The effect on the recovery profile of a change from enflurane to desflurane during the latter part of anaesthesia.

This study compared emergence and recovery characteristics after either enflurane anaesthesia or crossover from enflurane to desflurane anaesthesia. At an estimated 1 h prior to the end of operation, enflurane was either reduced (group E, n = 23) or replaced with desflurane (group X, n = 23). At the end of the operation, emergence and recovery characteristics of the two groups were compared. The crossover technique accelerated recovery compared with enflurane anaesthesia. The time taken for the eyes to open in response to painful pinching or a verbal command, and to regain awareness of age and name, were significantly shorter after crossover anaesthesia than after enflurane anaesthesia. The digit symbol substitution test and serial seven test scores were significantly better in patients subjected to crossover anaesthesia than in those subjected to enflurane anaesthesia. We conclude that, during surgery, the substitution of enflurane with desflurane in the latter part of anaesthesia can improve recovery.

Weak polar interactions confer albumin binding site selectivity for haloether anesthetics.

BACKGROUND: Enflurane and isoflurane are structural isomers with different anesthetic potencies and side effects. It is not clear whether these differences are produced by differing occupancy of common protein binding sites or by occupancy of different sites, but the very similar molecular properties make the latter possibility unlikely. In this study, the authors examined binding site selectivity of these anesthetics in human serum albumin (HSA). METHODS: Binding of isoflurane and enflurane with HSA was determined with isothermal titration calorimetry. Competition with known ligands (propofol) allowed localization of binding sites within the HSA molecule. Molecular properties of isoflurane and enflurane were calculated. RESULTS: Isoflurane binds HSA with higher affinity but smaller total enthalpy than enflurane. Enthalpogram analysis suggested that isoflurane bound a single site, whereas enflurane bound two. Competition experiments indicated that enflurane and isoflurane share one binding site, which also binds propofol. The additional enflurane site binds propofol but not isoflurane. Increased salt concentration decreased the affinity for isoflurane but not for enflurane. The dipole moment of isoflurane is higher than that of enflurane, and the isoflurane binding site is more polar. CONCLUSION: These data indicate two binding sites of different character for the haloether anesthetics on HSA. One site is more polar and prefers isoflurane, presumably because of its larger dipole. The second site prefers the less polar enflurane. Therefore, weak polar interactions confer considerable selectivity, and differences in drug action may arise from occupancy of different protein sites.

Fluoride metabolism in smokers and non-smokers following enflurane anaesthesia.

BACKGROUND: Inorganic fluoride is released by the metabolism of enflurane and the increased serum fluoride concentrations may impair renal function. Tobacco smoke consists of numerous reactive compounds that can either induce or inhibit drug metabolism. Studies on the interaction of smoking with anaesthetic drug metabolism and possible toxicity are warranted. METHODS: Sixteen non-smoking and 17 smoking (>10 cigarettes day(-1)) generally healthy women undergoing elective gynaecological surgery were given 1 MAC (minimum alveolar concentration)-hour standardized anaesthesia with enflurane in oxygen-air mixture. The serum inorganic fluoride and renal function markers beta(2)-microglobulin, tumour-associated trypsin inhibitor (TATI) and serum creatinine were measured for 48 h. RESULTS: The greatest inorganic fluoride concentration was between 8.4 and 21.0 (mean 13.8 (SD 3.4)) micromol litre(-1) in the non-smokers and between 8.6 and 38.0 (18.7 (7.0)) micromol litre(-1) in the smokers; the mean difference was 4.9 micromol litre(-1) (95% confidence interval (CI) 1.0-8.8, P<0.05). Serum beta(2)-microglobulin, TATI and creatinine were not increased. Serum inorganic fluoride concentrations were significantly greater in the smokers compared with the non- smokers 1, 2, 3 and 6 h after 1 MAC-hour inhalation with enflurane (P<0.05). Inorganic fluoride concentrations were still increased 24 h after anaesthesia in both groups. Urine beta(2)-microglobulin and TATI creatinine ratio remained at low values during the whole 48-h period in both groups. CONCLUSIONS: Regular smoking is associated with an increase in serum inorganic fluoride concentration after anaesthesia with enflurane, but there are no signs of renal damage.

Antagonism of inhalant and volatile anesthetic enhancement of glycine receptor function.

Recent studies suggest that alcohols, volatile anesthetics, and inhaled drugs of abuse, which enhance gamma-aminobutyric acid, type A, and glycine receptor-activated ion channel function, may share common or overlapping molecular sites of action on these receptors. To investigate this possibility, these compounds were applied singly and in combination to wild-type glycine alpha(1) receptors expressed in Xenopus laevis oocytes. Data obtained from concentration-response curves of the volatile anesthetic enflurane constructed in the presence and absence of ethanol, chloroform, or toluene were consistent with competition for a common binding pocket on these receptors. A mutant glycine receptor, insensitive to the enhancing effects of ethanol but not anesthetics or inhalants, demonstrated antagonism of anesthetic and inhalant effects on this receptor. Although ethanol (25-200 mm) had no effect on its own in this receptor, it was able to inhibit reversibly the enhancing effect of enflurane, toluene, and chloroform in a concentration-dependent manner. These data suggest the existence of overlapping molecular sites of action for ethanol, inhalants, and volatile anesthetics on glycine receptors and illustrate the feasibility of pharmacological antagonism of the effects of volatile anesthetics.

Modulation of cytochrome P-450 dependent monooxygenases in streptozotocin-induced diabetic hamster: I. Effects of propofol on defluorination and cytochrome P-450 activities.

BACKGROUND: Diabetes mellitus could induce polymorphic alterations of metabolic activities of cytochrome P-450 dependent monooxygenases in chemical-induced diabetic animals. The purpose of this study is to define the functional impact of clinical concentrations of propofol on the metabolic activities of cytochrome P-450 in the diabetic animals. METHODS: In order to validate the effect of propofol on cytochrome P-450 activities, especially the cytochrome P-450 2E1 and its defluorination activity, we applied NADPH-generating system to measure the metabolizing activities of cytochrome P-450 isozymes of streptozotocin-induced diabetic hamsters within the microsomes preincubated with various concentrations of propofol. The extent of defluorination and activity of cytochrome P-450 2E1 were assessed by reacting the propofol-treated microsomes in NADPH-generating system with enflurane and aniline as substrates respectively. Drug metabolizing activities of cytochrome 1A1, 2B1, and 3A4 were evaluated by metabolizing specific substrates, benzo(a)pyrene, pentoxyresorufin and erythromycin, within the microsomes of diabetic hamsters preincubated with various concentrations of propofol. RESULTS: The hepatic and renal defluorination of enflurane was significantly inhibited by 0.05 and 0.10 mM propofol in the microsomes of diabetic hamster (P < 0.05). The activities of aniline hydroxylase (cytochrome 2E1), pentoxyresorufin O-dealkylase (cytochrome 2B1) and benzo(a)pyrene hydroxylase (cytochrome 1A1) were inhibited by propofol in a concentration-dependent manner from 0.05 to 0.10 mM. However, propofol showed no significant effect to the erythromycin N-demethylase (cytochrome 3A4) at its concentration of 0.05-0.10 mM in the diabetic hamsters. CONCLUSIONS: Our data demonstrated that propofol in therapeutic concentrations of 0.05 and 0.10 mM, could inhibit both liver and kidney defluorination and cytochrome P-450's activities of the diabetic hamsters in vitro of different extent. This should remind clinicians of propofol's potential drug-to-drug interactions in the diabetic patients.

Modulation of cytochrome P450-dependent monooxygenases in streptozotocin-induced diabetic hamster: II. Reverse role of insulin in P450 activity and defluorination.

BACKGROUND: Metabolic activities of cytochrome (cyt) P450-dependent monooxygenase could be modulated by diabetic state in experimental diabetic animals. The purpose of this study is to validate the effect of insulin on the modulation of the metabolic activity of cyt P450 and the defluorination ability to inhalational anesthetics in diabetic animals. METHODS: Diabetic state in golden Syrian hamsters was achieved by intraperitoneal injection of streptozotocin 40 mg/kg once a day for 4 days. After stabilization of diabetic state for 6 weeks, a regimen of insulin treatment given subcutaneously was carried out. Metabolic activities of cyt P450 were assessed by the reaction with benzo(a) pyrene, pentoxyresorufin, aniline and erythromycin (specific substrates). The metabolic activities of cyt 1A1, 2B1, 2E1 and 3A4 respectively in a NADPH-generating system in microsomal preparations of the diabetic hamsters were observed before and after insulin treatment, and were compared with the control group. The ability of defluorination was evaluated by measuring the free fluoride metabolites after incubating the microsomes with enflurane in diabetic and insulin-treated hamsters. Contents of cyt P450 isozymes were measured by electrophoresis and immunoblotting before and after insulin treatment. Pathological features of hepatocytes in diabetic hamsters were evaluated microscopically before and after insulin treatment. RESULTS: The defluorination of enflurane and activity of aniline hydroxylase (cyt 2E1) were successfully induced by diabetic state (P < 0.01). The pentoxyresorufin O-dealkylase (cyt 2B1) was inhibited nearly 50% in the diabetic hamster liver when compared with that of control (P < 0.01). While the activities of benzo(a)pyrene hydroxylase (cyt 1A1) and the erythromycin N-demethylase (cyt 3A4) were basically unaffected by diabetes, alterations in content of cyt P450 were parallel to the alterations in enzyme activities. Microscopically, diabetes induced vacuolization with fatty droplets in the hepatocytes. After treatment with insulin injection, the enzyme activities, protein content and pathologic features returned to the baseline similar to the control. CONCLUSIONS: Our data demonstrated that under diabetic state, metabolic activities of cyt P450 and its extent of defluorination would be polymorphically modulated. After administration of insulin, the activities of cyt P450 and defluorination of enflurane returned to baseline as the blood sugar level had been normalized. This could remind the clinicians of the importance of insulin treatment in the potential drug-to-drug interactions in the diabetic patients.

Propofol inhibits renal cytochrome P450 activity and enflurane defluorination in vitro in hamsters.

PURPOSE: To determine the effect of propofol on renal cytochrome P450 activity and defluorination of enflurane. METHODS: Renal microsomes were prepared by homogenization and differential centrifugation from pooled hamster kidneys. Defluorination of enflurane was assessed by measuring free fluoride metabolites after reacting enflurane with renal microsomes incubated with various concentrations, 0.05 - 1.0 mmol x L(-1) propofol in the NADPH-generating system. Drug metabolizing activities of renal cytochrome P450 mono-oxygenase enzymes were evaluated within microsomes preincubated with propofol and reacted with the specific marker substrates, aniline, benzo(a)pyrene, erythromycin and pentoxyresorufin, for cytochrome P450 2E1, 1A1, 3A4 and 2B1, respectively. RESULTS: Renal defluorination of enflurane was inhibited by clinical concentrations, 0.05 mmol x L(-1) of propofol (P < 0.05). Dose-dependent inhibition of defluorination, aniline and benzo(a)pyrene hydroxylase within kidney microsomes was related to propofol concentration. Propofol demonstrated a profound inhibition of renal pentoxyresorufin dealkylase activity even at low concentrations, 0.05 mmol x L(-1) (P < 0.01). Propofol did not exhibit inhibition of erythromycin N-demethylation of kidney microsomes except at high concentration, 1.0 mmol x L(-1). Spectral analyses of key coenzymes of renal cytochrome P450 monooxygenase, cytochrome b5 and cytochrome c reductase, demonstrated an inhibition when incubated with high concentrations of propofol (P < 0.05). CONCLUSION: In an in vitro study in an NADPH-generating system of hamster kidney microsomes, propofol, in clinical concentrations, exhibited a broad-spectrum of inhibition to renal monooxygenase activities and enflurane defluorination.

Metabolic characteristics and enflurane defluorination of cytochrome P450-dependent monooxygenases in human hepatocellular carcinoma.

BACKGROUND: Xenobiotic metabolism and defluorination capacity of microsomal monooxygenases were investigated in vitro through the surgical specimens of liver resected from patients with hepatocellular carcinoma and patients of extrahepatic pathology as control. METHODS: In microsomes of hepatocellular carcinoma tissues, the activities of cytochrome P450-dependent monooxygenase isoenzymes 1A1, 2B1, and 2E1 were evaluated in vitro by reacting with the specific marker substrates benzo(a)pyrene, benzphetamine and aniline, respectively, in the generating incubation system. The distant normal liver tissues and tissues from control patients with extrahepatic lesion were also investigated for comparison. The ability of enflurane defluorination was assessed by Orion combined for detection of free fluoride ion production. RESULTS: Concentrations of P450 total content, cytochrome b5, and NADPH-cytochrome c reductase showed parallel and marked reduction in tumor tissues when compared with its distant normal regions or normal livers. The monooxygenase functions displayed significant decreases within the tumor tissues as benzo(a)pyrene hydroxylation > or = benzphetamine demethylation > aniline hydroxylation in magnitude. Defluorination of enflurane also markedly decreased in tumor tissues comparing with normal livers. CONCLUSIONS: These marked reductions in the compositions and in vitro metabolic activities, including defluorination of anesthetics, in the cytochrome P450-dependent monooxygenases within the tumor tissues characterize the unique pattern of xenobiotic metabolism in patients with hepatocellular carcinoma.

Mass spectrometry provides warning of carbon monoxide exposure via trifluoromethane.

BACKGROUND: The chemical breakdown of isoflurane, enflurane, or desflurane in dried carbon dioxide absorbents may produce carbon monoxide. Some mass spectrometers can give false indications of enflurane during anesthetic breakdown. METHODS: During clinical anesthesia with isoflurane or desflurane, the presence of carbon monoxide in respiratory gas was confirmed when enflurane was inappropriately indicated by a clinical mass spectrometer that identified enflurane at mass to charge ratio = 69. In vitro, isoflurane, enflurane, or desflurane in oxygen was passed through dried carbon dioxide absorbents at 35, 45, and 55 degrees C. Gases were analyzed by gas chromatography and by mass spectrometry. RESULTS: Mass spectrometry identified several clinical incidents in which 30-410 ppm carbon monoxide was measured in respiratory gas. Trifluoromethane was produced during in vitro breakdown of isoflurane or desflurane. Although these inappropriately indicated quantities of "enflurane" correlated (r2 > 0.95) to carbon monoxide concentrations under a variety of conditions, this ratio varied with temperature, anesthetic agent, absorbent type, and water content. CONCLUSIONS: Trifluoromethane causes the inappropriate indication of enflurane by mass spectrometry, and indicates isoflurane and desflurane breakdown. Because the ratio of carbon monoxide to trifluoromethane varies with conditions, this technique cannot be used to quantitatively determine the amount of carbon monoxide to which a patient is exposed. If any warning of anesthetic breakdown results from this technique then remedial steps should be taken immediately to stop patient exposure to carbon monoxide. No warning can be provided for the breakdown of enflurane by this technique.

Stereoselective metabolism of enflurane by human liver cytochrome P450 2E1.

The toxicity of the chiral fluorinated volatile anesthetics halothane, enflurane, and isoflurane is closely related to their metabolism by hepatic cytochrome P450. Although individual anesthetic enantiomers have been shown to exhibit a difference in anesthetic efficacy, metabolism of anesthetic enantiomers has not been reported. Human liver enflurane metabolism to difluoromethoxydifluoroacetic acid (DFMDFA) and inorganic fluoride is catalyzed in vivo and in vitro by cytochrome P450 2E1. The purpose of this investigation was to characterize enflurane racemate and enantiomer metabolism to test the hypothesis that fluorinated ether anesthetic metabolism by cytochrome P450 2E1 exhibits substrate stereoselectivity. Enflurane metabolism by microsomes from three human livers and by microsomes containing cDNA-expressed human P450 2E1 was measured at saturating enflurane concentrations. DFMDFA was quantitated with gas chromatography-mass spectrometry by detection of the ethanolamide derivative. In microsomes from all three livers, (R)-enflurane metabolism was significantly greater than that of (S)-enflurane, whereas rates of racemic enflurane metabolism were generally between those seen for the R- and S-enantiomers. The ratio of (R)-enflurane to (S)-enflurane metabolism in the three livers studied was 2.1:1, 1.9:1, and 1.4:1. (R)-, (S)-, and racemic enflurane were all metabolized by expressed P450 2E1. The ratio of (R)-enflurane to (S)-enflurane metabolism was 1.9:1. The metabolic enantiomeric selectivity of human liver P450 2E1 for (R)-enflurane suggests that enflurane metabolism by P450 2E1 occurs by direct substrate oxidation, rather than indirectly through the generation of a P450-dependent reactive oxygen species, and supports the hypothesis that the P450 2E1 active site is somewhat restrictive and capable of stereochemical discrimination.

Halogenated anesthetics form liver adducts and antigens that cross-react with halothane-induced antibodies.

Two halogenated anesthetics, enflurane and isoflurane, have been associated with an allergic-type hepatic injury both alone and following previous exposure to halothane. Halothane hepatitis appears to involve an aberrant immune response. An antibody response to a protein-bound biotransformation product (trifluoroacetyl adduct) has been detected on halothane hepatitis patients. This study was performed to determine cross-reactivity between enflurane and isoflurane with the hypersensitivity induced by halothane. The subcellular and lobular production of hepatic neoantigens recognized by halothane-induced antibodies following enflurane and isoflurane, and the biochemical nature of these neoantigens was investigated in two animal models. Enflurane administration resulted in neoantigens detected in both the microsomal and cytosolic fraction of liver homogenates and in the centrilobular region of the liver. In the same liver, biochemical analysis detected fluorinated liver adducts that were up to 20-fold greater in guinea pigs than in rats. This supports and extends previous evidence for a mechanism by which enflurane and/or isoflurane could produce a hypersensitivity condition similar to that of halothane hepatitis either alone or subsequent to halothane administration. The guinea pig would appear to be a useful model for further investigations of the immunological response to these antigens.

Clinical enflurane metabolism by cytochrome P450 2E1.

BACKGROUND: Fluorinated ether anesthetic hepatotoxicity and nephrotoxicity are mediated by cytochrome P450-catalyzed oxidative metabolism. Metabolism of the volatile anesthetic enflurane to inorganic fluoride ion by human liver microsomes in vitro is catalyzed predominantly by the cytochrome P450 isoform CYP2E1. This investigation tested the hypothesis that P450 2E1 is also the isoform responsible for human enflurane metabolism in vivo. Disulfiram, which is converted in vivo to a selective inhibitor of P450 2E1, was used as a metabolic probe for P450 2E1. METHODS: Twenty patients undergoing elective surgery were randomized to receive disulfiram (500 mg orally; n = 10) or nothing (control subjects; n = 10) the evening before surgery. All patients received a standard anesthetic of enflurane (2.2% end-tidal) in oxygen for 3 hours. Blood enflurane concentrations were measured by gas chromatography. Plasma and urine fluoride concentrations were quantitated by ion-selective electrode. RESULTS: Patient groups were similar with respect to age, weight, gender, duration of surgery, and blood loss. Total enflurane dose, measured by cumulative end-tidal enflurane concentrations (3.9 to 4.1 MAC-hr) and by blood enflurane concentrations, was similar in both groups. Plasma fluoride concentrations increased from 3.6 +/- 1.5 mumol/L (baseline) to 24.3 +/- 3.8 mumol/L (peak) in untreated patients (mean +/- SE). Disulfiram treatment completely abolished the rise in plasma fluoride concentration. Urine fluoride excretion was similarly significantly diminished in disulfiram-treated patients. Fluoride excretion in disulfiram-treated patients was 62 +/- 10 and 61 +/- 12 mumol on days 1 and 2, respectively, compared with 1090 +/- 180 and 1200 +/- 220 mumol in control subjects (p < 0.05 on each day). CONCLUSIONS: Disulfiram prevented fluoride ion production after enflurane anesthesia. These results suggest that P450 2E1 is the predominant P450 isoform responsible for human clinical enflurane metabolism in vivo.

Anesthetic and hemodynamic interactions of dexmedetomidine and fentanyl in dogs.

BACKGROUND: Anesthetic doses of dexmedetomidine (DMED), a highly selective alpha 2 agonist, are not well tolerated hemodynamically. The combination of an opioid with DMED might reduce the dosage requirements for each drug and thereby allow the same anesthetic depth to be achieved with lesser degrees of their individual side effects. METHODS: Dogs were anesthetized with enflurane. One group (n = 5) received intravenous doses of DMED from 0.1 to 10 micrograms/kg. Two other groups of five dogs each received fentanyl 15 micrograms/kg plus 0.05 microgram.kg-1.min-1 or fentanyl 45 micrograms/kg plus 0.2 micrograms.kg-1.min-1. Thereafter, they received DMED doses of 0.03-3 micrograms/kg. After the effects of the last DMED dose were measured, atipamezole 0.3 mg/kg was infused intravenously and all measurements were repeated. Then, naloxone (1 mg/kg) was injected intravenously and a final set of measurements obtained. Anesthetic effects were assessed by determining enflurane minimum alveolar concentration (MAC). Hemodynamics and plasma fentanyl concentrations were measured at each determination of MAC. RESULTS: DMED and fentanyl individually produced dose-related reductions of enflurane MAC. During the lower rate infusion of fentanyl (plasma fentanyl concentration 1.0 +/- 0.3 ng/ml), DMED reduced enflurane MAC more than could be attributed to a simple additive interaction. During the higher rate infusion of fentanyl (plasma fentanyl concentration 4.4 +/- 0.7 ng/ml), DMED reduced enflurane MAC to greater degrees than were achievable by fentanyl alone. DMED caused a dose-dependent increase in arterial pressure concomitantly with a decrease in cardiac output, and these changes were not modified by fentanyl. The bradycardia following DMED was augmented by fentanyl. CONCLUSIONS: There was a positive interaction, additive or synergistic, between DMED and fentanyl with respect to their enflurane-sparing effects. The interaction allowed the same depth of anesthesia to be achieved by lower doses of all three drugs, potentially limiting the intensity of their individual side effects. However, the presence of fentanyl increased the degree of bradycardia induced by DMED.

Intrapulmonary gas mixing and pulmonary gas exchange in artificially ventilated dogs.

To investigate the effect of incomplete gas mixing between tidal air and residual gas on pulmonary gas exchange, anaesthetized dogs were ventilated artificially with breathing patterns with different durations of the post-inspiratory apnoea (ta = 0, 0.5, 1.0 and 2.0 s), where tidal volume, breathing frequency, inspiratory and expiratory flow patterns were kept constant. We determined the alveolar ventilations (VA) of He and SF6 from the product of end-expiratory lung volume (VL,E') and specific ventilation (VA/VL,E'). VL,E' was determined by the dilution technique and the specific ventilations of the two gases were obtained from their multiple-breath washout. Further, tracer amounts of acetone, ether and enflurane were infused continuously into a peripheral vein and a bolus of a gas mixture of krypton, Freon12 and SF6 was introduced into the peritoneal cavity. We determined the Excretion (E) and Retention (R) of these six gases according to the multiple-inert-gas-elimination technique (MIGET). VA increased with increasing ta, where VA,He was about 14% larger than VA,SF6. For both gases, however, the increase in VA relative to control (VA for ta = 0) was virtually the same: 9, 11 and 19% (mean values) for ta = 0.5, 1.0 and 2.0 s respectively. For all dogs the E/R curve shifted to larger E values with increasing ta. E for the most soluble tracer gas (acetone) increased by 11, 21 and 25% for ta = 0.5, 1.0 and 2.0 s respectively. VA, determined with MIGET from the ventilation/perfusion distribution, increased by almost the same percentages.(ABSTRACT TRUNCATED AT 250 WORDS)

Inhibitory effects of propofol on cytochrome P450 activities in rat hepatic microsomes.

The effects of propofol on cytochrome P450 activity in rat hepatic microsomes were evaluated to determine the potential influence of this anesthetic on the metabolism of coadministered agents. In microsomes from untreated and isoniazid-treated rats, propofol was a weak inhibitor of enflurane metabolism, inhibiting activity only at 0.35 mM propofol. In contrast, toluene, a related compound, effectively impaired enflurane defluorination in microsomes from untreated, and isoniazid- and phenobarbital-treated rats at concentrations as low as 0.025 mM. Propofol, in contrast to toluene, was an effective inhibitor of benzphetamine demethylation where it inhibited this activity at propofol concentrations as low as 0.025 mM in microsomes from phenobarbital-treated rats. In microsomes from phenobarbital-treated rats, propofol potently inhibited the metabolism of aniline. Sixty-four percent inhibition was achieved at 0.03 mM propofol, whereas toluene had no effect at 1 mM. These data demonstrate that propofol does not effectively inhibit enflurane metabolism performed by the isoniazid-inducible cytochrome P450IIE1 but effectively impairs activities of the phenobarbital-inducible cytochrome P450 isozymes.

Intrathecal morphine during thoracotomy, Part I: Effect on intraoperative enflurane requirements.

The ability of intrathecal morphine to reduce the anesthetic requirements during thoracotomy was investigated. Twenty-four patients scheduled for thoracic surgery were studied. Anesthesia was induced with thiamylal sodium, 4 mg/kg, fentanyl, 100 micrograms, and 100 mg of succinylcholine. Prior to skin incision, 12 patients received intrathecal injection of 12 micrograms/kg of preservative-free morphine sulfate (ITM), while the remaining 12 patients served as controls. The ITM was given undiluted at the L3-4 or L4-5 level. Anesthesia was maintained solely with enflurane, titrated to keep mean arterial pressure within 15% of the preoperative values. Vecuronium was given as required for relaxation. No additional narcotics were administered. Throughout the procedure, end-tidal (ET) enflurane concentration was recorded at 15-minute intervals from the mass spectrometer (Perkin Elmer). The intraoperative mean ET concentration of enflurane was significantly reduced in the ITM group beginning 1 hour after the injection (1.19 +/- .45% in the control group versus 0.73 +/- 0.08% in the ITM group). The enflurane requirements, expressed as percent end-tidal enflurane/hour, were significantly less in the ITM group for the duration of the procedure (0.8 +/- .17 v 1.08 +/- .22, respectively). In conclusion, when administered prior to skin incision for post-thoracotomy pain control, intrathecal morphine reduces intraoperative enflurane requirements.