Advertisement

Predictive ability of renal cortex enhancement in dynamic CT for residual renal function after nephroureterectomy: Comparative study with renography

Login to Access Video or Poster Abstract: MP71-15
Sources of Funding: none

Introduction

The precise prediction of perioperative change in renal function would enable the appropriate management including perioperative systemic chemotherapy for patients with upper tract urothelial carcinoma (UTUC). The renal cortex enhancement in CT could reflect renal function. We have reported the usefulness of renal cortex enhancement to predict residual single kidney function after nephroureterectomy (NU) [AUA 2015]. The predictive ability of renal cortex enhancement in CT was evaluated in comparison with that of nuclear renography.

Methods

Forty-seven patients with UTUC in whom both dynamic CT and Tc-99m diethylenetriamine pentaacetic acid renography were performed before NU were subjects of this study. Pre- and postoperative estimated glomerular filtration rates (eGFRs) were calculated using serum creatinine measured before and 1-3 months after NU, respectively. Renal cortex radiodensity was measured in Hounsfield units (HU) from a round region of interest placed over the highest enhancing area on corticomedullary images. CT-based split renal function (SRF) of the intact contralateral kidney was calculated by splitting pre-eGFR: pre-eGFR × percent of total HU. The same formula was used for nuclear SRF. The correlations between observed and calculated post-eGFR using CT and nuclear renography were assessed with a calibration plot.

Results

The median pre-eGFR was 77 (range: 41-124) ml/min/1.73 m2, and decreased to 53 (27-89) after NU. The median percent of total HU for the contralateral kidney was 52% (48-68), and that of the total nuclear SRF was 56% (43-100). CT-based SRF showed linear correlation with the observed post-eGFR (r = 0.80), which was equivalent to the nuclear SRF (r = 0.78). From the conversion factors derived from a linear regression, an updated prediction formula was developed by parallel translation, adding 11.7 to the CT-based SRF and 8.8 to the nuclear SRF. The calibration plot in which the observed post-eGFR was plotted against each updated SRF showed equally good agreement (Fig.).

Conclusions

The predictive ability of CT enhancement for postoperative renal function is feasible and comparable to that of renography. This CT model could be useful in clinical practice because CT is an essential diagnostic tool for patients with UTUC.

Funding

none

Authors
Yuma Waseda
Kazutaka Saito
Masaharu Inoue
Masaya Ito
Toshiki Kijima
Soichiro Yoshida
Minato Yokoyama
Junichiro Ishioka
Yoh Matsuoka
Kazunori Kihara
Yasuhisa Fujii
back to top