Pulmonary Disposition of Propofol in Surgical Patients

This study was approved by the Institutional Review Board of Osaka University Medical School, and written informed consent was obtained from all patients enrolled in the study. The characteristics of patients are shown in table 1.

Nine patients scheduled for elective surgery with general anesthesia were enrolled in the study. After the induction of general anesthesia with intravenous administration of thiopental (250 mg) and vecuronium (8 mg), central venous and radial artery catheters were inserted. General anesthesia was maintained with sevoflurane (0.5–1.0%). A finger probe, which is connected to the integrated pulse-spectrophotometry monitoring system, was fixed on the tip of forefinger to detect the blood concentrations of indocyanine green (ICG) based on pulse spectrophotometry. 17–19Before propofol administration, 0.5 ml of blood was drawn into a heparinized syringe to measure the hemoglobin concentration that is necessary for calculating the ICG blood concentration. After baseline measurement, propofol (5 mg; Diprivan, Zeneca Pharmaceuticals, Osaka, Japan) and ICG (15 mg; Diagnogreen, Daiichi Pharmaceutical Company, Tokyo, Japan) were simultaneously administered via  the central venous catheter within 1 s. Sequential arterial blood samples were collected from the radial artery at 1-s intervals up to 45 s and at 5-s intervals up to 1 min by directing the spontaneous outflow of the arterial catheter into heparinized tubes. Blood samples were immediately centrifuged at 15,000 rpm for 5 min at 4°C, and the plasma samples were stored at 4°C until subsequent analysis within 48 h.

The cardiac output was estimated from the area under the first dilution curves (AUC) of ICG:

The concentration data of propofol and ICG were transformed to the unit mass flow, termed frequency output 20:

where C(t) is the concentration of either propofol or ICG at time t, Q is equal to the cardiac output, and f(t) is the unit response of the system in terms of the fraction of injected dose per unit time. Plotted in such a manner, the AUC from time zero to the end of the first pass is equal to the availability of an injected dose through the lungs on the first passage (F^pul) 21,22:

where t is the end of the first pass determined by the last point of the linear portion on a semilogarithmic plot of the descending part of the ICG dilution curve.

The pulmonary extraction ratio (E^pul) was calculated from F^pulas follows 22:

The mean pulmonary transit time (MTT^pul) was calculated by:

The apparent distribution volume (V^app) between the injection site and the sampling site can be estimated from 23,24:

For ICG, V^appis the intravascular volume between the site of injection (the central venous catheter) and the site of sampling (the tip of a forefinger). The difference between the V^appof propofol and ICG is the pulmonary tissue volume of propofol.

For comparative purposes, the conventional instantaneous extraction ratio at each time point was calculated by the following formula 11: where F(t) represents the fractions of injected dose of each drug per milliliter at time t, respectively.

Eleven patients undergoing coronary artery bypass graft surgery were included in this study. After the induction of anesthesia with fentanyl (0.2–0.4 mg), midazolam (5–10 mg) and vecuronium (8 mg), jugular venous, radial arterial, and pulmonary arterial catheters were inserted. Anesthesia was maintained with sevoflurane (1–2%) in oxygen. Propofol was administered via  an intravenous cannula placed in the right internal jugular vein by an infusion pump (Graseby 3500, Graseby Medical Limited Inc., Hartford, United Kingdom) at a rate of 50 μg · kg⁻¹· min⁻¹. Blood samples were simultaneously collected from pulmonary and radial arteries at 0, 2, 3.5, 5, 7, 10, 12, 15, and 20 min and at 10-min intervals up to 60 min. Blood samples were immediately centrifuged, and the plasma samples were stored at 4°C pending assay. The AUCs for propofol plasma concentration in pulmonary and radial arteries were calculated based on the trapezoidal rule.

Several groups have reported that no difference was detected between whole blood and plasma concentrations of propofol 13,25–27; thus, the plasma concentration of propofol was measured in the present study. Plasma propofol concentration was measured using modified high-performance liquid chromatography method reported previously with fluorescence detection. 28The high-performance liquid chromatography (LC-10AD; Shimadzu, Kyoto, Japan) is composed of a variable wavelength ultraviolet detector (SPD-10AD, Shimadzu) and fluorimetric detector (RF-10AD, Shimadzu), degasser (DGU-14A, Shimadzu), autoinjector (SIL-10AD, Shimadzu), and column oven (CTO-10AS, Shimadzu). Briefly, a Symmetry C18 column (3.5 mm, 4.6 × 100 mm; Waters, Osaka, Japan) was equilibrated at 1.5 ml/min with the mobile phase containing acetonitrile and water (60:40, vol). The excitation and emission wavelengths for fluoremetric detection of propofol were set at 276 and 310 nm, respectively. The limit of sensitivity of the assay was 0.05 μg/ml in plasma. The intraday and interday coefficients of variation at the concentration of 1.0 μg/ml were < 5%.

Blood ICG concentration was monitored noninvasively with pulse spectrophotometry. 18,19The integrated pulse-spectrophotometry monitoring system is composed of a finger probe, a monitoring device, and a computer for recording and printing the results (DDG1001, Nihon Kohden Inc., Tokyo, Japan). The blood ICG concentration was measured at every pulse interval, which was less than 1 s, according to the patient’s heart rate. At the end of each study, hemoglobin concentration was entered into the computer, and the final ICG blood concentration–time course was printed out automatically. The data files of ICG blood concentration–time course were downloaded for further data analysis. The blood concentrations of hemoglobin were measured with a radiometer (ABL 620, Radiometer, Copenhagen, Denmark) immediately after sampling.

Data are expressed as mean values ± SD. Differences in parameters between ICG and propofol were analyzed with a paired t  test. Differences in propofol concentrations between pulmonary and radial arteries at each time point were analyzed with the one-way repeated-measures analysis of variance. If the analysis of variance was found to be significant, Bonferroni correction was performed to compare the difference in propofol concentrations between pulmonary and radial arteries. A P  value < 0.05 was considered statistically significant.

Figure 1shows the representative frequency outflow curves of propofol and ICG during the first pass through the lungs. Plotted in such a manner, the AUC from time zero to the end of the first pass is equal to the availability of an injected dose. A pronounced difference between the shape of the dilution curve of propofol and that of ICG was observed. ICG showed a typical shape of intravascular space with a sharp peak, whereas that of propofol showed a much lower peak and displayed slowly tailing, suggesting a reversible distribution of propofol into the lungs. The cardiac output estimated based on the ICG dilution curve was 4.91 ± 1.50 l/min for the patients participating in the first-pass study. The pulmonary uptake of propofol during the first pass through the human lung was estimated to be 28.4 ± 11.6% based on equation 4. The mean pulmonary transit time of propofol (31.3 ± 6.0 s) was significantly longer than that of ICG (22.4 ± 2.7 s;P < 0.01), indicating that some of the propofol trapped by lungs during the first pass returned to the circulation by back diffusion. The apparent distribution volume for propofol estimated based on equation 6was 2.52 ± 0.79 l, which showed a significant larger value than that of ICG (1.67 ± 0.37 l;P < 0.001). The pulmonary tissue volume of propofol estimated from the difference between the apparent distribution volume for propofol and that for ICG was 0.866 ± 0.448 l, demonstrating an extensive distribution of propofol into the lung tissue. The instantaneous extraction ratios of propofol at each sampling time were calculated based on equation 7. The initial extraction of propofol was as high as 99% and then decreased to 18% at the end of first pass through the lungs.

Figure 2illustrates the plasma concentrations of propofol in pulmonary and radial arteries during a constant infusion at 50 μg · kg⁻¹· min⁻¹. The plasma concentrations of propofol in both pulmonary and radial arteries increased progressively and reached pseudo–steady state at 20 min. Peak plasma concentrations were observed at 60 min (1.14 ± 0.27 mg/ml and 1.16 ± 0.28 mg/ml). No significant differences were observed with the plasma concentrations of propofol between pulmonary and radial arteries except for that at 2 min. The significant higher plasma concentration of propofol in pulmonary artery than that in radial artery at 2 min (0.37 ± 0.26 μg/ml vs.  0.29 ± 0.23 μg/ml;P < 0.05) might be attributed to the pulmonary uptake of propofol at the initial phase or lag time between the two sampling sites. The AUC of pulmonary and radial arterial concentration curves to 60 min were 59.1 ± 14.8 μg · ml⁻¹· min⁻¹and 56.8 ± 12.5 μg · ml⁻¹· min⁻¹, respectively. No significant difference was observed with the AUC between the pulmonary and radial arterial plasma concentration curves.

The first-pass pulmonary uptake of fentanyl, meperidine, morphine, diazepam, and thiopental have been studied in humans. 8–11These studies suggest that the extent of pulmonary uptake of drugs is related to physicochemical properties such as a basic amine moiety and lipid solubility and to physiological factors such as plasma protein binding and cardiac output. The first-pass uptake of a drug through the lungs has been calculated at the time when 95% of the injected ICG has passed through the lungs in the anesthetic field. 8–11,13,14In the present study, the first pass uptake of propofol and ICG were separately calculated based on their frequency output curves. The concept of “frequency output” was proposed to interpret the dilution curves in rat liver by Roberts and Rowland in 1986. 21The advantage of this expression is that the distribution properties of an indicator can be envisaged directly from the shape of the normalized frequency output curves. In addition, the availability of a drug through an organ can be directly estimated from the integral of the frequency output dilution curve in an isolated perfused organ preparation where the perfusion flow is under control.

In this study, the first-pass uptake of propofol through human lungs was estimated to be 28.4 ± 11.6%. The pulmonary extraction of propofol during the first pass through human lungs (28.4 ± 11.6%) was higher than that of thiopental (13.8 ± 1.8%), 10similar to that of diazepam (30.4 ± 3.7%), 10and lower than that of fentanyl (75.2 ± 3.2%) 8and meperidine (64.5 ± 7.8%). 8The pulmonary uptake of diazepam has been shown to be dependent on the albumin concentrations in perfusate in rat lung, where the binding of diazepam to plasma albumin limited its availability for pulmonary uptake. 29Roerig et al.  10,11suggested that such a mechanism may also be possible for neutral and anionic lipophilic drugs such as thiopental, i.e. , binding to plasma albumin limits their partitioning into pulmonary tissue. Propofol is a weakly acidic compound that is extensively bound to the plasma proteins (97–99%). 30–32Drugs of this type are generally believed to bind to albumin in plasma. Irrespective of the high lipophilicity of propofol, the binding to plasma protein may limit its availability for uptake into lung tissue.

Kuipers et al.  16recently demonstrated that 30% of the propofol was eliminated during the first pass through the lungs in awake sheep, although they acknowledged that pulmonary clearance or distribution into a deep-tissue compartment could not be discerned. It is worth noting that the first-pass pulmonary uptake of propofol in awake sheep is similar to that in human lungs (28.4%) reported in the present study. On the other hand, a substantial removal of propofol by the cat lung was demonstrated (61%), which was significantly decreased to 39% and 32% in cats exposed to 1.5% halothane and pretreated with fentanyl (1 mg/kg) 30 s before propofol administration, respectively. 13,14In our human study, sevoflurane (0.5–1.0%) used for maintaining the anesthesia might have some effects on the pulmonary uptake of propofol. However, considering that 1.0% halothane (1.33 minimum alveolar concentration) did not influence the pulmonary uptake of propofol in cats (60%), a significant effect by 0.49 minimum alveolar concentration sevoflurane (1%) on the pulmonary uptake of propofol does not seem to be plausible. A species difference would also have to be kept in mind because the enzyme involved in the glucuronidation of propofol is uridine diphosphate–glucuronosyltrasferase, and a notable species difference in this enzyme was indicated. 33–35 

In this study, we demonstrated that propofol was significantly taken up during the first pass through the human lung. A significant first-pass uptake in the lungs can be a result of either metabolism or distribution into lung tissue. Although extensive pulmonary uptake of propofol in cats, rats, and sheep has been reported, 13–16none of these studies can discriminate between pulmonary elimination and distribution into lung tissue. In addition to the first-pass study, we measured both the radial and pulmonary arterial propofol concentrations during constant-rate infusion. In contrast to the first-pass study, pulmonary extraction at steady state should reflect the metabolism rather than distribution. By combining the results from the first-pass study and that from the constant-infusion study, we were capable of distinguishing between pulmonary metabolism and pulmonary tissue distribution for the first time. In the first-pass study, the significantly longer pulmonary mean transit time of propofol than that of ICG and the decreasing instantaneous extraction ratio indicate the back diffusion of the extracted propofol during the first pass out of the lungs into blood. In the infusion study, no significant differences between the propofol concentrations in pulmonary and radial arteries were observed after 2 min, suggesting that the distribution process of propofol into lung tissue reached equilibrium within a short time of infusion. Pulmonary extraction at the steady state is likely to result from metabolism rather than distribution. The AUC to 60 min under the pulmonary arterial plasma concentration of propofol was not significantly different from that under the arterial plasma concentration curve, suggesting that the initial uptake of propofol into human lungs involves temporary sequestration, after which the propofol is released back into the circulation, probably unmetabolized.

Therefore, we conclude that it is the pulmonary distribution of propofol that resulted in a significant amount of pulmonary uptake by the human lungs during the first pass, and the lungs do not appear to contribute to the extrahepatic metabolism of propofol in the human. This is consistent with and supported by an investigation by Guellec et al. , 34who demonstrated a lack of significant metabolism of propofol in human lung tissue. They also investigated the glucuronidation of propofol in microsomal fractions from various tissues and species, including humans, and found no activity in human lung microsomes. 34They concluded that kidney rather than lung might play an important role in the extrahepatic metabolism of propofol in humans. Consistent with their speculation, an efficient ability of human kidney microsomes to form propofol glucuronide has been reported. 36Therefore, it is reasonable to speculate that the significant pulmonary uptake of propofol during the first pass is the result of pulmonary distribution rather than metabolism.

In summary, a significant amount of propofol was taken up during the first pass through the human lung. Most of the propofol that undergoes pulmonary uptake on the first pass was released back into the circulation by back diffusion when the blood concentrations decreased. No pulmonary extraction of propofol was observed at the pseudo–steady state, suggesting that metabolism was not involved in the pulmonary uptake in human lungs. Therefore, the lungs do not seem to be a site contributing to the extrahepatic metabolism of propofol in humans.