| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
BACKGROUND: We hypothesized that pairs of inhaled anesthetics having divergent potencies [one acting weakly at minimum alveolar anesthetic concentration (MAC); one acting strongly at MAC] on specific receptors/channels might act synergistically, and that such deviations from additivity would support the notion that anesthetics act on multiple sites to produce anesthesia.
METHODS: Accordingly, we studied the additivity of MAC for 11 anesthetic pairs divergently (one weakly, one strongly) affecting a specific receptor/channel at MAC. By "divergently," we usually meant that at MAC the more strongly acting anesthetic enhanced or blocked the in vitro receptor or channel at least twice (and usually more) as much as did the weakly acting anesthetic. The receptors/channels included: TREK-1 and TASK-3 potassium channels; and RESULTS: All combinations produced a result within 10% of that which would be predicted by additivity except for the combination of isoflurane with nitrous oxide where infraadditivity was found. CONCLUSIONS: Such results are consistent with the notion that inhaled anesthetics act on a single site to produce immobility in the face of noxious stimulation.
A present consensus holds that ligand-gated and/or voltage-gated ion channels mediate the actions of inhaled anesthetics, including the capacity of these anesthetics to produce immobility. Plausible candidates include amino butyric acid (GABAA), glycine, serotonin, nicotinic acetylcholine, various glutamate [e.g., NMDA (N-methyl-d-aspartic acid) and AMPA-kainate (AMPA is -amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid)], adrenergic, and opioid receptors, and potassium and sodium channels. We previously examined the relevance of most of these to mediate the capacity of inhaled anesthetics to produce immobility in the face of noxious stimulation (i.e., their relevance to MAC, the minimum alveolar concentration of inhaled anesthetic that produces immobility in 50% of subjects given a noxious stimulation).1 Each channel appeared to have, at most, a modest relevance as a mediator of immobility.1 Thus, if such channels explain anesthesia, they must act in concert and probably must produce more than simply an additive effect. Synergy is required because the modest anesthetic effects on individual channels seem insufficient to produce immobility through an additive summation of effects. And yet, as noted in a companion article,2 additivity or infraadditivity (antagonism) appears to be the rule for combinations of inhaled anesthetics. Inhaled anesthetics vary greatly in their in vitro actions on specific ion channels. In a companion article, Shafer et al.3 find that divergent potencies at two or more receptors (i.e., different potencies at two distinct receptors at MAC) would likely produce synergy or antagonism. Indeed, the more complex the circuitry, the more difficult it is to understand how anesthetics acting at different locations could behave in an additive manner. Here is a specific example of the reasoning: anesthetizing concentrations of cyclopropane minimally affect glycine or GABAA receptors whereas isoflurane and halothane cause major enhancements.4–6 In contrast, cyclopropane depresses the NMDA receptor7 far more than halothane and isoflurane. If the synergistic summation of small effects at multiple ion channels mediates anesthesia, then cyclopropane should demonstrate synergy, not additivity, when combined with isoflurane or halothane. The companion manuscript by Hendrickx and colleagues2 extends this view beyond inhaled anesthetics, proposing that additivity is the exception for the interaction of drugs that target different receptors. In a third companion essay, Jenkins et al.8 show that anesthetics with divergent effects (different potencies at MAC) on a single receptor do not produce synergy when they act on that receptor alone. These thoughts and findings lead to the notion that combinations of anesthetics with divergent potencies on two or more receptors may produce an effect different from additivity, whereas divergent potencies do not produce synergy for their action on a single receptor (two or more receptors in combination are needed). Thus, we hypothesize that anesthetic pairs having divergent potencies on a specific receptor will act synergistically in the whole animal. We assume that if an anesthetic weakly affects one receptor, then to produce anesthesia (specifically immobility or MAC) it must compensate by strongly affecting another. Conversely, if an anesthetic strongly affects one receptor, then it probably weakly affects another. That is, two anesthetics that differ greatly in their effects on a given receptor/channel also will likely have conversely different effects on other receptors/channels. The proponents of the argument that two or three or four small effects on as many channels would combine to produce a major effect might predict that the concurrent administration of such anesthetic pairs would have synergistic effects.
With approval of the Committee on Animal Research of the University of California, San Francisco, we studied approximately 260 male specific-pathogen-free, Sprague- Dawley rats (Crl:CD(SD)BR) weighing 300–450 g obtained from Charles River Laboratories (Hollister, CA). Each animal was caged with up to as many as two additional rats, and all had continuous access to standard rat chow and tap water before study. We studied 11 anesthetic pairs defined by differences in potency for a given ligand- or voltage-gated channel (Table 1). The first column of anesthetics listed in Table 1 had much less effect in vitro (were less potent) than the second column of anesthetics on the indicated receptor/channel. We also studied one anesthetic pair (benzene-cyclopropane) defined by a difference in responsiveness to the NMDA blocker MK-801.14 We studied four additional pairs of anesthetics because of previous reports suggesting a deviation from additivity for the combination of nitrous oxide and isoflurane.15 The pairs were nitrous oxide plus cyclopropane, desflurane, isoflurane, or sevoflurane. Three of these pairs also differed in their capacities to block NMDA receptors at MAC; desflurane, isoflurane, and sevoflurane have a smaller capacity than does nitrous oxide.7 We also studied sevoflurane with halothane or isoflurane because these pairs of anesthetics have opposite effects on release of calcium from sarcoplasmic reticulum.16 Finally, we attempted to determine the additivity of the combination of n-hexane plus 2,4-hexadiene to test whether the combination of rigid and flexible anesthetics17 produced a deviation from additivity. We considered that the greater flexibility of n-hexane might allow it to enter a site that excluded the rigid 2,4-hexadiene and thus would, as with the other pairs, test the effect of actions on separate sites.
We determined MAC using electrical stimulation18 in groups of 5–12 rats. Two to four rats were studied at any one time. Studies of a potent inhaled anesthetic plus xenon or nitrous oxide were conducted in a hyperbaric chamber in which the oxygen partial pressure always exceeded 0.6 atm, and usually was approximately 1 atm. Each rat supplied a single value for either the MAC of one of the pairs of study anesthetics or for the combination. Each rat was used once. A period of 45–55 min was used to provide equilibration with the initial anesthetic concentration or combination of concentrations. The less soluble anesthetic was added at approximately 50% of its MAC concentration and was sustained at that concentration for the remainder of the experiment. The second anesthetic was added to produce 25%–30% of its MAC concentration at the end of the 45–55 min period of initial equilibration. All rats initially moved in response to stimulation at the end of this equilibration. The concentration of the second anesthetic then was increased in steps that produced a 10%–20% increase in its MAC contribution at the end of the equilibration times of 25–35 min. The longer times were used for more soluble anesthetics. At the bracketing concentrations [highest concentration(s) permitting and lowest concentration(s) preventing movement in response to stimulation], the step size increases equaled 10%–15% of the preceding value for the sum of the normalized fractional MAC values for the 2 anesthetics. Rectal temperature was measured and maintained between 36°C and 39°C. Anesthetic concentrations were analyzed using gas chromatography as described previously.19 A thermal conductivity detector was used to analyze nitrous oxide and xenon, and a flame ionization detector was used to analyze the remaining anesthetics. The chromatograph was calibrated with either primary or secondary standards. MAC was calculated as the average of the bracketing concentrations. A priori, we defined additivity as occurring when the sum of the fractional MAC values for 2 anesthetics that produced immobility in 50% of subjects lay within 10% of 1.0. That is, when
where ConcA plus ConcB are the concentrations of anesthetics A and B that combined produced immobility in 50% of subjects, and MACA and MACB are their MAC values. ConcA/MACA and ConcB/MACB are the fractional MAC contributions of anesthetics A and B. We a priori defined synergy as occurring when an anesthetic pair produced a result <90% of that which would indicate additivity. That is, there was synergy when the sum of the fractional MAC contributions that produced immobility was <0.9. Similarly, we a priori defined infraadditivity (antagonism) as occurring when an anesthetic pair produced immobility with a sum exceeding 110% of that which would indicate additivity. That is, infraadditivity arose when the sum of the fractional MAC contributions that produced immobility exceeded 1.1.
No anesthetic pair produced synergistic results (Tables 2 and 3; Fig. 1). The combination of nitrous oxide with isoflurane, but not with other potent inhaled anesthetics, produced infraadditivity (Table 3). Results of the tests of the additivity of n-hexane and 2,4-hexadiene were compromised by experimental difficulties. Of the four rats tested, one had frank convulsions, two had preconvulsive activity, and three died during the determinations of MAC.
We failed to confirm our hypothesis that anesthetic pairs having divergent potencies on a given receptor would act synergistically. None of 16 test pairs produced synergy, and all but one (where we found infraadditivity) produced additive results (Tables 2 and 3). These findings appear to conflict with those of a companion paper which finds that in vivo synergy is common for combinations of anesthetics with divergent receptor potencies,2 and with a second companion paper which, from a theoretical consideration of potential interactions with receptors, argues that synergy should be common.3 Our results largely confirm earlier in vivo work with inhaled anesthetics. Previous investigations testing combinations also primarily found additivity (Table 4). Those that did not usually applied an anesthetic whose value for MAC had to be estimated by extrapolation. In rats, DiFazio et al.22 found additivity for three anesthetic pairs where they could measure the MAC of both pairs (fluroxene-halothane, cyclopropane-teflurane, and cyclopropane-halothane). When they made estimates by extrapolation, they sometimes found modest infraadditivity (cyclopropane-nitrous oxide, cyclopropane-ethylene). Similarly, Cole et al. found infraadditivity for combinations of nitrous oxide with enflurane, halothane, and isoflurane in rats.26 These studies may have under-estimated nitrous oxide MAC [e.g., Difazio et al. found an extrapolated value of 1.3–1.4 atm in rats whereas direct measurement of MAC gives a value between 2.2 and 2.4 atm20] (different as a function of different rat strains), and their MAC for ethylene was variable (1.3 atm in one study and 2.0 atm in a second).
Russell and Graybeal also may have under-estimated MAC for nitrous oxide with/without isoflurane, but not because they arrived at MAC by extrapolation. Although they measured MAC directly in a hyperbaric chamber, they used only one set of electrodes and applied a voltage (50 V) that may have burned the electrode site and thereby lessened the perception of stimulation.15 Nonetheless, the results of the present study qualitatively support their finding that combinations of nitrous oxide with a potent inhaled anesthetic produce infraadditivity. However, the extent of infraadditivity found in the present study (19% ± 6%; i.e., the sum of the MAC contributions of nitrous oxide and isoflurane was 1.19 ± 0.06) was half that found by Russell and Graybeal (35%). In humans, where MAC for nitrous oxide has been measured,27 additivity with potent inhaled anesthetics is found.28–34 These results differ from those found with rats (above). However, one study of MAC-Awake found infraadditivity for sevoflurane or isoflurane combined with nitrous oxide and additivity for the combination of these potent anesthetics plus xenon.35 Similarly, we found that nitrous oxide plus isoflurane act infraadditively for both MAC-Awake (an 18% infraadditivity) and learning and memory (5% infraadditivity).36 We have no explanation for the infraadditivity found with the isoflurane-nitrous oxide combination in the present or previous studies. Our present data suggest that both isoflurane and nitrous oxide contribute to the infraadditivity. As indicated above, the combination of isoflurane and nitrous oxide produced significant infraadditivity, whereas nitrous oxide plus cyclopropane, desflurane, or sevoflurane only produced additivity (Table 3). Similarly, isoflurane combined with any anesthetic other than nitrous oxide only produced additivity. However, when either nitrous oxide or isoflurane were combined with a given anesthetic, the resultant average MAC sum exceeded the sum for that anesthetic with any anesthetic other than nitrous oxide or isoflurane. For example, sevoflurane plus isoflurane produced a value of 1.087 ± 0.087, whereas sevoflurane plus halothane gave 1.014 ± 0.110 and sevoflurane plus xenon gave 0.972 ± 0.062. Similarly, cyclopropane plus nitrous oxide gave 1.012 ± 0.072, whereas cyclopropane plus benzene gave 0.979 ± 0.054 and cyclopropane plus halothane gave 0.909 ± 0.061. The chance of all six combinations of nitrous oxide with anesthetic "A" producing a value that exceeded the value for "A" with any anesthetic other than isoflurane is 1 in 32 (P = 0.03). Thus, previous studies all suggest additivity or infraadditivity of inhaled anesthetic combinations. Consistent with those findings, no present study pair demonstrated synergy (Table 1). The greatest downward deviation from unity was found with cyclopropane plus halothane, which gave a value of 0.909 ± 0.061. A previous report for this combination found additivity.22 Could we have overlooked a few crucial synergistic combinations? We cannot exclude that possibility, but we did not limit our test to a single pair of anesthetics with divergent potencies on one receptor/channel. We explored interactions between several inhaled anesthetic pairs with known divergent potencies for diverse ion channels. If additivity is unusual for drug combinations other than inhaled anesthetics, as suggested by Hendrickx et al.2 and Kissin,37 then we should have found synergy in some of our studies because of the multiple anesthetic pairs and receptors/channels we investigated. As noted, combinations of injected anesthetics often, but not inevitably, produce synergistic effects [Hendricks et al.2 and Kissin37]. Synergy also may result from combinations of inhaled anesthetics with injected anesthetics. Fentanyl (thought to act primarily on opioid receptors) synergistically decreases the MAC of sevoflurane in humans,38 and sufentanil has a similar effect in rats.39 Thiopental (thought to act primarily on GABAA receptors) synergistically decreases the MAC of halothane in rats.40 If GABAA were involved in mediating immobility for inhaled anesthetics, then combinations of inhaled anesthetics with divergent potencies on GABAA should also show synergy. In rats, the intrathecal infusion of muscimal (a GABAA agonist) and either AP-5 or YM872 (NMDA and AMPA blocking drugs) decrease MAC for sevoflurane.32 Administration of 10 µg muscimal decreases MAC by 50%, 10 µg AP-5 by 30%, and 10 µg YM872 by 40%. But, the combination of 10 µg each of muscimal plus AP-5 or muscimal plus YM872 only decreases MAC by 50%. That is, these combinations act infraadditively rather than synergistically despite their divergent receptor potencies. One of the pairs we used, benzene-cyclopropane, does not appear to fit our criteria of divergent effects on a single receptor. These two anesthetics have similar in vitro effects on the NMDA receptor, both strongly blocking the receptor at 1 MAC.7 We chose to study this pair because we found that infusion of the NMDA blocker MK-801 decreased the MAC for benzene by only 4% but decreased that for cyclopropane by 58%,14 implying a different interaction of benzene versus cyclopropane with the NMDA receptor. Nonetheless, the combination of benzene with cyclopropane produced additive results. Similarly, we studied pairs of xenon plus halothane and xenon plus isoflurane because two reports for these pairs in swine gave significantly different extrapolated values for the MAC of xenon,42,43 suggesting synergy, infraadditivity, or both. For the sevoflurane-xenon combination, the extrapolated value for the xenon MAC was 166% of an atmosphere whereas with the halothane-xenon combination it was 119% of an atmosphere. Xenon produces a greater in vitro blockade of NMDA receptors than halothane or isoflurane7 but produces less enhancement of the response of the GABAA receptor to GABA.44 Thus, there is a potential for synergy. Despite these observations, we found neither synergy nor infraadditivity (Table 1). The Introduction to this report suggested that synergy would be required to explain anesthesia because the modest effects on six well-studied individual ion channels (GABAA, 5-HT, glycine, NMDA, and acetylcholine receptors, and TREK potassium channels) seem insufficient to produce immobility through an additive summation of effects. But what about the possibility of a larger, perhaps 20, such channels, each contributing 5% to produce a sum equaling a 100% effect? We cannot exclude this possibility, but considered what it would imply. Given that the spinal cord mediates the immobility produced by inhaled anesthetics,45 each of the 20 receptors would have to be present in appreciable numbers in the spinal cord, or, if in small numbers, such a receptor would have to contribute mightily. Each of the 20 receptors would have to be affected by anesthetics at relevant (anesthetizing) concentrations in a way that would plausibly explain anesthesia. Also, note the enormous variation in the differences in functional potencies: acetycholine receptors nearly completely blocked at 0.1–0.2 MAC46 and NMDA receptors affected, the majority blocked, by some anesthetics only at concentrations 2–3 times MAC.7 Somehow the receptors that are affected at sub-MAC concentrations would contribute less than those requiring supra-MAC concentrations so that at MAC each would contribute just 5% to the MAC. That is, fortuitously, each would, indeed, make a 5% contribution without any one receptor providing a large, obvious, contribution (i.e., one that would exceed the 5% average contribution, and therefore be measurable). That is, the 20-receptor theory would have to argue that no one receptor would stand out, even for one anesthetic. The 20 receptors would have to add to each other's effect and not be synergistic (lest we see wide differences in potency/MAC among anesthetics that did not reflect the Meyer–Overton correlation). Thus, we see the 20-receptor theory as possible, but unlikely. The notion that immobility results from a synergistic summation of multiple small effects is not supported by the additivity or infraadditivity demonstrated in this study. Indeed, a synergistic summation would appear to be an unlikely hypothesis in light of these data. Thus, our results do not support a multisite theory of narcosis wherein anesthesia results from a synergistic effect of anesthetics on multiple sites. Our findings are consistent with a unitary theory of narcosis, one in which inhaled anesthetics act on a single site yet to be defined. Whether this is a nonspecific site parallel to the lipid-like site envisioned by the Meyer–Overton hypothesis, or an as-yet-to-be-determined receptor or channel remains unclear.
Supported by NIH grant 1PO1GM47818 (UCSF). Dr. Eger is a paid consultant to Baxter Healthcare Corp. Baxter Healthcare Corp. donated the desflurane and isoflurane used in these studies. Dr. Steven L. Shafer, Editor-in-Chief, was recused from all editorial decision related to this manuscript. Dr. Flood is the wife of Dr. Shafer, Editor-in-Chief of Anesthesia & Analgesia. This manuscript was handled by James Bovill, former Section Editor of Anesthetic Pharmacology, and Dr. Shafer was not involved in any way with the editorial process or decision.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|