Abstract Background and Aims The ZEUS study was a multi-center open-label randomized controlled trial designed to investigate the effect of an early conversion from a calcineurin-inhibitor-based immunosuppression to a mammalian-target-of-rapamycin inhibitor-based regimen on graft function twelve months post-transplantation. Persistent improvement of renal function up to five years together with albeit higher rejection rates were found in the intervention group. In the context of an investigator-initiated local sub-study, functional magnetic resonance imaging (fMRI) was prospectively performed in a subset of the study population in order to non-invasively assess differences in renal graft oxygenation, diffusion and perfusion between groups and time-points using blood-oxygen level-dependent (BOLD)- and diffusion-weighted (DWI) MRI respectively. Method In the ZEUS trial, 300 de novo kidney graft recipients were randomized to continue ciclosporin (CsA) or be converted to everolimus (EVE) at 4.5 months post-transplant. Concomitant immunosuppression contained enteric-coated mycophenolic acid, steroids and basiliximab. From the 36 subjects included in the ZEUS study at our study center, sixteen were enrolled in the MRI sub-study. FMRI including BOLD-MRI and DWI was prospectively performed at month 4.5 and 12 post-transplantation on a 3 Tesla and 1.5 Tesla (n=3) MR scanner. Blinded image analysis was done by placing up to 24 regions of interest in cortex and medulla. BOLD-MRI yielded the relaxation rate R2*, a parameter indirectly proportional to tissue oxygen content, DWI yielded the perfusion-cleared apparent diffusion coefficient ADCD, a marker of tissue diffusivity, and the perfusion fraction FP, a marker of microperfusion. All data were analyzed using in-house custom-scripts written in IDL® and MATLAB®. Results After exclusion of subjects due to image quality, outlier values or missing data, BOLD data was analyzed in eight, DWI in ten subjects. Mean absolute values and standard deviations corresponded to published values. A clear cortico-medullary gradient was shown for R2* and FP values, but not for ADCD, corresponding to previous results. For all fMRI parameters, there was no change for the whole population from 4.5 to 12 months post-transplantation in cortex and medulla. FMRI parameters did not differ significantly between both medication groups at month 4.5. In BOLD, R2* values increased (decreased tissue oxygen) over time in the CsA-treated and decreased in the EVE-treated group in cortex and medulla. Similarly, R2* values at month 12 were higher in the CsA- vs. EVE-treated group; the low number of subjects having completed BOLD-MRI at month 12 precluded formal statistical testing. In DWI, ADCD decreased in the CsA-group over time, whereas it increased in the EVE-group in cortex and medulla (p=0.046). Likewise, ADCD at month 12 was lower by trend in the CsA- vs. EVE-treated group in both medulla and cortex. The change in ADCD from month 4.5 to month 12 significantly differed in cortex (p=0.033) and medulla (p=0.019) between treatment groups. There was no significant difference for FP between medication groups nor over time. Baseline renal function and blood pressure were comparable in both groups. Conclusion The R2* values obtained by BOLD point to reduced graft oxygenation on CsA and ameliorated graft oxygenation on EVE, possibly due to the known vasoconstrictor effect of calcineurin inhibitors. The observed decreased diffusivity by DWI on CsA might be due to increased tissue density in the context of beginning calcineurin inhibitor-induced interstitial fibrosis. There was no clear trend for microperfusion as assessed by FP. Limitations of this study are the low number of subjects available for analysis. In conclusion, this prospective sub-study of the ZEUS trial suggests an impact of immunosuppressive regimen on fMRI parameters of the renal graft.
Read full abstract