Human Kidneys Play an Important Role in the Elimination of Propofol

After institutional approval (Graduate School of Medicine, Gunma University, Showa-machi, Maebashi, Japan), informed consent was obtained from 10 patients undergoing nephrectomy under the diagnosis of renal cell carcinoma (5 right and 5 left nephrectomy; 8 males and 2 females; age, 66 ± 9.2 yr; height, 165 ± 7.7 cm; weight, 64 ± 7.8 kg). Individuals who had severe hepatic insufficiency, significant hemodynamic instability, or a known allergy to eggs, propofol, or para -aminohippurate (PAH) were excluded from the study.

Before induction of anesthesia, 3 ml blood was taken to obtain a blank sample of plasma for the measurement of PAH. Thereafter, 5% PAH infusion was started at a rate of 30 ml/h for the initial 5 min and 10 ml/h until the end of the operation. Anesthesia was induced with vecuronium (0.1 mg/kg) and propofol (2 mg/kg) and then maintained with nitrous oxide (60%) and sevoflurane (1∼2%) in oxygen with an infusion of propofol (2 mg · kg⁻¹· h⁻¹). Blood samples were collected from a cannula inserted in the radial artery every 30 min. Because it has been reported that the concentration of propofol reached more than 85% of the final steady state value,3the level of propofol at least 2 h after the start of constant infusion was regarded as steady state. The renal venous blood was obtained immediately after the ligation of the renal vein for nephrectomy. At the same time, the radial arterial blood was also obtained. Because we could not obtain a blood sample from the renal vein of the other side, the extraction ratio of both sides of the kidney was assumed to be the same. After ligation of the renal vein, blood samples of the radial artery were collected every 30 min in the same way until the end of the operation.

Unbound fraction of propofol to arterial plasma protein was estimated using equilibrium dialysis method. Plasma samples (1 ml each) were dialyzed against an isotonic buffer containing 0.067 m sodium phosphate and 0.05 m sodium chloride (pH 7.4) for 10 h at 37°C using a dialysis membrane with a molecular weight cutoff of 6,000 (VB-8, Sanplatec, Japan). In the preliminary experiments, the concentrations of propofol in both sides of the dialysis membrane reached equilibrium after 5 h, and 80% of the added amount of propofol was recovered after dialysis. After dialysis, the propofol concentrations in the plasma and the dialyzed fractions were analyzed separately. The unbound fractions of propofol in plasma (fu) and blood (fB) were calculated according to the following equations:

Five milliliters heparinized blood was centrifuged immediately, and then obtained plasma samples were stored at −20°C until subsequent analysis. The plasma concentration of PAH was measured using the method of Brun.17 

The propofol concentrations in whole blood from artery and renal vein, plasma from artery, and dialysate and retentate after dialysis were measured using high-performance liquid chromatography as reported previously.2 

Renal plasma flow (RPF) was determined by the steady state infusion and extraction technique using PAH as an indicator. RPF and renal blood flow (RBF) were calculated by equations 1 and 2, respectively.

where Ca,PAH and Cv,PAH are the steady state concentrations of PAH in plasma at the radial artery and renal vein, respectively.

The renal extraction ratio of propofol (E(r)) was calculated as

where Ca,pro and Cv,pro are the steady state concentrations of propofol in whole blood in the radial artery and renal vein, respectively.

The total body clearance (Cl tot) and renal clearance (Cl r) of propofol were calculated as follows:

Usually, renal clearance of a drug is expressed as the sum of glomerular filtration, tubular secretion, and reabsorption. However, most of propofol is eliminated by renal metabolism, and only 0.3% of propofol is excreted in urine in unchanged form.1Therefore, the renal intrinsic clearance of propofol (Cl int,r) was calculated using a formula similar to that for hepatic intrinsic clearance.

where fB is the unbound fraction of propofol in blood.

Data are expressed as mean ± SD. The differences in propofol concentrations between the radial artery and the renal vein were analyzed by paired t  test. The differences in propofol concentration and renal blood flow before and after removal of the kidney were analyzed by paired t  test. A P  value of less than 0.05 was considered statistically significant.

All patients tolerated the surgery without complications and were clinically improved at the time of discharge from hospital. Ligation of the renal vein for nephrectomy was performed at 2.47 ± 0.34 h (mean ± SD; range, 2.1–3.1 h) after the start of the continuous infusion of propofol.

The propofol concentration in the radial artery (1.43 ± 0.18 μg/ml) was higher than in the renal vein (0.60 ± 0.21 μg/ml) (P < 0.0001; table 1). The renal extraction ratio of propofol was 0.58 ± 0.15.

Total body clearance was 1.53 ± 0.25 l/min. The renal clearance of propofol just before binding the renal artery was 0.41 ± 0.15 l/min, or 27 ± 9.9% of total body clearance. The unbound fraction of propofol in blood was 0.024 ± 0.0084, and renal intrinsic clearance was 73 ± 79 l/min.

Although renal blood flow decreased from 0.74 ± 0.15 l/min to 0.51 ± 0.11 l/min (P < 0.0001) after the ligation of the renal artery for nephrectomy, there was no significant change of propofol concentration in blood from the radial artery (P = 0.19; fig. 1).

The pharmacokinetics of propofol have been studied in healthy subjects18,19and groups of patients with renal20and hepatic insufficiency.21Propofol was extensively distributed and rapidly cleared from the body in each condition. Extrahepatic clearance of propofol has been suggested because total body clearance exceeds hepatic blood flow. We have directly demonstrated the disappearance of propofol in the kidneys.

Glucuronidation is the major metabolic pathway of propofol in humans.1,11Raoof et al.  16reported that human liver, kidney, and small intestine were capable of forming propofol glucuronide. In the current study, we showed that renal clearance of propofol was 0.41 ± 0.15 l/min, or 27 ± 9.9% (mean ± SD) of total body clearance. Therefore, it became clear that the human kidneys play an important role in the elimination of propofol.

Icks et al.  20reported that total body clearance of propofol was similar in individuals with end-stage renal disease and control subjects. They concluded that the kidneys do not contribute significantly to the extrahepatic clearance of propofol. However, a considerable amount of propofol disappeared in the human kidneys. Usually, renal clearance of a drug is expressed as the sum of glomerular filtration, tubular secretion, and reabsorption. However, most of propofol is eliminated as metabolites, and only 0.3% of propofol is found as unchanged form in urine.1Therefore, renal intrinsic clearance of propofol (Cl int,r) could be calculated using a formula similar to that for hepatic intrinsic clearance. In this study, renal intrinsic clearance of propofol was calculated as 73 ± 79 l/min. The product of renal intrinsic clearance and the unbound fraction in blood was 1.48 ± 1.24 and exceeded renal blood flow (0.74 ± 0.15 l/min). It might be considered that renal intrinsic clearance is so high that the renal clearance of propofol depends on renal blood flow and is unaffected by renal failure.

After nephrectomy, renal blood flow decreased by one third. If 27% of total body clearance were renal, total body clearance was expected to reduce by 9% after nephrectomy. Such a small difference in clearance rate is difficult to detect. We found no significant difference in total body clearance of propofol before and after nephrectomy. Although the kidneys are an important site of propofol elimination, our results suggest that total body clearance is not appreciably influenced by nephrectomy. Therefore, propofol can be used safely for patients undergoing nephrectomy.

In this study, renal extraction ratio and total body clearance of propofol was calculated at the pseudo–steady state—at least 2 h after the start of continuous infusion. We cannot entirely rule out the possibility that the apparent clearance of propofol in the kidneys results from a distribution of the drug into tissues, which would tend to overestimate total body clearance. True steady state was not established during this study because the elimination half-life of propofol is quite long (4–6 h). However, given that the rapid distribution half-life is short (1–3 min) and the distribution clearance is large,22,23the concentration of propofol reaches pseudo–steady state at 20 min after a constant infusion and increases slowly until true steady state is reached.9The contribution of rapid distribution, second elimination (30–50 min), and terminal elimination half-life (4–6 h) to the changes in concentration are 94.6, 4.9, and 0.57%, respectively.24There is a slight contribution of terminal half-life to the increase in concentration of propofol, but this is probably clinically irrelevant. Therefore, the pseudo–steady state of propofol is almost equal to the true steady state. Propofol concentrations were reasonably stable (fig. 1). Furthermore, the distribution process of propofol into well-perfused organ tissue, such as the lungs, reaches equilibrium within a short time of constant infusion.9In contrast, less well-perfused organs, such as fat, approach steady state conditions quite slowly. Therefore, we believe that renal extraction of propofol at pseudo–steady state does not reflect distribution.

In conclusion, human kidneys play an important role in the elimination of propofol. The renal intrinsic clearance is so high that the renal clearance of propofol depends on renal blood flow. Total body clearance of propofol is not significantly influenced by nephrectomy.