Intra- and inter-observer variability of functional MR urography (fMRU) assessment in children

Functional MR urography (fMRU) provides comprehensive functional data that can be subject to variability. To interpret the results of fMRU, it is essential to know the intra- and inter-observer variability of the measured parameters. To define the range of variability in fMRU, particularly that of the differential renal function based on volume (volumetric differential renal function) and Patlak differential renal function measurements in children. We included 15 fMRU studies, 10 of non-duplicated and 5 of unilateral duplex kidneys. We recruited six observers with a range of fMRU experience, including two MRI technologists, one resident, one fellow, one pediatric radiologist and one pediatric urologist. The observers underwent intensive training in using the Children's Hospital of Philadelphia (CHOP)-fMRU freeware for analysis. They conducted the fMRU analysis on each case twice, at least 1 week apart. Mean and standard deviation were calculated for each set of absolute volume, absolute Patlak, volumetric differential renal function and Patlak differential renal function. We calculated the statistical significance of these deviations using the student's t-test. We also calculated interclass correlations for intra-observer and inter-observer agreement of both volume and Patlak measurements using SPSS software. Intra- and inter-observer variability did not differ significantly, measuring 6% and 4% for relative volume (volumetric differential renal function: P > 0.05) and 5% and 3% for relative function (Patlak differential renal function: P > 0.05). Absolute values of parameters showed more variability than the relative values. Intra- and inter-observer agreement was well above 0.90 (P < 0.001) for all volume measures except for duplex upper pole intra-observer measurements (0.80, P < 0.01). Intra- and inter-observer agreement for Patlak values were also above 0.90 (P < 0.001) except for duplex upper pole measurements, which were 0.54 (P = 0.13) and 0.81 (P < 0.01), respectively. Functional MRU analysis using CHOP-fMRU software is reproducible, with overall intra- and inter-observer variability rates of 5% for volumetric differential renal function and 4% for Patlak differential renal function. There was higher variability in volume and function measurements between upper and lower pole moieties of duplicated kidneys and for absolute volume and function values overall. A range of 45-55% for relative values of volumetric differential renal function and Patlak differential renal function could serve as the normal range.