Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Factors affecting the early arterial drug concentration versus time profile influence the intensity and timing of the onset of drug effect for rapidly acting drugs such as intravenous anesthetics (Krejcie and Avram, 1999). These factors include intravascular mixing of the drug and its distribution to highly perfused tissues by flow and diffusion (Riggs, 1963; Krejcie et al., 1997). Our recirculatory pharmacokinetic model describes these processes by referencing them to the disposition of markers of intravascular space [ICG (Henthorn et al., 1992)], extracellular fluid space [inulin (Henthorn et al., 1982; Krejcie et al., 1996a)], and body water [antipyrine (Krejcie et al., 1996a)].

Physiologically based pharmacokinetic models describe measured blood and tissue drug concentration histories by apportioning cardiac output (CO), hence drug distribution, among tissues or tissue groups with similar perfusion and drug solubility characteristics (Stanski, 1981). In constructing physiologic models, many assumptions are made to interpret tissue concentration data and to adjust the models for different conditions of body composition and blood flow. Because these physiologic models generally assume tissue blood flow is homogeneous, some models assume all blood leaving a tissue is in equilibrium with it (Price et al., 1960), whereas others interpret their tissue data as demonstrating diffusion barriers to flow-limited drug uptake (Ebling et al., 1994). When simulating drug disposition in the presence of altered physiology, physiologic models adjust regional blood flows in direct proportion to changes in CO (Davis and Mapleson, 1993), arbitrarily and independently (Price, 1960), or on the basis of radioactive microsphere regional blood flow measurements (Benowitz et al., 1977) while maintaining the assumption of either complete tissue equilibration with exiting blood or diffusion barriers to flow-limited drug uptake. In addition, the physiologic model does not account for measured drug blood, or tissue, concentrations in the first minutes after drug administration when intravenous anesthetics and other rapidly acting drugs produce their maximum effect (Sear, 1993).

Because traditional multicompartmental pharmacokinetic models lack an anatomic and physiologic basis, they cannot be used to describe the effect of alterations in CO on drug disposition. The present recirculatory multicompartmental model of the disposition of physiologic markers based on frequent early arterial blood sampling (Krejcie et al., 1996a; Avram et al., 1997) retains the best aspects of the traditional multicompartmental model and the physiologically based model in addition to offering significant advantages over both. Only the recirculatory model is able to describe drug disposition from the moment of injection. The recirculatory multicompartmental model makes no assumptions of blood flow distribution due to altered physiology but estimates blood flow to tissue compartments nondestructively in individuals based on calculated intercompartmental clearances of a flow-limited tissue distribution marker. The recirculatory model neither assumes that all tissues are in equilibrium with the blood leaving it nor invokes the concept of diffusion barriers for flow-limited drug uptake. Instead, the recirculatory model describes nondistributive blood flow, or clearance, that returns blood to the central circulation after minimal tissue equilibration due to arteriovenous anastomoses or significant diffusion barriers.

Our earlier canine studies of various paradigms of altered CO and blood flow distribution, including different levels of halothane anesthesia (Avram et al., 1997), different levels of isoflurane anesthesia (Avram et al., 2000), and volume loading as well as mild and moderate hypovolemia (Krejcie et al., 1999), have demonstrated that not only CO but also its peripheral distribution affect early physiological marker concentration history after rapid intravenous administration. These studies have also demonstrated that changes in early antipyrine distribution are not proportional to changes in CO, because regional blood flow changes depend not only on the altered CO but also on the physiologic circumstances leading to these changes in CO.

The purpose of the present study was to examine the pharmacokinetic effect that additional paradigms of altered CO and peripheral blood flow have on the disposition of physiologic markers (drug surrogates) by mixing, flow, and diffusion. The cardiovascular (CV) perturbations were induced by infusions of the nonselective -adrenergic agonist isoproterenol, the nonselective vasodilator nitroprusside, and the ₁-adrenergic agonist phenylephrine in awake dogs.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Experimental Protocol. The design of this pharmacokinetic study entailed 16 individual experiments. Four purpose-bred male coonhounds, weighing 32 to 42 kg (36.6 ± 4.1 kg, see Table 1), were studied on four occasions each in this Institutional Animal Care and Use Committee-approved study. Approximately 1 month before being studied, a Vascular-Access-Port (Access Technologies, Skokie, IL) was implanted with its catheter tip positioned near the aortic bifurcation via a femoral artery of each dog to facilitate frequent percutaneous arterial blood sampling (Garner and Laks, 1985).

Analytical Methods. Plasma ICG concentrations of all samples obtained up to 20 min were measured on the study day by the HPLC technique of Grasela et al. (1987) as modified in our laboratory (Henthorn et al., 1992). Plasma [¹⁴C]inulin concentrations of all samples were determined in duplicate by liquid scintillation counting, using an external standard method for quench correction (Bowsher et al., 1985). Plasma antipyrine concentrations of all samples were measured in duplicate using a modification of an HPLC technique developed in our laboratory (Krejcie et al., 1994, 1996a).

Pharmacokinetic Model. The pharmacokinetic modeling methodology (Fig. 1) has been described in detail previously (Krejcie et al., 1996a; Avram et al., 1997). It is based on the approach described by Jacquez (1996) for obtaining information from outflow concentration histories, the so-called inverse problem. Inulin and antipyrine distributions were analyzed as the convolution of their intravascular behavior, determined by the pharmacokinetics of concomitantly administered ICG, and tissue distribution kinetics (Krejcie et al., 1996a).

View larger version (26K): [in this window] [in a new window]   Fig. 1.   The general model for the recirculatory pharmacokinetics of ICG, inulin, and antipyrine (Krejcie et al., 1996a). CO flows through the central circulation, which is defined by the delay elements (VC). All delay elements are represented generically by rectangles surrounding four compartments, although the number of compartments needed in a delay varied between 2 and 30. The pulmonary tissue volume (VT-P), a subset of VC, is calculated for antipyrine by subtracting the VC of ICG from that of antipyrine. Beyond the central circulation, CO distributes to numerous circulatory and tissue pathways, which lump, on the basis of their blood volume to flow ratios or tissue volume to distribution clearance ratios (MTTs), into fast (CLND-F, VND-F) and slow (CLND-S, VND-S) peripheral blood circuits (ICG) or nondistributive peripheral pathways and fast (CLT-F, VT-F ) and slow (CLT-S, VT-S) tissue volume groups (inulin and antipyrine). ICG, which distributes only within the intravascular space, does not have fast and slow tissue volumes. The nondistributive flows for ICG and inulin were resolved into fast and slow components; antipyrine does not have an identifiable second nondistributive peripheral circuit. The elimination clearance (CLE) of all three markers are modeled from the arterial sampling site without being associated with any particular peripheral circuit.

Area under the Blood Concentration versus Time Relationship (AUC). The AUC was determined for both the first-pass fit (AUC_first-pass, calculated for the sum of two parallel Erlang functions) and for the full recirculatory model (Krejcie et al., 1999). The AUCs for the full model were calculated for the interval 0 to 3 min (AUC_0-3 min), because most intravenous drugs used in the practice of anesthesia (e.g., hypnotics and muscle relaxants) have demonstrable onset within this time. AUC_0-3 min is the sum of AUC_first-pass, which is determined by CO (i.e., AUC_first-pass = dose/CO), and the AUC resulting from marker recirculation (AUC_recirc). To resolve the factors influencing AUC_0-3 min, both AUC_first-pass and AUC_recirc were determined.

Statistical Analysis. The effects of treatment as well as the order of treatment on observed pharmacokinetic parameters were assessed using a general linear model analysis of variance for a Latin square experimental design (NCSS 6.0.2 Statistical System for Windows, Number Cruncher Statistical Systems, Kaysville, UT). Post hoc analysis was carried out using Fisher's least significant difference test. The criterion for rejection of the null hypothesis was p < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

The infusions of the vasoactive drugs reached the targeted minimum 50% decrease in SVR (isoproterenol and nitroprusside) or doubling of SVR (phenylephrine) (Table 1). During the isoproterenol infusion (0.6-2.0 µg/min), SVR decreased to an average of 30% of control, CO more than doubled, HR nearly doubled, and MAP decreased modestly. The nitroprusside infusion (240-400 µg/min) decreased SVR to an average of 43% of control, increased CO slightly, decreased MAP nearly 25%, and increased HR nearly 50%. SVR increased to an average of 204% of control during the phenylephrine infusion (50-120 µg/min), CO decreased by more than one-third, MAP increased nearly one-third, and HR decreased modestly.

Blood ICG, inulin, and antipyrine concentration versus time relationships were well characterized by the models from the moment of injection (Figs. 2-4) The one-sample runs test confirmed that there were no systematic deviations of observed data from calculated values.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Arterial blood ICG concentration histories for the first 1.5 min (illustrating the first- and second-pass peaks) and for 15 min (inset) following right atrial injection in one dog when it received no vasoactive drug (control, closed circles, solid line), during an isoproterenol infusion (upright triangles, long dashed line), during a nitroprusside infusion (inverted triangles, short dashed line), and during a phenylephrine infusion (diamonds, dotted line). The symbols represent drug concentrations; the lines represent concentrations predicted by the models.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Arterial blood inulin concentration histories for the first 1.5 min (illustrating the first- and second-pass peaks) and for 360 min, or to the limit of detection (inset), following right atrial injection in one dog when it received no vasoactive drug (control, closed circles, solid line), during an isoproterenol infusion (upright triangles, long dashed line), during a nitroprusside infusion (inverted triangles, short dashed line), and during a phenylephrine infusion (diamonds, dotted line). The symbols represent drug concentrations; the lines represent concentrations predicted by the models.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Arterial blood antipyrine concentration histories for the first 1.5 min (illustrating the first- and second-pass peaks) and for 360 min, or to the limit of detection (inset), following right atrial injection in one dog when it received no vasoactive drug (control, closed circles, solid line), during an isoproterenol infusion (upright triangles, long dashed line), during a nitroprusside infusion (inverted triangles, short dashed line), and during a phenylephrine infusion (diamonds, dotted line). The symbols represent drug concentrations; the lines represent concentrations predicted by the models.

Hematocrit increased an average of 16% and 41% during the isoproterenol and phenylephrine infusions, respectively, but was unaffected by nitroprusside (Table 1). Blood volume defined by the V_SS of ICG was increased nearly 15% by isoproterenol but neither extracellular fluid volume defined by the inulin V_SS nor total body water defined by the antipyrine V_SS were affected by infusions of the vasoactive agents (Tables 1-4). The ICG and inulin CL_E values were unaffected by vasoactive drug infusions, but the CL_E of antipyrine was increased more than 20% by the infusion of isoproterenol (Tables 1-4).

ICG. Of the blood volume estimated by ICG disposition (Table 2), V_C contained approximately one-third, whereas 15% was in V_ND-F and nearly half was in V_ND-S under the control condition as well as during both the isoproterenol and phenylephrine infusions. Despite the more than doubling of CO during the isoproterenol infusion and the 40% decrease in CO during the phenylephrine infusion, CO during those infusions was distributed between the fast and slow peripheral vascular circuits in a manner very similar to its distribution under control conditions, during which more than half of CO was represented by CL_ND-F and approximately 40% was represented by CL_ND-S.

Inulin. The volumes of the recirculatory inulin pharmacokinetic model and the fractions of V_SS they represent were largely unaffected by infusions of the vasoactive drugs (Table 3). V_C, V_ND-S, and the volumes of the rapidly and slowly equilibrating tissue volumes (V_T-F and V_T-S, respectively), which represented nearly 80% of the total inulin volume of distribution, were unaffected by infusions of the vasoactive drugs. The fraction of the total inulin distribution volume represented by V_ND-F nearly doubled during the isoproterenol infusion but was unchanged under the other conditions.

Antipyrine. The antipyrine distribution volume most affected by the vasoactive drugs was V_T-F (Table 4), which represented approximately one-third of peripheral tissue distribution volume (V_T-F + V_T-S) under control conditions but increased during the isoproterenol infusion to represent half of peripheral tissue distribution volume and decreased during the nitroprusside and phenylephrine infusions to represent approximately 20% of peripheral tissue distribution volume. The only peripheral nondistributive volume that could be independently resolved in the antipyrine model, V_ND (Krejcie et al., 1996a), increased significantly during the isoproterenol infusion but was unaffected by either nitroprusside or phenylephrine.

AUC. The AUCs for at least the first 3 min after drug administration were decreased by isoproterenol, unaffected by nitroprusside, and increased by phenylephrine (Figs. 2-4 and Table 5). Because ICG is distributed only within the intravascular space and inulin distributes only to the extracellular fluid space, the changes in AUC in the first minutes after ICG and inulin administration are due to changes in first-pass AUC resulting solely from alterations in CO (AUC_first-pass = Dose/CO). Despite the more extensive tissue distribution of antipyrine, because the increase in CL_ND was proportional to the increase in CO during the isoproterenol infusion, the antipyrine AUC_0-3 min decreased due to a smaller AUC_first-pass (increased CO). In contrast, because of the fractional increase in nondistributive blood flow during the phenylephrine infusion despite a significant decrease in CO, the antipyrine AUC_0-3 min increased due to increases in both AUC_first-pass and AUC_recirc.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The cardiovascular effects of isoproterenol, nitroprusside, and phenylephrine (Table 1) were consistent with their expected effects (Hoffman and Lefkowitz, 1996; Oates, 1996). Isoproterenol decreased SVR by relaxing vascular smooth muscle and increased CO through positive inotropic and chronotropic effects; the net effect of these changes was a decrease in MAP. Nitroprusside decreased SVR, hence MAP, by dilating arterioles and venules, despite a modest reflex increase in HR and CO. Phenylephrine increased SVR and MAP as a result of marked arterial vasoconstriction and caused a reflex slowing of the heart rate and decrease in CO. Phenylephrine increased hematocrit as a result of splenic contraction.

The nondistributive peripheral pathway in the antipyrine model represents blood flow that returns the lipophilic marker to the central circulation after minimal apparent tissue distribution (Krejcie et al., 1996a; Avram et al., 1997). Antipyrine V_T-F represents splanchnic tissues, whereas V_T-S represents nonsplanchnic tissues (primarily muscle) (Sedek et al., 1989; Krejcie et al., 1996a). The ability of our model to describe changes in physiologic marker disposition due to altered CO and peripheral blood flow distribution can be evaluated by comparing blood flow-dependent antipyrine disposition during a given experimental condition with blood flow measurements made with radioactive microspheres under similar conditions. The studies most suitable for comparison are those conducted with microspheres (8-10, or 15 µm), which distribute like erythrocytes (but with capillary trapping) and quantify arteriovenous anastomoses in terms of the fraction of microspheres trapped in the lungs (Heymann et al., 1977). Unless evaluating anesthetic effects, studies being `compared should have been conducted in awake animals, because anesthesia affects both CO and its distribution (Forsyth and Hoffbrand, 1970; Gelman et al., 1984; Yano and Takaori, 1994).

High dose isoproterenol infusions in conscious rhesus monkeys increased the fraction of CO flowing through the portal vein 18% as measured by microspheres (Hoffbrand et al., 1973), which is nearly identical to the 20% increase in antipyrine CL_T-F we observed in awake dogs (Table 4), with no change in the fraction of CO represented by either arteriovenous anastomotic flow or CL_ND, respectively. However, although in our study antipyrine CL_T-F (splanchnic) increased at the expense of CL_T-S (muscle), the fraction of CO received by skeletal muscle blood flow in the monkey was not affected by high dose isoproterenol infusions. This apparent discrepancy may be explained by our observation that slowly equilibrating tissue volume under control conditions appeared rapidly equilibrating during the isoproterenol infusion. Because pharmacokinetic heterogeneity exists within single tissue types (Crone, 1979), the shift from slow to fast could be explained by a change in the balance of tissue blood supply of the dual circulation of muscle (Renkin, 1955) that is not apparent in microsphere studies, because they measure only total tissue blood flow.

Just as in our study, microsphere studies of regional blood flow distribution during nitroprusside infusions in awake dogs (Dumont et al., 1983) and young lambs (Kuipers et al., 1984) found no change in the fraction of CO represented by flow to the splanchnic system (CL_T-F), muscle (CL_T-S), or arteriovenous anastomoses (CL_ND).

Phenylephrine nearly doubled the fraction of CO represented by CL_ND in the present study (Table 4) and the fraction of CO flowing through arteriovenous anastomoses in microsphere studies in the pithed rat (Hiley and Thomas, 1987). Although in our study the phenylephrine-induced increase in CL_ND occurred at the expense of antipyrine CL_T-F, phenylephrine did not change the fraction of CO flowing through the hepatosplanchnic circulation of the rat. This apparent discrepancy may be explained by the observation that in our study rapidly equilibrating tissue volume under control conditions appeared to be slowly equilibrating during the phenylephrine infusion, a change in the balance of tissue blood flow that is not apparent in microsphere studies.

Other studies likewise suggest that the increased fraction of CO represented by antipyrine CL_ND is due to opening of arteriovenous anastomoses. Gelman et al. (1984) found that in dogs anesthetized with 1 and 2 minimum alveolar concentration (MAC) halothane the percentage shunting of microspheres was 2.5 and 4 times that during awake control, respectively, increases that were similar to increases in the fraction of CO represented by antipyrine CL_ND in our halothane-anesthetized dogs (Avram et al., 1997). Yano and Takaori (1994) reported an 8.9% control microsphere shunt rate that is consistent with the 9.6% of CO that antipyrine CL_ND represented in our awake dogs, whereas their 19.9% and 17.4% shunt rates in dogs anesthetized with 1% and 2% isoflurane, respectively, are consistent with the 21.6% of CO that antipyrine CL_ND represented in our dogs when they were anesthetized with 1.7% isoflurane (Avram et al., 2000).

The importance of CO in determining dose requirements of drugs with a rapid onset of effect was recognized by Price (1960). He explained reduced thiopental dose requirements of patients with decreased CO on the basis of a higher fraction of the dose received by the brain and a slow rate of removal from the brain due to decreased blood flow to indifferent tissues. Price also predicted that patients with increased CO would require larger thiopental doses, because a smaller fraction of the intravenous dose would appear in the brain due to increased fractional blood flow to indifferent tissues.

An important observation of our work with various paradigms of perturbed physiology is that not only CO but also its distribution affect early drug concentrations, as reflected in the AUC in the first minutes after intravenous administration (Avram et al., 1997, 2000; Krejcie et al., 1999; Table 5) and suggested previously by Upton et al. (1999; Krejcie and Avram, 1999). The fraction of CO represented by CL_ND and the balance of blood flow between CL_T-F and CL_T-S are important determinants of early drug concentrations. Increased arterial drug concentrations resulting from a larger fractional CL_ND increases drug exposure of the sites of action of drugs with a rapid onset of effect, for which antipyrine is a surrogate (Renkin, 1952; Krejcie et al., 1996a) and would be expected to produce a more profound and prolonged effect.

Although halothane, isoflurane, phenylephrine, and hypovolemia all reduced CO and increased antipyrine AUC in the first minutes after drug administration due to altered CO and peripheral blood flow distribution, their mechanisms were different and, consequently, increases in antipyrine AUC differed. CL_ND increased both absolutely and as a fraction of CO during halothane anesthesia (Avram et al., 1997), whereas during isoflurane administration CL_ND was unchanged but increased as a percentage of the decreased CO (Avram et al., 2000). At the highest halothane concentration studied, CL_ND equaled total distribution clearance (CL_T-F + CL_T-S) because of a profound decrease in CL_T-F and a lesser decrease in CL_T-S, while at the highest isoflurane concentration studied CL_ND was only half of total distribution clearance, although isoflurane decreased both V_T-F and CL_T-F. Like isoflurane, phenylephrine increased CL_ND only as a fraction of CO and at the expense of the fraction of CO represented by CL_T-F to a smaller V_T-F (Table 4). Hypovolemia increased the fraction of CO represented by CL_ND, because CL_ND did not change, whereas CO decreased, but this fractional increase affected AUC less than that observed during isoflurane anesthesia and the phenylephrine infusion, because it was only at the expense of a fractional decrease in flow to the slowly equilibrating tissues (CL_T-S) (Krejcie et al., 1999).

Both isoproterenol and volume loading increased CO and decreased antipyrine AUC in the first minutes after drug administration. However, isoproterenol decreased AUC only as a result of increased CO (Table 5), whereas volume loading decreased AUC as a result of alterations of both CO and peripheral blood flow distribution (Krejcie et al., 1999). Although isoproterenol more than doubled CO, because the fraction of CO represented by both CL_ND and CL_T-F remained virtually unchanged, AUC decreased only as a result of increased CO (Table 5). Volume loading decreased antipyrine AUC due to increases in both CO and CL_T-F, but the decrease in AUC was not as profound as might be expected, because CL_ND also increased (Krejcie et al., 1999).

Nitroprusside produced the only cardiovascular perturbation studied that had no significant effect on antipyrine AUC (Table 5). Despite profound effects on MAP and SVR (Table 2), nitroprusside had little effect on CO and its distribution; antipyrine clearances remained proportional to CO during the nitroprusside infusion (Table 4). Thus, the nitroprusside perturbation was the only one that behaved in a manner consistent with modeling altered physiology in a physiologic kinetic model by adjusting regional blood flow in proportion to changes in CO (Davis and Mapleson, 1993).

Our recirculatory pharmacokinetic model describes the effect of altered CO and blood flow distribution on drug disposition nondestructively in awake subjects without the assumptions made in physiologic modeling and is thus suitable for studying the effect of altered physiology on early drug disposition in both animals and humans. The data of the present study provide further evidence that both CO and its distribution affect early physiologic marker concentrations after rapid intravenous administration and that changes in antipyrine [a prototype for many lipophilic drugs (Renkin, 1952; Krejcie et al., 1996a)] blood flow-dependent distribution are not proportional to changes in CO. Alterations in CO and its distribution are likely to provide the pharmacokinetic bases of interindividual differences in response to drugs with a rapid onset of effect.