Nonsurgical management of the symptomatic shoulder with MDI is the initial step in treatment, consisting of extensive patient education and physical therapy. It has been the prevailing thought that most patients could be treated successfully by patient education, rotator cuff and shoulder girdle strengthening exercises, and activity modification. In 1956, Rowe (3) reported that a majority of patients with atraumatic shoulder instability did well with physical therapy during short-term follow-up. In 1992, Burkhead and Rockwood (4) reported good or excellent results in 88% of patients treated for MDI with physiciandirected rehabilitation at an average follow-up of 48 months. However, in their report, it is not possible to determine any specific details about the MDI patients such as age, specific symptoms, duration of symptoms, and magnitude of shoulder laxity. More recently, in a much more defined population, Misamore and co-workers (5) have shown that only 17 of 57 patients had satisfactory outcomes from the nonsurgical management based on stability and Rowe scores at 7- to 10-year follow-up.

Neer and Foster (6) were the first to report on the inferior capsular shift for MDI in 1980. They reported on 40 shoulders in 36 patients. In their group 26 of 29 patients reported a moderate or severe traumatic event with their initial dislocation. Five of their patients that were included had Bankart lesions. The patient population varied from age 15 to 55 years, and only 17 of the patients were described as having generalized ligamentous laxity. Some of these patients appeared to fit the definition of MDI, but not all of them. They reported only one unsatisfactory result in 32 shoulders that were followed up for more than 1 year. Nine patients in their group returned to competitive sports. Three patients suffered axillary nerve injury.

In 1991, Altcheck and colleagues (7) reported on the surgical treatment of 40 patients for MDI. In this group 39 of 40 patients suffered a traumatic onset with 36 patients suffering a frank dislocation. Thirty-eight of 40 shoulders had a Bankart lesion and 24 had a Hill-Sachs lesion. This group of patients does not seem to fit the definition of MDI. Thirty-six of the 40 patients with at least 2-year follow-up had relief of instability. They also reported that 33 of 40 patients had a full return to sports; although they did not describe or quantify their activity level.

Cooper and Brems (8) reported on 43 shoulders in 38 patients over 6 years. Thirty-two of 43 shoulders had no history of trauma of a magnitude normally associated with dislocation, although for 18 the onset of painful instability could be traced to relatively minor trauma. Twenty-eight of 36 patients had laxity in the opposite shoulder and 29 patients had generalized ligamentous laxity. In this group only seven patients were found to have Bankart lesions. This group of patients more closely represents an MDI population. Thirty-nine shoulders (91%) functioned well with no recurrence of instability. Of these patients, 29 shoulders were thought to be stable enough to allow the patient to return to the same or a more physically demanding job. Sixteen shoulders were stable enough for the patients to return to the same sport and play at the same or a reduced level of activity. However, none of these patients had been elite athletes.

Lebar and Alexander (9) reported on 10 patients surgically treated for MDI over 10 years. In their group, all patients had inferior laxity but only three had generalized ligamentous laxity. Four patients in this group had an initial traumatic onset. At surgery, three Bankart and two Hill-Sachs lesions were found. Also six patients had undergone previous shoulder surgery with four having procedures to address instability. From this information, it is difficult to determine what type of instability these patients had. Eight of 10 patients at 1-year follow-up were happy with their surgery, reporting an overall improvement. One half of the patients received a medical discharge from the military with a disability. Additionally, of the five patients who returned to active duty, only two patients could function without restrictions.

Pollock and co-workers (10) reported on a longer term follow-up for the capsular shift procedure as a treatment of MDI. This study included 52 shoulders in 49 patients. The follow-up ranged from 24 to 132 months, over an average of 61 months. Average age of the patients was 23 years (range 16 to 42). Twenty-five shoulders had a history of a major episode of trauma with their onset of instability and there were 10 Bankart lesions and two anterior glenoid rim fractures identified at the time of surgery. Forty-six shoulders (96%) had good/excellent results. Forty-seven shoulders (96%) remained stable at their most recent follow-up. Recurring anterior instability developed in two shoulders within the first year after surgery. Thirty-one shoulders (86%) were able to return to participation in their premorbid sports activities; however, only 25 shoulders (69%) were able to return at the same level. With longer term follow-up and the activity level of this group of patients, there were no late failures demonstrating that an open capsular shift is a durable procedure.

Bak and co-workers (11) attempted to look at a population of strictly athletes surgically treated with capsular shift for MDI. Their study included 26 shoulders in 25 athletes evaluated at an average follow-up of 54 months (range 25 to 113). Their ages ranged from 16 to 35 years with a median age of 23. In this group, 16 shoulders had symptoms that were initiated by trauma, with one patient having an incomplete Bankart lesion at the time of surgery. Twenty-one patients (84%) returned to their preinjury sport at the same level. Of 21 patients involved in sports utilizing overhead motions, 16 (76%) returned to their previous sport, but there was no comparison made to their preoperative ability. Five shoulders were unstable at follow-up and 2 (8%) were considered failures.

In summary, it would appear that satisfactory results were obtained in approximately 90% of cases (Neer and Foster (6), 97%; Altchek and colleagues (7), 90%; Cooper and Brems (8), 91%; Lebar and Alexander (9), 80%; Pollock and co-workers (10), 96%; Bak and colleagues (11), 84%). However, even with this high percentage of successful results, it does still not seem that many shoulders were functioning well enough to allow a return to stressful activity. Although these reports of surgical treatment of MDI with an inferior capsular shift describe satisfactory results in approximately 90% of patients, the results are difficult to interpret because of varied patient populations. Also, the criteria for satisfactory results are not clearly defined or consistent. Because of this, the limited reports on the surgical treatment of MDI must be viewed cautiously and critically.

The literature on arthroscopic treatment of MDI is limited with only short-term results. The arthroscopic treatment of MDI includes two different techniques consisting of capsular plication with sutures and thermal capsulorrhaphy.

Duncan and Savoie (12) published a report on arthroscopic treatment utilizing a transglenoid suture technique. Ten consecutive patients were evaluated with 1- to 3-year follow-up. Although MDI had been diagnosed by the authors, 9 of 10 patients had previous dislocations and four Bankart lesions were identified at the time of surgery, suggesting they were not typical MDI. They reported an average Bankart score of 90 (range 75 to 95) and all patients resumed activities of daily living without discomfort. Four patients were able to return to sporting activities at their preinjury levels.

Treacy and co-workers (13) reported on 25 patients with an average follow-up of 60 months (range 36 to 80) using the same technique. The average age of this patient group was 26.4 years (range 15 to 39). Eleven patients had sustained a prior dislocation, 14 patients had a Bankart lesion identified at the time of surgery, and five patients had evidence of generalized ligamentous laxity. The average Bankart score was 95 (range 50 to 100) and 21 patients (88%) had a satisfactory result according to the Neer system. Three of the patients had episodes of recurrent instability and required further surgical treatment.

McIntyre and colleagues (14) also described a transglenoid suture technique but placed sutures anteriorly as well as posterior to the glenoid. This series consisted of 19 consecutive shoulders treated for MDI. Their average age was 23 years (range 15 to 52). Fourteen of the 19 patients were initially injured during athletic activity. At the time of surgery, seven patients were identified who had both an anterior and posterior Bankart lesion, two had an anterior Bankart lesion, and one had posterior labral fraying. Six patients were identified to have an anterior or posterior Hill-Sachs lesion. At an average follow-up of 34 months postoperatively, the average outcome score was 91 of 100. There were 13 excellent, five good, and one fair result. One patient had repeated episodes of instability that required a repeat arthroscopic capsular shift.

Gartsman and colleagues (15) prospectively evaluated 47 patients treated with an arthroscopic capsular shift for MDI. Their technique consisted of an anatomic repair of the labrum and a capsular placation of the glenohumeral ligaments to the labrum. No transosseus sutures were used. The average age of the patients was 30 years (range 15 to 56). Twenty-seven patients had a history of recurrent subluxation and 20 patients had recurrent dislocation before surgery. Instability after a single traumatic event developed in 21 patients and instability without trauma developed in 26. At the time of surgery, there were 10 Bankart and five Hill-Sachs lesions identified. At an average follow-up of 35 months (range 26 to 67), 44 of 47 shoulders were rated as excellent according to the Rowe score. One patient had episodes of recurrent instability and required a second operative procedure. Twenty-two of 26 patients were able to return to their desired level of athletic participation.

Lyons and colleagues (16) reported on a group of 27 shoulders in 26 patients treated with a combination of laser-assisted capsulorrhaphy and rotator interval suture plication. The average age of the patients was 25 years (range 16 to 46). Four of the patients developed instability from a single traumatic event. Patients who were found to have labral pathology requiring repair were excluded from the study. The follow-up averaged 27 months (range 24 to 35). Twenty-six of 27 shoulders remained stable. Three patients (12%) had an unsatisfactory rating using the Neer criteria and one patient required a reoperation. No perioperative complications were noted.