- Email: [email protected]
The arthroscopic treatment of posterior shoulder instability: two-year results of a multiple suture technique
The Arthroscopic Treatment of Posterior Shoulder Instability: Two-Year Results of a Multiple Suture Technique Louis F. Mclntyre, M.D., Richard B. Caspari, M.D., and Felix H. Savoie III, M.D.
Summary: This study was undertaken to review the results of an arthroscopic posterior capsular shift procedure. Twenty consecutive shoulders in 19 patients were treated with an arthroscopic posterior capsular shift for symptomatic posterior shoulder instability. Patients underwent the procedure if they exhibited a posterior Bankart lesion or had complaints of posterior instability and evidence of increased posterior joint laxity on physical examination and examination under anesthesia. Twelve of the 20 patients were injured during athletic activity. All surgeries were performed in an outpatient setting. Twelve of the 20 patients had posterior Bankart lesions and 10 had anterior Hill-Sachs lesions. The procedure entails releasing the posterior labroligamentous structures from the posterior glenoid and freshening the glenoid neck with a bur. A suture punch is used to place multiple absorbable monofilament stitches in the ligament complex. The stitches are brought through a supraclavicular portal and tied over the clavicle or scapular spine. All 20 shoulders were evaluated at an average of 31 months postoperatively with a minimum follow-up of 24 months. Based on the outcome scale described by Tibone and Bradley, the average postoperative score was 83 out of a possible 100, with 15 excellent, 2 good, 1 fair, and 3 poor results. There were two recurrent dislocations and three subluxations for an overall recurrence rate of 25%. All the recurrences occurred in patients with posterior Bankart lesions and four of the five had a voluntary component to their instability. There were no neurovascular complications or infections. Arthroscopic evaluation facilitated the diagnosis of posterior instability with the visualization of intra-articular pathology that is difficult to identify during open procedures. Although the majority of patients were able to return to vigorous activities, a recurrence rate of 25% is disturbing and consistent with recurrence rates for open procerdures. Key Words: Posterior shoulder instab i l i t y - Arthroscopy--Multiple suture.
houlder instability affects young, athletic people and can be quite debilitating in this population in regard to athletic endeavors. Although the treatment of traumatic anterior instability has become relatively straightforward, other instability patterns remain more
S
From Westchester Orthopaedic Associates, White Plains, New York (L.F.M.); Tuckahoe Orthopaedic Associates, Glen Allen, Virginia (R.B.C.); and Mississippi Sports Medicine and Orthopaedic Center, Jackson, Mississippi (F.H.S.), U.S.A. Address correspondence and reprint requests to Louis F. Mclntyre, M.D., Westchester Orthopaedic Associates, 222 Westchester Ave, Suite 101, White Plains, NY 10605, U,S.A. © 1997 by the Arthroscopy Association of North America 0749-8063/97/1304-148253.00/0
426
difficult to manage. This is especially true of posterior shoulder instability (PI), which is fairly uncommon, has a variable presentation, and an historically unpredictable response to conservative and surgical management. 1-4 Interest in this condition in recent years has increased awareness of the incidence, etiology, and successful treatment of this complex problem. B o y d and Sisk ~ "strongly indicated" operative intervention based on a history of recurrent, traumatic posterior dislocation. At an average follow-up o f 28 months (range, 11 to 48 months) no recurrences were reported. Hawkins et al. 2 reported on a retrospective survey of 50 shoulders in 35 patients with recurrent posterior
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 13, No 4 (August), 1997: pp 426-432
A R T H R O S C O P I C TREATMENT OF POSTERIOR INSTABILITY
instability in 1984. Patients were included in the study if a review of their hospital records indicated a diagnosis of posterior instability. Many different orthopaedic surgeons were involved in the care of these patients and, as such, no single criterion for the diagnosis of posterior instability was given. At an average followup of 7 years, the overall recurrence rate in patients treated surgically was 50%. In contrast, 20 of the 24 patients treated conservatively had few complaints and only 4 had to modify their sports activity. There is a growing body of literature documenting experience with the arthroscopic treatment of anterior and multidirectional shoulder instability but little has been written on the role of the arthroscope in the treatment of posterior shoulder instability. 5'6 This report is a summary of our results in treating a consecutive group of patients with posterior instability with an arthroscopic posterior capsular shift.
MATERIALS AND METHODS
427
An additional 1 patient fell while skiing. There were 4 college athletes, 6 high school athletes, and 1 professional. Five patients injured their shoulders while driving and 3 of these sustained dislocations while their hands were on the steering wheel with their shoulders in a flexed and internally rotated position. Two patients had had a fall that caused their symptoms and 1 patient had no history of trauma. Six patients had a voluntary component to their instability and could demonstrate posterior subluxation in the forward flexed and internally rotated position. A total of 8 patients had sustained a posterior dislocation. Twelve patients complained of pain and instability, 4 of instability alone, and the remaining 4 of pain alone. Four patients had complaints of pain in the opposite shoulder, 1 as the result of trauma. The average duration of symptoms was 16 months (range, 2 to 72 months). Six patients had had previous surgery including four arthroscopies, one open staple capsulorrhaphy, and one arthroscopic stapling.
Physical Examination
Twenty shoulders in 19 consecutive patients were treated surgically for PI between February 1991 and June 1992. All the patients were operated on by two of the authors (R.B.C. and F.H.S.) in an outpatient setting. Indications for surgery included a history of posterior dislocation, complaint of posterior instability or pain associated with increased posterior laxity on physical examination or evidence of a posterior Bankart lesion at arthroscopy. All of the patients had failed conservative therapy, which included monitored physical therapy or a home exercise program. All the patients were contacted an average of 31 months postoperatively (range, 24 to 44 months). Fifteen were examined by one of the authors and 4 were contacted by telephone. All were graded according to the Athletic Shoulder Outcome Rating Scale described by Tibone and Bradley 7 (Table 1). Preoperative data were gathered from office records.
Three patients exhibited generalized ligamentous laxity on preoperative physical examination. 8 All but 3 had full range of motion in the affected shoulder. These 3 had forward elevation limited to 100 °. Two of these also had limited internal rotation to the iliac crest only. Twelve shoulders had 3+ posterior laxity, 3 had 2+ laxity, 4 had 1 + laxity, and 1 had 0 laxity. Anterior laxity of 2+ was seen in 3 shoulders. A sulcus sign 9 of 2+ or greater was seen in 11 shoulders. Fifteen shoulders had a positive jerk test. 3'4 Fourteen had a positive posterior relocation test during an examination under anethesia. During this maneuver, the affected shoulder is placed in the forward flexed, internally rotated position while a posteriorly directed force is applied to the humerus. The shoulder is then gently extended, reducing the subluxated joint. A palpable clunk heralds the reduction.
Patient Data
Pathology at Surgery
There were 15 male and 4 female patients. One female patient had involement of both shoulders; one of traumatic etiology and one atraumatic. Their average age was 22 years with a range of 15 to 36 years. Twelve shoulders were injured during athletic activity. An additional 2 patients participated in athletics but were not injured during this activity. Five patients were hurt while playing football and sustained an injury to flexed, internally rotated shoulders, with two of these injuries resulting in dislocations. Two were baseball pitchers, 2 were bench pressing, and 2 were gymnasts.
Twelve shoulders had posterior Bankart lesions at arthroscopy. Posterior Bankart lesions were defined as detachment of the posterior labrum from the glenoid (Fig 1). These were associated with anterior HillSach's lesions in 8 shoulders. An additional 2 shoulders had anterior Hill-Sach's lesions without associated posterior Bankart lesions. Three shoulders without posterior Bankart lesions showed posterior labral fraying. The remaining 3 shoulders were judged to have had posterior capsular laxity. There were no rotator cuff tears.
L. F. MCINTYRE E T AL,
428
TABLE 1. Athletic Shoulder Outcome Rating Scale Name Age Dominant Hand (R) _ _ (L) _ _ (Ambidextrous) Date of Examination Surgeon Type of Sport Position Hayed Years Played Prior Injury Activity Level l) Professional (major league) 2) Professional (minor league) 3) College 4) High School 5) Recreational (full time) 6) Recreational (occasionally)
..
Sex__
Diagnosis 1) Anterior instability 2) Posterior instability 3) Multidirectionalinstability 4) Recurrent dislocations 5) Impingement syndrome 6) Acromioclavicularseparation 7) Acromioclaviculararthrosis 8) Rotator cuff repair (partial) 9) Rotator cuff repair (complete) 10) Biceps tendon rupture 11) Calcific tendinitis 12) Fracture
Subjective (90 points) I Pain No pain with competition Pain after competing only Pain while competing Pain preventing competing Pain with ADLs Pain at rest II Strength/Endurance No weakness, normal competition fatigue Weakness after competition, early competition fatigue Weakness during competition, abnormal competition fatigue Weakness or fatigue preventing competition Weakness or fatigue with ADLs Weakness or fatigue preventing ADLs III Stability No looseness during competition Recurrent subluxations while competing Dead-ann syndrome while competing Recurrent subluxafionsprevent competition Recurrent subluxations during ADLs Dislocation
Points 10 8 6 4 2 0 Points 10 8 6 4 2 0 10 8 6 4 2 0
IV Intensity Preinjury vs. postinjury hours of competition (100%) Preinjury vs. postinjury hours of competition (less than 75%) Preinjury vs, postinjury hours of competition (less than 50%) Preinjury vs, postinjury hours of competition (less than 25%) Preinjury and postinjury hours of ADLs (100%) Preinjury and postinjury hours of ADLs (less than 50%) V Performance At the same level, same proficiency At the same level, decreased proficiency At the same level, decreased proficiency, not acceptable athlete Decreased level with acceptable proficiency at that level Decreased level, unacceptable proficiency Cannot compete, had to switch sport Objective (10 points) Range of motion Normal external rotation 900-90° position; normal elevation < 5° loss of external rotation; normal elevation < 10° loss of external rotation; normal elevation < 15° loss of external rotation; normal elevation < 20° loss of external rotation; normal elevation > 20° loss of external rotation, or any loss of elevation
10 8 6 4 2 0 50 40 30 20 10 0 Points 10 8 6 4 2 0
Overall Results Excellent 90-100points Good 70-89 points Fair 50-69 points Poor <50 points
Adapted and reprinted with permission.7 Abbreviation: ADL, activity of daily living.
Surgical Technique The patient is placed in the lateral decubitus position and a thorough examination under anesthesia is performed. Special attention is given to the a m o u n t of anterior, posterior, and inferior translation of the joint. It is sometimes helpful to examine the patient before placement in the lateral position. This allows comparison examination of the opposite shoulder. Overhead traction is used and a posterior portal is used to gain access to the shoulder. All j o i n t structures are then examined. A n anterior portal facilitates exami-
nation of the posterior structures; the arthroscope should be placed in the anterior portal to view the posterior structures. Both the anterior and posterior labra must be assessed for the presence of Bankart lesions. The anterior and posterior aspects of the humeral head must be e x a m i n e d for the presence of HillSach's lesions. R e d u n d a n c y of the glenohumeral ligaments should he documented. The arthroscope is placed in the anterior portal and the posterior capsule is released from the glenoid with an arthroscopic blade through the posterior portal (Fig
ARTHROSCOPIC TREATMENT OF POSTERIOR INSTABILITY
429
FIG 1. Posterior ligaments are released form the glenoid to the 6 o'clock position.
l). The posterior ligament and capsule must be released completely from the glenoid so that it can be shifted superiorly. This is accomplished when the fibers of the infraspinatus muscle can be seen between the ligament and the glenoid. A bur is used to abrade the posterior glenoid neck. Multiple stitches (4 to 6) are placed into the posterior ligament down to the 6 o'clock position with the suture punch (Fig 2). The stitches are brought through the posterior portal and are tagged. A supraclavicular portal is then developed and a loop of nonabsorbable suture is brought into the joint with a grasping forceps through an operative cannula in this portal. Another grasping device is used to pull the loop of suture from the joint and through the posterior portal. The PDS stitches are then placed through this loop and pulled into the joint and up through the supraclavicular portal and tagged (Fig 3). Care is taken to place the
FIG 3. tal.
The stitches are pulled up through the supraclavicular por-
supraclavicular cannula so it enters the joint in the posterior gutter, so that the sutures, when tensioned will not cross the face of the glenoid. The arm is taken out of traction, placed in neutral rotation, and the stitches are separated into two bundles. One bundle is brought through a drill hole in the clavicle or the scapular spine, whichever is closer. The two bundles are then tied over bone. Postoperative rehabilitation is outlined in Table 2. RESULTS
FIG 2. punch.
Multiple sutures are placed in the ligament with a suture
Based on the outcome scale described by Tibone and Bradley, 7 the average postoperative score was 83 out of a possible 100, with 15 excellent, 2 good, 1 fair, and 3 poor results. Five shoulders experienced recurrent instability postoperatively, all between 12
430
L. F. MCINTYRE E T AL. TABLE 2. Rehabilitation After Arthroscopic Multidirectional Shoulder Stabilization Phase I Activity
Time
Abduction pillow. Encourage use of uninvolved joints and ice TID. Gentle pain-free isometric exercises. Begin mobilization out of pillow, passive ROM, mobilization and supine wand exercises (avoid aggressive anterior and inferior humeral glides, avoid external rotation initially). Active assistive ROM against gravity (wand, wall climb, pulley etc; start into flexion and gradually progressive to abduction). Active ROM against gravity, same considerations as active assistive ROM. Limited arc resistive exercises, avoid internal rotation beyond 0°, emphasis to posterior shoulder musculature; low reistance, high repetition program initially. Phase II Activity
Phase III Activity
Start at 3-4 weeks Start at 4 weeks Start at 4-5 weeks
Start at 5-6 weeks Start at 6 weeks Start at 8 weeks Time
Vigorous strengthening program. High load, low repetition as indicated and tolerated. Full ROM if indicated for function. This may require rather vigorous mobilization and strengthening. Functional progression.
Analysis o f Failures There were two recurrent dislocations and three subluxations in this series of patients. The average age of the patients who failed was 24 years (range, 17 to 34 years); 4 were male and 1 female. All were operated on during the first 12 months of the study period. All the shoulders had a history of trauma before the surgery; 4 were injured during athletic activity and 1 in a motor vehicle accident. Four complained of pain and instability while 1 complained of instability alone. Only 1 had a dislocation. Three of the patients had been operated on previously. All of the patients had posterior Bankart lesions and 3 had associated anterior Hill-Sachs lesions.
Start at 3 weeks
Time
Progressive ROM. Begin work into internal rotation with arm at side, as appreciation for end-feel is essential; full flexion should be obtained by 10-12 weeks. Increase rigor of strengthening and exercises, avoiding combined position of internal rotation and abduction. Address both muscle isolation and synergy. We utilize a lot of manual resistance initially, then progress to isokinetic and finally isotonic exercises emphasizing eccentric mode for throwers. Mobilize into external and internal rotation with arm abducted.
and 24 months after surgery for an overall recurrence rate of 25%. Seventeen of the patients were able to return to vigorous physical activity and 12 out of the 14 athletes returned to their sports at their preinjury level of competition. Two patients lost 5 ° of external rotation, 1 lost 10 ° of external rotation, and 1 lost 10 ° of internal rotation. One patient, rated as a failure, lost 20 ° of forward elevation. Only 1 o f the 15 patients with stable shoulders complained of pain after vigorous activity or competition. All of the 5 patients with recurrent instability had pain that prohibited them from vigorous physical activity. There were no infections or neurovascular complications.
0-3 weeks Start at 1-2 weeks
Start at 10-12 weeks Obtain by 14-16 weeks Start at 16 weeks
All of the patients with a recurrence had a voluntary component to their instability. One patient was involved in litigation. Only 2 of the patients were able to return to vigorous activity, 1 at an unacceptable level. Three o f the patients were rated as failures, 1 as fair, and 1 as good according to the outcome scale. As of this writing, 1 patient has undergone a repeat posterior stabilization procedure and another is contemplating one.
DISCUSSION Treatment of posterior shoulder instability remains problematic for the surgeon. Perhaps the most difficult problem resides in making the diagnosis itself. These patients present with varied, sometimes vague complaints and represent a variety of pathologies, all with the supposed unifying feature of increased, pathological posterior translation of the glenohumeral joint. As of yet, however, no single typical pathological change has been found to explain posterior instability of the shoulder. 3 While history and physical examination remain the most accurate method of categorizing these patients, our knowledge of this entity still lacks in identifying a universally accepted, accurate constellation of historical events and physical findings capable of identifying all patients with posterior instability. The
ARTHROSCOPIC TREATMENT OF POSTERIOR INSTABILITY
most straightforward patients are those who present with a history of frank posterior dislocation. Unfortunately, these patients are few and far between, and represent only 1% to 5 % 3'4 of all patients with shoulder instability. Those patients with recurrent posterior subluxation, voluntary posterior instability, and painful shoulders with increased posterior laxity remain more difficult to diagnose and treat. Complicating the situation further is the association of posterior instability with multidirectional instability. ~°'11 In addition, increased posterior laxity was found in 63 of 75 asymptomatic schoolchildren (average age, 15 years) in a study by Emery and Mullaji. ~2 All these factors lead to significant subjectivity in the diagnosis of posterior instability. A review of the indications for surgery, pathological findings, and results cited by authors reporting the outcomes of the surgical treatment of posterior instability highlights inconsistencies in the diagnosis and treatment of this disorder. Fronek et al.l~ documented the treatment of 24 patients with posterior instability. All of these patients complained of pain, instability, or both with their shoulders in a forward flexed and internally rotated position. None of these patients was reported as having had a posterior dislocation. All showed posterior subluxation of the shoulder with the joint in a provactive (forward flexion and internal rotation) position on physical examination. Sixty-three percent of the patients in the nonoperative group improved after a physical therapy program, and of the 6 who failed, 3 elected surgical treatment. Six of the 10 patients who improved with therapy were able to return to vigorous activities; 91% of the patients treated surgically experienced improvement. All of the operative patients returned to sports; 70% at a diminished level. One of these patients experienced recurrent instability but continued to participate in athletics. Tibone and Ting ~° treated a group of 20 athletes with recurrent posterior subluxation of the shoulder with a staple capsulorrhaphy. Eight of the patients had anterior instability in addition to their posterior component. All the patients had some type of pathology involving the posterior labrum or capsule at surgery, described as labral cracking or fraying with capsular redundancy. Posterior instability recurred in 6 of these patients and complications occured in 5, including a painful staple, postoperative adhesions, and ectopic bone formation. Sixteen of the patients were satisfied with their result but only one returned to throwing at an elite level after surgery. The authors concluded that staple capsulorrhaphy is not an acceptable treatment for posterior subluxation of the shoulder.
431
Recent experimental work has shed some light on the pathology of posterior shoulder dislocation and the role of the posterior and anterior glenohumeral ligaments in posterior instabilityJ 3'~4 A single, unifying mechanism explaining this condition remains, unfortunately, elusive. Weber and Caspari 15 created posterior dislocations in cadavera with the shoulders in forward flexion and internal rotation. They found posterior Bankart or capsular lesions in all specimens tested and concluded that posterior instability is the result of incompetence of the posterior inferior glenohumeral ligament and posterior capsule. Warren et al. 16 found increased posterior translation of the adducted, internally rotated shoulder with sectioning of the posterior portion of the inferior glenohumeral ligament complex. For actual posterior dislocation to occur, however, the anterior, superior capsule had to be incised from the 12 to the 3 o'clock position. This led to the "circle concept" of shoulder stability in which increased translation and instability is possible with ligament injuries to one side of the joint while frank dislocation requires injury to ligaments on both sides. In keeping with this concept, closure of the rotator interval has been advocated as a method to decrease pathological posterior translation of the glenohumeral joint. 17 These studies and the clinicopathologic findings of clinical studies indicate some role for the posterior glenohumeral ligaments in the pathomechanics of posterior instability. The static role of the posterior musculature ~8 has been studied and found to contribute to posterior stability of the joint but the dynamic role of the musculature remains largely unstudied and may contribute significantly to our understanding of posterior instability patterns. The patients treated in our series are similar to other published series in that most were male and were injured during athletic activity. Many injured their shoulders in a position of forward flexion and internal rotation. Only 12 of 19 actually complained of instability and most of them could demonstrate their instability on demand. Eleven of the shoulders had a positive sulcus test, indicative of some component of multidirectional instability. Our finding of posterior Bankart lesions in 12 of 20 shoulders with 8 anterior Hill-Sachs lesions is significantly higher than any previously published series and probably represents increased sensitivity of the arthroscope in detecting these lesions. Our patients are also similar in their high incidence of postoperative recurrence of posterior instability. Surprisingly, all of the patients with recurrences had posterior Bankart lesions at arthroscopy and our initial bias was that these patients would
432
L. F. MCINTYRE ET AL.
do well because of diffinitive pathological findings. These patients also had a high incidence of voluntadsm (80%) compared with an incidence of 6% for those whose surgery was successful. In this regard, our patients were similar to those reported by Tibone but very different from those reported by Fronek et al. who did not experience an increased recurrence rate in those patients with a voluntary component to their instability. Therefore, patients with voluntary posterior instability may be better treated by an open capsulorrhaphy with or without bone block. Educating these patients to avoid positions of provocation may also help decrease recurrences. The problem of posterior instability remains one of diagnosis. The cursory review of published surgical results above shows the diversity of this patient population. As such, it is clear that those who sustain a dislocation or subluxation of their glenohumeral joint in forward flexion and internal rotation represent a distinct clinical entity as do overhead athletes with repetitive trauma and increased posterior laxity. Those with voluntary instability and with atraumatic instability should also be considered separately. By separating these groups, treatment regimens may be better evaluated and compared. The arthroscope may help in this separation by allowing better visualization of intraarticular pathology of this condition. Finally, further study is necessary to elucidate the pathomechanics of all patterns of posterior instability. REFERENCES 1. Boyd HB, Sisk D. Recurrent posterior dislocation of the shoulder. J Bone Joint Surg Am 1972;54:779-786. 2. Hawkins RJ, Koppert G, Johnston G. Recurrent posterior instability of the shoulder. J Bone Joint Surg Am 1984;66:169-174.
3. Matsen FA, Thomas SC, Rockwood CA. Glenohumeral instability. In: Rockwood, CA, Matsen FA, eds. The Shoulder. Philadelphia: WB Saunders, 1990;526-622. 4. Pollack RG, Bigliani LU. Recurrent posterior shoulder instability. Clin Orthop 1993;291:85-96. 5. Eakin CL, Wolf EM. Arthroscopic management of posterior shoulder instability. Presented at the Fourteenth Annual Meeting of the Arthroscopy Association of North America, San Fransisco, CA, 1995. 6. Maki N. Posterior shoulder instability: Current status of arthroscopic treatment. Presented at the Thirteenth Annual Fall Course, Arthroscopy Association of North America, Palm Desert, CA, 1994. 7. Tibone JE, Bradley JP. Evaluation of treatment outcomes for the athletic shoulder. In: Matsen FA, Fu FH, Hawkins RJ, eds. The shoulder: A balance of mobility and stability. American Academy of Orthopedic Surgeons, Rosemont, IL, 1993;519530. 8. Marshall JL, Johansen N, Wickiewicz TL. Joint looseness: A function of the person and the joint. Med Sci Sports Exerc 1980; 12:189-194. 9. Neer CS, Foster CR. Inferior capsular shift for involuntary inferior and multidirectional instability of the shoulder. A preliminary report. J Bone Joint Surg Am 1980;62:897-908. 10. Tibone J, Ting A. Capsulorrhaphy with a staple for recurrent posterior subluxation of the shoulder. J Bone Joint Surg Am 1990;72:999-1002. 11. Fronek J, Warren RF, Bowen M. Posterior subluxation of the glenohumeral joint. J Bone and Joint Surg Am 1989;71:205215. 12. Emery JH, Mullaji AB. Glenohumeral joint instability in normal adolescents. J Bone Joint Surg Br 1991;73:406-408. 13. Schwartz E, Warren RF, O'Brien SJ, Fronek J. Posterior shoulder instability. Orthop Clin North Am 1987; 18:409-419. 14. Schwartz RE, O'Brien SJ, Warren RF, Torzilli PF. Capsular restraints to anterior-posterior motion of the shoulder. Orthop Trans 1988; 12:727. 15. Weber SC, Caspari RB. A biomechanical evaluation of the restraints to posterior shoulder dislocation. Arthroscopy 1989;5: 115-121. 16. Warren RF, Komblatt IB, Marchaud R. Static factors affecting posterior shoulder stability. Orthop Trans 1984;8:89. 17. Harryman DT, Sidles JA, Harris SL, Matsen FA. The role of the rotator interval in passive motion and stability of the shoulder. J Bone and Joint Surg Am 1992;74:53-66. 18. Ovensen J, Nielsen S. Posterior instability. A cadaver study. Acta Orthop Scand 1986;57:436-439.
Summary: This study was undertaken to review the results of an arthroscopic posterior capsular shift procedure. Twenty consecutive shoulders in 19 patients were treated with an arthroscopic posterior capsular shift for symptomatic posterior shoulder instability. Patients underwent the procedure if they exhibited a posterior Bankart lesion or had complaints of posterior instability and evidence of increased posterior joint laxity on physical examination and examination under anesthesia. Twelve of the 20 patients were injured during athletic activity. All surgeries were performed in an outpatient setting. Twelve of the 20 patients had posterior Bankart lesions and 10 had anterior Hill-Sachs lesions. The procedure entails releasing the posterior labroligamentous structures from the posterior glenoid and freshening the glenoid neck with a bur. A suture punch is used to place multiple absorbable monofilament stitches in the ligament complex. The stitches are brought through a supraclavicular portal and tied over the clavicle or scapular spine. All 20 shoulders were evaluated at an average of 31 months postoperatively with a minimum follow-up of 24 months. Based on the outcome scale described by Tibone and Bradley, the average postoperative score was 83 out of a possible 100, with 15 excellent, 2 good, 1 fair, and 3 poor results. There were two recurrent dislocations and three subluxations for an overall recurrence rate of 25%. All the recurrences occurred in patients with posterior Bankart lesions and four of the five had a voluntary component to their instability. There were no neurovascular complications or infections. Arthroscopic evaluation facilitated the diagnosis of posterior instability with the visualization of intra-articular pathology that is difficult to identify during open procedures. Although the majority of patients were able to return to vigorous activities, a recurrence rate of 25% is disturbing and consistent with recurrence rates for open procerdures. Key Words: Posterior shoulder instab i l i t y - Arthroscopy--Multiple suture.
houlder instability affects young, athletic people and can be quite debilitating in this population in regard to athletic endeavors. Although the treatment of traumatic anterior instability has become relatively straightforward, other instability patterns remain more
S
From Westchester Orthopaedic Associates, White Plains, New York (L.F.M.); Tuckahoe Orthopaedic Associates, Glen Allen, Virginia (R.B.C.); and Mississippi Sports Medicine and Orthopaedic Center, Jackson, Mississippi (F.H.S.), U.S.A. Address correspondence and reprint requests to Louis F. Mclntyre, M.D., Westchester Orthopaedic Associates, 222 Westchester Ave, Suite 101, White Plains, NY 10605, U,S.A. © 1997 by the Arthroscopy Association of North America 0749-8063/97/1304-148253.00/0
426
difficult to manage. This is especially true of posterior shoulder instability (PI), which is fairly uncommon, has a variable presentation, and an historically unpredictable response to conservative and surgical management. 1-4 Interest in this condition in recent years has increased awareness of the incidence, etiology, and successful treatment of this complex problem. B o y d and Sisk ~ "strongly indicated" operative intervention based on a history of recurrent, traumatic posterior dislocation. At an average follow-up o f 28 months (range, 11 to 48 months) no recurrences were reported. Hawkins et al. 2 reported on a retrospective survey of 50 shoulders in 35 patients with recurrent posterior
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 13, No 4 (August), 1997: pp 426-432
A R T H R O S C O P I C TREATMENT OF POSTERIOR INSTABILITY
instability in 1984. Patients were included in the study if a review of their hospital records indicated a diagnosis of posterior instability. Many different orthopaedic surgeons were involved in the care of these patients and, as such, no single criterion for the diagnosis of posterior instability was given. At an average followup of 7 years, the overall recurrence rate in patients treated surgically was 50%. In contrast, 20 of the 24 patients treated conservatively had few complaints and only 4 had to modify their sports activity. There is a growing body of literature documenting experience with the arthroscopic treatment of anterior and multidirectional shoulder instability but little has been written on the role of the arthroscope in the treatment of posterior shoulder instability. 5'6 This report is a summary of our results in treating a consecutive group of patients with posterior instability with an arthroscopic posterior capsular shift.
MATERIALS AND METHODS
427
An additional 1 patient fell while skiing. There were 4 college athletes, 6 high school athletes, and 1 professional. Five patients injured their shoulders while driving and 3 of these sustained dislocations while their hands were on the steering wheel with their shoulders in a flexed and internally rotated position. Two patients had had a fall that caused their symptoms and 1 patient had no history of trauma. Six patients had a voluntary component to their instability and could demonstrate posterior subluxation in the forward flexed and internally rotated position. A total of 8 patients had sustained a posterior dislocation. Twelve patients complained of pain and instability, 4 of instability alone, and the remaining 4 of pain alone. Four patients had complaints of pain in the opposite shoulder, 1 as the result of trauma. The average duration of symptoms was 16 months (range, 2 to 72 months). Six patients had had previous surgery including four arthroscopies, one open staple capsulorrhaphy, and one arthroscopic stapling.
Physical Examination
Twenty shoulders in 19 consecutive patients were treated surgically for PI between February 1991 and June 1992. All the patients were operated on by two of the authors (R.B.C. and F.H.S.) in an outpatient setting. Indications for surgery included a history of posterior dislocation, complaint of posterior instability or pain associated with increased posterior laxity on physical examination or evidence of a posterior Bankart lesion at arthroscopy. All of the patients had failed conservative therapy, which included monitored physical therapy or a home exercise program. All the patients were contacted an average of 31 months postoperatively (range, 24 to 44 months). Fifteen were examined by one of the authors and 4 were contacted by telephone. All were graded according to the Athletic Shoulder Outcome Rating Scale described by Tibone and Bradley 7 (Table 1). Preoperative data were gathered from office records.
Three patients exhibited generalized ligamentous laxity on preoperative physical examination. 8 All but 3 had full range of motion in the affected shoulder. These 3 had forward elevation limited to 100 °. Two of these also had limited internal rotation to the iliac crest only. Twelve shoulders had 3+ posterior laxity, 3 had 2+ laxity, 4 had 1 + laxity, and 1 had 0 laxity. Anterior laxity of 2+ was seen in 3 shoulders. A sulcus sign 9 of 2+ or greater was seen in 11 shoulders. Fifteen shoulders had a positive jerk test. 3'4 Fourteen had a positive posterior relocation test during an examination under anethesia. During this maneuver, the affected shoulder is placed in the forward flexed, internally rotated position while a posteriorly directed force is applied to the humerus. The shoulder is then gently extended, reducing the subluxated joint. A palpable clunk heralds the reduction.
Patient Data
Pathology at Surgery
There were 15 male and 4 female patients. One female patient had involement of both shoulders; one of traumatic etiology and one atraumatic. Their average age was 22 years with a range of 15 to 36 years. Twelve shoulders were injured during athletic activity. An additional 2 patients participated in athletics but were not injured during this activity. Five patients were hurt while playing football and sustained an injury to flexed, internally rotated shoulders, with two of these injuries resulting in dislocations. Two were baseball pitchers, 2 were bench pressing, and 2 were gymnasts.
Twelve shoulders had posterior Bankart lesions at arthroscopy. Posterior Bankart lesions were defined as detachment of the posterior labrum from the glenoid (Fig 1). These were associated with anterior HillSach's lesions in 8 shoulders. An additional 2 shoulders had anterior Hill-Sach's lesions without associated posterior Bankart lesions. Three shoulders without posterior Bankart lesions showed posterior labral fraying. The remaining 3 shoulders were judged to have had posterior capsular laxity. There were no rotator cuff tears.
L. F. MCINTYRE E T AL,
428
TABLE 1. Athletic Shoulder Outcome Rating Scale Name Age Dominant Hand (R) _ _ (L) _ _ (Ambidextrous) Date of Examination Surgeon Type of Sport Position Hayed Years Played Prior Injury Activity Level l) Professional (major league) 2) Professional (minor league) 3) College 4) High School 5) Recreational (full time) 6) Recreational (occasionally)
..
Sex__
Diagnosis 1) Anterior instability 2) Posterior instability 3) Multidirectionalinstability 4) Recurrent dislocations 5) Impingement syndrome 6) Acromioclavicularseparation 7) Acromioclaviculararthrosis 8) Rotator cuff repair (partial) 9) Rotator cuff repair (complete) 10) Biceps tendon rupture 11) Calcific tendinitis 12) Fracture
Subjective (90 points) I Pain No pain with competition Pain after competing only Pain while competing Pain preventing competing Pain with ADLs Pain at rest II Strength/Endurance No weakness, normal competition fatigue Weakness after competition, early competition fatigue Weakness during competition, abnormal competition fatigue Weakness or fatigue preventing competition Weakness or fatigue with ADLs Weakness or fatigue preventing ADLs III Stability No looseness during competition Recurrent subluxations while competing Dead-ann syndrome while competing Recurrent subluxafionsprevent competition Recurrent subluxations during ADLs Dislocation
Points 10 8 6 4 2 0 Points 10 8 6 4 2 0 10 8 6 4 2 0
IV Intensity Preinjury vs. postinjury hours of competition (100%) Preinjury vs. postinjury hours of competition (less than 75%) Preinjury vs, postinjury hours of competition (less than 50%) Preinjury vs, postinjury hours of competition (less than 25%) Preinjury and postinjury hours of ADLs (100%) Preinjury and postinjury hours of ADLs (less than 50%) V Performance At the same level, same proficiency At the same level, decreased proficiency At the same level, decreased proficiency, not acceptable athlete Decreased level with acceptable proficiency at that level Decreased level, unacceptable proficiency Cannot compete, had to switch sport Objective (10 points) Range of motion Normal external rotation 900-90° position; normal elevation < 5° loss of external rotation; normal elevation < 10° loss of external rotation; normal elevation < 15° loss of external rotation; normal elevation < 20° loss of external rotation; normal elevation > 20° loss of external rotation, or any loss of elevation
10 8 6 4 2 0 50 40 30 20 10 0 Points 10 8 6 4 2 0
Overall Results Excellent 90-100points Good 70-89 points Fair 50-69 points Poor <50 points
Adapted and reprinted with permission.7 Abbreviation: ADL, activity of daily living.
Surgical Technique The patient is placed in the lateral decubitus position and a thorough examination under anesthesia is performed. Special attention is given to the a m o u n t of anterior, posterior, and inferior translation of the joint. It is sometimes helpful to examine the patient before placement in the lateral position. This allows comparison examination of the opposite shoulder. Overhead traction is used and a posterior portal is used to gain access to the shoulder. All j o i n t structures are then examined. A n anterior portal facilitates exami-
nation of the posterior structures; the arthroscope should be placed in the anterior portal to view the posterior structures. Both the anterior and posterior labra must be assessed for the presence of Bankart lesions. The anterior and posterior aspects of the humeral head must be e x a m i n e d for the presence of HillSach's lesions. R e d u n d a n c y of the glenohumeral ligaments should he documented. The arthroscope is placed in the anterior portal and the posterior capsule is released from the glenoid with an arthroscopic blade through the posterior portal (Fig
ARTHROSCOPIC TREATMENT OF POSTERIOR INSTABILITY
429
FIG 1. Posterior ligaments are released form the glenoid to the 6 o'clock position.
l). The posterior ligament and capsule must be released completely from the glenoid so that it can be shifted superiorly. This is accomplished when the fibers of the infraspinatus muscle can be seen between the ligament and the glenoid. A bur is used to abrade the posterior glenoid neck. Multiple stitches (4 to 6) are placed into the posterior ligament down to the 6 o'clock position with the suture punch (Fig 2). The stitches are brought through the posterior portal and are tagged. A supraclavicular portal is then developed and a loop of nonabsorbable suture is brought into the joint with a grasping forceps through an operative cannula in this portal. Another grasping device is used to pull the loop of suture from the joint and through the posterior portal. The PDS stitches are then placed through this loop and pulled into the joint and up through the supraclavicular portal and tagged (Fig 3). Care is taken to place the
FIG 3. tal.
The stitches are pulled up through the supraclavicular por-
supraclavicular cannula so it enters the joint in the posterior gutter, so that the sutures, when tensioned will not cross the face of the glenoid. The arm is taken out of traction, placed in neutral rotation, and the stitches are separated into two bundles. One bundle is brought through a drill hole in the clavicle or the scapular spine, whichever is closer. The two bundles are then tied over bone. Postoperative rehabilitation is outlined in Table 2. RESULTS
FIG 2. punch.
Multiple sutures are placed in the ligament with a suture
Based on the outcome scale described by Tibone and Bradley, 7 the average postoperative score was 83 out of a possible 100, with 15 excellent, 2 good, 1 fair, and 3 poor results. Five shoulders experienced recurrent instability postoperatively, all between 12
430
L. F. MCINTYRE E T AL. TABLE 2. Rehabilitation After Arthroscopic Multidirectional Shoulder Stabilization Phase I Activity
Time
Abduction pillow. Encourage use of uninvolved joints and ice TID. Gentle pain-free isometric exercises. Begin mobilization out of pillow, passive ROM, mobilization and supine wand exercises (avoid aggressive anterior and inferior humeral glides, avoid external rotation initially). Active assistive ROM against gravity (wand, wall climb, pulley etc; start into flexion and gradually progressive to abduction). Active ROM against gravity, same considerations as active assistive ROM. Limited arc resistive exercises, avoid internal rotation beyond 0°, emphasis to posterior shoulder musculature; low reistance, high repetition program initially. Phase II Activity
Phase III Activity
Start at 3-4 weeks Start at 4 weeks Start at 4-5 weeks
Start at 5-6 weeks Start at 6 weeks Start at 8 weeks Time
Vigorous strengthening program. High load, low repetition as indicated and tolerated. Full ROM if indicated for function. This may require rather vigorous mobilization and strengthening. Functional progression.
Analysis o f Failures There were two recurrent dislocations and three subluxations in this series of patients. The average age of the patients who failed was 24 years (range, 17 to 34 years); 4 were male and 1 female. All were operated on during the first 12 months of the study period. All the shoulders had a history of trauma before the surgery; 4 were injured during athletic activity and 1 in a motor vehicle accident. Four complained of pain and instability while 1 complained of instability alone. Only 1 had a dislocation. Three of the patients had been operated on previously. All of the patients had posterior Bankart lesions and 3 had associated anterior Hill-Sachs lesions.
Start at 3 weeks
Time
Progressive ROM. Begin work into internal rotation with arm at side, as appreciation for end-feel is essential; full flexion should be obtained by 10-12 weeks. Increase rigor of strengthening and exercises, avoiding combined position of internal rotation and abduction. Address both muscle isolation and synergy. We utilize a lot of manual resistance initially, then progress to isokinetic and finally isotonic exercises emphasizing eccentric mode for throwers. Mobilize into external and internal rotation with arm abducted.
and 24 months after surgery for an overall recurrence rate of 25%. Seventeen of the patients were able to return to vigorous physical activity and 12 out of the 14 athletes returned to their sports at their preinjury level of competition. Two patients lost 5 ° of external rotation, 1 lost 10 ° of external rotation, and 1 lost 10 ° of internal rotation. One patient, rated as a failure, lost 20 ° of forward elevation. Only 1 o f the 15 patients with stable shoulders complained of pain after vigorous activity or competition. All of the 5 patients with recurrent instability had pain that prohibited them from vigorous physical activity. There were no infections or neurovascular complications.
0-3 weeks Start at 1-2 weeks
Start at 10-12 weeks Obtain by 14-16 weeks Start at 16 weeks
All of the patients with a recurrence had a voluntary component to their instability. One patient was involved in litigation. Only 2 of the patients were able to return to vigorous activity, 1 at an unacceptable level. Three o f the patients were rated as failures, 1 as fair, and 1 as good according to the outcome scale. As of this writing, 1 patient has undergone a repeat posterior stabilization procedure and another is contemplating one.
DISCUSSION Treatment of posterior shoulder instability remains problematic for the surgeon. Perhaps the most difficult problem resides in making the diagnosis itself. These patients present with varied, sometimes vague complaints and represent a variety of pathologies, all with the supposed unifying feature of increased, pathological posterior translation of the glenohumeral joint. As of yet, however, no single typical pathological change has been found to explain posterior instability of the shoulder. 3 While history and physical examination remain the most accurate method of categorizing these patients, our knowledge of this entity still lacks in identifying a universally accepted, accurate constellation of historical events and physical findings capable of identifying all patients with posterior instability. The
ARTHROSCOPIC TREATMENT OF POSTERIOR INSTABILITY
most straightforward patients are those who present with a history of frank posterior dislocation. Unfortunately, these patients are few and far between, and represent only 1% to 5 % 3'4 of all patients with shoulder instability. Those patients with recurrent posterior subluxation, voluntary posterior instability, and painful shoulders with increased posterior laxity remain more difficult to diagnose and treat. Complicating the situation further is the association of posterior instability with multidirectional instability. ~°'11 In addition, increased posterior laxity was found in 63 of 75 asymptomatic schoolchildren (average age, 15 years) in a study by Emery and Mullaji. ~2 All these factors lead to significant subjectivity in the diagnosis of posterior instability. A review of the indications for surgery, pathological findings, and results cited by authors reporting the outcomes of the surgical treatment of posterior instability highlights inconsistencies in the diagnosis and treatment of this disorder. Fronek et al.l~ documented the treatment of 24 patients with posterior instability. All of these patients complained of pain, instability, or both with their shoulders in a forward flexed and internally rotated position. None of these patients was reported as having had a posterior dislocation. All showed posterior subluxation of the shoulder with the joint in a provactive (forward flexion and internal rotation) position on physical examination. Sixty-three percent of the patients in the nonoperative group improved after a physical therapy program, and of the 6 who failed, 3 elected surgical treatment. Six of the 10 patients who improved with therapy were able to return to vigorous activities; 91% of the patients treated surgically experienced improvement. All of the operative patients returned to sports; 70% at a diminished level. One of these patients experienced recurrent instability but continued to participate in athletics. Tibone and Ting ~° treated a group of 20 athletes with recurrent posterior subluxation of the shoulder with a staple capsulorrhaphy. Eight of the patients had anterior instability in addition to their posterior component. All the patients had some type of pathology involving the posterior labrum or capsule at surgery, described as labral cracking or fraying with capsular redundancy. Posterior instability recurred in 6 of these patients and complications occured in 5, including a painful staple, postoperative adhesions, and ectopic bone formation. Sixteen of the patients were satisfied with their result but only one returned to throwing at an elite level after surgery. The authors concluded that staple capsulorrhaphy is not an acceptable treatment for posterior subluxation of the shoulder.
431
Recent experimental work has shed some light on the pathology of posterior shoulder dislocation and the role of the posterior and anterior glenohumeral ligaments in posterior instabilityJ 3'~4 A single, unifying mechanism explaining this condition remains, unfortunately, elusive. Weber and Caspari 15 created posterior dislocations in cadavera with the shoulders in forward flexion and internal rotation. They found posterior Bankart or capsular lesions in all specimens tested and concluded that posterior instability is the result of incompetence of the posterior inferior glenohumeral ligament and posterior capsule. Warren et al. 16 found increased posterior translation of the adducted, internally rotated shoulder with sectioning of the posterior portion of the inferior glenohumeral ligament complex. For actual posterior dislocation to occur, however, the anterior, superior capsule had to be incised from the 12 to the 3 o'clock position. This led to the "circle concept" of shoulder stability in which increased translation and instability is possible with ligament injuries to one side of the joint while frank dislocation requires injury to ligaments on both sides. In keeping with this concept, closure of the rotator interval has been advocated as a method to decrease pathological posterior translation of the glenohumeral joint. 17 These studies and the clinicopathologic findings of clinical studies indicate some role for the posterior glenohumeral ligaments in the pathomechanics of posterior instability. The static role of the posterior musculature ~8 has been studied and found to contribute to posterior stability of the joint but the dynamic role of the musculature remains largely unstudied and may contribute significantly to our understanding of posterior instability patterns. The patients treated in our series are similar to other published series in that most were male and were injured during athletic activity. Many injured their shoulders in a position of forward flexion and internal rotation. Only 12 of 19 actually complained of instability and most of them could demonstrate their instability on demand. Eleven of the shoulders had a positive sulcus test, indicative of some component of multidirectional instability. Our finding of posterior Bankart lesions in 12 of 20 shoulders with 8 anterior Hill-Sachs lesions is significantly higher than any previously published series and probably represents increased sensitivity of the arthroscope in detecting these lesions. Our patients are also similar in their high incidence of postoperative recurrence of posterior instability. Surprisingly, all of the patients with recurrences had posterior Bankart lesions at arthroscopy and our initial bias was that these patients would
432
L. F. MCINTYRE ET AL.
do well because of diffinitive pathological findings. These patients also had a high incidence of voluntadsm (80%) compared with an incidence of 6% for those whose surgery was successful. In this regard, our patients were similar to those reported by Tibone but very different from those reported by Fronek et al. who did not experience an increased recurrence rate in those patients with a voluntary component to their instability. Therefore, patients with voluntary posterior instability may be better treated by an open capsulorrhaphy with or without bone block. Educating these patients to avoid positions of provocation may also help decrease recurrences. The problem of posterior instability remains one of diagnosis. The cursory review of published surgical results above shows the diversity of this patient population. As such, it is clear that those who sustain a dislocation or subluxation of their glenohumeral joint in forward flexion and internal rotation represent a distinct clinical entity as do overhead athletes with repetitive trauma and increased posterior laxity. Those with voluntary instability and with atraumatic instability should also be considered separately. By separating these groups, treatment regimens may be better evaluated and compared. The arthroscope may help in this separation by allowing better visualization of intraarticular pathology of this condition. Finally, further study is necessary to elucidate the pathomechanics of all patterns of posterior instability. REFERENCES 1. Boyd HB, Sisk D. Recurrent posterior dislocation of the shoulder. J Bone Joint Surg Am 1972;54:779-786. 2. Hawkins RJ, Koppert G, Johnston G. Recurrent posterior instability of the shoulder. J Bone Joint Surg Am 1984;66:169-174.
3. Matsen FA, Thomas SC, Rockwood CA. Glenohumeral instability. In: Rockwood, CA, Matsen FA, eds. The Shoulder. Philadelphia: WB Saunders, 1990;526-622. 4. Pollack RG, Bigliani LU. Recurrent posterior shoulder instability. Clin Orthop 1993;291:85-96. 5. Eakin CL, Wolf EM. Arthroscopic management of posterior shoulder instability. Presented at the Fourteenth Annual Meeting of the Arthroscopy Association of North America, San Fransisco, CA, 1995. 6. Maki N. Posterior shoulder instability: Current status of arthroscopic treatment. Presented at the Thirteenth Annual Fall Course, Arthroscopy Association of North America, Palm Desert, CA, 1994. 7. Tibone JE, Bradley JP. Evaluation of treatment outcomes for the athletic shoulder. In: Matsen FA, Fu FH, Hawkins RJ, eds. The shoulder: A balance of mobility and stability. American Academy of Orthopedic Surgeons, Rosemont, IL, 1993;519530. 8. Marshall JL, Johansen N, Wickiewicz TL. Joint looseness: A function of the person and the joint. Med Sci Sports Exerc 1980; 12:189-194. 9. Neer CS, Foster CR. Inferior capsular shift for involuntary inferior and multidirectional instability of the shoulder. A preliminary report. J Bone Joint Surg Am 1980;62:897-908. 10. Tibone J, Ting A. Capsulorrhaphy with a staple for recurrent posterior subluxation of the shoulder. J Bone Joint Surg Am 1990;72:999-1002. 11. Fronek J, Warren RF, Bowen M. Posterior subluxation of the glenohumeral joint. J Bone and Joint Surg Am 1989;71:205215. 12. Emery JH, Mullaji AB. Glenohumeral joint instability in normal adolescents. J Bone Joint Surg Br 1991;73:406-408. 13. Schwartz E, Warren RF, O'Brien SJ, Fronek J. Posterior shoulder instability. Orthop Clin North Am 1987; 18:409-419. 14. Schwartz RE, O'Brien SJ, Warren RF, Torzilli PF. Capsular restraints to anterior-posterior motion of the shoulder. Orthop Trans 1988; 12:727. 15. Weber SC, Caspari RB. A biomechanical evaluation of the restraints to posterior shoulder dislocation. Arthroscopy 1989;5: 115-121. 16. Warren RF, Komblatt IB, Marchaud R. Static factors affecting posterior shoulder stability. Orthop Trans 1984;8:89. 17. Harryman DT, Sidles JA, Harris SL, Matsen FA. The role of the rotator interval in passive motion and stability of the shoulder. J Bone and Joint Surg Am 1992;74:53-66. 18. Ovensen J, Nielsen S. Posterior instability. A cadaver study. Acta Orthop Scand 1986;57:436-439.