> Skip repeated content

HSS Manual Ch. 8 – Arthrocentesis, Intra-articular Injection and Synovial Fluid Analysis

From the HSS Manual of Rheumatology and Outpatient Orthopedic Disorders

Key Points

  • Aspiration of a joint should always be performed on an initial evaluation or whenever a diagnosis is in doubt or in the case of monoarticular arthritis when infection must be excluded.
  • Fluid aspirated should always be analyzed for cell count, gram stain, culture and the presence of crystals at minimum.
  • A known diagnosis of rheumatoid arthritis, gout or pseudogout does not exclude infection especially when one joint is disproportionately affected.
  1. Arthrocentesis, or aspiration of fluid from a joint, is a safe and relatively easy procedure that plays both diagnostic and therapeutic roles in the management of arthritis. It should be included in the initial evaluation of every patient with a joint effusion, especially those with monarthritis.
  2. Synovial fluid analysis following aspiration may help in differentiating a primary inflammatory process such as rheumatoid arthritis (RA) from a non-inflammatory process such as osteoarthritis. It can provide a specific diagnosis in crystalline arthritis.
    1. Diagnostic indications:
      1. As part of an initial evaluation.
      2. To rule out infection or hemorrhage.
      3. To rule out crystalline disease.
    2. Therapeutic indications:
      1. Drainage of an effusion to relieve pain and restore range of motion.
      2. Instillation of medication (e.g. steroids, viscosupplementation).
      3. Drainage of a septic joint.
      4. Drainage of a hemarthrosis.
    3. Absolute Contraindications: Infection in overlying skin or surrounding soft tissue.
    4. Relative Contraindications: Coagulation disorder, especially if severe.
    5. Intra-articular injection is primarily used to deliver intra-articular corticosteroids to treat inflamed joints (a similar technique is employed for injection of inflamed soft-tissues such as bursae, or tendons). Contraindications are the same as for arthrocentesis. Corticosteroids should not be injected into a joint until infection has been excluded. In general, joints should not be injected more than three to four times per year. Injection of corticosteroids directly into a tendon or tendon insertion can sometimes result in tendon rupture.

Preparation for Aspiration/Injection

  1. Materials for aseptic skin preparation:
    1. Sterile gloves.
    2. Iodine solution.
    3. Alcohol solution.
    4. Sterile gauze pads.
  2. Materials for local anesthesia:
    1. One percent lidocaine for skin, subcutaneous tissues, and joint structures.
    2. Ethyl chloride spray for skin.
  3. Needles. Sterile 18- to 25-gauge needles, depending on the size of the joint. Inflamed joint fluids may be thick and require a large-bore needle for removal. Sterile 30-gauge needles may be used for local anesthetic instillation, and for injection into small joints such as the proximal interphalangeal joints and when no fluid aspiration is anticipated.
  4. Syringes, 1cc to 50cc in size, depending on the joint and amount of effusion.
  5. Tubes for synovial fluid analysis:
    1. Hematology tube for cell count and differential.
    2. Sterile tubes for Gram stain, cultures and smears.
    3. Heparinized tube for crystal analysis. Powdered anticoagulant may interfere with crystal identification.
    4. Cytology bottle (if neoplasm is suspected).
  6. Intra-articular medications. At Hospital for Special Surgery, methylprednisolone acetate (Depomedrol), a long-acting, insoluble corticosteroid preparation is used. The dose varies with the size of the joint. Betamethasone (Celestone) may be used when the goal is avoidance of a joint flare reaction and a shorter 2-4 week duration of action is acceptable (e.g. with attacks of crystal disease) Doses and appropriate needle sizes are summarized in Table 8-1.

Table 8-1 Intra-articular Therapy Regimens
Table 8-1 Intra-articular Therapy Regimens

Joint-Specific Techniques

The most important maneuver before aspirating a joint is to locate the appropriate landmark. This can be best done by making a skin impression with the round end of a non-opened pen point or pen mark. Local anesthesia of the overlying skin and subcutaneous tissues is recommended. In general, if it is important to obtain fluid for diagnostic purposes, a larger needle should be used (so as to avoid the need for re-aspiration if the fluid is too thick to be aspirated by a small needle).

  1. Shoulder. The shoulder can be entered either anteriorly or posteriorly.
    1. Anterior approach (Fig. 8-1). With the patient’s hand in the lap and the shoulder muscles relaxed, the glenohumeral joint can be palpated by placing the fingers between the coracoid process and the humeral head. As the shoulder is internally rotated, the humeral head can be felt turning inward and the joint space can be felt as a groove just lateral to the coracoid process. When the skin over this area is anesthetized, a 20- or 22-gauge needle can be inserted lateral to the coracoid. (the thoraco-acromial artery lies medial to the coracoid). The needle is directed dorsally and medially into the joint space. The needle should be directed slightly superiorly to avoid the neurovascular bundle.

      Illustration of arthrocentesis of the shoulder, anterior approach
      Figure 8-1. Arthrocentesis of the shoulder - anterior approach.
    2. Posterior approach (Fig. 8-2). The posterior aspect of the shoulder joint is identified with the patient’s arm internally rotated maximally. This position is achieved by placing the patient’s ipsilateral hand on the opposite shoulder. The humeral head can then be palpated by placing a finger posteriorly along the acromion while the shoulder is rotated. A 20- or 22-gauge needle is inserted about 1 cm inferior to the posterior tip of the acromion and directed anteriorly and medially.

      Illustration of arthrocentesis of the shoulder, posterior approach
      Figure 8-2. Arthrocentesis of the shoulder - posterior approach.
  2. Elbow. The elbow joint (Fig. 8-3) can be identified by placing the patient’s relaxed arm in the lap. With the palm facing the patient, flex the elbow to a 45-degree angle. Place your finger on the lateral epicondyle and note the shallow depression distal to it, which represents the elbow joint. A 22-gauge needle is introduced perpendicular to the joint.

    Illustration of arthrocentesis of the elbow
    Figure 8-3. Arthrocentesis of the elbow.
  3. Wrist (Fig. 8-4) aspiration is performed on the dorsal aspect just distal to the radius or ulna as indicated by clinical examination.
    1. Radial entry. The hand and wrist are relaxed in a slightly flexed position. The joint space can be located by palpating the edge of the distal radius just medial to the thumb extensor tendon. A 22-gauge needle should be directed into the joint from the dorsal aspect.
    2. Ulnar entry. Keep the wrist in the same relaxed position. The joint space can be identified by palpating just distal to the distal ulna. The 22-gauge needle is directed in a volar and radial direction.

      Illustration of arthrocentesis of the wrist, medial and lateral approaches
      Figure 8-4. Arthrocentesis of the wrist - medial and lateral approaches.
  4. Ankle. For both approaches (Fig. 8-5), the foot is first placed at about a 45-degree angle of plantar flexion.
    1. Medial approach. A 22-gauge needle is placed about 1 in. proximal and lateral to the distal end of the medial malleolus. The flexor hallucis longus tendon is just lateral to this point. The needle is directed 45 degrees posteriorly, slightly upward, and laterally.
    2. Lateral approach. A 22-gauge needle is placed about 1/2 in. proximal and medial to the distal end of the lateral malleolus. The needle should be directed 45 degrees posteriorly, slightly upward, and medially.

      Illustration of arthrocentesis of the ankle, medial and lateral approaches
      Figure 8-5. Arthrocentesis of the ankle - medial and lateral approaches.
  5. Knee. The knee (Fig. 8-6) is the easiest joint to enter. It may be entered either medially or laterally. The patient should be supine with the knee comfortably extended. A 19-22-gauge needle is introduced in a direction parallel to the plane of the posterior surface of the patella in a medial position (in patients with a known coagulopathy, or taking anticoagulant drugs, a 25g needle can be used). With thick exudative effusions, a larger-bore needle may be required. Drainage of the knee suprapatellar bursa can be facilitated by compressing the suprapatellar pouch during aspiration. With large knee effusions, the distended suprapatellar pouch can be aspirated directly from either the medial or lateral aspect of the quadriceps muscle mass.

    Illustration of arthrocentesis of the knee, medial approach
    Figure 8-6. Arthrocentesis of the knee - medial approach
  6. Small joints of the hands and feet may be difficult to enter. Occasionally, the effusion bulges and facilitates aspiration. Often, a corticosteroid injection can be performed just adjacent to the joint rather than within it; this may result in an equivalent clinical response.
    1. The metacarpophalangeal (MCP) joint can be easily palpated on its dorsal, lateral aspect with the finger slightly flexed and relaxed. The joint is entered on the dorsal-lateral aspect with a 22 to 25-gauge needle. Because this is a ball (distal metacarpal) and cup (first phalanx) joint, the needle should not be directed at a 90-degree angle but rather distally at about a 60-degree angle.
    2. The proximal interphalangeal (PIP) joint margin is barely palpable but may be felt on its dorsal aspect just distal to the skin crease. The joint is entered from the dorsal aspect with a 25 to 30-gauge needle that is directed slightly distally.
    3. The distal interphalangeal (DIP) joint is extremely small and difficult to enter. The technique is the same as for aspirating the PIP joint.  If there is no suspicion of infection, a 30g needle may be the least uncomfortable for the patient and easiest for joint entry.
    4. The metatarsophalangeal (MTP) joint is aspirated in a fashion similar to that for the MCP joint.
  7. Other joints. There are external landmarks that can direct aspiration and injection of the hip joint, but success in this venture requires some experience. Because the hip joint lies deep within the pelvis, aspiration can be more readily performed under fluoroscopic or ultrasound guidance. Ultrasound in particular enables direct visualization, accurately defines the presence and size of an effusion and allows for successful and safe aspiration to be performed if need be. The injection of corticosteroids for treatment can also be done with greater precision in administration. The spinal and sacroiliac joints often demand fluoroscopic or CT guidance.


  1. “Post-injection flare”: Long-acting intra-articular steroid preparations like methylprednisolone acetate may induce a crystal synovitis 24 hours after the injection. Application of ice at the onset of a post injection flare may be helpful.
  2. Bruising: The needle itself may traumatize the joint, especially if the joint is small; for this reason, the patient should be warned of possible short-term aggravation of symptoms in the injected joint and receive appropriate instructions regarding analgesia.
  3. Skin atrophy can sometimes occur if corticosteroids are injected close to the undersurface of the skin.
  4. Tendon rupture: Injection of corticosteroids directly into a tendon or tendon insertion can cause rupture. If tendon injection is desired, ultrasound-guided injections directly into the paratenon (not the tendon proper) may be both effective and safe. Ultrasound-guided injections, e.g. into the retro-calcaneal bursa, can minimize the risk of injection into the Achilles tendon itself.

Synovial Fluid Analysis

Fluid analysis is an extremely useful diagnostic tool in the evaluation of rheumatic diseases. It should be included in the initial evaluation of most arthritic conditions that generate effusions. It can yield a specific diagnosis in infectious and crystal-induced arthritis and can be helpful in categorization, definition and the diagnosis of other arthritides.

  1. Synovial fluid studies (Table 8-2):
    1. Gross examination can be helpful in establishing the nature of a joint fluid. After air bubbles are allowed to clear, a heparinized specimen is examined for the following:
      1. Color. Normal synovial fluid is straw-colored. Inflammatory fluids range from yellow to greenish yellow. Bloody fluid to one degree or another occurs in patients with coagulation disorders, trauma, neoplasms, and tuberculous arthritis and in patients receiving anticoagulant therapy.
      2. Clarity. Normal synovial fluids are clear enough that print can be read through them. As inflammation increases from mild to marked, the fluid becomes first translucent and then opalescent.
      3. Viscosity. Synovial fluid viscosity is tested by allowing a drop of fluid to fall from the needle tip. Normal synovial fluids are quite viscous, and a “string” of fluid will form. Because viscosity is decreased in inflammatory synovial fluids, no string sign is seen.
    2. Cell Count. A cell count for both white blood cells (WBC), including a differential, and red blood cells (RBC) should be performed. Inflammatory fluids generally have WBC counts in the thousands, while non-inflammatory fluids have WBC counts under 1000. Remember the ratio of RBC/WBC is ~750/1. This can be important in hemorrhagic fluids that are inflammatory or infected. The WBC differential can be useful diagnostically, with greater than 50% polymorphonuclear (PMN) cells common in inflammatory effusions and greater than 90% PMN cells commonly seen with infection.
    3. Crystal examination can be performed using polarized microscopy of a specimen of heparinized fluid (see Chapters 43, 44). Urate crystals parallel to the polarizer axis appear yellow, and urate crystals perpendicular to the polarizer axis appear blue. The opposite is true for the calcium pyrophosphate crystals of pseudogout. Urate crystals are needle-shaped and calcium pyrophosphate crystals are rhomboid. Remember that the finding of crystals does not rule out the possibility of an infection.
    4. Microbiologic studies:
      1. Stains should include both Gram’s and acid-fast methods if infection is suspected or fluid appears turbid or cloudy.
      2. Cultures should include routine bacterial studies. Fungal and mycobacterial cultures are ordered if clinically indicated. Some fastidious or slower-growing organisms may need to be watched for growth for up to 4 weeks. Synovial fluids suspicious for gonococci or Borrelia burgdorferi may be specifically sent for PCR analysis if the diagnosis is in question.
    5. Biochemical studies
      1. Glucose. Determination of synovial fluid glucose, when interpreted with a simultaneous serum value, is helpful in diagnosing infectious arthritis. In bacterial infection or tuberculosis, the synovial fluid glucose will be less than half the serum value. Occasionally, low values may be seen in RA.
      2. Protein determination does not provide additional useful information and should not be routinely ordered.
      3. Complement may be decreased in RA, but the test is rarely helpful for diagnosis because synovial fluid complement is usually normal in early RA.
  2. Diagnosis by fluid group. Synovial fluid can be divided into three groups based on the degree of inflammation (Table 8-2):
    1. Group 1 fluids are clear and transparent and have few white cells on cell count. They include normal, osteoarthritic, and systemic lupus erythematosus (SLE) joint fluids.
    2. Group 2 fluids generally have a higher WBC count and are not as reactive as group 1 fluids; they appear translucent. This group includes fluids from most non-infectious, inflammatory arthritic conditions such as gout, pseudogout, psoriatic arthritis, Reactive arthritis and RA. Leukemia or lymphoma occasionally presents in this category, but the differential count reveals more than 90% mononuclear cells.
    3. Group 3 fluids are opalescent or purulent. Group 3 fluids include those from bacterial infections and tuberculosis (although joint fluid from gonococcal arthritis can be either group 2 or group 3). Group 3 fluids typically have 50,000 to 300,000 WBC per milliliter; these are mostly neutrophils. Occasionally, the synovial fluid from a patient with an inflammatory arthritic condition such as RA or gout may have as many as 50,000 to 75,000 white cells per milliliter and appears opalescent or even purulent. As Table 8-1 shows, there is considerable overlap between the various arthritic diseases; this table is meant to serve as a guideline rather than provide a rigid set of criteria.

Table 8-2. Synovial Fluid Analysis thumbnail
Click Thumbnail to Enlarge
Table 8-2. Synovial Fluid Analysis


Headshot of Theodore R. Fields, MD, FACP
Theodore R. Fields, MD, FACP
Attending Physician, Hospital for Special Surgery
Professor of Clinical Medicine, Weill Cornell Medical College
Headshot of Jessica R. Berman, MD
Jessica R. Berman, MD
Assistant Program Director, Rheumatology Fellowship, Hospital for Special Surgery
Associate Attending Physician, Hospital for Special Surgery
Richard Stern, MD
Attending Physician
Hospital for Special Surgery-New York Presbyterian Hospital
Clinical Associate Professor
Weill Medical College of Cornell University

Related Professional Content