Article Text

Download PDFPDF

A systematic review of ultrasonography in gout and asymptomatic hyperuricaemia
  1. Priya V Chowalloor1,2,
  2. Helen Isobel Keen1,2
  1. 1Department of Rheumatology, Royal Perth Hospital, Perth, Western Australia, Australia
  2. 2School of Medicine and Pharmacology, Royal Perth Hospital Unit, University of Western Australia, Perth, Western Australia
  1. Correspondence to Dr Helen Isobel Keen, School of Medicine and Pharmacology, Royal Perth Hospital, Level 3, MRF building, Rear 50 Murray Street, Perth, WA 6000, Australia; helen.keen{at}uwa.edu.au

Statistics from Altmetric.com

Introduction

Gout is the most common inflammatory joint disease in males.1 It is the result of a disorder in purine metabolism leading to hyperuricaemia, and deposition of monosodium urate (MSU) crystals in articular, periarticular and soft tissues. This induces acute episodes of inflammation and also chronic inflammation that is associated with progressive joint destruction. Additionally, uncontrolled gout and hyperuricaemia are associated with tophi formation, renal failure, cardiovascular disease2–6 and increased morbidity and mortality.5 ,7 Thus it is imperative that gout is diagnosed appropriately and the associated inflammation is well controlled. Several studies have shown that gout is suboptimally treated in the community and there is a paucity of validated markers of disease activity.8 ,9

Imaging modalities, such as ultrasound (US), MRI, and office-based MRI, may aid in the diagnosis, monitoring and management of gout. Advantages of US over other imaging modalities include the lack of ionising radiation, relatively low cost, patient acceptability and ease of access.10 Ultrasonography is well accepted to be sensitive to the presence of articular pathology in rheumatoid arthritis (RA), particularly synovial and cortical changes.11–14 The role of US in the diagnosis and management of gout and asymptomatic hyperuricaemia (AH) is not well established. This manuscript aims to review published literature with regards to the evidence of US as an assessment tool in gout and AH, with a focus on validity, responsiveness, reproducibility and feasibility.15

Methods

A systematic review was undertaken with the objective of identifying all studies published from 1950 till February 2012 using US in subjects with gout and AH. Medline and PubMed databases were searched using the search terms gout, asymptomatic hyperuricaemia, US, ultrasonography, Doppler, colour; ultrasonography interventional; ultrasonography, Doppler, pulsed and power Doppler and gout, asymptomatic hyperuricaemia and US, respectively. Studies were included if they used US to image subjects with gout or AH. Publications were excluded if they were: not published in English, review articles, case reports, letters to editor that are purely commentary or non-human studies. Search results were screened to avoid duplicates. Manuscripts were reviewed and data extracted and categorised into those involving the joint and subcutaneous tissue. Extracted data addressed aspects of study design, US methodology and reproducibility, comparator technique, pathology imaged and the study findings. The references of published review articles were screened for additional manuscripts that met our inclusion and exclusion criteria. As this manuscript focuses on the validity, responsiveness, reproducibility and feasibility of US in gout and AH, review articles were not included in the data extraction or discussed in the results, but were sourced, read and are referenced in the discussion.

Results

Sixty-seven studies were identified in Medline, and 169 articles in PubMed.

The reasons for excluding articles are presented in figure 1. There were a total of 18 articles included in this review. Fourteen studies were in gout, three studies were in AH and one study in gout and AH.

Figure 1

Search methods and reasons for exclusion of studies from the review. AH, asymptomatic hyperuricaemia.

The studies included are listed in table 1, along with a description of the study design, disease studied and comparator groups, and comparator used to assess US. Table 2 provides information on the region imaged, type of Doppler used and study design. The pathologies studied by US included tophi (table 3), articular cartilage (table 4), soft tissue abnormalities (table 5) and bony lesions (table 6). Only 10 (55%) studies reported the reproducibility of the imaging technique. Only one study of the 18 (5.5%) was powered.

Table 1

Summary of studies included in this review

Table 2

Summary of studies detailing imaged region, US techniques and reproducibility

Table 3

Summary of studies detailing US imaging of tophi in gout and AH

Table 4

Summary of studies detailing articular cartilage involvement in gout

Table 5

Summary of studies detailing soft tissue changes in gout

Table 6

Summary of studies detailing bony changes in gout

Tophi

Table 3 provides details of the information extracted from manuscripts focused on imaging tophi. In gout, most studies using US to examine tophi are descriptive (n=12),16–19 ,22 ,23 ,26 ,28–32 focusing on describing or categorising tophi. Tophus-like lesions are variably referred to as tophi, or by a description of their US appearance which is varied (see table 3). The majority of tophi are hyperechoic, with either a heterogeneous appearance, or heterogeneous appearance with calcification. They are often grouped, and have a poorly defined border and demonstrate postacoustic shadowing.16

With regards to the prevalence of tophi in gout, nearly half of the patients with gout in one prospective study had tophi in at least one of the examined joints,30 and they are more common in subjects with a higher uric acid29 or subjects not on urate lowering therapy (ULT).29 Two studies in AH have reported the presence of tophi.22 ,32 A significant number of people with AH are also reported to have tophi.32 These were present in tendons, synovium and soft tissues.22 ,32

Two studies focused on the construct validity of US to assess tophi using MRI as the standard.17 ,23 US compared well against MRI, detecting more tophi in one study (non-significant)23 and detecting 90% of the lesions reported to be tophi by MRI in another.17 Criterion validity is reported, with 83% of detected tophi sampled demonstrating MSU crystals on histology.17 One study compared the subjects with gout to healthy controls. There was US evidence of tophaceous material in all of the symptomatic joints examined in gout subjects, but not in any asymptomatic joints of healthy controls.18 Tophus was present equally in clinically affected and unaffected metatarsophalangeal (MTP) joints in gout.29

Responsiveness was addressed in only one study, demonstrating a reduction in size of tophi as measured by US in 20 of the 38, and resolution in 9 out of 38 tophi detectable by US in response to ULT. The majority of subjects who demonstrated a reduction in tophi had uric acid level <6 mg/ml.17

Reliability was addressed in only five studies, four of them reported interobserver to be very good to excellent17 ,23 ,29 ,30 and substantial in the fifth study.26 Only one of these studies17 examined interoccasion reliability, the rest reread stored images. Intraobserver reliability (interoccasion) was reported in only one study as very good.17

Feasibility was not reported in any of the studies.

Articular cartilage

Table 4 provides details of the studies focusing on articular cartilage in gout and AH.

Most of the studies focused on the presence of a double contour (DC); a term used to describe a hyperechoic irregular band over the articular cartilage, best seen on the dorsal side of the MTP joints.31 MSU crystals are demonstrated to create this DC sign on the cartilage surface. Hyperechoic spots within the cartilage are associated with calcium pyrophosphate deposition disease.27

The DC was demonstrated in up to 60% of joints imaged in gout, in majority of joints with intra-articular tophi and also in clinically unaffected joints.29 ,30 The DC is also reported in AH.32 ,33 The DC was found to be present in subjects with gout and AH statistically more often than in subjects without gout or AH.19 ,18 ,32 The sensitivity and specificity of the DC in diagnosing gout is estimated to be 43.7% and 99%, respectively.24

The responsiveness of US to cartilage changes in gout has been demonstrated by the disappearance of the DC sign in response to ULT.25

Inter-reader reliability assessing the DC sign was reported in three studies of gout24 ,2930 and one of AH,29 and was found to be excellent in all. One of these addressed interoccasion reliability.24

Feasibility was not reported in any of the studies.

Soft tissue changes

Table 5 addresses soft tissue changes in gout. These include joint effusion, synovial hypertrophy, intra-articular Doppler signal, intra-articular hyperechogenicity, tendon lesions and soft tissue oedema.

Joint effusion was seen in several inflammatory joint diseases, but rarely in healthy controls.19 US detected more joint effusions than clinical examination in gout.20 ,31 The validity and responsiveness of US in detecting effusions in gout or AH has not been assessed. Feasibility has also not been addressed.

Synovial hypertrophy is not specific and is seen in gout and other inflammatory joint diseases but not in healthy controls.19 ,21 The validity of US in detecting soft tissue changes has rarely been studied. Only one study compared the validity of US findings with MRI, with MRI detecting more synovial pannus in symptomatic joints than US.23 Additionally, US is able to detect synovial pathology and power Doppler signals, in the absence of clinical evidence of acute gout or swelling.19 ,20 Other synovial signs reported include intrasynovial hyperechogenicity, described variably as bright stippled foci, hyperechoic soft tissue and hyperechoic spots in gout, and hyperechoic cloudy areas in AH.19 ,21 ,33 These lesions are seen relatively commonly in gout and AH, but while Rettenbacher and de Miguel report that hyperechoic areas and brightly stippled foci are very specific,21 ,33 Wright found that hyperechoic spots <1 mm were seen no more commonly in gout than in controls.19 Responsiveness of US detected synovial pathology in gout has not been reported. Interobserver agreement is reported as fair for synovial pannus.23 Feasibility is not reported.

Five studies looked for the presence of Doppler signal (power Doppler n=4 and colour Doppler n=1) reflecting the vascularisation of synovial tissue. Power Doppler sonography detected significantly more inflamed joints than clinical examination20 and is seen in inflammatory joint disease but not in healthy joints.19 The validity of US-detected Doppler signal has not been compared against other imaging techniques or pathology in gout or AH. The presence of Doppler signal as a diagnostic test was addressed in one study, finding that Doppler signal was sensitive but not specific for the diagnosis of gout.21 Responsiveness, reliability and feasibility are not reported in any of the studies.

The validity of US in detecting soft tissue oedema was examined in one study.23 MRI did not detect significantly more soft tissue oedema than US in symptomatic and asymptomatic joints. Responsiveness of US to changes in soft tissue oedema has not been reported. Interobserver reliability was poor in the only study addressing it.23 Feasibility has not been reported.

Bony changes

Table 6 addresses bone changes in gout. The validity of US in detecting erosions, compared with MRI was examined in one study.23 MRI detected significantly more erosions than US, however US performed better than conventional radiography (CR), particularly when erosions were small.19–21 US detected erosions in gout are more common in subjects with longer disease duration, increased frequency of attacks and the presence of US detected tophi.19 Erosions are most commonly found in the first MTP joint (medial surface), and metacarpophalangeal joints.19 ,18 ,31 They are commonly found in MTP joints that have never been clinically affected by gout.19 When subjects with gout were compared with other populations, significantly more US detected erosions were seen in the MTP joints of patients with gout (67%) than in other inflammatory joint diseases (43%) or healthy controls (6%) (p<0.001).19 Using US detected erosions as a diagnostic test for gout has been demonstrated to be moderately specific (69%) but not sensitive (24%).21

Responsiveness of US to erosions has not been reported. Inter-reader reliability assessing erosion was reported in two of the studies and was found to be excellent.19 ,23 Feasibility is not reported in any of the studies.

Discussion

This manuscript is the first systematic review that systematically examines the literature on US-detected pathologies in gout.

US-detected pathology found in gout can largely be grouped into tophi, cartilage changes (namely the DC sign), soft tissue pathology and erosions. Soft tissue oedema has also been reported, but in only one study with poor reliability. Pictorial and review articles, not included in the results section, report pathological findings similar to those reported in the tables in the results section of this manuscript, and also report bursal involvement and a snowstorm appearance suggestive of MSU crystals in synovial fluid.34 ,35

US appears to be valid in detecting tophi, using MRI and histology as standards, is responsive to change and is reliable; there is a large amount of heterogeneity in the descriptions of what has been designated as tophi for the purpose of the results section of this manuscript (table 3). These descriptive terms often avoid the term ‘tophi’, making it uncertain whether all authors are referring to the same pathological lesion. Standardisation of the description of what is considered an US-detected tophi may better facilitate the development of a useful outcome tool, for diagnosis and monitoring of disease.

The DC sign is widely considered to be useful in diagnosing gout. In this manuscript, a review of the evidence suggests that the DC is relatively common in gout, in affected and asymptomatic joints, and also in AH. While it has been documented to be uncommon in controls, the validity of the US-detected lesion has not been compared with other imaging techniques, or against histology, and none of the studies addressed potential confounders such as the fluid-cartilage interface sign.

Reliability is reported as excellent, although reliability was usually assessed by rereading stored images. Additionally, the DC sign is reported to be specific, but not particularly sensitive; false negatives are common. This US-detected sign may prove to be a useful diagnosis and monitoring tool in gout, however further work on the validity and reliability, particularly with regards to intraoccasion reading, may be required.

Synovial effusion, hypertrophy and Doppler signal were commonly seen in gout, with only one study comparing US-detected synovial hypertrophy to MRI. The validity of US in detecting synovial pathology has been extensively studied in RA, and, again there is no reason to suspect that the validity will be substantially different between RA and gout. The soft tissue changes of most interest in the literature focus on findings that may indicate intra-articular MSU crystal deposition. While these are generally to be considered highly suggestive of gout (de Migeul finding the presence of the DC sign in combination with intrasynovial hyperechoic cloudy areas were strongly associated with the presence of uric acid crystals in synovial fluid), Wright found that very small hyperechoic spots were particularly not found more commonly in the first MTP joint of gout subjects than in diseased and healthy controls. Potential confounders are rarely discussed with regards to these US signs that are widely accepted to be suggestive of gout, and it has been suggested that the snowstorm appearance reported in pictorial reviews 35 could also be due to gas bubbles formed as a result of fluid agitation following joint movement.36 As these US lesions are of potential diagnostic utility, defining, standardising and validating US lesions thought to be specific in gout is likely to be of clinical benefit, in the way the Outcomes Measures in Rheumatology (OMERACT) US definitions have been beneficial in the study of RA.37 A combination of US pathologies may also be of diagnostic interest. In particular, the combination of DC sign and a cloudy synovial fluid was found to be specific, and another study demonstrated that reading control and gouty images side by side, with a focus on the presence of tophi, DC sign, effusions and erosions allowed correct identification of gouty joints in 97% cases.18

US detected erosions were valid compared with MRI, were detected more frequently than with CR,19–21 and had excellent reliability. Responsiveness of US, specifically to changes in erosions in gout, has not been reported, but there is no particular reason to think this would be different from the ability of US to detect erosions in other inflammatory joint diseases, such as RA.38 ,39 However, the diagnostic specificity of US is less than CR,21 perhaps due to the lower sensitivity of CR, particularly to small lesions. It may be that the potential role of US-detected erosions in gout is in monitoring response to therapy.

None of the studies have reported the feasibility of US imaging in gout, however it is widely accepted that US feasibility is acceptable, relative to other imaging techniques. The biggest issues are likely to be reliability, high user dependence and the time required to image multiple joints or lesions.

This review is necessarily limited by the quantity and quality of published manuscripts. No studies address potential confounders and reasons for false positives and false negatives, and the lack of standard definitions to date of US-detected pathology makes comparison of manuscripts difficult. The number of studies in this area is limited, particularly if the number of validity studies concerning US in RA is considered. The number of patients in most of the studies was small. Most of the studies were prospective and comparative in nature. There were no randomised control studies in this review. Additionally studies are not uniform and it is therefore difficult to compare them. Comparators used were different in different studies. While some used another imaging technique as comparator, others used other inflammatory diseases and healthy people. Some of the studies have no standard definition of the pathology assessed.

Only two studies looked at responsiveness to treatment and there were no feasibility studies.

Only two databases were searched. Therefore some of the published studies may have been missed, although review articles were examined to try to identify any articles that may have been missed.

In conclusion, US is a promising tool in the diagnosis and management of gout with a view to improving outcomes for our patients, however further work is required, particularly with regards to defining pathology (such as tophi), establishing the validity of the US-detected signs (especially the DC sign, snowstorm or cloudy soft tissue), identifying which lesions are optimal for monitoring disease and progression (synovitis, tophi or erosions) and to balance feasibility against responsiveness and validity, before the full clinical and research utility of US is established.

References

View Abstract

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.