journal of orthopaedic & sports physical therapy | volume 39 | number 2 | february 2009 | 135
[ CLINICAL COMMENTARY ]
1Musculoskeletal Team Leader, Good Shepherd Penn Partners, Philadelphia, PA. 2Professor, Arcadia University, Glenside, PA. 3Advanced Clinician II, Good Shepherd PennPartners, Philadelphia, PA. Address correspondence to Dr Martin J. Kelley, Good Shepherd Penn Partners, Penn-Presbyterian Medical Center, 39th and Market Sts, Philadelphia,PA 19104.
MARTIN J. KELLEY, PT, DPT, OCS¹ PT, PhD² PT, DPT, OCS³
Frozen Shoulder: Evidence and a ProposedModel Guiding Rehabilitation
Frozen shoulder, or adhesive capsulitis, describes the commonshoulder condition characterized by painful and limited activeand passive range of motion (ROM). Frozen shoulder is reportedto affect 2% to 5% of the general population,4,13,64,88 increasing
to 10% to 38% in patients with diabetes and thyroid disease.4,5,13,64,71,88
Individuals with primary frozen shoulder are commonly between 40and 65 years old,79,82,83 and the incidence appears higher in femalesthan males.4,9,43,64,71,109 The occurrence of frozen shoulder in 1 shoulderincreases the risk of contralateral shoulder involvement by 5% to
history, clinical presentation, and recov-ery. Codman22 described frozen shoulderas “a condition difficult to define, difficultto treat, and difficult to explain from thepoint of view of pathology.” Nevaiser80
introduced the term adhesive capsulitisto describe the inflamed and fibrotic con-dition of the capsuloligamentous tissue.The term frozen shoulder will be used,because it encompasses both primaryfrozen shoulder (adhesive capsulitis) andsecondary frozen shoulder related to sys-temic disease and extrinsic or intrinsicfactors, excluding cerebral vascular ac-cident, proximal humeral fracture, andcausative rotator cuff or labral pathol-ogy. This paper will present an overviewof the classification, etiology, pathology,examination, and plan of care for frozenshoulder.
The absence of standardized
nomenclature for frozen shouldercauses confusion in the literature.
Lundberg64 first described a classificationsystem identifying primary frozen shoul-der as idiopathic and secondary frozenshoulder as posttraumatic. Nash and Ha-zelman77 expanded the classification sys-tem by including diseases such as diabetesmellitus, myocardial infarction, or vari-ous neurologic disorders under secondaryfrozen shoulder. Zuckerman128 proposeda classification schema where primary
34%, and simultaneous bilateral shoul-der involvement occurs as often as 14%of the time.16,39,64,107
To date, the etiology of frozen shoul-der remains unclear; however, patientstypically demonstrate a characteristic
Frozen shoulder or adhesive cap-sulitis describes the common shoulder conditioncharacterized by painful and limited active andpassive range of motion. The etiology of frozenshoulder remains unclear; however, patientstypically demonstrate a characteristic history,clinical presentation, and recovery. A classifica-tion schema is described, in which primary frozenshoulder and idiopathic adhesive capsulitis areconsidered identical and not associated with asystemic condition or history of injury. Secondaryfrozen shoulder is defined by 3 subcategories:systemic, extrinsic, and intrinsic. We also proposeanother classification system based on thepatient’s irritability level (low, moderate, and high),that we believe is helpful when making clinicaldecisions regarding rehabilitation intervention.Nonoperative interventions include patient educa-tion, modalities, stretching exercises, joint mobili-zation, and corticosteroid injections. Glenohumeralintra-articular corticosteroid injections, exercise,and joint mobilization all result in improved short-and long-term outcomes. However, there is strong
evidence that glenohumeral intra-articular corti-costeroid injections have a significantly greater4- to 6-week beneficial effect compared to otherforms of treatment. A rehabilitation model basedon evidence and intervention strategies matchedwith irritability levels is proposed. Exercise andmanual techniques are progressed as the patient’sirritability reduces. Response to treatment is basedon significant pain relief, improved satisfaction,and return of functional motion. Patients whodo not respond or worsen should be referredfor an intra-articular corticosteroid injection.Patients who have recalcitrant symptoms anddisabling pain may respond to either standard ortranslational manipulation under anesthesia orarthroscopic release.
136 | february 2009 | volume 39 | number 2 | journal of orthopaedic & sports physical therapy
[ CLINICAL COMMENTARY ]
frozen shoulder and idiopathic adhesivecapsulitis are considered identical andnot associated with a systemic conditionor history of injury.128 Secondary frozenshoulder was defined by 3 subcategories:systemic, extrinsic, and intrinsic (1).128 The 3 subcategories for second-ary frozen shoulder identify a relation-ship between some disease process andshoulder symptoms. Systemic secondaryfrozen shoulder is more common amongthese patients, due to the related under-lying systemic connective tissue diseaseprocesses.13,14,88 Extrinsic secondary fro-zen shoulder includes patients whosepathology is not directly related to the
shoulder, and intrinsic secondary frozenshoulder describes patients with a knownpathology of the glenohumeral joint softtissues or structures. Specific causes ofsecondary frozen shoulder may influenceprognosis. For instance, individuals withsecondary frozen shoulder related to in-sulin-dependent diabetes are more likelyto have a more protracted and difficultclinical course.85,86,88
We also propose another classifica-tion system based on the patient’s irri-tability level (low, moderate, and high),that we believe is helpful when makingclinical decisions regarding rehabilita-tion intervention ( ). Irritability is
determined based on pain, range of mo-tion (ROM), and extent of disability. Pa-tients with low irritability have less painand have capsular end feels with little orno pain; therefore, active and passive mo-tion are equal and disability lower. Thesepatients typically report stiffness ratherthan pain as a chief complaint. Patientswith high irritability have significant painresulting in limited passive motion (dueto muscle guarding) and greater disabil-ity. These patients typically report painrather than stiffness as a chief complaint.While these criteria are not time based,most commonly, patients in early-stagefrozen shoulder have a high level of irri-tability, while patients in later stages havelow irritability.
The benefit of Zuckerman’s clas-
sification system is that it organizesthe following previously described
possible etiologies of frozen shoulder intosubcategories: rotator cuff contracture,69
Although the precise etiology remainsunclear, recent evidence identifies el-evated serum cytokine levels as part ofthe process.19,49,101 Cytokines and othergrowth factors facilitate tissue repair andremodeling as part of the inflammatoryprocess. Elevated cytokine levels appearpredominately involved in the cellularmechanisms of sustained inflammationand fibrosis in primary and some sec-ondary frozen shoulder.19,49,75,101 Althoughthe initial stimulus is unknown, Bun-ker et al19 postulated that a minor insultcould initiate an inflammatory healingresponse leading to excess accumulationand propagation of fibroblasts releasingtype I and type III collagen. Fibroblastsdifferentiate into myofibroblasts, caus-ing contraction of newly laid-down typeIII collagen. He proposed an imbalancebetween aggressive fibrosis and a loss ofnormal collagenous remodeling may leadto protracted stiffening of the capsule
Frozen Shoulder
Primary (idiopathic) Secondary (knowndisorders)
Systemic:
mmm
Extrinsic:y
diseasec
A
Intrinsic:
s
ss
Classification system. Reprinted with permission from Coumo F. Diagnosis, classification, andmanagement of the stiff shoulder. In Iannotti JP, Williams GR, eds. Disorders of the Shoulder: Diagnosis andManagement. Philadelphia, PA: Lippincott, Williams & Wilkins; 1999.
Irritability Classification
Abbreviations: 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 HandQuestionnaire; PROM, passive range of motion; PSS, Penn Shoulder Score; ROM, range of motion.
journal of orthopaedic & sports physical therapy | volume 39 | number 2 | february 2009 | 137
and ligaments.19 Using new histologicaland immunocytochemical analysis tech-niques, Hand et al42 found that patientswith frozen shoulder had both chronicinflammatory cells and fibroblast cells,indicating both an inflammatory processand fibrosis.
Frozen shoulder is typically consideredan inflammatory process; however, thisconcept is being challenged. No signifi-cant inflammatory cells in the capsulartissue have been identified upon histo-logical examination.18,19,64,115 Numerousinvestigators report the visual presenceof synovitis consistent with inflamma-tion,43,79,83,123 yet focal vascularity and
synovial angiogenesis (increased papil-lary growth), rather than synovitis, aredescribed by others.18,50,124 In addition toconfirmation of angiogenesis, frequentpositive staining for nerve cells was foundin patients with frozen shoulder.42 How-ever, if the synovial pathology is angio-genesis or synovitis, there is agreementthat pain accompanies the change. Clini-cally, the idea that frozen shoulder occursin the absence of inflammation is difficultto accept, especially because corticoster-oid injections have been shown to havesuch a significant positive short-termeffect.3,15,20,58,104,108,117
There is little disagreement regarding
significant capsuloligamentous complex(CLC) fibrosis and contracture, which areconsistently found in open or arthroscopicshoulder surgery and histologic examina-tions in patients with frozen shoulder.80,115
Contracture of the rotator cuff interval(RCI) is prevalent in patients with frozenshoulder.50,78,86,87,115,116,124 The RCI formsthe triangular-shaped tissue between theanterior supraspinatus tendon edge andupper subscapularis border, and includesthe superior glenohumeral ligament andthe coracohumeral ligament. The intervalacts as an anterior-superior hammock, re-stricting external rotation with the arm atthe side and preventing inferior transla-tion.94 Imbrication of the RCI resulted in a50% loss of external rotation with the armat the side,45 and RCI release in patientswith frozen shoulder leads to an imme-diate and dramatic increase in shoulderexternal rotation ROM.43,78,86,87 Othershave noted significant subacromial scar-ring,50,80 loss of the subscapular recess,64,81
and inflammation of the long head of thebiceps tendon and its synovial sheath123 inpatients with frozen shoulder. Cliniciansattempting to regain shoulder externalrotation should perform stretching andjoint mobilization techniques to target theRCI as well as the anterior CLC.
Reeves96 elaborated on the natural his-tory of frozen shoulder and distinguished3 sequential stages: the painful stage, thestiff stage, and the recovery stage. Han-nafin and Chiaia43 described 4 stagesincorporating the arthroscopic stagesdescribed by Nevaiser,83 the clinical ex-amination, and the histologic findings( ). Stage 1, the preadhesive stage,demonstrates mild erythematous synovi-tis. Patients present with mild end-rangepain and are often misdiagnosed as hav-ing rotator cuff impingement. Stage 2,the acute adhesive or “freezing” stage, ischaracterized by a thickened red synovi-tis. Patients frequently have a high level ofdiscomfort and a high level of pain nearend-range of movement. Even though thisphase is represented by pain, examinationunder anesthesia reveals connective tissuechanges resulting in loss of motion. Stage
Stages of Adhesive Capsulitis*
* Reprinted with permission from Hannafin JA, Chiaia T. Adhesive Capsulitis. Clin Ortho Rel Res.2000;372:95-109.
Duration of symptoms: 0 to 3 months
Pain with active and passive ROM
Limitation of forward flexion, abduction, internal rotation, external rotation
Examination with the patient under anesthesia: normal or minimal loss of ROM
Arthroscopy: diffuse glenohumeral synovitis, often most pronounced in the anterosuperior capsule
Significant limitation of forward flexion, abduction, internal rotation, external rotation
Examination with the patient under anesthesia: ROM essentially identical to ROM when patient is awake
Arthroscopy: diffuse pedunculated synovitis (tight capsule with rubbery or dense feel on insertion of arthroscope)
Pathologic changes: hypertrophic, hypervascular synovitis with perivascular and subsynovial scar, fibroplasias and scarformation in the underlying capsule
Duration of symptoms: 9 to 15 months
Minimal pain except at end ROM
Significant limitation of ROM with rigid end feel
Examination with the patient under anesthesia: ROM identical to ROM when patient is awake
Arthroscopy: no hypervascularity seen, remnants of fibrotic synovium can be seen. The capsule feels thick in insertion ofthe arthroscope and there is a diminished capsular volume
Pathologic changes: “burned-out” synovitis without significant hypertrophy or hypervascularity. Underlying capsule showsdense scar formation
138 | february 2009 | volume 39 | number 2 | journal of orthopaedic & sports physical therapy
[ CLINICAL COMMENTARY ]
the night indicates less irritability. It alsoindicates that the painful synovitis/an-giogenesis is resolving as consistent withstage 3. The second factor is whether painor stiffness is the predominant symptom.The patient experiencing more stiffnessthan pain likely has less symptomaticsynovitis/angiogenesis and more fibrosis.The third factor is whether the symptomshave been improving or worsening overthe last 3 weeks. Improving symptomsmay indicate that the patient is advanc-ing from stage 2 into stage 3, and that theirritability level is decreasing. Recogniz-ing the extent of tissue irritability has adirect influence on the plan of care.
A full upper-quarter examination is per-formed to rule out cervical spine andneurological pathologies. With frozenshoulder, the examination of the shouldertypically reveals significant limitation ofboth active and passive elevation, usuallyless than 120°18,80,95,100; but motion limita-tions are stage dependent. Scapular sub-stitution frequently accompanies activeshoulder motion.103,120 Passive motionsshould be assessed supine to appreciatethe quality of the resistance to motion atthe end of passive movement (end feel).Frequently, passive glenohumeral mo-tions are very restricted due to pain at orbefore end range, and muscle guardingcan often be appreciated at end range. Webelieve that muscle guarding can mas-querade as a capsular end feel. The firstauthor has had the opportunity to ex-amine 6 patients prior to manipulation,both preanesthesia and postanesthesia.
All were felt to have a capsular end feelwhile awake, yet 5 of 6 patients had anincrease in passive motion of 10° to 30°when anesthetized. Partial improvementin motion related to diminished pain, andmuscle guarding has been reported afterlocal or regional anesthetic.109
Cyriax24 described a capsular patternhe believed diagnostic for adhesive cap-sulitis. The capsular pattern is definedas greater limitation of external rotationthan abduction and less-limited internalrotation. Although the capsular pattern isoften encountered, it is not consistentlyseen in patients with frozen shoulder whenobjectively measured.103 A greater than50% reduction in passive external rota-tion or less than 30° of external rotation,when measured with the arm at the side,is a common finding in individuals withfrozen shoulder.8,15,18,24,39,78,96,100,107,119,121
Although authors of textbooks havedescribed patients with frozen shoulderas having normal strength and pain-less resisted motions,24 authors of recentstudies, using handheld dynamometry,have revealed significant weakness of theshoulder internal rotators53,59 and eleva-tors53,59,111 in these patients. The shoulderinternal rotators were significantly weakerin patients with frozen shoulder comparedto patients with rotator cuff tendinopa-thy; however, significant weakness of theexternal rotators and abductors was alsofound relative to the uninvolved side.59
Special tests, such as impingement signsand Jobe’s test, are not helpful in differen-tiating frozen shoulder from rotator cufftendinopathy because they require pain-ful end-range positioning.
3, the fibrotic or “frozen” stage, is charac-terized by less synovitis but more matureadhesions. Patients note significant stiff-ness with less pain. These patients havemotion limited by established contractureas opposed to pain based on examinationunder anesthesia, which reveals equalpassive motion compared to when awake.Severe capsular restriction without appar-ent synovitis defines stage 4, the “thaw-ing” phase. Patients in this phase presentwith painless stiffness and motion thattypically improves by remodeling.
Arthroscopic staging clarifies the con-tinuum of frozen shoulder and, althoughinitially considered a 12- to 18-monthself-limited process, mild symptomsmay persist for years, depending on theextent of fibroplasia and subsequent re-sorption.15,21,22,39,41,107 Authors report mo-tion restrictions in 90% of patients at 6months15 and up to 50% of patients atgreater than 3 years.8,21,107 Mild symptomspersisted in 27% to 50% of patients at anaverage of 22 months to 7 years.39,107
Although specific diagnostic cri-teria do not exist, patients with pri-mary frozen shoulder demonstrate
a consistent history and clinical examina-tion ( ).80,83,96 Primary frozen shoul-der and some secondary frozen shoulder(eg, secondary to diabetes mellitus), ischaracterized by an insidious onset, aprogressive increase in pain, and gradualloss of motion. A minor traumatic eventmay coincide with the patient’s first rec-ognition of symptoms. Pain, specificallysleep disturbing night pain, frequentlymotivates the patient to seek medical ad-vice. Most patients are comfortable withthe arm at the side or with mid-rangeactivities, but often describe a sudden,transient, excruciating pain with abruptor end-range movements.
Three specific factors from the historymay be useful in determining the stageor irritability level of the patient’s condi-tion. First, the ability to sleep through
Characteristic of Primary
Frozen Shoulder
Patient age, 40-65 years
Insidious or minimal, event resulting in onset
Significant night pain
Significant limitations of active and passive shoulder motion in more than 1 plane
50% or greater than 30° loss of passive external rotation
All end ranges painful
Significant pain and/or weakness of the internal rotators
journal of orthopaedic & sports physical therapy | volume 39 | number 2 | february 2009 | 139
Significant loss of passive external ro-tation with the arm at the side, as well asloss of active and passive motion in otherplanes of movement, differentiates frozenshoulder from other pathologies. How-ever, other pathologies resulting in sig-nificant loss of external rotation with thearm at the side include proximal humeralfracture, severe osteoarthritis, acute cal-cific bursitis/tendinitis, and a locked pos-terior dislocation. Early frozen shouldermay be difficult to differentiate from ro-tator cuff tendinopathy because motionmay be minimally restricted and strengthtesting may be normal. The patient witha slight loss of passive external rotationmotion at the side and relatively full mo-tion in all other directions should be cau-tioned to return for further evaluation ifthe patient experiences a rapid progres-sion of shoulder pain and stiffness.
Diagnosing frozen shoulder is oftenachieved by physical examination alone,but imaging studies can further confirmthe diagnosis and rule out underlying pa-thology. Radiography rules out pathologyto the osseous structures. Arthrographyhas been used to determine decreasedglenohumeral joint volume associatedwith adhesive capsulitis.81,118,120 AlthoughBinder et al9 observed that over 90%of patients with frozen shoulder dem-onstrated an increased uptake on thediphosphonate scans (bone scan), theyconcluded bone scans possess little di-agnostic or prognostic value for frozenshoulder.9 Magnetic resonance imaging(MRI) helps with the differential diag-nosis by identifying soft tissue abnor-malities of the rotator cuff and labrum.56
MRI has identified abnormalities of thecapsule and RCI in patients with frozenshoulder.56,70 Recently, ultrasonographyhas gained favor because it can help dif-ferentiating rotator cuff tendinopathyfrom frozen shoulder. A recent study re-vealed fibrovascular inflammatory softtissue changes in the RCI in 100% of 30patients with frozen shoulder.57
A comprehensive history and exami-nation should include a patient-orientedshoulder functional outcome measure.
Multiple shoulder-specific outcomemeasures are available, such as the Dis-abilities of the Arm, Shoulder, and HandQuestionnaire (DASH),68 Simple Shoul-der Test (SST),62 Penn Shoulder Score,60
American Shoulder and Elbow Surgeons(ASES) score,98 and the Constant-Murleyscore.23 These forms typically includequestions relative to the patient’s painand function and some include impair-ment data, such as ROM and strengthmeasurements. To date, research hasnot identified a specific outcome tool orspecified score range that is optimal forindividuals with frozen shoulder.
The definitive treatment for
frozen shoulder remains un-clear even though multiple
interventions have been studied in-cluding oral medications,10,32,97 corti-costeroid injections,3,8,20,25,46,58,95,104,117
exercise,3,15,20,25,28,39,46,55,58,72,95 jointmobilization,15,52,84,118,119,127 disten-sion,37,73 acupuncture,114 manipula-tion,30,44,47,48,76,89,102 nerve blocks,26 andsurgery.1,7,38,43,50,85-87,90,93,106,123,124 Unfortu-nately, varied inclusion criteria, differenttreatment protocols, and various out-come assessments render study compari-son difficult. One of the major difficultiesin assessing efficacy is success criterion.Often success is defined by return of“normal” motion rather than pain-freefunctional motion. It may be implausiblefor conservative treatment to rapidly re-store full pain-free motion, consideringthe presence of dense fibrotic CLC tissueand the months of collagen remodelingrequired to regain soft tissue length. Evenif an intra-articular corticosteroid injec-tion relieves pain in someone with stage 3frozen shoulder, the fibrotic/contracturedtissue continues to limit motion. Estab-lishing treatment effectiveness is alsodifficult because the majority of patientswith frozen shoulder significantly im-prove in approximately 1 year; therefore,natural history must be considered.
Although multiple studies demon-strate improved outcomes with physicaltherapy, these outcomes are not alwayssuperior to other interventions.3,15,20,46,117
Additionally, the optimal use of commonphysical therapy interventions (modali-ties, exercise, joint mobilization), fre-quency and timing of visits, and dischargecriteria have not been established. Theproposed physiologic effect and support-ing literature for using modalities, exer-cises, and manual techniques in physicaltherapy will be discussed in the followingsections.
Patient education about the natural his-tory of frozen shoulder is probably animportant treatment aspect, thoughno studies have specifically addressedthis component. Explaining the insidi-ous nature of frozen shoulder allays thepatient’s fear of more serious diseases.Discussing how the painful synovitis/angiogenesis progresses into fibroplasiaand restricts motion prepares the patientfor an extended recovery. Instruction inperforming a consistent home exerciseprogram (HEP) is important, becausedaily exercise is effective in relievingsymptoms.15,20,55
Little data exist to supporting the use offrequently employed modalities such asheat, ice, ultrasound, or electric stimula-tion. Modalities are suggested to influencepain and muscle relaxation; therefore,they might enhance the effect of exercisesand manual techniques. Hot packs can beapplied before or during ROM exercises.Application of moist heat in conjunctionwith stretching has been shown to im-prove muscle extensibility.51 This may oc-cur by a reduction of muscle viscosity andneuromuscular-mediated relaxation.105,121
Gursel et al40 demonstrated the lack of ef-ficacy of ultrasound, as compared to shamultrasound, in treating shoulder soft tis-sue disorders. Transcutaneous electricalnerve stimulation (TENS), together witha prolonged low-load stretch, resulted
140 | february 2009 | volume 39 | number 2 | journal of orthopaedic & sports physical therapy
[ CLINICAL COMMENTARY ]in less pain and improved motion in pa-tients with frozen shoulder.99
The basic strategy in treating structuralstiffness is to apply appropriate tissuestress.74 It is helpful to think of the totalamount of stress being applied as the“dosage,” in much the same way that dos-age applies to medication. The primaryfactors that guide this process are painand ROM. Adjusting the dose of tissuestress results in the desired therapeuticchange (increased motion without in-creased pain). Three factors should beconsidered when calculating the dose, ortotal amount of stress delivered, to a tis-sue: intensity, frequency, and duration.The total end range time (TERT)34,66 is thetotal amount of time the joint is held ator near end-range position. TERT is cal-culated by multiplying the frequency andduration of the time spent at end rangedaily, and is a useful way of measuring thedose of tissue stress.34,66 Intensity remainsan important factor in tensile stress dosebut is typically limited by pain. Tradition-al ROM exercises are considered lowerforms of tensile stress, while the highesttensile stress doses are achieved by low-load prolonged stretching (LLPS), be-cause TERT is maximized. Therefore, thegoal with each patient is to determine thetherapeutic level of tensile stress.
Applying the correct tensile-stressdose is based upon the patient’s irritabil-ity classification ( ). In patients withhigh irritability, low-intensity and short-duration ROM exercises are performedto simply alter the joint receptors’ input,reduce pain, decrease muscle guarding,and increase motion.126
show commonly performed exercises forpatients with high irritability. Stretchesmay be held from 1 to 5 seconds at therelatively pain-free range, 2 to 3 times aday. A pulley may be used, depending onthe patient’s ability to tolerate the exer-cise. These exercises primarily influencedifferent regions of the synovial/CLCand have been used in supervised physi-cal therapy programs and an HEP in
journal of orthopaedic & sports physical therapy | volume 39 | number 2 | february 2009 | 141
patients with frozen shoulder.20,28,39,55,104
Aggressive stretching beyond the painthreshold resulted in inferior outcomesin patients with frozen shoulder, particu-larly if performed in the early phase of thecondition.28 As the synovitis/angiogenesisand pain reduce, the fibrotic connectivetissue wall is reached (stage 3). Tissuestress is progressed primarily by increas-ing stretch frequency and duration, whilekeeping the intensity in tolerable limits.The patient may be asked to hold thestretch for longer periods and increase thenumber of sessions per day. The patientis instructed to avoid excessive scapularcompensation while performing exercisesto minimize carryover of abnormal move-ment patterns as motion returns.
As the patient’s irritability level be-comes low, more-intense stretching andLLPS using a pulley or device ( )are performed to influence tissue remod-eling. Tissue remodeling refers to a physi-cal rearrangement of the connective tissueextracellular matrix (fibers, crosslinks,and ground substance). Collagenous tis-sues respond to increased tensile loadingby increasing the synthesis of collagenand other extracellular components.36,74,125
The collagen is oriented parallel to thelines of stress, and tensile strength is in-creased. It is important to note that bio-logic remodeling occurs over long periods(months), in contrast to mechanically in-duced change, which occurs within min-utes.2 Brand12 describes this phenomenonas “growth,” not stretch, of the contractedtissue. This growth process is consistentwith the recovery process seen in primaryfrozen shoulder. Commercially availabledevices, such as the Dynasplint (Dynas-plint Systems Inc, Severna Park, MD),and continuous passive motion units canprovide LLPS; however, these devices re-quire specific positioning and dedicatedtime during the day. Sustained end-rangepositioning devices are typically not tol-erated in the patients with high or mod-erate irritability. McClure and Flowers67
have described a simple abduction splintfabricated from thermoplastic materialsthat provides a LLPS.
Outcomes have been reported in pa-tients with frozen shoulder treated pri-marily with exercise in physical therapy.Diercks and Stevens28 prospectively fol-lowed 77 patients with idiopathic frozenshoulder for 24 months to compare theeffects of “intensive physical therapy” to“supervised neglect.” The intensive physi-cal therapy group performed active exer-cises up to and beyond the pain threshold,passive stretching, glenohumeral jointmobilization, and an HEP. The “super-vised neglect” group was instructed not toexercise in excess of their pain threshold,to do pendulum exercises and active ex-ercises within the painless range, and toresume all activities that were tolerated.These authors found both groups madesignificant improvement in ROM andpain; however, 89% of the “supervisedneglect” group achieved a Constant scoreof greater than 80, compared to only 63%of those in the intense-physical-therapygroup. A conclusion of this study wasthat aggressive stretching beyond a painthreshold could be detrimental, espe-cially if applied in the early phase of thecondition.
As mentioned earlier, criteria for suc-cessful treatment is pain reduction andimproved functional motion and patient
satisfaction. Patient satisfaction may ul-timately be the most important measure.Griggs et al39 reported that 90% of 75patients (mean follow-up, 22 months),classified with stage 2 idiopathic frozenshoulder, demonstrated good outcomeswith an exercise program in a prospectivefunctional outcome study. All patientswere referred to physical therapy andperformed HEP of passive stretching ex-ercises in forward elevation, external ro-tation, horizontal adduction, and internalrotation. Ten percent of the patients werenot satisfied with the outcome, and 7%of these patients underwent manipula-tion and/or arthroscopic release. Patientswith the worst perceptions of their shoul-der before treatment tended to have theworst outcomes.
Levine et al61 reported that 89.5% of 98patients with frozen shoulder respondedwith nonoperative management.61 Reso-lution of symptoms occurred in 52.4%with physical therapy and nonsteroidalanti-inflammatory drugs (NSAIDs),while 37.1% resolved with NSAIDs, phys-ical therapy, and 1 or more corticosteroidinjections. The average time to successfultreatment was 3.8 months. An impres-sive finding among several studies is thatpatients placed on a therapist-directedHEP had the same outcomes at short-(4-6 months) and long-term (12 months)follow-ups as those treated with other in-terventions.15,20,55,104 Kivimaki55 comparedpatients treated with an HEP to thosewho underwent manipulation under an-esthesia and HEP. Other than a slightincrease in ROM, the group performingjust an HEP did not differ at any follow-up (6 weeks, and 3, 6, and 12 months) inpain or working ability.
Many authors and clinicians advocatejoint mobilization for pain reductionand improved ROM.31,54,65,84,118,119 Unfor-tunately, little scientific evidence existsto demonstrate the efficacy of joint mo-bilization over other forms of treatmentfor frozen shoulder. However, patientstreated with joint mobilization, with or
Specific joint mobilization techniquesare believed to selectively stress certainparts of the joint capsule; for example,an inferior glide with the arm at the side,while in external rotation, would stressthe RCI ( ). While this may betrue, it may be more beneficial to viewthe CLC through the circle concept. Thecircle concept refers to all regions of theCLC providing stability in all directions(ie, anterior structures providing anterioras well as posterior stability).112 When thisconcept is applied to the shoulder withlimited glenohumeral motion, improvedextensibility of any portion of the CLCresults in improved motion in all planes.This concept appears supported by thefindings of Johnson et al,52 who foundsignificant improvement in external ro-tation motion in patients with frozenshoulder after performing posterior glidemobilizations sustained for 1 minute atend range of abduction and external ro-tation. High-grade joint mobilizations(grades III and IV) are used to promoteelongation of shortened fibrotic soft tis-sues. High-grade mobilizations should beperformed with the joint positioned at ornear its physiologic end range. It shouldbe noted that immediate ROM gainsmade with manual techniques ( joint mo-bilization or end-range stretching) repre-sent transient tissue preconditioning12,35
and must be reinforced by an HEP. Jointmobilization techniques may be com-bined with hold-relax stretching methodsto maximize relaxation, so that tensileload may be applied to the affected CLC.An example is performing a submaximalisometric contraction of the internal ro-tators, preceding an anterior glide, whileat external rotation end range.
Several studies have examined the ef-fect of joint mobilization in patients withfrozen shoulder.15,52,84,118,119 Nicholson84
compared a group of patients who re-ceived joint mobilization and active exer-cise to a group receiving exercise alone.They found significantly improved mo-tion and pain reduction in both groups,
but the mobilization group had greaterimprovement only in passive abductionover the exercise group. Vermulen118 pre-sented a case series of 7 patients with fro-zen shoulder treated using only intenseend-range mobilization techniques (noexercise or modalities) over a 3-monthduration. They reported significant im-provement in active and passive motion,pain, and joint volume. Vermullen119 alsoperformed a randomized prospectivestudy comparing high-grade mobiliza-tion techniques to low-grade mobiliza-tion techniques (grades I and II). Patientswere treated over 12 weeks (24 sessions)and followed for 12 months. They foundsignificant improvement in motion anddisability for both groups and the greatestamount of improvement occurred in thefirst 3 months. The high-grade mobiliza-tion group did better, but only a minorityof comparisons reached statistical signifi-cance and the overall differences betweenthe 2 interventions was small.119 Bulgen etal15 found that patients treated with jointmobilization and an HEP significantly im-proved in the first 4 weeks but not morethan patients receiving intra-articularand subacromial corticosteroid injections.At 6 months, the mobilization group sig-nificantly improved in motion return andpain reduction, but no difference wasnoted compared to the other treatmentgroups, even the group performing justpendulum exercises. Yang et al127 per-formed a multiple-treatment trial usingcombinations of end-range mobilization,midrange mobilization, and mobiliza-tion with motion in patients with frozen
shoulder. They found improved motionand function at 12 weeks, and concludedthat end-range mobilization and mobili-zation with motion were more effectivethan midrange mobilization in increasingmotion and functional mobility.
Corticosteroid injections have been usedto manage inflammatory processes formany years. The proposed effect of cor-ticosteroids is to quell the inflammation,resulting in symptom reduction. Often,corticosteroid injections are adminis-tered with either a short- or long-actinglocal anesthetic lasting 30 minutes to 6hours, respectively. An immediate gain inmotion following a glenohumeral intra-articular corticosteroid injection is attrib-utable to the anesthetic effect of reducingpain and thereby muscle guarding.109
Over subsequent days, the corticoster-oid’s anti-inflammatory effect diminishesthe painful synovitis/angiogenesis.76
Multiple studies have investigated theuse of corticosteroids alone (either intra-articular or subacromial), in conjunc-tion with supervised physical therapy,or with an HEP. Lee et al58 found that all3 exercise groups (2 receiving differentsite-specific corticosteroid injections),treated over a 6-week period, improvedequally, though better than a fourthgroup treated only with analgesics. Thegroups performed active assisted ROMand active ROM exercises and resistiveexercise. The authors noted the greatestmotion improvement in the first 3 weeksof treatment. Hazelman46 comparednonspecified physical therapy, cortisoneinjections, and manipulation under an-esthesia and analgesics for 130 patientsretrospectively. Although they found nosignificant recovery difference amongstthe groups, 28% of the patients in thephysical therapy group had symptom ex-acerbation. Arslan and Celiker3 randomlyallocated patients to receive either an in-tra-articular glenohumeral joint injectionand home exercise, or physical therapyand a nonsteroidal anti-inflammatorydrug. Physical therapy consisted of hot
Inferior glide with the arm at the side andin external rotation.
journal of orthopaedic & sports physical therapy | volume 39 | number 2 | february 2009 | 143
packs, ultrasound, passive glenohumeralstretching exercises and wall climb. ROMand a pain scale were used for outcomemeasures. Patients in both groups im-proved similarly at 2 and 12 weeks. Theauthors concluded that corticosteroidinjections and home exercise were as ef-fective as physical therapy, but injectionswere much cheaper.
Several studies have also advancedthe argument that intra-articular injec-tions may be superior to therapy. Van derWindt et al117 compared intra-articularinjections to physiotherapy in a prospec-tive randomized study on 109 patientswith a stiff, painful shoulder (capsularsyndrome). Physiotherapy consisted oftwelve 30-minute sessions involvingpassive joint mobilization and exercises.Thermal modalities and electrostimula-tion could be used at the therapist’s dis-cretion. At 7 weeks, 77% of the patientstreated with injections were consideredtreatment successes, compared to only46% treated with physiotherapy, andsignificant differences were found innearly all outcome measures. The maindifferences between groups were relatedto faster initial relief of symptoms withinjections.
Bulgen et al15 compared paired intra-articular and subacromial injections,joint mobilization, ice/proprioceptiveneuromuscular facilitation (PNF), andno treatment (pendulum exercise), ina prospective randomized study. Painand ROM significantly improved by thefourth week of treatment for all groupsand continued until 6 months. Improve-ment was most obvious in the corticoster-oid injection group, reaching statisticalsignificance for motion, but not pain,during the first 4 weeks. No significantdifferences were seen among the groupsat 6 months. The study concluded thatthere is little long-term advantage of onetreatment over the other; however, cor-ticosteroid injections may best improvepain and ROM in the first 4 weeks.
Carette et al20 confirmed the benefit ofintra-articular corticosteroid injectionsin treating frozen shoulder in a well-
controlled randomized prospective study(n = 90). This study compared 4 groups,glenohumeral intra-articular corticos-teroid injection with HEP, glenohumeralintra-articular corticosteroid injectionwith physical therapy and HEP, intra-articular saline injection with physicaltherapy, and intra-articular saline injec-tion with HEP. To control for the inac-curacy of blind intra-articular injections,which can be as high as 42%,33 fluoros-copy was used to ensure the accurate lo-cation of the injections of corticosteroid.At 6 weeks, the corticosteroid injection/physical therapy/HEP and corticoster-oid injection/HEP groups demonstrat-ed the largest change in Shoulder Painand Disability Index (SPADI) score andwere improved significantly over thenoncorticosteroid groups. At 6 months,the SPADI scores were similar amongthe groups; however, active and passivemotion were better in the corticosteroidinjection/physical therapy/HEP group.This study concluded that at 6 weeks, anintra-articular injection alone, or in con-junction with physical therapy, was moreeffective than supervised physical therapyor an HEP; however, there was no benefitof one intervention over the others at 12months.
Ryans et al104 also investigated theeffect of corticosteroid injections butperformed both an intra-articular andsubacromial injection. Their methodswere similar to those of Carette et al20 (4groups), except they did not use fluorosco-py-guided injections, and only 8 sessions(over 4 weeks) of physical therapy weredelivered instead of 12. The physical ther-apy program included PNF, mobilization,interferential electrical stimulation, andexercise. At 6 weeks, the injection groupssignificantly improved in the ShoulderDisability Questionnaire (SDQ) com-pared to the other groups; but patientstreated in supervised physical therapygained significantly more external rota-tion motion. All groups significantly im-proved by 16 weeks, but no difference waspresent between the groups. The authorsrecommended an intra-articular and
subacromial corticosteroid injection forrelieving shoulder disability and physicaltherapy for improving external rotationmotion.104
Glenohumeral intra-articular corti-costeroid injections, exercise, and jointmobilization all result in improved short-and long-term outcomes. However, thereis strong evidence that glenohumeralintra-articular corticosteroid injectionshave a significantly greater 4- to 6-weekbeneficial effect compared to other formsof treatment.
e believe that rehabilita-
tion should be guided by theevidence in the literature, the
extent of tissue irritability (as defined in), and the response to treatment. shows basic rehabilitation strate-
gies matched with the level of irritabil-ity. While there is not strong evidencesupporting the use of modalities, theymay be useful in some patients with highor moderate irritability if there is a cleardecrease in pain with their application.Patients with high irritability should betreated with short-duration, relativelypain-free stretching and low-grade jointmobilization to reduce symptoms andavoid exacerbation of pain and inflam-mation. Exercise found to be too pain-ful or resulting in a prolonged painfulresponse is held from the program andreintroduced when irritability reduces.Patients with low irritability shouldbe given longer-duration stretchingtechniques and high-grade mobiliza-tions performed with the joint near endrange. The core exercises include pen-dulum exercise, passive supine forwardelevation, passive external rotation withthe arm in approximately 40° abductionin the plane of the scapula, and activeassisted ROM in extension, horizontaladduction, and internal rotation ( -
). Patients with moderateirritability may be instructed in pulleyuse for elevation. As the irritability level
144 | february 2009 | volume 39 | number 2 | journal of orthopaedic & sports physical therapy
[ CLINICAL COMMENTARY ]reduces, progressive end-range stretch-ing and mobilization may be performed.The authors encourage reassessment ofmotion and end-range discomfort ateach session to determine the patient’sresponse to treatment. Patients classi-fied as having low irritability may beinstructed in the same exercises andthe use of pulleys, but will hold at endrange for up to 30 seconds. More pro-vocative stretching positions are used,such as stretching into external rotationwith the arm in adduction (to isolate thestretch of the RCI) or with the arm inextension and adduction ( ). Webelieve that strengthening and aggres-sive functional activity should be avoid-ed when high and moderate irritabilityis present, and introduced graduallywhen individuals have low irritability;however, regaining motion should al-ways be emphasized.
There is no clear evidence to deter-mine which patients may need formalsupervised therapy rather than simplya home program. Therefore, we recom-mend this decision be made based onthe physician and patient preference,with input from the therapist after initialevaluation. Factors that may favor use ofsupervised therapy may be greater dis-ability, more comorbidities, lower socialsupport, lower educational level, or highfear and anxiety. Patients may initiallybe offered an intra-articular corticoster-oid injection, and clearly those who failto progress within approximately 3 to 6weeks should be offered this option. Areturn visit to the referring physician fora corticosteroid injection should be facili-tated if the patient’s symptoms worsen.
There is also no clear evidence tosuggest proper frequency of supervisedtherapy visits. We make decisions aboutfrequency of visits based on a patient’swithin-session and between-session re-sponses to treatment over the first severalweeks. In general, patients with moder-ate or high irritability who demonstratepain reduction and within-treatmentROM changes of greater than 10° to 15°,are seen more frequently, typically 2
times per week. Patients with low irrita-bility who have achieved pain reductionbut minimal changes in motion are seenless frequently, typically once every weekor 2, with emphasis on the home pro-gram as long as they are able to adhere toit appropriately. Success of treatment isnot necessarily based on the restorationof normal motion but, rather, symptomreduction and patient satisfaction. Com-monly, patients are discharged when thefollowing occur: significant pain reduc-tion, stagnant motion gains between ses-sions, improved functional motion, andimproved satisfaction.
If the symptoms and motion are un-responsive to the various levels of treat-ment over time (3-6 months) and qualityof life is compromised, a manipulationunder anesthesia or surgical capsular re-lease should be considered. If the patientis unwilling to have a manipulation orsurgery, the patient is discharged but en-couraged to continue with a daily stretch-ing program.
-
Manipulation under anesthe-
sia remains a reasonable treat-ment for patients who have not
responded to conservative treatmentand are capable of adhering to a post-manipulation program of stretching andtherapy.30,48,83,90 The anesthesia, eithergeneral or a local brachial plexus block,completely relaxes the shoulder muscles,
ensuring that the force applied by thesurgeon reaches the capsuloligamen-tous structures. Potential complicationsinclude glenoid, scapular, and humeralfractures, dislocations, postmanipula-tion pain, hemarthrosis, rotator cuff tear,labral tears, and traction injuries of thebrachial plexus or a peripheral nerve.63,76
Manipulation under anesthesia is con-trolled, forced, end-range positioning ofthe humerus relative to the glenoid inan anesthetized patient. Surgeons try touse short lever arms to minimize poten-tial fracture risk. Frequently, the surgeonfirst forcefully abducts the shoulder bystabilizing the scapula against the thorax,while elevating the humerus to releasethe inferior capsule. Next, the surgeontypically manipulates the shoulder intoexternal and then internal rotation.47
Audible and palpable release of the tissuesuggests a good prognosis.63 Arthroscopicexamination following manipulation re-veals significant bleeding into the jointdue to tearing of the CLC.63 The use of aglenohumeral joint intra-articular injec-tion of corticosteroid following manipu-lation likely minimizes postmanipulationjoint irritability.48,90 Contraindications tomanipulation of a frozen shoulder in-clude a history of fracture or dislocations,moderate bone loss, or inability to followthrough with postprocedure care. Studiesassessing manipulation under anesthesiareport success rates ranging from 75% to100%, due to varied inclusion criteria,intraoperative procedures, and outcomemeasures.1,30,48,90,93
Instead of traditional manipulationunder anesthesia, Roubal et al102 per-formed translational manipulationfor the treatment of frozen shoulder.Translational manipulation differs fromtraditional manipulation in its use oftranslational (gliding) techniques andstatic end-range capsular stress, with ashort-amplitude high-velocity thrust, ifneeded, as opposed to angular stretch-ing forces. The translational techniques
Stretch to target the rotator cuff interval.The patient’s hand remains fixed and the elbow ismoved toward the table.
journal of orthopaedic & sports physical therapy | volume 39 | number 2 | february 2009 | 145
are identical to joint mobilization tech-niques (anterior, posterior, and inferiorgliding). The authors determined thatpostmanipulation average increase inflexion was 68°, abduction 77°, externalrotation 49°, and internal rotation 45°.Placzek et al92 used the same techniquesand found significant improvement inoutcomes at both short- (5.3 weeks) andlong-term (14.4 months) follow-ups.Boyles et al11 used translational glid-ing as already described on 4 patientswith frozen shoulder; however, theyperformed additional mobilization/ma-nipulation into directions of perceivedrestrictions. Translational manipulationunder anesthesia appears to be a safeand efficacious alternative for the treat-ment of frozen shoulder.11,92,102
Few reports of open surgical re-
lease for frozen shoulder exist.80,110
Complete or near complete returnof motion has been described with openrelease directed toward the RCI and cora-cohumeral ligament.86,87
Arthroscopic surgery has replaced opencapsular release as the preferred surgi-cal treatment of primary frozen shoul-der. Initially, arthroscopic surgery wasused only after manipulation failed; butnow it is typically performed alone or ac-companies the manipulation. However,most clinicians still reserve arthroscopicsurgery for patients with painful, dis-abling frozen shoulder unresponsive toat least 6 months of conservative treat-ment.6,38,43,91,123 With arthroscopy, thesurgeon can identify and address anyintra-articular and subacromial pathol-ogy.63,123 The surgeon selectively releasespathologic fibrosis in a controlled man-ner, versus manipulation, which rup-tures capsuloligamentous structuresnonspecifically.6,43,85,91,123
Debate continues about which struc-tures should be arthroscopically re-
leased. Several authors believe the RCIand the contracted coracohumeral liga-ment are the only structures requiringrelease.7,86,87,124 Berghs7 demonstratedimpressive short- and long-term results(mean follow-up, 14.8 months) in 25 pa-tients with primary frozen shoulder whohad just the RCI and coracohumeralligament released. Other authors se-lectively release additional portionsof the CLC, such as the superior andmiddle glenohumeral ligament,6,50 infe-rior glenohumeral ligament,38,85,91,122 theintra-articular component of the sub-scapularis tendon,29,85,91,122 and the poste-rior capsule.38,122,123
Postoperative protocols can vary fromusing a continuous passive motion deviceand exercise50 to a an initial daily com-prehensive physical therapy program.123
In 37% of the patients, a follow-up intra-articular cortisone injection was requiredat approximately 4.5 weeks.123
Frozen shoulder is a commonly
treated musculoskeletal problem,yet the etiology remains uncertain.
Patients present with a characteristic his-tory, physical examination, and naturalcourse of recovery. Multiple interventionshave been investigated assessing short-and long-term outcomes. Corticosteroidintra-articular injections demonstrateshort-term (4-6 weeks) benefits and arefavored in patients with high irritabilityor those who have not responded well torehabilitation. Applying the correct ten-sile stress dose (intensity, frequency, andduration) while stretching is based on thepatient’s irritability classification. Themajority of patients will respond to con-servative interventions by achieving sig-nificant pain relief, return of functionalmovement, and patient satisfaction. CLCremodeling occurs over a prolonged pe-riod, resulting in functional motion. Thepatient with a recalcitrant frozen shoul-der has the option of manipulation and/or capsular release, if conservative treat-ment fails.
1. Andersen NH, Sojbjerg JO, Johannsen HV,Sneppen O. Frozen shoulder: arthroscopy andmanipulation under general anesthesia andearly passive motion. J Shoulder Elbow Surg.1998;7:218-222.
2. Arem AJ, Madden JW. Effects of stress on heal-ing wounds: I. Intermittent noncyclical tension.J Surg Res. 1976;20:93-102.Arslan S, Celiker R. Comparison of the efficacyof local corticosteroid injection and physicaltherapy for the treatment of adhesive capsuli-tis. Rheumatol Int. 2001;21:20-23.Aydeniz A, Gursoy S, Guney E. Which muscu-loskeletal complications are most frequentlyseen in type 2 diabetes mellitus? J Int Med Res.2008;36:505-511.Balci N, Balci MK, Tuzuner S. Shoulder adhesivecapsulitis and shoulder range of motion in typeII diabetes mellitus: association with diabeticcomplications. J Diabetes Complications.1999;13:135-140.Beaufils P, Prevot N, Boyer T, et al. Arthroscopicrelease of the glenohumeral joint in shoulderstiffness: a review of 26 cases. French Societyfor Arthroscopy. Arthroscopy. 1999;15:49-55.
8. Binder AI, Bulgen DY, Hazleman BL, Roberts S.Frozen shoulder: a long-term prospective study.Ann Rheum Dis. 1984;43:361-364.
9. Binder AI, Bulgen DY, Hazleman BL, Tudor J,Wraight P. Frozen shoulder: an arthrographicand radionuclear scan assessment. Ann RheumDis. 1984;43:365-369.
10. Binder AI, Hazleman BL, Parr G, Roberts S. Acontrolled study of oral prednisolone in frozenshoulder. Br J Rheumatol. 1986;25:288-292.
11. Boyles RE, Flynn TW, Whitman JM. Manipula-tion following regional interscalene anestheticblock for shoulder adhesive capsulitis: a caseseries. Man Ther. 2005;10:80-87. http://dx.doi.org/10.1016/j.math.2004.05.002
12. Brand PW. The forces of dynamic splinting: tenquestions before applying a dynamic splintto the hand. In: Hunter JM, Mackin EJ, Cal-lahan AD, eds. Rehabilitation of the Hand. C. V.Mosby; 1995:1581-1587.Bridgman JF. Periarthritis of the shoulder anddiabetes mellitus. Ann Rheum Dis. 1972;31:69-71.Bruckner FE, Nye CJ. A prospective study ofadhesive capsulitis of the shoulder (“frozenshoulder’) in a high risk population. Q J Med.1981;50:191-204.Bulgen DY, Binder AI, Hazleman BL, Dutton J,Roberts S. Frozen shoulder: prospective clinicalstudy with an evaluation of three treatmentregimens. Ann Rheum Dis. 1984;43:353-360.Bulgen DY, Binder AI, Hazleman BL, Park JR.
20. Carette S, Moffet H, Tardif J, et al. Intraarticularcorticosteroids, supervised physiotherapy, ora combination of the two in the treatment ofadhesive capsulitis of the shoulder: a placebo-controlled trial. Arthritis Rheum. 2003;48:829-838. http://dx.doi.org/10.1002/art.10954
21. Clarke GR, Willis LA, Fish WW, Nichols PJ. Pre-liminary studies in measuring range of motionin normal and painful stiff shoulders. Rheuma-tol Rehabil. 1975;14:39-46.
22. Codman EA. The Shoulder: Rupture of theSupraspinatus Tendon and Other Lesions in orAbout the Subacromial Bursa. Boston, MA: T.Todd Co; 1934.Constant CR, Murley AH. A clinical method offunctional assessment of the shoulder. ClinOrthop Relat Res. 1987;160-164.Cyriax J. Textbook of Orthopaedic Medicine:Diagnosis of Soft Tissue Lesions. 7th ed. Balti-more: Williams & Wilkins; 1978.Dacre JE, Beeney N, Scott DL. Injections andphysiotherapy for the painful stiff shoulder. AnnRheum Dis. 1989;48:322-325.Dahan TH, Fortin L, Pelletier M, Petit M,Vadeboncoeur R, Suissa S. Double blind ran-domized clinical trial examining the efficacy ofbupivacaine suprascapular nerve blocks in fro-zen shoulder. J Rheumatol. 2000;27:1464-1469.
27. DePalma A. Surgery of the Shoulder. Philadel-phia, PA: Lippincott; 1973.
28. Diercks RL, Stevens M. Gentle thawing of thefrozen shoulder: a prospective study of super-vised neglect versus intensive physical therapyin seventy-seven patients with frozen shouldersyndrome followed up for two years. J ShoulderElbow Surg. 2004;13:499-502. http://dx.doi.org/10.1016/S1058274604000825
29. Diwan DB, Murrell GA. An evaluation of theeffects of the extent of capsular release andof postoperative therapy on the temporaloutcomes of adhesive capsulitis. Arthroscopy.2005;21:1105-1113. http://dx.doi.org/10.1016/j.arthro.2005.05.014Dodenhoff RM, Levy O, Wilson A, Copeland SA.Manipulation under anesthesia for primaryfrozen shoulder: effect on early recovery andreturn to activity. J Shoulder Elbow Surg.2000;9:23-26.Donatelli RA, Greenfield B. Case study: rehabili-tation of a stiff and painful shoulder: a biome-chanical approach. J Orthop Sports Phys Ther.1987;9:118-126.Duke O, Zecler E, Grahame R. Anti-inflamma-tory drugs in periarthritis of the shoulder: a
double-blind, between-patient study of naprox-en versus indomethacin. Rheumatol Rehabil.1981;20:54-59.Eustace JA, Brophy DP, Gibney RP, BresnihanB, FitzGerald O. Comparison of the accuracyof steroid placement with clinical outcome inpatients with shoulder symptoms. Ann RheumDis. 1997;56:59-63.Flowers KR, LaStayo P. Effect of total end rangetime on improving passive range of motion. JHand Ther. 1994;7:150-157.Frank C, Amiel D, Woo SL, Akeson W. Normalligament properties and ligament healing. ClinOrthop Relat Res. 1985;15-25.Frost HM. Skeletal structural adaptations tomechanical usage (SATMU): 4. Mechanicalinfluences on intact fibrous tissues. Anat Rec.1990;226:433-439. http://dx.doi.org/10.1002/ar.1092260405Gam AN, Schydlowsky P, Rossel I, Remvig L,Jensen EM. Treatment of “frozen shoulder” withdistension and glucorticoid compared with glu-corticoid alone. A randomised controlled trial.Scand J Rheumatol. 1998;27:425-430.Gerber C, Espinosa N, Perren TG. Arthroscopictreatment of shoulder stiffness. Clin OrthopRelat Res. 2001;119-128.Griggs SM, Ahn A, Green A. Idiopathic adhesivecapsulitis. A prospective functional outcomestudy of nonoperative treatment. J Bone JointSurg Am. 2000;82-A:1398-1407.Gursel YK, Ulus Y, Bilgic A, Dincer G, van derHeijden GJ. Adding ultrasound in the manage-ment of soft tissue disorders of the shoulder: arandomized placebo-controlled trial. Phys Ther.2004;84:336-343.Haggart GE, Dignam RJ, Sullivan TS. Manage-ment of the frozen shoulder. J Am Med Assoc.1956;161:1219-1222.Hand GC, Athanasou NA, Matthews T, CarrAJ. The pathology of frozen shoulder. J BoneJoint Surg Br. 2007;89:928-932. http://dx.doi.org/10.1302/0301-620X.89B7.19097Hannafin JA, Chiaia TA. Adhesive capsulitis.A treatment approach. Clin Orthop Relat Res.2000;95-109.Harmon PH. Methods and results in the treat-ment of 2,580 painful shoulders, with specialreference to calcific tendinitis and the frozenshoulder. Am J Surg. 1958;95:527-544.Harryman DT, 2nd, Sidles JA, Harris SL, MatsenFA, 3rd. The role of the rotator interval capsulein passive motion and stability of the shoulder.J Bone Joint Surg Am. 1992;74:53-66.Hazleman BL. The painful stiff shoulder. Rheu-matol Phys Med. 1972;11:413-421.Helbig B, Wagner P, Dohler R. Mobilization offrozen shoulder under general anaesthesia.Acta Orthop Belg. 1983;49:267-274.Hill JJ, Jr., Bogumill H. Manipulation in thetreatment of frozen shoulder. Orthopedics.1988;11:1255-1260.Hutchinson JW, Tierney GM, Parsons SL, DavisTR. Dupuytren’s disease and frozen shoulderinduced by treatment with a matrix metal-
loproteinase inhibitor. J Bone Joint Surg Br.1998;80:907-908.Ide J, Takagi K. Early and long-term results ofarthroscopic treatment for shoulder stiffness. JShoulder Elbow Surg. 2004;13:174-179. http://dx.doi.org/10.1016/S1058274603002799Jarvinen TA, Jarvinen TL, Kaariainen M, KalimoH, Jarvinen M. Muscle injuries: biology andtreatment. Am J Sports Med. 2005;33:745-764.http://dx.doi.org/10.1177/0363546505274714Johnson AJ, Godges JJ, Zimmerman GJ,Ounanian LL. The effect of anterior versusposterior glide joint mobilization on externalrotation range of motion in patients with shoul-der adhesive capsulitis. J Orthop Sports PhysTher. 2007;37:88-99. http://dx.doi.org/10.2519/jospt.2007.2307Jurgel J, Rannama L, Gapeyeva H, ErelineJ, Kolts I, Paasuke M. Shoulder function inpatients with frozen shoulder before and after4-week rehabilitation. Medicina (Kaunas).2005;41:30-38.Kaltenborn FM, Robinson BE. Manual Therapyfor the Extremity Joints: Specialized Tech-niques, Tests and Joint-Mobilisation. Oslo: OlafNorlis Bokhandel; 1976.Kivimaki J, Pohjolainen T, Malmivaara A, etal. Manipulation under anesthesia with homeexercises versus home exercises alone in thetreatment of frozen shoulder: a randomized,controlled trial with 125 patients. J ShoulderElbow Surg. 2007;16:722-726. http://dx.doi.org/10.1016/j.jse.2007.02.125Lee JC, Guy S, Connell D, Saifuddin A, LambertS. MRI of the rotator interval of the shoulder.Clin Radiol. 2007;62:416-423. http://dx.doi.org/10.1016/j.crad.2006.11.017Lee JC, Sykes C, Saifuddin A, Connell D. Adhe-sive capsulitis: sonographic changes in the ro-tator cuff interval with arthroscopic correlation.Skeletal Radiol. 2005;34:522-527. http://dx.doi.org/10.1007/s00256-005-0957-0Lee M, Haq AM, Wright V, Longton EB. Peri-arthritis of the shoulder: a controlled trial ofphysiotherapy. Physiotherapy. 1973;59:312-315.Leggin B, Kelley MJ, Pontillo M. Impairmentsand Function in Patients with Frozen ShoulderCompared to Patients with Rotator Cuff Ten-donopathy. Second International Congress ofShoulder Therapists. Bahia, Brazil: 2007.Leggin BG, Michener LA, Shaffer MA, Bren-neman SK, Iannotti JP, Williams GR, Jr. ThePenn shoulder score: reliability and validity.J Orthop Sports Phys Ther. 2006;36:138-151.http://dx.doi.org/10.2519/jospt.2006.2090Levine WN, Kashyap CP, Bak SF, Ahmad CS,Blaine TA, Bigliani LU. Nonoperative manage-ment of idiopathic adhesive capsulitis. J Shoul-der Elbow Surg. 2007;16:569-573. http://dx.doi.org/10.1016/j.jse.2006.12.007Lippit S, Harryman DT, 2nd, Matsen FA, 3rd.A practical tool for evaluating function: thesimple shoulder test. In: Matsen FA, 3rd, Fu FH,Hawkins RJ, eds. The Shoulder: A Balance ofMobility and Stability. Park Ridge, IL: American
journal of orthopaedic & sports physical therapy | volume 39 | number 2 | february 2009 | 147
Academy of Orthopaedic Surgeons; 1993:Loew M, Heichel TO, Lehner B. Intraarticularlesions in primary frozen shoulder after ma-nipulation under general anesthesia. J ShoulderElbow Surg. 2005;14:16-21. http://dx.doi.org/10.1016/j.jse.2004.04.004Lundberg J. The frozen shoulder. Clinical andradiographical observations. The effect of ma-nipulation under general anesthesia. Structureand glycosaminoglycan content of the jointcapsule. Local bone metabolism. Acta OrthopScand. 1969;Suppl 119:111-159.Maitland GD. Peripheral Manipulation. London:Butterworths; 1977.McClure PW, Blackburn LG, Dusold C. The useof splints in the treatment of joint stiffness:biologic rationale and an algorithm for makingclinical decisions. Phys Ther. 1994;74:1101-1107.McClure PW, Flowers KR. Treatment of lim-ited shoulder motion: a case study based onbiomechanical considerations. Phys Ther.1992;72:929-936.McConnell S, Beaton D, Bombardier C. TheDASH Outcome Measure User’s Manual. To-ronto, CA: Institute for Work and Health; 1999.McLaughlin HL. Lesions of the musculotendi-nous cuff of the shoulder I. The exposure andtreatment of tears with retraction. J Bone JointSurg Am. 1944;26:31-51.
71. Milgrom C, Novack V, Weil Y, Jaber S, Radeva-Petrova DR, Finestone A. Risk factors foridiopathic frozen shoulder. Isr Med Assoc J.2008;10:361-364.
72. Miller MD, Wirth MA, Rockwood CA, Jr. Thaw-ing the frozen shoulder: the “patient” patient.Orthopedics. 1996;19:849-853.Morgan B, Jones AR, Mulcahy KA, FinlayDB, Collett B. Transcutaneous electric nervestimulation (TENS) during distension shoul-der arthrography: a controlled trial. Pain.1996;64:265-267.Mueller MJ, Maluf KS. Tissue adaptationto physical stress: a proposed “PhysicalStress Theory” to guide physical therapistpractice, education, and research. Phys Ther.2002;82:383-403.Mullett H, Byrne D, Colville J. Adhesive capsuli-tis: human fibroblast response to shoulder jointaspirate from patients with stage II disease. JShoulder Elbow Surg. 2007;16:290-294. http://dx.doi.org/10.1016/j.jse.2006.08.001Murnaghan JP. Adhesive capsulitis of the shoul-der: current concepts and treatment. Orthope-dics. 1988;11:153-158.
78. Neer CS, 2nd, Satterlee CC, Dalsey RM, FlatowEL. The anatomy and potential effects of con-tracture of the coracohumeral ligament. ClinOrthop Relat Res. 1992;182-185.
79. Neviaser JS. Adhesive capsulitis and the stiff
and painful shoulder. Orthop Clin North Am.1980;11:327-331.
80. Neviaser JS. Adhesive capsulitis of the shoul-der, a study of the pathological findings inperiarthritis of the shoulder. J Bone Joint SurgAm. 1945;27:211-222.
81. Neviaser JS. Arthrography of the shoulderjoint: study of the findings in adhesive cap-sulitis of the shoulder. J Bone Joint Surg Am.1962;44:1321-1330.
82. Neviaser RJ. Painful conditions affecting theshoulder. Clin Orthop Relat Res. 1983;63-69.Neviaser RJ, Neviaser TJ. The frozen shoulder.Diagnosis and management. Clin Orthop RelatRes. 1987;59-64.Nicholson GG. The effects of passive jointmobilization on pain and hypomobility associ-ated with adhesive capsulitis of the shoulder. JOrthop Sports Phys Ther. 1985;6:238-246.Ogilvie-Harris DJ, Biggs DJ, Fitsialos DP, Mac-Kay M. The resistant frozen shoulder. Manipula-tion versus arthroscopic release. Clin OrthopRelat Res. 1995;238-248.Omari A, Bunker TD. Open surgical releasefor frozen shoulder: surgical findings andresults of the release. J Shoulder Elbow Surg.2001;10:353-357. http://dx.doi.org/10.1067/mse.2001.115986
87. Ozaki J, Nakagawa Y, Sakurai G, Tamai S.Recalcitrant chronic adhesive capsulitis ofthe shoulder. Role of contracture of the cora-cohumeral ligament and rotator interval inpathogenesis and treatment. J Bone Joint SurgAm. 1989;71:1511-1515.
88. Pal B, Anderson J, Dick WC, Griffiths ID. Limita-tion of joint mobility and shoulder capsulitis ininsulin- and non-insulin-dependent diabetesmellitus. Br J Rheumatol. 1986;25:147-151.
89. Parker RD, Froimson AI, Winsberg DD, ArshamNZ. Frozen shoulder. Part II: Treatment bymanipulation under anesthesia. Orthopedics.1989;12:989-990.
90. Pearsall AWt, Holovacs TF, Speer KP. Theintra-articular component of the subscapularistendon: anatomic and histological correlationin reference to surgical release in patientswith frozen-shoulder syndrome. Arthroscopy.2000;16:236-242.
91. Pearsall AWt, Osbahr DC, Speer KP. An ar-throscopic technique for treating patients withfrozen shoulder. Arthroscopy. 1999;15:2-11.http://dx.doi.org/10.1053/ar.1999.v15.0150002
92. Placzek JD, Roubal PJ, Freeman DC, Kulig K,Nasser S, Pagett BT. Long-term effectiveness oftranslational manipulation for adhesive capsuli-tis. Clin Orthop Relat Res. 1998;181-191.Pollock RG, Duralde XA, Flatow EL, BiglianiLU. The use of arthroscopy in the treatment ofresistant frozen shoulder. Clin Orthop Relat Res.1994;30-36.Pouliart N, Somers K, Eid S, Gagey O.Variations in the superior capsuloligamentouscomplex and description of a new ligament. JShoulder Elbow Surg. 2007;16:821-836. http://dx.doi.org/10.1016/j.jse.2007.02.138
Quin CE. “Frozen Shoulder”. Evaluation of Treat-ment with Hydrocortisone Injections and Exer-cises. Ann Phys Med. 1965;10:22-25 PASSIM.Reeves B. The natural history of the frozenshoulder syndrome. Scand J Rheumatol.1975;4:193-196.
97. Rhind V, Downie WW, Bird HA, Wright V, EnglerC. Naproxen and indomethacin in periarthritisof the shoulder. Rheumatol Rehabil. 1982;21:51-53.
98. Richards RR, An K, Bigliani LU. A standardmethod for the assessment of shoulder func-tion. J Shoulder Elbow Surg. 1994;3:347-352.
99. Rizk TE, Christopher RP, Pinals RS, Higgins AC,Frix R. Adhesive capsulitis (frozen shoulder): anew approach to its management. Arch PhysMed Rehabil. 1983;64:29-33.
101. Rodeo SA, Hannafin JA, Tom J, Warren RF,Wickiewicz TL. Immunolocalization of cytokinesand their receptors in adhesive capsulitis ofthe shoulder. J Orthop Res. 1997;15:427-436.http://dx.doi.org/10.1002/jor.1100150316
102. Roubal PJ, Dobritt D, Placzek JD. Glenohumeralgliding manipulation following interscalenebrachial plexus block in patients with adhe-sive capsulitis. J Orthop Sports Phys Ther.1996;24:66-77.Rundquist PJ, Anderson DD, Guanche CA,Ludewig PM. Shoulder kinematics in subjectswith frozen shoulder. Arch Phys Med Rehabil.2003;84:1473-1479.Ryans I, Montgomery A, Galway R, KernohanWG, McKane R. A randomized controlled trialof intra-articular triamcinolone and/or physio-therapy in shoulder capsulitis. Rheumatology(Oxford). 2005;44:529-535. http://dx.doi.org/10.1093/rheumatology/keh535Safran MR, Garrett WE, Jr., Seaber AV, Glis-son RR, Ribbeck BM. The role of warmup inmuscular injury prevention. Am J Sports Med.1988;16:123-129.Segmuller HE, Taylor DE, Hogan CS, Saies AD,Hayes MG. Arthroscopic treatment of adhesivecapsulitis. J Shoulder Elbow Surg. 1995;4:403-408.
107. Shaffer B, Tibone JE, Kerlan RK. Frozen shoul-der. A long-term follow-up. J Bone Joint SurgAm. 1992;74:738-746.
108. Shah N, Lewis M. Shoulder adhesive capsulitis:systematic review of randomised trials usingmultiple corticosteroid injections. Br J GenPract. 2007;57:662-667.
148 | february 2009 | volume 39 | number 2 | journal of orthopaedic & sports physical therapy
[ CLINICAL COMMENTARY ]
@ MORE INFORMATIONWWW.JOSPT.ORG
tromyogr Clin Neurophysiol. 2007;47:205-213.112. Terry GC, Hammon D, France P, Norwood LA.
The stabilizing function of passive shoulderrestraints. Am J Sports Med. 1991;19:26-34.
113. Travell JG, Simons DG. Myofascial Pain andDysfunction: The Trigger Point Manual. Williams& Wilkins; 1983.
114. Tukmachi ES. Frozen shoulder: A comparisonof western and traditional chinese approachesand a clinical study of its acupuncture treat-ment. Acupuncture Med. 1999;17:9-22.
115. Uhthoff HK, Boileau P. Primary frozenshoulder: global capsular stiffness versuslocalized contracture. Clin Orthop Relat Res.2007;456:79-84. http://dx.doi.org/10.1097/BLO.0b013e318030846d
116. Uitvlugt G, Detrisac DA, Johnson LL, Austin MD,Johnson C. Arthroscopic observations beforeand after manipulation of frozen shoulder. Ar-throscopy. 1993;9:181-185.
117. van der Windt DA, Koes BW, Deville W, BoekeAJ, de Jong BA, Bouter LM. Effectiveness ofcorticosteroid injections versus physiotherapyfor treatment of painful stiff shoulder in primary
care: randomised trial. BMJ. 1998;317:1292-1296.
118. Vermeulen HM, Obermann WR, Burger BJ, KokGJ, Rozing PM, van Den Ende CH. End-rangemobilization techniques in adhesive capsulitisof the shoulder joint: A multiple-subject casereport. Phys Ther. 2000;80:1204-1213.
119. Vermeulen HM, Rozing PM, Obermann WR, leCessie S, Vliet Vlieland TP. Comparison of high-grade and low-grade mobilization techniquesin the management of adhesive capsulitis ofthe shoulder: randomized controlled trial. PhysTher. 2006;86:355-368.
120. Vermeulen HM, Stokdijk M, Eilers PH, MeskersCG, Rozing PM, Vliet Vlieland TP. Measurementof three dimensional shoulder movement pat-terns with an electromagnetic tracking devicein patients with a frozen shoulder. Ann RheumDis. 2002;61:115-120.
122. Warner JJ, Allen A, Marks PH, Wong P. Ar-throscopic release for chronic, refractory ad-hesive capsulitis of the shoulder. J Bone Joint
Surg Am. 1996;78:1808-1816.123. Watson L, Dalziel R, Story I. Frozen shoulder:
a 12-month clinical outcome trial. J ShoulderElbow Surg. 2000;9:16-22.
124. Wiley AM. Arthroscopic appearance of frozenshoulder. Arthroscopy. 1991;7:138-143.
125. Woo SL, Buckwalter JA. Injury and Repair of theMusculoskeletal Soft Tissues. American Acad-emy of Orthopaedic Surgeons; 1988.
126. Wyke B. The neurology of joints. Ann R CollSurg Engl. 1967;41:25-50.
127. Yang JL, Chang CW, Chen SY, Wang SF, Lin JJ.Mobilization techniques in subjects with frozenshoulder syndrome: randomized multiple-treatment trial. Phys Ther. 2007;87:1307-1315.http://dx.doi.org/10.2522/ptj.20060295
JOSPT subscribers and APTA members of the Orthopaedic and SportsPhysical Therapy Sections can help the environment by “opting out” ofreceiving the Journal in print each month as follows. If you are:
· A JOSPT subscriber: Email your request to [email protected] or call the Journal office toll-free at 1-877-766-3450 and provide your name and subscriber number.· An APTA Orthopaedic or Sports Section member: Go to www.apta.org
and update your preferences in the My Profile area of myAPTA. Select “myAPTA” from the horizontal navigation menu (you’ll be asked to login, if you haven’t already done so), then proceed to “My Profile.” Click on the “Email & Publications” tab, choose your “opt out” preferences and save.
Subscribers and members alike will continue to have access to JOSPTonline and can retrieve current and archived issues anytime and anywhereyou have Internet access.
1. Mark W. ButlerCommon Shoulder Diagnoses 219-255. [CrossRef]2. Rebecca Lowe. 2013. Adhesive capsulitis of the hip: A case report: An entity in question. Manual Therapy 18:6, 594-597.
[CrossRef]3. Martin J. Kelley, Michael A. Shaffer, John E. Kuhn, Lori A. Michener, Amee L. Seitz, Tim L. Uhl, Joseph J. Godges, Philip
4. Anthony Delitto, Steven Z. George, Linda Van Dillen, Julie M. Whitman, Gwendolyn Sowa, Paul Shekelle, Thomas R. Denninger,Joseph J. Godges. 2012. Low Back Pain. Journal of Orthopaedic & Sports Physical Therapy 42:4, A1-A57. [Citation] [Full Text][PDF] [PDF Plus]
5. Katherine BeissnerConservative pain management for the older adult 395-411. [CrossRef]6. Jason B. Lunden, Mike Muffenbier, M. Russell Giveans, Cort J. Cieminski. 2010. Reliability of Shoulder Internal Rotation Passive
Range of Motion Measurements in the Supine Versus Sidelying Position. Journal of Orthopaedic & Sports Physical Therapy 40:9,589-594. [Abstract] [Full Text] [PDF] [PDF Plus]
7. Angela R. Tate, Philip W. McClure, Ian A. Young, Renata Salvatori, Lori A. Michener. 2010. Comprehensive Impairment-BasedExercise and Manual Therapy Intervention for Patients With Subacromial Impingement Syndrome: A Case Series. Journal ofOrthopaedic & Sports Physical Therapy 40:8, 474-493. [Abstract] [Full Text] [PDF] [PDF Plus]
