Shoulder Subacromial Injection 

Updated: Feb 16, 2021
Author: Brett J Rothaermel, MD, PT; Chief Editor: Erik D Schraga, MD 



Impingement syndrome and rotator cuff disorders are common causes of shoulder pain.[1]  These conditions are difficult to differentiate clinically. Both are chronic, painful conditions that result from inflammation, damage, or both to the structures that lie within the subacromial space (including the subacromial bursa and the rotator cuff tendons).

The two conditions are treated in the same manner initially. Nonoperative treatment should begin with measures to reduce pain and inflammation, including activity modification, modalities, and nonsteroidal anti-inflammatory drugs (NSAIDs).[2]  If this is not effective, some patients may benefit from a subacromial corticosteroid injection.

Corticosteroid injections can decrease inflammation and improve function,[3]  thereby permitting improved range of motion (ROM) and facilitating rehabilitative and strengthening exercises.[4]  In general, subacromial injection with corticosteroids should not be performed more than four times per year. Combining such injections with exercise may enhance their effectiveness.[5]

A study by Lee et al tested botulinum toxin type B for subacromial injections and concluded that injections with this agent may be an alternative to steroid injections.[6]

In a study of 40 patients with shoulder impingement syndrome with findings of rotator cuff tendinitis or subacromial bursitis on magnetic resonance imaging (MRI), three 20-mg injections of the nonsteroidal anti-inflammatory drug (NSAID) tenoxicam at weekly intervals yielded results comparable to those of a single 40-mg injection of methylprednisolone acetate.[7]

However, a study comparing single-dose subacromial injections of betamethasone (7 mg) and lornoxicam (8 mg) in 70 patients with subacromial impingement syndrome found that whereas the patients in the NSAID group achieved rapid functional recovery that partially extended into the intermediate term, the results were inferior to those seen in the steroid group.[8] The authors suggested that a single lornoxicam injection may be an alternative only in cases where corticosteroids are contraindicated.

Injection of platelet-rich plasma (PRP) has also been employed for rotator cuff disease. In a study by Shams et al that included 40 patients with symptomatic partial rotator cuff tears, PRP injection yielded better early results than corticosteroid injection, though after 6 months, there was no longer a statistically significant improvement.[9] A study by von Wehren et al found similar results.[10]

The accuracy of a blind anatomic approach to subacromial steroid injections is a subject of investigation. In a small study that employed a mix of depot methylprednisolone and iopamifodol (a radiographic contrast medium), followed by radiographic assessment 6 weeks later to determine intra-or extra-articular placement of the steroid, only two of the 20 patients who received shoulder steroid injections were found to have intra-articular placement.[11] Radiologic placement was uncertain for 12 of the 20, and the remaining six were found to have extra-articular placement. Another study found the blind approach to be only 29% accurate for subacromial bursa injections.[12]

Ultrasonographic (US) guidance may improve the accuracy of the injection. A randomized controlled clinical trial showed US-guided injections to be substantially more accurate (63%) than blind injections (40%).[13]


Like most joint and soft-tissue injections with corticosteroids or anesthetics, subacromial injections can be either diagnostic or therapeutic.[2]

Diagnostic joint injection, with or without a corticosteroid, can help determine whether shoulder pain involves the structures that lie within the subacromial space (ie, subacromial bursa or rotator cuff). This can help differentiate impingement syndrome from other shoulder disorders, such as osteoarthritis of the glenohumeral or acromioclavicular joints and labral tears. If pain resolves or decreases after injection, then the pain is likely to be attributable to inflammation of either the subacromial bursa or the rotator cuff.

Furthermore, subacromial injection is particularly helpful in differentiating between shoulder weakness caused by impingement, in which shoulder strength improves after injection, and a true rotator cuff tear (see the images below), in which no change in strength is noted after injection.[4]

Presence of contrast medium in subdeltoid-subacrom Presence of contrast medium in subdeltoid-subacromial bursa signals presence of complete rotator cuff tear.
Complete rotator cuff tear with presence of contra Complete rotator cuff tear with presence of contrast medium in subacromial-subdeltoid bursa. Also note multiple irregularities in synovial fluid, shown as multiple filling defects.

Therapeutic joint injection provides pain relief and functional improvement in symptomatic subacromial impingement syndrome, rotator cuff disorders, and adhesive capsulitis.[14] This is accomplished through delivery of the corticosteroid to either the subacromial bursa or the rotator cuff.


Bacteremia, cellulitis of overlying skin, and adjacent osteomyelitis are often considered absolute contraindications for subacromial injection because of the potential risk of seeding the joint with bacteria. In these situations, the procedure should only be performed when septic arthritis is strongly suspected as the cause of overlying inflammatory changes, and only after consultation with an orthopedist.

Relative contraindications include chronic infection distant from the injection site, allergy to the injectate, diabetes mellitus, and uncontrolled coagulopathy.


Periprocedural Care

Patient Education and Consent

Before the procedure, the risks, benefits, and alternatives should be discussed with the patient. Informed consent should be obtained.


Equipment employed for subacromial injection includes the following:

  • Sterile gloves
  • Bactericidal skin preparation solution
  • Needle, 1.5 in., 22-25 gauge
  • Syringe, 10 mL
  • Lidocaine 1% without epinephrine (or similar local anesthetic), 5 mL
  • Sterile gauze
  • Sterile adhesive

Additional equipment for either indirect or direct ultrasound-guided injections includes the following:

  • Ultrasound device with 7.5-15 MHz linear-array broadband probe 
  • Sterile (direct approach) or nonsterile (indirect skin-marking approach) ultrasound gel

Patient Preparation


Cooling spray or local anesthetic may be used. If a local anesthetic is chosen, use a separate 25-gauge needle to raise a wheal at the site of the injection by inserting the needle about 1 cm below the skin surface. (See Local Anesthetic Agents, Infiltrative Administration.)


The patient should be seated upright in a comfortable position with the arm hanging unsupported by the patient's side.



Injection Into Subacromial Space

Opinions differ as to whether a single-needle or a two-needle injection technique is preferable. The single-needle technique is less painful. The two-needle technique prevents the possibility of flocculation of steroid crystals in the local anesthetic after mixing. Proponents of the two-needle technique also suggest that injection of the local anesthetic first numbs the area, making the subsequent injection more tolerable. The single-needle technique is described below.

Single-needle technique

Seat the patient upright in a comfortable position, with the arm hanging unsupported by his or her side. Palpate and mark the distal posterolateral edge of the acromion. Prepare and drape the site in a sterile manner with povidone-iodine, chlorhexidine gluconate, or isopropyl alcohol. Wear sterile gloves during the procedure.

Insert the needle inferior to the posterolateral edge of the acromion, directing it medially and slightly anteriorly; this places the needle tip beneath the acromion. Always aspirate before injecting to confirm that the needle tip is not placed intravascularly.

Next, inject the 6-7 mL corticosteroid–local anesthetic preparation. The injectate should flow freely without any significant resistance. (See the video below.) Never inject if significant resistance is encountered. Reposition the needle, and reattempt insertion until minimal resistance is encountered. If the patient has a contraindication for the use of a corticosteroid, then a diagnostic block may be performed with only local anesthetic. 

In 77-year-old woman with long history of chronic right shoulder impingement, posterior approach for subacromial injection is performed. Video courtesy of James R Verheyden, MD.

After the injection, patients with impingement syndrome experience temporary relief of symptoms and increased range of motion and strength. In the setting of a rotator cuff tear, corticosteroid injections should be used judiciously. Such injections may decrease inflammation and provide short-term pain relief, but they also weaken the involved tendon.[3] If no improvement is observed after injection, further imaging is indicated.

In a systematic review and meta-analysis by Aly et al, ultrasound (US)-guided subacromial injection was found to be as accurate as landmark-guided injection and to have superior efficacy (ie, significant reduction in pain and improvement in function).[15]

Dress the injection site with a sterile adhesive bandage. Encourage the patient to ice the area immediately after the injection and to avoid strenuous activity with the involved shoulder for the remainder of the day.

Treatment of impingement or rotator cuff syndrome generally includes physical therapy.[4]  However, a 2016 Cochrane review did not find high-quality evidence indicating that manual therapy and exercise yielded significant improvements on patient-important outcomes for patients with rotator cuff disease.[16]

Ultrasound-guided injection: lateral approach

The lateral approach is the most common approach to US-guided injection into the subacromial-subdeltoid bursa. The following is a description of the technique preferred by Molini et al.[17] Their approach is to have the patient seated on the examination table with the back turned to the physician so as to keep the needle outside of the patient’s field of view. They prefer a 5-cm (2-in.) 23-gauge needle. 

Applying sterile gel, position the probe parallel to the long axis of the supraspinatus. Apply disinfectant to the area to be punctured, and puncture the skin about 2-3 cm from the probe so as to avoid contaminating the needle with the probe. When the needle has penetrated the subcutaneous tissues, it will be visible as a hyperechoic structure with posterior comet-tail artifact when the beam is perpendicular to the long axis. (See the image below.)

Ultrasonogram shows needle penetrating bursa (righ Ultrasonogram shows needle penetrating bursa (right side of image) prior to injection of medication. Courtesy of Nicholas Goyeneche, MD, Ochsner Health System.

The long-axis scan permits real-time visualization of the needle entering the subacromial bursa. Once the needle has entered the bursa, inject a small amount of fluid, and visualize a real-time spread of hyperechoic echoes in the bursa to confirm correct positioning. With correct positioning confirmed, inject the rest of the corticosteroid, and then remove the needle.

Molini et al request that the patient avoid loading the shoulder for 2-3 days after the injection.[17] They also inform the patient of the possibility of renewed pain in the hours following the injection secondary to localized trauma.


Complications are uncommon and often insignificant but include the following:

  • Iatrogenic infection - The risk of inducing joint infection is low when sterile technique is used
  • Injection of corticosteroids directly into a tendon or tendon insertion can sometimes result in tendon rupture
  • Corticosteroids may cause a transient rise in blood glucose levels in patients with diabetes mellitus
  • Subcutaneous corticosteroid injection may also cause skin hypopigmentation and fat atrophy

Questions & Answers