Skip navigation

Uric acid crystals in synovial fluid for diagnosing gout


Clinical bottom line

The evidence for using uric acid crystals for diagnosing gout is not extensive. It may be a useful test, but good quality evidence is lacking.


JB Segal & D Albert. Diagnosis of crystal-induced arthritis by synovial fluid examination for crystals: lessons from an imperfect test. Arthritic Care and Research 1999 12: 376-380.

G Kerolus et al. Is it mandatory to examine synovial fluids promptly after arthrocentesis? Arthritis and Rheumatism 1989 32: 271-278.

Systematic review

Segal & Albert conducted a MEDLINE search up to November 1996 for articles about the accuracy of crystal identification by microscopy.

There were four articles. Two of these studies four samples only, one 11 and the other 96 samples.

The largest study (96 samples) used synthetic crystals suspended in synovial fluid at various concentrations, and presented them to technicians and physicians at various stages of training. The average sensitivity for urate crystals was 69% and average specificity was 95%. This gives a positive likelihood ratio of 14 and a negative likelihood ratio of 0.3.

The next largest study (11 samples) sent samples of uric acid crystals to a reference and two test laboratories. Levels of agreement were low in one laboratory compared to the reference laboratory, and moderate in the other.

The two other studies sent samples to a number of laboratories for analysis. Most correctly identified uric acid crystals.

Identification of calcium phosphate crystals was uniformly poor in any study that examined it.

Stability study

One study (Kerolus et al) has looked at stability of urate crystals in synovial fluid. samples stored at room temperature had reduced numbers of urate crystals within 2-3 days, with large reductions in some samples by one week. Samples stored at 4äC mostly maintained the number of urate crystals for several weeks.

Calcium phosphate crystals ere reduced in numbers rapidly (hours) at room temperature and when refrigerated.


The amount of evidence we have for the use of microscopic detection of uric acid crystals in synovial fluid is not great. It looks as if different laboratories or observers mostly agree, and that it is helpful in diagnosing gout. If clinical suspicion before the test was, say, 40%, then with a likelihood ratio of 14, the post-test probability would be over 90%. There should be hesitation, though, as the likelihood ratio is based on artificial samples, and not real samples on real patients against a gold standard diagnosis of gout. This is a real gap in our knowledge that needs to be filled.

Clearly care on storage is an issue if samples are not examined immediately, and especially so for calcium phosphate crystals.