I. Problem/Challenge.

Arthrocentesis is the procedure of aspirating synovial fluid out of a joint. Arthrocentesis is similar in technique to joint injection – these two procedures can be performed during the same encounter in appropriate cases.

II. Identify the Goal Behavior

Accurate differential diagnosis of joint effusions followed by proficient preparation, technical performance, sample analysis and post-procedural care for joint aspiration and injection.

III. Describe a Step-by-Step approach/method to this problem.

Effusions generally indicate an intra-articular process. Patients with joint effusions may describe pain, pressure, sensations of joint instability or restricted range of motion. Joint effusions are apparent depending on the size of the effusion and depth of the joint under the skin (e.g., a large knee effusion will likely be apparent on visual inspection, whereas a small glenohumeral effusion may be difficult to diagnose even with palpation). An understanding of the underlying anatomy is critical to joint examination and to successful arthrocentesis or joint injection because effusions may disrupt the normal anatomic landmarks (e.g., loss of tibial condyles and knee joint line to visual inspection in a large knee effusion). Effusions are fluctuant and located around the joint space. You may be able to massage the patient’s joint fluid into a smaller or more apparent space (“milking the effusion” or “ballotment” tests).

Continue Reading

It is important to differentiate between effusions (intra-articular fluid) and enlarged bursa (extra-articular fluid) or other swollen or inflamed structures (e.g., fat pads, tendonitis), because arthrocentesis or corticosteroid injections into a joint are unlikely to improve extra-articular pathology and can cause complications (e.g., fat pad atrophy due to corticosteroid effect, tendon rupture from injection into a tendon). In cases where the physical exam is indeterminate, ultrasound can be helpful in diagnosing effusions and locating anatomic landmarks, however, ultrasound guidance has not been shown to consistently increase clinical efficacy of intraarticular injection.

When your patient has a new joint effusion consider the following: Is there a risk of infection that may have seeded the joint? Does the patient have a pre-existing condition that predisposes them to effusions, e.g., crystalline disease (gout, pseudogout/CPPD), rheumatologic disease (RA, systemic lupus erythematosus), or osteoarthritis? Is there a history of new trauma or coagulopathy as risk factors for hemarthrosis?

  • When the cause of an effusion is unclear, arthrocentesis allows synovial fluid analysis which can assist in confirming or ruling out a pathologic diagnosis (e.g., septic joint versus crystalline disease).

  • When an effusion is causing pressure and/or pain, arthrocentesis may be indicated as therapy with or without a concomitant corticosteroid/local anesthetic injection.

  • Absolute:

    Periarticular or overlying skin and soft tissue infection in the area of the joint or a systemic infection due to risk of seeding joint with infection.

    Corticosteroid or local anesthetic injection are contraindicated in the presence of a documented allergy to the medications planned for injection (N.B. amide anesthetics display cross-reactivity).

  • Relative:

    Disrupted or denuded skin, especially if chronic (e.g., in dermatoses such as psoriasis, venous stasis ulcers) due to risk of chronic bacterial colonization.

    Anatomic disruption (dislocation, fracture, tendon tear or rupture, prosthetic joint).

  • Considerations:

    If the patient is on full strength anticoagulation, consideration of risks:benefits should be given. There is evidence that joint aspiration/injection is safe in patients on warfarin with INR up to 4.5.

    If the patient has severe immunosuppression, consideration of risks:benefits should be given relative to the indication (e.g., strong indication for arthrocentesis in concern for septic joint versus low urgency for arthrocentesis and corticosteroid injection for chronic DJD with a small effusion).

Except for septic joint, symptomatic effusions and their underlying disease processes may be treated with injection of corticosteroids (as an anti-inflammatory agent), local anesthetics (for analgesia) or a combination of the two in the joint. The evidence for benefit of corticosteroid injections is mixed depending on indication, site, technique and type of medication used. To date the highest quality evidence for corticosteroid injection is for degenerative joint disease, which shows moderate benefit for short term pain relief (6-12 weeks) with minimal additional adverse effect versus placebo.

Septic arthritis or systemic infections are contraindications to corticosteroid injection. If septic joint is at all possible, you should perform arthrocentesis first and wait for results of a STAT cell count and gram stain prior to administering intraarticular medications. If septic joint is ruled out based on the synovial analysis, intra-articular medications may be considered. Acute inflammatory arthritis (crystalline and rheumatologic) and degenerative joint disease from osteoarthritis may be treated with corticosteroid injection with or without arthrocentesis of the effusion, especially in patients in whom non-steroidal anti-inflammatory drugs (NSAIDs) are relatively contraindicated.

While high quality evidence on long term safety is limited, most review articles recommend performing corticosteroid injections no more often than 4 times yearly. Hyaluronic acid (synthetic synovial fluid) injection is also used for degenerative joint disease – it is exponentially more expensive than corticosteroid preparations, and there is minimal evidence for increased benefit over corticosteroids or non-invasive therapy. Other treatments for arthralgia depending on the patient may include “RICE” (rest, ice, compression and elevation), physical therapy/graded exercise to strengthen surrounding muscle groups, oral analgesics (e.g., NSAIDs, APAP), topical analgesics (e.g., menthol cream), and systemic corticosteroids in certain cases of disabling acute inflammatory arthritis.

  • Gloves (clean, non-sterile gloves may be used. The injection site should not be touched after cleaning/sterilizing. This is called “no touch” technique and is supported by multiple published studies as a best practice).

  • Iodine swabs x 3 or chlorhexidine swab

  • Gauze pads

  • Needles

    18 gauge for larger joint aspiration, 23 gauge for smaller joint aspiration

    25 gauge for injection only including pre-arthrocentesis local anesthetic injection

    Length: 2” needle for medium and large joints, 1-1.5” for small joints

  • Syringes: 5 mL+ for aspiration, for injection use size appropriate for volume

  • Adhesive bandage/dressing

  • Ethyl chloride spray or topical anesthetic ointment for skin anesthesia (if needed)

  • Corticosteroid (if needed)

  • 1% or 2% lidocaine without epinephrine or other local anesthetic (if needed)

  • Ultrasound (if needed)

See Table I for a brief guide to injectable medications.

Table I.
Joint size (examples) Triamcinolone acetonide dose* (vol of 40 mg/1mL) Hydrocortisone equivalent (if converting to other steroid)** 1% Lidocaine volume in mL (dose in mg)***


Shoulder – glenohumeral or subacromial space

Hip – greater trochanteric bursa

40 mg (1mL) 200 mg 4-10 mL (50-100 mg)


Wrist – carpal

Ankle – tibiotalar

10-20 mg (0.25-0.5 mL) 50-100 mg 3-5 mL (30-50 mg)

Hand – CMC, MCP

Foot – TMT, MTP

Digits – IPJ

1-10 mg (0.05-0.25 mL) 5-50 mg 1-3 mL (10-30 mg)

A rule of thumb we use in our practice – starting at large proximal joint, decrease dose of steroid by half with each joint distally (i.e. Triamcinolone use in leg: knee – 40 mg, ankle – 20 mg, MTP’s 5-10 mg, IPJ’s 1-5 mg)

Methyl prednisolone (Depo-Medrol) and triamcinolone acetonide (Kenalog) are reported to cause less post-injection flare; triamcinolone acetonide and triamcinolone hexacetonide (Aristospan) are reported to have longer clinical activity.

Lidocaine toxicity (for preparations without epinephrine) occurs around total dose 4.5 mg/kg, roughly 300 mg (30 mL) in a 70 kg person.

For most effusions without clear diagnosis, send the following:

  • Cell count and differential

  • Protein and glucose

  • Bacterial culture and gram stain

  • Microscopy (e.g., polarized light microscopy in suspected gout or pseudogout)

  • Consider testing for acid-fast bacillus (AFB) and fungus if clinically indicated

Step 1: Preparation

Perform an accurate physical examination, focusing on palpation of the joint landmarks. Mark the area for needle entry into the skin with a surgical (permanent) marking pen.

Sterilize an area 5 cm in diameter around the puncture site using 3 swabs of iodine with concentric outward circles or with a single chlorhexidine swab x 60 seconds. Allow both substances to dry – this is important for their bactericidal/static activity. (There is no need to wipe the cleansing agent off prior to puncture.) Once you have sterilized the area, only the needle touches the marked site (“no touch technique”). If you touch the area with your hand or a non-sterile object, re-sterilize as above prior to proceeding.

Prepare the syringes and needles based on the planned procedure:

  • For arthrocentesis only, use an 18 gauge (medium and large joints) or 22 gauge needle (small joints) connected to a 5-60 cc syringe depending on the size of the joint and volume of fluid needed. 5-10 cc is likely to be adequate for routine testing. For therapeutic arthrocentesis, a larger syringe may be indicated.

  • For arthrocentesis followed by corticosteroid injection, use needle and syringe above for the arthrocentesis and prepare a separate 5-10 cc syringe for medications (e.g., for the knee: a 5 cc syringe containing 1 cc of 40 mg/mL triamcinolone acetonide mixed with 4 cc of plain lidocaine).

  • For corticosteroid injection use 25 gauge needle with steroid mixture in a syringe prepared as above.

Step 2: Anesthetize the area (optional depending on preference and procedure)

Skin anesthesia/analgesia can be accomplished in multiple methods. For arthrocentesis with a large needle, skin anesthesia may be preferable, whereas it may be unnecessary if performing a joint injection with a 25 gauge needle. Discuss options with patient in advance:

  • Ice

  • Stretching skin (counter-traction which limits discomfort from skin distention)

  • Ethyl chloride spray

  • Topical anesthetic (e.g., EMLA) administration

  • Cutaneous anesthesia with 1-3 mL lidocaine and a 25G needle

If using ice, ethyl chloride or topical anesthetics, apply these prior to skin prep. If using lidocaine injection for local anesthesia (e.g., prior to large needle arthrocentesis), inject ~1-3ml lidocaine with a 25 gauge needle subcutaneously at the marked site. Wait a few minutes following the injection of lidocaine for anesthetic effect prior to arthrocentesis.

Step 3: Aspirate / Inject

For arthrocentesis, apply counter-traction to the skin outside the sterilized area and enter the skin with needle at the marked site, advancing toward the joint space while pulling back syringe plunger with gentle consistent force. You will create a small vacuum. When you enter the joint, the syringe will begin to fill with synovial fluid.

Once fluid enters the syringe, stop advancing the needle and try to maintain position for the rest of the aspiration. Once in the joint space, aspirate synovial fluid to the volume needed, based on intended studies or need for therapeutic drainage.

If the needle moves and/or flow stops prior to the desired volume, try to advance and retract it slightly, without radically repositioning it.

If no aspirate is found, pull the needle back until you are in the soft tissue, but not out of the skin. Quickly reassess what direction is optimal to reach the joint space. Redirect the needle right or left, superiorly or inferiorly, depending on your goal location. Advance again, until the joint fluid is found.

If you are performing arthrocentesis followed by joint injection (sometimes called “two syringe, one needle technique”), hold the hub firmly between the thumb and index finger, with the little finger/ulnar aspect of the hand resting against the patient’s skin for stability. Unscrew the syringe and exchange syringes for the syringe containing medications for injection. Once the new syringe is in place, inject the medication. If resistance is felt, you may need to retract or change position slightly if the needle bore is obstructed.

If performing corticosteroid and/or local anesthetic intra-articular injection alone, apply counter-traction to the skin outside the sterilized area and enter the skin with needle at the marked site, advance until you are in the joint space (depth depending on the size and depth of the joint and size of patient.) Inject all of the medication at the desired depth. Avoid injecting into tendons (this is usually noted as resistance to injection, retract until medication flows freely in this case) and avoid injection into the dermis (injection of steroids can cause dermal fat atrophy.)

When aspiration and/or injection are completed, remove the needle, apply pressure with sterile gauze, wipe away iodine with alcohol and apply an adhesive bandage.

See Figure 1, Figure 2, Figure 3, Figure 4, Figure 5, Figure 6, Figure 7 and Figure 8 for a series of photos and brief descriptions of approaches to arthrocentesis and joint injection for common injections: shoulder glenohumeral joint and subacromial (extra-articular) space, elbow, knee, ankle tibiotalar joint, and 1st MTP joint. The approaches here are adapted from the excellent series of articles by Tallia and Cardone, and Zuber in AFP 2002-2003.

Figure 1

Anterior bent-knee approach to knee tibiofemoral joint. Enter in the joint line, direct needle at 45 degrees to the skin toward midline of patella, needle parallel to floor. (Black oval = patella; red arcs = joint line). NB. While successful for joint injection, this technique has been reported to be less successful for arthrocentesis vs suprapatellar approach.

Figure 2

Lateral suprapatellar approach to knee patellofemoral joint. Enter at supralaterlal or supramedial aspect of patella. Direct needle at a 45 degree angle to the skin pointing underneath and toward the midpoint of the patella. (Black oval = patella; red arcs = joint line)

Figure 3

Anterior approach to tibiotalar (ankle). Direct needle posterolaterally, needle parallel to the plantar surface or slightly toward plantar surface. joint (Red line = medial malleolus, black line = tibialis anterior tendon)

Figure 4

Dorsal approach to 1st MTP joint. Direct needle inferiorly between metatarsal head and proximal aspect of proximal phalanx. (Red arc = joint line.)

Figure 5

Lateral approach to elbow joint. Direct needle in the middle of three bony landmarks (RH = radial head, LE = lateral epicondyle, OP = olecranon process.)

Figure 6

Anterior approach to glenohumeral joint. Enter skin 1cm lateral to the coracoid process, direct needle posterior and slightly lateral. (Blue circle = humeral head, red oval = coracoid process)

Figure 7

Posterior approach to approach to glenohumeral joint. Enter skin 3 cm inferior to posteriolateral aspect of acromion. Direct needle anteriorly and slightly medial. (Blue triangle = scapula)

Figure 8

Posterior approach to subacromial space. Enter skin inferior to posterolateral aspect of acromion. Direct needle toward ipsilateral coracoid process and angle slightly superior. (Blue triangle = scapula, red arrow = path of needle)

  • Decrease to normal level of activity (no strenuous activity) for a few days following corticosteroid injection, following by a gradual return to usual activity.

  • Discuss signs and symptoms of infection and precautions – return to primary care, urgent care or hospital emergency department immediately.

  • Discuss signs of post-injection flare in patients receiving corticosteroid injection and plan of care – ice, NSAID’s/APAP, relative rest.

  • In case of joint injection with corticosteroid and local anesthetic, patients may experience immediate relief from anesthetic for hours. This effect will wane, usually followed by slower onset of pain relief in 48 hours as the steroid begins to take effect.

  • Stretching/strengthening/resting as indicated based on underlying diagnosis.

See Table II for Interpretation of synovial fluid analysis.

Table II.
Normal Non-inflammatory (e.g., DJD/OA) Inflammatory (e.g., autoimmune, crystal disease) Septic
Total WBC/mm3 <100 100-2,000 2,000-50,000 >25,000-50,000
PMN (%) <25 <25 <50 >50-90
Culture Negative Negative Negative Often positive

IV. Common Pitfalls.

What are the possible complications of arthrocentesis and/or joint injection?

Arthrocentesis and joint injection:

  • Iatrogenic infection (1/10000 to 1/100,000) – this generally takes around 12-48 hrs to develop) – requires urgent referral to orthopedics for joint lavage

Corticosteroid/local anesthetic injection only:

  • Post-injection flare in 1-6%, thought due to inflammatory response to corticosteroid crystals (generally begins 2 hours-2 days after injection with corticosteroids) – self-limited over 3-5 days and responds to NSAID’s)

  • Impaired glucose control (for approximately 1-2 weeks in patients with pre-existing diabetes or impaired glucose tolerance. Hyperglycemia occurs in a dose-responsive manner to potency and amount of corticosteroid used)

  • Tendon rupture

  • Fat necrosis/calcification

V. National Standards, Core Indicators and Quality Measures.

No national standards/benchmarks are established yet.

VI. What's the evidence?

Barr, L. “Basic Ultrasound-guided Procedures”. Crit Care Clin. vol. 30. 2014. pp. 275-304.

Bettencourt, RB, Linder, MM.. “Arthrocentesis and Therapeutic Joint Injection: An Overview for the Primary Care Physician”. Prim Care Clin Office Pract. vol. 37. 2010. pp. 691-702.

Bloom, JE. “Image-guided versus blind glucocorticoid injection for shoulder pain”. Cochrane Database of Systematic Reviews. 2012.

Buchbinder, S, Green, S, Youd, JM.. “Corticosteroid injections for shoulder pain”. Cochrane Databate of Systematic Reviews. 2003.

Cardone, DA, Tallia, AF.. “Diagnostic and Therapeutic Injection of the Elbow Region”. American Family Physician. vol. 66. 2002. pp. 2097-2101.

Foster, ZJ. “Corticosteroid Injections for Common Musculoskeletal Conditions”. American Family Physician. vol. 92. 2015. pp. 694-699.

Jüni, P. “Intra-articular corticosteroid for knee osteoarthritis”. Cochrane Database of Systematic Reviews. 2015.

Monseau, AJ, Nizran, PS.. “Common Injections in Musculoskeletal Medicine”. Prim Care Clin Office Pract.. vol. 40. 2013. pp. 987-1000.

Tallia, AF, Cardone, DA.. “Diagnostic and Therapeutic Injection of the Shoulder Region”. American Family Physician. vol. 67. 2003. pp. 1271-1278.

Tallia, AF, Cardone, DA.. “Diagnostic and Therapeutic Injection of the Ankle and Foot”. American Family Physician. vol. 68. 2003. pp. 1356-1362.

Wittich, CM. “Concise Review for Clinicians: Musculoskeletal Injection”. Mayo Clin Proc.. vol. 84. 2009. pp. 831-837.

Zuber, TJ.. “Knee Joint Aspiration and Injection”. American Family Physician. vol. 66. 2002. pp. 1497-1500.

Pemberton, R.. Arthritis and rheumatoid conditions. Their nature and treatment. 1935.

Hollander, JL.. “Hydrocortisone and cortisone injected into arthritic joints. Comparative effects of and use of hydrocortisone as a local antiarthritic agent”. Journal of the American Medical Association. vol. 147. 1951. pp. 1629

Roberts, WN, Babcock, EA, Breitbach, SA, Owen, DS, Irby, WR.. “Corticosteroid injection in rheumatoid arthritis does not increase rate of total joint arthroplasty”. J Rheumatol.. vol. 23. 1996. pp. 1001-4.

Raynauld, JP, Buckland-Wright, C, Ward, R, Choquette, D, Haraoui, B, Martel-Pelletier, J. “Safety and efficacy of long-term intra-articular steroid injections in osteoarthritis of the knee: a randomized, double-blind, placebo-controlled trial”. Arthritis Rheum.. vol. 48. 2003. pp. 370-7.

McCarty, DJ.. “Treatment of rheumatoid joint inflammation with triamcinolone hexacetonide”. Arthritis Rheum.. vol. 15. pp. 157

Gaffney, K, Ledingham, J, Perry, JD.. “Intra-articular triamcinolone hexacetonide in knee osteoarthritis: factors influencing the clinical response”. Ann Rheum Dis. vol. 54. May 1995. pp. 379-81.