Aspiration of the glenohumeral joint can be accomplished via either an anterior or a posterior approach. The posterior approach allows the patient to be blinded from the procedure, and it mimics the approach used in arthroscopy of the joint.
Shoulder arthrocentesis can be performed either diagnostically (to identify the etiology of acute arthritis) or therapeutically (to drain an effusion). With the same technique, the joint can be injected therapeutically with corticosteroids, anesthetics, or both.
Diagnostically, shoulder arthrocentesis is indicated in patients with shoulder pain for injection of anesthetic (with or without corticosteroid) with the aim of determining whether the glenohumeral joint is the source of the patient’s pain. By permitting joint aspiration, arthrocentesis aids in the diagnosis of the underlying pathologic process through synovial fluid analysis.[1, 2]
Therapeutically, shoulder arthrocentesis is indicated for providing pain relief and functional improvement in patients who have glenohumeral osteoarthritis, rheumatoid arthritis, or adhesive capsulitis (this may be performed through drainage of an effusion, septic joint, or hemarthrosis or through instillation of medication).
Bacteremia, cellulitis of overlying skin, and adjacent osteomyelitis are often considered absolute contraindications for shoulder arthrocentesis because of the potential risk of seeding the joint with bacteria. In these situations, the procedure should be performed only if the clinician strongly suspects septic arthritis as the cause of overlying inflammatory changes, and then only after consultation with an orthopedist.
Relative contraindications include glenohumeral joint infection, lack of effusion seen on ultrasonography (US), chronic infection distant to the injection site, prosthetic joints, allergy to the injectate, diabetes mellitus, and uncontrolled coagulopathy. Emergency arthrocentesis of a sterile joint may be required if a sterile environment is not available.
“Allergies” to povidone-iodine, lidocaine, or latex are not absolute contraindications. The injection site can be cleaned with chlorhexidine or pHisoHex followed by alcohol pads. Latex allergies are typically local reactions. In severe latex allergies, the rubber stopper on the lidocaine must be removed.
The practitioner performing the procedure should be familiar with the anatomy of the glenohumeral joint and the surrounding structures so as to minimize the risk of complications.
The glenohumeral joint of the shoulder is formed by the humeral head and the glenoid fossa of the scapula. It is bounded by the acromion. The subdeltoid bursa lies under the deltoid muscle and covers the lateral and superior aspect of the proximal humerus. The neurovascular bundle lies medially in the axilla.
Three glenohumeral ligaments exist, as follows:
Superior glenohumeral ligament (SGHL)
- Middle glenohumeral ligament (MGHL)
- Inferior glenohumeral ligament (IGHL)
The SGHL has a variable origin and inserts on the humerus near the lesser tubercle; it resists inferior translation of the humeral head in the adducted shoulder. The MGHL originates from the labrum and inserts on the humerus medial to the lesser tubercle; it resists inferior translation in the adducted and externally rotated shoulder. The IGHL originates from the labrum and the adjacent glenoid neck and inserts on the anatomic neck of the humerus; it resists humeral head anterior and posterior translation. (See Shoulder Joint Anatomy.)
The shoulder accounts for only 5-12% of all cases of septic arthritis. The overall incidence of septic arthritis of the shoulder is approximately 3200 cases per year in the United States. In older patients with substantial comorbidities, shoulder arthrocentesis may be considered when a septic joint is suspected. In younger patients, surgical drainage is optimal; septic arthritis of the shoulder has been associated with poor prognosis and recurrent effusion, drainage, subluxation, dislocation, osteomyelitis, septicemia, and septic shock.
Dry taps with obvious effusions may be caused by the following:
Rice bodies (infarcted pieces of synovium) in the needle lumen
- Poor technique
- High-viscosity synovial fluid
Equipment for shoulder arthrocentesis consists of the following:
- Skin-cleansing agent (chlorhexidine or povidone-iodine)
- Gloves (clean nonsterile gloves are sufficient if no contact is made with skin or needle)
- Local anesthetic
- Specimen tubes
- Needles – 1.5-in., 25-gauge for anesthetic injection; 18- to 22-gauge for aspiration
- Syringes – 3-5 mL for anesthetic injection; 5-20 mL for aspiration
A spinal needle may be needed for obese or muscular patients. If increased viscosity is suspected, a larger needle may be required. Larger needles make it easier to distinguish changes in resistance from tendons.
Adequate vacuum probably will not be generated from a syringe smaller than 5 mL. In some cases, a syringe larger than 50 mL may be required, but the practitioner should be cautious of excess suction. Excess suction may retrieve material other than fluid and may prevent aspiration.
It is recommended that the practitioner use one needle and one syringe size, so as to become accustomed to a standard degree of resistance.
The injection site should be inspected, palpated, and sterilized. If the site is incorrect, sterilization should be repeated at the new injection site. The site should not be touched after sterilization. Because the site has been sterilized, sterile gloves are not required. Although ethyl chloride spray is labeled as nonsterile, it has been found to maintain the sterility of injection sites.
Local anesthesia is recommended. (See Local Anesthetic Agents, Infiltrative Administration.) After skin preparation and identification of the needle insertion site, a 25-gauge needle is used to make a small skin wheal with local anesthetic into the subcutaneous tissue and then along the anticipated needle pathway.
The patient should be seated in a comfortable position. For the anterior approach, the patient’s hand should be rested on his or her lap so that the shoulder is internally rotated. For the posterior approach, the patient’s hand should be placed on the contralateral shoulder.
Aspiration of Synovial Fluid From Shoulder
Steps in the performance of shoulder arthrocentesis are as follows:
Obtain informed consent
- Palpate and mark the needle insertion site according to the approach chosen (see below)
- Sterilize the skin surrounding the insertion site, and allow it to dry
- Administer local anesthetic with a 25-gauge needle and a 3- or 5-mL syringe
- Insert an 18-gauge needle, and aspirate on needle advancement into the joint space until synovial fluid is obtained
- If injection is warranted, exchange the syringe for a corticosteroid-filled syringe and inject
- Remove the needle, and apply pressure and a bandage
- Divide the fluid among the specimen tubes
Needle placement accuracy rates appear to be significantly higher for the posterior approach to the glenohumeral joint than for the anterior approach.
Prevention of unnecessary arthrocentesis is possible through confirmation of the effusion with ultrasonography (US). Procedural success is increased, and potential damage to important structures (vessels, tendons, ligaments, and cartilage) is minimized with visualization of the needle tip. An ultrasound probe sheath, chlorhexidine for site preparation, sterile gloves, sterile gel, and sterile drape are required.
With US guidance, the posterior approach accessing the posterior glenohumeral recess is preferred. After the transducer is aligned in the long axis with the infraspinatus, the needle is advanced from lateral to medial (or vice versa) until its tip is positioned near the surface of the humeral head hyaline cartilage.
There are advantages to both in-plane and out-of-plane approaches. An in-plane approach does not allow neighboring structures to be identified, but it does provide the advantage of visualizing the entire length of the needle. An out-of-plane approach makes visualizing the needle tip difficult, but it allows improved visualization of the surrounding structures. The choice of approach is determined by the preferences of the provider and the depth of the joint. The curvilinear probe is ideal for the shoulder.
Accuracy rates are also higher when imaging is used in conjunction with injection and aspiration. In a randomized trial comparing US-guided with blind steroid injection in patients with adhesive capsulitis of the shoulder, Raeissadat et al found that US-guided injections yielded improved accuracy, pain relief, range of motion, and function; however, these improvements were not statistically significant. In addition, the use of US guidance was associated with higher cost and was more time-consuming.
Palpate the coracoid process and the humeral head. As the arm is internally rotated, the joint space can be felt as a groove lateral to the coracoid process (see the image below).
Circle represents coracoid process.
Insert the needle medial to the head of the humerus and just below the tip of the coracoid process (see the image below).
Shoulder arthrocentesis. Insert needle medial to head of humerus and just below tip of coracoid process.
Direct the needle slightly laterally and superiorly into the scapulohumeral joint space (see the image below).
Shoulder arthrocentesis. Direct needle slightly laterally and superiorly.
Insert the needle 1-2 cm inferior and medial to the posterior tip of the acromion. Direct the needle anteriorly and medially toward the coracoid.
Switching of syringe
If the practitioner wants to change the syringe, then the needle hub can be clasped with a hemostat. The hub should be clasped distally because the proximal portion can distort the opening. This technique is useful if corticosteroid injection follows aspiration or if an additional syringe is needed for complete aspiration.
Complications are uncommon and often insignificant but include the following:
Hematoma and intravascular injection are possible because of the close proximity of the axillary vessels
- Iatrogenic infection is possible; 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
- Vasovagal syncope
Laboratory Medicine Summary
Synovial Fluid Analysis
The following may be included in analysis:
Gram stain and culture should be performed on all samples suspicious for infection.
- Total Leukocyte count and differential support differentiation between noninflammatory and inflammatory joint conditions.
- If macrophage predominance (>80%): consider “Milwaukee shoulder”
- Polarized microscopy evaluates for the presence of pathological crystals.
- If inflammation or infection is not present, normal synovial fluid glucose is within 20 mg% of serum value.
- Inflammation increases normal synovial fluid protein, and the average is approximately 2 mg% (33% of the serum total protein).
- Ordering the following tests is not recommended due to low clinical significance: lactate dehydrogenase, uric acid, pH, electrolytes, and immunological studies.
Synovial fluid should be analyzed immediately after arthrocentesis. Delayed analysis may cause:
Appearance of artifactual crystals
- Decrease in leukocyte count (from cell disruption)
- Decrease in number of crystals (especially calcium pyrophosphate dihydrate) 
Please refer to Joint Fluid Interpretation / Overview/Joint Fluid Interpretation for additional information.
- Lin HM, Learch TJ, White EA, Gottsegen CJ. Emergency joint aspiration: a guide for radiologists on call. Radiographics. 2009 Jul-Aug. 29 (4):1139-58. [Medline].
- Fields TR, Berman JR. Arthrocentesis, intraarticular injection and synovial fluid analysis. Paget SA, Gibofsky A, Beary JF III, Sculco TP, eds. Hospital for Special Surgery Manual of Rheumatology and Outpatient Orthopedic Disorders. 5th ed. Philadelphia: Lippincott Williams & Wilkins; 2005. Chap 8.
- Koh KH. Corticosteroid injection for adhesive capsulitis in primary care: a systematic review of randomised clinical trials. Singapore Med J. 2016 Dec. 57 (12):646-657. [Medline]. [Full Text].
- Baruffi S, Bhimji SS. Arthrocentesis, Shoulder. Treasure Island, FL: StatPearls; 2018. [Full Text].
- Hobbs K. Arthrocentesis and synovial fluid analysis. West SG, ed. Rheumatology Secrets. 3rd ed. Philadelphia: Elsevier Mosby; 2015. 58-62.
- Jiang JJ, Piponov HI, Mass DP, Angeles JG, Shi LL. Septic Arthritis of the Shoulder: A Comparison of Treatment Methods. J Am Acad Orthop Surg. 2017 Aug. 25 (8):e175-e184. [Medline]. [Full Text].
- Polishchuk D, Gehrmann R, Tan V. Skin sterility after application of ethyl chloride spray. J Bone Joint Surg Am. 2012 Jan 18. 94 (2):118-20. [Medline]. [Full Text].
- Daley EL, Bajaj S, Bisson LJ, Cole BJ. Improving injection accuracy of the elbow, knee, and shoulder: does injection site and imaging make a difference? A systematic review. Am J Sports Med. 2011 Mar. 39 (3):656-62. [Medline].
- Jacobson JA. Interventional techniques. Fundamentals of Musculoskeletal Ultrasound. 3rd ed. Philadelphia: Elsevier; 2018. 407-43.
- Taylor T et al. Rheumatology. Sherman SC, ed. Simon’s Emergency Orthopedics. 8th ed. Chicago: McGraw-Hill; 2019. 45-74. [Full Text].
- Raeissadat SA, Rayegani SM, Langroudi TF, Khoiniha M. Comparing the accuracy and efficacy of ultrasound-guided versus blind injections of steroid in the glenohumeral joint in patients with shoulder adhesive capsulitis. Clin Rheumatol. 2017 Apr. 36 (4):933-940. [Medline].
- Walsh NE, Rogers JN. Injection procedures. DeLisa JA, Gans BM, Walsh NE, eds. Physical Medicine & Rehabilitation: Principles and Practice. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2005. Chap 14.
- Partin WR. Emergency procedures. Stone CK, Humphries RL, eds. Current Diagnosis & Treatment: Emergency Medicine. 6th ed. New York: McGraw-Hill; 2008. Chap 6.
- Braddom R. Peripheral joint, soft tissue and spinal injection techniques. Braddom RL, Buschbacher RM, Chan L, et al, eds. Physical Medicine & Rehabilitation. 3rd ed. Philadelphia: Elsevier; 2007. 541-62.