Is the Frozen Shoulder Classification a Reliable Assessment?

Article information

Clin Shoulder Elb. 2018;21(2):82-86
Publication date (electronic) : 2018 June 1
doi : https://doi.org/10.5397/cise.2018.21.2.82
1Department of Orthopaedic Surgery, Gyeongsang National University Changwon Hospital, Changwon, Korea
2Department of Orthopedic Surgery, All India Institute of Medical Science, Jodhpur, Rajasthan, India
3Department of Orthopaedic Surgery, Gyeongsang National University School of Medicine, Changwon, Korea
Correspondence to: Hyung Bin Park Department of Orthopaedic Surgery, Gyeongsang National University Changwon Hospital, Gyeongsang National University School of Medicine, 11 Samjeongja-ro, Seongsan-gu, Changwon 51472, Korea Tel: +82-55-214-3744, Fax: +82-55-214-3259, E-mail: hbinpark@gnu.ac.kr
Received 2017 December 7; Revised 2018 February 13; Accepted 2018 March 1.

Abstract

Background

Although a common shoulder disease, there are no accepted classification criteria for frozen shoulder (FS). This study therefore aimed to evaluate the accuracy of the conventionally used FS classification system.

Methods

Primary FS patients (n=168) who visited our clinic from January 2010 to July 2015 were included in the study. After confirming restrictions of the glenohumeral joint motion and absence of history of systemic disease, trauma, shoulder surgery, shoulder muscle weakness, or specific x-ray abnormalities, the Zuckerman and Rokito’s classification was employed for diagnosing primary FS. Following clinical diagnosis, each patient underwent a shoulder magnetic resonance imaging (MRI) and blood tests (lipid profile, glucose, hemoglobin A1c, and thyroid function). Based on the results of the blood tests and MRIs, the patients were reclassified, using the criteria proposed by Zuckerman and Rokito.

Results

New diagnoses were ascertained including blood test results (16 patients with diabetes, 43 with thyroid abnormalities, and 149 with dyslipidemia), and MRI revealed intra-articular lesions in 81 patients (48.2%). After re-categorization based on the above findings, only 5 patients (3.0%) were classified having primary FS. The remaining 163 patients (97.0%) had either undiagnosed systemic or intrinsic abnormalities (89 patients), whereas 74 patients had both.

Conclusions

These findings demonstrate that most patients clinically diagnosed with primary FS had undiagnosed systemic abnormalities and/or intra-articular pathologies. Therefore, a modification of the Zuckerman and Rokito’s classification system for FS may be required to include the frequent combinations, rather than having a separate representation of systemic abnormalities and intrinsic causes.

Introduction

The definition of frozen shoulder (FS) remains largely unchanged since it was first expounded in 1934 by Codman [1]. He described FS as a painful disease characterized by slow onset, restricted movements at the shoulder joint, and a grossly normal radiograph, and considered it to be a self-resolving disorder of unknown etiology [1]. Codman’s description of FS as a disease of uncertain etiology has been supported by the American Academy of Orthopaedic Surgery [2], which defines primary FS as a condition of uncertain etiology, characterized by restriction of both active and passive glenohumeral ranges of movement in the absence of any underlying causes but having normal radiographs [2,3]. However, investigators realized later that many cases of FS have underlying causes; Lundberg [4] coined the term secondary FS to refer to such cases. Zuckerman and Rokito [3] proposed a modification to the original Lundburg classification based on the etiology, and the secondary FS were sub-classified as intrinsic, extrinsic, and systemic subtypes.

However, debates still exist regarding diagnostic criteria and classification for FS. A uniform and accurate classification is mandatory for any disease to characterize the nature of the disorder and guide the treatment. Furthermore, the prognosis of the disease progression needs to be anticipated, so that consistent reporting of the treatment and outcome is possible. This allows for accurate comparison of outcomes from different studies [5]. A good classification system requires good inter- and intra-observer reliability, and enough validity to correctly describe the etiology. However, the validity of previously reported classifications for FS have not been completely evaluated [5]. The Zuckerman and Rokito’s FS classification system [3] is based on clinical examination, history and radiographs used to identify the etiology. Their diagnostic criteria, which have traditionally been used in clinical practices, are unable to fully diagnose the intrinsic or systemic causes, since simple radiographs and medical history are insufficient to accurately diagnose these factors.

We hypothesized that most primary FS, as classified by Zuckerman and Rokito’s FS criteria, are actually secondary FS, but cannot be identified with the traditional diagnostic methods of clinical examination. Therefore, the current study was undertaken to determine the accuracy of Zuckerman and Rokito’s FS classification using blood tests and shoulder magnetic resonance imaging (MRI).

Methods

This study was approved by the institutional review board of the Gyeongsang National University Hospital (GNUH 2015-05-013). Medical records of patients diagnosed with primary FS from January 2010 to July 2015 were retrospectively reviewed. Of the 465 patients reviewed, we excluded 70 patients who did not have the result of shoulder MRI, 62 patients who did not have laboratory results, 53 patients who did not have results of the physical examination, and 112 patients who had previous history of shoulder surgery, trauma, and systemic disease. The remaining 168 patients included in the study underwent blood tests (lipid profile, glucose level, glycosylated hemoglobin A1c [HbA1c], and thyroid function tests) and a shoulder MRI.

The initial diagnosis of primary FS was based on observations of the clinical examination which showed restriction in both the active and passive glenohumeral movements during flexion, abduction, and internal rotation (associated with >50% decrease in external rotation with arm at side), and on the basis of normal radiographic findings of the affected shoulders in true anteroposterior, outlet, and axillary lateral views. Additionally, the diagnosis was based on a medical history of no underlying disease, systemic abnormality, shoulder surgery, or shoulder trauma. All the clinical assessments were carried out by the senior author (HBP).

The blood test results were analyzed in accordance with the standard criteria established for diagnosis of the respective diseases. Dyslipidemia was defined when any of following criteria of lipid profiles was positive: hypercholesterolemia (cholesterol≥200 mg/dl), hyper-low-density lipoproteinemia (≥100 mg/dl), hyper-triglyceridemia (≥150 mg/dl), hypo-high-density lipoproteinemia (HDL≤40 mg/dl in male, ≤50 mg/dl in female), and hyper-non-HDLemia (non-HDL≥130 mg/dl) [6]. Diabetes was diagnosed when plasma levels of HbA1c were >6.4%, fasting plasma glucose was >125 mg/dl, or plasma glucose>199 mg/dl after two hours of a 75 g oral glucose load [7]. Hyper- and hypo-thyroidism were based on the results of thyroid function tests, in which the serum free T4 levels>1.70 ng/dl indicated hyper-thyroidism, and <0.93 ng/dl indicated hypo-thyroidism [8].

Most patients (142/168, 84.5%) underwent MRIs at our institute, with a 1.5 T scanner (Siemens Medical Systems, Erlangen, Germany); the remaining 26 patients performed MRIs outside our institute. All the MRI images included in this study, whether performed at our institute or elsewhere, were interpreted by a single experienced musculoskeletal radiologist who was blind to the clinical findings. All but 5 patients had their MRI examinations within two months of the outpatient visit. After compiling the results of the blood tests and MRI findings, we re-categorized the patients into appropriate sub-classifications, as proposed by Zuckerman and Rokito [3]. We used the IBM SPSS Statistics ver. 21 Developer software (IBM Co., Armonk, NY, USA) to perform the frequency analysis of the data and to calculate the distribution of patients in the various sub-classifications.

Results

A total of 168 patients, who were initially diagnosed with primary FS and who met the aforementioned inclusion criteria, were enrolled in this study. These included 66 males (39.3%) and 102 females (60.7%), with an average age of 53.5 ± 8.3 years. The right shoulders of 91 patients (54.2%) were affected, and the left shoulders of 77 patients (45.8%) (Table 1).

Summary of Demographic Data (Gender and Systemic Disease after Blood Investigations)

Based on the analyses of the blood tests, the newly diagnosed afflictions were 16 cases of diabetes (9.5%), 43 cases of thyroid abnormalities (25.6%) (4 hyper-thyroidism and 39 hypothyroidism), and 149 cases of dyslipidemia (88.7%). A total of 156 patients (92.9%) were found to have one or more of the above systemic abnormalities, with dyslipidemia being the most common (149/168, 88.7%). Dyslipidemia was present in 68.8% (11/16) of the diabetic patients, and in 88.4% (38/43) of patients with thyroid dysfunction (Table 2).

Cases of Various Causes of Systemic Frozen Shoulder as Assessed after Blood Investigations

MRI examinations revealed 81 patients (48.2%) had intraarticular lesions of the shoulder joint, with the most commonly found lesion being a tear of the supraspinatus tendon present in 53 patients (31.5%). The spectrum of lesions detected on MRI is summarized in Table 3.

Lesions Observed in the MRI Findings of the Affected Shoulders

After compilation of the results and reclassification according to the Zuckerman and Rokito’s classification [3], only 5 patients (3.0%) were classified as primary FS. The remaining 163 patients (97.0%) had previously undiagnosed systemic and/or intrinsic abnormalities. Of the 163 newly diagnosed secondary FS patients, 74 (45.4%) had both intrinsic lesions and systemic abnormalities (Table 4).

Classification of Patients Based on Clinical and Radiographic Evaluations, with and without Considering Blood Tests and MRI Findings

Discussion

This study was undertaken to determine the accuracy of the Zuckerman and Rokito’s FS classification [3], after confirming blood tests and shoulder MRI outcomes. In accordance with our hypothesis, the results revealed that most of the primary FS were reclassified as secondary FS, having intrinsic lesions and/or systemic disease.

Zuckerman and Rokito [3] differentiated secondary FS from primary FS using systemic, intrinsic, and extrinsic factors. The secondary FS group was later expanded by Kelley et al. [9] and Nash and Hazleman [10] to include diabetes, myocardial infarction, and other neurological disorders that were associated with FS. Robinson et al. [11] used a different classification system, wherein they divided the primary FS into idiopathic and systemic diseases. They separated diabetic FS from secondary FS, due to the former’s prognosis being worse than that of other afflictions, and the incidence of FS being high in diabetes. Among the traditional definitions and diagnostic methods, Zuckerman and Rokito’s classification [3] (an etiology-based classification) is simple and helpful to diagnose FS. However, this system is not universally accepted among shoulder surgeons; when polled to determine their opinions of the classification system, 34% of the respondents expressed either disapproval or no opinion regarding appropriateness of secondary FS sub-classification. In our patient group, this classification led to a gross over-diagnosis of primary FS, and was not applicable to many patients in our study because of overlapping etiologies of combined systemic and intrinsic abnormalities. We found that physical examination, simple radiological evaluation, and history evaluation was insufficient to identify 93% of our patients, who were eventually diagnosed to have systemic abnormalities. Therefore, we suggest some modifications to the currents FS classification system to accommodate the possibilities of simultaneous lesions and multiple etiologies.

Among the various systemic risk factors known to be associated with FS, diabetes [12,13], hyper-thyroidism [14], hypo-thyroidism [15], and dyslipidemia [16,17] have a relatively high prevalence, and can be easily diagnosed by blood tests. Early diagnosis of diabetes permits early intervention, which helps reduce the progressive stiffness [18,19], worsening of shoulder pain, and disability associated with poor glycemic control [20,21]. Similarly, thyroidectomy and normalization of thyroid hormone levels are reported to resolve shoulder stiffness [14,22]. Patients enrolled in the current study were initially diagnosed with primary FS, but blood test diagnosed 16 patients (9.5%) with diabetes, 43 patients (25.6%) with thyroid abnormalities, and 149 (88.7%) with dyslipidemia. This suggests that most patients clinically diagnosed with primary FS probably have undiagnosed systemic abnormalities and have lost the opportunity for early intervention. In view of this wealth of evidence, we believe blood tests to be essential for the initial diagnostic evaluation of FS to enable optimal early intervention.

In the current study, any intra-articular pathologies detected on MRI were considered to be intrinsic causes of FS [3]. Of the few intra-articular lesions found, the most notable were the supraspinatus tear, followed by the subscapularis tear and the SLAP lesion. Although literature reports various MRI findings as factors associated with FS, there is no consensus as to whether MRI-detected intra-articular lesions are instrumental in causing FS [23,24]. In a study using MR arthrography of primary FS patients, Yoo et al. [25] reported findings similar to those of the current study: 61.7% of their patients had supraspinatus tendon pathologies and 40% had rotator cuff tears. Since most of the lesions are commonly found in aged shoulders [26-28], it remains unclear whether they are causative factors of FS. With many authors advocating against routine use of MRI [29,30], no consensus has emerged regarding the inclusion of MRI for initial evaluations of FS. Longitudinal follow-up studies are required to determine whether MRI-detected intrinsic lesions are causative factors of FS. The current study found that 74 patients (44.0%) had both intrinsic lesions detected on MRI and systemic abnormalities found in blood tests, a circumstance not addressed by the Zuckerman and Rokito’s classification [3].

This study has several limitations. First, the evaluation was confined to three systemic disease entities and did not include other systemic factors which are known to be associated with FS, for example, adrenocorticotropic hormone deficiency. Second, this is a cross-sectional observation study; hence, it was not possible to identify the causative relationships between various factors and FS, particularly whether FS is merely age-related or whether MRI-detected intrinsic lesions trigger FS. Third, because we only included patients initially diagnosed with primary FS based on clinical findings, we were unable to evaluate any associations between systemic causes and extrinsic causes. These limitations need to be addressed in future studies.

Conclusion

Findings of the current study demonstrate that most patients who were clinically diagnosed with primary FS had undiagnosed systemic abnormalities and/or intra-articular pathologies. We suggest a modification of the Zuckerman and Rokito’s classification system for FS to include frequent combinations, rather than a separate presentation of systemic abnormalities and intrinsic causes.

Notes

Research Ethics

IRB approval: Gyeongsang National University Hospital (No. GNUH 2015-05-013).

Financial support

None.

Conflict of interest

None.

References

1. Codman EA. The shoulder: rupture of the supraspinatus tendon and other lesions in or about the subacromial bursa Boston (MA): T. Todd Company; 1934. liii, 51, 29.
2. Zuckerman JD, Cuomo FC. Frozen shoulder. In : Matsen FA III, Fu FH, Hawkins RJ, eds. The shoulder: a balance of mobility and stability Rosemont (IL): American Academy of Orthopaedic Surgery; 1993. p. 253–67.
3. Zuckerman JD, Rokito A. Frozen shoulder: a consensus definition. J Shoulder Elbow Surg 2011;20(2):322–5.
4. Lundberg BJ. The frozen shoulder. Clinical and radiographical observations. The effect of manipulation under general anesthesia. Structure and glycosaminoglycan content of the joint capsule. Local bone metabolism. Acta Orthop Scand Suppl 1969;119:1–59.
5. Garbuz DS, Masri BA, Esdaile J, Duncan CP. Classification systems in orthopaedics. J Am Acad Orthop Surg 2002;10(4):290–7.
6. Grundy SM, Cleeman JI, Merz CN, et al. Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation 2004;110(2):227–39.
7. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2010;33(Suppl 1):S62–9.
8. Baskin HJ, Cobin RH, Duick DS, et al. American Association of Clinical Endocrinologists medical guidelines for clinical practice for the evaluation and treatment of hyperthyroidism and hypothyroidism. Endocr Pract 2002;8(6):457–69.
9. Kelley MJ, McClure PW, Leggin BG. Frozen shoulder: evidence and a proposed model guiding rehabilitation. J Orthop Sports Phys Ther 2009;39(2):135–48.
10. Nash P, Hazleman BL. Frozen shoulder. Baillieres Clin Rheumatol 1989;3(3):551–66.
11. Robinson CM, Seah KT, Chee YH, Hindle P, Murray IR. Frozen shoulder. J Bone Joint Surg Br 2012;94(1):1–9.
12. Arkkila PE, Kantola IM, Viikari JS, Rönnemaa T. Shoulder capsulitis in type I and II diabetic patients: association with diabetic complications and related diseases. Ann Rheum Dis 1996;55(12):907–14.
13. Thomas SJ, McDougall C, Brown ID, et al. Prevalence of symptoms and signs of shoulder problems in people with diabetes mellitus. J Shoulder Elbow Surg 2007;16(6):748–51.
14. Wohlgethan JR. Frozen shoulder in hyperthyroidism. Arthritis Rheum 1987;30(8):936–9.
15. Bowman CA, Jeffcoate WJ, Pattrick M, Doherty M. Bilateral adhesive capsulitis, oligoarthritis and proximal myopathy as presentation of hypothyroidism. Br J Rheumatol 1988;27(1):62–4.
16. Bunker TD, Esler CN. Frozen shoulder and lipids. J Bone Joint Surg Br 1995;77(5):684–6.
17. Hand GC, Athanasou NA, Matthews T, Carr AJ. The pathology of frozen shoulder. J Bone Joint Surg Br 2007;89(7):928–32.
18. Sattar MA, Luqman WA. Periarthritis: another duration-related complication of diabetes mellitus. Diabetes Care 1985;8(5):507–10.
19. Ogilvie-Harris DJ, Biggs DJ, Fitsialos DP, MacKay M. The resistant frozen shoulder. Manipulation versus arthroscopic release. Clin Orthop Relat Res 1995;(319):238–48.
20. Ramchurn N, Mashamba C, Leitch E, et al. Upper limb musculoskeletal abnormalities and poor metabolic control in diabetes. Eur J Intern Med 2009;20(7):718–21.
21. Laslett LL, Burnet SP, Redmond CL, McNeil JD. Predictors of shoulder pain and shoulder disability after one year in diabetic outpatients. Rheumatology (Oxford) 2008;47(10):1583–6.
22. Oldham BE. Periarthritis of the shoulder associated with thyrotoxicosis. A report of five cases. N Z Med J 1959;58:766–70.
23. Mengiardi B, Pfirrmann CW, Gerber C, Hodler J, Zanetti M. Frozen shoulder: MR arthrographic findings. Radiology 2004;233(2):486–92.
24. Sofka CM, Ciavarra GA, Hannafin JA, Cordasco FA, Potter HG. Magnetic resonance imaging of adhesive capsulitis: correlation with clinical staging. HSS J 2008;4(2):164–9.
25. Yoo JC, Ahn JH, Lee YS, Koh KH. Magnetic resonance arthrographic findings of presumed stage-2 adhesive capsulitis: focus on combined rotator cuff pathology. Orthopedics 2009;32(1):22.
26. Pfahler M, Haraida S, Schulz C, et al. Age-related changes of the glenoid labrum in normal shoulders. J Shoulder Elbow Surg 2003;12(1):40–52.
27. Sher JS, Uribe JW, Posada A, Murphy BJ, Zlatkin MB. Abnormal findings on magnetic resonance images of asymptomatic shoulders. J Bone Joint Surg Am 1995;77(1):10–5.
28. Yamamoto A, Takagishi K, Osawa T, et al. Prevalence and risk factors of a rotator cuff tear in the general population. J Shoulder Elbow Surg 2010;19(1):116–20.
29. Manton GL, Schweitzer ME, Weishaupt D, Karasick D. Utility of MR arthrography in the diagnosis of adhesive capsulitis. Skeletal Radiol 2001;30(6):326–30.
30. Enders NK, ElHassan B, Higgins LD, Warner JJP. The stiff shoulder. In : Rockwood CA Jr, Matsen FA III, Wirth MA, Lippitt SB, Fehringer EV, Sperling JW, eds. The shoulder 4th edth ed. Philadelphia: Saunders Elsevier; 2009. p. 1405–36.

Article information Continued

Table 1.

Summary of Demographic Data (Gender and Systemic Disease after Blood Investigations)

Enrolled subject Percentage Mean age (yr)
Total enrolled subjects (n=168) 53.5 ± 8.3
 Male 39.3 (66/168) 54.8 ± 5.9
  Diabetes 7.6 (5/66) 55.3 ± 6.2
  Abnormal thyroid function 18.2 (12/66) 54.2 ± 5.1
  Dyslipidemia 83.3 (55/66) 54.3 ± 8.1
 Female 60.7 (102/168) 52.4 ± 6.8
  Diabetes 10.8 (11/102) 54.2 ± 7.1
  Abnormal thyroid function 30.4 (31/102) 53.8 ± 6.2
  Dyslipidemia 92.2 (94/102) 53.0 ± 5.3
Affected side
 Right 54.2 (91/168) 54.2 ± 7.2
 Left 45.8 (77/168) 53.0 ± 5.3

Values are presented as percent (number/total number) or mean ± standard deviation.

Table 2.

Cases of Various Causes of Systemic Frozen Shoulder as Assessed after Blood Investigations

Variable Diabetes Abnormal thyroid function Dyslipidemia All three abnormalities
Diabetes 2 3 9 -
Abnormal thyroid function 2 36 -
Dyslipidemia 102 -
All three abnormalities 2

Table 3.

Lesions Observed in the MRI Findings of the Affected Shoulders

MRI finding Specific lesions Percentage
Intra-articular lesion 48.2 (81/168)
Supraspinatus lesion Articular side partial tear 13.7 (23/168)
Bursal side partial tear 4.8 (8/168)
Interstitial partial tear 12.5 (21/168)
Full thickness tear 0.6 (1/168)
Subscapularis lesion Articular side partial tear 8.3 (14/168)
SLAP lesion 6.0 (10/168)
Biceps tendon lesion 1.8 (3/168)
Subscapularis partial tear and SLAP lesion 0.6 (1/168)
Negative MRI finding 51.8 (87/168)

Values are presented as percent (number/total number).

MRI: magnetic resonance imaging, SLAP: superior labral tear from anterior to posterior.

Table 4.

Classification of Patients Based on Clinical and Radiographic Evaluations, with and without Considering Blood Tests and MRI Findings

Variable Group classification
Without considering blood tests and MRI findings After including blood tests and MRI findings
Primary frozen shoulder 168 5 (3.0)
Secondary frozen shoulder 163 (97.0)
Systemic cause 156 (92.9)
Intrinsic cause 81 (48.2)

Values are presented as number only or number (%).

MRI: magnetic resonance imaging.