Background The diagnosis of calcium pyrophosphate crystal (CPP) deposition disease (CPPD) is mainly based on the synovial fluid analysis and Xrays. US has demonstrated high sensitivity and specificity values for diagnosing CPPD compared to synovial fluid analysis as the gold standard (1), but less is known about sensitivity and specificity of synovial fluid analysis itself.

Objectives to compare ultrasonography and synovial fluid analysis performances in the diagnosis of CPPD using a real gold standard.

Methods We enrolled in our study all patients waiting to undergo knee replacement surgery due to severe osteoarthritis. Each patient underwent US examination of the knee, focusing on the menisci and the hyaline cartilage, the day prior to surgery, scoring each site according to the presence/absence of CPP as defined previously (1). The day of the surgery, synovial fluid of the knee (if present) was aspirated by the surgeon. After surgery, the menisci, condyles and the synovial fluid were retrieved and examined microscopically. Synovial fluid analysis was performed on wet preparations. For the meniscus and cartilage microscopic analysis, six samples were collected, either from the surface and from the internal of the structure trying to cover a large part of it. All slides were observed under transmitted light microscopy and by compensated polarised microscopy. A dichotomous score was given for the presence/absence of CPP. US and microscopic analysis were performed by different operators, blind to each other's findings. Sensitivity and specificity of US and synovial fluid were calculated using microscopic findings of the menisci and cartilage as the gold standard.

Results we enrolled in the study 32 patients (9 males), mean age of 74 years old (±7). Synovial fluid has been collected from 24 patients. If we consider all the structures examined with US (both menisci and cartilage of both condyles), were positive for CPP 22 patients while synovial fluid analysis was positive for 11 patients. At microscopic examination of the speciments, 21 patients were positive for CPP in at least one of the structures examined. US demonstrated a sensitivity of 95% (CI95: ±0.01) with PPV =0.91 and specificity of 81% (CI95: ±0.23) and a NPV =0.90 while respective values for synovial fluid microscopic analysis were 73% (CI95: ±0.22) with PPV of 1 and 100% (CI95: ±0) with NPV of 0.69.

Conclusions US demonstrated higher sensitivity values for identifying CPP deposits in the knee joint than synovial fluid analysis. Specificity values on the other had were higher for the microscopic analysis as expected. Globally we believe that for its intrinsic characteristics, the non invasive nature, for the high values of both specificity and specificity, and last but not least, for the capability to address differential diagnosis US should be the first exam ti be performed when CPPD disease is suspected. As this study demonstrates, the presence of CPP crystals in the synovial fluid, definitely confirms the diagnosis but a negative microscopic exam does not exclude it.

References

1. Frediani B, Filippou G, Falsetti P, Lorenzini S, Baldi F, Acciai C, et al. Diagnosis of calcium pyrophosphate dihydrate crystal deposition disease: ultrasonographic criteria proposed. Ann Rheum Dis. 2005 Apr;64(4):638-40

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.1621