ASSESSMENT OF LUPUS NEPHRITIS DISEASE ACTIVITY USING NON-CONTRAST MRI: A PILOT STUDY

Background: Current non-invasive approaches to monitoring disease activity in patients with lupus nephritis (LN) include serial measurement of glomerular filtration rate(GFR) and assessment of proteinuria. Both are influenced by renal damage and drugs hence, may not give a true reflection of LN activity. There is a need for a more accurate non-invasive biomarker for LN disease activity. Quantitative renal magnetic resonance imaging(MRI) techniques, such as, arterial spin labelling (ASL), diffusion tensor imaging (DTI) and T2Star sequences, require no exogenous contrast agents and measurements correlate with GFR in healthy volunteers and those with chronic renal disease. ASL provides a measure of microperfusion; DTI parameters are associated with early nephropathy changes and fibrosis. T2star values reflect microhaemorrhage. Objectives: We aimed to study quantitative renal MRI measurements in patients with SLE, to assess whether they are associated with GFR and /or levels of proteinuria. We also explored how these measures perform in SLE patients with or without biopsy-proven nephritis. Methods: A pilot study of SLE patients and healthy controls was conducted. Patients from 18 to 70 years old with 4 or more revised ACR criteria for SLE, were recruited from North West England. LN patients were required to have biopsy confirmed active LN, within 36 months of inclusion. Patients with a GFR <30 mL/min/1.73m2 or other known renal pathology were excluded. Healthy volunteers were age and sex matched. Clinical evaluation was undertaken including measurement of weight, urinary protein:creatinine ratios(PCR) and serum creatinine levels. GFR was calculated using the MDRD formula. MRI was performed on a 1.5 Tesla scanner. ASL, DTI and T2star sequences were acquired. Median values were calculated for cortex and medulla. Statistical analysis was performed using non-parametric tests. Results: 21 subjects completed the study (6 LN patients, 8 SLE patients without nephritis(LNN) and 7 healthy controls), the group included 18(88%) females. In the whole group the median (IQR) age, GFR and urinary PCR were 34(28,41) years,87(78.5, 96.4) mL/min/1.73m, and 10(6,18) respectively. The median (IQR) disease duration of SLE was 7(2,12) years and the median time from renal biopsy to MRI was 17.5 (7,100) weeks. DTI and T2star sequences were acquired in all subjects. Due to an imaging protocol error, ASL data was missing for 3 LN and 1 LNN subjects. In the whole group GFR correlated significantly with DTI values in the medulla (rho=0.47, p=0.03) and correlation with medullary ASL values approached significance (rho=0.458, p=0.06). PCR correlated strongly with DTI values in the renal cortex and moderately with T2star measurements (rho=-0.71, p<0.001 and rho=-0.53, p=0.013 respectively). In the LN group (n=3), ASL flow values in the medulla were significantly higher than in other groups (p=0.04) and this group also had a lower cortex signal on DTI which approached statistical significance (p=0.07). Conclusions: A number of MRI measures correlated with both GFR and PCR. We also noted increased renal perfusion in LN patients. Quantitative MRI may provide a novel, non-invasive approach to monitoring LN disease activity as an alterative to more invasive approaches such as serial biopsy. Disclosure of Interest: S. Skeoch: None Declared, M. Dobbs: None Declared, P. Hubbard: None Declared, J. Naish: None Declared, N. Woodhouse: None Declared, M. Ho: None Declared, J. Waterton: None Declared, G. Parker Shareholder of: Director of Bioxydyn Imaging, I. Bruce: None DeclaredCitation: , volume 72, supplement s3, year 2013, page Session: Abstract session: Imaging in diagnosis of rheumatic diseases ( )