Table 1

Chronological review of studies showing that magnetic resonance imaging bone marrow edema predicts erosive progression in rheumatoid arthritis




Association with erosion progression

Outcome measure (XR, MRI, or CT)

McQueen et al. [22]


In this NZ cohort, 42 patients with early RA were enrolled at presentation, within 6 months of symptom onset.

Baseline bone edema score predicted MRI erosion score at 1 year (OR = 6.47, P <0.001).

MRI erosion score at dominant wrist

Savnik et al. [35]


Danish cohort of 22 patients had RA for less than 3 years. Patients were followed at 1 year.

Presence of bone edema in wrist bones at baseline was the strongest individual predictor of bone erosions at 1 year (P <0.007).

MRI erosions at dominant wrist

McQueen et al. [23]


NZ cohort was reviewed at 6 years. Patients were treated with non-biologic DMARDs. MRI outcome data on 31 patients and XR outcome data on 34 patients are available.

Baseline bone edema score was only MRI feature on multivariate analysis to predict 6-year Sharp score (R2 = 0.20, P = 0.01). At each bone, OR (95% CI) for bone edema predicting MRI erosion was 6.5 (2.78 to 18.1).

XR erosion score at hands and feet and MRI erosion score

Palosaari et al. [33]


Twenty-seven patients with early RA had a disease duration of less than 12 months. Contrast-enhanced MRI data on 24 patients at 2 years are available.

Bone edema score was the only baseline variable to predict erosive progression at 2 years on multivariate regression (OR = 4.2, 95% CI = 1.3 to 13.8). At each bone, ORs (95% CIs) of predicted erosion at 1 and 2 years were 28 (11.7 to 67.1) and 14.9 (6.3 to 34.9), respectively.

MRI erosion score wrist

Haavardsholm et al. [28]


Eighty-four patients with RA had a disease duration of less than 1 year. Follow-up was at 1 year. Traditional DMARD therapy (plus anti-TNFs in two patients by 1 year) was used.

Baseline MRI bone edema score of more than 2 RAMRIS units was an independent predictor of both XR (OR = 2.77, 95% CI = 1.06 to 7.21) and MRI erosive progression.

MRI erosive progression at dominant wrist and XR erosive progression at the hands

Hetland et al. [29]


One hundred thirty patients with early RA had a disease duration of 3.3 years. Combination non-biologic DMARD therapy, including ciclosporin, or placebo was used. Two-year follow-up data are available.

Baseline bone edema score was the only independent predictor of 2-year change in Sharp score (multivariate linear regression): coefficient = 0.75 (95% CI = 0.55 to 0.94, P = 0.001).

XR erosion score at hands and feet

Mundwiler et al. [21]


Forty-six patients had a disease duration of less than 5 years. MRI and XR of 3rd, 4th, and 5th MTP joints of both feet were performed. Patients were treated with traditional and biologic DMARDs; 1- and 2-year data are available.

Bone edema predicted MRI erosions at 6 and 12 months with PPVs of 0.25 and 0.50 and NPVs of 0.99 and 0.99, respectively. At each bone, ORs for bone edema being followed by erosion were 34.2 and 68.0 at 6 and 12 months, respectively.

MRI erosion score at 3rd, 4th, and 5th MTP joints bilaterally

Hetland et al. [30]


In the same cohort as above (Hetland et al. [29]), 139 patients completed 5 years of follow-up. No treatment influence on erosion progression was noted. No biologics were used.

MRI bone edema was an independent predictor of XR progression (coefficient = 0.82, CI = 0.50 to 1.13, P <0.001). Bone edema explained 23% of the variation in the progression of the TSS (Pearson's r = 0.48).

Change in XR erosion score at hands and feet (TSS progression rate)

Døhn et al. [57]


Fifty-two patients with biologic-naïve RA and a disease duration of 7 years were followed for 12 months on anti-TNF therapy (adalimumab/methotrexate).

When baseline MRI bone edema was present versus not present, RR for erosive progression in the same bone on CT at 12 months was 3.8 (95% CI = 1.5 to 9.3, P = 0.004). If bone edema was 'ever present' versus 'never present', RR was 14.8 (95% CI = 4.3 to 50.7).

CT erosions at dominant wrist and 2nd to 5th MCP (site-by-site).

Bøyesen et al. [34]


Eighty-four patients with RA (same cohort as that of Haavardsholm et al. [28])

Baseline total MRI bone edema score predicted MRI erosive progression at 1 year with an OR of 1.28 (95% CI = 1.01 to 1.64, P = 0.04).

MRI erosive progression

anti-TNF, anti-tumor necrosis factor; CI, confidence interval; CT, computed tomography; DMARD, disease modifying antirheumatic drug; MCP, metacarpophalangeal; MRI, magnetic resonance imaging; MTP, metatarsophalangeal; NPV, negative predictive value; NZ, New Zealand; OR, odds ratio; PPV, positive predictive value; RA, rheumatoid arthritis; RAMRIS, rheumatoid arthritis magnetic resonance imaging score; RR, relative risk; TSS, total Sharp score; XR, x-ray.

McQueen Arthritis Research & Therapy 2012 14:224   doi:10.1186/ar4035