Ultrasound-guided subacromial-subdeltoid bursa injection for rotator cuff impingement syndrome
Subacromial pain syndrome (encompassing impingement, rotator cuff tendinopathy, and subacromial bursitis) is the most common cause of shoulder pain in adults, accounting for an estimated 44–65% of all shoulder complaints presenting to primary care. When conservative management fails, ultrasound-guided subacromial corticosteroid injection is a well-established next step.
Diagnosis is largely clinical, supported by imaging:
• History: pain with overhead activity, night pain (classically when lying on the affected side), gradual onset in repetitive overhead workers/athletes • Provocative tests: Neer sign (pain with passive forward flexion), Hawkins-Kennedy test (pain with forced internal rotation at 90° forward flexion), painful arc sign (60–120° of abduction) • Strength testing: weakness with resisted abduction (supraspinatus) or external rotation (infraspinatus) raises suspicion for a partial- or full-thickness cuff tear rather than isolated impingement/bursitis • Ultrasound: dynamic assessment during abduction can show mechanical impingement of the supraspinatus and bursa under the coracoacromial arch; static images assess tendon echotexture (tendinosis = heterogeneous thickened tendon), tears (focal hypoechoic/anechoic defect), and bursal thickness (>2 mm with or without Doppler signal = bursitis)
MRI or MR arthrography is reserved for suspected full-thickness tears, labral pathology, or failed conservative treatment being considered for surgery.
Indications: persistent pain and functional limitation despite NSAIDs, activity modification, and a structured physiotherapy program (typically 6–12 weeks), especially with sonographic bursitis or tendinosis without a large full-thickness tear.
Relative caution: full-thickness rotator cuff tears (corticosteroid does not address the structural defect and repeated injections may impair tendon healing or surgical repair quality), diabetes (transient hyperglycemia), anticoagulation (technique-dependent, generally low risk with small-gauge needles), and prior injection within the last 3 months at the same site.
Absolute contraindications mirror other joint/bursa injections: overlying infection, systemic sepsis, and known allergy to the injectate.
The subacromial-subdeltoid (SASD) bursa is a thin synovial-lined potential space between the deltoid muscle and the rotator cuff tendons/greater tuberosity, extending from beneath the acromion laterally over the proximal humerus. Correct positioning brings the bursa and cuff footprint out from under the acromial arch, maximizing the sonographic window and safe needle access.
Patient seated with the examiner/injector standing or seated behind or beside the shoulder. The modified Crass position — hand placed on the ipsilateral back pocket/hip with the elbow pointed posteriorly — internally rotates and extends the humerus, rotating the supraspinatus tendon and its overlying bursa anteriorly out from under the acromion, exposing more of the footprint for both scanning and injection.
Alternative neutral or slight-extension positions are used if internal rotation is limited by pain (as in adhesive capsulitis) or a large tear is suspected, since excessive rotation can retract a torn tendon further under the acromion.
Long-axis (parallel to tendon fibers) scanning over the anterolateral shoulder identifies, from superficial to deep:
• Skin and subcutaneous fat • Deltoid muscle (medium-echogenicity, pennate fiber pattern) • Peribursal fat (thin echogenic line) and the SASD bursa (normally a thin, <2 mm anechoic-to-hypoechoic stripe, often not clearly separable from peribursal fat when normal) • Supraspinatus tendon (fibrillar, convex "bird-beak" contour) inserting on the greater tuberosity • Humeral head cortex (bright hyperechoic line, posterior acoustic shadow)
Bursal pathology appears as focal or diffuse hypoechoic/anechoic thickening >2 mm, sometimes with internal septations, debris, or Power Doppler hyperemia in acute bursitis. Dynamic scanning during passive abduction can reveal mechanical "catching" of the thickened bursa/tendon under the coracoacromial ligament, correlating with the clinical painful arc.
Two commonly taught and validated approaches allow safe, accurate access to the SASD bursa: an in-plane approach along the long axis of the supraspinatus tendon (posterior or posterolateral entry) with continuous needle visualization, and an out-of-plane (short-axis) direct lateral approach beneath the acromion, favored by some proceduralists for its shorter needle path.
With the probe in long axis over the supraspinatus tendon, the needle is introduced from the posterolateral edge of the probe and advanced in-plane, parallel to the ultrasound beam, so the entire needle shaft and tip remain visible throughout. The needle is directed just superficial to the tendon surface and deep to the deltoid, entering the bursal potential space. This approach allows precise control of needle depth relative to the acromion above and the tendon below, minimizing risk of intratendinous injection (which can weaken collagen architecture).
Alternatively, with the probe in short axis over the lateral subacromial space, the needle is introduced out-of-plane (perpendicular to the beam) directly beneath the lateral edge of the acromion. Only a cross-sectional "dot" of the needle shaft/tip is visualized at any moment, requiring careful stepwise advancement and confirmation of the tip position by observing tissue displacement or the appearance of a bright dot at the expected depth just superficial to the tendon.
Both techniques have shown comparable accuracy (>90–95%) in comparative studies (Sethi et al., Zwar et al.), with technique choice generally guided by operator preference, patient body habitus, and the need to avoid the acromion or a lateral bony prominence blocking the in-plane trajectory.
Regardless of approach, the injector should feel minimal resistance during injection — high resistance suggests the needle tip is within the substance of the tendon (intratendinous), not the bursal space, and the needle should be repositioned before injecting.
Once the needle tip is confirmed within the SASD bursal space, the corticosteroid-anesthetic mixture is injected under direct real-time visualization. A successful injection is seen as the bursa distending with anechoic-to-mixed fluid, cleanly separating the deltoid from the underlying rotator cuff surface.
A typical mixture is 1 mL triamcinolone acetonide 40 mg/mL (or an equivalent dose of betamethasone or methylprednisolone) combined with 4–6 mL of local anesthetic, for a total volume of 5–7 mL — the SASD bursa is a relatively large potential space and accommodates greater volume than smaller joint recesses.
Because the local anesthetic component produces rapid symptom relief within minutes if correctly placed, the injection also serves a diagnostic function: a positive "anesthetic response" (marked reduction in pain with resisted abduction/impingement maneuvers immediately post-injection) supports the subacromial bursa/cuff as the true pain generator, as originally described in the impingement injection test (Neer test).
Real-time sonographic signs of correct placement include:
• Progressive anechoic-to-mixed-echogenicity distension of the bursal potential space as fluid is injected • A clean separation/lifting of the deltoid muscle away from the underlying supraspinatus tendon surface • Low injection resistance • Absence of intratendinous spread (which would appear as fluid tracking within the fibrillar tendon architecture rather than along its surface)
Studies directly comparing image-guided versus blind (palpation-based) subacromial injection consistently find substantially higher accuracy with ultrasound guidance — Eustace et al. (1997) found only ~70% accuracy for blind injection via arthrography confirmation, compared with 95–100% for ultrasound-guided injection in subsequent series (Naredo et al. 2004; Chen et al.).
Ultrasound guidance meaningfully improves the accuracy of subacromial injection compared with landmark-based technique, and accurate placement correlates with better short-term clinical outcomes. As with knee injections, benefit tends to be time-limited, and dosing frequency should be limited given theoretical and demonstrated effects on tendon integrity.
A landmark study by Naredo et al. (Ann Rheum Dis 2004) randomized patients with subacromial impingement to ultrasound-guided versus blind subacromial injection and found significantly better clinical outcomes (pain and function, Constant score) in the ultrasound-guided group at 6 weeks. Subsequent systematic reviews (e.g., Aly et al. 2015) confirm image guidance improves injection accuracy substantially (pooled accuracy ~90–99% vs. 70–83% for blind injection) and is associated with modestly superior short-term outcomes, though the difference in longer-term (>3 month) outcomes is less consistent.
The Cochrane review on subacromial corticosteroid injection for rotator cuff disease (Coombes et al.) found subacromial injections produce modest short-term benefit over placebo, with effects typically diminishing by 8–12 weeks, similar in magnitude to structured physiotherapy alone over the longer term — supporting a role as an adjunct to, rather than a replacement for, rehabilitation.
Ultrasound guidance for subacromial injection should be viewed primarily as an accuracy and diagnostic-clarification tool: it ensures the corticosteroid actually reaches the bursa (rather than deltoid muscle or subcutaneous tissue), which is essential both for therapeutic effect and for correctly interpreting the anesthetic response as a diagnostic test.
• Post-injection flare: 2–5%, self-limited within 24–48 hours • Facial flushing: up to 10%, more common with triamcinolone • Skin atrophy/hypopigmentation: more common with subcutaneous leakage or repeated superficial injection than true intrabursal placement • Tendon weakening/rupture: theoretical and case-report-level concern with repeated corticosteroid exposure directly into or near tendon tissue; minimized by confirming purely bursal (not intratendinous) placement under direct visualization • Transient hyperglycemia in diabetics: 24–48 hours • Infection: rare (<1 in 10,000) with sterile technique
As with knee injections, patients are counseled to limit injections to no more than 3 per year at the same site and to report any increasing pain, redness, warmth, or fever promptly.