Systematic review and quality analysis of emerging diagnostic measures for calcium pyrophosphate crystal deposition disease

Abstract

Objectives: Calcium pyrophosphate crystal deposition disease (CPPD) is common, yet prevalence and overall clinical impact remain unclear. Sensitivity and specificity of CPPD reference standards (conventional crystal analysis (CCA) and radiography (CR)) were meta-analysed by EULAR (published 2011). Since then, new diagnostic modalities are emerging. Hence, we updated 2009-2016 literature findings by systematic review and evidence grading, and assessed unmet needs.

Methods: We performed systematic search of full papers (PubMed, Scopus/EMBASE, Cochrane 2009-2016 databases). Search terms included CPPD, chondrocalcinosis, pseudogout, ultrasound, MRI, dual energy CT (DECT). Paper selection, data abstraction, EULAR evidence level, and Quality Assessment of Diagnostic Accuracy Studies (QUADAS)-2 bias and applicability grading were performed independently by 3 authors.

Results: We included 26 of 111 eligible papers, which showed emergence in CPPD diagnosis of ultrasound (U/S), and to lesser degree, DECT and Raman spectroscopy. U/S detected CPPD crystals in peripheral joints with sensitivity >80%, superior to CR. However, most study designs, though analytical, yielded low EULAR evidence level. DECT was marginally explored for CPPD, compared with 35 published DECT studies in gout. QUADAS-2 grading indicated strong applicability of U/S, DECT and Raman spectroscopy, but high study bias risk (in ∼30% of papers) due to non-controlled designs, and non-randomised subject selection.

Conclusions: Though CCA and CR remain reference standards for CPPD diagnosis, U/S, DECT and Raman spectroscopy are emerging U/S sensitivity appears to be superior to CR. We identified major unmet needs, including for randomised, blinded, controlled studies of CPPD diagnostic performance and rigorous analyses of 4 T MRI and other emerging modalities.

Keywords: Chondrocalcinosis; Qualitative research; Quality Indicators; Synovial fluid; Ultrasonography.

Figure 1

Systematic review study flow. A total of 111 articles were reviewed. Among articles identified, 65 were excluded after review of the title and/or abstract, 19 were excluded after review of manuscript, and 1 duplicate was excluded. A total of 26 articles were included in the analysis.

Figure 2

Graphical display for QUADAS-2 results. We assessed methodological quality by the QUADAS-2 tool, to evaluate the risk of bias and applicability of primary diagnostic accuracy studies, as described in the text. As the data show in the figure, the overall quality of the studies was satisfactory, with most QUADAS-2 grading in the low-risk range. The QUADAS-2 grading strongly supported applicability of U/S, DECT, and Raman spectroscopy, but there was moderate study bias risk overall; specifically ∼30% studies lacked proper knowledge of the results of the reference standard; and ∼25% studies without proper patient selection due to non-controlled study designs, and non-randomized subject selection. DECT, dual-energy CT; QUADAS, Quality Assessment of Diagnostic Accuracy Studies; U/S, ultrasound.