Therapist’s Management of the Frozen Shoulder


  • The key clinical examination finding for frozen shoulder (FS) is the loss of at least 50% of passive external rotation (ER) at the side.

  • Successful treatment is evidenced by significant symptom reduction, improved patient satisfaction, and restoration of functional motion.

Rehabilitating patients with frozen shoulder (FS) can be a challenging but rewarding endeavor. The therapist must understand the natural history of this unique pathology, and the patient must have patience. FS, or adhesive capsulitis, results in painful and limited active and passive shoulder range of motion (ROM). FS is reported to affect 2% to 5% of the general population increasing up to 10% and 38% of patients with thyroid disease and diabetes mellitus, respectively. Individuals with primary FS are commonly between 40 and 65 years old, and the incidence appears higher in females than males. The occurrence of FS in one shoulder increases the risk of contralateral shoulder involvement by 5% to 34%, and simultaneous bilateral shoulder involvement occurs as often as 14% of the time.


Zuckerman proposed a classification schema defined by primary and secondary FS ( Fig. 90-1 ). Primary FS and idiopathic adhesive capsulitis are considered identical and not associated with a systemic condition or history of injury. Secondary FS was defined by three subcategories: systemic (e.g., diabetes, thyroid disease), extrinsic (e.g., myocardial infarction, cervical disk disease), and intrinsic (e.g., rotator cuff/biceps tendinopathy (see Fig. 90-1 ).

Figure 90-1

Classification system. AC, acromioclavicular; CVA, cerebrovascular accident.

(Reprinted with permission from Cuomo F. Diagnosis, classification and management of the stiff shoulder. In: Iannotti JP, Williams GR, eds. Disorders of the Shoulder: Diagnosis and Management. Philadelphia: Lippincott-Williams & Wilkins; 1999.)

Another proposed classification system is based on the patient’s irritability level (low, moderate, and high) since most often the intervention is chosen based on tissue irritability ( Table 90-1 ). Irritability is determined based on pain, ROM, and extent of disability. Patients with low irritability have lower disability, less pain, and capsular end-feels with little or no pain with over pressure; therefore, active and passive motion may be equal. These patients typically report stiffness rather than pain as a chief complaint. Patients with high irritability have significant pain, resulting in limited passive motion (due to muscle guarding) and greater disability. These patients typically report pain rather than stiffness as a chief complaint. Although these criteria are not time-based, most commonly patients with early-stage FS have a high level of irritability, and those with later stages have low irritability.

Table 90-1

Irritability Classification

High Irritability Moderate Irritability Low Irritability

  • High pain (≥7/10)

  • Consistent night or resting pain

  • High disability on DASH, ASES, Penn Shoulder Score

  • Pain prior to end ROM

  • AROM < PROM, secondary to pain

  • Moderate pain (4–6/10)

  • Intermittent night or resting pain

  • Moderate disability on DASH, ASES, Penn Shoulder Score

  • Pain at end ROM


  • Low pain (≤3/10)

  • No resting or night pain

  • Low disability on DASH, ASES, Penn Shouler Score

  • Minimal pain at end ROM with overpressure


AAROM, active assisted range of motion; AROM, active range of motion; ASES, American Shoulder and Elbow Surgeons score; DASH, Disabilities of the Arm, Shoulder and Hand Questionnaire; PROM, passive range of motion; ROM, range of motion.


Hannafin and Chiaia described four stages incorporating the arthroscopic stages described by Neviaser and Neviaser, the clinical examination, and the histologic findings. Stage 1, the preadhesive stage, demonstrates mild erythematous synovitis. Patients present with mild end-range pain and are often misdiagnosed as having rotator cuff impingement. Stage 2, the acute adhesive, or “freezing,” stage, is characterized by a thickened red synovitis or angiogenesis (vascular proliferation). Patients frequently have a high level of discomfort and a high level of pain near end-range of movement. Even though this phase is represented by pain, examination under anesthesia reveals connective tissue changes resulting in loss of motion. Stage 3, the fibrotic, or “frozen,” stage, is characterized by less synovitis but more mature adhesions. Patients note significant stiffness with less pain. Based on examination under anesthesia, these patients have motion limited by established contracture as opposed to pain, as evidenced by equal passive motion to that when awake. Severe capsular restriction without apparent synovitis defines stage 4, or the “thawing,” phase. Patients present with painless stiffness and motion that typically improves by remodelling.


In addition to taking a comprehensive history, the following three questions help to determine the current stage of the patient’s FS. First, “Can you sleep through the night?” The ability to sleep well indicates low irritability. Secondly, “Do you have more pain or stiffness?” Greater pain reflects active synovitis, whereas stiffness indicates resolving synovitis and stage progression from stage 2 to 3. The third question, “Is it better or worse in the last 3 weeks?,” can help determine if the patient is moving into or out of stage 2.

A full upper quarter examination is performed to rule out cervical spine and neurologic pathologies. The examination typically reveals significant global limitations of both active and passive ROM with elevation usually less than 120 degrees, but motion limitations are stage-dependent. Scapular substitution frequently accompanies active shoulder motion.

Loss of motion of greater than 25% in at least two planes, a 50% loss of passive ER, or less than 30 degrees of ER are indicative of a patient with FS. Passive motions should be assessed supine to appreciate the quality of the resistance to motion at the end of passive movement—that is, the end-feel. Frequently, passive glenohumeral motions are restricted due to pain at or before end-range, and muscle guarding can often be appreciated at end-range. I believe muscle guarding can masquerade as a capsular end-feel. I examined six patients prior to manipulation both before and after administration of anesthesia. The examinations revealed an increase in passive motion of 10 to 30 degrees during anesthetization in five of six patients. A capsular end-feel was appreciated in all patients before anesthesia; however, this was a false interpretation. Partial improvement in motion related to diminished pain and muscle guarding has been reported after local or regional anesthetic.

Patients with FS are considered to have normal strength and painless resisted motions. However, markedly reduced shoulder isometric maximal voluntary force has been reported in patients with FS. Objective shoulder strength measurements using a hand-held dynamometer revealed significant weakness of the shoulder internal rotators and elevators in patients with FS.

Special tests such as impingement signs and Jobe’s test are not helpful in differentiating FS from rotator cuff tendinopathy since they require painful end-range positioning.


Multiple interventions have been studied; however, definitive treatment remains unclear. Unfortunately, comparison between studies is difficult due to varied inclusion criteria, different protocols, and varied outcome assessment tools. One of the major difficulties in assessing efficacy is defining success criteria because the majority of patients with FS significantly improve in approximately 1 year regardless of how they were treated. Successful treatment should not be defined by the return of “normal” motion, but rather as significant symptom reduction, improved patient satisfaction, and improved functional motion. Therefore the initial goal is pain reduction by influencing synovial tissue. Typically, as pain lessens, the patient moves well within the passive limits and “feels” greatly improved. The capsuloligamentous complex (CLC) tissue length is restored, and motion returns as dense fibrotic collagen remodeling occurs over time.


Little data support the use of frequently employed modalities such as heat, ice, ultrasound, or electric stimulation. Modalities may influence pain and muscle relaxation, and therefore they might enhance the effect of exercises and manual techniques. Hot packs can be applied before or during ROM exercises. Application of moist heat in conjunction with stretching has been shown to improve muscle extensibility. This may occur by a reduction of muscle viscosity and neuromuscular-mediated relaxation. Gursel and colleagues demonstrated the lack of efficacy of ultrasound compared with sham ultrasound in treating shoulder soft tissue disorders. Transcutaneous electrical nerve stimulation (TENS) together with a prolonged low-load stretch resulted in less pain and improved motion in patients with FS.


Outcomes have been reported in patients with FS treated primarily with stretching exercises in therapy. Stretching appears to influence pain and improve motion over time. Studies have shown that exercise results in improved ROM and outcomes; however, aggressive stretching may be detrimental. An impressive finding among multiple studies is symptom improvement and minimal or no difference in outcomes (at 3–6 months) in patients treated with a therapist-directed home exercise program compared with other interventions.

Applying the correct “dose” of stretching is important and based on the stage of FS and the patient’s irritability classification ( Table 90-2 ). In patients with high irritability, low intensity and short-duration ROM exercises are performed to simply alter the joint receptors input, reduce pain, decrease muscle guarding, and increase motion. Three factors should be considered when calculating “dose” or total amount of stress delivered to a tissue: intensity, frequency, and duration. The total end-range time (TERT) is the total amount of time the joint is held at or near end-range position. TERT is calculated by multiplying the frequency and duration of the time spent at end-range daily and is a useful way of measuring the dose of tissue stress. Intensity remains an important factor in tensile stress dose but is typically limited by pain. Traditional ROM exercises are considered lower forms of tensile stress, whereas the highest tensile stress doses are achieved by low-load prolonged stretching (LLPS) because TERT is maximized. Therefore, the goal with each patient is to determine the therapeutic level of tensile stress required based on the patient’s irritability level and response to treatment.

Table 90-2

Treatment Strategies Based on Irritability Level

High Irritability Moderate Irritability Low Irritability
Modalities Heat/ice/electrical stimulation Heat/ice/electrical stimulation
Activity modification yes yes
ROM/stretch Short duration (1–5 sec) Pain-free passive, AAROM Short duration (5–15 sec) Passive, AAROM → AROM

  • End-range/overpressure

  • Increase duration

  • Cyclic loading

Manual techniques Low-grade mobilization Low- → high-grade mobilization High-grade mobilization/sustained hold at end ranges

  • Low → high resistance

Functional activities Basic High demand
Patient education + + +
Other Intra-articular steroid injection

AAROM, active assisted range of motion; AROM, active range of motion; ROM, range of motion.

Joint Mobilization

Several studies have examined the effect of joint mobilization in patients with FS, and although there is evidence that it may be beneficial there is little evidence that it is more efficacious than other forms of treatment. One study treated one group with high-grade (III and IV) and another with low-grade (I and II) mobilizations without any other intervention for three sessions a week for 12 weeks. Outcome measures for both groups improved, and although the high-grade mobilization group did better, only a minority of comparisons reached statistical significance and the overall differences between the two interventions was small. In other words, just performing low-grade mobilizations (for pain) brought about significant improvement in motion and function.

Clearly some patients appear to react very favorably to joint mobilization, demonstrated by their response to treatment. Further research is required to determine if certain patient characteristics respond to specific interventions such as joint mobilization.


Corticosteroids can be administered by injection to dampen the inflammatory response in patients with FS. Significant evidence demonstrates that intra-articular corticosteroid injections provide significant improvement of symptoms in the first 4 to 6 weeks of the intervention. Although no long-term differences can be attributed to corticosteroids, the belief is that the patient can be made more comfortable at rest and with functional movement and the synovitis stage possibly shortened, thereby hastening the time needed to achieve end-range stretching of fibrotic tissue.

Proposed Intervention Algorithm

An algorithmic approach offering conservative and surgical interventions can effectively address the FS continuum ( Fig. 90-2 ). The patient is seen by the shoulder service surgeon, physician, or therapist and diagnosed with FS. The patient is given one of four treatment options based on the known pathology, and evidence-based literature. The patient is given the option to receive an intra-articular corticosteroid injection. Whether he or she receives the injection or not, the patient is referred to physical therapy for either a home exercise program (HEP) or supervised therapy (ST). There is not clear evidence to determine which patients may need formal ST rather than simply a HEP. Therefore the decision is made based on the physician’s and patient’s preference with input from the therapist after initial evaluation. Factors that may favor use of ST may be increased irritability, greater disability, more comorbidities, lower social support, lower educational level, or high fear and anxiety.

Apr 21, 2019 | Posted by in PHYSICAL MEDICINE & REHABILITATION | Comments Off on Therapist’s Management of the Frozen Shoulder
Premium Wordpress Themes by UFO Themes