Brain qMRI after Liver Transplantation (Hannover)

This entry covers the brain MRI cohort of a single-center cross-sectional study at Hannover Medical School in Germany. The study asked whether long-term treatment with calcineurin inhibitors (cyclosporine A or tacrolimus) after orthotopic liver transplantation leaves measurable traces in brain tissue, and whether the effect depends on the dose. It combined neurological and psychometric testing with quantitative MRI. The images are not released. The paper reports group results only, and its figshare record holds two supporting tables of test statistics.

Composition

The qMRI analysis includes 85 liver transplant recipients, seen a median of about ten years after surgery, and 31 healthy controls matched for age, sex and education. Patients are split by their immunosuppression at the time of the visit: 19 without calcineurin inhibitors, 36 on a reduced dose and 30 on a standard dose. Most patients who left the standard dose did so because their kidney function declined. Enrollment ran from February 2014 to February 2016. Of 91 enrolled patients, 85 met all criteria. Twelve patients had a history of hepatic encephalopathy before transplantation.

Acquisition

All participants were scanned on one Siemens Verio 3T scanner with a 12-channel head coil. The protocol had an axial triple-echo turbo spin echo sequence, an axial triple-echo gradient echo sequence, both with 1 x 1 x 3 mm voxels, and a single-shot echo-planar diffusion tensor sequence with 12 gradient directions at b = 1500 s/mm2. The scanner computed T2 and T2* maps from the echo series; ADC and FA maps came from the diffusion data. T2' was derived from T2 and T2*.

Annotations

Values were read from circular 15 mm2 regions of interest placed by hand in 17 gray and white matter structures per hemisphere, and left and right values were averaged. Two senior neuroradiologists reviewed the T2-weighted images for structural abnormalities. No masks or per-subject values are published.

Known limitations

• Images, parameter maps and region values are not shared, and no terms or request route are published.
• An earlier paper on the same patients (Pflugrad et al. 2018) uses slightly different group sizes.
• Age is reported only as a mean per group, so no age bins are listed.