Propofol is a short-acting drug with a large volume of distribution and a high total body clearance . Monitoring of propofol concentrations is useful for pharmacokinetic and pharmacodynamic purposes; the assay conventionally uses high-performance liquid chromatography (HPLC) [2,3]. Most methods are time consuming and require more than 1 mL of blood for each analysis; therefore, a better method might be beneficial for routine clinical monitoring. Recently, gas chromatography (GC) was developed for measurement of propofol concentrations in plasma . This method requires only 200 micro Liter of sample and takes no more than 10 min. However, it has not been used to determine whole blood concentrations and has not been compared with conventional HPLC methods. Considering the possible influence of the heating process in the GC, the first objective of our study was to compare both methods in different specimens.
Although several HPLC methods have been reported for monitoring concentrations in plasma or whole blood, less attention has been given to the differences in these two specimens. The presence of propofol metabolites in urine during liver transplantation suggests that extrahepatic metabolism occurs [5,6]. It seems doubtful whether rapid intra- and extrahepatic clearance would result in lower plasma concentrations than that of whole blood if there is a lag of redistribution across blood cell membranes. Furthermore, if propofol binding capacity within blood cells is different from that of plasma, propofol concentrations by plasma and whole blood samples should be different after equilibration. In general, plasma concentrations are more directly related to the activity of a drug in the target organ than are whole blood concentrations, but whole blood collection is easier and more convenient for the clinician. The second part of this study was designed to compare and analyze the differences in concentrations between these two specimens.
After institutional approval and informed consent, blood samples were collected from 15 adult patients. Ten patients received a bolus injection and five received continuous infusion of propofol. Patients with blood loss greater than 500 mL intraoperatively, or those with renal, hepatic, or cardiac dysfunction were excluded from the study.
To compare both methods, 10 mL of heparinized blood was collected at 1, 3, 5, 10, 30, 60, 120, and 240 min after the bolus injection of propofol 2 mg/kg for induction. Isoflurane was used to maintain anesthesia. Each blood sample was divided into four aliquots: 1) blood for HPLC analysis; 2) blood for GC analysis; 3) plasma for HPLC analysis, and 4) plasma for GC analysis. To prepare the plasma samples, one-half of the blood was centrifuged immediately; the other half was centrifuged after storage at 4 degrees C for 1 h to allow equilibration between blood cells. For the infusion group, 4 mL of blood was collected at 1, 3, 5, 10, 30, and 60 min after bolus injection of propofol 2 mg/kg followed by propofol infusion 10 mg centered dot kg-1 centered dot h-1. The samples were divided into two parts (whole blood and plasma samples). The plasma samples were prepared as noted previously. All plasma samples were stored at -70 degrees C and whole blood samples at 4 degrees C until analysis within 2 days after collection.
To analyze equilibration with red blood cells with different storage times and at different temperatures, 60 mL of blood were obtained from three healthy volunteers and divided among six tubes. Three tubes were stored at 25 degrees C and the others at 4 degrees C. Propofol (100 ng/mL, 1 micro gram/mL, and 10 micro gram/mL) was added to each tube and shaken gently, continuously. Then 2-mL aliquots of blood were taken at 3, 5, 10, 30, and 60 min after mixing, for concentration measurement and blood gas analysis.
The GC method was performed according to the report of Yu et al. . The apparatus (Model GC-14A; Shimazu, Kyoto, Japan) consisted of a split-splitless capillary inlet system and a flame ionization detector. Whole blood samples were analyzed by the same method with a minor change in the extraction process. Blood samples were spun on a vortex mixer with an equal volume of chloroform (containing an internal standard). After centrifugation, a 200-micro Liter chloroform layer was washed with the same volume of 0.5 N sodium hydroxide solution to remove the interfering substances in GC analysis. The HPLC method reported by Plummer  was applied in the present study. The instrument consisted of an integrator (Model 3396A; Hewlett-Packard Co., Avondale, PA), a fluorescence detector (Model RF-535; Shimazu) and a reverse phase column (Lichrosphere RP-18 100, 4 mm times 125 mm, 5-micro meter particle size; Merck, Darmstadt, Germany).
Data are summarized as mean +/- SD. Propofol concentrations resulting from both assay methods and between different samples were analyzed by linear regression analysis. The former differences were compared by paired Student t-test. Comparison of propofol concentrations in different specimens and with different storing times was analyzed by repeated-measures analysis of variance followed by post hoc Student-Newman-Keuls test. Comparison in groups with different concentrations and storing temperature in vitro study was also analyzed by this method. A value of P <or=to 0.05 was considered significant.
The interassay or intrassay coefficients of variation were smaller than 10% for HPLC (20 ng/mL to 15 micro gram/mL) and GC assay (10 ng/mL to 20 micro gram/mL). Propofol concentrations by both methods correlated well with each other either in whole blood or plasma (HPLCwhole blood = 0.13 + 0.95 times GCwhole blood; r2 = 0.99, P < 0.001; HPLCplasma = 0.03 + 0.98 times GCplasma; r2 = 0.98, P < 0.001).
Although whole blood versus plasma concentrations also revealed significant correlation, significant biphasic differences were observed after bolus injection Table 1. When plasma was separated immediately after collection, the concentrations were 5%-10% higher during the first 10 min, but approximately 10% lower at terminal phase, than whole blood concentrations Figure 1. If the blood samples were stored for 1 h at 4 degrees C before centrifugation, plasma concentrations remained 5%-10% higher than those of whole blood Figure 1 and showed significant differences with plasma concentrations which had immediate centrifugation.
For the infusion group, plasma concentrations from immediate centrifugation were 30% higher than that of whole blood throughout the entire infusion period. It still remained 10% higher than whole blood concentrations even after storing for 1 h before centrifugation Figure 2. In the in vitro study for propofol distribution between red blood cells, plasma concentrations were higher than those of whole blood throughout the experiment both at 4 and 25 degrees C with different concentrations. Propofol in plasma is 20% higher at the first 10 min, then decreased to 10% afterward. No significant differences in percentage of concentration difference between plasma and whole blood samples were found in groups at different concentrations and temperatures Table 2. The pH of blood samples did not change significantly during the storage of blood samples. The simultaneous loss of CO2 in the blood samples and consumption of bicarbonate by metabolism in red blood cells maintained an almost unchanged pH Table 2.
The results showed that the two methods correlated well with each other. The biphasic differences in propofol concentrations for different specimens after bolus injection may influence the calculation of pharmacokinetic variables. The higher plasma concentrations during the first 10 min were due to the lag in propofol distribution into red blood cells. On the other hand, the lower plasma at the terminal phase was caused by slow redistribution of propofol from blood cells to plasma against rapid plasma clearance. If the separating time was delayed, or propofol was added to the blood in vitro study to prevent the intra- or extrahepatic clearance, plasma concentrations were kept higher than in whole blood samples. The almost unchanged pH values in blood samples excluded the influence of pH on the distribution across cell membranes.
Greater differences were observed in the infusion group. Plasma concentrations remained 30% higher than whole blood concentrations during infusion periods. The infusion of propofol seemed to surpass the rate of clearance. If whole blood instead of plasma concentrations were used during propofol infusion, the concentration could be underestimated. Even after storing for 1 h before centrifugation, the plasma was still 10% higher than whole blood concentrations, but closer. Evidence from the in vitro study showed the same degree of discrepancy in different storing temperatures and duration to separate the plasma samples. It excluded the possibilities that drug metabolism within red blood cells was responsible for the discrepancy.
From a pharmacokinetic point of view, the equilibrium of drug concentrations between tissue and plasma must be more direct than that between tissue and whole blood when the equilibrium between plasma and blood cells is not instant. Therefore, plasma concentrations should be more informative than blood concentrations for propofol. In conclusion, although using less blood and simpler procedures, GC can establish as accurate an assay as HPLC for measurement of propofol concentrations. During propofol infusion, plasma concentration monitoring is preferred, but immediate centrifugation is needed to prevent underestimation of propofol concentrations. When a bolus injection is given, a biphasic difference is expected and may influence the precise calculation of pharmacokinetic data.
Technical assistance was provided by Mithra Bioindustry Co. The authors wish to thank Shih-hsun Ngai, MD, for his revision of the manuscript and Jiun-Kai Liau, MSc, for his help throughout this study.
1. Sebel PS, Lowdon JD. Propofol: a new intravenous anesthetic. Anesthesiology 1989;71:260-77.
2. Adam HK, Douglas EJ, Plummer GF, Cosgrove MB. Estimation of ICI 35,868 (Diprivan Registered Trademark) in blood by high-performance liquid chromatography, following coupling with Gibbs' reagent. J Chromatogr 1981;223:232-7.
3. Plummer GF. Improved method for the determination of propofol in blood by high-performance liquid chromatography with fluorescence detection. J Chromatogr 1987;421:171-6.
4. Yu HY, Liau JK. Quantitation of propofol in plasma by capillary gas chromatography. J Chromatogr 1993;615:77-81.
5. Veroli P, O'Kelly B, Bertrand F, et al. Extrahepatic metabolism of propofol in man during the anhepatic phase of orthotopic liver transplantation. Br J Anaesth 1992;68:183-6.
6. Mather LE, Selby DG, Runciman WB, McLean CF. Propofol, assay and regional mass balance in the sheep. Xenobiotica 1989;19:1337-47.