Purpose or Case Report: ARPKD causes diffuse microcysts in the kidney parenchyma, and cyst burden cannot be quantified by standard image segmentation methods. We sought to assess whether apparent diffusion coefficient (ADC) measured by DWI could serve as a non-invasive biomarker of ARPKD severity, with the hypothesis that ADC would be higher in cystic vs. non-cystic parenchyma due to higher extracellular water content. We examined whether ADC could distinguish healthy kidneys from those affected by ARPKD, and whether a threshold ADC value could be established to differentiate cystic vs. non-cystic parenchyma.
Methods & Materials: 12 individuals with ARPKD (age 14.4±5.9y) and 10 healthy controls (age 12.5±4.0y) were compared cross-sectionally. DWI was acquired in a fasting state with parameters: TR 7100ms; TE 67ms; slice thickness 4mm, b-values: 0, 200, 500, 800, and 1000 s/mm² (Siemens Skyra 3T). Whole-kidney regions of interest were drawn manually, excluding the collecting system and large blood vessels, and ADC was calculated using a mono-exponential decay model (pMRI software, parametricmri.com). Mean and median kidney ADC were compared in healthy controls vs. ARPKD (Mann-Whitney U test). Histograms of voxel-wise ADC frequencies were examined in control and ARPKD groups to determine if the distribution was skewed towards higher ADC in cystic parenchyma.
Results: Mean kidney ADC values were similar in the ARPKD group (3.42±0.40 x 10^-3mm²/s) and healthy controls (3.44±0.47 x 10^-3mm²/s) (p =0.9). On histogram analysis, voxel-wise ADC frequencies were not normally distributed in either control or ARPKD groups, with significantly right-skewed distribution with long tails towards high ADC values and a large number of high-ADC outliers in both groups (ARPKD vs. control skewness 3.6 vs. 4.0 and kurtosis 21.5 vs. 25.6). Given the non-normal distribution of ADC values, we also compared median kidney ADC between the ARPKD and control groups and found no statistically significant difference (p = 0.5).
Conclusions: ADC values obtained from DWI do not appear to distinguish between cystic and non-cystic kidney parenchyma in individuals with ARPKD, possibly due to confounding effects of kidney fibrosis on ADC measurements in individuals with ARPKD. Further studies are needed to determine if other MRI sequences could identify potential biomarkers of kidney disease severity in ARPKD.