19TH AMERICAN AND SURGEONS OPEN MEETING

San Francisco, CA, USA February 8, 2003

1 PREDICTORS OF THE HUMERAL HEAD ISCHEMIA FOL- humerus fracture as part of a massive upper extremity injury. None LOWING FRACTURE OF THE PROXIMAL HUMERUS of the nerves recovered after primary repair. All eight intact ex- R. Hertel*, A. Hempfing, M. Stiehler, M. Leunig, Department of plored nerves and nine of ten unexplored nerves recovered, the , Inselspital, University of Berne, Switzerland only non-recovery occurring in a patient treated with closed in- The purpose of this study was to search for predictors of fracture tramedullary rod fixation that may have had iatrogenic nerve induced humeral head ischemia. Material and Methods: Between injury. The average time to initial signs of recovery was 7 weeks February 1998 and December 2001, 100 intracapsular fractures (range 1 to 25 weeks). The average time to full recovery was six of the proximal humerus, referred to the senior author for open months (range 1 to 21 months). Discussion: Transection of the surgery, were included in a prospective surgical evaluation proto- radial nerve is associated with open fractures of the humerus that col (mean age 60 y., min. 21, max. 88; 45 male; 57 right). are part of a massive upper extremity injury. The results of primary Fracture morphology was described using a binary description nerve repair in this circumstance are poor, likely related to an system. Perfusion was assessed intraoperatively by observation of extensive zone of injury and the need for nerve grafting. Intact backflow after drilling a borehole in the central part of the head in nerves and nerve palsies that are part of a closed fracture nearly all and by intraosseous laser-Doppler flowmetry in 46 . always recover, even after high-energy injuries. Because the first Results: 55/100 heads were ischemic. Good predictors of isch- signs of nerve recovery and complete recovery of the nerve can be emia were: the basic fracture pattern (for fractures comprising the anatomic neck: accuracy 0.7; PPV 0.71; NPV 0.68), the length of quite delayed, patience is merited prior to considering tendon the metaphyseal extension of the head (for calcar segment Ͻ8mm: transfers. accuracy 0.84; PPV 0.77; NPV 1), the integrity of the medial hinge (for disrupted hinge: accuracy 0.79; PPV 0.83; NPV 0.75). Mod- erate and poor predictors of ischemia were fractures with three 3 HUMERAL HEAD REPLACEMENT FOR ACUTE PROXI- fragments (accuracy 0.38, PPV 0.43, NPV 0.34), the angulation of MAL HUMERUS FRACTURES IN PATIENTS LESS THAN 60 the head (for angulations over 45°: accuracy 0.62; PPV 0.79, NPV YEARS OLD 0.55), the amount of dislocation of the tuberosities (for dislocations Gregory N. Lervick, MD*, Steven J. Klepps, MD, Mauricio Herrera, MD, over 10mm: accuracy 0.61; PPV 0.63; NPV 0.58), gleno-humeral Theodore A. Blaine, MD, William N. Levine, MD, Evan L. Flatow, MD, dislocation (accuracy 0.49; PPV 0.6; NPV 0.46) and head split Louis U. Bigliani, MD, New York Presbyterian Hospital, Department components (accuracy 0.49; PPV 0.61; NPV 0.46). Combining the of Orthopaedic Surgery, New York, NY above criteria (anatomic neck, short calcar, disrupted hinge) pos- Introduction: itive predictive values of up to 97% could be obtained. Conclusion: The treatment of comminuted four-part fractures of The most relevant predictors of ischemia were: the location of the proximal humerus is controversial in patients under the age of fracture planes (binary description system); the length of the calcar 60. The purpose of this study was to evaluate the functional segment and the integrity of the medial hinge. outcome of humeral head replacement (HHR) for acute fractures of the proximal humerus (PH) in patients less than 60 years old. Methods: We reviewed 22 patients between 1988 and 2000. There were 13 males and 9 females with a mean age of 51.7 years 2 RADIAL NERVE PALSY ASSOCIATED WITH HUMERAL (32-60). All patients were treated within three weeks of the injury. SHAFT FRACTURE: SHOULD HIGH-ENERGY INJURIES BE EX- The patients completed the American Shoulder and Elbow Surgery PLORED? (ASES) form, Simple Shoulder Test (SST), and a questionnaire David Ring, MD*, Kingsley Chin, MD, Amir H. Taghinia, BA, regarding work and recreational activities. Results: Average fol- Jesse B. Jupiter, MD, Massachusetts General Hospital, Boston, MA low-up was 82 months (24-172). Overall, there were 5 excellent Introduction: The management of radial nerve palsy associated (23%), 11 satisfactory (50%), and 6 unsatisfactory (27%) results. with fracture of the humeral diaphysis is disputed. Some recent The average ASES score was 68.6 (41.7-100), pain score 2.2 data suggest that high-energy closed fractures merit exploration. (0-5), and SST 7.9 (4-12). Active ROM averaged 134° (60-160) FE Methods: Twenty-four patients with a high-energy, diaphyseal frac- ture of the humerus and a complete radial nerve palsy were and 40° (20-70) ER. 18/21 (86%) returned to work at an average reviewed retrospectively. There were eighteen men and six women of 9.3 weeks after the surgery. Overall, 19/22 (86%) of patients with an average age of 27 years (range 16 to 41 years). Eleven had little or no pain that interfered with daily function. Of 14 fractures were open (Gustilo Type 2 in 5 and Type 3 in 6). Six of patients who were active before injury, 4 (29%) quit or altered their the 11 open injuries were considered ‘massive‘ injuries, defined as recreational activities after the surgery. Radiographs were avail- an open humerus fracture combined with either multiple ipsilateral able in 10 patients at average 80 months follow-up. 7 of these 10 upper extremity fractures and/or additional neurovascular injury patients demonstrated glenoid arthrosis, and 2 had advanced (near amputation). Five humerus fractures were part of a floating superomedial glenoid erosion. Discussion: We conclude that HHR elbow injury and eight patients had polytrauma. Nine fractures is a viable treatment option for selected PH fractures in this age were isolated injuries. All 11 patients with open fractures and 3 of group. The majority of patients are able to return to their occupa- 13 patients with closed injuries had radial nerve exploration. tions and activities, and have good pain relief and function with Results: All six patients with a transected radial nerve had open an intermediate follow-up.

1 2 Abstracts J Shoulder Elbow Surg

4 PREVIOUSLY UNRECOGNIZED DEFICITS FOLLOWING 6 LIGAMENTOUS RECONSTRUCTION FOR POSTEROLAT- CONSERVATIVE TREATMENT OF DISPLACED, MID-SHAFT ERAL INSTABILITY OF THE ELBOW USING AN AUTOGE- FRACTURE OF THE CLAVICLE DETECTED BY PATIENT-BASED NOUS TENDON GRAFT: 2 TO 8 YEAR RESULTS OUTCOME MEASURES AND OBJECTIVE MUSCLE STRENGTH David E. Kim, MD, FRCSC*, William D. Regan, MD, FRCSC, UBC TESTING Hospital, Vancouver, British Columbia, Canada Michael D. McKee, MD, FRCSC*, Beth Pederson, BSc, Introduction: The long-term results of reconstructing the lateral Caroline Jones, BSc, PT, Emil H. Schemitsch, MD, FRCSC, St ulnar collateral ligament of the elbow using an autogenous tendon Michael’s Hospital, Toronto, Ontario, Canada graft for posterolateral rotatory instability were reviewed. Methods: Purpose: We used a patient based outcome questionnaire and We reviewed a consecutive series of 13 patients with posterolat- objective muscle strength testing to evaluate a series of patients eral rotatory instability of the elbow treated with reconstruction of following conservative care of a displaced, midshaft fracture of the the lateral ulnar collateral ligament using either a palmaris tendon clavicle. Methods: We identified twenty-five patients (17 males, 8 autograft (12 patients) or a gracilis tendon autograft (1 patient). females, mean age 39 years) who had sustained a displaced Patients were seen both pre operatively and post operatively with a midshaft fracture of the clavicle. All patients were treated conser- mean follow up of 4.75 years (range 2-8 years). Results: All vatively. Mean follow-up was 54 months, with a minimum of 10 demonstrated at least one pre operative active apprehension sign: months. Outcome measures included the Constant shoulder score push up or chair. All had a positive posterolateral pivot shift test and the DASH questionnaire. Patients also had objective shoulder while under anaesthesia. Excellent or good results were obtained muscle strength testing performed on the Baltimore Therapeutic in 85% according to the Mayo Elbow Performance Score and in Equipment Work Stimulator. Results: was well 77% according to the rating system of Nestor et al. Stability was maintained, with flexion of 170° Ϯ 20°, and abduction of 165° Ϯ obtained in 10/13 (elimination of the active apprehension signs 25°. Strength of the injured compared to the uninjured side was: and pivot shift). Conclusion: Reconstruction of the lateral ulnar flexion (maximal) 81%, flexion (endurance) 74%, abduction (max- collateral ligament of the elbow using autogenous tendon graft is imal) 81%, abduction (endurance) 66%, external rotation 80%, successful in the treatment of posterolateral rotatory instability of the external rotation (endurance) 84%, and internal rotation (maximal) elbow. Active apprehension signs are eliminated. 86%. The mean Constant score was 69, and the mean DASH score was 25.1, indicating significant residual disability. Discussion and Conclusion: Traditionally, good results with minimal functional def- 7 MUSCLES AND FOREARM ROTATION EFFECT VALGUS icit have been reported following clavicle fractures treated conser- LAXITY OF THE ELBOW vatively. However, surgeon-based methods of evaluation may be insensitive to residual deficits. We detected residual deficits in Marc R. Safran, MD*, Michelle H. McGarry, MS, shoulder strength, especially endurance strength, in this patient Thay Q. Lee, PhD, University of California, San Francisco, CA Introduction: population. This may relate to the significant level of dissatisfaction Accurately diagnosing the presence and degree detected by patient-based outcome measures following conserva- of ulnar collateral ligament (UCL) injury is often difficult, even for tive care of displaced midshaft fractures of the clavicle. experienced clinicians. Valgus laxity of the elbow is difficult to confirm clinically with various reports recommending examination in pronation, supination and neutral rotation and varying degrees 5 A BIOMECHANICAL COMPARISON OF FOUR RECON- of elbow flexion. There have been no studies performed assessing STRUCTION TECHNIQUES FOR THE MEDIAL COLLATERAL valgus laxity with the UCL intact and cut in varying degrees of LIGAMENT OF THE ELBOW flexion and rotation with the entire upper extremity musculature April D. Armstrong, MD, FRCSC*, Cynthia E. Dunning, PhD, intact. The purpose of this study was to determine the effect of Chris Harwood, BEng, Louis M. Ferreira, BEng, Ken J. Faber, MD, FRCSC, forearm rotation and elbow flexion on valgus laxity. Methods: James A. Johnson, PhD, Graham J. W. King, MD, FRCSC, Bioengineer- Twelve fresh frozen cadaveric upper extremities with no radio- ing Research Laboratory, Hand and Upper Limb Centre, London, graphic evidence of arthritis, previous fracture or surgery of the Ontario, Canada elbow were used. The entire upper extremity was used to maintain We compared the initial strength of the intact medial collateral the origins and insertions of all muscles that cross the elbow that ligament (MCL) of the elbow and four reconstruction techniques; the may provide passive stability. The thawed arms were held in a Jobe, the docking, a single strand reconstruction utilizing interfer- custom instrumented elbow laxity testing device that permits the ence screws, and a single strand reconstruction which employed measurement of varus/valgus laxity of the elbow while applying a an endobutton for ulnar fixation. Twenty (ten pairs) unpreserved pre-specified valgus torque. Valgus laxity was measured using a cadaveric upper extremities were mounted in a custom jig with the Microscribe digitizing system. Using a muscle splitting approach to elbow at 90°, and a valgus force was applied 12 cm from the limit injury to other structures and maintain their effect on elbow elbow . The specimens were loaded starting at 20 N with the stability, the anterior band (AB) of the anterior oblique ligament load increased in increments of 10N (200 cycles at each load), (AOL) and the posterior band (PB) of the AOL were sectioned until either complete ligament failure ora5mmincrease in the randomly and sequentially. Valgus laxity was directly measured (in distance between the attachment sites of the MCL (i.e. 5 mm joint degrees) with the forearm in pronation (P), supination (S) and gapping). The average peak load to failure or 5 mm of joint neutral rotation (NR) at 30°, 50°, and 70° of elbow flexion using 1 gapping was 142.5 Ϯ 39.4 N for the intact, 53.0 Ϯ 9.5 N for the Nm and 2 Nm valgus torque with the ligament intact, joint vented, docking, 52.5 Ϯ 10.4 N for the endobutton, 41.0 Ϯ 16.0 N for following cutting of the anterior or posterior half of the AOL, and the interference screw, and 33.3 Ϯ 7.1 N for the Jobe reconstruc- following complete sectioning of the AOL. ANOVA was used for tions. The peak load to failure was higher for the intact specimens statistical analysis with pϽ0.05 set for statistical significance. compared to any of the reconstructions (pϽ0.001). The docking Results: In all cases, NR had more valgus angular displacement reconstruction showed higher peak loads than the Jobe or interfer- than P for all degrees of elbow flexion with the ligament intact, AB ence screw reconstruction, and the endobutton reconstruction transected and complete cutting of the AOL at 1 and 2 Nm of showed higher peak loads than the Jobe reconstruction (pϽ0.004). torque (pϽ0.05). With the PB only cut, N had more angular There was no difference in peak loads between the docking and displacement at 30° flexion with 1Nm torque, and at 30° and 70° endobutton reconstructions (pϾ0.05). The peak loads of the MCL with 2Nm force (pϽ0.05). NR revealed more valgus displacement reconstructions were considerably inferior compared to the intact than S in all degrees of elbow flexion with the AOL intact and ligament. The optimal fixation method for a single strand MCL completely transected at 2 Nm (pϽ0.05), while at 1Nm of torque, reconstruction may require improved interference screws or per- NR resulted in more displacement than S at 30° and 50° for both haps a modified endobutton and docking procedure. intact and complete AOL transection (pϽ0.05). With the AB only J Shoulder Elbow Surg Abstracts 3

cut, NR resulted in more displacement than S at all degrees of 9 SHOULD ACUTE ANTERIOR SHOULDER DISLOCA- elbow flexion and both torque forces (except 50° with 1 Nm force) TIONS BE IMMOBILIZED IN EXTERNAL ROTATION? and with the PB only cut, (pϽ0.05) NR resulted in a similar pattern Bruce S. Miller, MD*, Cameron Hatrick, MD, Sean O’Leary, MD, when compared with N versus P (pϽ0.05). Supination resulted in Jerome Goldberg, MD, David H. Sonnabend, MD, more forearm valgus displacement as compared with P at 30° and William R. Walsh, PhD, Orthopaedic Research Laboratories, Uni- 50° of elbow flexion with an intact ligament at both 1Nm and 2Nm versity of New South Wales, Sydney, Australia Ͻ torque (p 0.05), however, there was no difference at either force Introduction: The high recurrence rate associated with anterior Ͼ when the AOL was completely cut (p 0.05). With the AB cut at shoulder dislocations may reflect inadequate healing of a Bankart 30° of flexion with 1Nm force and at 50° of flexion with 2Nm force Ͻ lesion when the arm is immobilized in internal rotation. This study S resulted in more valgus displacement than P (p 0.05), otherwise investigates the effect of external rotation (ER) of the humerus on the there was no statistically significant difference between pronation glenoid-labrum contact parameters of a Bankart lesion in a human and supination at any other degree of flexion and no difference at cadaveric model. Methods: Ten human cadaveric shoulder girdles any degree of flexion-extension with AB only or PB only cut were fixed to a testing jig, a Bankart lesion was created arthro- (pϾ0.05). At 1Nm of force, 30° of flexion resulted in more valgus angular displacement than 50° with AOL cut and in S, (pϽ0.05) scopically, and a force sensor was placed between the detached while more displacement was noted with complete AOL transection labrum and the glenoid rim. The contact force between the glenoid in NR versus 70° (pϽ0.05). There was no statistical difference in labrum and the glenoid was measured in 60° of internal rotation, valgus displacement with flexion angle at 2 Nm testing (pϾ0.05). neutral rotation, and 45° of ER. The measurements were repeated There was no difference in cutting the PB or AB first in testing three times in each position and the mean contact force calculated (pϽ0.05). In all cases, 2 Nm force resulted in more displacement for each position of rotation. Results: There was no detectable than 1 Nm force (pϽ0.05), and fully transected AOL resulted in contact force with the arm in internal rotation. The contact force more angular displacement than the intact ligament (pϽ0.05). increased as the arm passed through neutral and reached a Discussion: The results of this study confirm the importance of maximum at 45 degrees of ER. The contact force returned to 0 forearm rotation in the assessment of valgus laxity of the elbow. when the arm was returned to neutral. The mean contact force at 45 Neutral rotation is the best position to examine the elbow for valgus degrees of ER was 83.5 g. Discussion and Conclusion: External laxity. The results from this study differ from previous published data rotation resulted in a significant increase in contact force between because the entire upper extremity was used and all musculature the labrum and glenoid. The improved soft-tissue contact parame- and their insertions were maintained. Although the muscles are not ters achieved in external rotation may influence the healing of a activated, they contribute to stability by passive muscle length and Bankart lesion. Immobilization of first-time anterior shoulder dislo- bulk effects. cations in an external rotation brace may reduce the high recur- rence rate associated with this injury. 8 A NEW METHOD OF IMMOBILIZATION AFTER DISLO- CATION OF THE SHOULDER: A PROSPECTIVE RANDOMIZED STUDY 10 GLENOHUMERAL ARTHROSIS IN ANTERIOR INSTABIL- Eiji Itoi, MD*, Yuji Hatakeyama, MD, Tadato Kido, MD, ITY BEFORE AND AFTER SURGICAL TREATMENT: INCIDENCE Takeshi Sato, MD, Hiroshi Minagawa, MD, Ikuko Wakabayashi, MD, AND CONTRIBUTING FACTORS Moto Kobayashi, MD, Akita University School of Medicine, Akita, 1 2 1 Japan Florent Buscayret, MD *, Istvan Szabo, MD , Gilles Walch, MD , 3 1 1 Purpose: Anterior dislocation of the glenohumeral joint is well T. Bradley Edwards, MD , Henri Coudane, MD , Lyon, France, 2 3 known of its high recurrence. Evidence has been provided that Pecs, Hungary, Minneapolis, MN conventional immobilization in internal rotation after shoulder dis- Introduction: Few large series of arthropathy related to ante- location is of no use in preventing further dislocations. We have rior glenohumeral instability are available in the orthopaedic shown in our MRI study that the Bankart lesion is better coapted in literature. We present a multicentric study of 570 patients with external rotation than in internal rotation in patients with initial operatively treated anterior glenohumeral instability evaluating dislocations. Based on this observation, we hypothesized that the incidence and the risk factors of preoperative and postop- immobilization in external rotation would decrease the recurrence erative glenohumeral arthritis. Methods: Five hundred seventy rate. The purpose of this study was to determine the effect of patients underwent an instability procedure, including 217 open immobilization in external rotation compared with internal rotation soft tissue procedures, 279 Bristow-Latarjets, and 74 arthro- on the recurrence rate after initial dislocation. Patients and Meth- scopic stabilizations. Clinical preoperative data was collected ods: Since January 2000, patients with initial anterior dislocation for these patients. We identified preoperative glenoid rim and of the glenohumeral joint were enrolled in this study. After the Hill Sachs lesions. Arthritis was evaluated preoperatively and shoulder was reduced, the patients were randomly assigned to two postoperatively with the Samilson classification (I, II, III, IV). The groups: immobilization in internal rotation (IR group) or in external mean age at surgery was 29.4 years (21 to 50 years). Follow-up rotation (ER group) for 3 weeks. As of March 2002, 40 patients averaged 6.5 years (3 to 16.5 years). Results: The preoperative were enrolled in this study with informed consents. There were 20 incidence of arthritis was 8.5% (I: 6.7%, II: 1.4%, III: 0.4%). patients in IR group (average age 38 yrs) and 20 patients in ER Arthritic risk factors were age at the initial dislocation and at group (average age 40 yrs). The average follow-up period was surgery, the length of time from the initial dislocation until 15.9 months for IR group and 13.7 months for ER group. The surgery, and the presence of osseous glenoid rim lesions recurrence rate and anterior apprehension sign were compared Ͻ between the groups using the chi-square test. Results: Recurrence (p 0.0001). Postoperative arthritis in patients without any pre- rate was 6/20 (30%) in IR group and 0/20 (0%) in ER group operative arthritis occurred in 19.9% (I: 14.4%, II: 3.4%, III: (pϭ0.0079). Among those who were younger than 29 years of 1.0%, IV: 1.1%) and was correlated with age at the initial Ͻ age, recurrence rate was 5/11 (45%) in IR group and 0/11 (0%) dislocation and at surgery (p 0.05), number of dislocations, Ͻ in ER group (pϭ0.0110). Anterior apprehension sign was positive and length of follow up (p 0.01). No statistical correlation in 2/14 (14%) in IR group without recurrence and 1/20 (5%) in ER existed between development of arthritis and the type of surgical group (pϭ0.35). Conclusion: Immobilization in external rotation technique employed. Discussion and Conclusion: Similar factors after shoulder dislocation is better than the conventional immobili- contribute to preoperative and postoperative arthritis in patients zation in internal rotation in terms of reducing recurrent disloca- with anterior glenohumeral instability, suggesting that surgery tions. does not prevent development of arthritis. 4 Abstracts J Shoulder Elbow Surg

11 CLINICAL RESULTS OF A PROSPECTIVE 15 YEAR MVC). During shoulder elevation tasks, symptomatics had signifi- STUDY ON 118 BRISTOW-LATARJET REPAIRS FOR RECUR- cantly greater supraspinatus (p Ͻ 0.03), infraspinatus (p Ͻ 0.05), RENT ANTERIOR DISLOCATION OF THE SHOULDER and upper trapezius (p Ͻ 0.04) muscle activation compared to Lennart Hovelius, MD*, Bjo¨rn Sandstrom, MD, Modolv Saeboe, MD, asymptomatics. During heavy elevation (8 lbs.), asymptomatics Kent Sundgren, MD, Orthopedic Department Gavle Hospital, Gavle, showed a trend toward increased activation (p Ͻ 0.06) of the Sweden subscapularis compared to symptomatics (34% MVC versus 21% Background: There are few studies reporting the long term MVC). Conclusion: Differential shoulder muscle firing patterns in results after different surgical repairs for recurrent anterior shoulder patients with massive pathology may play a role in the dislocations. In this prospective study on the Bristow Latarjet repair, presence or absence of symptoms. Asymptomatic subjects demon- that started in 1980 and ended in year 2001, we report the strated increased firing of the intact subscapularis, while symptom- outcome of 118/124 shoulders where the patients have been atic subjects continued to rely on torn rotator cuff tendons and followed for fifteen years. Six patients had died during the fol- peri-scapular muscle substitution resulting in compromised function. low-up period. Methods: We prospectively analyzed 118 shoul- ders in 113 patients two and fifteen years (mean 15.1, range 14.3-16.6 yrs) after a Bristow-Latarjet repair for traumatic recurrent 13 AN IN VIVO COMPARISON OF THE MODIFIED MA- anterior dislocation of the shoulder. The study was based on a SON ALLEN VERSUS A HORIZONTAL MATTRESS STITCH ON , scoring according Rowe et al (1978) and TENDON HEALING TO : A BIOMECHANICAL AND HIS- the patients’ subjective assessment of the operative result. Results: TOLOGIC STUDY IN SHEEP After two years one of the 118 shoulders had redislocated and 98% were satisfied with the operative repair. At fifteen years Theodore F. Schlegel, MD*, Richard J. Hawkins, MD, follow-up one patient had underwent revision surgery due to recur- Chad Lewis, BS, A. Simon Turner, BVSc, MS, Steadman Hawkins rence of the instability. One patient had had one recurrence and Sports Medicine Foundation/Colorado State University, Denver, one patient reported three recurrences three years postoperatively. CO Objectives: This patient has had no redislocations during the last twelve years. The modified Mason Allen stitch has been favored Further one patient had had two recurrences nine and twelve years for open rotator cuff repairs. This stitch pattern is extremely chal- after surgery. Subluxations occurred occasionally once (four pa- lenging to place using the current arthroscopic techniques. The tients) and several times in seven patients. The subluxations had simple and horizontal stitch patterns are technically more feasible ceased by time in some of these. Further one patient reported when performing arthroscopic rotator cuff repairs. The purpose of posterior subluxations at follow-up. Mean and medium score this study was to determine whether a horizontal stitch pattern that (Rowe) was 95 and 89.4 points respectively. Ninety patients (76%) we employ with special instrumentation is as secure as the modified were ‘very satisfied‘ with the operative result, twenty-six ‘satisfied‘ Mason Allen technique when repairing tendon to a bone trough. Methods: (22%) and one did not know. The patient with revision surgery was After performing a pre hoc power analysis, eighteen considered as unsatisfied. Bilateral involvement of shoulder insta- skeletally mature sheep had the right infraspinatus tendon reat- bility increased from twenty-two out of 118 (19%) at surgery to tached into a bone trough. Half of the sheep were randomly forty-one out of 117 (35%) fifteen years after surgery. Conclusions: assigned to the modified Mason Allen stitch group while the other The overall clinical results with a satisfaction rate of 98% and a half had placement of a horizontal mattress suture. Postoperatively, recurrence rate of 3.4% were as good as the best results reported the sheep were immobilized by fixing a soft ball underneath the after any operative method for recurrent anterior shoulder disloca- foot of the operative extremity and placing the animal into a small tion. stall to restrict movement. After 6 weeks, the ball was removed and the sheep were allowed unrestricted activity. At 26 weeks following the surgery, the animals were humanely euthanized and the shoul- ders were harvested for biomechanical and histologic testing. Six 12 DIFFERENTIAL PATTERNS OF MUSCLE ACTIVATION IN sheep from each group were prepared for biomechanical testing PATIENTS WITH SYMPTOMATIC AND ASYMPTOMATIC RO- by potting the humerus and attaching the ends of the infraspinatus TATOR CUFF TEARS to a brass cryogenically cooled grip. Using an MTS testing ma- Bryan T. Kelly, MD*, Riley J. Williams, MD, Frank A. Cordasco, MD, chine, the bone tendon interface were loaded to failure at a rate of Sherry I. Backus, MA, PT, James C. Otis, PhD, Daniel E. Weiland, MD, 500mm/minute. The load to failure (N) and the stiffness (slope, David W. Altchek, MD, Edward V. Craig, MD, Thomas L. Wickiewicz, MD, N/cm) were determined for each group. Statistical analysis was Russell F. Warren, MD performed using the paired student t-test. For the histologic evalu- Introduction: Treatment of massive rotator cuff tears (operative ation, three specimens from each group were fixed in 10% formalin versus non-operative) remains controversial. Our purpose was to and then 7␮ sections were obtained. The specimens were evalu- evaluate the differential firing patterns of rotator cuff, deltoid, and ated by a single pathologist blinded to treatment for assessment of scapular stabilizer muscle groups in normal controls and in patients the integrity of the tendon insertion site, the degree of collagen with symptomatic and asymptomatic two tendon rotator cuff tears. ingrowth, and the contribution of periosteal fibers to the organiza- Methods: Eighteen subjects were evaluated: six normals and twelve tion of the tendon attachment. Results: At 26 weeks, the mean load with two tendon cuff tears (six asymptomatic and six symptomatic). to failure of the modified Mason Allen group was 3223 Ϯ 1077 N Subjects were grouped based upon shoulder examination, and while the mean for the horizontal stitch was 3853 Ϯ 486 N. The outcomes questionnaires. All cuff tear patients had MRI scans mean stiffness of the modified Mason Allen group was 3243 Ϯ documenting a two tendon tear (supraspinatus and infraspinatus); 816 N/cm and 3543 Ϯ 412 N/cm for the horizontal mattress all normals had an ultrasound examination confirming the absence group. There were no statistically significant differences between of cuff pathology. Electromyographic activity from twelve muscles the modified Mason Allen stitch pattern and the horizontal mattress and kinematic data were collected simultaneously during ten func- group for load to failure or stiffness. The histologic findings were tional tasks. Results: Both symptomatic and asymptomatic cuff similar between the two treatment groups. Regardless of treatment, subjects demonstrated a trend toward increased muscle activation the infraspinatus tendon inserted into the bone through a zone of during all tasks compared to normals. During the internal rotation fibrocartilage. The junction between the tendon and the bone was tasks, asymptomatic patients had significantly greater (p Ͻ 0.05) complete and appeared normal. Conclusions: This in vivo study subscapularis activity than symptomatic patients (65% MVC versus would indicate that the use of a horizontal mattress suture pattern is 42% MVC). During the carrying task, asymptomatic patients dem- equal to the traditional modified Mason Allen stitch when evaluat- onstrated significantly less (p Ͻ 0.03) upper trapezius muscle ing the long term biomechanical properties of the tendon healing in activation than the symptomatic patients (16% MVC versus 50% a sheep. There are no discernible histologic differences detected J Shoulder Elbow Surg Abstracts 5

when comparing the two suture patterns. Therefore the horizontal painful tears were on the dominant side (pϽ.01). For intact cuffs, mattress suture, which is technically easier to perform arthroscopi- avg thickness equaled 4.7 mm and was not affected by age, cally, is appropriate for rotator cuff repair. gender, or symptoms. Discussion: The high incidence of asymptom- atic and bilateral tears with age suggests that for some, cuff disease is intrinsic; however, the association of symptoms with hand dom- 14 FATTY INFILTRATION OF THE ROTATOR CUFF IN A inance and increased tear size suggest that extrinsic factors may be RABBIT MODEL important in the development of pain. In pt. with an intact cuff, a L. Joseph Rubino, MD*, Harold F. Stills, Jr, DVM, decrease in thickness was not seen as a function of age thus, Lynn A. Crosby, MD, Wright State University School of Medicine degeneration does not appear inevitable for all and may be limited Objectives: The purpose of this investigation is to document to about 50% of the population. quantitatively and qualitatively the progression of fatty infiltration into atrophic rotator cuff muscle that was surgically detached to simulate a chronic . Methods: We surgically de- 16 THE CLINICAL OUTCOME FROM A RANDOMIZED tached the left supraspinatus tendon from its insertion in 20 New CONTROLLED STUDY OF ROTATOR CUFF REPAIRS RE- Zealand white rabbits. The contralateral rotator cuff tendon served VIEWED AT 6 AND 12 MONTHS as our control. At time intervals of six weeks, three months, six W. Angus Wallace, FRCSEd(Orth)*, Ian G. Kelly, MD, Notting- months and one year, the rabbits underwent whole body perfusion ham Shoulder & Elbow Unit, Nottingham, United Kingdom and the was harvested bilaterally from the Introduction: This study was carried out to investigate the out- rabbits. The specimens were measured, weighed and sectioned come of rotator cuff repair surgery in 14 centers in a randomized into five symmetrical segments, with segment one and the muscu- controlled trial. It also looked at a comparison of a long-acting lotendinous junction and section five at the muscle origin. There absorbable suture (Panacryl) and a non-absorbable suture (Ethi- were 200 total muscle specimens for histological evaluation. All bond). Methods: All patients were treated with open repair of their histological slides were converted to computer files for evaluation rotator cuff tear with modified Mason-Allen sutures used in 83% of using digital analyzing software. The percentage of fat in the cases. 171 patients were included in the analysis. Patients had specimens was calculated by subtraction of the area of muscle and Constant scores carried out pre-operatively, 6 and 12 months as fibrous tissue from the total area of the specimen. All 200 images well as ultrasound real time dynamic scans at 8 weeks, 6 and 12 were randomized before measurement to eliminate observer bias. months. Results: The results are shown in Tables I and II for the Two different individuals performed the measurements at different ‘Intention to treat‘ patients. There was no significant difference in times to detect interobserver error. Our results were analyzed using the analysis carried out for ‘Per-protocol patients.‘ Discussion and ANOVA and Pearson’s correlation coefficient. Results: We ob- Conclusions: There is a significant improvement in the Constant served a consistent decrease in muscle weight in the test specimens Score after rotator cuff repair surgery. However for large tears, the as compared to the controls. Fatty infiltration was evident 6 weeks re-tear rate at 6 months is approximately 50% as assessed by after detachment of the supraspinatus tendon, and increased over experienced ultrasonographers and deteriorates further by 12 time as a percentage of muscle volume (pϭ0.002; ANOVA). We months after surgery. An individual analysis of the outcomes for also observed a progression of fatty infiltration from the musculo- each of the 14 centers (anonymised) will be presented. Despite this tendinous junction proximally towards the muscle origin over time high re-tear rate there was still a good benefit from surgery. Is the (Pearson correlation r ϭ -0.51: pϽ0.0001). Conclusion: Fatty improvement in those cases with a re-tear a consequence of the infiltration is a progressive, infiltrative process that increases over sub-acromial decompression (SAD) and what would have been the time. It appears that fatty infiltration occupies space left by atrophic outcome with an SAD alone? muscle, and this infiltrative process progresses over time in the unrepaired rotator cuff tendon tear. 17 ARTHROSCOPIC REPAIR OF MASSIVE, CONTRACTED ROTATOR CUFF TEARS USING SINGLE AND DOUBLE INTER- 15 THE DEMOGRAPHICS AND MORPHOLOGY OF ROTA- VAL SLIDES TOR CUFF DISEASE: A COMPARISON OF ASYMPTOMATIC Ian K. Y. Lo, MD, FRCSC*, Stephen S. Burkhart, MD, The San AND SYMPTOMATIC SHOULDERS Antonio Orthopaedic Group, San Antonio, TX Konstantinos Ditsios, MD*, Sharlene A. Teefey, MD, Introduction: Prior reports of arthroscopic debridement of mas- William D. Middleton, MD, Charles F. Hildebolt, DDS, PhD, sive, contracted rotator cuff tears have demonstrated improvements Leesa M. Galatz, MD, Ken Yamaguchi, MD, Washington Univer- in pain. However, increases in strength, motion or function, were sity, St Louis, MO often not achieved. The purpose of this study was to review the Introduction: Little comparative information is available regard- results of arthroscopic repair of massive, severely contracted rota- ing asymptomatic and symptomatic rotator cuff tears. This data tor cuff tears using an interval slide technique. This technique may lend insight to the Natural History. The purpose of this study sufficiently mobilizes severely contracted massive rotator cuff tears was to compare the morphology and incidence of rotator cuff to the lateral bone bed, allowing a tension free repair to bone. disease in people who presented with unilateral painful shoulders. Methods: Between January 1999 to December 2000, the senior Methods: 588 consecutive shoulder ultrasounds were reviewed for author performed arthroscopic rotator cuff repair on 94 massive patients presenting with unilateral pain. Ultrasound has been pre- rotator cuff tears. Of these, 9 (9.6%) were massive, severely viously validated in our institution as highly accurate. There were contracted rotator cuff tears and required repair using an interval 212 pt/bilat intact cuffs, 191 pt/unilateral tears, and 185 pt/ slide technique. The mean tear size was 5.1 cm X 6.2 cm with a bilateral tears. Data for age, cuff thickness, tear size, and symp- mean tear area of 31.8 cm2. A single anterior interval slide was toms were tested for statistical associations. Results: Cuff disease used in 6 patients (releasing the interval between the supraspinatus increased with average age, (no tear ϭ 48.7yo; unilateral tear ϭ tendon and the rotator interval) and a double interval slide was 58.7yo; bilateral tears ϭ 67.8yo). Logistic regression showed a used in 3 patients [anterior interval slide ϩ posterior interval slide 50% likelihood of bilateral tears after age 66 (pϽ.01). Overall, a (releasing the interval between the supraspinatus and infraspinatus pt. with a full tear on the symptomatic side had a 35.5% incidence tendons)]. In 4 patients a complete repair was obtained and in 5 of a contralateral asymptomatic tear. In contrast, presentation of a patients a partial repair was obtained. All patients were evaluated symptomatic nl. or partial tears had only a 0.5% or 4.3% incidence pre-operatively and post-operatively using a modified UCLA scor- of contralateral tear. In patients with bilateral tears, the symptom- ing system. Results: At a mean follow-up of 17.9 Ϯ 4.0 months atic side was larger by an average of 25% (pϽ.01). 65% of (range: 10-24 months), 8 of 9 patients were satisfied with the Abstracts 6

Table I Results

CONSTANT PAIN SCORES TOTAL CONSTANT SCORES

Pre-op 6/12 Post-op 12/12 Post-op Pre-op 6/12 Post-op 12/12 Post-op

Suture All <5cm >5cm All <5cm >5cm All <5cm >5cm All <5cm >5cm All <5cm >5cm All <5cm >5cm

Panacryl (n) 6.3 (75) 6.3 (50) 6.4 (25) 11.4 (75) 11.4 (50) 11.4 (25) 12.1 (75) 11.9 (50) 12.4 (25) 45.6 (75) 46.0 (50) 44.8 (25) 64.1 (75) 65.2 (50) 62.1 (25) 70.7 (75) 72.9 (50) 66.3 (25) Ethibond (n) 6.0 (68) 6.3 (50) 4.9 (18) 11.5 (68) 11.9 (50) 10.4 (18) 11.5 (68) 11.9 (50) 10.4 (18) 47.1 (67) 48.2 (49) 43.9 (18) 67.0 (67) 70.4 (49) 57.7 (18) 72.4 (67) 75.5 (49) 64.2 (18)

Table II Results

Re-Tear Rates

At 8 weeks At 6 months At 12 months

Suture All <5cm >5cm All <5cm >5cm All <5cm >5cm hudrEbwSurg Elbow Shoulder J

Panacryl (n) 12 (15%) (83) 4 (8%) (60) 8 (30%) (23) 25 (30%) (83) 11 (20%) (60) 14 (48%) (23) 32 (38%) (83) 15 (27%) (60) 17 (57%) (23) Ethibond (n) 9 (12%) (88) 1 (2%) (57) 8 (36%) (31) 16 (20%) (88) 5 (8%) (57) 11 (50%) (31) 20 (24%) (88) 7 (12%) (57) 13 (59%) (31) J Shoulder Elbow Surg Abstracts 7

procedure. The mean UCLA score increased from 10.0 Ϯ 2.2 were found to have a defect in the RCR at some point post-op. pre-operatively to 28.3 Ϯ 4.2 post-operatively (p Ͻ 0.00001). The Defects were seen in 50% pts at T#1, 45% at T#2, and 43% at mean pain score improved from 2.1 Ϯ 0.8 pre-operatively to T#3. Majority of RC defects were classified as Type 1 B. Average 8.7 Ϯ 1.7 post-operatively (p Ͻ 0.00005). All patients demon- defect size at each time point was 1 x1 cm2. 33% of asymptomatic strated at least some improvement in motion, strength or function. controls were found to have a defect consistent with RCT, 7.6x7.1 Active forward flexion (108° Ϯ 36.5° to 146.1° Ϯ 20.1°; pϭ mm. Discussion/Conclusion: Outcomes tools demonstrate consis- 0.025) and external rotation (24.4° Ϯ 22.1° to 35.0° Ϯ 19.5°; tently but not statistically significantly worse short-term result with pϭ 0.04) increased significantly. There was also a significant ARCR compared to MORCR and ORCR. Whether or not poor increase in strength grade (2.2 Ϯ 1.4 to 3.6 Ϯ 0.7; pϽ 0.005) short-term results are a function of surgical learning curve or and function (2.5 Ϯ 0.7 to 7.1 Ϯ 1.5; pϽ 0.0005). There was no whether result is dependent on number of suture anchors and significant difference in patients who achieved a partial versus subsequent suture strands crossing the repair is unclear. complete repair. There were no significant complications to the procedure. Discussion and Conclusions: The interval slide technique provides a method of mobilization of severely contracted massive 19 OUTCOMES OF COMBINED ARTHROSCOPIC CAPSU- rotator cuff tears allowing repair of previously irreparable tears. LAR RELEASE, , AND DISTAL CLAVICLE RE- While the anterior interval slide improves supraspinatus tendon SECTION FOR END-STAGE GLENOHUMERAL ARTHRITIS mobility approximately 1-2 cm, the function of the posterior interval Robert J. Nowinski, DO*, Wayne Z. Burkhead, MD slide is two fold. First, it increased supraspinatus tendon mobility Introduction: End-stage glenohumeral arthritis in younger pa- over and above that of an anterior interval slide (approximately 2-3 tients and those older patients looking for an alternative to replace- cm) and second, the posterior interval slide also improves the ment presents a formidable challenge during clinical mobility of the residual posterior rotator cuff. This improves superior decision-making. The purpose of this study was to evaluate the and lateral mobility of the infraspinatus/teres minor and may allow results of combined arthroscopic capsular release, acromioplasty, repair of these tendons, particularly the inferior infraspinatus ten- and distal clavicle resection in patients with end-stage arthritis of don, in a biomechanically favorable position. This is essential in the glenohumeral joint and to determine the factors associated with re-establishing a normal glenohumeral fulcrum and is likely the its success. Methods: Thirty-three patients with end-stage glenohu- reason that partial repairs fared as well as complete repairs in this meral arthritis were treated with arthroscopic debridement and small series. In conclusion, arthroscopic single and double interval lavage, capsular release, acromioplasty, distal clavicle resection, slides in conjunction with arthroscopic rotator cuff repair may biceps tenotomy or tenodesis, and manipulation under anesthesia. provide improvements in pain, motion, strength and overall shoul- All patients had radiographic diagnosed end-stage osteoarthritis der function in patients with massive, severely contracted rotator with osteophyte formation, subchondral sclerosis, and complete cuff tears. collapse of the joint space, with a flat or biconcave glenoid. From a radiographic standpoint, these patients would normally be can- didates for shoulder arthroplasty. Results: Patients had an average age of 62 years old (range 36-69) and were followed for at least 18 SHORT TERM RESULTS COMPARING ROTATOR CUFF 24 months. Standardized data collection was performed at the REPAIR TECHNIQUES: PROSPECTIVE DOUBLE BLIND STUDY initial office visit and at most recent follow-up. Overall outcome was USING ULTRASOUND AND CLINICAL OUTCOMES analyzed by the Simple Shoulder Test and with regard to patients’ Stephen Fealy, MD*, Ronald S. Adler, PhD, MD, self-assessment of pain, function, improvement, satisfaction, and Mark C. Drakos, BA, Anne M. Kelly, MD, Answorth A. Allen, MD, duration of pain relief. Excellent results were reported in 15 pa- Frank A. Cordasco, MD, Stephen J. O’Brien, MD, Department of tients (46%) and good results in 10 patients (31%). Overall, 77% Sports Medicine & Shoulder Service, Hospital for Special Surgery, achieved a satisfactory outcome (excellent or good result). Five New York, NY patients (15%) reported fair results and three patients (8%) reported Introduction: Several studies have compared mini-open poor results. One of the poor results did not initially have a distal (MORCR) to open rotator cuff repair (ORCR). Arthroscopic RCR clavicle resection performed. Three patients required revision to (ARCR) has encouraging short-term results. We sought to objec- arthroplasty at an average of 28 months (range 12-48) after tively compare early results between techniques in a prospective . All three cases initially presented with a biconcave double-blind fashion. Methods: 50pts undergoing RCR were en- glenoid as diagnosed by axillary radiographic. The mean patient rolled; 28 MORCR, 14 ORCR, 8 ARCR. 20 shoulders served as pain score (0 ϭ no pain; 10 ϭ worst pain) improved from 6.3 controls. Avg. pre-op RCT size was 12x20mm; age of the RCT was preoperatively to 2.8 at final follow-up, with 80% of patients consistent for groups. Outcomes: Patients underwent physical ex- indicating that they would have the surgery again. Average post- amination and completed UCLA, L’Insalata, and ASES Shoulder operative motion increased for forward elevation by 42 degrees questionnaires. Vascular: Pts underwent Power Doppler sonogra- and for external rotation by 32 degrees. Workers’ compensation phy (PDS) of affected shoulders at 6wks (T#1), 3mos (T#2), and patients obtained inferior subjective results overall; however, im- 6mos (T#3) post-op. Results were blinded. Anatomic: Gray scale provements in pain and function were found in all patients. Con- features documented well-marginated intrasubstance, partial or clusions: In well-selected patients with end-stage arthritis of the full-thickness defects, thinning of repair, location of anchors. De- glenohumeral joint, significant improvements in pain relief and fects categorized according to Harryman classification. Results: function can follow arthroscopic treatment. The addition of the Outcomes: Ultrasound findings of a defect in the cuff repair did not concomitant procedures of acromioplasty, distal clavicle resection, correlate with functional assessment and outcome at 1 year when and biceps tenotomy/tenodesis after arthroscopic debridement compared to those without defect. UCLA, L’Insalata, and ASES and capsular release demonstrated a positive impact on the func- score was 28.8, 80.6, and 81. ARCR did slightly worse at 6mos tional results. Patients with a biconcave glenoid, however, tend to than ORCR or MORCR, according to questionnaire. Average of be associated with return of pain and failure of this procedure. 2.2 anchors used in the ARCR group, 3.2 for MORCR, 5 for ORCR. Vascular: There is a significant and predictable decrease in RC vascularity over time regardless of surgical technique. Average 20 EFFICACY OF ARTHROSCOPIC DEBRIDEMENT FOR vascular score was 11.6/T#1, 8.3/T#2, 7/T#3, and 2.4 for TREATMENT OF GLENOHUMERAL ARTHRITIS controls. The most robust flow was at peritendinous region at each Daniel D. Feldmann, MD*, John F. Orwin, MD, University of time; lowest vascular flow was at anchor site/cancellous trough. Wisconsin, Madison, WI There was no significant difference in vascular score between RCR Osteoarthritic changes of the glenohumeral joint are often a with a documented defect and those without. Anatomic: 48% of pts challenge to treat. Non-operative management of known arthritis of 8 Abstracts J Shoulder Elbow Surg

the shoulder is limited to physical therapy and pharmacologic between L4 and L5. Three of the children who had an isolated intervention. Traditionally, shoulder arthroplasty, either hemi or release developed a recurrent contracture requiring further treat- total, has been the surgical treatment of choice. Recent clinical ment. None of the children that had a latissimus transfer developed research has suggested that arthroscopic debridement may be of a recurrent contracture. MRI at 2-year’s F/U available in 12 some value in the management of this problem. The purpose of this children showed reversal of pre-operative deformity in all but one study was to perform a retrospective chart review and long term child, again the oldest child. Conclusions: Arthroscopic release is follow-up of patients having undergone arthroscopic debridement effective in achieving passive external rotation. Most young chil- of the shoulder for known glenohumeral arthritis. Twenty five pa- dren have sufficient external rotation strength at early follow-up to tients where identified as undergoing arthroscopic debridement for maintain a centered glenohumeral joint and normalize glenoid shoulder arthritis. Nineteen where available for long-term follow-up development. Improvements in elevation are modest and loss of of 3.8 yrs (1.8-8 yrs). Average age was 52.9 yrs (40-68 yrs). There internal rotation is common. were 17 men and 2 women. Surgical technique consisted of extensive debridement of damaged grade IV glenohumeral osteo- chondral lesions and labral structures in all cases. Additional 22 UNSTABLE NONUNIONS OF THE DISTAL HUMERUS ϭ procedures included bursectomy (n 5), biceps tendon release David Ring, MD*, Lawrence Gulotta, BA, Jesse B. Jupiter, MD, ϭ (n 5) and subacromial decompression/distal clavicular resection Massachusetts General Hospital, Boston, MA ϭ (n 1). No capsular release or manipulation was performed. Data Introduction: Some nonunions of the distal humerus are so collection involved a review of the and a detailed unstable that the hand and forelimb cannot be supported against follow-up clinic visit. Pre and post operative range of motion, gravity. The purpose of this retrospective study was to analyze the strength, pain and function scores were obtained. A modified results of open reduction and internal fixation, joint contracture version of the UCLA shoulder assessment was utilized. Duration of release, and autogenous in the treatment of these pain relief and satisfaction with surgery was also recorded. Anal- unstable nonunions of the distal humerus. Methods: Fifteen patients ysis of the data revealed that there was significant pain relief (UCLA with unstable nonunions of the distal humerus were treated with ϭ assessment, 2.17 to 3.7, p .0023) at final follow-up in 11 of 19 excision of fibrous and synovial tissues, opening of sclerotic frac- patients. Sixty eight percent were satisfied with their pain relief at ture surfaces, internal fixation with multiple plates and screws, and final follow-up. Range of motion in all planes measured decreased autogenous bone grafting. There were thirteen females and two ϭ significantly (forward flexion: 148’ to 132’, p .008; supine exter- males with an average age of sixty years. The average time from ϭ nal rotation: 55’ to 30’, p .01; supine internal rotation, 24’ to the original fracture to the index treatment of the nonunion was ϭ 10’, p .02). Function as measured by the UCLA score did not eleven months. Vascularized fibular grafts and supplemental exter- change. Three patients (16%) went on to hemi arthroplasty. No nal fixation were necessary in two patients with large bone defects complications were reported. This is the first study to our knowledge after debridement of prior infection. Results: Three nonunions failed which specifically reviews the outcome of arthroscopic debride- to heal and were treated with total elbow arthroplasty. Twelve ment of known shoulder arthritis; previous studies have reviewed nonunions healed, but five of the twelve required additional sur- patient outcomes following an intraoperative diagnosis of shoulder gery to address painful implants, ulnar neuropathy, or elbow osteoarthritis. Based on our current outcomes, we believe arthro- contracture. At an average follow-up of fifty-one months (range scopic debridement is a valuable intermediate step between non- twenty-four to 130 months), the remaining twelve patients had an operative management and arthroplasty in the treatment of osteo- average of ninety-five degrees of ulnohumeral motion (range sixty arthritis of the shoulder. It can significantly provide lasting pain to 130 degrees) with an average flexion of 115 degrees (range relief. Subacromial decompression and distal clavicular resection ninety to 135 degrees) and an average flexion contracture of do not seem to be necessary for pain relief. We have not found it twenty degrees (range, zero to thirty-five degrees). According to to restore or maintain ROM of the shoulder joint, nor have we found the Mayo Elbow Performance Index, the functional result was rated it to improve overall function. The procedure has low morbidity and excellent in two patients, good in nine patients, and fair in one should not interfere with future arthroplasty. patient. Conclusions: Unstable nonunion of the distal humerus can be successfully treated in most active, healthy patients by rigid internal fixation, joint contracture release, and bone grafting. 21 ARTHROSCOPIC RELEASE OF SHOULDER INTERNAL ROTATION CONTRACTURES SECONDARY TO BIRTH PALSY: EARLY FOLLOW UP OF 26 CHILDREN 23 RECONTOURING FOR THE TREATMENT Michael L. Pearl, MD*, Bradford W. Edgerton, Paul A. Kazimiroff, OF EXTENSION MALUNIONS OF THE DISTAL HUMERUS Raoul J. Burchette, Karyn Wong, Kaiser Permanente Los Angeles Theodore A. Blaine, MD1*, Shawn W. O’Driscoll, PhD, MD2, Medical Center, Los Angeles, CA 1Columbia University Center for Shoulder, Elbow, and Sports Introduction: Internal rotation contractures secondary to birth Medicine, New York, NY, 2Mayo Clinic, Rochester, MN palsy frequently lead to glenohumeral deformity. Treatment recom- Introduction: Surgical restoration of elbow flexion in patients mendations range from open release to primary latissimus dorsi with extension-type malunions of the distal humerus has tradition- transfer. This is a report of 26 arthroscopic contracture releases ally been performed by distal humeral and internal with F/U ranging from 1 to 2.5 years. Methods: Arthroscopic fixation. Recontouring ostectomy (RCO) is an alternative procedure contracture releases were performed on 26 children, aged 0.8 to that restores elbow flexion without osteotomy and internal fixation, 12 years. Older children (mean age 7.3 years) also received a and may be performed through arthroscopic or open techniques. latissimus dorsi transfer; younger children (mean age 1.6 years) Methods: Nine patients with an average age 20 years had RCO of only a release. Pre-operative passive external rotation averaged the distal humerus to restore elbow flexion after distal humeral Ϫ33° and 0° for the two groups, respectively. Eleven children malunion. All patients had prior surgical treatment of distal humeral received a latissimus transfer in addition to a release, 15 a release fractures. The average interval between the index procedure and only. Pre-operative arthrograms were done in all cases, MRIs in 24. RCO was 74 months. CPM was used post-operatively in all pa- Results: Arthroscopic release was successful in achieving passive tients. Patients were evaluated for elbow range of motion at aver- external rotation and a centered position of the glenohumeral joint age one-year follow-up. Statistical differences were determined at the time of surgery in all but one case, the oldest child with severe using ANOVA (pϽ0.05). Results: Flexion improved an average of deformity. At F/U, the average increase in external rotation was 23 degrees from 104 degrees preop to 127 degrees postop 86° for the latissimus transfer group and 55° for the release group. (pϽ0.05). 78% (7/9) had flexion to 125 degrees or greater. The average gain in elevation was 7°. Internal rotation averaged Extension also improved from 44 to 19 degrees postoperatively J Shoulder Elbow Surg Abstracts 9

(pϽ0.05). The total arc of motion improved significantly from 60 1988 to 2000 were identified. There were 9 males and 7 degrees pre-operatively to 108 degrees post-operatively (pϽ0.05). females. The average age was 52 years (29-77). The average 56% (5/9) patients had a functional 100 degree arc of motion with interval from index surgery to revision TSA was 3.5 years (11 extension to at least 30 and flexion to at least 130 degrees months-10.5 years). One patient was lost to follow-up. Patients post-operatively. Discussion/Conclusions: RCO of the distal hu- were evaluated clinically and with standard questionnaires eval- merus is a successful surgical procedure for restoring adequate (but uating functional outcome and pain. Results: At an average of not full) motion to with post-traumatic extension malunions. 5.5 years (2-14) postoperatively, there were 3/15 (20%) excel- Based on our experience with this procedure, as well as with lent, 5/15 (33%) satisfactory, and 7/15 (47%) unsatisfactory corrective osteotomy, it is less likely to result in serious complica- results according to Neer’s criteria. Two patients had subse- tions and much easier to perform. quent revision arthroplasty for glenoid loosening and were considered failures. One patient was diagnosed with a deep infection necessitating a two-stage reconstruction to a total 24 OSTEOCAPSULAR ARTHROPLASTY FOR TREATMENT shoulder arthroplasty. The average American Shoulder and OF PRIMARY OSTEOARTHRITIS OF THE ELBOW: DESCRIP- Elbow Surgeons score was 73.6 (46.7-95). The average pain TION OF PROCEDURE AND EARLY RESULTS score was 2.4 (0-6). The simple shoulder test score averaged Joseph Mileti, MD*, Mauricio Largacha, MD, Shawn W. O’Driscoll, PhD, MD, 9.3 (4-12). Improvement in forward elevation averaged 50°. Mayo Clinic, Rochester, MN Improvement in external rotation averaged 31°. 11/15 (73%) Introduction: Primary osteoarthritis of the elbow typically in- of the patients were satisfied with their result. Discussion: Our volves three pathologic entities: 1) loose bodies, 2) osteophytes in data shows that revision TSA for failed HHR due to glenoid and around the olecranon, coronoid and radial fossae, as well as arthrosis provides inconsistent pain relief and unsatisfactory on the tip and sides of the olecranon and coronoid, 3) capsular results in a substantial number of patients. We conclude that contracture. Current standards of treatment including the ulnohu- revision of a failed HHR to a TSA for glenoid arthrosis is a meral arthroplasty (Outerbridge-Kashiwaga arthroplasty) address salvage procedure whose results are inferior to those of primary some, but not all, of these changes. The purpose of this preliminary TSA. report is to describe our technique of osteocapsular arthroplasty and its early results for correcting all 3 types of pathological changes. Methods: Seventeen patients with primary osteoarthritis of the elbow underwent arthroscopic osteocapsular arthroplasty 26 MINIMUM 15 YEAR FOLLOW-UP OF NEER HEMIAR- and have at least one year follow-up. All patients used a continuous THROPLASTY AND TOTAL SHOULDER ARTHROPLASTY IN passive motion machine postoperatively for 2-6 weeks. Range of PATIENTS 50 YEARS OLD AND YOUNGER motion was measured and prospectively recorded until most recent John W. Sperling, MD1*, Robert H. Cofield, MD1, Ϯ follow-up. Results: Preoperatively, the total arc of motion was 76 and Charles M. Rowland, MS2, 1Department of Orthopedic Sur- Ϯ Ϯ 13 degrees with extension to 36 7 degrees and flexion to 116 gery, Mayo Clinic, Rochester, MN, 2Department of Biostatistics, Ϯ 10 degrees. Final arc of motion was 124 13 degrees from an Mayo Clinic, Rochester, MN Ϯ Ϯ average extension of 12 9 degrees to flexion of 136 7 Introduction: We reviewed our experience with Neer hemi- degrees. Improvements in flexion (pϽ0.001), extension Ͻ Ͻ arthroplasty (HHR) and total shoulder arthroplasty (TSA) in pa- (p 0.001) and total arc of motion (p 0.001) were statistically tients 50 years old or less to determine the results, risk factors for significant. A functional arc of motion (i.e. extension to 30 degrees an unsatisfactory outcome, and the rates of failure. Methods: or better and flexion to 130 degrees or more) was obtained in 15 Between 1976 and 1985, 78 Neer HHR and 36 Neer TSA were of 17 cases. Pre-operatively, all patients had moderate to severe performed by the senior surgeon in patients who were 50 years pain at the end-points of motion. Post-operatively thirteen patients old or less. Sixteen patients died and seven were lost to follow- had no pain, three patients had mild pain at the endpoint of up. The remaining 62 HHR and 29 TSA were followed for a extension and one patient had moderate pain in full extension. The minimum of fifteen years (mean, 16.7 years) or until revision only complication in this series was the development of a seroma in Results: one patient that resolved with aspiration. However, more recently, surgery. Shoulder arthroplasty resulted in significant long-term pain relief, improvement in active abduction, and we have learned that ulnar nerve compression can limit flexion Ͻ postoperatively or even develop into a neuropathy that requires a external rotation (p 0.01). There was not a significant differ- transposition. Conclusion: Osteocapsular arthroplasty corrects the ence between the two procedures with respect to these vari- 3-dimensional pathologic changes in the bony anatomy of the ables. Radiographs were available for 53 HHR and 25 TSA with elbow, thereby eliminating impingement. Resection of the con- a minimum ten year follow-up. Humeral peri-prosthetic lucency tracted capsule permits restoration of close to normal motion. Early was present more frequently after TSA (60%) compared to HHR ϭ results are encouraging and appear to be better than the senior (34%) (p 0.0079). Glenoid erosion was present in 38 of 53 author’s results after open treatment of primary elbow osteoarthri- HHR (72%). Glenoid peri-prosthetic lucency was present in 19 of tis. 25 TSA (76%). According to a modified Neer result rating system, among the HHR, there were 6 excellent, 19 satisfactory, and 37 unsatisfactory results. Among the TSA, there were 6 25 GLENOID ARTHROSIS AFTER HEMIARTHROPLASTY: excellent, 9 satisfactory, and 14 unsatisfactory results. The RESULTS OF REVISION TO TOTAL SHOULDER ARTHRO- estimated survival of the HHR was 82% at ten years and 74% at PLASTY twenty years. Patients who underwent HHR for the sequelae of Raymond M. Carroll, MD1*, Michael Vazquez, MD2, trauma had a significantly higher rate of revision surgery ϭ Rolando Izquierdo, MD3, William N. Levine, MD3, (p 0.04). The estimated survival of the TSA was 97% at ten Theodore A. Blaine, MD3, Louis U. Bigliani, MD3, 1Washington, years and 84% at twenty years. Conclusions: The data from this DC, 2Boston, MA, 3New York-Presbyterian Medical Center, Co- study indicate there is marked long term pain relief and improve- lumbia University, College of Physicians and Surgeons, New York, ment in motion with shoulder arthroplasty; however, when ap- NY plying a result rating system, nearly half of all young patients We report our results of sixteen consecutive patients who who undergo shoulder arthroplasty have unsatisfactory results. underwent revision total shoulder arthroplasty (TSA) for failed Great care must be exercised, and alternative methods of treat- hemiarthroplasty (HHR) due to glenoid arthrosis. Methods: Six- ment considered, prior to offering shoulder arthroplasty to teen patients who underwent revision TSA for failed HHR from young patients. 10 Abstracts J Shoulder Elbow Surg

27 MODIFIED IMPACTION BONE GRAFTING: A TECH- der Test (SST) and Visual Analogue Scale (VAS). Results: Motion NIQUE FOR HUMERAL COMPONENT FIXATION IN SHOUL- improved an average of 28 degrees with forward elevation, 22 DER ARTHROPLASTY degrees with external rotation, and two spinal segments with Michael A. Wirth, MD1,2,3*, Charles A. Rockwood, Jr, MD1,2,3, internal rotation. SST improved from 4.1 positive responses to 9.1. Moon Sup Lim, MD1,2,3, T. Kenneth Kaar, MD1,2,3, VAS scores for pain and function improved from a mean of 6.0 to Rebecca Loredo, MD1,2,3, 1University of Texas Health Science 1.6. Only two patients required revision to a total shoulder arthro- Center, San Antonio, TX, 2Carleton Southworth, MS, 3Warsaw, IN plasty. Discussion: The long-term results of this study demonstrate Introduction: Difficulties with fixation of the glenoid component that excellent pain relief and improvement in range of motion can account for most of the complications related to loosening of total be achieved and sustained with time. There was no significant shoulder prostheses. However, radiographic evaluations have change in results when comparing the 2 vs. 5-10 year follow-up. shown that complete radiolucent lines about the humeral implant This may be explained by employing hemiarthroplasty in patients and subsidence of the component are not common. Although press with concentric glenoids, as well as surgeon controlled variables fitting of humeral components usually results in satisfactory fixation such as selecting anatomic component sizing, soft tissue balancing, at the time of surgery, cement is occasionally required to provide and ensuring proper version. secure fixation. The purpose of this study was to describe our technique of modified impaction bone grafting, which was used in place of cement when press-fit implantation failed to provide secure 29 GLENOHUMERAL FOR FAILED ASEPTIC fixation of the humeral prosthesis. We also describe our early PROSTHETIC ARTHROPLASTY WITH SEVERE BONE, ROTA- prospective outcome results and the radiographic findings in the TOR CUFF AND DELTOID DEFICIENCY proximal humerus. Methods: Between January 1992 and July Joseph P. Iannotti, MD, PhD*, Edwin Spencer, MD, The Cleveland 1997, seventy-three shoulders (sixty-eight patients) had a shoulder Clinic Foundation, Cleveland, OH replacement arthroplasty with impaction bone grafting of the prox- Introduction: Surgical management of severe pain and disabil- imal humerus. There were thirty-eight males and thirty females with ity associated with aseptic failure of a prosthetic shoulder arthro- an average age of sixty years, (range, twenty-seven to eighty-three plasty associated with bone loss, irreparable rotator cuff tears and years). Bone graft appearance was classified as showing no deltoid deficiency present a significant challenge. Surgical options change, localized resorption, or incorporation with trabecular include resection arthroplasty and arthrodesis. This is a retrospec- remodeling. Humeral component subsidence and change in varus/ tive review of the results and complications associated with arthro- valgus alignment were measured. Outcome was assessed with six desis in this patient population. Methods: Nine patients, ages visual analog scales (VAS), twelve patient administered pain and 35-60 with 3-7 prior surgical procedures, underwent arthrodesis of functional assessment questions, and the American Shoulder and the shoulder using vascularized fibula when there was complete Elbow Surgeons Shoulder Score Index (SSI). Results: At an average loss of the tuberosities (3 cases) and large structural allografts when follow-up of fifty-six months (range, two to seven years) all patients the tuberosities were partly intact (6 cases). The patients were had reported an improvement in pain and function. All Visual followed an average of 36 months or until bone union and com- Analog Scales and the Shoulder Score Index demonstrated signif- pletion of there rehabilitation. Results: Average pre-operative and Ͻ icant improvement (p 0.05). Radiographs demonstrated no post-operative ASES scores were 18 and 50 respectively, pain change in varus/valgus alignment. Subsidence of the humeral scores improved from 8.5 to 1.8. Pre-operatively all patients were implant was noted in one asymptomatic patient with progressive limited to painful waist level function and post-operatively all had radiolucent lines at the bone-cement interface of the glenoid four stable pain free shoulder or chest level function. All patients ulti- years following a total shoulder arthroplasty. This was the only mately had a successful fusion. Four of the nine patients required shoulder that demonstrated bone graft resorption. Conclusion: The revision bone grafting and ORIF for incomplete union, one of these technique of impaction grafting is well described in revision hip also had a post-operative infection. Two additional patients re- arthroplasty literature. In the present report, modifications of this quired hardware removal for pain. One additional patient with a technique were applied to arthroplasty when successful fusion had a superficial infection requiring debridement press-fit implantation of the humeral component failed to provide and IV antibiotics. Conclusions: Arthrodesis in this patient popula- stability of the implant-bone construct with the advantage of avoid- tion has the potential for several complications but fusion can be ing the use of cement. In conclusion, the impaction autogenous achieved and satisfactory functional results can be achieved. The bone grafting technique provided good clinical results and stable results are substantially improved from pre-operative level and offer implant fixation without subsidence in 72 of 73 shoulders at an an advantage over resection in patients requesting shoulder level average follow-up of 56 months. function.

28 THE LONG-TERM RESULTS OF TREATING GLENOHU- 30 ANTERIOR INSTABILITY AS A COMPLICATION OF MERAL ARTHRITIS WITH A HEMIARTHROPLASTY: A 5-10 SHOULDER ARTHROPLASTY YEAR PROSPECTIVE OUTCOME STUDY Pascal Boileau, MD*, Philip Ahrens, MD, Sumant G. Krishnan, MD, Robert Stacy Tapscott, MD1,2,3*, Michael A. Wirth, MD1,2,3, Gilles Walch, MD, Dept of Orthopaedic Surgery, University of Charles A. Rockwood, Jr, MD1,2,3, 1The University of Texas Health Nice, Nice, France Science Center at San Antonio, San Antonio, TX, 2Carleton South- Introduction: The purpose of this study was to identify the worth, MS, 3Warsaw, IN etiologies and document the outcomes of treatment of anterior Introduction: A review of the literature reveals few studies with instability following shoulder arthroplasty. Material and Methods: long-term results of glenohumeral hemiarthroplasty. The purpose of Fifty-one patients (42 primary and 9 revision) consecutive patients this study is to evaluate our long-term results with this procedure. with anterior instability following glenohumeral replacement were Methods: From July 1991 to November 1997, 62 shoulders in 56 retrospectively evaluated at a minimum 2 year follow-up (mean 41 patients were treated at our institution with hemiarthroplasty. Forty- months). Diagnoses included osteoarthritis (29 cases), rheumatoid seven of these were available for long-term follow-up. There were arthritis (7 cases), and fracture (15 cases). Twenty-nine patients 26 males and 21 females with an average age of 64 years (range (57%) underwent total shoulder replacement and 22 (43%) under- 44-83). There were 28 shoulders with primary OA; 13 with sec- went hemiarthroplasty. Anterior glenohumeral prosthetic instability ondary OA; 4 with AVN; and 2 with CTA. The mean follow-up was developed within 6 weeks in 23 cases and after 6 weeks in 28 7 years (range 5-10 years). Patient directed measures of shoulder cases. Seven cases suffered trauma leading to instability; 21 cases pain and function were obtained using the validated Simple Shoul- developed atraumatic instability. Thirty-eight patients (75%) dem- J Shoulder Elbow Surg Abstracts 11

onstrated dislocation and 13 (25%) experienced painful subluxa- 32 IMPROVEMENTS IN UNCEMENTED NOTTINGHAM tion. Closed reduction/immobilization was performed in 16 cases TOTAL SHOULDER PROSTHESIS SURVIVAL WITH HYDROXY- and revision surgery in 35 cases. Results: Subscapularis rupture or APATITE COATING OF THE GLENOID BASE-PLATE incompetence was identified in 87% of cases. Technical errors in N. Rosenberg, MD*, W. A. Wallace, FRCSEd(Orth), prosthetic placement (including prosthetic malrotation) occurred in L. Neumann, FRCSEd, A. Modi, FRCS, I. J. Mersich, MD, Notting- 47% of cases. Associated complications included glenoid loosen- ham Shoulder & Elbow Unit, Nottingham, United Kingdom ing (24%), dissociation of the polyethylene from a metal-backed This paper presents improvement in the survivorship of the glenoid component (10%), infection (10%), and humeral fracture Nottingham Total Shoulder prosthesis with hydroxyapatite coating (4%). Fifty-five percent of patients were disappointed or dissatisfied of the glenoid base-plate. We present the long term survival rates with their prosthetic replacement. No shoulder was stable following for the uncemented Nottingham Total Shoulder prostheses with a specific focus on the improvement in the short-term survivorship conservative treatment. Prosthetic revision also gave disappointing following a change of the design of the glenoid component. One results with a 51% recurrence of instability. Conclusion: Anterior hundred and ninety three uncemented Nottingham Total Shoulder instability is a serious complication following shoulder arthroplasty. Replacement prostheses were inserted in one shoulder unit between Treatment of this difficult problem may still lead to poor functional 1989 - 1997 (Group 1). The main cause of failure of prostheses in results. Failure of subscapularis repair appears to be the principal this group of patients was loosening of the glenoid component. cause. Therefore a change in the prosthetic design was made by coating the glenoid component base-plate with hydroxyapatite. In 1998 and 1999 additional 34 of the re-designed prostheses were im- 31 INDICATIONS FOR AND OUTCOMES OF REVISION planted (Group 2). Survivorship analysis of both groups of patients TOTAL SHOULDER REPLACEMENT was performed. Failure of the prosthesis was determined by a revision operation, where components were removed or ex- Stephen Fealy, MD*, Joshua S. Dines, MD, Eric Strauss, MS, changed. The survival rates for the Group 1 was 83.2% and Russell F. Warren, MD, Edward V. Craig, MD, 75.2% at four and twelve years respectively. The four-year survival Answorth Allen, MD, David M. Dines, MD, Hospital for Special rate for Group 2 patients was 93.1%. The long-term survival rate of Surgery, New York, NY the initial design of the Nottingham Total Shoulder prostheses was Introduction: Multiple etiologies contribute to the need for revi- comparable with the previously reported data on cemented de- sion of humeral hemiarthroplasty (HA) and primary total shoulder signs. Furthermore this addition of hydroxyapatite coating of the replacement (TSR). Infection, glenoid arthrosis, trauma, aseptic glenoid base-plate has improved short-term survivorship. loosening, rotator cuff pathology, and soft tissue contractures about the shoulder have been cited as reasons for failure of the index procedure. The hypothesis of this study is that outcome following 33 FUNCTIONAL SYMPTOMATOLOGY FOLLOWING BI- revision shoulder arthroplasty (rTSR) can be can be predicted CEPS TENODESIS COMPARED TO BICEPS TENOTOMY OR based on the surgical indication for the revision procedure. Mate- RUPTURE rials and Methods: 77 shoulders (75pts; 40F, 35M; mean age Robin V. West, MD*, Melissa Koenig, MD, Robert J. Neviaser, MD, 62yo; 3 surgeons) that underwent either revision of HA to TSR or George Washington University, Washington, DC revision involving a primary TSR were retrospectively evaluated at Introduction: The purpose of this study was to compare the a minimum of 2 years post-op. All patients were evaluated with the functional symptomatology during supination activities following tenotomy and tenodesis of the long head of the biceps. Methods: ASES Shoulder Assessment Form, UCLA, and L’Insalata question- Fifty-six patients who underwent treatment at our institution by a naires. Outcome results were assessed for the entire group and single surgeon for biceps tendinitis from 1999-2001 were ran- separate cohorts within the group. 7 separate cohorts of patients domly selected for review. Fifteen patients were treated with a were identified within the 77 shoulders. Results: The seven cohorts tenotomy, 31 patients underwent a tenodesis, and ten patients had were as follows: 1) revision or resection of the glenoid component a biceps rupture that was treated non-operatively. The average age ϭ (N 20), 2) revision from HA to TSR for both trauma and osteoar- of the patients who had a tenotomy was 61 years, a tenodesis was ϭ throsis (N 16), 3) arthroscopy, debridement and rotator cuff re- 52 years, and a rupture was 62 years. The minimum follow-up was ϭ pair following TSR (N 14), 4) revision of the humeral head com- 12 months (range 12-48). Patient evaluation included an interview, ponent following TSR (Nϭ10), 5) instability, including component a physical examination, completion of an ASES shoulder survey, malpositioning, following TSR (Nϭ9), 6) trauma requiring ORIF and review of the medical record. Results: The average ASES without component revision (Nϭ5), 7) infection requiring resection shoulder index, pain, and ADL scores were 72/2/20 after a arthroplasty following TSR (Nϭ3). 36pts (47%) of the entire pop- tenotomy, 74/2.5/21 following a tenodesis, and 68/2.7/17 ulation were found to have good to excellent results; whereas 41pts after a rupture. Eighteen patients (72%) with a functional tenotomy (53%) had a fair to poor outcome based on questionnaire. The complained of pain and cramping in the biceps muscle when cohort requiring revision or resection of the glenoid component performing activities that required repetitive or resisted supination. alone faired the best of those defined. Not surprisingly, the worst Conclusion: Tenotomies are less surgically demanding and require outcomes could be expected following infection and subsequent a shorter rehabilitation. With the high incidence of cramping resection arthroplasty. A surprising finding in this study was that following a tenotomy, we recommend reserving this procedure for both the cohort that underwent modular humeral head revision and less active patients who have low functional demands. The criteria the cohort that underwent arthroscopy and rotator cuff repair for deciding between a tenodesis or a tenotomy lie more in the consistently yielded fair to poor results. Conclusion: Our findings functional demands of the patient rather than the age of the patient. indicate that there are distinct cohorts of patients who may require revision shoulder arthroplasty. Outcome may be generally pre- 34 SUPRASCAPULAR NERVE ENTRAPMENT AT THE SU- dicted based on the indication for revision. Patients who need an PRASCAPULAR NOTCH: RESULTS OF SURGICAL DECOM- isolated glenoid revision or resection consistently performed better PRESSION AND DESCRIPTION OF A NEW TECHNIQUE than those who required a soft tissue procedure or exchange of the Brian E. Koch, MD*, John F. Orwin, MD, John R. Turnbull, MD, humeral head. This knowledge may help in surgical planning and University of Wisconsin-Madison, Department of Orthopedics and decision making for the TSR patient with rotator cuff pathology or Rehabilitation one in whom a large humeral head is thought to be the cause of Suprascapular nerve (SSN) entrapment at the suprascapular persistent pain and disability. notch has been well described previously in the literature. Contro- 12 Abstracts J Shoulder Elbow Surg

versy exists as to whether this condition should be treated surgically access to the ligament for decompression than previously de- or non-operatively. Twenty-six patients with confirmed SSN entrap- scribed approaches. Many patients found complete relief of pain ment at the suprascapular notch were treated surgically at our with return to normal function. 70% of patients in our study would institution between 1990 and 2001. 25/26 patients were male have the surgery again. Reasonable surgical expectations and and all had nerve entrapment confirmed by EMG studies done at results should be communicated with each patient to improve our institution and read by the same neurologist who has extensive postoperative satisfaction. experience with diagnosis of this condition. Most patients had physical findings of shoulder pain, tenderness, atrophy, and weak- 35 PATIENT-ORIENTED FUNCTIONAL OUTCOME FOL- ness of the rotator cuff (primarily the supraspinatus and infraspina- LOWING REPAIR OF DISTAL BICEPS TENDON RUPTURES tus) prior to surgical release. All patients had failed a course of USING A SINGLE-INCISION TECHNIQUE non-operative treatment consisting of formal physical therapy. The Michael D. McKee, MD, FRCSC*, Rahim Hirji, BSc, average duration of symptoms prior to surgery was 25 months Emil H. Schemitsch, MD, FRCSC, Lisa Wild, BSCn, RN, (range 2-60). The operative procedure consisted of a diagnostic James P. Waddell, MD, FRCSC, St Michael’s Hospital, Toronto, arthroscopy with debridement of intra-articular abnormalities fol- Ontario, Canada lowed by open release of the suprascapular ligament to decom- Purpose: Using a patient-oriented outcome questionnaire in press the nerve. Various anterior and posterior surgical ap- addition to standard outcome measures, we sought to determine proaches for decompression have been previously described in the the outcome of patients who had repair of a complete rupture of literature. We describe a new approach based off the posterior their distal biceps tendon using a single-incision anterior technique border of the distal clavicle which allows easier access to the with suture anchors. Method: We identified 68 patients who were ligament for decompression and earlier shoulder range of motion treated by a single surgeon over a 8 year period with a diagnosis and postoperative rehabilitation. 15/26 patients were able to of rupture of the distal biceps tendon. Six patients had a partial return for interview, clinical examination, and assessment with rupture or declined surgical intervention, and eleven could not be objective scoring of shoulder function utilizing the American Shoul- located for final follow-up. Fifty-one patients (75%) participated in der and Elbow Surgeons and UCLA scores. Four additional pa- the study. All patients were male, and their mean age was 42 tients were able to be contacted by phone and returned a detailed years. Results: All repairs were performed using a single anterior questionnaire of their pain, functional capabilities, and satisfaction incision utilizing two suture plugs in the bicipital tuberosity, on an with the surgery. Therefore the study group consisted of 19 patients out-patient basis. There were few complications: one wound infec- with an average follow up of 4.1 years (range 0.75-11.2). Sub- tion, two transient paraesthesiae in the lateral cutaneous nerve jectively, 11/19 rated their result as good or excellent, 6 fair, and distribution and one posterior interosseous nerve palsy that re- 2 poor. 13/19 stated that they would go back and have the solved in six weeks (no re-operations). There were no re-ruptures, surgery again, 4 said that they would not, and 2 were not sure. All and no patient lost more than 5° of elbow flexion-extension or patients were able to return to their preoperative level of work and forearm rotation. All patients completed the DASH questionnaire. activities. For the group as a whole, both pain and function scores The mean DASH score was 8.2 Ϯ 11.6 (95% CI 5.2 to 11.9). In improved as did rotator cuff strength and atrophy. No patient was a published study, the mean DASH score of population controls worse after the surgery than before. There were no surgical com- was 6.2 (DASH User Manual). Discussion and Conclusions: The plications. In conclusion, SSN entrapment is challenging to diag- strengths of our study include the consistent surgical technique by a nose early and is often not recognized until months of ongoing single surgeon, the large number of patients, and the use of a weakness and atrophy has persisted. This creates a difficult situa- patient-oriented outcome measure. Distal biceps tendon rupture tion to try to improve patients’ strength, pain, and function. When repair using a single-incision technique with suture anchors was non-operative treatment fails to resolve symptoms, surgical decom- effective in restoring injured arms to normal, as measured by pression should be considered. Our newly described technique limb-specific patient-oriented measures, with minimal morbidity resulted in minimal postoperative morbidity and allows easier and a low complication rate.