ADHESIVE CAPSULITIS, first termed frozen shoulder

Transcription

1 ORIGINAL ARTICLE Randomized Controlled Trial for Efficacy of Intra-Articular Injection for Adhesive Capsulitis: Ultrasonography-Guided Versus Blind Technique Hong-Jae Lee, MD, Kil-Byung Lim, MD, PhD, Dug-Young Kim, MD, Kyung-Tae Lee, MD ABSTRACT. Lee H-J, Lim K-B, Kim D-Y, Lee K-T. Randomized controlled trial for efficacy of intra-articular injection for adhesive capsulitis: ultrasonography-guided versus blind technique. Arch Phys Med Rehabil 2009;90: Objective: To evaluate the clinical effect of ultrasonography (US)-guided intra-articular injections compared with a blind (unguided) technique for the treatment of adhesive capsulitis. Design: Randomized controlled trial. Setting: Outpatient rehabilitation clinic. Participants: Patients (N 43) diagnosed as having adhesive capsulitis after clinical examinations and radiologic and ultrasonographic study. Intervention: Under either US-guided or a blind technique, patients received a 20-mg intra-articular injection of triamcinolone mixed with 1.5mL 2% lidocaine and 4mL normal saline in the first week followed by 5 weekly injections of sodium hyaluronate. Main Outcome Measures: A visual analog scale for pain intensity, range of motion (ROM) of the shoulder (flexion, abduction, external rotation, and internal rotation), and general shoulder function during daily activities at preinjection as a baseline and then every week after injection for 6 weeks for each patient. Results: Twenty patients out of 22 in the blind injection group and 20 out of 21 in the US-guided group finished the entire 6-week study period. The improvement in pain intensity, ROM, and shoulder function score was significantly greater in the US-guided injection group than in the blind injection group by the second week postinjection (P.05). However, there were no further significant differences in the improvement between the 2 groups beyond the third week. Conclusions: US-guided intra-articular injections may offer advantages over a blind technique for the treatment of adhesive capsulitis and may deliver clinical benefits during the first few weeks of treatment. This finding suggests that the improved targeting to the intra-articular space by using US can result in better treatment of adhesive capsulitis. Key Words: Injections; Intra-articular; Rehabilitation; Ultrasonography by the American Congress of Rehabilitation Medicine From the Department of Rehabilitation Medicine, Ilsan-Paik Hospital, Inje University, and Institute of Sports Rehabilitation, Inje University, Goyang city, Gyonggido, South Korea. Presented to the Annual Assembly of American Academy of Physical Medicine and Rehabilitation, November 22, 2008, San Diego, CA. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Hong-Jae Lee, MD, Dept of Rehabilitation Medicine, Ilsan-Paik Hospital, Inje University, 2240 Daewha-dong, Ilsan-gu, Goyang city, Gyonggi-do, South Korea, , honglee@paik.ac.kr /09/ $36.00/0 doi: /j.apmr ADHESIVE CAPSULITIS, first termed frozen shoulder by Codman, has various causes and brings about sleep disorders, motor abnormality, and difficulty in daily activities because of its gradually increasing pain and the limitations imposed on the ROM of the joint. 1,2 A number of treatment techniques exist, including diathermy, electrotherapy, therapeutic exercises, drug treatment, manipulation under local anesthesia, local injection of steroids, injection of sodium hyaluronate, hydraulic distension, and surgical treatments. 3-8 In case of injection procedures, it is important to increase the accuracy rate of the injection to get a better clinical outcome. For intra-articular injection into the glenohumeral joint, the blind technique, which determines the injection site through palpation, and the fluoroscopy-guided intra-articular injection have both been used. More recently, a US-guided technique has been introduced. Because the US-guided technique has no danger of the radiation exposure typical of the fluoroscopyguided intra-articular injection and is more accurate than the blind technique, its use as a treatment aid is increasing. Yi et al 9 have reported that the US-guided technique was more accurate in intra-articular injection than the blind technique. However, the clinical treatment effects of the US-guided technique on frozen shoulder patients have not yet been addressed. Therefore, we aimed to evaluate the clinical effect of USguided intra-articular injection compared with a blind (unguided) technique. It was the hypothesis of this study that the US-guided intra-articular injection is more effective in clinical outcomes than a blind technique in patients with adhesive capsulitis. METHODS Participants We selected the subjects from patients who visited the Department of Rehabilitation Medicine of our hospital between October 2006 and May 2007 with the main complaint of limited ROM of the shoulder joint and pain. The subjects in this study were patients diagnosed as having clinical disease stage II idiopathic adhesive capsulitis 10 after clinical and physical examinations, x-ray test, and US. Patients with the diagnosis of other disorders that can cause similar clinical symptoms were excluded. Cervical examination showed no significant dysfunction suggestive of the cervical disk disorder or radiculopathy. X-ray tests of the shoulder joint were otherwise normal. In addition, US showed no particular problems such as rotator cuff injury, calcific tendinopathy, and subacromial bursitis. We ANOVA ROM US VAS List of Abbreviations analysis of variance range of motion ultrasonography visual analog scale 1997

2 1998 ULTRASONOGRAPHY-GUIDED INTRA-ARTICULAR INJECTION, Lee explained this study to all of the subjects and received their signed consent on the explanatory notes. They were alternately assigned to either US-guided injection or blind injection groups in the sequence in which they joined the study. This study was performed with the prior approval of The Ethics Committee of Inje University. Clinical Evaluations We measured the intensity of shoulder joint pain expressed by the patient by a VAS for daytime and just before sleep separately. For ROM of the shoulder, the passive ROMs were measured for flexion, abduction, external rotation, and internal rotation. We followed the standard technique, and we used a universal goniometer, which has 2 arms. 11 The flexion, extension, and abduction were checked while the patient was in a sitting position, and the external rotation and internal rotation were checked while the patient was in the supine position. For general shoulder functions, the 10 functions by Cho et al 6 were used, which included putting on clothes, combing hair with the affected arm, using a spoon and chopsticks during a meal, writing or typing on keyboards, carrying a handbag, cleaning the anus, catching something overhead, putting a hand on the back, doing routine sports, and sleeping on the painful side. This scale was developed for evaluating pain intensity while the patient was performing each common functional activity in daily life. This scale provides the specific pain intensity during specific activity, whereas the VAS in daytime and at night provides a global impression of the patient s pain. A scale of 0 to 40 was used for the ability to perform, with 4 points for the normal condition, 3 for a slight difficulty, 2 for a considerable difficulty, 1 for the condition that required help, and 0 for impossibility to perform each function. One tester, who did not know which patient belonged to which group, performed the evaluations before and after treatment. Evaluations were performed just before injection and then during the weekly outpatient follow-up observations. Injection Methods All patients received a 0.5-mL intra-articular injection of triamcinolone (20mg) mixed with 1.5mL of 2% lidocaine and 3mL of normal saline in the first week. This was followed by a 2.5-mL injection of low molecular-weight sodium hyaluronate (25mg) in the second week and then once weekly for the next 4 weeks for a total of 5 hyaluronate injections. Because it was believed that one 20-mg injection of triamcinolone does not seriously change the blood sugar of diabetes patients, 12 we explained this to the diabetes patients and received their consent before injecting the drugs. For injection into the shoulder joint, we used the posterior approach for both groups, as suggested by Yi et al. 9 The US-guided technique is as follows. For equipment, we used the LOGIQ 5 a and a 10-MHz linear probe. The patients were seated on a chair, with the affected shoulder bent and adducted. The probe touched the lower part of the acromion sideways, and a long needle (23G, 7cm) was inserted to the posterior articular surface of the humeral head and positioned inside the articular capsule (fig 1). The expansion of the articular capsule was checked while the fluid was being injected (fig 2). The reason for using a 7-cm-long needle was because if the needle is short, the ultrasound probe can contact the syringe body and interfere with the operation. For blind injection, we seated the patients on a chair, and their affected shoulder was bent and adducted. We palpated the acromion of the scapula and inserted the needle 1cm inferior to the tip of the acromion. We directed the needle toward the Fig 1. The intra-articular injection techniques used in this study. (A) The US-guided injection and (B) the blind injection. direction of the coracoid process and advanced it into the articular capsule, after which we injected the drug. To mimic injection under the guidance of US, we first contacted the patients in the blind injection group by an ultrasound probe at the skin under the acromion before the intra-articular injection. The blind injection was conducted by 1 doctor who has 7 years of experience in this type of operation but who has never used US, whereas the US-guided injection was conducted by another doctor who had 2 years of experience in the US-guided injection of the shoulder joint. After injection, we taught the patients exercises for increasing the joint ROM including stretching forward and bending down to a desk, Codman exercise, and a wall-climbing exercise with the fingers. We told them to steadily practice these at home and checked and encouraged them to keep up with these exercises every time they visited the hospital. We used SPSS 10.0 software b for statistical analysis. A post hoc repeated-measures ANOVA, which provides a better estimate of the true effect of the new injection technique, was used to compare the degree of improvement between the US-guided and blind groups in terms of VAS, ROM of flexion, abduction, external rotation and internal rotation of the shoulder, and the general shoulder functions. The significance level was defined as a P value lower than.05.

3 ULTRASONOGRAPHY-GUIDED INTRA-ARTICULAR INJECTION, Lee 1999 Fig 2. US showing (A) the needle tip in the articular capsule and (B) the capsular distension. RESULTS Clinical Characteristics of the Patients Twenty patients out of 22 in the blind injection group and 20 out of 21 in the US-guided group finished the entire 6-week study period. The average ages of the 2 groups were similar at 53.1 and The man to woman ratio was 10:10 for the blind injection group and 9:11 for the US-guided injection group. The weight and height were similar between the 2 groups. The average morbid period of shoulder pain was 10.6 months for the blind injection group and 8.5 months for the US-guided group, and there was no statistical difference between the 2 groups. The morbid ratio between left and right shoulders was 1:1. The number of patients with a history of diabetes was 3 in the blind injection group and 5 in the US-guided injection group (table 1). Comparison of Improvements in Clinical Symptoms Changes of the visual analog scale. Before treatment, the intensity of pain during daytime activities was 5.4 for the US-guided group and 6.0 for the blind group, and the intensity of pain before sleep was 5.6 for the US-guided group and 6.1 for the blind group. None of these differences was statistically significant. First, in the changes of VAS during daytime activities from baseline in the 2 groups, the intensity of pain significantly decreased from the first week onward (P.001). Regarding weekly change in a group, between the pain at a week and that at the last week, the intensity of pain decreased significantly every week except the third and sixth week in the US-guided group while that decreased significantly every week except for the sixth week in the blind group (table 2 and fig 3). In a post hoc repeated-measures ANOVA for the comparison between the 2 groups, the degree of pain reduction was more significant statistically in the US-guided group than the blind group during the first (P.001) and second week (P.001). However, there were no significant differences between the 2 groups from the third week until the end of the study (see table 2 and fig 3). Second, the intensity of pain before sleep significantly decreased in both groups from the first week on (P.001). Regarding weekly change, the intensity of pain decreased significantly every week except the sixth week in the US-guided group; it decreased significantly every week except the sixth week in the blind group. In a post hoc repeated-measures ANOVA for the comparison between the 2 groups, the amount of pain reduction was more significant statistically in the USguided group than the blind group at the first week (P.008) and the second week (P.001). However, there were no significant differences between the 2 groups from the third week until the end of the study (see table 2 and fig 3). Changes in range of motion of the shoulder joint. First, in the changes of shoulder flexion range, the ROM significantly increased from the first week on in both groups (P.001). Regarding weekly change in a group, between the range at a week and the range at the last week, the flexion range increased significantly for the first 3 weeks in the US-guided group, whereas it increased significantly the first 4 weeks in the blind group. In a post hoc repeated-measures ANOVA for the comparison between the 2 groups, the amount of increasing flexion range was more significant statistically in the US-guided group than in the blind group for the first 3 weeks (P.039 in the first week, P.001 in the second week, P.025 in the third week). However, there were no significant differences between the 2 groups from the fourth week until the end of the study (table 3). Second, in the changes of shoulder abduction range, the ROM significantly increased from the first week on in both groups (P.001). Regarding weekly change in a group, the abduction range increased significantly for the first 4 weeks in the US-guided group, whereas it increased significantly for the first 5 weeks in the blind group. In the comparison between the 2 groups, the amount of increasing abduction range was more Table 1: General Characteristics of Subjects Characteristics US-Guided Group Blind Group Number of cases Age (y) Sex (male/female) 9/11 10/10 Height (cm) Weight (kg) Duration of symptoms (months) Affected side (right/left) 10/10 10/10 Cases of diabetes mellitus 5 3 NOTE. Values are mean SD.

4 2000 ULTRASONOGRAPHY-GUIDED INTRA-ARTICULAR INJECTION, Lee Table 2: Comparison of VAS After Treatment Before After 1wk After 2wk After 3wk After 4wk After 5wk After 6wk VAS in daytime US guided * * Blind VAS before sleep US guided * * Blind NOTE. P values were obtained after the repeated-measures ANOVA. P values correspond to the mean difference between 2 groups. Values are mean SD. *P.01. significant statistically in the US-guided group than in the blind group for the first 2 weeks (P.005 in the first week, P.017 in the second). However, there were no significant differences Fig 3. A comparison of changes in the VAS of US-guided injection and conventional blind injection techniques. NOTE. P values were obtained after the repeated-measures ANOVA. P values correspond to the mean difference between the 2 groups after 1 week and 2 weeks. between the 2 groups from the third week until the end of the study (see table 3). Third, in the changes of shoulder external rotation range, the ROM significantly increased from the first week on in both groups (P.001). Regarding weekly change, the external rotation range increased significantly for the first 2 weeks in the US-guided group, whereas it increased significantly the first 5 weeks in the blind group. In the comparison between the 2 groups, the amount of increasing external rotation range was significant statistically during the second week only in the US-guided group (P.011). There were no significant differences between the 2 groups during the rest of the weeks (see table 3). Fourth, in the changes of shoulder internal rotation range, the ROM significantly increased from the first week on (P.001) in both groups. Regarding weekly change, the internal rotation range increased significantly for the first 4 weeks in the USguided group, whereas it increased significantly the first 5 weeks in the blind group. In the comparison between the 2 groups, the amount of increasing internal rotation range was more significant statistically in the US-guided group than in the blind group for the first 4 weeks (P.001 in the first and second weeks, P.001 in the third week, P.048 in the fourth week). However, there were no significant differences between the 2 groups during the fifth and last week (see table 3). Evaluations of Shoulder Functions In the changes of shoulder function score, the score significantly increased from the first week on in the US-guided group (P.003), whereas it increased from the second week in the blind group. Regarding weekly change in a group, between the score at 1 week and the score at the last week, the change increased significantly for the first 3 weeks in the US-guided group, whereas it increased significantly for the middle 4 weeks except the first and the last week in the blind group. In the comparison between the 2 groups, the amount of score improved was more significant statistically in the US-guided group than in the blind group for the first 3 weeks (P.001 in the first week, P.005 in the second week, P in the third week). However, there were no significant differences between the 2 groups from the fourth week until the end of the study (table 4). DISCUSSION Also referred to as frozen shoulder, adhesive capsulitis brings about gradually increasing pain and restricted joint ROM of the affected shoulder. Patients often complain of serious limitations in external rotation and abduction. The intra-articular injection of steroids for adhesive capsulitis has shown treatment effects through its reduction of the inflammatory reaction in the pathologic progress and has led to a relatively rapid improvement of symptoms. 13,14 The injection

5 ULTRASONOGRAPHY-GUIDED INTRA-ARTICULAR INJECTION, Lee 2001 Table 3: Comparison of ROM of the Shoulder After Treatment Before After 1wk After 2wk After 3wk After 4wk After 5wk After 6wk Flexion US guided * * Blind Abduction US guided * Blind External rotation US guided Blind Internal rotation US guided * Blind NOTE. Values are mean SD. P values were obtained after the repeated measures ANOVA. P values correspond to the mean difference between 2 groups. *P.05; P.01. of sodium hyaluronate has also been shown to have effects such as pain reduction, anti-inflammatory effects, increased joint ROM, restriction of the degeneration of cartilage tissues, and assistance in the normalization of abnormal joint fluid. 5,7,15 Previous studies 5,6 showed the usefulness of 5 consecutive injections of sodium hyaluronate. In addition, the repeated injection of hyaluronate has fewer side effects than the repeated steroid injection. Actually, the 5 consecutive weekly intraarticular injections of low molecular-weight sodium hyaluronate are performed for the patients under reimbursement of the National Health Insurance System in South Korea. So, the steroid intra-articular injection followed by 5 hyaluronate injections is a popular regimen for patients with frozen shoulder in our country. We adopted the common treatment regimen in our clinical situation for this current study. Also, we chose 20mg triamcinolone because we wanted to minimize the possible side effects of steroid, although the dosage of 40mg is commonly used for joint injections. A previous study 16 has suggested that the accuracy of injection is more important than the volume with regards to the outcome of corticosteroid injection treatment. We selected patients who were clinically in stage II of this disease because stage I patients could be improved dramatically after just the first steroid intra-articular injection and stage III patients could show little response to the treatment with any anti-inflammatory substance. Two different operators performed each injection procedure in this study. This could have led to interindividual variability in accuracy, which may have confounded the interpretation of the results. The ideal method for the current study would be that one operator who is skillful for both techniques injects to patients while being blinded. However, we were not able to find such an operator, so we chose the next best method; each skillful operator to each technique injects to patients who are assigned to him. Because both of these treatments require direct injection of the therapeutic agent into the affected joint, aids such as US for guiding the insertion of the injection needle into the articular capsule can be expected to be of overall benefit to the patient. Ultrasound can be used for obtaining images of the patients shoulder joint in real time, and, unlike radiologic methods, US requires no radiation exposure. Furthermore, it can be conducted within a relatively short time and in different postures as desired by the tester, making it more adaptable than other techniques such as computed tomography scans and magnetic resonance imaging. Yi et al 9 reported that the accuracy of intra-articular injection was only 33.3% and 46.6%, even when performed by 2 experienced operators, when the blind technique was used, but it increased to 93.3% when the US-guided technique was used in 30 patients with frozen shoulder. In this study, the success of the blind technique was determined by the existence of the injected fluid with contrast media in the intra-capsular space on an x-ray, and the success of the US-guided technique was determined by capsular distention during the injection in real time. In the current study, we injected therapeutic substances to the patients with either the US-guided technique or the blind technique. For both groups, the reduction in pain, the joint ROM, and general shoulder functions greatly improved after treatment. Also, we found that in both groups there was a significant improvement during the first 2 weeks. This could have been caused by the anti-inflammatory effect of triamcinolone. However, the comparison of the 2 groups revealed that the US-guided group showed a faster improvement than the blind group during the early stages of treatment. The pain relief was more significant during the first 2 weeks, and improvements in joint ROM and general shoulder functions were more significant during the first 3 weeks in the US-guided group. There can be a question about clinical significance of the VAS change of 1.8 in the US-guided group and 1.2 in the blind Table 4: Comparison of Total Scores of Functional Activities After Treatment Before After 1wk After 2wk After 3wk After 4wk After 5wk After 6wk US-guided * * * Blind NOTE. P values were obtained after the repeated-measures ANOVA. P values correspond to the mean difference between 2 groups. Values are mean SD. The total scores of functional activities were made by summations of 10 subitems. *P.01.

6 2002 ULTRASONOGRAPHY-GUIDED INTRA-ARTICULAR INJECTION, Lee group 1 week after the injections. If these changes were in the acute stage of the disease, it would have little clinical significance. However, all subjects were in the chronic stage of the capsulitis when there were very few chances of spontaneous reduction in patient s pain intensity in a short period. So, we think that in the chronic stage VAS changes of about 2 points in the first week after the intervention and 3 points in the second week have sufficient clinical significance. We think the changes in ROM and the shoulder function score in the chronic stage also have sufficient significance. The fact that the US-guided group showed more rapid improvements during the early stage of treatment could be attributed to the increased accuracy of the intra-articular injections. Eustace et al 16 reported that accurate steroid placement may be one of the important factors for the clinical outcome of patients with shoulder pain. Also, previous work 9 reported that the US-guided technique achieved a higher success rate in intraarticular injection than the blind technique. Therefore, we can say that the accurate intra-articular injection of drugs can be the major factor in the treatment effects we observed in the current study. Shortcomings of the blind technique arise because it is impossible to check the position of the needle. It would be difficult to ensure proper needle positioning into a shrunk capsule in the chronic stage. The needle may slide out of the articular capsule because of swelling of the articular cavity during injection. In this case, the drug may be injected outside the cavity because the needle may fail to be positioned between the head of the humerus and the glenoid fossa. US-guided injection offers the advantage of being able to accurately position the needle in the articular cavity. However, its minor disadvantage is that the doctor performing the procedure also needs to be experienced in the use of US. To see the needle properly, the probe needs to be accurately stuck to the patient s skin, and knowledge of the sectional structure and images of shoulder joint have to be sufficient. However, if sufficient knowledge and experience are available, the US-guided technique does seem to be advantageous. This brings up the question as to why there were no differences of improvements between the 2 treatments beyond about the third week. We believe that there are 2 possible explanations for this occurrence. First, there was a total of 6 injections performed on each patient. Although the accuracy of intraarticular injection is low in the blind technique, the probability of accurate injection of drugs into the articular cavity may increase if several successive operations are performed. Second, although the drugs may be injected outside the articular capsule, the drugs will spread out and come into contact with the articular capsule, which could result in some degree of anti-inflammatory effect. Regarding the cost-effectiveness of the US-guided technique, this technique may be less cost-effective in the circumstance of high medical fee for US guidance or of being a skillful operator with blind intra-articular injection. However, the cost may not be a big burden to patients in the circumstance of low medical fee for US. Moreover, this technique will be beneficial to physicians in case the physician is not good at the blind procedure or is a beginner in the procedure because the US can guide the needle tip into the intra-articular space accurately in real time. Study Limitations The limitation of this study is the relatively small sample size in each group. For more confirmative recognition of the usefulness of the US-guided technique, several similar studies with a larger number of patients in multiple centers or studies with 2 or 3 serial injections of 1 substance under US guidance are needed. CONCLUSIONS We found that for intra-articular injection for patients with adhesive capsulitis, the US-guided technique offers faster reduction of pain and higher improvement of the range of joint motion and general shoulder functions during the early stage of treatment compared with the blind technique. Therefore, we believe that the US-guided injection technique can be a useful treatment that leads to earlier improvements in patients with adhesive capsulitis. References 1. Anthon HA. Frozen shoulder. Can Fam Physician 1993;39: Awan R, Smith J, Boon AJ. Measuring shoulder internal rotation range of motion: a comparison of 3 techniques. Arch Phys Med Rehabil 2002;83: Neviaser TJ. Adhesive capsulitis. Orthop Clin North Am 1987; 18: Rizk TE, Gavant ML, Pinals RS. Treatment of adhesive capsulitis (frozen shoulder) with arthrographic capsular distension and rupture. Arch Phys Med Rehabil 1994;75: Blaine T, Moskowitz R, Udell J, et al. Treatment of persistent shoulder pain with sodium hyaluronate: a randomized, controlled trial. A multicenter study. J Bone Joint Surg Am 2008;90: Cho KH, Song JY, Lee H, Kim JS, Rhee YG. The effect of subacromial bursa injection of hyaluronate in patients with adhesive capsulitis of shoulder joint multicenter, prospective study. J Korean Acad Rehab Med 2002;26: Kim HW, Kim HS, Ahn KH. The effect of intraarticular injection of hyaluronic acid and steroid in adhesive capsulitis of shoulder. J Korean Acad Rehab Med 1999;23: Park SK, Nam MH, Yune SH. Hydraulic management of frozen shoulder. J Korean Acad Rehab Med 1992;16: Yi TI, Kim ST, Kim DH, Kim JS, Park JS, Lee JH. Comparison of blind technique and ultrasonography guided technique of intraarticular injection of the shoulder. J Korean Acad Rehab Med 2006;30: Finnoff JT. Musculoskeletal problems of the upper limb. In: Braddom RL, Bushbacher RL, editors. Physical medicine and rehabilitation. 3rd ed. Philadelphia: Saunders; p O Dell MW, Lin CD, Panagos A, Fung NQ. The physiatric history and physical examination. In: Braddom RL, Bushbacher RL, editors. Physical medicine and rehabilitation. 3rd ed. Philadelphia: Saunders; p Yoon JS, Kim SJ, Park EM. The change of blood sugar level after steroid injection in shoulder pain with diabetes mellitus. J Korean Acad Rehab Med 2003;27: De Jong BA, Dahmen R, Hogemwg JA. Intra-articular triamcinolone acetonide injection in patients with capsulitis of the shoulder: a comparative study of two dose regimens. Clin Rehabil 1998;12: Park GY, Hwang SE. Comparison of intraarticular steroid injection with and without capsular distension in adhesive capsulitis of the shoulder. J Korean Acad Rehab Med 2000;24: Rovetta G, Monteforte P. Intraarticular injection of sodium hyaluronate plus steroid versus steroid in adhesive capsulitis of the shoulder. Int J Tissue React 1998;20: Eustace JA, Brophy DP, Gibney RP, Bresnihan B, FitzGerald O. Comparison of the accuracy of steroid placement with clinical outcome in patients with shoulder symptoms. Ann Rheum Dis 1997;56: Suppliers a. GE Healthcare Bio-Sciences Corp, 800 Centennial Ave, PO Box 1327, Piscataway, NJ b. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL