Nonoperative Treatment of Frozen Shoulders

Abstract & Commentary

Synopsis: Phase II frozen shoulder can effectively be treated with a therapy program. However, significant differences persist in motion and pain compared to the unaffected shoulder.

Source: Griggs S, et al. Idiopathic adhesive capsulitis: A prospective functional outcome study of nonoperative treatment. J Bone Joint Surg Am 2000;82-A:1398-1407.

Griggs and colleagues evaluated 75 consecutive patients (77 shoulders), mean age 53 years, with idiopathic capsulitis (phase II) treated conservatively with a rehabilitation exercise protocol. The rehabilitation program consisted of structured home therapy, and for 91% of the patients, an associated in-house physical therapy program. To be included in this study, patients had to have no history of trauma, significant pain at all extremes of motion, marked loss of active and passive shoulder motion, globally limited glenohumeral translation and normal radiographs. Patients with systemic disorders (such as diabetes) or associated pathology (such as cervical spine arthritis) were included, but patients with glenohumeral arthritis or reflex sympathetic dystrophy were excluded. Posttreatment evaluation consisted of subjective and objective assessments including measurements of pain, range of motion, and function, as well as the Disability of the Arm, Shoulder, and Hand Questionnaire (DASH) and the Short Form-36 Health Survey (SF-36).

Seven (10%) patients were not satisfied with their shoulders following treatment, and five (7%) required surgical intervention (manipulation or arthroscopic capsular release). Patients were candidates for surgical intervention only if they did not improve after at least three months of treatment. Griggs et al wrote, "Patients who had the worst perceptions of the shoulder before treatment tended to have the worst outcomes." Sixty-four (90%) patients were satisfied with their results. However, in these satisfied patients, DASH scores were not comparable with reported normals, although the SF-36 scores were.

Despite the program being rated a success based on patient self-assessment, Griggs et al report measurable objective differences between the affected and unaffect- ed shoulders. Results in the satisfied group of patients demonstrated significant increases in all ranges of motion; however, at final evaluation the range of motion of the affected shoulder was significantly less than the unaffected shoulder (P < 0.0001). Moreover, even though patients at the final evaluation felt comfortable with their arm, 27% still reported mild to moderate pain with activity.

Nonetheless, Griggs et al felt that in this study, which defined outcome in terms of patients’ self-assessment rather than "categorical ranking," the results demonstrated that the vast majority of patients with phase II idiopathic capsulitis can be effectively treated with a shoulder-stretching exercise program.

Comment by Letha Y. Griffin, MD, PhD

Adhesive capsulitis is a poorly understood syndrome whose treatment remains controversial. Duplay, who initially described this syndrome in 1872, felt manipulation under anesthesia was necessary for successful treatment in most cases.1 Codman, who in 1934 termed the condition "frozen shoulder," felt that most cases spontaneously resolve in two years.1 Although Neviaser1 used the term "adhesive capsulitis" to describe the increased vascularity and fibrosis, which he felt represented a reparative inflammatory process associated with this disease, Connolly1 has stated that this is a misnomer. He felt the underlying pathophysiology is fibroplasia, not inflammation, resembling the fibroplasia of Dupuytren’s contracture. He recommended a treatment regimen of physical therapy modalities and passive stretching exercises with closed manipulation or arthroscopic release when patients show no improvement after 6-8 weeks. In contrast, Rockwood and colleagues2 believed the syndrome is self-limited and that the treatment of choice is a physician directed, self-administered, rehabilitation program that they term "orthotherapy." In their report of 50 patients treated with this program, functional motion returned at an average of 14 months.

Vad and Hannafin3 believe that this condition is both inflammatory and fibrosing depending on the stage of the disease as described by Neviaser: Stage I—hypervascular synovitis with little scarring; Stage II—hypervascular synovitis with a proliferative fibroblastic capsular response; Stage III—mild synovial hyperplasia with significant capsular fibrosis; Stage IV—the thawing stage where no pathologic specimens are available for review since patient’s symptoms are resolving and surgical intervention is not needed. These researchers have suggested, in addition to physical therapy, significant benefits are gained by the judicious early use of injectable steroids in stage one and two (i.e., the inflammatory stages).

In summary, adhesive capsulitis is a perplexing enigma with myriad suggested treatments including benign neglect, oral or injectable corticosteroids, non-steroid anti-inflammatory agents, physical therapy modalities, exercise, ultrasound, acupuncture, nerve stimulation, infiltration brisement, manipulation under anesthesia, and arthroscopic and open releases. Griggs et al highlight an intriguing concept: that is, success of treatment should be based on the patient’s self assessment of satisfaction rather than purely on an evaluator’s objective measurements.


1. Connolly JF. Unfreezing the frozen shoulder. Journal of Musculoskeletal Medicine 1998;1:47-57.

2. Miller MD, Wirth MA, Rockwood CA, Jr. Thawing the frozen shoulder. Orthopaedics 1996;19:849-853.

3. Vad VB, Hannafin JA. Frozen shoulder in women: Evaluation in management. Journal of Musculoskeletal Medicine 2000;1:13-28.