Effectiveness of High-Volume Corticosteroid Injection in Shoulder Impingement

Subacromial impingement syndrome (SIS) is commonly treated conservatively with corticosteroid (CS) injections into the subacromial bursa, often combined with local anesthetics (LA). Despite widespread use, there remains no consensus regarding the optimal injection technique, volume, or drug formulation. Some studies suggest high-volume injections may provide superior pain relief, while others show no significant difference. To clarify this, a prospective, double-blind, randomized controlled trial was designed to compare the efficacy and safety of high-volume versus low-volume CS injections in patients with SIS.

Material and methods:

This study is a prospective, double-blind, randomized controlled trial. The study procedure was discussed with the patients, and each participant provided written informed consent.

All patients who were admitted to the outpatient clinic with unilateral shoulder pain within the study period, were screened to assess eligibility for the study The following 3-step checklist was completed for an accurate diagnosis of SIS: history of anterolateral arm pain and exacerbation of symptoms upon raising the shoulder to 90° or higher, a combination of the Neer test, Hawkins-Kennedy test, painful-arc test, and infraspinatus muscle test produced at least 2 positive results and confirmation of the diagnosis by an ultrasound (US) evaluation of the affected shoulder.

The same physiatrist (who had 5 years of experience in musculoskeletal US) performed the complete standard US evaluation of the affected shoulder in each patient, according to the EURO-MUSCULUS/USPRM shoulder scanning protocol. All patients also underwent dynamic US evaluation for SIS . A 3-16-mHz linear probe was used in B mode (Samsung HM70A Plus, Seoul, South Korea). For the US diagnosis of SIS, the subacromial-subdeltoid bursa must have been thicker than 2 mm (bursitis) and a sign of impingement must have been seen on the dynamic evaluations. Any pathological abnormalities associated with SIS (tendinosis, partial-thickness rotator cuff tear, etc.) were also noted.

Outcome Measures:

Detailed in the Outcome Measurements section.

Randomization and blinding After baseline assessment, patients were randomly assigned (1:1) to high-volume (H-VCI) or low-volume (L-VCI) corticosteroid injection groups using computer-generated randomization by an independent secretary. A blinded research assistant prepared the injections. Patients and outcome assessors were blinded to group allocation, but the injector was not blinded due to the differing injection volumes.

Interventions Patients in the H-VCI group received 1 mL of 40 mg triamcinolone acetonide (TA), 1 mL of 2% lidocaine, and 8 mL of normal saline mixture (10 mL total). The patients in the L-VCI group received 1 mL of 40 mg TA, 1 mL of 2% lidocaine, and 1 mL of normal saline mixture (3 mL total).

The physiatrist administered all injections following a standard procedure with US . The patient was taken to the operating room and seated in a chair. The affected arm was placed in the modified Crass position, and the injection site was sterilized and draped. The transducer was placed on the anterosuperior aspect of the shoulder to capture the short axis of the supraspinatus tendon. The subacromial-subdeltoid bursa was visualized between the deltoid muscle and the supraspinatus tendon. Using the in-plane technique, a 21G, 38-mm needle was advanced from lateral to medial to reach the subacromial-subdeltoid bursa. After confirming that the needle tip was inside the bursa, the injectate was administered and its distribution appeared.

Injection-related adverse reactions were monitored and recorded over the following 1-2 h prior to discharge. Following the injections, patients were allowed to take paracetamol (and naproxen 750 mg daily if needed) and were asked to note the dosages that they consumed. Patients were instructed not to take analgesics the day before the follow-up appointment, as this could affect the study results.

All patients were trained once by an experienced physical therapist for a home exercise regimen and were then given a leaflet. The regimen consisted of a combination of range of motion, stretching, scapular stabilization, and rotator cuff resistance exercises. Each patient was instructed to begin this exercise program 4 days after the injection.

Data collection and assessment All patients completed standard questionnaires that collected demographic (age, gender, hand dominance, weight, and height) and current shoulder-pain-related data.

The primary outcome measure was the Shoulder Pain and Disability Index (SPADI) score, which evaluates shoulder pain and disability.

Secondary outcomes included Visual Analog Scale (VAS: pain at night, rest, and movement), Western Ontario Rotator Cuff Index Score (WORC) , and active range of motion (AROM: abduction, flexion, internal/external rotation).

All assessments were performed by a blinded researcher. Evaluations were conducted at baseline, 1 hour, 4 weeks, and 12 weeks post-injection. VAS was assessed at all time points; SPADI, WORC, and AROM were assessed at 4 and 12 weeks.

At 4 and 12 weeks, exercise adherence (never, sometimes, always) was also recorded.

Statistical Analysis Sample size was calculated using G*Power (v3.1.9.4) based on a 10-point minimal clinically important difference in SPADI, with a standard deviation of 14, α = 0.05, and power (β) = 80%, requiring 64 patients. To account for 10% attrition, 70 patients were enrolled.

Analyses were performed using SPSS 22.0 (IBM Corp., Armonk, NY, USA). Data were presented as mean ± SD or median (IQR) for continuous variables, and frequency (%) for categorical variables. Normality was assessed using the Shapiro-Wilk test. Independent t-test or Mann-Whitney U test was used for between-group comparisons, and chi-square test for categorical data.

All outcomes were analyzed based on the intention-to-treat principle, with missing data handled via multiple imputation. Within-group changes were analyzed using the Friedman test. Percent changes from baseline to 4 and 12 weeks were calculated and compared between groups. Post-hoc pairwise comparisons were conducted with the Dunn-Bonferroni test. Statistical significance was set at p < 0.05.