid component and then varied 6° in all directions surrounding the central position, for a total of 27 loading

e329

J Shoulder Elbow Surg Volume 25, Number 10

Table 1

Peak global displacements of neoglenoid bone under varying net joint load vector magnitudes/directions

Net Load Vector Direction

P-S P P-I S Center I A-S A A-I

Peak Global Displacements (μm) Posterior-Step

Posterior-Wedge

Full-Wedge

500 N

750 N

1000 N

500 N

750 N

1000 N

500 N

750 N

1000 N

57 37 27 18 16 14 27 27 26

82 54 40 27 24 20 43 41 40

105 71 53 38 33 26 56 55 53

22 14 11 14 11 13 23 22 25

32 21 15 19 15 20 36 35 36

42 27 20 27 22 28 47 47 49

30 21 16 13 10 10 20 21 23

44 31 24 19 16 15 31 32 33

58 42 32 24 22 20 41 42 45

P, Posterior; A, Anterior; S, Superior; I, Inferior.

contact areas for posteriorly directed loads, while the full-wedge showed greater compressive contact area for central and anterior directed loads. Greater implant liftoff in the anterior hemisphere of the glenoid and greater neoglenoid bone displacements were observed with posteriorly directed loads. The posterior-wedge implant resulted in the lowest amount of reactive posterior glenoid bony displacement. Similarly, with anterior directed loads, anterior liftoff was decreased and neoglenoid bone displacements were reduced in all implants. Table 1 presents peak neoglenoid bone displacements as a function of net joint load vector magnitude/direction. Discussion: This study examined three designs of augmented glenoid components and compared them with respect to compressive contact, global reactive bone displacement, and implant lift off. These simulations found the posteriorstep implant displayed the best compressive contact with the remaining glenoid bone under posteriorly-directed net joint load vectors. However, the posterior-step also showed the highest displacements of the neoglenoid under posterior loading than the other two implants. This is likely due to the increased bone removal required to insert the posterior-step, which creates a posterior diving board-type phenomenon. In our model, the wedged implants preserved more neoglenoid bone, which likely decreased the global bone displacements observed in this region. The results suggest that compressive contact areas, global neoglenoid bone displacements, and implant liftoff are a function of the magnitude and direction of the applied joint load and the backside morphology of augmented implants. This study introduces the loading characteristics of augmented implants to simulated joint loads in B2 glenoids. Figure 1 Finite element results for three augmented glenoid component designs – posterior-step (top), posterior-wedge (middle) and full-wedge (bottom). The compressive contact area represents the load transfer area from the implant to the underlying bone. The global displacement is the magnitude of bone displacement with brighter green representing increasing displacement. The implant liftoff is shown by complete contact (blue), no contact (grey), and partial contact (light blue to green). Figures are for a 750 N compressive load directed six degrees posteroinferior in a right shoulder.

conditions. The compressive contact area, global neoglenoid bone displacement, and implant liftoff were measured (Fig. 1). Results: Compressive contact area was found to increase linearly with applied load and as the net joint load vector moved from anterior to posterior. The posterior-step and full-wedge implants had similar compressive

Paper #24 COMPARISON OF REVERSE TOTAL SHOULDER ARTHROPLASTY OUTCOMES WITH AND WITHOUT REPAIR OF THE SUBSCAPULARIS

Richard J. Friedman, MD, FRCSCa, Pierre-Henri Flurin, MDb, Thomas W. Wright, MD c , Joseph D. Zuckerman, MD d , Christopher Roche, MSE, MBAe, aMedical University of South Carolina, Charleston, South Carolina, USA; bBordeaux-Merignac Clinic, Bordeaux, France; cDepartment of Orthopaedics, University of Florida, Gainesville, Florida, USA; dNYU Hospital for Joint Diseases, New York, New York, USA; eExactech, Inc., Gainesville, Florida, USA Introduction: Repair of the subscapularis with reverse shoulder arthroplasty (rTSA) is controversial. Recent studies have presented conflicting reports about the necessity of repair on joint stability. Additionally, biomechanical studies have suggested that deltoid, posterior cuff, and joint reaction forces can be reduced by not repairing the