Author | Feature (method) | Structural severity or progression outcome | Adjustment for confounders | Association (magnitude) crude | Association (magnitude) adjusted | Association | Quality (score %) |
---|---|---|---|---|---|---|---|
Hand MRI bone marrow lesion case series | |||||||
Haugen 2014 [141] | BMLs - semi-quantitative at 2nd to 5th IPJs (C) | Progression of hand ROA (JSN, KL grade or new erosion) (L) | Age, sex, BMI, | OR 2.73, 95 % CI 1.29 to 5.78 | NR | + | High (61) |
Bigger the BML, the more the JSN | |||||||
Hand MRI bone marrow lesion cross-sectional studies | |||||||
Haugen 2012 Abstract [143] 299 | BML (Oslo MRI hand score) (C) IPJs | Radiographic JSN grade IPJ (OARSI atlas) (C) | Age, sex, | OR 10.0, 95 % CI 4.2 to 23 | OR 4.4, 95 % CI 2.2 to 9.0 | + | Low (43) |
BML score association with more JSN | |||||||
Haugen 2012 [142] | BML (Oslo MRI hand score) (C) IPJs | Hand KL grade of IPJs (C) | Age, sex | NR | OR (95 % CI) | + | High (64) |
BMLs 11 (5.5 to 21) | |||||||
p <0.001 | |||||||
Hand MRI osteophyte cross-sectional studies | |||||||
Haugen 2012 [142] | Osteophyte (Oslo MRI hand score) (C) IPJs | Hand KL grade of IPJs (C) | Age, sex | NR | OR (95 % CI) | + | High (64) |
osteophytes | |||||||
415 (189 to 908) | |||||||
p <0.001 | |||||||
Hand MRI attrition cross-sectional studies | |||||||
Haugen 2012 [142] | Attrition (Oslo MRI hand score) (C) IPJs | Hand KL grade of IPJs (C) | Age, sex | NR | OR (95 % CI) attrition 87 (37 to 204) | + | High (64) |
p <0.001 | |||||||
Hand MRI bone cyst cross-sectional studies | |||||||
Haugen 2012 [142] | Cyst (Oslo MRI hand score) (C) IPJs | Hand KL grade of IPJs (C) | Age, sex | NR | OR (95 % CI) | Nil | High (64) |
cysts 2.0 (0.6 to 6.3) | |||||||
p = 0.26 | |||||||
Hip MRI BML cross-sectional studies | |||||||
Neumann 2007 [46] | Semi-quantitative BMLs (C) | Semi-quantitative cartilage lesions (C) | Nil | R = 0.44, p ≤0.001 | NR | + Correlation between BML and cartilage lesions | Low (43) |
Dawson 2013 Abstract [69] | Femoral head BMLs (MRI) (C) | 1. Presence of hip OA. 2. Femoral head cartilage volume (MRI) (C) | Age, sex, BMI | NA | OA hip presence | + | Low (14) |
BMLs associated with diagnosis of hip OA | |||||||
OR (95 % CI) | |||||||
5.32 (1.78 to 15.9) | |||||||
p = 0.003 | BMLs inversely associated with cartilage volume | ||||||
cartilage volume | |||||||
regression coefficient (95 % CI) | |||||||
−245.7 mm3 | |||||||
(−456 to −36) p = 0.02 | |||||||
Hip CT bone morphometry cross-sectional studies | |||||||
Chiba 2011 [33] | Acetabular and femoral head subchondral trabecular morphometry: bone volume fraction, trabecular thickness, number, separation (CT) (C) | Hip joint space volume (CT) (C) | Nil | Femoral head Bone volume fraction r = −0.691, p <0.001 | NR | Joint space narrowing is associated with increased bone volume fraction, trabecular thickening. trabecular number and spacing decrease | Low (57) |
Hip DXA BMD cross-sectional studies | |||||||
Chaganti 2010 [30] | Femoral neck BMD (C) DXA | Hip ROA Modified Croft score (categorical 0–4) (C) | Age, BMI, height, activity level, race, 6-m walk pace, Nottingham muscle strength, inability to do chair stands, and clinic site, | NR | p <0.0001 | + | High (64) |
Higher BMD for higher grade of OA of hip | |||||||
Antoniades 2000 [32] | DXA BMD of the femoral neck of left (nondominant) hip with ROA (C) | Radiographic OA (Croft score) (C) | BMI, lifetime physical activity, menopausal status, use of oestrogen, and smoking | OR 1.63, 95 % CI 1.06 to 2.50) | OR 1.80, 95 % CI 1.05 to 3.12 | + Association between BMD and hip ROA grade in the index hip | High (64) |
Higher OA grade means higher BMD | |||||||
2D Hip bone shape longitudinal studies | |||||||
Agricola 2013 [146] | Baseline 2D femoral and acetabular shape modes (segmented by statistical shape modelling) (C) | THR at or within 5 years (L) | Age, sex, BMI, shape modes | 5 modes were associated with THR OR 1.71 to 2.01, p ≤0.001 | 3 modes were associated with THR OR 1.78 to 2.10, p ≤0.001 | + Increasing femoral head asphericity is associated with THR | High (72) |
Agricola 2013 [147] | Baseline alpha angle (2D femur shape) dichotomous abnormal >60 °, normal ≤60 ° (C) | Incident ROA hip (KL >1), incident end-stage ROA hip (KL >2 or THR) at or within 5 years (L) | Age, sex, BMI, KL grade | OR (95 % CI) Incident ROA hip 6.82 (3.55 to 13.10) p <0.0001 | OR (95 % CI) incident ROA hip 2.42 (1.15 to 5.06) p = 0.02, incident severe ROA or THR 3.67 (1.68 to 8.01) p <0.0001 | + Elevated alpha angle is associated with incident end-stage OA hip | High (67) |
Agricola 2013 [148] | Baseline 2D centre edge angle (acetabular shape): 25 ° <normal <40 °, undercoverage <25 °, overcoverage >40 ° (C) | Incidence within 5 years of: 1. ROA hip (KL >1), 2. end-stage OA (KL >2 or THR) | Age, sex, BMI, KL grade | OR (95 % CI) overcoverage 0.52 (0.19 to 1.43) p = 0.21, undercoverage 3.64 (1.91 to 6.99) p = 0.00 | OR (95 % CI) overcoverage 0.34 (0.13 to 0.87) p = 0.025, undercoverage 5.45 (2.40 to 12.34) p = 0.00 | Overcoverage is protective against OA incidence (−). Undercoverage is associated with greater odds of OA incidence and end-stage OA (+) | High (67) |
2D and 3D hip bone shape cross-sectional studies | |||||||
Gosvig 2010 [149] | Categorical hip 2D deformity: 1. normal, 2.‘pistol grip’, 3) deep acetabular socket (C) | Presence of radiographic hip OA (JSW ≤2 mm) (C) | Age, sex, BMI, other hip deformities | NR | RR (95 % CI) pistol grip 2.2 (1.7 to 2.8) p <0.001, deep acetabular socket 2.4 (2.0 to 2.9) p <0.001, normal (p >0.05) | + | Low (50) |
Reichenbach 2011 [27] | The presence or absence of any 3D semi-quantitative MRI-defined cam-deformity (C) | Combined femoral and acetabular cartilage thickness (C) | Age, BMI (NB all participants were young men) | Unadjusted mean cartilage thickness difference with CAM deformity −0.24 mm (95 % CI −0.46 to −0.03) | Adjusted mean cartilage thickness difference with CAM deformity −0.19 mm (95 % CI −0.41 to 0.02) | NC | High (64) |
2D hip bone shape case control studies | |||||||
Doherty 2008 [43] | Non-spherical femoral head 2D shape assessment: 1. appearance of ‘pistol grip deformity’ (C), 2. maximum femoral head diameter ratio to minimum parallel femoral neck diameter (C) | Presence of radiographic hip OA (JSW ≤2.5 mm) (C) | Age, sex, BMI, BMD, physical activity, history of hip injury, type 3 hand (index finger shorter than ring finger), hand nodes, and center-edge angle | OR (95 % CI) pistol grip deformity 5.75 (4.00 to 8.27). Femoral head-to-neck ratio 10.45 (7.16 to 15.24) | OR (95 % CI) pistol grip deformity 6.95 (4.64 to 10.41). Femoral head-to-neck ratio 12.08 (8.05 to 18.15) | + | Low (53) |
Barr 2012 [150] | 2D Shape measures of centre-edge angle (acetabular shape) (C) | THR vs no radiographic progression over 5 years (L) | Age, gender, BMI KL grade, use of walking stick, WOMAC function, duration of pain | OR (95 % CI) mode 2 0.74 (0.50 to 1.10) p >0.05 | OR (95 % CI) Mode 2 0.17 (0.04 to 0.71) p <0.05 | NB, this model association is inverse and correlates with acetabular shape | High (76) |
Nicholls 2011 [39] | CAM deformity; mean modified triangular index height, alpha angle. 2D acetabular dysplasia; mean lateral center edge angle, (C) | Total hip replacement (L) | BMI, age | OR (p value) triangular index 1.131 (0.021). Alpha angle 1.056 (<0.0005). Centre edge angle 0.906 (0.004) | OR (p value) Triangular index 1.291 (0.011). Alpha angle 1.057 (<0.0005). Centre- edge angle 0.887 (0.002) | + Association of hip replacement with CAM impingement and acetabular dysplasia indicated by these results | High (71) |
Ankle scintigraphic subchondral bone cross-sectional studies | |||||||
Kraus 2013 [60] | Ipsilateral late phase bone scintigraphy, retention presence in tibiotalar joint (C) | Tibiotalar ROA KL grade and JSN (C) | Age, gender, BMI | NR | KL grade r = 0.49, p <0.0001. JSN r = 0.35, p <0.0001 | + | High (71) |
Knupp 2009 [24] | Late phase bone scintigraphy, semi-quantitative retention scoring of tibiotalar joint (C) | Tibiotalar ankle joint JSN. (modified Takakura score) (C) | Nil | 0.62 to 0.75 (p <0.01) | NR | + | Low (57) |