Halothane impairs the hemodynamic influence of endothelium-derived nitric oxide.

BACKGROUND: The endogenous vasodilator endothelium-derived nitric oxide (EDNO) contributes to the regulation of vascular tone and organ perfusion. It has been suggested that some volatile anesthetics may diminish the influence of EDNO and thereby decrease regional blood flow. METHODS: Radioactive microspheres were used to determine regional hemodynamics in rats. The authors tested the hypothesis that halothane inhibits EDNO and, therefore, should diminish the response to nitric oxide synthesis inhibition by NW-nitro-L-arginine methyl ester (L-NAME) compared with either conscious or barbiturate-anesthetized rats. RESULTS: NW-nitro-L-arginine methyl ester decreased blood flow to the brain by 23% (P < 0.005) in conscious rats to a level similar to that seen with either anesthetic agent. In both conscious and barbiturate-anesthetized rats, L-NAME increased blood pressure (BP) by 24 +/- 2 (P < 0.001) and 20 +/- 1 (P < 0.001) mmHg and total peripheral resistance (TPR) by 132% (P < 0.001) and 105% (P < 0.001), respectively. In contrast, during halothane anesthesia, both the pressor response (only 7 +/- 1 mmHg) and the increase in TPR (only 22%) were greatly diminished (P < 0.001). NW-nitro-L-arginine methyl ester decreased cardiac output (CO) by 47% (P < 0.001) and heart rate (HR) by 28% (P < 0.001) in conscious rats. In barbiturate-anesthetized rats, L-NAME decreased CO by 38% (P < 0.005) and HR by 13% (P < 0.001). In halothane-anesthetized rats, L-NAME changed neither CO nor HR. Thus halothane anesthesia largely eliminated the systemic response to EDNO synthesis inhibition. In conscious rats, L-NAME decreased blood flow to the heart (30%) and kidneys (47%). In barbiturate-anesthetized rats, L-NAME did not alter blood flow to the heart but decreased renal blood flow by 35% (P < 0.005). In halothane-anesthetized rats, L-NAME did not alter blood flow to either the heart or the kidneys. Overall, halothane blunted or blocked the systemic and regional hemodynamic responses to EDNO synthesis inhibition seen in conscious and barbiturate-anesthetized rats. CONCLUSIONS: Halothane anesthesia greatly diminished or eliminated all systemic and regional hemodynamic responses to L-NAME. These data indicate that halothane anesthesia inhibits EDNO-mediated regulation of systemic and organ hemodynamics.

Vascular effects of halothane and isoflurane: cGMP dependent and independent actions.

This study investigated the effects of halothane and isoflurane on cGMP-dependent and independent regulation of vascular contraction of the isolated rat aorta and on NO-stimulated soluble guanylate cyclase (sGC) isolated from the perfused rat liver. For the studies of the aorta, isometric tension of isolated rings, with and without, endothelium was recorded and cGMP content measured. ACh was used to initiate endothelial-dependent relaxation of norepinephrine (NE)-contracted rings while NO was used to directly stimulate isolated aortic ring sGC which catalyzes the isolated aortic ring formation of cGMP. Both halothane and isoflurane interfered with ACh and NO relaxations and with NO-stimulated increases in cGMP. Halothane was more potent, having significant attenuating effects at 0.34 mM (1 MAC) and 0.72 mM (2 MAC) while isoflurane had effects only at 0.53 mM (2 MAC). For the isolated sGC studies, a soluble liver fraction was prepared from perfused rat livers. In the absence of NO stimulation, neither halothane nor isoflurane modified the activity of the sGC. However, during NO-stimulation halothane produced significant, concentration-dependent, inhibition of sGC activity over a wide range of NO concentrations. Isoflurane also inhibited sGC activity, but to a lesser extent than halothane. The mechanism whereby the anesthetics could interfere with sGC from liver and blood vessels is unknown. It could result from anesthetic interaction at hydrophobic sites that may exist in GC. However, the results of both the aorta and liver sGC enzyme studies support the suggestion that these anesthetics can compete with NO for its binding site on the ferrous heme of sGC, with chemical structural differences accounting for the potency variations. Both anesthetics also had cGMP independent effects, causing concentration dependent relaxations of NE-contracted vessels without endothelium. Isoflurane was about 5 times more effective at 1 MAC than halothane. Therefore, the net effects of these anesthetics involve the sum of two opposite effects on tension of vessels with intact endothelium: 1) interference with NO-stimulated cGMP relaxation and 2) direct stimulation of relaxation (not dependent on changes in cGMP).

Time course of 72-kilodalton heat shock protein induction and appearance of trifluoroacetyl adducts in livers of halothane-exposed rats.

Previous studies have shown that exposure of phenobarbital-pretreated rats to halothane in 10% O2 causes centrilobular necrosis, induces expression of the 72-kDa heat shock protein (HSP72), and produces several trifluoroacetylated adducts. In the present study the time course of development of the centrilobular lesion, as measured by histochemistry, was compared with the time course of appearance of both trifluoroacetylated adducts and HSP72, as measured by Western blotting. One group of 20 rats was pretreated with phenobarbital for 5 days, whereas a second group of two rats was left as untreated controls. Ten phenobarbital-pretreated rats were exposed for 2 hr to 1% halothane in 10% O2 and 10 were exposed to 1% halothane in 20% O2. At either 2, 4, 6, or 24 hr after exposure, livers were excised and frozen without fixation. Thin sections stained with hematoxylin and eosin demonstrated that centrilobular lesions occurred at 6 hr and became extensive at 24 hr in rats pretreated with phenobarbital and exposed to 1% halothane in 10% O2. The time course of appearance of both trifluoroacetylated adducts and HSP72 was determined by Western blotting. Trifluoroacetylated adducts appeared in all rats exposed to halothane by 2 hr, lasted until 6 hr, and then diminished by 24 hr. In contrast, HSP72 was induced only in the rats pretreated with phenobarbital and exposed to 1% halothane in 10% O2. HSP72 appeared in both the nuclear and supernatant fractions at 6 hr after exposure and was intense 24 hr after exposure.

Nuclear translocation of heat shock protein 72 in liver cells of halothane-exposed rats.

Immunocytochemical studies have revealed that one of the major heat shock proteins, HSP72, is induced in livers of rats that have been pretreated with phenobarbital and then administered halothane in a hypoxic gas mixture of 10% oxygen. To determine the sub-cellular localization of HSP72 in the livers of these rats 24 hr after halothane administration, cytoplasmic and nuclear fractions were prepared and separated by PAGE electrophoresis. Western blotting with a mouse monoclonal anti-HSP70 IgG antibody, which recognizes both the constitutive (HSP73) and inducible (HSP72) forms, revealed that HSP72 was induced and translocated into the nucleus in only those rats exposed to halothane under hypoxia following phenobarbital pretreatment. Nuclear translocation of HSP72 under the latter conditions was confirmed by immunocytochemical staining using gold-conjugated secondary antibodies followed by digital laser microscopy with Nomarski optics. Neither phenobarbital pretreatment alone nor phenobarbital plus hypoxia treatment induced HSP72. No alteration in amount of HSP73 was observed under any of these conditions.

Effect of halothane on hypoxic toxicity and glutathione status in cultured rat hepatocytes.

BACKGROUND: In hypoxic rats, halothane causes hepatotoxicity at oxygen levels that would cause minimal hepatotoxicity in the absence of halothane. Using a model that excludes systemic and extrahepatic effects of halothane, the authors tested the hypothesis that halothane hepatotoxicity in the whole-rat model is caused by a direct hepatotoxic effect of halothane, which is mediated by halothane-derived free radicals. METHODS: Rat hepatocyte monolayer cultures were exposed to defined gas phases for 2 h. Three experimental variables were present or absent: hypoxia (1% O2), halothane (2%), and cytochrome P-450 induction (by phenobarbital). Two experimental outcomes were measured: aspartate aminotransferase release, a measure of cell death, and reduced glutathione, an endogenous free radical scavenger whose levels are decreased by physiologically significant free radical injury. RESULTS: As anticipated, hypoxia increased cell death. Cytochrome P-450 induction by itself increased cell death during hypoxia. However, halothane had no effect on cell death during hypoxia, with or without cytochrome P-450 induction. Halothane had no toxic effect, even when glutathione was depleted before the onset of hypoxia. Glutathione was decreased moderately by hypoxia alone. Neither halothane nor cytochrome P-450 induction had any effect on glutathione levels. CONCLUSIONS: Halothane was not toxic, and it did not generate a physiologically significant free radical insult during hypoxia in the isolated rat hepatocyte under the experimental conditions used in testing.

Effects of halothane on EDRF/cGMP-mediated vascular smooth muscle relaxations.

BACKGROUND: Halothane has been reported to inhibit endothelium-dependent relaxation in a variety of vessels. These studies were done to determine whether this inhibition is caused by interference with synthesis, release, or action of endothelium-derived relaxing factor (EDRF) on cyclic guanosine monophosphate (cGMP) levels within the vascular smooth muscle. METHODS: Rat aortic rings were suspended in aerated Krebs solution (37 degrees C) and were contracted to a stable plateau with EC60-70 norepinephrine (NE). Relaxations caused by acetylcholine (ACh; 1 x 10(-8)-1 x 10(-6) M), nitric oxide (NO; 5 x 10(-9)-1 x 10(-6) M), or nitroglycerin (NG; 2 x 10(-9)-3 x 10(-7) M) in rings contracted with NE were compared in the presence and absence of halothane. Tissue cGMP contents were measured using a radioimmunoassay method. RESULTS: In the presence of halothane (0.5, 1.0, and 2.0 MAC), the ACh-induced relaxations were significantly attenuated in a concentration-dependent manner, an effect that was reversible. Halothane (2 MAC) significantly attenuated NO-induced relaxations at all concentrations and NG-induced relaxations at low concentrations (5 x 10(-9)-3 x 10(-8) M) but not at higher concentrations (1 x 10(-9)-3 x 10(-7) M) in denuded vessels. Nitric oxide-stimulated (5 x 10(-8)-5 x 10(-6) M) cGMP content was significantly attenuated by halothane (2 MAC) at NO concentrations between 1 x 10(-7) and 5 x 10(-6) M. CONCLUSIONS: Nitric oxide, either endogenous or exogenous, interacts with the enzyme guanylate cyclase to stimulate the production of cGMP. Halothane interfered with the relaxations caused by NO (in rings without endothelium) and decreased the NO-stimulated cGMP content. These results suggest that the site of action of halothane in attenuating endothelium-dependent relaxation in the rat aorta is within the vascular smooth muscle, rather than on the synthesis, release, or transit of the EDRF from the endothelium and that its action may involve an interference with guanylate cyclase activation.

The volatile anesthetic isoflurane attenuates Ca++ mobilization in cultured vascular smooth muscle cells.

Isoflurane is a volatile anesthetic which decreases vascular tone. Experiments were designed to determine whether isoflurane attenuated agonist-induced signaling in cultured vascular smooth muscle cells (A7r5). Cells were preincubated for 15 to 20 min with clinically relevant concentrations of isoflurane--0.5 to 2% in the gas phase and stimulated with 10(-9) or 10(-7) M vasopressin or with 3.3 x 10(-9) M platelet-derived growth factor. The two agonists are believed to act via differing signaling pathways. Total inositol phosphate formation was measured by column chromatography. Apparent intracellular free Ca++ concentration (1) [Ca++]i was estimated using indo-1 and flow cytometry. Isoflurane attenuated increases in [Ca++]i evoked by both agonists. Isoflurane 2.0% inhibited [Ca++]i responses evoked by vasopressin by 35 to 41%. Responses due to Ca++ release from intracellular stores were particularly sensitive to inhibition by isoflurane. The anesthetic attenuated inositol phosphate generation evoked by vasopressin and platelet-derived growth factor, suggesting a mechanism for isoflurane action on Ca++ release. Surprisingly, the anesthetic only modestly inhibited increases in [Ca++]i due to Ca++ entry. Isoflurane's effect on Ca++ influx after emptying of Ca++ stores was probed using thapsigargin. Inhibition of Ca++ influx was modest. It is suggested that isoflurane attenuates total inositol phosphate formation and Ca++ release evoked by vasopressin and platelet-derived growth factor while having limited effects on agonist-induced Ca++ entry.

The effects of sevoflurane on intracellular Ca2+ regulation in rat hepatocytes.

The effects of sevoflurane, a new volatile anesthetic agent undergoing clinical trial, on the mobilization of intracellular Ca2+ in isolated rat hepatocytes was studied. This agent produced a dose-dependent release of 45Ca2+ from internal, non-mitochondrial stores of permeabilized hepatocytes (saponin treated). However, the administration of sevoflurane to aequorin-loaded intact hepatocytes had little or no effect on intracellular [Ca2+] (i.e., short transient or no increases in luminescence: no toxic effect). These data may indicate that because of the low solubility of sevoflurane, it has a selective effect on endoplasmic reticulum, i.e., mobilizing internal stores of Ca2+ relative to increasing transmembrane fluxes.

Immunocytochemical detection of the 72-kDa heat shock protein in halothane--induced hepatotoxicity in rats.

Liver sections removed from phenobarbital induced rats 24 to 48 hours after a 2 hour exposure to 1.0% halothane with 10% oxygen and subjected to immunocytochemical treatment showed evidence of centrilobular damage as well as evidence of the production of a protein which has immunoreactivity with anti HSP 72 antibodies. The cells showing evidence of immunoreactivity were within the area of the centrilobular lesion. The level of immunoreactive protein varied directly with the intensity of the lesion. Liver sections from animals treated with phenobarbital alone, phenobarbital plus 10% oxygen, or phenobarbital plus 20% oxygen and 1.0% halothane all were without lesions as well as the immunoreactive protein.

Transient increases of intracellular Ca2+ induced by volatile anesthetics in rat hepatocytes.

The affects of volatile anesthetics on mobilization of intracellular Ca2+ was monitored in primary cultures of rat hepatocytes using the fluorescent Ca2+ probe Fura-2. The use of Fura-2 was limited by several factors which complicated the quantitative analysis of the results, such as: (i) a high rate of dye leakage; (ii) changes in the redox state of the hepatocytes which interfered with the fluorescence produced by the dye at various excitation wavelengths; (iii) compartmentalization of the dye producing high local intracellular concentrations; and, of particular importance for this study, (iv) enhanced photobleaching of the dye in the presence of halothane. To aid in the interpretation of the Fura-2 data, the Ca2(+)-sensitive photoprotein aequorin was also used to monitor changes in [Ca2+]i. The aequorin and Fura-2 techniques qualitatively yielded the same result, that the volatile anesthetic agents halothane, enflurane, and isoflurane induce an immediate and transient increase of [Ca2+]i. The durations of these transients were approximately between 5 and 10 min and were not related to any evident acute cell toxicity. The [Ca2+]i increases induced by the volatile anesthetic agents were dose-dependent, with halothane the most potent. The exact mechanism governing these increases in [Ca2+]i induced by these anesthetics in rat hepatocytes is unknown, but is likely to involve effects on both the cell surface membrane and endoplasmic reticulum components of the signal transducing system.

The effects of volatile anesthetics on Ca++ mobilization in rat hepatocytes.

This study provides direct evidence that in hepatocytes, intracellular Ca++ is released from internal stores by halothane, enflurane, and isoflurane. Hepatocytes isolated from rat livers were used fresh or treated with saponin and then incubated in 45Ca++ media. The uptake of 45Ca++ by hepatocytes was maximal following 13-16 min of incubation (untreated or saponin-treated) and the effects of various agents on the release of 45Ca++ was studied following maximal loading. The agents used included halothane, enflurane, isoflurane, and several putative intracellular second messengers. The anesthetics, to various degrees, all stimulated a significant release of 45Ca++ from internal stores at concentrations that were at or less than clinical concentrations. The release of intracellular 45Ca++ by each of the anesthetic agents was dose-dependent with halothane and enflurane being equally potent at concentrations equivalent to 1 MAC exposure. The halothane-induced release was only somewhat suppressed by preincubation in either 2 mM LaCL3 or 10 microM dantrolene, both suggested Ca++ channel blockers. Transient increases in intracellular Ca++ regulates a number of enzyme systems, including glycogenolysis, while prolonged elevation in Ca++ concentrations have been implicated in the mechanism of hepatotoxicity.

Purification and characterization of human kidney cytosolic cysteine conjugate beta-lyase activity.

The central role of cysteine conjugate beta-lyase (beta-lyase) in the bioactivation of nephrotoxic halogenated hydrocarbons and the possibility of human exposure to these chemicals has focused interest on the beta-lyase from human kidney. Human kidney tissue was collected as surgical waste material, and subcellular fractions were isolated by differential centrifugation. Human beta-lyase activity, determined with S-(2-benzothiazolyl)-L-cysteine (BTC) as the substrate, was present in the cytosolic, mitochondrial, and microsomal fractions, but was highest in the cytosolic fraction. Activity in human kidney cytosol was about 10% of that present in rat kidney cytosol. Human kidney cytosolic beta-lyase activity, with BTC as the substrate, was not stimulated by pyridoxal phosphate or by exogenous 2-keto acids. Cytosolic beta-lyase activity was purified 280-fold with a yield of 12% from human kidneys unsuitable for transplantation. The beta-lyase activity copurified with cytosolic glutamine transaminase K and exhibited a molecular mass of 85 kDa on a Sephacryl 5300 column and a subunit molecular mass of 45 kDa by gel electrophoresis. Both BTC and its homocysteine analogue S-(2-benzothiazolyl)-L-homocysteine were excellent substrates, exhibiting Km and kcat values of 0.97 mM and 2.78 mM and 9.35 min-1 and 6.90 min-1, respectively. beta-Lyase activity was inhibited by aminooxyacetic acid, indicating that the human cytosolic enzyme contains pyridoxal phosphate, and by the nephrotoxins S-(1,2-dichlorovinyl)-L-cysteine and S-(1,2-dichlorovinyl)-L-homocysteine, which serve as alternative substrates.(ABSTRACT TRUNCATED AT 250 WORDS)

Comparison of analgesic requirements after liver transplantation and cholecystectomy.

In a prospective study of 10 patients who underwent liver transplantation and 10 patients who underwent cholecystectomy, we analyzed the postoperative analgesic requirements and the resultant plasma morphine concentrations. Analgesia was more intense, with less medication, and the plasma morphine concentration was significantly lower in the liver transplant group than in the cholecystectomy group. This finding is most likely attributable to endogenous factors rather than to altered morphine pharmacokinetics.

Anesthesia approach to hepatic transplantation.

Anesthesia support for patients undergoing orthotopic liver transplantation can be complicated because of multiple medical problems in such patients and rapid hemodynamic, metabolic, and coagulation changes intraoperatively. Preoperative assessment should include careful review of the cardiovascular, respiratory, and hematologic systems. Use of isoflurane as the main anesthetic agent will minimize toxicity to the liver. During liver transplantation, hemodynamic monitoring and immediate laboratory studies should be available. In our experience during the first 100 liver transplantations performed at our institution, use of a rapid infusion pump and venovenous bypass has helped normalize hemodynamic and renal function.

Hemodynamic and metabolic changes in hepatic transplantation.

In this study, we retrospectively analyzed the intraoperative hemodynamic, laboratory, and coagulation data on the first 83 patients who underwent an initial liver transplantation procedure at our institution. The major hemodynamic changes at the time of reperfusion of the donor liver were significant decreases in arterial blood pressure, systemic vascular resistance, and pulmonary artery temperature and significant increases in cardiac output and pulmonary capillary wedge pressure. The alterations in laboratory values reflected intraoperative therapeutic manipulations. Citrate toxicity is a concern, and the amount of calcium chloride administered reflected the volume of blood transfused. On reperfusion, the fibrinogen concentration decreased and both the prothrombin time and the activated partial thromboplastin time increased. This coagulopathy was also evident in the thromboelastographic values. Aggressive monitoring and prompt intervention are necessary to maintain hemodynamic and metabolic homeostasis in these patients.

Resonance Raman study of the cytochrome P-450 LM2-halothane intermediate complex.

Resonance Raman (RR) and absorption spectroscopic studies of purified rabbit liver cytochromes P-450 show that the form 2 isomer (LM2) but not the form 4 isomer (LM4) forms a long-lived complex with halothane after dithionite reduction, absorbing light at 470 nm, in which ferric 6-coordinated heme iron in the low-spin configuration is liganded to 2-chloro-1,1-difluoroethylene. The RR data exclude the possibility that the CF3CHCl- carbanion is a ligand and are consistent with the involvement of an active-site pocket in the cytochrome P-450 polypeptide.

Reductive metabolism of halothane by purified cytochrome P-450.

The reductive metabolism of halothane was determined using purified RLM2, PBRLM4 and PBRLM5 forms of rat liver microsomal cytochrome P-450. The metabolites, 2-chloro-1,1,1-trifluoroethane (CTE) and 2-chloro-1,1-difluoroethylene (CDE), were determined. All three forms of cytochrome P-450 produced CTE with relatively small differences in its production among the various forms. There were major differences, however, in the production of CDE, with PBRLM5 being the most active. PBRLM5 was also the only form to show the development of a complex between halothane and cytochrome P-450. This complex absorbed light maximally at 470 nm. The complex formation and the production of CDE by PBRLM5 were stimulated by the addition of cytochrome b5. Cytochrome b5 had no effect on CDE production by PBRLM4 and inhibited the production of both CTE and CDE by RLM2. These results show that the two-electron reduction of halothane by cytochrome P-450 was catalyzed by the PBRLM5 form and that cytochrome b5 stimulated the transfer of the second electron to halothane through PBRLM5, but not RLM2 or PBRLM4.

Isoflurane causes endothelium-dependent inhibition of contractile responses of canine coronary arteries.

The authors sought to determine if isoflurane would attenuate effects of three different types of vasoconstrictors on isolated segments of canine epicardial coronary arteries removed from healthy dogs. As the endothelium has a major role in regulating epicardial coronary artery tone, and as it modulates the effect of many vasoactive substances, experiments were conducted both on normal rings and on rings whose endothelium had been mechanically removed. In addition, the endothelium is thought to be damaged in human atherosclerosis. Rings were suspected in organ chambers filled with modified Krebs-Ringer bicarbonate solution, aerated with 95% oxygen and 5% carbon dioxide, and connected to strain gauges for the measurement of isometric tension. Isoflurane 2.3% (1.5 MAC in the dog) was added to the aerating gas mixture in half the preparations, while the other rings served as control. The vasoconstrictors serotonin, phenylephrine, or prostaglandin F2 alpha were added in increasing concentrations to the bath solution. In the presence of endothelium, vasoconstrictor evoked contractions were attenuated by isoflurane. Maximal tension generated by prostaglandin F2 alpha in untreated rings was 114 +/- 18% (mean +/- SEM) of a reference contraction, while, following isoflurane, it was 46 +/- 8% (P less than 0.005). In the absence of endothelium, isoflurane attenuated neither prostaglandin F2 alpha nor serotonin evoked contraction, and had decreased effectiveness against phenylephrine mediated contraction (P less than 0.001). It is concluded that isoflurane attenuates vasoconstrictor-evoked contraction of isolated canine epicardial coronary arteries, and that this effect is mediated by the endothelium.

Human liver and conjugation of catecholamines.

To investigate the role of the liver in conjugation of catecholamines we measured the concentrations of free and conjugated norepinephrine, epinephrine, and dopamine in plasma of patients with severe liver disease who were undergoing liver transplantation. Comparisons were made with catecholamine levels in plasma of euhepatic patients who were undergoing abdominal aortic aneurysmectomy. We were also able to determine the importance of the liver in conjugation of exogenous dopamine because this compound was given to both groups of patients. The concentrations of conjugated amines were within the normal range in the patients undergoing liver transplantation, and administered dopamine was conjugated to a similar extent in the two groups of patients. The data suggest that the liver is not indispensable for the conjugation of circulating catecholamines.