Objectives: The aim of this review is to assess the evidence for the accuracy and safety of tests to diagnose gout in patients with no prior diagnosis of gout. The review also assesses factors that affect accuracy of diagnostic tests. Tests include algorithms that combine clinical signs and symptoms, dual-energy computed tomography (DECT), ultrasound, and plain x ray, with particular emphasis on tests that can be conducted in primary and acute (urgent and emergent) care settings.
Data sources:
We searched Medline® (from 1946), Embase® (from 1972), the Cochrane Library (from 1945), and the Web of Science™ (from 1980) to November 7, 2014, for published studies. We also searched
Review methods: We reviewed published and unpublished prospective cohort, cross-sectional, and case-control studies, as well as prior systematic reviews on the accuracy (sensitivity and specificity) of diagnostic tests for gout compared with a validated reference standard in patients without a prior gout diagnosis. We also reviewed studies and prior reviews of factors affecting the accuracy of monosodium urate crystal assessment in synovial fluid. We reviewed prospective cohort, cross-sectional, and case-control studies; case reports of any size; and systematic reviews that reported adverse events associated with diagnostic tests for gout and outcomes of gout misdiagnosis. A standardized protocol with predefined criteria was used to extract details on study design, interventions, outcomes, and study quality, and to assess the strength of evidence for each conclusion.
Results: Six clinical algorithms comprising clinical signs and symptoms have been tested for diagnostic accuracy against the presence of monosodium urate crystals in synovial fluid aspirated from affected joints. Most studies were conducted with small groups of patients in academic rheumatology departments. Two recently developed clinical algorithms, the Diagnostic Rule, which is the only one developed and validated with primary care physicians and patients, and the Clinical Gout Diagnosis (CGD), demonstrated sensitivities of 88 percent and 97 percent, respectively, and specificities of 75 percent and 96 percent, respectively, in patients with shorter (2 years or less) and longer durations of symptoms, and they are simple to administer. However, the strength of evidence supporting their use is low, as validation of these tools remains limited. Three studies of DECT that enrolled patients without a previous gout diagnosis revealed sensitivities ranging from 85 percent to 100 percent and specificities ranging from 83 percent to 92 percent in diagnosing gout; the strength of evidence regarding the use of DECT for gout diagnosis is low. Four studies of ultrasound that enrolled patients without a previous diagnosis showed sensitivities ranging from 37 percent to 100 percent and specificities ranging from 68 percent to 97 percent, depending on the ultrasound signs assessed; the strength of evidence is low for the utility of ultrasound in diagnosing gout. A small number of studies examined factors that affected the accuracy of tests for the diagnosis of gout. The accuracy of monosodium urate analysis in synovial fluid varies widely among practitioners, but evidence on the effects of skill and experience is insufficient. No studies examined differences among practitioners in the rate of successful joint aspiration. No studies reported adverse events directly associated with techniques used to diagnose gout. However in one small study, missed gout diagnosis resulted in unnecessary surgery, longer hospital stays, and delay in appropriate treatment.
Conclusions: Promising diagnostic clinical algorithms such as the Diagnostic Rule and CGD need to be validated more broadly in primary and urgent care settings. A clinical algorithm with high diagnostic accuracy ideally can form part of a diagnostic decision tree, with referral of more clinically challenging cases to rheumatologists for more invasive tests or imaging. Research is needed to assess the incremental value of synovial fluid monosodium urate crystal analysis and imaging over that of a diagnostic clinical algorithm.