Parenchymal volume analysis to assess longitudinal functional decline following partial nephrectomy

Abstract

Objective: To identify factors associated with longitudinal ipsilateral functional decline after partial nephrectomy (PN). Patients and Methods: Of 1140 patients managed with PN (2012–2014), 349 (31%) had imaging/serum creatinine levels pre-PN, 1–12 months post-PN (new baseline), and >3 years later necessary for inclusion. Parenchymal-volume analysis was used to determine split renal function. Patients were grouped as having significant renal comorbidity (CohortSRC: diabetes mellitus with insulin-dependence or end-organ damage, refractory hypertension, or severe pre-existing chronic kidney disease) vs not having significant renal comorbidity (CohortNoSRC) preoperatively. Multivariable regression was used to identify predictors of annual ipsilateral parenchymal atrophy and functional decline relative to new baseline values post-PN, after the kidney had healed. Results: The median follow-up was 6.3 years with 87/226/36 patients having cold/warm/zero ischaemia. The median cold/warm ischaemia times were 32/22 min. Overall, the median tumour size was 3.0 cm. The preoperative glomerular filtration rate (GFR) and new baseline GFR (NBGFR) were 81 and 71 mL/min/1.73 m2, respectively. After establishment of the NBGFR, the median loss of global and ipsilateral function was 0.7 and 0.4 mL/min/1.73 m2/year, respectively, consistent with the natural ageing process. Overall, the median ipsilateral parenchymal atrophy was 1.2 cm3/year and accounted for a median of 53% of the annual functional decline. Significant renal comorbidity, age, and warm ischaemia were independently associated with ipsilateral parenchymal atrophy (all P < 0.01). Significant renal comorbidity and ipsilateral parenchymal atrophy were independently associated with annual ipsilateral functional decline (both P < 0.01). Annual median ipsilateral parenchymal atrophy and functional decline were both significantly increased for CohortSRC compared to CohortNoSRC (2.8 vs 0.9 cm3, P < 0.01 and 0.90 vs 0.30 mL/min/1.73 m2/year, P < 0.01, respectively). Conclusions: Longitudinal renal function following PN generally follows the normal ageing process. Significant renal comorbidities, age, warm ischaemia, and ipsilateral parenchymal atrophy were the most important predictors of ipsilateral functional decline following establishment of NBGFR.