Table 2

Point to consider 1: Summary of included studies comparing imaging with CE in the detection of joint inflammation

UltrasoundMRICR
US knees vs CE
13 studies131433343876–83
MRI knees vs CE
studies13–15367284–87
CR knees vs CE
3 studies131585
Detection rate, mean (range)
US vs CE
Detection rate, mean (range)
MRI vs CE
Detection rate, mean (range)
CR vs CE
Synovitis/effusion (12 studies)13 14 34 38 76–831.19-fold (0.14–3.67-fold)Synovitis vs clinical swelling (3 studies)14 85 861.02-fold (0.96–1.12-fold)Joint distension vs swelling (3 studies)13 15 850.69-fold (0.45–1.0-fold)
Effusion vs swelling (5 studies)13–15 36 851.07-fold (0.75–1.33-fold)
Effusion (1 study)33Agreement k=0.54
CE missed a significant no. of effusions
Effusion vs pain (1 study)131.45-fold (1.33–1.57-fold)
PD vascularity (2 studies)38 821.63-fold (0.96–2.71-fold)Synovial volume vs CRP (1 study)87r=0.51–0.80
p=0.000–0.036
Joint distension vs pain (1 study)131.57-fold
Synovial hypertrophy vs pain (1 study)84r=0.68–0.74
US hip vs CE
5 studies1316788889
MRI hip vs CE
5 studies131590–92
CR hip vs CE
1 study15
Synovitis/effusion (5 studies)13 16 78 88 890.85-fold (0.13–1.39-fold)MRI inflammation (4 studies)13 15 90 910.88-fold (0.50–1.78-fold)Joint distension vs clinical effusion (1 study)150.80-fold
Synovitis/effusion vs LOM (1 study)89Association
p=0.006
Synovitis/effusion vs pain (1 study)89Association
p=0.103
Synovial enhancement (1 study)910.94-fold
US hands/wrists vs CE
4 studies38409394
MRI hands/wrists vs CE
2 studies1995
CR hands/wrists vs CE
1 study93
Synovitis/effusion (3 studies)38 93 940.93-fold (0.47–1.33-fold)Synovitis volume vs total hand swelling score (1 study)95r=0.52–0.72
p<0.05
Joint distension vs clinical effusion (1 study)930.63-fold
PD vascularity (2 studies)38 400.96-fold
GS synovitis had weaker correlation with clinical disease activity than PD
Synovitis volume vs LOM (1 study)95r=0.76
p<0.05
Flexor/extensor tenosynovitis (1 study)40Significant association with clinical disease activitySynovitis score vs wrist swelling score (1 study)19MRI score significantly higher with higher swelling score
p<0.00001
US ankles/feet vs CE
5 studies387996–98
MRI ankles/feet vs CE
1 study99
Synovitis/effusion (3 studies)38 79 960.97-fold (0.86-1.04-fold)Tibiotalar synovitis (1 study)991.00-fold
PD vascularity (1 study)380.57-foldSubtalar synovitis (1 study)993.33-fold
US TMJ vs CE3 studies25100101MRI TMJ vs CE
8 studies254142102–106
Synovitis/effusion (2 studies)25 10011.7-fold (0.35–23.0-fold)Synovitis (6 studies)25 41 42 104–1062.46-fold (1.10–5.91-fold)
Synovitis vs reduced MIO (4 studies)25 102 103 105Significantly correlated
Reduced MIO best predictor of active MRI changes
Acute changes (1 study)4171% asymptomatic
63% normal CE
US enthesitis vs CE
3 studies107–109
MRI enthesitis vs CE
1 study110
Enthesitis (3 studies)107–1090.79-fold (0.53–1.09-fold)Enthesitis (1 study)1100.50-fold
US VARIOUS MULTIPLE JOINTS vs CE
9 studies485064111–117
MRI VARIOUS MULTIPLE JOINTS vs CE
1 study110
CR VARIOUS MULTIPLE JOINTS vs CE
1 study118
Synovitis/effusion (6 studies)48 111–114 1161.85-fold (1.00–3.33-fold)Synovitis/effusion (1 study)1101.08-foldSoft tissue swelling vs clinical swelling (1 study)1181.05-fold
Association US changes vs swelling (1 study)48SH: r=0.63
Effusion: r=0.66
PD: r=0.50
Association synovitis vs CE (2 studies)50 117Swelling: r=0.50
LOM: r=0.40
Pain: r=0.21
CE missed inflammation in 25.2% joints
 MRI cervical spine vs CE
1 study20
 
Synovitis/SH (1 study)204.25-fold
MRI SIJ vs CE
2 studies23 119
Sacroiliitis (2 studies)23 1190.93-fold
CE was normal in 22.9% patients with MRI sacroiliitis
  • CE, clinical examination; CR, conventional radiography; CRP, C reactive protein; GS, grey scale; LOM, limitation of movement; MIO, maximal incisional opening; PD, power Doppler; SIJ, sacroiliac joint; SH, synovial hypertrophy; TMJ, temporomandibular joint; US, ultrasound.