The Enlarged Joint: How to Use Joint Taps as a Simple Diagnostic Tool in Solving Joint Problems
World Small Animal Veterinary Association Congress Proceedings, 2018
College of Veterinary Medicine, Murdoch University, Murdoch, WA, Australia

Learning Objectives

At the end of this session, you will be able to:

  • Recognise non-inflammatory joint fluid from its gross characteristics
  • Use simple cytological interpretation of synovial fluid to differentiate inflammatory and non-inflammatory fluid

The Role of Joint Fluid Analysis in Diagnosis of Joint Diseases

Synovial fluid is a dialysate of plasma that contains glycoproteins from the synovial membrane type B cells. Polysulfated glycosaminoglycans (PGAGs) contribute to the viscosity of the synovial fluid. Disease processes of the synovial membrane or articular cartilage can alter the characteristics of synovial fluid.

Osteoarthritis (OA) is by far the most common type of joint disease in dogs. Inflammatory joint disease, infection and immune-mediated diseases are an infrequent but significant cause of joint pain and lameness.

Synovial fluid analysis can be used to differentiate between:

  • Inflammatory and non-inflammatory arthropathies
  • Infectious and immune-mediated inflammatory arthropathies
  • Traumatic arthropathies
  • Acute and chronic inflammation

Since the treatment of OA is very different from that used for inflammatory joint disease, an accurate diagnosis is important. The single most useful test in distinguishing between OA and inflammatory joint disease is synovial fluid analysis.

Arthrocentesis for synovial fluid analysis is a relatively simple procedure to perform, though it remains a surprisingly under-utilised tool in veterinary medicine.

What is Needed for Arthrocentesis?

  • Needles, 20–22 gauge. Length depends on the distance of the joint cavity from the skin surface and the size of the animal. For the majority of joints in most small to large dogs, 40 mm/1.5” length is ideal; 25 mm is sufficient for the carpus and hock joints, while for the shoulder and hip joint in giant breeds >40 mm/1.5” may be necessary.
  • Syringes, 2–5 ml
  • Microscope slides
  • EDTA tubes
  • (Heparin tubes +/-, only necessary if a mucin clot test is to be performed)
  • Sterile gloves
  • Bottle of blood culture medium if synovial fluid is to be cultured; this requires ideally 5 ml of synovial fluid. This is incubated for 24 hours before plating to blood agar plates. Culture of a small amount of fluid is unreliable.

Collection Technique

General anesthesia or sedation is recommended. The sites for arthrocentesis should be clipped and surgically prepared. Draping is not necessary; however, the use of sterile surgical gloves is recommended.

Successful arthrocentesis is easier if the operator has a good “mind’s eye” picture of the local joint anatomy when making the needle puncture.

The needle puncture should be made with the needle attached to the syringe. Apply gentle negative pressure to the syringe. After collection, release the negative pressure, disconnect the syringe from the needle to prevent sucking blood into the syringe when you withdraw the needle through the joint capsule, and then prepare the fluid as described below.

If blood appears in the needle hub, disconnect the syringe before it contaminates the fluid in the syringe if possible, as contamination with peripheral blood will alter the synovial fluid analysis. If this is not possible and blood contamination has occurred, make a note of this so that it may be taken into account when interpreting the results.

Iatrogenic contamination with blood may be differentiated from hemorrhagic synovial fluid. Hemorrhagic fluid is usually uniformly discolored, whereas iatrogenic contamination can be seen as discrete blood swirling in the fluid.

Shoulder Joint

A lateral or a cranial approach is possible. The lateral approach is made by inserting the needle just distal to the acromion process of the scapula and immediately caudal to the greater tubercle.

The cranial approach is made by inserting the needle medial to the greater tubercle of the humerus and directing it caudally, ventral to the supraglenoid tubercle of the scapula.


Many options exist for arthrocentesis of the elbow joint. The easiest approach is the caudolateral approach. The elbow is flexed to a normal standing angle.

The needle is inserted just medial to the lateral epicondylar ridge and directed cranially, medially and slightly distally into the olecranon fossa (toward the tip of the anconeal process). This is the largest joint space in the elbow.


This is probably the easiest joint to sample. There are three joint levels in the carpus: the antebrachiocarpal (ABC) joint, the middle carpal joint and the carpometacarpal joint. The ABC joint is the most proximal of the carpal joints located between the radius and the radial and ulna carpal bones. The middle carpal joint lies between the radial and ulna carpal bones and the numbered carpal bones. The carpometacarpal joint is the most distal of the three joint levels.

The middle carpal joint and the carpometacarpal joint communicate with each other while the ABC joint is separate. Despite this, sampling is routinely made from the ABC joint, as it is the largest of the carpal joint spaces and it is rare that joint disease should be localised specifically to the distal two joint levels.

The carpus is flexed and the needle is inserted into the craniomedial part of the ABC joint. A depression can be visualised and palpated at the joint space. The accessory cephalic vein is also easily visualised in this location and avoided.

Coxofemoral Joint

The coxofemoral joint is more difficult to sample and is not routinely sampled unless specific indications exist. A lateral or a ventral approach is possible.

For the lateral approach, the femur is abducted and slightly externally rotated (supinated). The needle is inserted craniodorsal to the greater trochanter and directed medially and slightly ventrally to penetrate the joint space.

For the ventral approach, the animal should be positioned in dorsal recumbency with the stifle joint fully abducted and the femur perpendicular to the long axis of the spine. The insertion of the pectineus muscle on the iliopectineal eminence of the pubis is palpated. The needle is inserted caudolateral to this eminence and directed cranially and dorsally into the joint.

Stifle Joint

When a palpable joint effusion exists, the easiest collection site in the stifle is the femoropatellar joint pouch, which is the largest of the joint pouches in the stifle joint. It should be noted that this is a potential space, so it is only suitable for collection when an effusion is present.

The needle is inserted immediately lateral to the patella and trochlea ridge at the level of the distal pole of the patella elevated about 30° from the skin surface. The needle is directed dorsally toward the non-articular part of the lateral femoral condyle.

In chronic disease with thickened joint capsule, you will feel a decrease in resistance when the needle tip passes through the capsule into the joint. If the condyle is contacted by the needle (this area is not covered in articular cartilage, so no damage is done), the needle is withdrawn slightly and the collection made.

The femorotibial joint space is the other option for collection from the stifle joint. It is a permanent space and so is easy to penetrate even when no effusion is present; however, collection of fluid can be complicated by interference by the fat pad.

The needle is inserted immediately lateral to the straight patellar ligament midway between the patella and the tibial tuberosity. The needle is directed medially and caudally through the fat pad into the intercondylar space.

Hock Joint

The hock joint can be penetrated medially or laterally and either cranial or caudal to the malleoli.

What to Do with the Synovial Fluid on Collection?

What you do with the synovial fluid will depend on the volume collected.

If small volumes (<0.2 ml) are obtained, these should be assessed in the syringe for color, clarity and viscosity prior to preparation of a direct smear. In these cases, there is not enough fluid to submit a sample in EDTA for total white blood cell count, so the sample is assessed visually for color and clarity, a smear is made, and the viscosity is subjectively assessed.

Smears should be made as soon as possible after collection to reduce the artefactual vacuolation and nuclear degeneration of large mononuclear cells. Smears are simply made by placing a drop on a slide and performing a “squash” preparation where a spreader slide is laid flat on the sample slide and the two slides are drawn slowly apart. Thin smears are easier to interpret and can be made by slowly pulling the two slides apart.

If larger volumes are collected, the fluid should be put into anticoagulant tubes. Unless a mucin clot test for viscosity is to be performed, place a drop on a slide as described above for a smear and put 1 ml into an EDTA tube.

Normal synovial fluid will not clot, as it does not contain fibrinogen. However, if there has been blood contamination during collection or intra-articular haemorrhage or protein exudation has occurred, clotting may occur.

EDTA is the preferred anticoagulant for cytological examination, while heparin is preferred for the mucin clot test and viscosity measurement. EDTA will cause degradation of hyaluronic acid and thereby affect viscosity. Both EDTA and heparin are suitable for assessment of color and clarity and measurement of protein.

If a septic arthropathy is suspected, synovial fluid should be submitted for aerobic culture in a blood culture bottle, or if anaerobic culture is required then anaerobic transport media is needed.


Synovial Fluid Analysis

Synovial fluid is assessed grossly for volume, color, clarity and viscosity, and a total and differential white cell count and protein content performed. Mucin clot test is usually not indicated.

Normal values are listed below:



The volume of synovial fluid collected from a normal joint varies and ranges from 0.1 to 0.5 ml. Joint effusion is usually apparent clinically +/- radiographically depending on the joint.


Normal synovial fluid should be colorless. A change in color suggests hemorrhage or inflammation. A yellow tinge indicates that hemorrhage into the synovial membrane has occurred and hemoglobin breakdown products are being released into the fluid. Iatrogenic contamination with blood is apparent from incomplete mixing of blood as evidenced in a clear synovial fluid sample that becomes streaked and blood-tinged during collection. Should iatrogenic contamination occur, this should be taken into consideration when assessing total and differential white cell counts and protein levels, as blood will falsely elevate all of these.


The clarity of synovial fluid is affected by the degree of cellularity. Normal synovial fluid is usually completely transparent. Synovial fluid may be graded as transparent, translucent or opaque. Transparent synovial fluid allows the print on the syringe to be read through it. The print appears as an area of darkness if the fluid is translucent and is indiscernible if the fluid is opaque. Inflammatory fluid has a high white cell count, and the fluid is usually opaque.


Viscosity may be assessed both subjectively and objectively. Viscosity is a function of the concentration and quality of hyaluronic acid and becomes poorer as the degree of joint inflammation increases due to depolymerisation of the hyaluronic acid by bacterial or inflammatory proteases.

Subjective assessment of viscosity is performed at the time of fluid collection by observing the length of the synovial fluid strand created by dropping some fluid from the collecting syringe to a slide or alternatively by placing a drop of fluid between your thumb and forefinger. The viscosity is normal if a strand of >2.5 cm is achieved before breaking (although much longer strands are typical). Fluid of normal viscosity will not run off a microscope slide when held vertical.

Normal synovial fluid has viscosity like oil. A “thin” fluid of poor viscosity is common in inflamed joints; however, it may also be seen in non-inflammatory arthropathies.

Objective assessment of viscosity by the mucin clot test may be performed on synovial fluid collected into either plain or heparinised tubes, although it is not routinely performed. EDTA is unsuitable for the mucin clot test, as it tends to degrade hyaluronic acid.

Total Cell Count

Total and differential cell counts of synovial fluid are important parameters of arthropathies. Reported total cell counts from normal canine joints can vary but are generally <3.0x109/L. Being a dialysate of plasma, there are only small numbers of nucleated cells and an absence of red blood cells.

If there is only sufficient fluid to prepare a smear, then a total cell count estimate should be done. The body of a smear should contain one to three nucleated cells per 400x magnification field. Cell count estimates are graded as normal or slightly, moderately or grossly increased.

Total white cell counts in non-inflammatory arthropathies are typically normal to slightly elevated (usually <5.0x109/L).

Total white cell counts in inflammatory arthropathies are typically moderately to markedly increased (>10.0x109/L).

Differential Cell Count

Normal synovial fluid contains >90% large mononuclear cells and lymphocytes (small mononuclear cells).

Lymphocyte numbers reported in normal synovial fluid range from 11% to 44% with the large mononuclear cells representing the majority of normal cells.

Normal joint fluid contains very low numbers of neutrophils. Typically neutrophils are <10% of the total white cell count.

Elevation of the relative proportion or absolute number of neutrophils in synovial fluid indicates either inflammation of the synovial membrane or contamination with peripheral blood. Elevation of the neutrophil percentage >10% regardless of the total white cell count is significant.

It is also important to assess the morphology of the neutrophils. It is widely reported that neutrophils from a septic inflammatory arthropathy will usually show evidence of toxic change or degeneration, whereas those from an immune-mediated arthropathy will appear more normal. This, however, has been shown to be unreliable. Joint infection is commonly seen without degenerate neutrophils being present. The presence of intracellular bacteria is indicative of infection.

A case series of dogs with septic arthritis (Marchevsky and Read 1998) would suggest that neutrophils in septic arthropathies are more typically non-degenerate. In a series of 19 dogs with septic arthritis, only one dog (5%) had degenerate neutrophils present in a synovial fluid smear. Only 7 of 13 dogs had bacteria visible on cytology.


As synovial fluid is a dialysate of plasma, the protein level in normal synovial fluid is usually low. Normal synovial protein levels in the dog are 20 to 25 g/L. Protein levels can be measured by either a refractometer or biochemical assay. Synovial fluid protein levels are a function of the local vascular permeability, the molecular size of the protein and its plasma concentration.

Synovial fluid protein levels will increase proportional to the degree of inflammation and may approach serum protein levels. Synovial fluid samples with very high protein levels may clot. (This can also happen in the presence of peripheral blood contamination.)



Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

College of Veterinary Medicine
Murdoch University
Murdoch, WA, Australia