Imaging

Ultrasound imaging for the rheumatologist XLV. Ultrasound of the shoulder in

L. Riente1, A. Delle Sedie1, E. Filippucci2, A. Iagnocco3, G. Sakellariou4, R. Talarico1, L. Carli1, L. Di Geso2, F. Ceccarelli3, S. Bombardieri1

1Rheumatology Unit, University of Pisa, ABSTRACT psoriasis which can involve peripheral Pisa, Italy; 2Cattedra di Reumatologia, Objectives. The aims of this study were and axial joints. Classified as a spondy- Università Politecnica delle Marche, and to investigate the prevalence of ultra- loarthropathy, it shares many clinical Ospedale A. Murri, Ancona, Italy; sound (US) pathologic abnormalities and radiological features with other 3Rheumatology Unit, Dipartimento di Medicina Interna e Specialità Mediche, in the shoulders of psoriatic arthritis rheumatic disorders in the spectrum of Sapienza Università di Roma, Policlinico (PsA) patients and to compare them spondyloarthritis, such as ankylosing Umberto 1°, Rome, Italy; 4Division of with the main clinical findings. spondylitis (AS) or . Rheumatology, IRCCS Policlinico San Methods. Ninety-seven PsA patients Enthesitis, dactylitis and arthritis with Matteo Foundation, Pavia, Italy. were enrolled in the study. The subacro- frequent involvement of sacroiliac Lucrezia Riente, MD mial/subdeltoid bursa, the sheath of the joints and spondylitis are observed in Andrea Delle Sedie, MD long biceps tendon, the glenohumeral PsA patients (1). It is known that shoul- Emilio Filippucci, MD joint and the acromion-clavicular joint der involvement occurs in up to 50% of Annamaria Iagnocco, MD were examined for the presence of syno- AS patients, but little is known on the Garifallia Sakellariou, MD Rosaria Talarico, MD vial effusions and synovial hypertrophy. prevalence and the outcome of shoul- Linda Carli, MD Rotator cuff tendons (supraspinatus, der joint involvement in PsA (2). Luca Di Geso, MD subscapularis, infraspinatus) were im- In the last decades, several radiographic Fulvia Ceccarellli, MD, PhD aged for tendinosis, calcifications and imaging modalities, including magnet- Stefano Bombardieri, MD, total or partial tears, while deltoid en- ic resonance imaging (MRI) and ultra- Professor of Rheumatology thesis were evaluated for local enthesi- sonography (US), have been explored Please address correspondence tis and the lesser and greater tuberos- for accurate and early diagnosis of ar- and reprint requests to: ity of the humerus for the presence of thritis, and particularly of rheumatoid Dr Lucrezia Riente, . U.O. Reumatologia, arthritis, and there is now growing in- Dipartimento di Medicina Clinica Results. Tendinosis represented the most terest for their use in spondyloarthritis e Sperimentale, frequent abnormal finding. Supraspina- (3-6). US is already established as an Università di Pisa, tus tendinosis was detected more often attractive imaging technique and has Via Roma 67, than subscapularis and infraspinatus been successfully used in the assess- Pisa 56126, Italy. tendinosis. When considering tendon ment of shoulder involvement in rheu- E-mail: [email protected] tear, supraspinatus was also the most matic disorders (7). However, to the Received on April 17, 2013; accepted frequently involved anatomical struc- best of our knowledge, so far, very few in revised form on April 19, 2013. ture. Clinical examination frequently studies have evaluated US examination Clin Exp Rheumatol 2013; 31: 329-333. failed to detect abnormalities in patients of shoulder joints in PsA patients (8). © Copyright CLINICAL AND in whom US examination showed patho- The aim of this multicentre study was to EXPERIMENTAL RHEUMATOLOGY 2013. logical findings. This is particularly true investigate the prevalence of US patho- for tendon involvement, i.e. effusion logic abnormalities in the shoulders of Key words: psoriatic arthritis, within the sheath of the biceps tendon PsA patients and to compare them with ultrasound, shoulder, , was imaged in 43 shoulders but clinical the main clinical findings. tendinosis, enthesitis assessment reported abnormalities only in 22 shoulders (p<0.0001). Patients and methods Conclusions. US examination appears The study included ninety-seven con- to be a useful and sensitive imaging secutive PsA patients (male/female ra- technique, specifically in identifying tio: 43/54; mean age 51±14 years, min: joint and tendon involvement of the 18, max: 83) attending the out-patient shoulder. and in-patient clinics of four Rheu- matology Units (University of Pisa, Introduction Università Politecnica delle Marche, Psoriatic arthritis (PsA) is a chronic in- University of Pavia and Sapienza Uni- Competing interests: none declared. flammatory associated with versity of Rome, Italy), examined from

329 IMAGING Shoulder ultrasound in psoriatic arthritis / L. Riente et al.

October 2012 to January 2013. The di- Table I. Demographic profile. agnosis of PsA was made according to Number of patients 97 the Classification Criteria for Psoriatic M/F 43/54 Arthritis (CASPAR) (9). The primary aim of the study was to assess the prev- Mean ± SD age (min-max) (years) 51±14 (18–83) alence of US pathologic abnormalities Mean ± SD disease duration (min-max) (months) 95±13 (2–576) in the shoulders of PsA patients and to Disease duration (months) Number of patients compare them with the clinical profile. <12 10 12–24 13 The secondary aim was to explore any 24–60 26 potential correlations between demo- >60 48 graphic features, clinical pattern and prevalence of US pathologic abnor- Table II. Shoulder clinical evaluation (provocative tests) in PsA patients. malities of the shoulders in the cohort examined. Therefore, data regarding Right Left Total joints sex, age, clinical features, US abnor- (n=97) (n=97) (n=194) malities, body mass index (BMI) (10) Hawkins’s test (n/%) 16/16.4 12/12.4 28/14.4 and quality of life (by means of vali- Patte’s test (n/%) 8/8.2 6/6.2 14/7.2 dated questionnaires) were collected Jobe’s test (n/%) 20/20.6 15/15.4 35/18 in an electronic database. Patients who Gerber’s test (n/%) 10/10.3 8/8.2 18/9.2 received corticosteroid injections in Speed test (n/%) 16/16.5 14/14.4 30/15.5 the shoulder within the 3 months prior to the study were excluded from the evaluation. Demographic and clinical Table III. US-detected features at the level of the rotator cuff tendons. characteristics of the cohort studied are Tendinosis Tears* summarised in Table I. Right Left Total Right Left Total n=97 n=97 n=194 n=97 n=97 n=194 Clinical assessment All the patients were assessed for his- Supraspinatus (n/%) 43/44 37/38 80/41 16/16 7/7 23/12 tory of shoulder pain and underwent Subscapularis (n/%) 23/24 20/21 43/22 1/1 2/2 3/1.5 a complete clinical assessment of the Infraspinatus (n/%) 16/16 13/13 29/15 0/- 0/- 0/- shoulder by an expert rheumatologist * for the presence/absence of pain and partial-thickness or complete tear. tenderness. The Hawkins test, Jobe test, Patte test, Gerber test and Speed by a single rheumatologist who was Rotator cuff tendons (supraspinatus, test were the provocative maneuvers well experienced in musculoskeletal subscapularis, infraspinatus) were im- performed to provide a more focused US. Before the start of the study, all the aged for the presence of tendinosis, evaluation for impingement syndrome, sonographers reached an agreement local calcifications and total or partial rotator cuff and biceps both on the scanning technique and the tears. brachi pathology, respectively (11). definition of the pathological findings The ACJ was examined for osteophytes All the patients underwent a comprehen- to detect. and erosions as well as fibrocartilage sive rheumatologic evaluation; DAS28 In all patients, US examination of both calcifications. The GHJ was also evalu- was used to evaluate the disease activ- shoulders was performed according to ated for the assessment of labrum cal- ity. The physicians conducting the clini- the EULAR guidelines (12). The su- cifications, intra-cartilagineous crystal cal and US examinations were blinded bacromial/subdeltoid bursa, the sub- material (double contour sign), topha- to the results of each other’s findings scapular bursa, the sheath of the long ceous deposits and erosions (with or biceps tendon, the axillary and poste- without local PD signal). Ultrasonographic assessment rior recesses of the gleno-humeral joint Finally, we imaged the deltoid en- In each of the 4 participating units in (GHJ) and the acromion-clavicular joint thesis at the acromion attachment for the study, US examinations (includ- (ACJ) were examined for the presence local enthesitis (defined according ing both grey-scale and power Doppler of synovial effusions (SE) and syno- to D’Agostino et al.), and lesser and examinations) were carried out using a vial hypertrophy (SH). When synovial greater tuberosity of the humerus for the MyLab TWICE XVG machine (Esa- hypertrophy was detected, power Dop- presence of local enthesophytes (8, 13). ote, Genoa, Italy) with a linear probe pler examination was performed and All abnormalities were studied ac- operating at 4-13 MHz and a Logiq 9 the following settings used: PRF 500 cording to the internationally accepted machine (General Electrics Medical Hz, Doppler frequency 7.1–7.5 MHz definitions and scored according to a Systems, Milwaukee, WI, USA) with and Doppler gain to avoid random dichotomous assessment (14, 15). In a linear probe operating at 9-14 MHz noise visualisation. addition, calcifications were assessed

330 Shoulder ultrasound in psoriatic arthritis / L. Riente et al. IMAGING for the presence of acoustic shadow- normalities of the rotator cuff tendons and 3/194 (1.5%) joints, respectively, ing and their maximal dimension was were frequently observed; in particu- while no infraspinatus tendon tear was measured. lar, tendinosis represented the most observed. Bilateral supraspinatus tear frequent abnormal finding (Table III). was detected in 4 patients. Statistical analysis Supraspinatus tendinosis was detect- Effusion within the sheath of the biceps All results are expressed in means ± ed more often than subscapularis and tendon was a frequent finding: it was standard deviation (SD). Chi-square infraspinatus tendinosis: in 80/194 observed in 43/194 shoulders (22%), test, t-test and ANOVA were used to (41%), 43/194 (22%) and 29/194 bilaterally in 13 patients, and synovial evaluate the differences between the (14.9%) shoulders, respectively. More- hyperthrophy was associated in 13 out subgroups. A p-value <0.05 was con- over, calcifications within supraspina- of 43 joints (30%) (PD signal was also sidered statistically significant. All cal- tus tendon were visualised more com- present in 4 of them). In 11 out 43 joints culations were made using StatView monly with respect to subscapularis (25%) effusion within the sheath of the program ver. 5.0. and infraspinatus tendons: in 35/194 biceps tendon was associated with su- (18%), 9/194 (4.6%) and 4/194 (2%) praspinatus tear (no association with Results shoulders, respectively. The dimen- subscapularis or infraspinatus tendon From a clinical point of view, 54 pa- sions of the rotator cuff tendon calci- tear was observed) tients showed active joint involvement fications were between 0.4–6.1 mm, Effusion within the subacromion- (defined according to a DAS28 >2.6) with acoustic shadowing present in 18 subdeltoid bursa and the subscapular at the time of US evaluation, 43% of (9%) out of 194 shoulders. Bilateral bursa was present in 28 (14%) and in whom with peripheral joint involve- tendinosis of the supraspinatus was im- 1 (0.5%) of the 194 shoulders, respec- ment, and 9% with axial joint involve- aged in 13 patients, of both supraspina- tively (Fig. 1). ment only; the other subjects were char- tus and subscapularis in 7 patients and Synovial hypertrophy was imaged in acterised by the concomitant presence of the three main components of rota- 57% of subacromion-subdeltoid bursa of axial and peripheral involvement. tor cuff, supraspinatus, subscapularis and in the only involved subscapular Moreover, monolateral pain and/or and infraspinatus in 9 patients. When bursa (PD signal was also detected in tenderness elicited by physical exami- considering tendon tear, the suprasp- 3 of them). nation was found in 21 patients, while inatus was also the most frequently in- Effusion in ACJ was detected in 51 in 20 cases shoulder pain was bilateral volved anatomical structure (Fig. 1). In out of 194 shoulders (26%). Synovial (Table II). fact, partial-thickness or complete tear hypertrophy was imaged in 19 out of 51 Globally, 194 shoulder joints of 97 of the supraspinatus and subscapula- ACJ (40%) associated to effusion and patients were studied. Using US, ab- ris were visualised in 23/194 (11.8%) only 1 out of them showed PD signal. Osteophytes at the ACJ were observed in 53 (27%) out of 194 shoulders and calcifications in only 2 joints. The GHJ was rarely involved; in fact joint effusion was found in only 4 shoulders by posterior and axillary US assessment and was associated to syno- vial hypertrophy in 3 cases with no PD signal (Fig. 1). Labrum calcifications were infrequently detected (4 out of 194 joints). Enthesophytes at the lesser and greater tuberosity of the humerus were visu- alised in 26 (13.4%) and in 15 (7.7%) shoulders, respectively; deltoid proxi- mal insertion enthesitis was found in 6 (3%) shoulders. Bone erosions were imaged in 29 (14.9%) shoulders, but no PD signal in the erosions was detected. No intra-cartilagineous crystal mate- rial, precisely “double contour sign”, or tophaceous deposits were imaged. Fig. 1. A. Axillary recess of the gleno-humeral joint with evidence of joint effusion*. h = humeral Humeral head hyaline cartilage calcifi- head. B. Distension of the sub-deltoid bursa (arrow) due to local effusion* and synovial proliferation°. cation was detected in only 1 shoulder. h = humeral head. C. and D. Transverse and longitudinal scan showing full thickness tear^ within the By dynamic US assessment, mechani- supraspinatus tendon (spt). h = humeral head. cal impingement of supraspinatus and

331 IMAGING Shoulder ultrasound in psoriatic arthritis / L. Riente et al. subscapular tendons was found in 6 authors concluded that DPIE appears jects (11), we can speculate that rotator shoulders. more frequently in PsA than in other cuff belongs to the clini- Patients with US-detectable shoulder spondyloarthritis. DPIE can mimic im- cal picture of PsA and may represent a involvement did not differ significantly pingement syndrome but US can dif- manifestation of the disease. Bilateral in terms of disease duration, disease ferentiate between the two conditions. tendinosis involving from one to three activity and BMI from patients with Plain radiographs have traditionally components of the rotator cuff was also normal US findings. been used to detect joint abnormali- imaged in 29 patients. Clinical examination frequently failed ties in patients with painful shoulder We frequently observed effusion of ACJ to detect abnormalities in patients in and findings suggestive of the presence in PsA. However, synovial hypertrophy whom US examination showed patho- of pathology can be seen. was associated to effusion in only 37% logical findings. This is particularly true However, plain radiography may be of the cases, with PD signal detectable for tendon involvement, i.e. effusion completely normal even in the presence exclusively in 1 shoulder. Osteophytes within the sheath of the biceps tendon of a complete rotator cuff disruption at ACJ level were observed in less than was imaged in 43 shoulders but clinical and is a poor indicator of inflammatory one third of the shoulders. It is prob- assessment reported abnormalities only changes in entheses. In 2007 Helliwell able that in the case of ACJ, beyond in 22 shoulders (p<0.0001). Moreover, et al. (17) reviewed the radiographic the classical signs of arthritis, we visu- supraspinatus tendinosis was detected evidence of obtained alised manifestations of osteoarthritis; by US in 80 shoulders but the Jobe from participants in the CASPAR study the clinical picture may thus be diffi- test was positive in only 35 out of 80 and the plain radiographic features of cult to interpret, due to the coexistence shoulders, subscapularis tendinosis peripheral enthesopathy at major sites of the two disorders. was visualised in 43 shoulders but the were compared between PsA, other While effusion within the sheath of the Gerber test was abnormal in 18 cases and rheumatoid biceps tendon represented a frequent and, finally, the Patte test was positive arthritis. Among the many and interest- finding, effusion within the subac- in 14 shoulders but infraspinatus tendi- ing observations reported in this study, romion-subdeltoid bursa and the sub- nosis was imaged in 29 shoulders. One the authors noted that entheseal erosion scapularis bursa and within the GHJ patient complained of bilateral pain but in the knee, elbow or shoulder of PsA was rarely detected and no significant no US abnormalities were disclosed. patients was uncommon and rarely de- differences with healthy subjects were Thus, it is worth noting that US re- tected by plain radiology. observed. This latter data may be unex- vealed pathological findings in patients Compared to MRI, US is a relatively pected and not easily explained. We can who did not complain of pain at the inexpensive imaging technique, which hypothesise that the prevalent damage provocative maneuvers at the time of provides a rapid diagnosis in specific in PsA shoulder is at the level of the ro- the clinical examination. clinical situations. A complete US ex- tator cuff tendons and that arthritis of The agreement between clinical and amination of the rotator cuff can be the gleno-humeral is a rare condition. US findings was indeed low, with a carried out in a relatively short time It should also be considered that the non-significant concordance on posi- and useful observations on soft tis- patients were enrolled consecutively tive or negative findings. sue structures during movement can and may not have had shoulder pain. be obtained. US imaging technique is, However, further studies are needed to Discussion therefore, a sensitive and accurate tool provide more information on the fluid Several studies have demonstrated the for evaluating the severity of shoulder topographic distribution in GHJ, also utility of US imaging in the evaluation joint involvement in the course of PsA using other imaging techniques (MRI). of joints and entheses in patients affect- and can help in planning an adequate Deltoid proximal insertion enthesitis ed by PsA, a disorder characterised by treatment of this disease. was found in a very limited number of a great variability in the clinical picture In using this technique we frequently PsA patients, which is in contrast with (3-6, 13, 16). identified inflammatory involvement, what Falsetti et al. reported (8). This PsA patients frequently complain of tendinosis and/or tendon tear of the ro- discrepancy may be due to the differ- shoulder pain but the prevalence and tator cuff tendons, with a significantly ent sample sizes in the cohorts studied, the features of shoulder arthropathy higher prevalence than that reported in but the frequency of enthesitis in PsA in the course of PsA are still unclear, the cohort of healthy subjects studied shoulder still needs to be examined in and has been the topic of very few by Iagnocco et al. (11), while no differ- depth, and we hope that it will be the investigations. In 2002 Falsetti et al. ences in the frequency of tendon cal- topic of further investigations. (8) performed a retrospective study of cifications were observed in the same It is interesting to observe the different clinical, sonographic and radiological cohorts studied. It is well-known that results provided by clinical examina- examinations of the shoulder in 100 the frequency of rotator cuff abnormal- tion and US assessment. Rotator cuff patients with spondyloarthritis (41 of ities progressively increase with age, tendinopathy or effusion within the whom were PsA patients) to evalu- but even if the mean age of our PsA pa- sheath of the biceps tendon were im- ate the frequency of deltoideal proxi- tients was higher, but not statistically aged in shoulders which were negative mal insertion enthesitis (DPIE). The relevant, than that of the healthy sub- for pain and/or tenderness during clini-

332 Shoulder ultrasound in psoriatic arthritis / L. Riente et al. IMAGING cal examination and the provocative 4. DELLE SEDIE A, RIENTE L, FILIPPUCCI E et in healthy individuals. Clin Exp Rheumatol maneuvers utilised. On the contrary, no al.: Ultrasound imaging for the rheumatolo- 2013; 31: 165-71. gist XXVI. Sonographic assessment of the 12. BACKHAUS M, BURMESTER GR, GERBER US abnormalities were detected in one knee in patients with psoriatic arthritis. Clin T et al.: Guidelines for musculoskeletal ul- patient who complained of bilateral Exp Rheumatol 2010; 28: 147-52. trasound in rheumatology. Ann Rheum Dis shoulder pain. These data confirm that 5. RIENTE L, DELLE SEDIE A, SCIRÈ CA et al.: 2001; 60: 641-9. clinical examination alone may not be Ultrasound imaging for the rheumatologist. 13. D’AGOSTINO MA: Ultrasound imaging in XXXI. Sonographic assessment of the foot in spondyloarthropathies. Best Pract Res Clin able to diagnose the exact features and patients with rheumatoid arthritis. Clin Exp Rheumatol 2010; 24: 693-700 the extent of joint involvement in PsA Rheumatol 2011; 29: 1-5. 14. WAKEFIELD RJ, BALINT PV, SZKUDLAREK patients, thus running the risk of com- 6. COATES LC, HODGSON R, CONAGHAN PG et M et al.: Musculoskeletal ultrasound Includ- al.: MRI and ultrasonography for diagnosis ing definitions for ultrasonographic pathol- promising the choice of a more effica- and monitoring of psoriatic arthritis. Best ogy. J Rheumatol 2005; 32; 2485-7. cious therapy (13, 18, 19). Pract Res Clin Rheumatol 2012; 26: 805- 15. FILIPPUCCI E, SCIRÈ CA, DELLE SEDIE A et In conclusion, PsA still remains an in- 22. al.: Ultrasound imaging for the rheuma- triguing and incompletely understood 7. IAGNOCCO A, FILIPPUCCI E, MEENAGH G et tologist. XXV. Sonographic assessment of al.: Ultrasound imaging for the rheumatolo- the knee in patients with gout and calcium condition in which US examination ap- gist. I. Ultrasonography of the shoulder. Clin pyrophosphate deposition disease. Clin Exp pears to be a useful and sensitive imag- Exp Rheumatol 2006; 24: 6-11. Rheumatol 2010; 28: 2-5. ing technique, specifically in identify- 8. FALSETTI P, FREDIANI B, FILIPPOU G et al.: 16. DELLE SEDIE A, RIENTE L, FILIPPUCCI E et ing joint and tendon involvement of the Enthesitis of proximal insertion of the deltoid al.: Ultrasound imaging for the rheumatolo- in the course of seronegative spondyloar- gist. XXXII. Sonographic assessment of the shoulder. thritis. An atypical enthesitis that can mime foot in patients with psoriatic arthritis. Clin impingement syndrome. Scand J Rheumatol Exp Rheumatol 2011; 29: 217-22. References 2002; 31: 158-62. 17. HELLIWELL PS, PORTER G; CASPAR STUDY 1. MEASE PJ: Psoriatic arthritis: update on 9. TAYLOR W, GLADMAN D, HELLIWELL P et GROUP: Sensitivity and specificity of plain ra- pathophysiology, assessment and manage- al.:. Classification criteria for psoriatic ar- diographic features of peripheral enthesopa- ment. Ann Rheum Dis 2011; 70 (Suppl. 1): thritis: development of new criteria from a thy at major sites in psoriatic arthritis. Skel- i77-84. large international study. Arthritis Rheum etal Radiol 2007; 36: 1061-6. 2. SIEPER J, BRAUN J RUDWEILET M et al.: 2006; 54: 2665-73 18. GALLUZZO E, LISCHI DM, TAGLIONE E et : an overview. Ann 10. LOVE TJ, ZHU Y, ZHANG Y et al.: Obesity al.: Sonographic analysis of the ankle in pa- Rheum Dis 2002; 61 (Suppl. 3): iii8-18. and the risk of psoriatic arthritis: a popula- tients with psoriatic arthritis. Scand J Rheu- 3. RIENTE L, DELLE SEDIE A, SAKELLARIOU G tion-based study. Ann Rheum Dis 2012; 71: matol 2000; 29: 52-5. et al.: Ultrasound imaging for the rheuma- 1273-7. 19. BALINT PV, KANE D, WILSON H et al.: tologist XXXVIII. Sonographic assessment 11. IAGNOCCO A, FILIPPUCCI E, SAKELLARIOU Ultrasonography of entheseal insertions in of the hip in psoriatic arthritis patients. Clin G et al.: Ultrasound imaging for the rheuma- the lower limb in . Ann Exp Rheumatol 2012; 30: 152-5. tologist XLIV. Ultrasound of the shoulder Rheum Dis 2002; 61: 905-10.

333