Arthrocentesis is generally a simple, relatively non-invasive way to gather critical information in patients of mono or polyarticular arthropathy.^1-3 Sampling of synovial fluid is among the most useful tests available to the clinician evaluating the patient.² Synovial biopsy procedure is also easy and carried out at the same time.⁴ It helps to distinguish between various inflammatory, non-inflammatory, traumatic and crystal induced arthropathies.⁵ In this study we have tried to predict the aetiology of a joint pathology by synovial fluid analysis as compared to synovial biopsy in order to determine the usefulness of simple tests done on synovial fluid.

A prospective study of synovial fluid analysis and synovial biopsy were carried out in cases of joint diseases from July 2001 to June 2002 at Terna Medical College, Nerul, Navi Mumbai.
Arthrocentesis was performed using aseptic techniques after the patient was kept fasting for eight hours.^5-7 Synovial fluid was aspirated from knee joints in 15 cases and analysed after collecting into three tubes:

a. Five ml in plain bulb for biochemical studies.

b. 2.5 ml in anticoagulant bulb (EDTA) for TLC, DLC, wet preparation and mucin clot and string tests.

c. Five to ten ml in sterile tube for microbiological studies.

Examination of synovial fluid was done as early as possible and following tests were done on each sample.^1,6,8,9

i. Gross examination : Volume, colour and clarity.

ii. Chemical analysis : Glucose estimation done by GOD-POD method. Protein estimation done by Biuret method.

iii. Mucin clot and String tests for viscosity.

iv. Microscopic examination : Total leucocyte count done by the standard haematological counting chamber method using isotonic saline with methylene blue as the diluent. Differential

Mucin clot test (MCT)^2,7 : It reflects the depolarization of hyaluronic acid and demonstrated by the precipitation of hyaluronate by acetic acid into the mucin clot. One part of synovial fluid is added to four parts of 2% acetic acid. After stirring briskly the nature of clot produced is noted and interpreted as shown in Table 1.

String Test (ST) / Falling drop test^1,6,7,9 : Synovial fluid is allowed to drop from a Pasteur pipette while noting the length of the string formed. Normal length of string is 4 to 6 cm (average 5 cm). Abnormal string less than 3 cm indicating low viscosity.

Other test parameters used to differentiate between inflammatory and noninflammatory categories^2,5,6,8 are given in Table 2.

Septic arthritis was considered when TLC was greater than 50,000 cell/cumm with greater than 75% neutrophils.2,3,6,10

ICrystals were not detected in wet preparation in any case. Grams and Ziehl-Neelsen stains were negative for organisms. Cultures failed to show any growth.

Synovial biopsies revealed Non-inflammatory conditions - 3 (all Baker’s cyst) and Inflammatory Conditions - 27 where (Non specific inflammation - 20, Tuberculosis - 4 and Rheumatoid arthritis - 3).

Of the 15 cases of synovial fluid analyzed, 3 were noninflammatory and the remaining 12 were inflammatory. Prompt examination of synovial fluid is required to avoid problems of misdiagnosing borderline inflammatory conditions, missing calcium pyrophosphatedihydrate crystals that dissolve with time or over interpreting new artifactual findings.3,11 Good clarity with good mucin clot test signifies non-inflammatory conditions whereas turbid appearance with fair to poor Mucin clot test were indicative of inflammatory conditions. These results were similar to that of Broderick et al and other workers.^1,2,6,12

Total leucocyte count ranging between 100 and 2100 cells/cumm with polymorphonuclear 18 to 23% in non-inflammatory joint diseases were in line with the observations of other workers.^2,6,8,12 However, there was a frequent overlap of TLC with that of various inflammatory arthropathies.

Findings in rheumatoid arthritis were similar to those of other workers.^2,3,6,12 Except in one case where Neutrophils were 76% of a total count of 30,000 cells/cumm. High neutrophil count is known in acute stage of rheumatoid arthritis similar to septic arthritis.^13,14 However total leucocyte count greater than 50,000 cells/cumm is required for the diagnosis of septic arthritis

Counts in cases of tuberculosis were similar to Wallace et al.¹⁵ In general the appearance of joint fluid in tuberculous arthritis depends on the stage of the disease with the diagnosis dependent on demonstration of acid fast bacilli in smear of culture.¹⁵ Wallace et al found 20% of patients of tuberculous arthritis to be positive for acid fast bacilli and 80% on synovial fluid culture.¹⁵ Both were negative in our study but surmised by typical appearances obtained on biopsy.

High neutrophil counts on DLC are sometimes obtained which were not categorized as septic arthritis in our study since TLC was less than the required number of 50,000 cells/cumm and no organisms on Gram stain or culture could be demonstrated.

Moderate reduction or normal glucose levels were seen in cases of inflammatory joint diseases but are usually linked to the levels in the serum. Glucose levels in synovial fluid are usually less than that of the serum by about 10 mg/dl in normal and in noninflammatory conditions but may be 20-60 mg/dl in septic arthritis. However, frequent overlaps are seen limiting the usefulness of glucose levels of synovial fluid.^2,3,6,8

Protein is not an accurate marker to distinguish inflammatory from noninflammatory diseases.^2,3,6,8 These findings are summarized in Table 3. Protein levels of synovial fluid involve multiple determinants like synovial microvessels, increased permeability and production and consumption of protein by the synovium. With other factors like local blood supply, lymphatic drainage and serum concentration.¹⁶

Rheumatoid arthritis may be RA positive but more characteristically shows typical histological appearances in the synovium including villous hypertrophy, proliferation of superficial synovial cells, infiltration with chronic inflammatory cells like lymphocytes and plasma cells, formation of lymphoid follicles, deposition of fibrin and foci of necrosis^4,5,13,14 (Fig. 1). Rheumatoid nodules may be present (Fig. 2). They were present in 7.6% cases of seropositive cases studied by Fassbender and 13% in other series.⁵

In the inflammatory group, the majority were chronic non-specific inflammation and showed mild to moderate villous hypertrophy, proliferated synovial cells and infiltration of chronic inflammatory cells. Accurate diagnosis of a particular disease in this broad group is difficult in routinely strained sections.

A better understanding of the pathogenesis may be achieved by immunohistochemical methods.⁵

In the non-inflammatory group, Baker’s cyst showed a typical histopathologic appearance and all were from the popliteal area.⁴

Synovial fluid analysis helps to identify inflammatory conditions by virtue of the following-

1. Total leucocyte count greater than 2000 cells/cumm
2. Polymorphonuclears greater than 25%
3. Mucin clot and string test showing fair to poor viscosity.

However, due to frequent overlap of synovial fluid findings, the above mentioned parameters are of limited usefulness to distinguish among the various aetiologies of the inflammatory group which still need to be biopsied to arrive at a specific diagnosis.