2

An Arthroscopic Technique for the Management of Patients with Frozen Shoulder

Albert W. Pearsall, IV, M.D.*Daryl C. Osbahr, B.S.Kevin P. Speer, M.D.#

*Department of Orthopaedic SurgeryUniversity of South Alabama

Mobile, Alabama 36617

#Division of Orthopaedic SurgeryDuke University Medical CenterDurham, North Carolina 27710

Address correspondence to: Kevin P. Speer, M.D.Division of Orthopaedic Surgery & Sports MedicineSection of Sports Medicine and RehabilitationDivision of Orthopaedic Surgery

` Duke University Medical CenterBox 3435

Durham, North Carolina 27710(919) 684-6431

Forty-three patients with a diagnosis of primary or secondary frozen shoulder who

had symptoms for an average of 12 months and failed conservative management of at least

12 weeks of physical therapy were treated with an arthrosopic capsular release. Upon

3

completion of a standard shoulder arthroscopy, an intraarticular cautery was used to

completely divide the anterior-inferior capsule, the intraarticular portion of the

subscapularis tendon, the middle glenohumeral, the superior glenohumeral, and the

coracohumeral ligaments. The subacromial space was inspected in all patients. Eighteen

patients had extensive subacromial fibrosis that required debridement. Subacromial

decompression was reserved for patients with evidence of an acromial spur visualized at the

time of arthroscopy. Post-operatively, all patients demonstrated substantial gains in shoulder

range of motion, as well as diminished shoulder pain. Thirty-five patients completed a

phone survey at an average of 22 months after surgery. The average modified shoulder

score was 19 (scale 13 to 65), with 83% of patients indicating that their shoulder was

normal or caused only mild symptoms. In conclusion, the authors believe that arthroscopic

capsular release is an effective and safe alternative to manipulation in the recalcitrant frozen

shoulder patient.

Key Words: Frozen shoulder, arthroscopic capsular release

4

The painful stiff shoulder has been the subject of numerous investigations to

elucidate the etiology and an effective treatment for this disabling condition. Codman

originally coined the term frozen shoulder to describe the features of slow onset shoulder

pain, localized discomfort near the deltoid insertion, an inability to sleep on the affected

side, restricted glenohumeral elevation and external rotation, and a normal radiological

appearance (1).

One of the initial objectives in the treatment of frozen shoulder is to restore motion

and thereby improve shoulder function. A number of therapeutic interventions have been

reported with mixed results. These treatments include analgesics, systemic steroids,

physical therapy, stretching and strengthening exercises, mobilization techniques, injections,

manipulation, distention arthrography, and arthroscopy (2,3). Among the numerous

published treatments for frozen shoulder, many techniques do not enable immediate range

of motion until the treatment had been completed (stretching, systemic steroids, analgesics)

or have the potential to precipitate significant pain after the intervention (manipulation,

mobilization techniques, distention arthrography) due to soft tissue trauma (2).

Prior to the onset of the current study, the senior author (KPS) treated patients with

recalcitrant frozen shoulder by glenohumeral manipulation under anesthesia. Post-

manipulation, many patients underwent diagonistic shoulder arthroscopy to document the

results of this procedure. Based upon these early arthroscopic findings, it was noted that

certain patterns of capsular tears correlated with specific pre-operative shoulder arc of

motion loss. Utilizing this clinical information, the authors developed an experimental

model to examine the specific contributions of the regional capsule to glenohumeral motion

(4). Incorporating this information with the patient’s preoperative motion loss, the authors

5

found that they could precisely direct their capsular release to maximize glenohumeral

motion improvement.

MATERIALS AND METHODS

Forty-six patients with a diagnosis of primary or secondary frozen shoulder who had

symptoms for an average of 12 months and failed conservative management of at least 12

weeks of physical therapy and non-steroidal and/or systemic steroid medications were

treated arthroscopically with the procedure described below. No patient had undergone a

previous surgical procedure for frozen shoulder. All procedures were performed over a 3-

year period and no patient underwent manipulation under anesthesia.

Preoperatively all patients had a side-to-side shoulder examination under anesthesia

performed with attention directed towards allowable passive glenohumeral range of motion.

Specifically, internal and external passive glenohumeral motion was evaluated at 0 and 90

degrees of abduction (Figs. 1-3). In addition, isolated glenohumeral elevation (IGHE) was

recorded in the scapular plane (Fig. 4). IGHE was measured by having the examiner

stabilize the scapula with one hand, while the other hand was used to elevate the ipsilateral

arm. The arc of motion of the glenohumeral joint was noted until movement of the scapula

upon the thorax was detected. The initiation of scapulothoraic motion delineated the

extreme of isolated glenohumeral motion. Based upon the authors’ previous work defining

the specific capsular contributions to glenohumeral motion, precise preoperative assessment

of glenohumeral motion loss determined the regional capsule that would be released (4).

Loss of external rotation at 0 degrees of abduction mandated a release of the anterosuperior

capsule. Loss of glenohumeral external rotation at 90 degrees of abduction merited a

6

release of the anteroinferior capsule. Diminished isolated glenohumeral elevation merited

release of the anterior-inferior capsule. Finally, loss of glenohumeral internal rotation

warranted a release of the posterior capsule.

Postoperatively, all patients were followed by the senior author (KPS). He

clinically examined all patients and documented their range of motion at the most recent

follow-up (5 months). If a patient's chart contained at least 1 of the motion parameters

recorded (IGHE, ER0, ER90, IR90), the data was used for statistical analysis. Three patient

records were incomplete and were not included, leaving 43 records for statistical analysis.

In addition, 35 patients were contacted by phone at an average of 22 months after their

surgical procedure. Each patient answered a modified shoulder questionnaire that addressed

pain, range of motion, instability, ability to work, and overall function (Fig. 5).

STATISTICS

Descriptive statistics including mean and standard deviations were calculated for

patients’ range of motion measurements. A Pearson correlation at the p<0.01 level of

significance was used to analyze the modified shoulder questionnaire variables.

SURGICAL TECHNIQUE

Interscalene anesthesia with Ropivicaine 0.5% and Epinephrine 1:200,000 was used

in all patients and the procedure was done as an outpatient. All patients received

intravenous sedation during the procedure. The interscalene anesthetic provided

approximately 12-18 hours of pain relief postoperatively.

7

After a side-to-side examination of allowable passive glenohumeral motion was

performed, the patient underwent a standard glenohumeral arthroscopy in the beach chair

position. The anterior portal was placed as cephalad as possible, so that the outflow cannula

entered the joint immediately anterior to the biceps tendon. A standard glenohumeral

arthroscopy was performed during which the biceps tendon, anterior and posterior glenoid

labrum, humeral head, inferior capsular recess, and rotator cuff were inspected.

Frequently, the intraarticular portion of the subscapularis tendon was difficult to visualize

secondary to proliferative synovitis. Any degenerative labral tears were debrided to a stable

rim with a motorized shaver. Upon completion of the preliminary joint arthroscopy, an

intraarticular cautery was placed through the anterior cannula. The arthroscopic cautery was

placed approximately 1 centimeter lateral to the glenoid and used to completely divide the

anterior capsule and the intraarticular portion of the subscapularis tendon until the

subscapularis muscle was visualized (Fig. 6). The anterior-superior capsular release

involved the middle glenohumeral, superior glenohumeral, and coracohumeral ligaments.

No attempt was made to release any of these ligaments individually. The endpoint of the

anterosuperior release was defined when the posterior aspect of the coracoid was visualized

and all tissue anterior to the biceps was divided. The extent of the inferior capsular release

was noted when the deltoid muscle began to twitch with the use of the intraarticular cautery,

indicating regional proximity to the axillary nerve. If a patient demonstrated preoperative

evidence of diminished glenohumeral internal rotation, the arthroscope was placed in the

anterior portal while the intraarticular cautery was brought into the posterior portal and a

posterior release or partial capsular debridement was performed. Next, a subacromial

arthroscopy was performed in all patients. If the subacromial space was noted to be free of

8

fibrosis and subacromial spurring, the arthroscope was removed. However, if indicated, a

subacromial soft tissue debridement with release of the coracoacromial ligament was

performed. In addition, if a prominent subacromial spur was present, it was removed at this

time.

Upon completion of the intraarticular capsular release and subacromial debridement,

the operated arm was placed through a range of motion, including glenohumeral external

rotation at zero degrees of abduction, as well as external and internal glenohumeral rotation

at 90 degrees of abduction. The patient’s post-surgical shoulder motion was compared to

preoperative glenohumeral motion measurements of the ipsilateral and contralateral

shoulders.

Postoperatively, all patients received narcotics for pain, cryotherapy, and a 21 day

tapered course of Prednisone (Table 1). All patients had their slings removed the day

following surgery to begin passive range of motion exercises with the therapist. Progressive

passive and active-assisted range of motion exercises including aquatics were instituted the

first week after surgery.

RESULTS

There were 22 males and 24 females with an average age of 49 years (range 21-76

years). Due to incomplete records, only 43 charts were used for statistical analysis. The

average follow-up periods for the functional and clinical assessments were 22 and 5 months

respectively. Twelve percent of patients had a history of diabetes mellitus and 49%

underwent at least 1 subacromial injection prior to surgery. Twenty-four charts of patients

who underwent an isolated anterior release were reviewed, in addition to nineteen charts of

9

patients who underwent an anterior and posterior capsular release. There were no

postoperative complications with regard to compartment syndrome or axillary nerve injury.

Eighty-three percent of patients had improvement in isolated glenohumeral

elevation (IGHE) post-operatively. At the latest follow-up, the average gain in IGHE for all

patients was 24 degrees. Among patients who underwent an anterior capsular release, 90%

demonstrated improvement in both external rotation at 0 degrees of abduction (ER0) and

external rotation at 90 degrees of abduction (ER90) (Table 2). The average gain in ER0 and

ER90 was 37 degrees and 36 degrees respectively. No patient who underwent an anterior

capsular release lost motion at the latest follow-up.

Among patients who underwent a posterior capsular release, 100% demonstrated

improvement in internal rotation at 90 degrees of abduction (IR90) (Table 2). The average

gain in IR90 was 30 degrees. No patient who underwent a posterior capsular release lost

motion at the latest follow-up.

Among the 35 patients who were interviewed by phone, the average shoulder

functional score was 19, with 13 indicating normal function and 65 completely disabled

(Table 3). The average score for pain was 1.6, while the average score for range of

motion was 1.9. Patients’ pain scores did not correlate with age or length of follow-up.

The patients’ perceived motion loss score averaged 1.9. The average score for instability

was 1.1, with 1 patient having a score greater than 2 (mild symptoms). The average

score for work capability was 1.9. A statistical correlation was observed between

patients’ functional score and motion loss score, pain score, sports score, and work score

(p < .010).

10

All patients demonstrated arthroscopic evidence of proliferative synovitis, in

addition to capsular and intraarticular subscapularis tendon thickening. The majority of

significant synovitis was noted anteriorly. Forty-one percent of patients had evidence of

significant subacromial fibrosis; however, this was not statistically associated with

decreased internal or external rotation motion loss post-operatively. Although all patients

had evidence of diminished joint volume as demonstrated by difficulty entering and

maintaining the arthroscope within the shoulder joint, no patient demonstrated significant

glenohumeral arthritis. Thirty-five percent of patients’ shoulders were observed to have

degenerative labral tears, none of which were unstable. The significance of these findings

was unclear.

DISCUSSION

The etiology of the frozen shoulder remains enigmatic. Investigators have proposed

a variety of etiological causes from autoimmune theories to systemic disease (5,6,7,8,9).

Lundberg subdivided the frozen shoulder syndrome into: 1) primary or idiopathic; and 2)

secondary (10). Others have simply defined frozen shoulder syndrome as a clinical loss of

glenohumeral motion preceded in many patients by an antecedent period of relative

immobilization (2,11). A variety of factors have been reported to be associated with frozen

shoulder, including fractures, soft tissue trauma, and neurologic injury (12). In addition, an

autoimmune theory has been postulated, with elevated levels of C-reactive protein and an

increased incidence of HLA-B27 histocompatibility antigen reported in patients with frozen

shoulder versus controls (5,6). DePalma proposed that muscular inactivity was a major

etiological factor, while Bridgman identified an increased incidence of frozen shoulders in

11

patients with diabetes mellitus (8,13). Frozen shoulder has also been associated with

cervical disease, hyperthyroidism, and ischemic heart disease (7,9,10).

The term adhesive capsulitis has also been used to describe the clinical entity of

frozen shoulder (14,15,16). Despite the use of this term by numerous authors, adhesive

capsulitis may not accurately reflect the syndrome of frozen shoulder (14, 15). Several

authors have demonstrated minimal inflammatory response in histological specimens taken

from patients with frozen shoulder (10,17,18,19,20).

Despite a lack of evidence linking frozen shoulder syndrome to a specific etiology,

there appear to exist various triggers that may predispose an individual to this problem. A

few of these reported etiologic agents include trauma, surgery including but not limited to

the shoulder, inflammatory disease, diabetes, regional conditions, and various shoulder

maladies that have been reported to have similar pathological findings (4,18,19). In the

current study, when the pathological findings observed within the subacromial space were

compared among patients, there were many similarities. Subacromial fibrosis with

hypertrophic synovium was noted in approximately 40% of all patients, regardless of the

preoperative etiology. Therefore, despite the small numbers in the current study, the

authors believe that the pathologic and arthroscopic similarities found within the

subacromial space among this study population may represent a secondary phenomenon of

frozen shoulder syndrome. The authors recommend a subacromial arthroscopy in all

recalcitrant frozen shoulder patients undergoing arthroscopic capsular release. However, we

only perform a subacromial debridement and decompression in patients with extensive

subacromial scarring and arthroscopic evidence of impingement.

12

Although various frozen shoulder series have reported fewer than 50% of patients

with long-term disability, for many patients a more aggressive management approach has

been described (21). Sharma and others have reported the results of glenohumeral capsular

distention with or without concomitant manipulation (22,23). Although good results have

been described with this technique, a large capsular tear can be created during the

procedure, which can cause significant bleeding that may impair immediate postoperative

mobilization. In addition, the risk of humeral fracture and rotator cuff injury is present (2).

Recently, arthroscopy of the shoulder has been shown to have a role in the diagnosis,

staging, and treatment of frozen shoulder. Pollock et al described the use of arthroscopy to

debride the subacromial space and selectively section the coracohumeral ligament to

improve glenohumeral motion (3). Warner et al reported on the use of arthroscopic

capsular release in patients with adhesive capsulitis (24). Ogilive-Harris reported on their

series of 40 patients with frozen shoulder treated with arthroscopy and manipulation or

arthroscopic sectioning of contracted capsular structures. In their study, all isolated

arthroscopy patients underwent complete division of all anterior capsular structures,

including the intraarticular portion of the subscapularis tendon. The inferior margin of the

release included the anterior inferior glenohumeral ligament and the anterior 50% of the

inferior capsule. The authors reported no post-operative complications related to nerve

injury or shoulder instabililty. Moreover, the authors found that arthroscopic division of

the anterior structures was significantly better than manipulation with regard to function,

pain, and overall outcome (18).

In the current study, the authors limited the inferior extent of the anterior-inferior

capsular release due to the proximity of the axillary nerve. In all cases, the extent of the

13

anterior capsular release involved a minimum of 50% of the anterior capsular structures.

Despite the lack of a complete release of the anterior-inferior capsule, symmetric

glenohumeral external rotation at 0 degrees of abduction and isolated glenohumeral

elevation were noted in 90% and 83% of patients respectively. The current study findings

concur with those reported by Ogilive-Harris et al that a release of the anterior capsular

structures is required to restore full glenohumeral external rotation and/or isolated

glenohumeral elevation (18).

In addition to the glenohumeral joint capsule, the intraarticular subscapularis

tendon has also been reported to contribute to limited abduction and/or external rotation

(25). The subscapularis is a broad flat muscle with intramuscular bands or septa that

converge laterally and insert into the lesser tuberosity of the humerus (25,26). Klapper

et al sectioned subscapularis specimens in 4 zones and described a constant histologic

pattern in the distribution of the tendinous bands within the muscle. The authors noted

superior migration of the bands as they traversed laterally, eventually coming to lie

within the superior one-third of the subscapularis at its insertion (26). Clark et al

reported the subscapularis capsule complex to consist of 5 to 6 bundles of collagen fibers

that entered from the subscapulairs muscle belly to the lesser tuberosity (27). Totterman

et al, in an MRI study of the shoulder, noted the subscapularis to have approximately 4 to

6 tendon slips that arose medially deep within the muscle. These slips converge laterally

to form a stout main tendon that had its insertion along the superior aspect of the lesser

tuberosity (28). Cooper et al observed the superior portion of the subscapularis tendon to

be intraarticular when viewed arthroscopically (29). Pearsall et al demonstrated that the

location of the intraarticular release of the subscapularis tendon is lateral to the muscle-

14

tendon junction. At this point, the sagittal diameter of the IASS averages 5 mm, which

represents approximately 86% of the entire subscapularis sagittal diameter (30). The

IASS constitutes only 25% of the entire cehpalad-caudad length of the subscapularis at

this location (Fig. 7). Upon IASS release, it was observed in cadaveric dissections that

the entire anterior muscle-tendon anatomy is undistorted and an anterior layer of muscle

is left intact (30) (Fig. 8).

Despite some authors advocating that the IASS be left intact during arthroscopic

shoulder capsular release, other authors have cited its release as an important step in

improving glenohumeral external rotation (18,24,31,32,33). Although several reports

have described rupture of the intraarticular subscapularis tendon after manipulation for

frozen shoulder, the authors are not aware of any reports describing significant symptoms

of instability after manipulation (18,32). In the current study, the authors incorporated a

complete release of the intraarticular portion of the subscapularis tendon (IASS). At an

average follow-up of 22 months after capsular and IASS release, all patients in the

current study described decreased pain, and improved motion and functional scores

compared to preoperatively. In addition, despite release of the IASS, 97% of patients

described minimal or no symptoms of instability. Based upon this work, we do not feel

that release of the intraarticular portion of the subscapularis tendon jeopardizes anterior

shoulder stability or function.

Selective arthroscopic capsular release enables the surgeon to precisely release

specific capsular structures without the non-specific capsular destruction and bleeding that

occur with manipulation. In addition, the procedure is safe. Ogilive-Harris et al cited the

safety of the procedure and the decreased postoperative morbidity noted in the

15

arthroscopically treated patients (18). In the current study there were no instances of

humeral fracture, nerve injury, or postoperative dislocations as a result of subscapular

tendon release. Another benefit is the capability to determine any concurrent diagnoses

during routine inspection of the joint. Similarly, the subacromial space can be inspected for

evidence of impingement and/or subacromial fibrosis. Finally, by means of a preoperative

interscalene block, the patient receives prolonged postoperative pain relief, in addition to a

sympathetic blockade. The latter may provide relief for any sympathetic pain component of

frozen shoulder that may exist.

In conclusion, the refractory frozen shoulder is a challenging problem, even to the

most experienced shoulder surgeon. Numerous treatment methodologies have been tried in

the past with mixed results. The authors believe that selected arthroscopic capsular release

is a technique that enables the shoulder surgeon to address the clinical problems of

diminished glenohumeral motion and associated shoulder pain, while minimizing

postoperative bleeding, the risk of humeral fracture, and rehabilitation time. With decreased

hemarthrosis and swelling, patients can then be mobilized more quickly to optimize

functional results.

16

REFERENCES

1. Codman EA. Tendinitis of the short rotators. In: Ruptures of the spraspinatous

tendon and other

lesions in or about the subacromial bursa. Boston: Thomas

Todd and Co., 1934.

2. Murnaghan JP. Frozen shoulder. In: Rockwood C, Matsen F, ed. The shoulder.

Philadelphia: W.B.

Saunders, 1990;2: 837-862.

17

3. Pollock RG, Duralde XA, Flatow EL, Bigliani LU. The use of arthroscopy in

the treatment of

resistant frozen shoulder. Clin Orthop 1994;304:30-36.

4. Speer KP. Anatomy and pathomechanics of shoulder instability. Operative

Tech in Sports Med

1993;1:252-255.

5. Bulgen DY, Hazleman BL, Voak D. HLA-B27 and frozen shoulder. Lancet

1976;1:1042-1044.

6. Bulgen DY, Binder A, Hazleman BL, et al. Immunological studies in frozen

shoulder. J.

Rheumatol 1982; 9:893-898.

18

REFERENCES

7. Askey JM. The syndrome of painful disability of the shoulder and hand

complicating coronary occlusion. Am Heart J 1961; 22:1-12.

8. Bridgman JF. Periarthritis of the shoulder and diabetes mellitus. Ann Rheum Dis

1972; 31:69-

71.

9. Wohlgethan JR. Frozen shoulder in hyperthyroidism. Arthritis Rheum 1987; 30:

936-939.

10. Lundberg BJ. The frozen shoulder. ACTA Orthop Scand (Suppl) 1969; 119:1-59.

11. Neviaser RJ. Painful conditions affecting the shoulder. Clin Orthop

1983;173:63-69. .

19

12. Waldburger M, Meier JL, Gobelet C. The frozen shoulder: diagnosis and

treatment. Prospective study of 50

cases of adhesive capsulitis. Clin Rheum. 1992;

11:364-368.

13. DePalma AF. Loss of scapulohumeral motion (frozen shoulder). Ann Surg 1952;

135:193-204.

REFERENCES

14. Bruckner FE, Nye CJ. A prospective study of adhesive capsulitis of the shoulder

(“frozen shoulder”)

in a high risk population. Q J Med 1981; 50:198:191-204.

20

15. Coombes WN. Distention-manipulation for the treatment of adhesive capsulitis

(frozen shoulder

syndrome) (letter). Clin Orthop 1984; 188:309-310.

16. Neviaser, RJ. Adhesive Capsulitis. Orthop Clin North Am 1987;18:439-443.

17. Bunker TD, Anthony PP. The pathology of frozen shoulder: a dupuytren-like

disease. J Bone Joint Surg

1995; 77(B): 677-683.

18. Ogilive-Harris DJ, Biggs DJ, Fitsialos DP, MacKay M. The resistant frozen

shoulder: manipulation

versus arthroscopic release. Clin Orthop 1995; 319:238-

248.

248

21

19. Ozaki J, Nakagawa Y, Sakurai G, Tamai S. Recalcitrant chronic adhesive

capsulitis of the shoulder. Role of

contracture of the coracohumeral ligament and

rotator interval in pathogenesis and treatment. J Bone Joint

Surg 1989;71A:1511-

1515.

REFERENCES

20. Withers RW. The painful shoulder: review of one hundred personal cases with

remarks on the

pathology. J Bone Joint Surg 1949;31:414-417.

22

21. Shaffer B, Tibone JE, Rerlan RK. Frozen shoulder. J Bone Joint Surg 1992:

738-740. .

22. Sharma RK, Bajekal RA, Bhan S. Frozen shoulder syndrome. A comparison of

hydraulic distention and manipulation. Int Orthop 1993;17:275-278.

23. Mulcahy KA, Baxter AD, Oni OO, Finlay D. The value of shoulder distention

arthrography with intraarticular injection of steroid and local anesthetic: a

follow-

up study.” Br. J Radiol. 67(795):263-266 1994, Mar.

up study. Br J Radiol 1994;67:263-266.

24. Warner J, Answorth A, Marks P, Wong. Arthroscopic release for chronic

refractory adhesive capsulitis of the shoulder. J Bone Joint Surg [Am]

1996;78:1808-1816.

25. Hollinshead W. Anatomy for Surgeons: the back and limbs. Philadelphia, Harper

and Row, 1982:313.

23

REFERENCES

26. Klapper R, Jobe F, Matsuura P. The subscapularis muscle and its glenohumeral

ligament-like bands. Am J Sports Med 1992;20:307-310.

27. Clark J, Harryman D: Tendons, ligaments, and capsule of the rotator cuff: gross

and microscopic anatomy. J Bone and Joint Surg [Am] 1992;74:713-725.

28. Totterman S, Miller, R, Meyers S. Basic anatomy of the shoulder by magnetic

resonance imaging. Topics in Magnetic Resonance Imaging 1994;6:86-93.

29. Cooper D, O’Brien S, Warren R. supporting layers of the glenohumeral joint: an

anatomic study. Clin Orthop 1993;289:144-159.

30. Pearsall A, Holovacs, T, Speer, K. Characterization of the intraarticular

component of the subscapularis tendon: an arthroscopic, anatomical, and

histologic study. Annual Meeting of the American Arthroscopy Association of

North America, 1998.

31. Burkhart S. Arthroscopic subcapularis tenolysis: a technique for treating

refractory glenohumeral stiffness following open reduction and internal fixation

of a displaced three-part proximal humerus fracture. Arthrosocopy 1996;12:87-

91.

24

REFERENCES

32. DePalma A. Surgery of the shoulder. Philadelphia, J.B. Lippincott, 1983.

33. Neer C, Satterlee C, Dalsey R, Flatow E. The anatomy and potential effects of

contracture of the coracohumeral ligament. Clin Orthop 1992;280:182-185.

25

Figure Legends

FIG. 1

Examination of passive glenohumeral external rotation at zero degrees of shoulder

abduction.

FIG. 2

Examination of passive glenohumeral external rotation at ninety degrees of shoulder

abduction.

FIG. 3

Examination of passive glenohumeral internal rotation at ninety degrees of shoulder

abduction.

FIG. 4

Examination of isolated glenohumeral elevation.

FIG. 5

Modified shoulder questionnaire.

26

Figure Legends

FIG. 6

View of the anterior structures of the glenohumeral joint as visualized from the

arthroscopic posterior portal. The arrow indicates the outflow cannula placed

immediately caudad to the intraarticular biceps tendon. The dotted line indicates the area

of capsular release.

FIG. 7

Anatomic photograph and drawing demonstrating the subscapularis muscle and tendon in

its entirety. The arrow indicates the area of arthroscopic intraarticular tendon release.

Note that significant muscle is present anteriorly, even after complete release of the

tendon.

FIG. 8

Anatomic drawing of the subscapularis musculotendinous junction viewed from anterior.

Despite complete intraarticular subscapularis tendon release, the musculotendinous

continuity has not been disturbed.

27

TABLE 1. Post-operative corticosteroid dosing in frozen shoulder patients.

Post-Op Day Medication Dose(Days 1-7) Prednisone 40 mg/day

(Days 8-14) Prednisone 30 mg/day

(Days 15-18) Prednisone 20 mg/day

(Days 19-21) Prednisone 10 mg/day

28

TABLE 2. Net change in glenohumeral motion after anterior arthroscopic capsular release.(Values in degrees. A negative value indicates lost motion)

Patient number IGHE* ER0** ER90+ IR90++1 25 45 70 302 15 45 30 03 15 25 20 254 20 30 20 105 0 55 10 156 25 45 60 07 0 13 20 158 10 50 10 09 30 30 75 010 10 15 20 511 30 55 30 3012 70 45 60 6013 50 50 55 1014 10 55 20 3015 0 60 60 3016 30 15 70 4017 10 20 45 1518 50 25 50 019 10 0 10 020 25 25 30 521 35 45 70 3022 10 35 70 3023 25 60 45 2024 15 35 30 0

N=24 X=22 X=37 X=40 X=17

* Isolated glenohumeral elevation** External rotation at 0 degrees of abduction+ External rotation at 90 degrees of abduction++ Internal rotation at 90 degrees of abduction

29

TABLE 3. Net change in glenohumeral motion after anterior and posterior arthroscopic capsular release.(Values in degrees. A negative value indicates lost motion)

Patient number IGHE* ER0** ER90+ IR90++1 -- 35 0 502 40 35 30 303 -- 55 70 254 30 55 75 105 65 50 40 506 45 65 45 457 35 35 10 508 20 35 50 309 25 50 25 5510 0 0 0 2011 40 65 15 1012 0 0 0 2013 15 65 40 1014 30 40 -- 3015 0 20 -- --16 25 30 20 1517 60 50 40 018 15 40 20 1019 0 0 0 20

N=19 X=26 X=38 X=34 X=27

* Isolated glenohumeral elevation** External rotation at 0 degrees of abduction+ External rotation at 90 degrees of abduction++ Internal rotation at 90 degrees of abduction

30

TABLE 4. Modified shoulder questionnaire post-operative scores.

Variable Score* RangeFunctional Activities 19 (+/-9) [13-65]

Instability 1.1 (+/-0.4) [1-4]

Limitation of Motion 1.9 (+/-0.9) [1-4]

Night Pain 1.4 (+/-0.8) [1-4]

Pain 1.6 (+/-0.8) [1-5]

Sports 1.7 (+/-0.8) [1-5]

Work 1.9 (+/-1.0) [1-4]

* standard deviations shown in parenthesis

31