Table 1

Value of ultrasonography in diagnosing early (rheumatoid) arthritis

Author (year) [ref] Phase*

Study population

Number; type of joints or tendons

US modality, definition of synovitis (A-F)

Outcome, Risk estimates (95% CI)

Authors' conclusions

Limitations^#

Filer (2011) [16]

3

58 IA

22 RA

38; MCP, PIP, MTP, wrists, elbows, shoulders, knees, ankles

GSUS & PDUS

A

US reclassifies 3 of 29 IA patients to RA (ACR 2010)

US of MCP, wrists and MTP improves optimally clinical models to predict RA

Q4E

Freeston (2010) [14]

3

30 patients with inflammatory symptoms and negative RF and aCCP tests

12; MCP, wrists and flexor tendons of fingers

GSUS & PDUS

B

US increases pretest probability for arthritis at one year from 6 to 94%

Inflammation at US predicts development of arthritis in patients with inflammatory symptoms and negative RF and aCCP tests

Q1D, Q2C

Ozgul (2009) [17]

3

51 IA

50; shoulders, elbows wrists, MCP, PIP, DIP, knees, ankles, MTP

GSUS

C

Kappa = 0.61 for US diagnosis of RA versus clinical diagnosis

Average agreement between US and clinical examination when diagnosing RA

Q1A, Q1E, Q4E

Salaffi (2010) [15]

3

149 UA

18; wrists, MCP2-5, PIP2-5

GSUS & PDUS

D

Sensitivity 0.35 (0.24, 0.48) and specificity 0.78 (0.67, 0.86) for US as baseline test, predicting RA at 1 year

Presence of PDUS signs predicts progression from UA to RA

Q1A, Q1C, Q1E, Q3B, Q4E

Stadt (2010) [18]

3

192 RF and/or aCCP-positive arthralgia patients, clinically without arthritis

Painful joint + adjacent and contralateral joints, mean 8 joints per patient

GSUS & PDUS

E

JE: OR = 3.07 (1.05, 8.9);

SH: OR = 5.5 (2.3, 12)

PD: OR = 5.5 (2.6, 12), all for future arthritis

Inflammation at US predicts future arthritis in aCCP-positive patients clinically without arthritis

Q4E

Wakefield (2004) [19]

3

80,

Oligoarthritis

First 40 patients: only symptomatic joints

Last 40 patients:

22; MTP, MCP, Knees

GSUS

F

1/3 of cases of oligoarthritis reclassified to polyarticular (>= 6 joints) arthritis

GSUS detects inflammation in more joints than clinical exam, but 8% of joints with clinical synovitis are normal at US

Q2C

A, according to Szkudlarek 2003 [30], 2006 [31] and Wakefield 2005 [32]. B, semiquantitative definitions based on binary definitions by Brown 2006 [11], Wakefield 2004 [19], Karim 2004 [33] and Newman 1996 [34]. C, (a) joint capsule elevated beyond normal range, or (b) synovial hypertrophy or effusion around the joints, or (c) erosions of the joint margins, or (d) change in the diameter of the tendon with or without peritendinous hypo-echogeneity supporting effusion. D, according to Wakefield 2005 [32] and Szkudlarek 2003 [30]. E, according to Szkudlarek 2003 [30]. F, abnormally hypo-echoic joint space, distinct from the intra-articular fat pad and non-compressible with the transducer. Ref, reference; US, ultrasonography; GSUS, greyscale ultrasonography; PDUS, power Doppler ultrasonography; aCCP, anti-cyclic citrullinated peptide; JE, joint effusion; IA, inflammatory arthritis; RA, rheumatoid arthritis; UA, undifferentiated arthritis; RF, rheumatoid factor; SH, synovial hypertrophy; OR, odds ratio. *Adaptation of phases according to Sackett and Haynes (Additional file 2, box 1): Phase 1. Do US results in patients with the condition differ from those without the condition? Phase 2. Are patients with certain US results more likely to have the condition? Phase 3. Do US results distinguish patients with and without the condition among those in whom it is clinically sensible to suspect the condition? Phase 4. Do patients undergoing US fare better in their ultimate health outcome than similar untested patients?. ^#See Additional file 3.

Ten Cate et al. Arthritis Research & Therapy 2013 15:R4   doi:10.1186/ar4132