Subacromial impingement syndrome is a common cause of pain in the shoulder22. Many non-operative treatments have been advocated, including rest; oral administration of non-steroidal anti-inflammatory drugs; physical therapy; and local modalities such as cryotherapy, ultrasound, electromagnetic radiation, and subacromial injection of corticosteroids9,15,24. While several investigators have attempted to evaluate the effects of subacromial injection of corticosteroids, the studies have involved heterogeneous groups of patients, with the causes of the pain in the shoulder including other etiologies in addition to subacromial impingement1,3,8,12,16,27,29,30. Furthermore, these studies were often not controlled for the use of other treatment modalities.
Subacromial injection of corticosteroids may be justified if it provides demonstrable clinical relief of subacromial impingement syndrome. If this is not the case, however, its use is not warranted, as tendon-weakening and possible destruction of the cartilage are associated risks8,10,13,15,17,19,20,25,26. The purpose of this prospective, randomized, controlled, double-blind study was to analyze the short-term efficacy of subacromial injection of corticosteroids for subacromial impingement syndrome.
Materials and Methods
The study was approved by the Institutional Review Board at our institution. All patients who were seen in the shoulder clinic from September 1992 to January 1993 with pain in the shoulder that was characteristic of subacromial impingement syndrome were asked to enroll in the study if they met these criteria: (1) the symptoms had lasted for at least three months, (2) a diagnosis of subacromial impingement syndrome had been made on the basis of the lidocaine injection test23, (3) the patient had not had previous subacromial injections of corticosteroids, (4) there was no evidence of os acromiale on plain radiographs, (5) the patient was not involved in a Workers' Compensation claim related to the shoulder, and (6) there was no clinical or radiographic evidence of a full-thickness tear of the rotator cuff. To identify and exclude patients who had a tear of the rotator cuff, double-contrast arthrography or magnetic resonance imaging was performed for all patients who had signs of muscular atrophy and weakness on forward elevation or external rotation after subacromial injection of lidocaine as well as for all patients who were more than sixty years old.
Forty patients (eight men and thirty-two women) who met the inclusion criteria agreed to participate in the study. Patients were randomized to receive either six milliliters of 1 per cent lidocaine without epinephrine (the control group) or two milliliters containing forty milligrams of triamcinolone acetonide per milliliter with four milliliters of 1 per cent lidocaine without epinephrine (the corticosteroid group). The corticosteroid group comprised nineteen patients (four men and fifteen women) whose mean age was fifty-six years (range, thirty-two to eighty years); the mean duration of symptoms before the injection was eight months (range, three to twenty-four months). The control group comprised twenty-one patients (four men and seventeen women) whose mean age was fifty-seven years (range, thirty-three to eighty-one years); the mean duration of symptoms before the injection was also eight months (range, three to twenty-two months). Fourteen dominant and five non-dominant shoulders were involved in the corticosteroid group, compared with thirteen dominant and eight non-dominant shoulders in the control group. Fourteen patients in the corticosteroid group described the onset of the symptoms as spontaneous and five, as following an injury. Sixteen patients in the control group had a spontaneous onset of pain and five had had a traumatic injury. All of the patients were able to localize the pain to the shoulder. In addition, three patients in the corticosteroid group and seven patients in the control group had neck pain, and five patients in each group had pain in the proximal part of the upper extremity on the affected side.
A functional assessment was performed by one of us (B. B.) to assess the performance of five activities of daily living. These included the ability to use a back pocket, to wash the opposite axilla, to eat with utensils, to wash or comb the hair, and to perform toilet functions. A score of 0 points indicated that the patient was unable to perform the activity; a score of 1 point, that the patient was able to but with difficulty; and a score of 2 points, that the patient was able to without difficulty. The mean scores, before the injection, for the corticosteroid group and the control group, were 0.8 and 0.8 point for using a back pocket, 1.2 and 1.2 points for washing the opposite axilla, 1.6 and 1.8 points for eating with utensils, 1.3 and 1.4 points for washing or combing the hair, and 1.5 and 1.7 points for performing toilet functions, respectively.
An over-all subjective assessment of pain was performed with use of a 4-point self-administered pain-appraisal score: 0 points indicated no pain; 1 point, mild pain; 2 points, moderate pain; and 3 points, severe pain. Mild pain was defined as occasional discomfort that occurred only after strenuous activity. Moderate pain was more constant and occurred with most overhead activity. Severe pain was constant and affected the performance of all activities of daily living. One patient in the corticosteroid group had mild pain, ten had moderate pain, and eight had severe pain before the injection, compared with two, ten, and nine patients, respectively, in the control group. With the numbers available, there was no significant difference between the mean pain score of 2.4 points for the corticosteroid group and that of 2.3 points for the control group. At the most recent follow-up examination, the patients were asked to rate the pain as decreased, unchanged, or worse compared with that before the injection.
A detailed physical examination was performed to document muscle atrophy, areas of localized tenderness, and the range of motion. The active range of motion of the shoulder was recorded with a goniometer. Forward elevation was measured as the angle between the arm and the trunk, external rotation was measured with the arm at the side, and internal rotation was measured as the most cephalad level on the back that the patient was able to reach with the thumb. The presence of an impingement sign, as described by Neer23, was determined.
All injections were administered by one of us (B. B.), who was independent of the examiner, in an opaque syringe with a 21-gauge needle. The needle was inserted under the anterolateral aspect of the acromion, with care taken to avoid injection directly into the tendons of the rotator cuff by documenting unrestricted flow of the solution through the syringe. After the injection, any immediate adverse reactions were noted.
All patients participated in a standardized program of physical therapy that consisted of passive, assisted, and active range-of-motion exercises for the shoulder in forward elevation, external rotation, and internal rotation as well as isometric exercises for the rotator cuff and the deltoid. As pain decreased and the range of motion increased, Theraband (Hygenics, Akron, Ohio) strengthening exercises for the rotator cuff muscles and deltoid were added. Physical therapy modalities such as cryotherapy, ultrasound, and electrical stimulation were specifically not used.
All patients were re-examined, with assessment of function and pain, every four weeks until completion of the study. Statistical analysis of the data was performed with the chi-square test and the Student unpaired t test, with a p value of less than 0.05 considered to be significant. Before the injection, there were no significant differences between the groups with regard to any parameters, including the gender and age of the patient, the duration of the symptoms, the shoulder involved, the nature of the onset of the symptoms, the location of the pain, the performance of activities of daily living, and the severity of the pain.
The mean duration of follow-up was thirty-three weeks (range, twelve to fifty-five weeks) for the corticosteroid group and twenty-eight weeks (range, twelve to fifty-two weeks) for the control group.
At the most recent follow-up examination, one patient in the corticosteroid group had severe pain, two had moderate pain, and sixteen had mild pain, compared with five, ten, and six patients, respectively, in the control group. The mean pain score after the injection was 1.2 points for the corticosteroid group and 2.0 points for the control group. This difference was significant (p < 0.005).
There was also a significant difference (p < 0.05) between the two groups with regard to the perception of pain after the treatment. Sixteen patients in the corticosteroid group stated that the pain had decreased after the treatment, while three stated that the pain was unchanged. Eight patients in the control group stated that the pain had decreased, twelve stated that it was unchanged, and one patient stated that it was worse (Fig. 1).
At the most recent follow-up examination, fifteen patients in the corticosteroid group had a negative impingement sign23, while only four patients in the control group had that finding. This difference was significant (p < 0.005). The active range of forward elevation had improved for fourteen patients in the corticosteroid group and for ten patients in the control group. The mean increases were 24 and 10 degrees for the corticosteroid and control groups, respectively. The active range of external rotation had improved for twelve patients in the corticosteroid group and for seven patients in the control group. The mean increases were 11 and 5 degrees in the corticosteroid and control groups, respectively. These differences were significant (p < 0.005). There was no significant difference between the two groups with regard to the change in internal rotation.
There was no significant difference between the two groups with regard to improvement in the performance of activities of daily living. The mean scores for the corticosteroid group and the control group were 1.7 and 1.4 points for using a back pocket, 1.8 and 1.7 points for washing the opposite axilla, 1.9 and 1.9 points for eating with utensils, 1.8 and 1.4 points for washing or combing the hair, and 1.9 and 1.9 points for performing toilet functions, respectively.
There were no complications related to the treatment in either group.
Subacromial impingement syndrome is caused by narrowing of the subacromial space with secondary impingement of the bursal surface of the rotator cuff on the undersurface of the acromion7,11,14,18,29. It does not appear to be self-limited given the fact that many patients are still symptomatic more than a year after first being seen by a physician because of pain and decreased function6.
In order to be successful, non-operative treatment must be directed toward the reduction of pain and the improvement of function. Toward this end, a multitude of therapies have been advocated, including rest, stretching and strengthening exercises, non-steroidal anti-inflammatory medications, and various local modalities; the reported results have varied widely9,15,24.
Several investigators have attempted to determine the efficacy of injecting corticosteroids as non-operative treatment for subacromial impingement syndrome. These studies, however, included patients who either had a tear of the rotator cuff8,12,29 or had a relatively short follow-up1,30.
One of the most striking findings in the present study was the absence of the Neer impingement sign23 at the latest follow-up examination in fifteen of the patients who had received corticosteroids compared with only four of those in the control group. This implies that the corticosteroid had a lasting effect.
It is interesting to note the strong placebo effect seen in the control group with regard to the subjective evaluation, especially with respect to pain. This is in contrast with the objective evaluation of improved motion and reversal of the impingement sign, which remained superior in most of the patients in the corticosteroid group.
Subacromial injection of corticosteroids does have potential side effects. Laboratory studies have shown temporary loss of strength after intratendinous injection of corticosteroids; however, few long-term deleterious effects have been noted15,19. Clinically, direct injection into the substance of tendons has been reported to cause late rupture10,13,20,26. Subcutaneous atrophy has also been reported after the injection of corticosteroids6,25. In addition, corticosteroids may have a deleterious effect on articular cartilage. However, this would occur either through an inadvertent injection into the glenohumeral joint instead of the subacromial space or as a result of a full-thickness tear of the rotator cuff allowing communication between the subacromial space and the glenohumeral joint. Injection of corticosteroids in the presence of a tear of the rotator cuff may also damage the tissue available for later repair. Bjorkenheim et al. found that a series of more than three preoperative injections of corticosteroids was associated with failure of repair of the rotator cuff5. Others have reported similar findings2,21. Watson noted that a series of four preoperative injections of corticosteroids or more was associated with a so-called softer residual cuff (as determined by an intraoperative pull-out suture test) and a poorer outcome28. Therefore, it is recommended that, at most, two subacromial injections of corticosteroids be given. Multiple injections should probably be avoided in patients who have a known history of a full-thickness tear of the rotator cuff.
A potential criticism of the present study is that both treatment groups had physical therapy. In order to isolate entirely the effects of the corticosteroid on subacromial impingement syndrome, ideally a group of patients who received a subacromial injection but did not have physical therapy should also have been included. Another problem with our study is the duration of follow-up. Although we followed our patients for a longer period than patients have been followed in previous studies1,30 on this subject of which we are aware, our study failed to provide information on the recurrence of symptoms of impingement after a mean of thirty weeks. In addition, we were unable to determine if patients who receive subacromial injection of corticosteroids are at increased risk for tears of the rotator cuff or, alternatively, if such injections decrease subacromial scarring, which may in turn decrease the risk of tears of the rotator cuff.
Although unlikely, lidocaine itself may have a therapeutic effect4. Infusion of corticosteroids with lidocaine is necessary to decrease the possibility of acute discomfort caused by the corticosteroids. As such, the placebo contained lidocaine in order for the study to remain double-blinded.
Finally, it has been demonstrated that patients with subacromial impingement syndrome for whom non-operative treatment has failed frequently continue to perform the inciting activities6. We did not attempt to determine if this was the case for our patients who were unresponsive to such treatment. If more patients in the control group than in the corticosteroid group continued to perform the inciting activities, then the results would have been skewed in favor of the corticosteroid group.
We realize that there was a wide range (twelve to fifty-five weeks) in the duration of follow-up, and a more objective study would present equal durations for each patient. The intent of this study, however, was to analyze the short-term efficacy of subacromial injection of corticosteroids for the treatment of subacromial impingement syndrome. Our results indicate that such injections are effective in the short term for alleviating pain and improving the range of motion of the shoulder.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
Investigation performed at The Shoulder Institute, the Hospital for Joint Diseases, New York City
1. Adebajo, A. O.; Nash, P.; and |and |Hazleman, B. L.: A prospective double blind dummy placebo controlled study comparing triamcinolone hexacetonide injection with oral diclofenac 50 mg TDS in patients with rotator cuff tendinitis. J. Rheumatol.
, 17: 1207-1210, 1990.
2. Bayne, O., and Bateman, J. E.: Long-term results of surgical repair of full-thickness rotator cuff tears. In Surgery of the Shoulder, pp. 167-171. Edited by J. E. Bateman and R. P. Welsh. Philadelphia, B. C. Decker, 1984.
3. Berry, H.; Fernandes, L.; Bloom, B.; Clark, R. J.; and |and |Hamilton, E. B.: Clinical study comparing acupuncture, physiotherapy, injection and oral anti-inflammatory therapy in shoulder-cuff lesions. Curr. Med. and Res. Opin.
, 7: 121-126, 1980.
4. Binder, A.; Parr, G.; Hazleman, B.; and |and |Fitton-Jackson, S.: Pulsed electromagnetic field therapy of persistent rotator cuff tendinitis. A double-blind controlled assessment. Lancet
, 1: 695-698, 1984.
5. Bjorkenheim, J. M.; Paavolainen, P.; Ahovuo, J.; and |and |Slatis, P.: Surgical repair of the rotator cuff and surrounding tissues. Factors influencing the results. Clin. Orthop.
, 236: 148-153, 1988.
6. Chard, M. D.; Sattelle, L. M.; and |and |Hazleman, B. L.: The long-term outcome of rotator cuff tendonitis—a review study. British J. Rheumatol.
, 27: 385-389, 1988.
7. Cofield, R. H.: Current concepts review. Rotator cuff disease of the shoulder. J. Bone and Joint Surg.
, 67-A: 974-979, July 1985.
8. Darlington, L. G., and |and |Coomes, E. N.: The effects of local steroid injection for supraspinatus tears. Rheumatol. and Rehab.
, 16: 172-179, 1977.
9. Downing, D. S., and |and |Weinstein, A.: Ultrasound therapy of subacromial bursitis. A double blind trial. Phys. Ther.
, 66: 194-199, 1986.
10. Ford, L. T., and |and |DeBender, J.: Tendon rupture after local steroid injection. Southern Med. J.
, 72: 827-830, 1979.
11. Hawkins, R. J., and |and |Brock, R. M.: Anterior acromioplasty: early results for impingement with intact rotator cuffs. Orthop. Trans.
, 3: 274, 1979.
12. Hollingworth, G. R.; Ellis, R. M.; and |and |Hattersley, T. S.: Comparison of injection techniques for shoulder pain: results of a double blind, randomised study. British Med. J.
, 287: 1339-1341, 1983.
13. Ismail, A. M.; Balakrishnan, R.; Rajakumar, M. K.; and |and |Lumpur, K.: Rupture of patellar ligament after steroid infiltration. Report of a case. J. Bone and Joint Surg.
, 51-B(3): 503-505, 1969.
14. Jackson, D. W.: Chronic rotator cuff impingement in the throwing athlete. Am. J. Sports Med.
, 4: 231-240, 1976.
15. Kennedy, J. C., and |and |Willis, R. B.: The effects of local steroid injections on tendons: a biomechanical and microscopic correlative study. Am. J. Sports Med.
, 4: 11-21, 1976.
16. Kessel, L., and |and |Watson, M.: The painful arc syndrome. Clinical classification as a guide to management. J. Bone and Joint Surg.
, 59-B(2): 166-172, 1977.
17. Lund, I. M.; Donde, R.; and |and |Knudsen, E. A.: Persistent local cutaneous atrophy following corticosteroid injection for tendinitis. Rheumatol. and Rehab.
, 18: 91-93, 1979.
18. Matsen, F. A., III, and Arntz, C. T.: Subacromial impingement. In The Shoulder, pp. 623-646. Edited by C. A. Rockwood, Jr., and F. A. Matsen, III. Philadelphia, W. B. Saunders, 1990.
19. Matthews, L. S.; Sonstegard, D. A.; and |and |Phelps, D. B.: A biomechanical study of rabbit patellar tendon: effects of steroid injection. J. Sports Med.
, 2: 349-357, 1974.
20. Melmed, E. P.: Spontaneous bilateral rupture of the calcaneal tendon during steroid therapy. J. Bone and Joint Surg.
, 47-B(1): 104-105, 1965.
21. Nasca, R. J.: Surgical treatment of complete rotator cuff tears. In Surgery of the Shoulder, pp. 149-153. Edited by J. E. Bateman and R. P. Welsh. Philadelphia, B. C. Decker, 1984.
22. Neer, C. S., II: Anterior acromioplasty for the chronic impingement syndrome in the shoulder: a preliminary report. J. Bone and Joint Surg.
, 54-A: 41-50, Jan. 1972.
23. Neer, C. S., II: Impingement lesions. Clin. Orthop.
, 173: 70-77, 1983.
24. Rocks, J. A.: Intrinsic shoulder pain syndrome: rationale for heating and cooling in treatment. Phys. Ther.
, 59: 153-159, 1979.
25. Rostron, P. K., and |and |Calver, R. F.: Subcutaneous atrophy following methylprednisolone injection in Osgood-Schlatter epiphysitis. J. Bone and Joint Surg.
, 61-A: 627-628, June 1979.
26. Sweetnam, R.: Corticosteroid arthropathy and tendon rupture [editorial]. J. Bone and Joint Surg.
, 51-B(3): 397-398, 1969.
27. Valtonen, E. J.: Double acting betamethasone (Celestone Chronodose) in the treatment of supraspinatus tendinitis: a comparison of subacromial and gluteal single injections with placebo. J. Internat. Med. Res.
, 6: 463-467, 1978.
28. Watson, M.: Major ruptures of the rotator cuff. The results of surgical repair in 89 patients. J. Bone and Joint Surg.
, 67-B(4): 618-624, 1985.
29. White, R. H.; Paull, D. M.; and |and |Fleming, K. W.: Rotator cuff tendinitis: comparison of subacromial injection of long acting corticosteroid versus oral indomethacin therapy. J. Rheumatol.
, 13: 608-613, 1986.
30. Withrington, R. H.; Girgis, F. L.; and |and |Seifert, M. H.: A placebo-controlled trial of steroid injections in the treatment of supraspinatus tendonitis. Scandinavian J. Rheumatol.
, 14: 76-78, 1985.