J Shoulder Elbow Surg (2013) 22, 595-601 www.elsevier.com/locate/ymse A double-blind randomized controlled trial comparing the effects of subacromial injection with corticosteroid versus NSAID in patients with shoulder impingement syndrome Kyong Su Min, MDa,*, Patrick St. Pierre, MDb, Paul M. Ryan, MDa, Bryant G. Marchant, MDa, Christopher J. Wilson, MDc, Edward D. Arrington, MDa aDepartment of Surgery, Orthopaedic Service, Madigan Healthcare System, Tacoma, WA, USA bDesert Orthopedic Center, Rancho Mirage, CA, USA cHand Surgery Associates, Inc., Sacramento, CA, USA Hypothesis: The objective of this study was to compare the efficacy of subacromial injection of triamcin- olone compared to injection of ketorolac in the treatment of external shoulder impingement syndrome. Methods: Thirty-two patients diagnosed with external shoulder impingement syndrome were included in this double-blinded randomized controlled clinical trial. Each patient was randomized into the steroid group or nonsteroidal anti-inflammatory drugs (NSAID) group. The steroid syringe contained 40 mg triam- cinolone; and the NSAID syringe contained 60 mg ketorolac. Each patient was evaluated in terms of arc of motion, visual analog scale (VAS) for evaluating pain, and the UCLA (The University of California at Los Angeles) shoulder rating scale. Results: At 1 month follow-up, both treatment arms resulted in increased range of motion and decreased pain. The steroid group decreased in active abduction while the NSAID group increased (steroid: 134, NSAID: 151, P ¼ .03). The mean improvement in the UCLA shoulder rating scale at 4 weeks was 7.15 for the NSAID group and 2.13 for the steroid group (P ¼ .03). Subgroup analysis of the UCLA scale demonstrated an increase in both forward flexion strength (P ¼ .04) and patient satisfaction (P ¼ .03) in the NSAID group. No significant difference could be seen in all other outcome measures. Conclusion: In this study, an injection of ketorolac resulted in greater improvements in the UCLA shoulder rating scale than an injection of triamcinolone at 4 weeks follow-up. While both triamcinolone and ketorolac are effective in the treatment of isolated subacromial impingement, ketorolac appears to have equivalent if not superior efficacy; all the while decreasing patient exposure to the potential side- effects of corticosteroids. This study was reviewed and approved by the Institutional Review Board *Reprint requests: Kyong Su Min, Department of Surgery, Orthopaedic at Madigan Army Medical Center (US Department of the Army, Combined Service, Madigan Healthcare System, 9040A Fitzsimmons Dr, Tacoma, Meeting of the Clinical Investigation Committee and the Human Use WA 98431, USA. Committee): Study No: 200125. E-mail address: [email protected] (K.S. Min). Therewas no external sources offunding for this investigation. This study was registered in Clinical Trials, a service of the United States National Institute of Health (clinicaltrials.gov), and the registry identifier is NCT01449448. 1058-2746/$ - see front matter Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.jse.2012.08.026 596 K.S. Min et al. Level of evidence: Level II, Randomized Controlled Trial, Treatment Study. Published by Elsevier Ltd. Keywords: Subacromial; external impingement; steroid; corticosteroid; NSAID; ketorolac; injection Subacromial impingement syndrome is a commonly to palpation about the acromion, positive Neer’s sign, positive treated shoulder condition.3,4 The pain associated with Hawkin’s sign, and pain exacerbated with the shoulder held in external impingement syndrome is thought to be secondary internal rotation (Table I). In addition, all patients had standard to subacromial bursitis and tendonitis of the rotator cuff radiographs of their affected shoulder to rule out glenohumeral resulting from encroachment of the coracoacromial arch on arthritis. Patients received education regarding the study, and they were asked to provide informed consent to participate. The the underlying bursa and supraspinatus tendon.15,24,25 study’s purpose, procedures, benefits, risks, inconveniences, Intrinsic rotator cuff dysfunction allowing superior discomforts, and alternative to participation were discussed with humeral head migration, and other factors such as scapular each patient. dyskinesis may also contribute to external shoulder In order to ensure a double-blinded evaluation, the following impingement. Many treatment modalities have been measures were taken: pre-prepared, unlabeled, and unidentifiable employed in attempts to relieve pain and restore function syringes of injectable solution were used; the syringe preparations of the affected shoulder.1,5,23,34 Nonoperative treatment were made by the pharmacy with a numbering system; each options for impingement syndrome have included rest, ice, syringe was randomly assigned a number by the preparing phar- physical therapy, ultrasound, electromagnetic radiation, macist; all of the syringes were individually covered in white tape corticosteroid injections, and systemic nonsteroidal anti- to blind the contents of the syringe; the contents of each syringe inflammatory drugs (NSAIDs). was kept in a sealed envelope; and neither the injector, evaluator, or patient knew the contents of the syringe. The control (steroid) Corticosteroid injections are widely utilized and are syringe contained 6 cc of 1% lidocaine with epinephrine and a well-accepted method of treatment for patients with 40 mg triamcinolone; and the test (NSAID) syringe contained 6 cc subacromial impingement syndrome who have failed of 1% lidocaine with epinephrine and 60 mg ketorolac. 6 more conservative therapies. Multiple studies have sup- Prior to receiving the injection, each patient completed ported its efficacy, although the exact mechanism of corti- a UCLA (The University of California at Los Angeles) shoulder costeroid injections is not completely understood.6,7 It is rating scale and visual analog scale (VAS) for evaluating pain. A commonly accepted that the relief attributed to subacromial hand held goniometer was used to measure the active and passive corticosteroid injections is related to its anti-inflammatory ranges of motion (ROMs) in forward flexion (FF) and abduction of properties.8 Unfortunately, the frequency of use of corti- the affected shoulder. Upon completion of the initial evaluation, costeroid injections is limited because of its potentially each qualified patient was given an injection into the subacromial serious side effects. Corticosteroids have been associated space with a standardized posterolateral approach. The landmarks for the injection were the acromion and the coracoid process. The with tendon rupture, subcutaneous atrophy, articular carti- 12,22,31 posterolateral shoulder was prepped with alcohol and the needle lage changes, and systemic effects like osteoporosis. was inserted 2 cm inferior and 1 cm lateral to the inferolateral If steroids are effective because of their anti-inflammatory edge of the acromion. The needle was directed from posterior to properties, there is an argument to use local injections of anterior in the direction of the coracoid process. A reduction of nonsteroidal anti-inflammatory drugs to decrease inflam- pain with increased active abduction after injection clinically mation in the subacromial space. There is evidence that confirmed accurate placement.25 Approximately 5 minutes after subacromial injections of NSAIDs may provide pain relief the injection, ROMs were reassessed. All of the patient were and restoration of function in shoulder impingement instructed to follow-up in 4 weeks.9 At the 4-week visit, the syndrome.16,19 The objective of this double-blind random- UCLA shoulder rating scale, VAS for evaluating pain, and ROMs ized controlled trial was to compare the efficacy of sub- were reassessed for the final time. acromial injection of triamcinolone compared to injection of The primary outcome measure was the UCLA shoulder rating scale.13,32 This scoring system consists of subjective assessments ketorolac. We hypothesize that the efficacy of subacromial of pain, function, and satisfaction, as well as objective measure- injection of triamcinolone will be equivalent to the efficacy ments of active forward elevation and strength in forward flexion. of ketorolac. Secondary outcome measures were the VAS for evaluating pain and goniometric measurements of forward flexion and abduction. A 2-tailed Student’s t test was used to compare the outcome Materials and methods measures of the 2 treatment groups and the level of statistical significance was set at P < .05. A power analysis was used to Patients were recruited from a sports medicine orthopaedic calculate the required sample size. Assuming a normal distribution, surgery clinic in a tertiary care medical center. Briefly, the a standard deviation of 6 points and with a power of 80%, a sample inclusion criteria were shoulder pain with passive and/or active size of 13 patients for each arm of the trial was required in order to abduction, diagnosis of subacromial bursitis based on tenderness detect an 8-point difference in the UCLA shoulder rating scale.20 Subacormial impingement: steroid vs NSAID injection 597 Table I Inclusion and inclusion criteria Inclusion criteria Exclusion criteria 1. Shoulder pain with passive and/or active abduction in the 1. Age <18 years 60-120 arc of motion 2. Symptoms less than one month 2. Positive Neer’s Test 3. Previous shoulder injections within the past 3 months 3. Positive Hawkin’s Test 4. Evidence of os-acromiale or other