Continuous infusion of low dose iohexol as an accurate and precise method for determining glomerular filtration rate in pigs

Background The pig is a suitable animal for the study of human acute kidney injury (AKI), yet precise methods for determining glomerular filtration rate (GFR) are rarely applied in dynamic physiological conditions. Iohexol is a preferred precise marker of GFR in humans and is validated in pigs. After equilibration, continuous infusion of low dose iohexol (CILDI) may be used to monitor changes in GFR in dynamic conditions, but a protocol should be established under stable conditions. Objectives This study aimed to develop a CILDI protocol tailored to anesthetized pigs with normal renal function. Methods Iohexol pharmacokinetics were studied in three anesthetized pigs. Nine anesthetized pigs of both sexes (29.5–37 kg) were instrumented before being given iohexol (Omnipaque 300 mg I/mL) as a bolus dose of 1.16 mL and a continuous infusion of 0.4 mL/h. Doses were tailored using pharmacokinetic data. Blood and urine samples were obtained during 8 h and iohexol was analysed by LC/MS-MS. Time to plateau was estimated by nonlinear fitting of concentration-time profiles and visual assessment of decay curves. Plasma iohexol clearance (mGFR plasma ) was compared to creatinine clearance (CrCl) and renal iohexol clearance using Bland-Altman analysis and paired t-tests. Accuracy was defined as < 10% mean difference between plasma and renal clearance. Precision was assessed as the intra-individual coefficient of variation (CV) for mGFR plasma and serum creatinine (s-Cr). Results Plateau concentration was achieved within 120 min. mGFR plasma was 53.7 ± 7.5 mL/min (mean±SD). No significant difference was observed between renal and plasma iohexol clearance (bias 2.15 mL/min, p = 0.37), while CrCl was significantly higher compared to plasma iohexol clearance (bias 13.29 mL/min, p = 0.004). The mean CV for mGFR plasma was 2.3%, whereas the mean CV for s-Cr was 5.6% (mean difference − 3.3% p = 0.004). Within-animal changes in mean arterial pressure were positively associated with iohexol plasma clearance ( p = 0.003), but not with creatinine ( p = 0.107). Conclusions The tailored CILDI protocols achieved plateau plasma concentrations by 120 min in pigs with normal renal function and provided accurate and precise measurement of GFR thereafter. If applied without adaptations, time to plateau would likely be longer in reduced renal function.