The

Magnuson-Stack procedure for treatment of recurrent glenohumeral dislocations* LAWRENCE S.

MILLER,†‡ MD,

JOHN R.

ROBERT P. GOOD,∥ MD, AND ARTHUR J. STAERK,∥ MS, ATC

DONAHUE,§ MD,

From the †Jefferson Medical College of Thomas Jefferson University , Philadelphia , , Pennsylvania , Bryn Mawr Pennsylvania , and § Pottstown , ∥ Bryn Mawr Hospital Pennsylvania

cedure has a place in the treatment of recurrent anterior dislocations of the shoulder. Also, the Cybex II isolated

ABSTRACT The

Magnuson-Stack procedure

was

developed

in

joint testing system provides an accurate reproducible method of comparing the postoperative results of all patients treated for dislocation of the shoulder. Hopefully, from such data, information can be obtained which will be valuable in selecting the appropriate operative procedure for each individual’s needs as well as assisting in the development of postoperative rehabilitation programs designed to maximize shoulder function while minimizing disability time.

1943 as a treatment for recurrent anterior dislocation of the glenohumeral joint. Since that time, very little literature has been published on this procedure. The purpose of our study is to evaluate its efficacy in preventing further dislocations and, also, introduce the Cybex Isokinetic Dynamometer as a quantitative tool in scientifically evaluating total shoulder function. At the Bryn Mawr (Pennsylvania) Hospital from 1971 to 1978, 43 patients underwent a Magnuson-Stack procedure. Twenty-nine returned detailed questionnaires ; 26 of these patients returned for clinical and roentgenographic examination by one of the authors, and 18 had their shoulder motion and power quantitatively compared to their contralateral normal arms with the Cybex Isokinetic Dynamometer. Eighty-five percent of shoulders did not dislocate postoperatively; 90% considered their results satisfactory. By gross physical examination, no shoulder atro-

Recurrent anterior

glenohumeral dislocation has long been perplexing problem for the orthopaedic surgeon. The exact etiology of the disability has yet to be defined. Magnuson and Stack,’ DePalma et aI.,4,,sand Symeonides11 strongly supported laxity of the subscapularis muscle as the major anatomical defect. Bankart’, 10 described injury to the anterior glenoid and its labrum. Other theories range from laxity of the anterior capsule to bony lesions as described by Hill a

and Sachs.6

phy was noted. Strength seemed to equal the uninjured shoulder. Range of motion showed a 10° loss of external rotation as measured with a hand-held goniometer. However, by testing range of motion with the Cybex

Many procedures have been described

rectly or indirectly while minimizing detrimental effects on glenohumeral function. The Putti-Platt procedure attempts to directly shorten both the subscapularis tendon and the anterior joint capsule. The Bristow procedure acts indirectly by adding anterior bone stock to the glenoid rim, similar to that described by Eden-Hybbinette. More importantly, this acts as a dynamic buttress to the inferior and anterior joint capsule when the arm is in abduction and external rotation by its &dquo;sling&dquo; effect on the subscapularis muscle. Since its original description almost 40 years ago, the Magnuson-Stack9 procedure has been a popular surgical treatment for recurrent anterior glenohumeral dislocation.

25° lack of external rotation was noted. Force values measured at 60° of arc/sec and 180° of arc/sec showed an 18 and 10% deficit, respectively, in external rotation. The authors believe that the Magnuson-Stack pro-

system,

*

a

Presented at the Annual

Sports Medicine, Tan-Tar-A,

for correction of

recurrent anterior dislocation. Several, but not all, repairs attempt to correct the major anatomical defect either di-

meeting of the American Orthopaedic Society for Lake of the Ozarks, Missouri, July 12 through 15,

1982. t Address correspondence and reprint requests to: Lawrence S. Miller, MD, Jefferson Medical College of Thomas Jefferson University, 1015 Walnut Street, Philadelphia, PA 19107.

133

134

Previous clinical studies have dealt primarily with describing and recurrence rates. 2,4,7 Dynamic shoulder

patient profiles

function could only be graded subjectively by the patient. With the recent development of isolated joint testing devices, it is now possible to quantitatively measure shoulder function. The purpose of this study is to report our results with the Magnuson-Stack procedure using the Cybex II Isokinetic Dynamometer.

MATERIALS AND METHODS Between 1971 and 1978, 43 patients with 46 involved shoulder joints underwent a Magnuson-Stack repair. Thirty-five patients were male and eight were female. No patients with history of neurologic disorders or disease associated with ligamentous laxity were included in the study group. The average age was 19 years with a range of 9 to 32 years at the onset of their disability. The mean duration from initial dislocation to surgery was 5.7 years. The average number of dislocations preoperatively was seven and one-half. Patient followup ranged from 2 to 10 years with an average of 5.5 years. No patients were operated upon prior to reaching skeletal maturity. Twenty-nine patients answered an extensive questionnaire, and 26 patients from this group with 29 involved shoulder joints returned for physical and roentgenographic examinations. From this group, 18 patients with 22 involved shoulders underwent extensive testing with the Cybex II isokinetic joint testing system.3 The remaining patients were unable to afford the time involved with Cybex testing. In all patients, significant trauma caused the initial dislocation. Football was incriminated in 80% of patients. Dominant and nondominant extremities were injured in equal frequency. Fifty percent of shoulders were initially immobilized less than 3 weeks with 25% receiving no postreduction treatment. Only 32% of patients were immobilized a full 6 weeks. One patient had a previously unsuccessful surgical procedure. All surgery was performed by or under the supervision of one of the senior staff surgeons. In all patients the subscapularis muscle tendon was transferred laterally to the greater tuberosity and then distally approximately 1 cm as described by DePalma and Silberstein.4 A bone staple was then used for osseous fixation. Postoperatively, the involved extremity was placed in a standard shoulder immobilizer for 3 weeks. Active range of motion limited to 90° of abduction was initiated at 3 weeks and free use of the extremity was begun 8 weeks postoperatively. No rigidly designed rehabilitation program was followed. Cybex testing system was not available for monitoring rehabilitation progress at Bryn Mawr Hospital until 1979. The Cybex II isokinetic dynamometer was used to assess range of motion as well as strength and power of the shoulder. Strength is defined as maximum slow-speed torque capability at 60° of arc/sec, while power is defined as fastspeed torque capability at 180° of arc/sec. The patient is placed on the upper extremity testing unit in the supine position. With the appropriate accessories, the shoulder may

be positioned to minimize the effects of possible body movement and, therefore, eliminate forces from muscle groups other than those being tested. The axis of rotation which will allow a smooth, comfortable range of motion is selected. After several practice runs, the patient is asked to move the arm to the extreme of flexion and extension. This is performed without resistance and, therefore, registers on the print-out unit as maximum range of motion. Extension values were disregarded as the testing arm came into contact with the lower support of the testing table prior to full extension with several patients. Shoulder force is then determined by performing three consecutive full power runs at testing speeds of 60° (strength) and 180° arc/sec (power). After testing in flexion-extension, a similar program was used for testing for internal and external rotation with the shoulder at 90° of abduction and the elbow at 90° of flexion (Fig. 1). Range of motion is tested by nonresisted maximum internal and external rotation with careful attention to measurements being taken without trunk rotation. Power and strength testing was performed by evaluating three consecutive full power runs at both testing speeds. Testing at 0° of abduction was not performed due to difficulty limiting abduction at the shoulder and trunk rotation during

testing. Physical

examinations were carried out by three of the authors. Patients were observed for muscle atrophy, gross strength, and cosmetic appearance. Range of motion as measured by goniometry was evaluated in flexion, extension, and internal-external rotation with the glenohumeral joint at both 0 and 90° of abduction.

RESULTS Questionnaire The detailed questionnaire allowed development of an indepth patient data base as well as serving as a subjective indicator of patient satisfaction. Five of 29 shoulders dislocated postoperatively yielding a recurrence rate of 17%. The initial postoperative dislocation in two patients occurred while playing football and in a third while performing a 360° turn on water skis. The other two patients had recurrences with reaching back to put on clothing or combing their hair. The temporal relationship to recurrence and surgery was random in all five patients. Only one of these five patients has had greater than one recurrence, all with minimal trauma. However, his operation was performed 15 years after his primary dislocation with greater than 10 dislocations in the interim. All redislocations were treated with sling immobilization for 6 weeks; no reoperations were necessary.

Three patients were unable to sleep in the prone position due to diffuse shoulder irritation. Two patients with similar complaints had localized pain over the staple and with subsequent removal reported complete relief of pain. From the 29 completed questionnaires, a subjective evaluation of long-term function was possible. Sixteen patients (55%) were aware of less power in the affected arm. Twenty-

135

of shoulder discomfort associated with vigorous activity or weather changes were classified as pain. Therefore, the large number of patients with shoulder discomfort that we are reporting is in an attempt to be rigorous. No narcotic anal-

gesics were required by any patients. The overall subjective patient rating showed 82% good to excellent results, 11% fair, and 7% poor.

Physical examination Gross physical examination showed no significant shoulder girdle atrophy. The surgical scars were cosmetically accepted by all patients, including females. Deficits in range of motion were performed by subtracting the range of motion of the injured extremity from the uninjured extremity. Average values showed a deficit of 3.5° of internal rotation and 10° of external rotation in the affected shoulders. There was no significant difference in extension and flexion. Goniometry proved unsatisfactory, however, because of unreliability in establishing the zero degree reference point, especially with internal and external rotation (Table 1).

Cybex results Cybex testing supplies a graphic print-out which can be calibrated for degrees of arc as in range of motion testing, and force in Newton-meters as in &dquo;power&dquo; and &dquo;strength&dquo; testing (Fig. 2). Since a zero degree reference point was set at each testing session, range of motion is quickly measured from the graphs. Deficits are calculated as follows:

Figure 1. A, the Cybex upper body exercise and testing table (UBXT). B, patient positioned to measure range of motion, strength, and power of shoulder flexion and extension. C, method of testing range of motion, strength, and power of shoulder rotation.

(76%) experienced a decreased range of motion. Only (52%) returned to their preinjury level of competitive athletics, but 27 (93%) stated they were presently involved in recreational sports. Four patients (15%) believed their shoulder disability had altered their choice of vocation; half of these patients had suffered at least one recurrence. Only nine patients (32%) were completely pain-free with six patients (21 % ) requiring occasional analgesics. Any complaints

Force evaluation must be analyzed as a percent deficit of the uninvolved extremities. This raises two problems. First, the patient with bilateral involvement has no normal shoulder with which to compare results and, therefore, is excluded from this portion of the study. Second, compensation for the strength of a dominant extremity, be it the involved or uninvolved shoulder, must be determined. Since none of the patients were professional athletes that would have excessively overdeveloped dominant extremities, such as jai-lai, tennis, or baseball players, a 10% compensation was made for each dominant extremity force value. Percent deficit may then be calculated with the compensated figures as:

Final calculations yielded an overall range of motion deficit of 15° flexion, 4° internal rotation, and 25° external rotation.

two

fifteen

Range

TABLE 1 of motion deficit

136

dynamometer has been measured to be accurate within 2 degrees, our results indicate that the range of motion data contained in previous studies performed without such a testing device are possibly of limited accuracy and value. This report reflects the long-term results of MagnusonStack procedure as applied to a general orthopaedic practice. The patient population at the time of injury was mostly composed of college students involved in extracurricular varsity and intramural level athletics. No &dquo;professional&dquo; athletes were treated and therefore the motivation for continued rehabilitation was not present. The average age of patients at the time of evaluation was 25 years. All patients were employed in areas not associated with daily physical activity. Our results from the questionnaire report only 52% of patients returned to the preinjury level of activity while still in college. However, it is more important to note that 93% of patients were involved in recreational sports on a routine basis. Significant in this study is the development of a technique for evaluating shoulder function using an isolated joint testing device. Range of motion as well as dynamic power and strength can be precisely measured. With the present availability of fitness centers and the increasing use of Cybex systems in general orthopaedic practice, a more comprehensive rehabilitation program is presently being employed. The patient’s progress can now be followed with Cybex testing at standard postoperative intervals. We would expect this postoperative rehabilitation to yield better short-term results ; however, evaluation of a group so treated at 5 years postoperatively will be needed to determine the true value of intensive and often costly rehabilitation.

CONCLUSIONS Isolated joint

Figure 2. A, graphic print-out for measurement of internal (I) and external (E) range of motion as measured from zero reference line. B, graphic print-out showing maximum flexion (F) and maximum extension (E) at 180° arc/sec.

testing devices, such as the Cybex II isokinetic dynamometer, can be useful in providing quantitative results of shoulder function. In this patient group, deficits in range of motion, especially external rotation, are greater than previously reported by use of goniometry. Strength and power measurements showed significant deficits in all patients. No patients had greater strength in the involved

Strength, or slow-speed torque, showed a deficit of 4% in flexion, 5% in extension, 11% in internal rotation, and 18%

The present study shows the Magnuson-Stack procedure, when employed for treatment in a young age group, to have a recurrence rate of 17%. Most of these recurrences were

in external rotation.

due to significant trauma.

shoulder

Power, high-speed torque, showed no deficit in flexion, and deficits of 8% in extension, 14% in internal rotation, and 10% in external rotation. (See Table 2.)

as

tested.

or

TABLE 2

Cybex testing resultsa

DISCUSSION The range of motion results obtained with the Cybex unit were quite different from those obtained with the hand-held goniometer. The ability to yield reproducible measurements of range of motion with the goniometer from one examiner to another was poor. Cybex testing showed easily reproducible data with minimal physician error. Since the isokinetic

° As recommended by Cybex, dominance was correct by decreasing the dominant extremity force value by 10% prior to calculating

deficits.

.

137

Similar studies, when performed on patients receiving alternative procedures, will allow direct, quantitative comparison of many parameters and, therefore, be helpful in determining the most effective method of managing recurrent anterior glenohumeral dislocation with minimal loss of function.

REFERENCES 1.Bankart ASB: Recurrent or habitual dislocation of the shoulder joint. Br J Med 2: 1132-1133, 1923 2. Bryan RS, DiMichele JD, Ford GL, et al: Anterior recurrent dislocation of the shoulder. Clin Orthop 63: 177, 1972 3. CYBEX Isolated-Joint Testing and Exercise. A Handbook for Using CYBEX II and the U.B.X.T., 1980 4. DePalma AF, Silberstein CE: Results following a modified Magnuson procedure in recurrent dislocation of the shoulder. Surg Clin N Am 43:

1651-1653,1963 5. DePalma AF, Cooke AJ, Prabhakar MP: The role of the subscapularis in recurrent anterior dislocations of the shoulder. Clin Orthop 54: 35, 1967 6. Hill HA, Sachs MD: The grooved defect of the humeral head. A frequently unrecognized complication of dislocations of the shoulder joint. Radiology

35: 690, 1940 7. Karadimas J, Rentis G, Varouchas G: Repair of recurrent anterior dislocation of the shoulder using transfer of the subscapularis tendon. J Bone Joint Surg 62A: 1147, 1980 8. Lombardo SJ, Kerlan RK, Jobe FW, et al: The modified Briston procedure for recurrent dislocation of the shoulder. J Bone Joint Surg 58A: 256-261, 1976 9. Magnuson PB, Stack JK: Recurrent dislocation of the shoulder. JAMA 123:

889, 1943 10. Rowe CR, Patel D, Southmayd W: The Bankart Procedure. J Bone Joint Surg 60A: 1, 1978 11. Symeonides PP: The significance of the subscapularis muscle in the

pathogenesis of recurrent anterior dislocation of the shoulder. J Bone Joint Surg 54B: 476, 1972

DISCUSSION R. Leach, MD, Boston, Massachusetts: This is an interesting discussion of the Magnuson-Stack procedure. It would appear that the recurrent dislocation rate is high in this series of patients. Despite the fact that a number of the patients had major athletic trauma to cause another dislocation, there still was an unacceptably high incidence of recurrent dislocations. We expect patients to return to athletics after their operative procedures for recurrent dislocations and they must do better than they did in this particular series. However, negative results can make a very worthwhile paper and I believe this paper is worthwhile. It does show the results of the Magnuson-Stack versus other procedures. Of more importance is the method used to test motion and strength of various muscle groups. It is a sophisticated method and one which everybody may not be able to follow. They show that muscle strength comes back slowly and muscle power comes back very slowly. I think their testing method deserves commendation and should be carefully studied, particularly when we are doing retrospective studies on patients with shoulder disabilities. This demonstrates the problem of having muscles react at high speeds when patients are engaged in athletic competition. I commend the authors for an honest report concerning their patient population and their method of testing.