Rotator Cuff Lesions in Patients with Stiff Shoulders

Transcription

1 1233 COPYRIGHT Ó 2015 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED Rotator Cuff Lesions in Patients with Stiff Shoulers A Prospective Analysis of 379 Shoulers Yusuke Uea, MD, Hiroyuki Sugaya, MD, Norimasa Takahashi, MD, Keisuke Matsuki, MD, Nobuaki Kawai, MD, Morihito Tokai, MD, Kazutomo Onishi, MD, an Shota Hoshika, MD Investigation performe at the Shouler & Elbow Center, Funabashi Orthopaeic Hospital, Funabashi, Chiba, Japan Backgroun: Iiopathic ahesive capsulitis is efine as a frozen shouler with severe an global range-of-motion loss of unknown etiology. The purpose of our stuy was to clarify the prevalence of rotator cuff lesions accoring to patterns an severity of range-of-motion loss in a large cohort of patients with stiff shoulers. Methos: Rotator cuff pathology was prospectively investigate with use of magnetic resonance imaging (MRI) or ultrasonography in a series of 379 stiff shoulers; patients with traumatic etiology, iabetes, or raiographic abnormalities were exclue. Eighty-nine shoulers emonstrate severe an global loss of passive motion ( 100 of forwar flexion, 10 of external rotation with the arm at the sie, an internal rotation not more cephala than the L5 level) an were classifie as having severe an global loss of motion (Group 1). The remaining 290 shoulers were ivie into two groups: those with severe but not global loss (Group 2; 111 shoulers) an mil to moerate limitation (Group 3; 179 shoulers). Results: Among all shoulers, imaging emonstrate an intact rotator cuff in 51%, a full-thickness tear in 35%, an a partial-thickness tear in 15%. In Group 1, 91% ha an intact rotator cuff an 9% ha a partial-thickness rotator cuff tear. No patient in this group emonstrate a full-thickness tear. In Group 2 an Group 3, respectively, 44% an 35% of the shoulers were intact, 17% an 16% ha a partial-thickness tear, an 39% an 50% ha a full-thickness tear. Conclusions: Shouler stiffness with severe an global loss of passive range of motion is not associate with fullthickness rotator cuff tears, although some patients may have a partial-thickness tear. Shoulers with severe an global loss of range of motion at a first visit are likely to be cases of iiopathic ahesive capsulitis an may not require further imaging stuies. Peer Review: This article was reviewe by the Eitor-in-Chief an one Deputy Eitor, an it unerwent bline review by two or more outsie experts. The Deputy Eitor reviewe each revision of the article, an it unerwent a final review by the Eitor-in-Chief prior to publication. Final corrections an clarifications occurre uring one or more exchanges between the author(s) an copyeitors. The term periarthritis was first use in 1872 to refer to shoulers that ha limite active an passive shouler motion 1. The term frozen shouler was introuce by Coman in Iiopathic ahesive capsulitis is now more narrowly efine as a frozen shouler with severe an global loss of range of motion (ROM) with unknown etiology 3-5. Despite this efinition an unerstaning, several authors recently reporte that rotator cuff lesions are strongly associate with frozen shouler, an some have even state that the rotator cuff lesion can be a cause of frozen shouler 5-8.However,because of the ambiguity of the efinition of a frozen shouler, the egree of ROM eficit often varies accoring to investigators 5,7-10. Robinson et al. efine frozen shouler as 90 of abuction an some limitation of internal an external rotation 5.Yoo et al. efine stage-2 ahesive capsulitis as 100 of forwar flexion, 10 of external rotation, an internal rotation not more cephala than the T12 level 7.Furthermore,Watsonetal. efine frozen shouler as 120 of abuction an 50% of Disclosure: None of the authors receive payments or services, either irectly or inirectly (i.e., via his or her institution), from a thir party in support of any aspect of this work. None of the authors, or their institution(s), have ha any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomeical arena that coul be perceive to influence or have the potential to influence what is written in this work. Also, no author has ha any other relationships, or has engage in any other activities, that coul be perceive to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitte by authors are always provie with the online version of the article. J Bone Joint Surg Am. 2015;97:

2 1234 TABLE I Range of Motion an Prevalence of Rotator Cuff Lesions by Group ä Severity of Motion Loss* Group FF ER IR Shoulers (no.) Male (no.) Female (no.) Age (yr) Group 1 (severe an global loss) S S S (35-80) Group 2 (severe but not global loss)# (38-83) S S M (41-76) S M S (43-70) M S S (44-77) S M M (53-78) M M S (38-83) M S M (45-77) Group 3 (mil to moerate limitation)# M M M (35-83) *FF = forwar flexion, ER = external rotation with arm at sie, IR = internal rotation, S = severe limitation ( 100 of FF, 10 of ER, an L5-level IR), an M = mil to moerate limitation ( 101 of FF, 11 of ER, an L4-level IR). Values are given as the mean with the range in parentheses. IR is inicate by the spine level reachable by the thumb; S = sacrum, B = buttock, an GT = greater trochanter. Values are given as the percentage with the number of shoulers in parentheses. PTT = partial-thickness tear, an FTT = full-thickness tear. Values are given as the mean with the range in parentheses. #A significant ifference in the prevalence of rotator cuff lesions compare with Group 1 was observe (p < 0.01, Steel- Dwass test). external rotation relative to the contralateral shouler 8.Thus, aconsensusonefinition is necessary when the pathology an etiology of frozen shouler are consiere. In a previous stuy, we investigate the characteristics of the ROM eficit in 155 patients with a full-thickness rotator cuff tear who unerwent surgical repair an foun that their ROM eficit was relatively mil compare with the group of patients with frozen shouler who unerwent arthroscopic capsular release for refractory severe an global loss of motion 11. Furthermore, although approximately one-thir of patients with a full-thickness rotator cuff tear emonstrate global loss of motion, there were no patients with severe loss of motion, which was efine as 100 of forwar flexion, 10 of external rotation with the arm at the sie, an internal rotation not more cephala than the L5 level 11. In the present stuy, we hypothesize that there woul be no full-thickness rotator cuff tears in stiff shoulers with severe an global loss of passive ROM. The purpose of this stuy was to investigate in a prospective fashion, using magnetic resonance imaging (MRI) or ultrasonography, the prevalence of rotator cuff lesions in patients with stiff shoulers, excluing patients with a traumatic etiology, iabetes, or raiographic abnormalities. Materials an Methos We receive institutional review boar approval prior to initiating this stuy an obtaine informe consent from all participating patients prior to imaging. Patient Selection an ROM Measurement Between June 2012 an July 2013, 2052 consecutive patients (2185 shoulers) thirty-five years of age or oler visite our institution with shouler pain associate with ROM eficit. Passive forwar flexion, external rotation with the arm at the sie, an internal rotation (the highest vertebral level reache by the thumb) were measure for both shoulers in all patients. Patients with unilateral involvement who ha a loss of 10 of forwar flexion, a loss of 10 of external rotation with the arm at the sie, an a loss of internal rotation of more than two spinal levels compare with the contralateral sie were inclue in this stuy. In aition, patients with bilateral shouler pain an stiffness were also inclue. Patients who ha a history of iabetes mellitus or who ha experience a traumatic event involving the shouler were exclue. Patients with raiographic evience of a calcifie eposit or osteoarthritic change were also exclue. Consequently, the stuy cohort consiste of 379 shoulers in 367 patients (162 male an 205 female) with a mean age of sixty-one years (range, thirty-five to eighty-three years). Passive ROM was measure first with the patient in the staning position. However, we also measure forwar flexion an external rotation with the patient in the supine position in orer to minimize the influence of pain. In the supine position, the scapula can be stabilize, an it is easier to maintain the humeral hea in an optimal position relative to the glenoi cavity by holing the patient s arm. Pain can therefore be reuce or eliminate because muscle tone aroun the shouler girle is reuce. Consequently, more precise measurement with less pain is possible compare with measurement when the patient is in the staning position (Figs. 1 an 2). Patterns an Severity of ROM Limitation The patients were ivie into three groups accoring to the severity an patterns of shouler stiffness. Group 1 ha severe an global loss, with severe limitation of passive ROM in all three measure irections corresponing with ahesive capsulitis: 100 of forwar flexion, 10 of external rotation, an internal rotation not more cephala than the L5 level. We use the same efinition in a previous stuy 11. Group 2 ha severe limitation in at least one irection but not global loss of motion. For example, this woul inclue a patient with 90 of forwar flexion, 40 of external rotation, an L5-level internal rotation. Group 3 inclue the remaining shoulers an was efine as those having mil to moerate limitation. Imaging Stuies Three hunre an sixty-five shoulers (355 patients) unerwent plain MRI with a 1.5-T imager (Intera 1.5T; Philips, Amsteram, the Netherlans) with a phase-array surface coil. Patients were positione with the humerus in a neutral position. T2-weighte MRI scans using an echo-train length of 10 were obtaine in axial, oblique coronal (parallel to the long axis of the supraspinatus

3 1235 TABLE I (continue) Range of Motion (eg) Rotator Cuff Lesions FF ER IR Intact PTT FTT 88 (70-100) 1 (215-10) B (GT-L5) 91% (81) 9% (8) 0% (0) 130 (80-170) 31 (25-80) L5 (GT-T7) 44% (49) 17% (19) 39% (43) 93 (80-100) 0 (25-5) L4 (L4-L4) 25% (1) 25% (1) 50% (2) 95 (90-100) 27 (15-45) S (GT-L5) 42% (5) 25% (3) 33% (4) 120 ( ) 3 (0-10) S (B-L5) 100% (12) 0% (0) 0% (0) 96 (85-100) 41 (20-70) L1 (L4-T7) 29% (2) 0% (0) 71% (5) 142 ( ) 41 (15-80) S (B-L5) 38% (25) 21% (14) 41% (27) 144 ( ) 7 (0-10) T12 (L4-T8) 40% (4) 10% (1) 50% (5) 154 ( ) 49 (20-90) T12 (L4-T6) 35% (62) 16% (28) 50% (89) tenon), an oblique sagittal (perpenicular to the long axis of the supraspinatus tenon) planes using a 3.5-mm slice thickness with a 1-mm gap between the slices. The MRI parameters of the T2 coronal scans were as follows: repetition time (TR), 5000 ms; echo time (TE), 100 ms; fiel of view (FOV), 160 mm; an matrix, For the T2 sagittal scans, parameters were TR, 4147 ms; TE, 100 ms; FOV, 160 mm; an matrix, For the T2 axial scans, they were TR, 4000 ms; TE, 100 ms; FOV, 160 mm; an matrix, In aition, T1- weighte oblique sagittal images (TR, 400 ms; TE, 10.5 ms; FOV, 160 mm; an matrix, ) were routinely obtaine in orer to assess rotator cuff muscle atrophy. The remaining fourteen shoulers (twelve patients), which i not unergo MRI because of cost issues, claustrophobia, or other reasons, unerwent high-resolution ultrasonography (EUB-7500; Hitachi Meical, Tokyo, Japan) performe by experience raiographic technicians with use of a highfrequency transucer (14 MHz). The examination was performe with the patient seate on a stool, with the examiner staning behin the patient. Images Fig. 1 Fig. 2 Fig. 1 Measuring external rotation in the staning position. Many patients complain of pain uring range-of-motion measurement, especially uring external rotation with the arm at the sie. This often forces examiners to unerestimate the range of motion. Fig. 2 Measuring external rotation in the supine position. In this position, the scapula can be stabilize an the pain can be reuce or eliminate uring measurement. As a result, a more accurate measurement may be possible.

4 1236 of the supraspinatus tenon were obtaine with 20 of shouler extension with neutral rotation. The transucer was oriente parallel to the tenon (approximately 45 between the coronal an sagittal planes). The transucer was rotate 90 in orer to examine the tenons in the transverse plane. Two physicians (one musculoskeletal raiologist an one shouler surgeon), who were unaware of the patients information, such as symptoms an ROM, investigate rotator cuff lesions using the T2-weighte MRI or ultrasonography images. The final iagnosis on ultrasonography was mae by these two physicians on the basis of reports by the experience raiology technicians. Reports regaring rotator cuff pathology were reviewe, an any ivergent finings were reevaluate until the two physicians reache consensus. Statistical Analysis We use the Kruskal-Wallis test for a comparison of the groups. The Steel- Dwass test was use for post-hoc analysis when the ifference was significant. A one-way analysis of variance (ANOVA) an a post-hoc test (Tukey-Kramer test) were also use for a comparison of age among the groups. The level of significance was set at p < All analyses were conucte with use of Statcel software (version 3; OMS Institute, Tokyo, Japan). Source of Funing No external funing was receive in support of this stuy. Results Group Characteristics Group 1 consiste of eighty-nine shoulers in eighty-five patients (twenty-six male an fifty-nine female). The mean age was fifty-seven years (range, thirty-five to eighty years). Group 2 consiste of 111 shoulers in 109 patients (fifty-five male an fifty-four female). The mean age was sixty years (range, thirty-eight to eighty-three years). Group 3 consiste of 179 shoulers in 173 patients (eighty-one male an ninety-two female). The mean age was sixty-three years (range, thirty-five to eighty-three years). One-way ANOVA an the post-hoc test foun that Group 1 was significantly younger than Group 3 (p < 0.01). Group 1 inclue a significantly higher percentage of female patients than Group 2 an Group 3 (p < 0.05). ROM by Group In Group 1 (severe an global loss), mean ROM measurements were 88 of forwar flexion (range, 70 to 100 ), 1 of external rotation with the arm at the sie (range, 215 to 10 ), an buttock-level internal rotation (range, greater trochanter to L5) (Table I). In Group 2 (severe but not global loss), mean ROM measurements were 130 of forwar flexion (range, 80 to 170 ), 31 of external rotation with the arm at the sie (range, 25 to 80 ), an L5-level internal rotation (range, greater trochanter to T7). In Group 3 (mil to moerate limitation), mean ROM measurements were 154 of forwar flexion (range, 105 to 180 ), 49 of external rotation with the arm at the sie (range, 20 to 90 ), an T12-level internal rotation (range, L4 to T6). Rotator Cuff Lesions Imaging stuies reveale that, in Group 1, eighty-one (91%) of the shoulers ha an intact cuff an eight (9%) ha a partialthickness tear (Table I). No patient in this group ha a fullthickness rotator cuff tear. In Group 2, forty-nine (44%) of the shoulers were intact, nineteen (17%) ha a partial-thickness tear, an forty-three (39%) ha a full-thickness tear. In Group 3, sixty-two (35%) of the shoulers were intact, twenty-eight (16%) ha a partial-thickness tear, an eighty-nine (50%) ha a full-thickness tear. Statistical analysis foun a significant ifference among the groups in the prevalence of rotator cuff lesions (p < 0.001). The post-hoc test reveale that Group 1 iffere significantly from both Group 2 an Group 3 (p < 0.01 for each). Discussion There is some controversy about the relationship between shouler stiffness, loss of motion, an rotator cuff tearing. This stuy clearly emonstrate that severe global loss of motion, which is consistent with iiopathic ahesive capsulitis, is not associate with full-thickness rotator cuff tearing. In contrast, shoulers with a rotator cuff tear may have stiffness, although it is not usually severe with global loss of motion. The most prominent fining in this stuy was that rotator cuff lesions were rare in the severe an global stiffness group (Group 1); no shoulers ha a full-thickness rotator cuff tear. In Group 2 an Group 3, we foun that 39% an 50% of the shoulers, respectively, ha a full-thickness rotator cuff tear. Because avance imaging stuies such as MRI or ultrasonography are not typically warrante in the early evaluation of frozen shouler 12-14, our results may obviate the use of avance imaging stuies for patients who emonstrate severe an global loss of motion. Clinical examination, incluing accurate measurement of ROM, an proper treatment accoring to symptoms are much more important than rushing into avance imaging stuies for these patients 15. In contrast, patients presenting with less severe shouler stiffness an loss of motion may have early or late/recovering ahesive capsulitis, or the stiffness they are experiencing might be seconary to rotator cuff isease. Therefore, avance imaging stuies in an early stage of a patients visit can help physicians make a proper iagnosis for these patients. Our criteria for classifying severe an global loss of ROM may have potential weaknesses because we use specific ranges that are somewhat arbitrary. However, in our experience, a typical patient with frozen shouler usually emonstrates aroun 80 to 90 of forwar flexion, 0 of external rotation, an buttocklevel internal rotation. In fact, a stanarize efinition of frozen shouler base on ROM oes not exist an may not be appropriate. Therefore, we believe that our criteria are appropriate when compare with those of other authors 5,7,8. The main strength of this stuy is that a broa spectrum of subjects with variable limitation of motion was inclue, an that enable us to perform a meaningful statistical analysis of the relationship between shouler stiffness an rotator cuff tearing. Another strength is that two physicians reviewe the imaging stuies in a bline fashion, an the rotator cuff lesions were iagnose through the consensus of these two physicians. In aition, ROM was evaluate carefully while also minimizing the effect of pain uring measurement. There were several limitations of this stuy. First, we i not take isease uration into account for the analyses, espite

5 1237 the fact that this might relate to the egree of stiffness. Secon, we i not evaluate muscle strength. However, this may have little effect on the results of this stuy, as weakness is not a typical fining of ahesive capsulitis an patients with rotator cuff tears who are weak generally have larger tears. Thir, we i not use intra-articular injections of anesthetics to completely eliminate pain when we examine passive ROM. Finally, we use both MRI an ultrasonography to evaluate rotator cuff lesions. Although sensitivity for the iagnosis of rotator cuff lesions can be ifferent between MRI an ultrasonography, we ha to use ultrasonography for some patients, for reasons incluing claustrophobia. At least one review article has emonstrate no significant ifferences in either sensitivity or specificity between MRI an ultrasonography in the iagnosis of partial or fullthickness rotator cuff tears 16. In conclusion, we foun that shouler stiffness with severe an global loss of passive ROM is not associate with a fullthickness rotator cuff tear, although some patients may emonstrate a partial-thickness tear. Shoulers with severe an global loss of ROM at the first visit are likely to be cases of iiopathic ahesive capsulitis an may have less necessity for further imaging stuies. In contrast, because the patients with less severe stiffness may have a higher risk of rotator cuff isease, further imaging can help etermine the proper treatment strategy. n NOTE: The authors thank Yuko Kohashi, MD, Momoko Matsumoto, an Hikaru Miyauchi for their contribution to this stuy involving the imaging an interpretation. Yusuke Uea, MD Hiroyuki Sugaya, MD Norimasa Takahashi, MD Keisuke Matsuki, MD Nobuaki Kawai, MD Morihito Tokai, MD Kazutomo Onishi, MD Shota Hoshika, MD Shouler & Elbow Center, Funabashi Orthopaeic Hospital, Hazama, Funabashi, Chiba , Japan. aress for H. Sugaya: hsugaya@nifty.com References 1. Duplay S. De la péri-arthrite scapulo-humérale et es raieurs e l'épaule qui en sont la conséquence. Arch Gen Me. 1872;20: Coman EA. The shouler. New York: G. Miller an Co. Meical Publishers; p Zuckerman JD, Cuomo F. Frozen shouler. In: Matsen FA, Fu FH, Hawkins RJ, eitors. The shouler: a balance of mobility an stability. Rosemont: American Acaemy of Orthopaeic Surgeons; p Kay NR, Slater DN. Fibromatoses an iabetes mellitus. Lancet Aug 8; 2(8241): Robinson CM, Seah KT, Chee YH, Hinle P, Murray IR. Frozen shouler. J Bone Joint Surg Br Jan;94(1): Runquist PJ, Luewig PM. Patterns of motion loss in subjects with iiopathic loss of shouler range of motion. Clin Biomech (Bristol, Avon) Oct;19(8): Yoo JC, Ahn JH, Lee YS, Koh KH. Magnetic resonance arthrographic finings of presume stage-2 ahesive capsulitis: focus on combine rotator cuff pathology. Orthopeics Jan;32(1): Watson L, Dalziel R, Story I. Frozen shouler: a 12-month clinical outcome trial. J Shouler Elbow Surg Jan-Feb;9(1): Pollock RG, Durale XA, Flatow EL, Bigliani LU. The use of arthroscopy in the treatment of resistant frozen shouler. Clin Orthop Relat Res Jul;304: Anersen NH, Søjbjerg JO, Johannsen HV, Sneppen O. Frozen shouler: arthroscopy an manipulation uner general anesthesia an early passive motion. J Shouler Elbow Surg May-Jun;7(3): Iwamoto W, Sugaya H, Takahashi N, Kawai N, Tanaka M, Kitayama S. Range of motion eficit in shoulers with full-thickness rotator cuff tears. Katakansetsu. 2013;37(2): Lee SY, Park J, Song SW. Correlation of MR arthrographic finings an range of shouler motions in patients with frozen shouler. AJR Am J Roentgenol Jan;198(1): Song KD, Kwon JW, Yoon YC, Choi SH. Inirect MR arthrographic finings of ahesive capsulitis. AJR Am J Roentgenol Dec;197(6):W Mengiari B, Pfirrmann CW, Gerber C, Holer J, Zanetti M. Frozen shouler: MR arthrographic finings. Raiology Nov;233(2): Epub 2004 Sep Loeffler BJ, Brown SL, D Alessanro DF, Fleischli JE, Connor PM. Incience of false positive rotator cuff pathology in MRIs of patients with ahesive capsulitis. Orthopeics May;34(5):362. Epub 2011 May e Jesus JO, Parker L, Frangos AJ, Nazarian LN. Accuracy of MRI, MR arthrography, an ultrasoun in the iagnosis of rotator cuff tears: a meta-analysis. AJR Am J Roentgenol Jun;192(6):