British Journal of Pharmacology (1995) 115, 503 509 - B 1995 Stockton Press All rights reserved 0007-1188/95 $12.00 M The use of microdialysis for the study of drug kinetics: some methodological considerations illustrated with antipyrine in rat frontal cortex 1P.N. Patsalos, 2W.T. Abed, M.S. Alavijeh & M.T. O'Connell Pharmacology and Therapeutics Unit, Epilepsy Research Group, University Department of Clinical Neurology, Institute of Neurology, Queen Square, London, WC1N 3BG 1 The neuropharmacokinetics of antipyrine, a readily dialysable drug, in rat frontal cortex were studied and the effect of sampling time and contribution of period sampling and dialysate dead volume investigated in relation to t,, Cm,,n, AUC and t1/2 values. 2 After i.p. administration, antipyrine (35 mg kg-, n = 5) concentrations rose rapidly in rat frontal cortex (tma,,, 12 min) and then declined exponentially. tmax, Cmax, AUC and t1/2 values were determined after 2 min dialysate sampling and compared to values obtained from simulated sampling times of 4, 6, 8, 10 and 20 min. 3 Antipyrine tmn and Cm,, values were directly dependent on sampling frequency. Thus, mean 2 min sampling tmnX and Cs,,, values were 63% lower and 27% higher, respectively, compared to 20 min sampling values. AUC and t1/2 values were unaffected. 4 Adjustment for dialysate dead volume (the volume of dialysate within the dialysis probe and sampling tube) reduced tmn. values significantly but did not affect the other neuropharmacokinetic parameters. 5 Contribution of period sampling on neuropharmacokinetic parameters were investigated by comparing plots of antipyrine concentration data at midpoint and at endpoint of sampling time interval. Only tmax values were affected with values decreasing with increasing sampling time interval. 6 In conclusion, although microdialysis is a useful method for monitoring events at the extracellular level and for kinetic studies, it is important to understand its inherent characteristics so that data can be interpreted appropriately. Sampling frequency, particularly during monitoring of periods of rapid change, is very important since Cm,,, and tmn,, values will be significantly underestimated and overestimated respectively, if sampling time is longer rather than shorter. These considerations are particularly important in relation to microdialysis studies of pharmacokinetic-pharmacodynamic interrelationships and modelling. Keywords: Microdialysis; antipyrine; neuropharmacokinetics; period analysis; pharmacodynamic modelling Introduction During the last 10-15 years the technique of intracerebral microdialysis has contributed significantly to our understanding of brain neurochemistry (Lehmann et al., 1983; Abercombie et al., 1988; Benveniste et al., 1989; Yamamoto & Davy, 1992; Millan et al., 1993; Yadid et al, 1993). By use of intracerebral microdialysis it has been possible to monitor the release of various neurotransmitters into the extracellular fluid of specific brain sites and, at the same time, to manipulate pharmacologically the dialysed areas under physiological conditions (Hernandez et al., 1991; Mizuno et al., 1991; Bustos et al., 1992). As the extracellular fluid is the liquid compartment where the traffic of compounds and the exchange of chemical information between cells takes place, monitoring of the fluid by microdialysis has been considered to reflect synaptic events and of events at the site of action or biophase of drugs. Consequently, neurochemical changes in the extracellular fluid have been correlated with various pharmacodynamic paradigms including behaviour (Abercombie et al., 1988; Hutson & Curzon, 1989; Stahle et al., 1990; Mizuno et al., 1991) and electrical activity (Delgado et al., 1984; Vezzani et al., 1985; Ludvig et al., 1992). Also numerous studies of the effects of a variety of drugs on neurochemical changes have been undertaken (Hurd et al., 1988; O'Connell et al., 1991; Semba et al., 1 Author 2 for correspondence. Present address: Jordan University for Women Faculty of Pharmacy & Medical Technology, PO Box 961343 Amman, Jordan. 1992). These studies have now been extended to man (Meyerson et al., 1990; Carlson et al., 1992; Persson & Hillered, 1992; During & Spencer, 1993). The use of the microdialysis technique to study peripheral pharmacokinetics and CNS neuropharmacokinetics of drugs has recently been advocated and numerous reports relating to studies in animals (Hurd et al., 1988; Sabol & Freed, 1988; Stahle et al., 1991b; Steele et al., 1991; Telting-Diaz et al., 1992; Alavijeh et al., 1993; Deleu et al, 1993; Dykstra et al., 1993; Golden et al., 1993; Kurata et al., 1993; O'Connell et al., 1993; Patsalos et al., 1993) and man (O'Connell et al., 1994; Scheyer et al., 1994a,b; Stahle et al., 1991a) have been published. Also published are a number of studies on methodological considerations, particularly the mathematical handling of microdialysis drug concentration data and the correlation between pharmacokinetic data derived from microdialysis and that derived from direct blood sampling (Scott et al., 1990; Morrison et al., 1991; Stahle, 1992; Dykstra et al., 1993; Stenken et al., 1993). Of additional concern has been the recovery characteristics of microdialysis probes and methodologies have been described to enable absolute concentrations to be determined. However, although methods are straightforward in relation to the blood compartment, they are rather tedious and cumbersome in relation to the brain (Scheller & Kolb, 1991; Menacherry et al., 1992; Sjoberg et al., 1992; Yokel et al., 1992; Justice, 1993; Stenken et al., 1993). A fundamental omission of all kinetic studies (both pharmacokinetic and neuropharmacokinetic) to date has been consideration of the effect of sampling rate and sample 504 P.N. Patsalos et al batching on kinetic profiles. Indeed the effect of sampling rate on neurotransmitter and other analytical profiles in general, using the microdialysis technique, has not been studied. This is particularly relevant since microdialysis sampling involves period sampling and not point sampling. In the present study we have investigated the effect of sampling time on the neuropharmacokinetics of antipyrine in rat frontal cortex. Also investigated was the contribution of the sampling methodology to the calculated neuropharmacokinetic constants of antipyrine. Antipyrine was selected because it distributes rapidly throughout body water, cerebrospinal fluid and cerebral cortex (Johanson & Woodbury, 1977), is unionised at physiological pH (Rall et al., 1959), is minimally bound to plasma proteins (Gumbleton & Benet, 1991) and has a low hepatic extraction ratio (Rane et al., 1977), suggesting that extracellular fluid antipyrine concentrations should not differ significantly between different organs. In addition, antipyrine is readily dialysable (Yokel et al., 1992; Kurata et al., 1993). Microdialysis and antipyrine neuropharmacokineics detector and a Chromjet integrator. A Lichrosorb RP8 10 gm column (Hichrom, Reading) and an acetonitrile: water: acetic acid (40: 48: 12) mobile phase (1.3 ml min-) were used. The column eluent was monitored at 270 nm and the retention time for antipyrine was 6.7 min. The lower limit of detection of antipyrine was 0.5 ymol 1-l and the inter-assay coefficient of variation was <2.5%. Neuropharmacokinetic analysis Five male Sprague-Dawley rats (B&K Universal Ltd, Hull) weighing 250-300 g with free access to a normal laboratory diet (SDS R and M number 1 expanded. Scientific Dietary Services, Witham, Essex) and water were used. Rats were individually housed in contiguous cages at constant temperature (220C) under a 12 h light-dark cycle (lights on 06 h 00 min). Using the non-linear least-square regression programme PCNONLIN 3.0, dialysate antipyrine concentration versus time data were analysed with a one compartment model and a weighting factor of 1/concentration (Metzler & Weiner, 1988). However, as the programme (and indeed most such programmes) requires initial parameter estimates, and the success of the least-square analysis often depends on how good the initial parameter estimates are, initial parameter estimates were obtained using JANA (Dunn, 1985). Both programmes were implemented on a Hewlett Packard Vectra 486 computer. The parameters computed were: area under the concentration versus time curve (AUC0_356 min) and elimination half-life (t1/2). As antipyrine was administered i.p. and not as a bolus (e.g. i.v.), the t1/2 values computed can be considered as apparent t1/2 values since they encompass both absorption and elimination processes. Time to maximum concentration (tmea) and maximum concentration (Cm,,) were obtained by visual inspection of the antipyrine versus time profiles. Intracerebral microdialysis Data and statistical analysis All animal procedures were in strict accordance with the Home Office guidelines and specifically licenced under the Animals (Scientific Procedures) Act of 1986. Anaesthesia was induced by sodium pentobarbitone (60 mg kg-', i.p.) and dialysis probes stereotactically implanted in the frontal cortex (3.2 mm anterior and 5.5 mm ventral to bregna, 2.5 mm lateral to midline) according to the atlas of Paxinos & Watson (1986). The dialysis probes were concentric in design and were the same as those previously described (Semba et al., 1992). The following day, the inflow line of each probe was connected to a syringe pump (model 22, Harvard Apparatus Ltd, Edenbridge) and artificial cerebrospinal fluid (composition mM: NaCl 125, KCl 2.5, MgCl2 1.18 and CaC12 1.26) was perfused through the probes at a flow rate of 2 yl min- . The outlet line of each probe was connected to a refrigerated fraction collector (CMA/170, Biotech Instruments Ltd, Luton) so as to enable automated dialysate collection. After an initial 30 min perfusion interval, animals were injected intraperitoneally with antipyrine (35 mg kg-; BDH Ltd, Poole) and sampling continued at 2 min intervals for 5.93 h. Dialysates were stored at -70'C until analysis of antipyrine content. Antipyrine relative recovery in vitro, for each microdialysis probe, was determined by a traditional in vitro recovery study at 370C. The five microdialysis probes used for implantation were first tested by immersing in a beaker of constantly stirred dialysate containing 80 ymol l-1 antypirine. After perfusing with dialysate at 2 M1 min- for 10 min, three 20 min dialysate samples were collected from each probe. These were analysed for antipyrine content and compared to known dialysate antipyrine concentration (80 yumol l-l in beaker dialysate) so as to quantitate recovery. In order to simulate dialysate sampling at 4, 6, 8, 10 and 20 min intervals, the antipyrine concentration data were summated accordingly. Thus, in order to simulate 4 min sampling, the first two 2 min antipyrine concentrations were averaged and also all subsequent two 2 min data sets were averaged. Similarly, in order to simulate 20 min sampling, the first ten 2 min antipyrine concentrations were averaged and all subsequent ten 2 min data sets were averaged. Since microdialysis sampling represents period sampling (e.g. 2 min or 10 min sampling) and consequently measured dialysate concentrations represent mean values during the sampling time period, the effect of plotting the antipyrine concentration versus time data at midpoint of sampling time (e.g. during 2 min sampling data are plotted at 1 min) as compared to plotting at endpoint of sampling period, on kinetic constants was investigated. These data are referred to as time-adjusted data. Finally, the contribution of a non-synchronized sampling protocol in relation to antipyrine administration on kinetic constants was investigated by adjusting antipyrine concentration versus time data to compensate for dialysate dead volume. The dialysate dead volume being the volume of dialysate within the dialysis probe and the tubing leading to the sampling tubes. These data are referred to as dead-volume adjusted data. Statistical analysis of the neuropharmacokinetic parameters were made by t test and paired t test. Methods Antipyrine analysis Antipyrine concentrations in the dialysis samples (4 jA) were determined by high performance liquid chromatography (h.p.l.c.) essentially as previously described (Patsalos et al., 1988). Briefly, 4 MI of dialysate were injected directly into a Spectra Physics liquid chromatogram with data processing capability (Spectra Physics, Maidenhead). The chromatogram comprised a P4500 pump, an A53000 autosampler, a UV2000 Results Mean i s.e.mean antipyrine relative recovery in vitro of the five probes was 23.3 i 1.5% at a dialysate flow rate of microdialysis 2 M1 min . Because of debate as to the justification of translating in vitro recovery determinations into those expected in vivo, the concentration data shown in this study are those measured in vivo and are not adjusted to take into consideration in vitro relative recovery. This does not affect data interpretation in any way. Figure 1 shows the mean frontal cortex dialysate concentration versus time profile of antipyrine after the in- traperitoneal administration of antipyrine (35 mg kg-, n = 5). P.N. Patsalos et al After a time lag of 16 min, antipyrine concentrations rapidly increased to peak 12 min later. Concentrations then fell exponentially. Figure 2a shows a comparison of frontal cortex antipyrine concentration versus time profiles for 2 min sampling and simulated 20 min sampling intervals in a typical rat and - Z E =L 0 r_ cu 0 4) 0 0 r_ 0. C 30 60 90 120 150 180 210 240 270 300 330 505 Microdialysis and aantipyrine neuropharmacokineffcs Figure 2b shows that of rat 5. The profiles are only shown up to 122 min since the terminal concentration versus time points are essentially superimposable and not dependent on sampling time interval. For clarity the simulated concentration versus time profiles for 4, 6, 8 and 10 min sampling are not included but are intermediate of the profiles for 2 and 20 min sampling. Figures 3 - 5 show the same profiles as for Figure 2a except that the data have been plotted at midpoint of sampling time interval to compensate for period sampling (Figure 3), adjusted to compensate for dialysate dead volume (Figure 4) and finally a combination of dead volume adjusted and plotted at midpoint of sampling time interval (Figure 5). Table 1 shows the apparent neuropharmacokinetic constants for individual rats together with the mean values as calculated from log concentration versus time plots. The microdialysate concentration data have not been adjusted to take into consideration microdialysate dead volume or period sampling analysis. The neuropharmacokinetic constants for individual rats showed moderate variability within each sampling time interval. Although sampling time interval did not affect AUC and t1/2 values, tm,, and Cm,,,a values were significantly affected. Thus, compared to 2 min sampling, tn.,, and C,,, values increased and decreased respectively with increase in sampling time interval. These differences were sig- lime (min) Figure 1 Antipyrine frontal cortex dialysate concentration versus time profile after i.p administration of 35mgkg-1 antipyrine. Values are mean of 5 rats. For clarity, s.e.bars are not shown but do not exceed 5%. 120 % L. -aE 100 q .2 80 a 120 r- 100 C 60 0 0 80 40 0) 0. >-2 60 _40 Time (min) Figure 3 A comparison of frontal cortex antipyrine concentration versus time profiles for 2min sampling (0) and simulated 20min sampling (0) time intervals where concentration values are plotted at midpoint of sampling time interval. Values are mean of 5 rats. For clarity, s.e.bars are not shown but do not exceed 5%. or 0 c 120_ +._ _ Cu c0. 100 _ 120 c * 80 E 100 60 .2 80C 0 40 -20 0 30 60 90 120 CL Time (min) Figure 2 (a) A comparison of frontal cortex antipyrine concentration versus time profiles for 2 min sampling (0) and simulated 20 min sampling (0) time intervals. Values are mean of 5 rats. For clarity, s.e.bars are not shown but do not exceed 5%. (b) A comparison of frontal cortex antipyrine concentration versus time profiles for 2min sampling (0) and simulated 20min sampling (0) time intervals for rat 5. The figure illustrates the largest difference in Cmax values seen in the study. For clarity, s.e.bars are not shown but do not exceed 5%. 0 20 40 60 80 100 120 Time (min) Figure 4 A comparison of frontal cortex antipyrine concentration versus time profiles for 2min sampling (0) and simulated 20min sampling (0) time intervals where concentration values are plotted so as to compensate for dialysate dead volume. Values are mean of 5 rats. For clarity, s.e.bars are not shown but do not exceed 5%. P.N. Patsalos et al 506 Microdialysis and andpyrine neuropharmacokineics nificantly different at all sampling time intervals for Cmax values and from 6 min for tmo values. Twenty minute sampling was therefore associated with a mean 63% increase in tang values and a mean 27% reduction in Cm. values compared to the corresponding 2 min values. However, for rat 5, Cmax values were 67% lower during 20 min sampling compared to 2 min sampling (Figure 2a). Table 2 shows a comparison of mean microdialysate neuropharmacokinetic constants for frontal cortex antipyrine before and after adjustment for microdialysis sampling methodology at 2, 4, 6, 8, 10 and 20 min sampling intervals. Within the 2 min sampling time interval, it can be seen that compared to unadjusted data, time adjusted data (i.e. plotting individual concentrations at midpoint of sampling time interval to compensate for period sampling) had no significant effect on any neuropharmacokinetic parameter studied. However, tma, values decreased, compared to unadjusted data, with increasing sampling time interval so that the reduction was 23% for 20 min sampling. However, this did not attain statistical significance. In contrast, adjustment for dialysate dead volume was associated with significant decreases in tmax values (Table 2). The decreases were particularly significant for the shorter sampling times with a decrease of approximately 50% observed for 2, 4, 6 and 8 min sampling and 31% and 18% for 10 and 20 min sampling respectively. However, adjustment for dialysate dead volume had no significant effect on C,a,, AUC and tl/2 values. Furthermore, although the combination of time adjustment and dead volume adjustment had no additional significant effect on the neuropharmacokinetic constants compared to dead volume adjustment only, tman values showed a tendency towards reduction with increasing sampling time. 120- E 100 - .2 80 60 60 40 -20 CL 0 20 40 60 80 100 120 Time (min) Figure 5 A comparison of frontal cortex antipyrine concentration versus time profiles for 2min sampling (0) and simulated 20min sampling (0) time intervals where concentration values are plotted at midpoint of sampling time interval and also compensated for dialysate dead volume. Values are mean of 5 rats. For clarity, s.e.bars are not shown but do not exceed 5%. Discussion With the increasing use of the microdialysis technique in in vivo monitoring of a wide range of analytes including neurotransmitters and drugs, it is important to characterize the Table 1 Antipyrine neuropharmacokinetic constants derived from 2 min microdialysate sampling and from projected sampling times up to 20min 2 Rat 1 2 3 4 5 Mean s.e.mean t,,. (min) 26 24 24 24 36 27b 2 Microdialysate sampling time (min) 4 6 8 28 28 28 24 40 30 2 10 20 30 30 30 30 40 32a 2 40 40 40 40 60 30 30 30 30 42 32 32 32 32 40 32a 34a 2 1 111.5 87.1 90.8 82.8 74.9 107.3 88.1 82.8 85.0 61.6 101.9 82.3 78.0 79.6 36.1 85.0c 75.6c 6.1 108.8 85.5 87.6 82.3 77.0 88.1c 5.3 7.3 10.8 174.2 166.2 151.9 148.1 67.4 141.6 19.1 174.2 165.7 151.3 148.1 67.4 141.3 19.1 173.6 165.1 151.3 147.6 67.4 141.0 19.0 172.0 163.0 149.2 146.5 66.9 139.5 18.7 86.8 131.1 105.0 97.9 82.4 100.6 7.7 86.8 130.9 105.0 98.0 79.4 100.0 7.9 86.6 128.8 105.3 96.9 80.6 99.6 7.5 86.9 129.0 105.0 96.8 87.8 101.1 6.9 44a 4 C,,.,x (mol 1-1) 2 3 4 5 Mean ± s.e.mean 2 3 4 5 Mean ± s.e.mean 2 3 4 5 Mean ± s.e.mean 118.9 96.6 101.4 91.9 108.8 103.5b 114.1 89.6 91.9 88.7 85.0 93.9C 5.2 4.7 A UC (pmol 1- min) 175.2 174.7 166.7 167.3 152.4 152.9 148.7 149.2 68.0 68.0 142.1 142.5 19.1 19.2 tl/2 (min) 86.8 86.7 130.9 129.2 105.0 103.9 97.0 97.1 82.3 79.8 100.4 99.4 7.7 7.6 89.4C tmax = time to maximum concentration; C.,,. =maximum concentration AUC = area under the dialsate concentration versus time curve; t4/2 = apparent Significantly different from , P<0.001; 'Significantly different from b, p <0.05 elimination half-life P.N. Patsalos et al 0% - -% 00 o 0 . 9 00 00 g c _; oi en o la 0% - 0 - a-. '.v -'.0 -a" -'. ogff t. tf . . N 4-6 to, 10 oo _-4 -_- 0 0. . .£-% I"- --j %6 - F _ e .O - .0 -. n- 0) 8- o6 qf 8 of IB 1- -I '.00 _ t 0 cr WIoofi ____~~~~__ 0 00 I'l _ IT 0% 110 t- Ilea ~a' %o0 Co le t rM en '311 r- C. No V 8t O O 0 0 0 A: a acd a C o D 0cd 'It 00 - Xt F w Q woqt w 4- a) _ 00 r0 'e _ le' . a 0.C ,' o 0s o f V~~~O It Cd m Cd .64 0 a a) aCd'A a 0 0 04 co~e he 0 t q Xt A t -O a) O 'N 00 ne-'N-N- _9 a) .0 1 as co A 4-r t co 'Tr _ en _ a __ as .0 eo _~ 0 Cq c " - en" Hcd - co0 .0 tv co t 00 " C4 en.) - a-434 oo>o .0 U. a- e Cd __ > a9 04 a~~C gD- cg > 1 S Microdialysis and antipyrine neuropharmacokinetics 507 methodology and to take into consideration its inherent properties which may impact significantly on data interpretation. The primary purpose of the present study was to investigate the effect of sampling time on the neuropharmacokinetics of antipyrine in rat frontal cortex. The second purpose was to ascertain the contribution of period sampling and of dialysate dead volume. The study design justifies some comment. Firstly, 2 min sampling intervals were chosen because preliminary experiments indicated that such sampling was within our quantitation range for h.p.l.c. antipyrine concentration measurement in dialysate sample volumes of 4 Ml. Secondly, antipyrine dialysate concentration data were simulated for 4, 6, 8, 10 and 20 min from the 2 min data, as opposed to generating the data from different experiments, in order to overcome potential variability in data sets consequent to factors such as variability in antipyrine administration, inter-rat handling of antipyrine, dialysate collection and analytical variability. Thus, the study design was optimumly focused to answer the questions posed. The major findings of this study are that tmax and Cmax are directly dependent on dialysate sampling frequency. Thus, mean 2 min sampling tax and Cmx values were 63% lower and 27% higher, respectively, compared with that of 20 min sampling values. However, in one rat (rat 5, Figure 2b) where apparently a 2 min sample exactly coinsided with the 'true' C.a. (i.e. dilution of antipyrine concentration due to period sampling was minimal), the C,,,, value was 67% higher during 2 min sampling compared to 20 min sampling. Scanning the microdialysis literature, the range of dialysate sampling frequency reported as being used varies from 5 min (e.g. Stahle, 1993; Golden et al., 1993) to 60 min (e.g. Chen & Steger, 1993; Kurata et al., 1993), with most authors using 20 or 30 min sampling time protocols (e.g. Anderson & DiMicco, 1992; Menacherry et al., 1992; Semba et al., 1992). All microdialysis studies report concentration changes of the substance analysed e.g. neurotransmitters or drugs relative to baseline as a percentage or as an absolute concentration. Clearly, from the present study, it can be concluded that during periods of rapid change e.g. neurotransmitter release and uptake or drug absorption, the concentration changes reported are underestimated if sampling time is longer rather than shorter. Furthermore, the present data suggest that studies should not be directly and quantitatively compared if the same sampling time protocol is not used. Perhaps the best approach to monitoring analyte by microdialysis, is to sample as frequently as possible during periods of rapid concentration changes and less frequently during slower concentration changes. With the continuing development of capillary liquid chromatography and the emergence of capillary electrophoresis as practical analytical techniques for quantitating (fmol) substances in very small volumes (nl), frequent sampling during microdialysis studies will be more widely undertaken. An extrapolation of the same arguments indicated above would apply for the time course of events (t,,,.) monitored by microdialysis. Data reported in this respect would be overestimated if sampling time is longer rather than shorter. As tI/2 and AUC are not discrete variables, they should not depend on sampling time and indeed microdialysate sampling time had no significant effect on these values. Because brain microdialysis monitors events at the extracellular level and is therefore considered to reflect events at the synaptic site of action, studies of drug pharmacokineticpharmacodynamic inter-relationships and modelling are feasible and indeed desirable. However, appreciation of the effects of sampling time on Cr,, and tm-- values is imperative in such studies. As one might expect, adjustment of data to take into consideration dialysate dead volume, resulted in significant reductions in tma, values but had no significant effect on the other neuropharmacokinetic parameters. Although conceptually P.N. Patsalos et al 508 dead volume adjustment should be a fundamental consideration in microdialysis studies, in fact publications rarely indicate its consideration. As microdialysis sampling represents period sampling and therefore concentrations of the substance under analysis in any particular sample represent a mean of the time sampled, there has been some debate as to whether microdialysis data in general should be represented on the time-axis by the endpoint, the starting point or the midpoint of the sampling interval since the concentration versus time curves will be different depending on representation chosen (Morrison et al., 1991; Stahle, 1993). In the present study, plotting antipyrine concentration data at the midpoint of the sampling interval compared to endpoint, resulted in some difference in the concentration versus time plots (Figures 3 and 5). However, only tma, values were affected with values decreasing with increasing sampling time. In contrast, Cma,, AUC and tl/2 values were unaffected. Thus, in most situations endpoint or midpoint representation of concentration data will not have a significant effect on kinetic values, except that for long sampling times tmn,, values will be lower when plotted at the midpoint of the sampling time interval. In order to avoid this potential problem, a midpoint sampling interval should be routinely used in plots of concentration versus time. The neuropharmacokinetic properties of antipyrine in rat frontal cortex, as characterized in the present study using microdialysis, in comparison to its pharmacokinetic and neuropharmacokinetic properties using traditional sampling methodologies (i.e. serial blood or cerebrospinal fluid sampling), need some comment. The pharmacokinetics of antipyrine in blood, as ascertained by blood sampling, has been extensively investigated with reported mean tl/2 values ranging from 71.0+ 7.5 min to 148.0 +9.6 min (Pei et al., 1986; Shaw Microdialysis and antipyrine neuropharmacokinedcs et al., 1986; Svensson & Liu, 1987; Tanaka et al., 1989; Van Bezooijen et al., 1989; Van Bree et al., 1989; Ben Zvi et al., 1991; Kurata et al., 1993). Also there appears to be no significant difference in blood antipyrine t112 values when traditional blood sampling and microdialysis blood sampling are directly compared (Kurata et al., 1993). As antipyrine readily and rapidly distributes throughout body water, cerebrospinal fluid and cerebral cortex, one might expect its central neuropharmacokinetics not to differ significantly from that of its peripheral pharmacokinetics. Indeed, in studies comparing antipyrine tl/2 values obtained by direct sampling of blood and cerebrospinal fluid, the two compartments were indistinguishable (85 ± 8 min versus 90 ± 4 min respectively; Johnson & Woodbury, 1977; Van Bree et al., 1989). Furthermore, mean frontal cortex t1/2 values obtained in the present study (99.4 + 7.6 min) are very comparable to values previously reported and confirm the suitability of microdialysis in studies of drug neuropharmacokinetics. Finally, although the present study involved antipyrine neuropharmacokinetics in rat frontal cortex, clearly the conclusions would be valid for drug blood pharmacokinetics and indeed should be a major consideration in such studies. This is particularly important as blood and tissue drug pharmacokinetic studies, using the microdialysis technique in man, are still in their infancy (Stahle et al., 1991a; Bolinder et al., 1993; O'Connell et al., 1994). We wish to thank The Great Britain Sasakawa Foundation for financial support and Nathalie Vomsheid for secretarial assistance. References ABERCOMBIE, E.D., KELLER, R.W. & ZIGMOND, M.J. (1988). Characterisation of hippocampal norepinephrine release as measured by microdialysis perfusion: pharmacological and behavioural studies. Neuroscience, 27, 897-904. ALAVIJEH, M.S., ABED, W.T., O'CONNELL, M.T. & PATSALOS, P.N. (1993). A comparison of classical and microdialysis sampling techniques: a study of carbamazepine (CBZ) pharmacokinetics. Epilepsia, 34, (Suppl. 2), 76. ANDERSON, J.J. & DIMICCO, J.A. (1992). The use of microdialysis for studying the regional effects of pharmacological manipulation on extracellular levels of amino acids - some methodological considerations. Life Sci., 51, 623 -630. BENVENNISTE, H., HANSEN, H.J. & OTTEOSEN, N.S. (1989). Determination of brain interstitial concentrations by microdialysis. J. Neurochem., 52, 1741-1750. BEN-ZVI, Z., HREASH, F. & KAPLANSKI, J. (1991). Disposition of hexobarbitone and antipyrine in DOCA-hypertensive rats. J. Pharm. Pharmacol., 43, 349-352. BOLINDER, J., UNGERSTEDT, U. & ARNER, P. (1993). Long-term continuous glucose monitoring with microdialysis in ambulatory insulin-dependent diabetic patients. Lancet., 342, 1080-1085. BUSTOS, G., ABARCA, J., FORRAY, M.I., GYSLING, K., BRADBER- RY, C.W. & ROTH, R.H. (1992). Regulation of excitatory amino acid release by N-methyl-D-aspartate receptors in rat striatum: in vivo microdialysis studies. Brain Res., 585, 105-115. CARLSON, H., RONNE-ENGSTROM, E., UNGERSTEDT, U. & HILLERID, L. (1992). Seizure related elevations of extracellular amino acids in human focal epilepsy. Neurosci. Lett., 140, 30 - 32. CHEN, Z. & STEGER, R.W. (1993). Plasma microdialysis. A technique for continuous plasma sampling in freely moving rats. J. Pharmacol. Toxicol. Methods, 29, 111- 118. DELEU, D., SARRE, S., MICHOTTE, Y. & EBINGER, G. (1994). Simultaneous in vivo microdialysis in plasma and skeletal muscle: a study of the pharmacokinetic properties of levodopa by non compartmental analysis. J. Pharmaceut. Sci., 83, 25-28. DELGADO, J.M.R., LERMA, J., MARTIN DEL RIO, R. & SOLIS, S.M. (1984). Dialytrode technology and local profiles of amino acids in the awake rat brain. J. Neurochem., 42, 1218-1228. DUNN, A. (1985). A new iterative polyexponential curve fitting program. Comput. Methods Biomed., 20, 601-608. DURING, M.J. & SPENCER, D.D. (1993). Extracellular hippocampal glutamate and spontaneous seizure in the conscious human brain. Lancet., 341, 1607-1610. DYKSTRA, K.H., ARYA, A., ARRIOLA, D.M., BUNGAY, P.M., MORRISON, P.F. & DEDRICK, R.L. (1993). Microdialysis study of zidovudine (AZT) transport in rat brain. J. Pharmacol. Exp. Ther., 267, 1227-1236. GOLDEN, P.L., BROUWER, K.R. & POLLACK, G.M. (1993). Assessment of valproic acid serum-cerebrospinal fluid transport by microdialysis. Pharmac. Res., 10, 1765-1771. GUMBLETON, M. & BENET, L.Z. (1991). Metabolism and laboratory anesthetic protocols in the rat: examination of antipyrine pharmacokinetics. Pharmacol. Res., 8, 544-546. HERNANDEZ, L., GUZMAN, N.A. & HOEBEL, B.G. (1991). Bidirectional microdialysis in vivo shows differential dopaminergic potency of cocaine, procaine and lidocaine for calibration of drug outward diffusion. Psychopharmacol., 105, 264- 268. HURD, J.L., KEHR, J. & UNGERSTEDT, U. (1988). In vivo microdialysis as a technique to monitor drug transport: correlation of extracellular cocaine levels and dopamine overflow in the rat brain. J. Neurochem., 51, 1314-1316. HUTSON, P.H. & CURZON, G. (1989). Concurrent determination of effects of p-chloramphetamine on central extracellular 5-hydroxytryptamine concentration and behaviour. Br. J. Pharmacol., 96, 801-806. JOHANSON, C.E. & WOODBURY, D.M. (1977). Penetration of 'tantipyrine and 14C- barbital into the choroid plexus and cerebrospinal fluid of the rat in vivo. Exp. Brain Res., 30, 65 - 74. JUSTICE, J.B. (1993). Quantitative microdialysis of neurotransmitters. J. Neurosci. Methods, 48, 263 - 276. KURATA, N., INAGAKI, M., KOBAYASHI, S., NISHIMURA, Y., OGUCHI, K. & YASUHARA, H. (1993). Antipyrine concentrations in liver and blood monitored by microdialysis of unrestrained conscious rats. Res. Commun. Chem. Pathol. Pharmacol., 79, 363-369. P.N. Patsalos et al LEHMANN, A., ISACSSON, H. & HAMBER, A. (1983). Effect of in vivo administration of kainic acid on extracellular amino acid pool in the rabbit hippocampus. J. Neurochem., 40, 1314-1320. LUDVIG, N., MISHRA, P.K., YAN, Q.S., LASLEY, S.M., BURGER, R.L. & JOBE, P.C. (1992). The combined EEG-intracerebral microdialysis technique: a new tool for neuropharmacological studies on freely behaving animals. J. Neurosci. Methods, 43, 129-137. MENACHERRY, S., HUBERT, W. & JUSTICE, J.B. (1992). In vivo calibration of microdialysis probes for exogenous compounds. Anal. Chem., 64, 577-583. METZLER, C.M. & WEINER, D.L. (1988). A User's Manual for PCNONLIN. Lexington, KY, USA: Statistical Consultants. MEYERSON, B.A., LINDEROTH, B., KARLSSON, H. & UNGERSTEDT, U. (1990). Microdialysis in the human brain: extracellular measurements in the thalamus of Parkinsonian patients. Life Sci., 46, 301-308. MILLAN, H., CHAPMAN, A.G. & MELDRUM, B.S. (1993). Extracellular amino acid levels in hippocampus during pilocarpineinduced seizures. Epilepsy Res., 14, 139-148. MIZUNO, T., ENDO, Y., ARITA, J. & KIMURA, F. (1991). Acetylcholine release in the rat hippocampus as measured by the microdialysis method correlates with motor activity and exhibits a diurnal variation. Neuroscience, 44, 607-612. MORRISON, P.F., BUNGAY, P.M., HSIAO, J.K., BALL, B.A., MEF- FORD, I.N. & DEDRICK, R.L. (1991). Quantitative microdialysis: analysis of transients and application to pharmacokinetics in brain. J. Neurochem., 57, 103-119. O'CONNELL, M.T., ABED, W.T., ALAVIJEH, M.S. & PATSALOS, P.N. (1993). A chronic neuropharmacokinetic study of carbamazepine (CBZ) in rat frontal cortex using microdialysis. Epilepsia, 34, (Suppl. 2), 88-89. O'CONNELL, M.T., PORTAS, C.M., SARNA, G.S. & CURZON, G. (1991). Effect of p-chlorophenylalanine on release of 5-hydroxytryptamine from the rat frontal cortex in vivo. Br. J. Pharmacol., 102, 831-836. O'CONNELL, M.T., TISON, F., QUINN, N. & PATSALOS, P.N. (1994). A newly developed microdialysis probe for monitoring substances in blood of man: pharmacokinetics of levodopa and its primary metabolite, 3-0-methyldopa in patients with Parkinson's disease, preliminary findings. In Monitoring Molecules in Neuroscience. ed. Louilot, A., Durkin, T., Spampinato, U. & Cador, M. pp. 53 - 54. Gradignan: Publi Typ. PATSALOS, P.N., ALAVIJEH, M.S., ABED, W.T., ELYAS, A.A. & O'CONNELL, M.T. (1993). Use of microdialysis to study brain kinetics: valproate and carbamazepine in rat frontal cortex. Epilepsia, 34, (Suppl. 6), 87. PATSALOS, P.N., DUNCAN, J.S. & SHORVON, S.D. (1988). Effect of the removal of individual antiepiletic drugs on antipyrine kinetics in patients taking polytherapy. Br. J. Clin. Pharmacol., 26, 253259. PAXINOS, G. & WATSON, C. (1986). The Rat Brain in Stereotactic Coodinates. (second edition). San Diego: Academic Press. PEI, Y.Y., BIALER, M. & LEVY, R.H. (1986). Effects of phenobarbital steady state levels on antipyrine clearance and distribution in the rat. Biopharm. Drug Dispos., 7, 11-19. PERSSON, L. & HILLERID, L. (1992). Chemical monitoring of neurosurgical intensive care patients using intracerebral microdialysis. J. Neurosurg., 76, 72-80. RALL, D.P., STABENAU, J.R. & ZUBROD, C.G. (1959). Distribution of drugs between blood and cerebrospinal fluid: general methodology and effect of pH gradients. J. Pharmacol. Exp. Ther., 125, 185- 193. RANE, A., WILKINSON, G.R. & SHAND, D.G. (1977). Prediction of hepatic extraction ratio from in vitro measurement of intrinsic clearance. J. Pharmacol. Exp. Ther., 200, 420-424. SABOL, K.E. & FREED, C.R. (1988). Brain acetaminophen measurement by in vivo dialysis, in vivo electrochemistry and tissue assay: a study of the dialysis technique in the rat. J. Neurosci. Methods, 24, 163-168. SCHELLER, D. & KOLB, J. (1991). The internal reference technique in microdialysis: a practical approach to monitoring dialysis efficiency and to calculating tissue concentrations from dialysate samples. J. Neurosci. Methods, 40, 31 -38. SCHEYER, R.D., DURING, M.J., HOCHHOLZER, J.M., SPENCER, D.D., CRAMER, J.A. &r MATTSON, R.H. (1994a). Phenytoin concentrations in the human brain: an in vivo microdialysis study. Epilepsy Res., 18, 227- 232. Microdialysis and antipyrine neuropharmacokinetics 509 SCHEYER, R.D., DURING, M.J., SPENCER, D.D., TOFTNESS, B.R. & MATTSON, R.H. (1994b). Measurement of carbamazepine and carbamazepine epoxide in the human brain using in vivo microdialysis. Neurology, 44, 1469-1472. SCOTT, D.O., STEELE, K.L. & LUNTE, C.E. (1990). Microdialysis sampling for pharmacokinetic studies. Curr. Separations, 10, 1115. SEMBA, J., DOHENY, M., PATSALOS, P.N., SARNA, G. & CURZON, G. (1992). Effect of milacemide on extracellular and tissue concentrations of dopamine and 5-hydroxytryptamine in rat frontal cortex. Br. J. Pharmacol., 105, 59-62. SHAW, P.N., TSETI, J., WARBURTON, S., ADEDOYIN, A. & HOUSTON, J.B. (1986). Inhibition of antipyrine metabolite formation. Steady state studies with cimetidine and metyrapone in rats. Drug Metab. Dispos., 14, 217-276. SJOBERG, P., OLOFSSON, I.M. & LUNDQUIST, T. (1992). Validation of different microdialysis methods for the determination of unbound steady-state concentrations of theophylline in blood and brain tissue. Pharmac. Res., 9, 1592- 1598. STAHLE, L. (1993). Microdialysis in pharmacokinetics. Eur. J. Drug Metab. Pharmacokin., 18, 89-96. STAHLE, L., ARNER, P. & UNGERSTEDT, U. (1991a). Drug distribution studies with microdialysis. III. Extracellular concentration of caffeine in adipose tissue in man. Life Sci., 49, 1853- 1858. STAHLE, L., SEGERSVARD, S. & UNGERSTEDT, U. (1990). Theophylline concentration in the extracellular space of the rat brain: measurement by microdialysis and relation to behaviour. Eur. J. Pharmacol., 185, 187-193. STAHLE, L., SEGERSVARD, S. & UNGERSTEDT, U. (1991b). Drug distribution studies with microdialysis. II. Caffeine and theophylline in blood, brain and other tissues in rats. Life. Sci., 49, 18431852. STEELE, K.L., SCOTT, D.O. & LUNTE, C.E. (1991). Pharmacokinetic studies of aspirin in rats using in vivo microdialysis sampling. Anal. Chim. Acta, 246, 181-186. STENKEN, J.A., TOPP, E.M., SOUTHARD, M.Z. & LUNTE, C.E. (1993). Examination of microdialysis sampling in a well-characterized hydrodynamic system. Anal. Chem., 65, 2324- 2328. SVENSSON, C.K. & LIU, L.L. (1987). Dose- and time-dependent effect of levamisole on the elimination of antipyrine in the rat. Drug Metab. Dispos., 15, 432-433. TANAKA, E., KOBAYASHI, S., UCHIDA, E., OGUCHI, K. & YASU- HARA, H. (1989). Antipyrine metabolism in female Lewis and dark Agousti strains of rats, which are extensive and poor metabolizers, respectively. Jpn. J. Pharmacol., 49, 433-435. TELTING-DIAZ, M., SCOTT, D.O. & LUNTE, C.E. (1992). Intravenous microdialysis sampling in awake, freely-moving rats. Anal. Chem., 64, 806-810. VAN BENZOOIJEN, C.F.A. & VAN OORSCHOT, R.M. (1989). The effect of age on antipyrine pharmacokinetics and metabolite formation in rats. J. Pharmacol. Exp. Ther., 251, 683 -686. VAN BREE, J.B.M.M., BALJET, A.V., VAN GEYT, A., DE BOER, A.G., DANHOF, M. & BREIMER, D.D. (1989). The unit impulse response procedure for the pharmacokinetic evaluation of drug entry into the central nervous system. J. Pharmacokin. Biopharmaceut., 17, 441-462. VEZZANI, A., UNGERSTEDT, U., FRENCH, E.D. & SCWARCZ, R. (1985). In vivo brain dialysis of amino acids and simultaneous EEG measurement following intra-hippocampal quinolinic acid injection: evidence for dissociation between neurochemical changes and seizures. J. Neurochem., 45, 335-344. YADID, Y., PACAK, K., GOLOMB, E., HARVEY-WHITE, J.D., LIEBERMANN, D.M., KOPIN, I.J. & GOLDSTEIN, D.S. (1993). Glycine stimulates striatal dopamine release in conscious rats. Br. J. Pharmacol., 110, 50-53. YAMAMOTO, B.K. & DAVY, S. (1992). Dopaminergic modulation of glutamate release in striatum as measured by microdialysis. J. Neurochem., 58, 1736- 1742. YOKEL, R.A., ALLEN, D.D., BURGIO, D.E. & MCNAMARA, P.J. (1992). Antipyrine as a dialyzable reference to correct differences in efficiency among and within sampling devices during in vivo microdialysis. J. Pharmacol. Toxicol. Methods, 27, 135-142. (Received October 20, 1994 Revised February 14, 1995 Accepted February 17, 1995)