1 of 16

5th International Renal Imaging Meeting���MRI: opportunities for functional renal imaging ��11th September 2023��Dr Iosif Mendichovszky

Department of Nuclear Medicine | Addenbrooke’s Hospital

Department of Radiology | University of Cambridge

2 of 16

The MRI toolbox

  • High-resolution MRI – macroscopic structural “roadmap” for pathology detection
  • Functional multiparametric MRI (mpMRI) – deriving and combining quantitative MRI parameters to assess (patho)physiological processes and tissue properties

1. Jiang K, Ferguson CM, Lerman LO. Noninvasive assessment of renal fibrosis by magnetic resonance imaging and ultrasound techniques. Transl Res. 2019 Jul;209:105-120.

2. Rankin AJ, Mayne K, Allwood-Spiers S, Hall Barrientos P, Roditi G, Gillis KA, Mark PB. Will advances in functional renal magnetic resonance imaging translate to the nephrology clinic? Nephrology (Carlton). 2022 Mar;27(3):223-230.

3. Francis ST, Selby NM, Taal MW. Magnetic Resonance Imaging to Evaluate Kidney Structure, Function, and Pathology: Moving Toward Clinical Application. Am J Kidney Dis. 2023 May 13:S0272-6386(23)00630-3.

4. Zhao K, Seeliger E, Niendorf T, Liu Z. Noninvasive Assessment of Diabetic Kidney Disease with MRI: Hype or Hope? J Magn Reson Imaging. 2023 Sep 7.�

​

​

​

3 of 16

Clinical conditions

  • Autosomal-dominant polycystic kidney disease (ADPKD)
  • Chronic kidney disease (CKD) staging and progression
  • Acute kidney injury (AKI)
  • Diabetic nephropathy
  • Renal transplantation
  • Renal artery stenosis
  • Systemic diseases affecting the kidneys (such as inflammatory / immune-mediated conditions)
  • Drug toxicity

4 of 16

Kidney Morphology�

  • T2-weighted or mDixon MRI for total kidney volume (TKV)
  • T1-weighted MRI for segmentation of the cortex and medulla
  • Limitations: time-consuming manual segmentation (prone to investigator error) for kidney, cortex and medulla
  • Opportunities: automatic accurate segmentation using deep learning (exploring pulse-sequence dependence and assembling large curated datasets for training and validation)

5 of 16

Tissue Microstructure�

  • Diffusion weighted (DWI), tensor (DTI), kurtosis (DKI) and magnetization transfer (MTI) imaging
  • Relaxometry mapping (T1, T1ρ and T2)
  • Magnetic Resonance Elastography (MRE)
  • High-field MRI (7T and above)
  • Limitations: susceptibility to artifacts & respiratory motion; dependence on b-value range; long acquisition times; non-standard acquisition (IVIM/DTI/DKI); large CoV (pseudo-diffusion and perfusion fraction)
  • Opportunities: accelerated acquisitions, T1ρ optimization, automated corrections for B0 and B1 effects, multi-vendor and multi-site validation, better modelling (DTI), better hardware and software (MRE, 7T)

6 of 16

Kidney Oxygenation�

  • Blood oxygen(ation) level-dependent (BOLD) and T2 relaxation under spin tagging (TRUST) MRI
  • Limitations: influence of scanner-related and non-renal factors (hampering the comparison of absolute R2* values across sites); BOLD, on its own, cannot distinguish between Δ renal oxygen supply and Δ renal oxygen consumption
  • Opportunities: elucidate the influence of technical factors, analysis methods, blood volume, blood flow, hydration status, age, hematocrit, sodium intake, pH, body temperature; increase TRUST MRI availability, clarify its biological underpinnings and perform clinical validation

​

​

​

7 of 16

Kidney Hemodynamics�

  • Phase-contrast MRI (PC-MRI) and arterial spin labelling (ASL)
  • Limitations: accurate slice positioning difficult for PC-MRI (must be perpendicular to the vessel direction) and needs good quality MRA; renal ASL not usually optimized or available on MR vendor platforms, low SNR
  • Opportunities: multi-vendor and multi-site validation in healthy controls and various clinical conditions, accelerated acquisitions with improved SNR, reduction in artefacts and development of (semi-)automated slice positioning methods

​

​

8 of 16

Kidney Metabolism�

  • Sodium (23Na), quantitative susceptibility mapping (QSM), Chemical Exchange Saturation Transfer (CEST) and 13C-hyperpolarized MRI
  • Limitations: small number of studies with limited number of patients across various pathologies (23Na MRI); complex technical setup and logistics (13C-hyperpolarized MRI); low SNR, long acquisition times and low spatial resolution
  • Opportunities: technical developments to reduce acquisition times and increase SNR; elucidate the relationships between MRI-derived measurements and biological processes; multi-vendor and multi-site validation / clinical trials

​

9 of 16

Data Acquisition�

  • Current status: (partly) informed by the PARENCHMA (RENALMRI.org) position and technical recommendations papers
  • Opportunities:
    • harmonized approaches to data collection across MRI vendors
    • assessment of the repeatability of kidney MRI measures
    • optimized pulse-sequences to specifically address, either independently or in combination, biological and/or clinical questions
    • reduce acquisition times, improve spatial resolution and SNR
    • new acquisition and reconstruction methods
    • interconnected image storage networks and QC procedures

​

​

10 of 16

Data Analysis�

  • Current status: (partly) informed by the PARENCHMA (renalmri.org) position and technical recommendations papers
  • Opportunities:
    • refine, implement and validate standardized and (semi)automatic post-processing methods
    • develop robust data handling procedures
    • design and adopt quality assurance standard operating procedures
    • create accurate and standardized methods for model-driven and/or ML/AI registration (respiratory and patient-related motion)
    • develop accurate, efficient and reproducible segmentation methods
    • use tailored data analysis pipelines

​

​

​

11 of 16

Clinical Interpretation�

  • Limitations: emerging area of clinical practice with scarce clinical reporting experience, mainly as part of research studies +/- case reports
  • Opportunities:
    • data sharing and imaging repositories
    • reporting workshops
    • structured reporting guidelines
    • multidisciplinary collaborations

​

​

12 of 16

The journey so far….

Systematic reviews and statement papers

​

13 of 16

The journey so far….

Consensus-based technical recommendations papers

​

14 of 16

The journey so far….

15 of 16

Clinical Viewpoint�

  • Although ever more complex methodological developments are welcome, we should first get the basics right
  • Clinical translation will succeed or fail based on high quality, reproducible and robust evidence becoming available that, together with health economics analyses, will facilitate incorporation into clinical guidelines, policy changes and funding approvals

16 of 16

The journey ahead (proposal)?

Continue to shape the field of renal MRI by:

    • “audit” the adoption of the consensus-based technical recommendations and statements by the research community
    • develop a set of standards (“checklists”) to facilitate the detailed and structured reporting of study methodologies to aid future comparisons and meta-analyses
    • develop recommendations and standards for emerging techniques (23Na, MRE, QSM, CEST, 13C-hyperpolarized MRI, etc) and their clinical translation
    • imaging biomarker “roadmap” for diffuse renal MRI studies statement paper

RenalMRI.org community initiative - working groups & topics to be decided

​

​