The paired availability design for historical controls.

BACKGROUND: Although a randomized trial represents the most rigorous method of evaluating a medical intervention, some interventions would be extremely difficult to evaluate using this study design. One alternative, an observational cohort study, can give biased results if it is not possible to adjust for all relevant risk factors. METHODS: A recently developed and less well-known alternative is the paired availability design for historical controls. The paired availability design requires at least 10 hospitals or medical centers in which there is a change in the availability of the medical intervention. The statistical analysis involves a weighted average of a simple "before" versus "after" comparison from each hospital or medical center that adjusts for the change in availability. RESULTS: We expanded requirements for the paired availability design to yield valid inference. (1) The hospitals or medical centers serve a stable population. (2) Other aspects of patient management remain constant over time. (3) Criteria for outcome evaluation are constant over time. (4) Patient preferences for the medical intervention are constant over time. (5) For hospitals where the intervention was available in the "before" group, a change in availability in the "after group" does not change the effect of the intervention on outcome. CONCLUSION: The paired availability design has promise for evaluating medical versus surgical interventions, in which it is difficult to recruit patients to a randomized trial.

Oxytocin and lysophosphatidic acid induce stress fiber formation in human myometrial cells via a pathway involving Rho-kinase.

The actin cytoskeleton is important for stress fiber formation and contributes to the initiation and maintenance of smooth muscle contraction. To determine if oxytocin and lysophosphatidic acid (LPA) induce stress fiber formation, cultured human myometrial cells were exposed to oxytocin (10(-5) M) or LPA (10(-6) M), and filamentous (F) and globular (G) actin pools were stained with fluorescein isothiocyanate-phalloidin and Texas red DNase I, respectively. The F- to G-actin fluorescent-staining ratio was measured by fluorescence microscopy. Oxytocin and LPA increased stress fiber formation, as indicated by an increase in the F- to G-actin fluorescent-staining ratio. The Rho-kinase inhibitor Y-27632 markedly attenuated this increase. Oxytocin-induced stress fiber formation was completely inhibited in the presence of the oxytocin antagonist compound VI. Tyrosine kinase inhibition with tyrphostin A23 partially blocked the increase induced by oxytocin but had no effect on LPA-induced stress fiber formation. Stress fiber formation was not blocked by pertussis toxin, mitogen-activated protein kinase, or protein kinase C inhibitors. Our results show that human myometrial cells respond to oxytocin and LPA with the formation of stress fibers that may be involved in the maintenance of uterine contractions. Rho-kinase appears to be a key signaling factor in this pathway.

Rethinking historical controls.

Inference from traditional historical controls, i.e. comparing a new treatment in a current series of patients with an old treatment in a previous series of patients, may be subject to a strong selection bias. To avoid this bias, Baker and Lindeman (1994) proposed the paired availability design. By applying this methodology to estimate the effect of epidural analgesia on the probability of Cesarean section, we made two important contributions with the current study. First, we generalized the methodology to include different types of availability and multiple time periods. Second, we investigated how well the paired availability design reduced selection bias by comparing results to those from a meta-analysis of randomized trials and a multivariate analysis of concurrent controls. The confidence interval from the paired availability approach differed considerably from that of the multivariate analysis of concurrent controls but was similar to that from the meta-analysis of randomized trials. Because we believe the multivariate analysis of concurrent controls omitted an important predictor and the meta-analysis of randomized trials was the gold standard for inference, we concluded that the paired availability design did, in fact, reduce selection bias.

Myometrial adenylyl cyclase protein decreases on the last day of pregnancy in the rat.

To determine whether gestation-related changes in responsiveness of the rat uterus to beta-adrenergic agonists are mediated at the level of adenylyl cyclase, we measured myometrial adenylyl cyclase activity and protein quantities during pregnancy and labor. In rat myometrial membranes, basal adenylyl cyclase activity increased from the nonpregnant state to mid (Days 12-14) and then late (Days 18-20) gestation and then decreased intrapartum (Day 22). Stimulated adenylyl cyclase activity, at the level of the beta-adrenergic receptor (isoproterenol, 10(-4) M), the G protein (GTP, 10(-5) M), or the adenylyl cyclase enzyme (MnCl(2), 20 mM), was similarly altered during gestation. Total adenylyl cyclase protein was quantified by [(3)H]forskolin binding assay in myometrial membranes from nonpregnant and pregnant (Day 14, Day 20, Day 21, and intrapartum Day 22) rats. Adenylyl cyclase protein increased progressively from nonpregnant rats to pregnant rats at mid (Day 14) and late (Day 20) gestation, but it decreased abruptly to nonpregnant levels on Day 21, the day before parturition, and remained at similar levels on Day 22 (intrapartum). The gestation-related increase in expression of myometrial adenylyl cyclase protein may facilitate uterine quiescence during pregnancy, and the abrupt decrease of adenylyl cyclase protein on the last day of pregnancy may be a contributing mechanism for the initiation of labor.

Hypocapnia-induced contraction of porcine airway smooth muscle.

Hypocapnia constricts peripheral airways in vivo. This study investigated the role of airway smooth muscle in this phenomenon and the mechanism of hypocapnia-induced contraction in vitro. Isometric tension, intracellular pH (pHi) and intracellular free calcium concentration ([Ca2+]i) were measured in porcine airway smooth muscles suspended in organ baths in the presence of 5% or 0% CO2. In tracheal strips precontracted with carbachol, hypocapnic challenge (0% CO2) produced increases in tension, pHi, and [Ca2+]i. In bronchial rings or tracheal strips precontracted with carbachol, nifedipine administered between consecutive contractions attenuated responses to hypocapnia (75+/-11% above carbachol-precontracted tension before nifedipine versus 39+/-9% after nifedipine, n=7 bronchial rings, p<0.05). Neither indomethacin (5 microM), nordihydroguaiaretic acid (10 microM) nor phenidone (10 microM) significantly altered responses. These data suggest that enhanced Ca2+ influx through voltage-dependent Ca2+ channels of airway smooth muscle cells is important in airway responses to hypocapnia.

Chronic oxytocin pretreatment inhibits adenylyl cyclase activity in cultured rat myometrial cells.

To determine whether chronic oxytocin pretreatment inhibits adenylyl cyclase, we compared adenylyl cyclase activity in membranes prepared from cultured, immortalized rat myometrial cells that were untreated or pretreated for 24 h with oxytocin. Chronic oxytocin pretreatment (1 x 10(-5) M for 24 h) attenuated basal, guanosine triphosphate (1 x 10(-5) M)-, isoproterenol (1 x 10(-4) M)-, forskolin (1 x 10(-5) M)-, MnCl2 (20 mM)- or NaF (1 x 10(-2) M)-stimulated adenylyl cyclase activity by 27 +/- 5% to 39 +/- 11% (n = 6, p < 0.05). Oxytocin pretreatment for 2 h (n = 5) did not produce a significant effect. To understand the mechanism by which oxytocin pretreatment decreased activity of the adenylyl cyclase pathway, we compared effects of pretreatment with either oxytocin or phenylephrine on adenylyl cyclase activity and determined the effects of Gi inhibition and protein kinase C (PKC) depletion. Chronic (24 h) phenylephrine pretreatment (1 x 10(-4) M) had effects similar to those of oxytocin pretreatment (1 x 10(-5) M). PKC depletion with phorbol 12-myristate 13-acetate (1 x 10(-6) M, 41 h) prevented attenuation of adenylyl cyclase activity by oxytocin pretreatment (1 x 10(-5) M for 24 h). Inhibition of Gi by pertussis toxin pretreatment (1.25 microg/ml, 41 h) had no significant effect. These findings suggest that chronic oxytocin pretreatment desensitizes the adenylyl cyclase pathway by a cross-regulatory mechanism that involves activation of Gq and PKC.

Adenylyl cyclase messenger ribonucleic acid in myometrium: splice variant of type IV.

Regulation of uterine tone during pregnancy and parturition involves G protein-coupled receptors that modulate cellular levels of cAMP. Cyclic AMP is synthesized by a family of at least nine adenylyl cyclase enzymes that have unique tissue distributions, basal activities, and modes of regulation. Little is known regarding the expression of adenylyl cyclase isoforms in myometrium. Reverse transcription-polymerase chain reaction (PCR) was performed using RNA extracted from freshly dissected rat uteri and a cell line of homogeneous rat myometrial cells. Using PCR primers specific for adenylyl cyclases I-IX, mRNA encoding adenylyl cyclases II-IX were identified in both fresh and cultured myometrial cells. A splice variant of adenylyl cyclase type IV was also identified in RNA from both sources. These data show that multiple isoforms of adenylyl cyclases are expressed at the mRNA level in myometrial cells and suggest that the regulation of cellular levels of cAMP likely involves a complex integration of cellular second messengers acting upon multiple isoforms of adenylyl cyclases.

Dermal and airway responses to monoclonal antibodies specific for canine IgE.

In order to understand mechanisms underlying variability of IgE-mediated responses in vivo, we compared effects of different monoclonal antibodies of IgE on dermal and airway responses in a group of atopic dogs. Using intradermal testing, fourteen antibodies were screened in Basenji-Greyhound dogs. For further comparisons between dermal and airway responses, we selected the two antibodies that stimulated the greatest and least dermal responses, respectively. These antibodies bound to IgE with similar affinities (4.1 +/- 0.2 x 10(9) and 1.5 +/- 0.2 x 10(10) M-1). Dose-response curves to intradermal testing were constructed for these two antibodies. On a separate occasion, peripheral airway resistance (Rp) was determined before and after aerosol challenge with an antibody or saline in the same dogs. For one antibody (affinity 4.1 +/- 0.2 x 10(9) M-1), Rp reached a maximum (407 +/- 142% above baseline; mean +/- SE, n = 6) 10 to 15 min after challenge, while maximum responses to saline (62 +/- 16% above baseline, p < 0.01) occurred immediately after aerosol delivery. Responses to the other antibody were similar (p = 0.068) to responses to saline. The magnitude of skin responses did not predict the magnitude of airway responses. These findings suggest that differences in affinities, alone, do not predict magnitude of responsiveness to the anti-IgE antibody and that mechanisms underlying skin and airway responses may differ qualitatively and/or quantitatively.

MgSO4 relaxes porcine airway smooth muscle by reducing Ca2+ entry.

Magnesium sulfate (MgSO4) is used clinically, but its mechanism of action is unknown. To determine whether MgSO4 relaxes airway smooth muscle and to investigate the pathways involved, we compared effects of MgSO4 in porcine tracheal and bronchial muscles contracted with either carbachol or KCl and measured the effects of MgSO4 on the concentration of intracellular free calcium ([Ca2+]i). Lungs were dissected after anesthesia and exsanguination. Tracheal strips and bronchial rings were suspended in tissue baths for measurement of isometric tension in the presence of different concentrations of MgSO4. In separate experiments, tracheal smooth muscle tension and [Ca2+]i were measured simultaneously, using the fluorescent dye fura 2. MgSO4 (1.2, 2.2, 9.2 mM) produced a concentration dependent rightward shift of contraction dose-response curves to KCl but not to carbachol. MgSO4 relaxed trachealis muscles precontracted with KCl or carbachol and simultaneously decreased [Ca2+]i. These findings indicate that MgSO4 directly relaxes airway smooth muscle and that the mechanism involves a decrease in [Ca2+]i. Because initiation and maintenance of contraction during KCl stimulation and maintenance of contraction during carbachol stimulation require Ca2+ entry through voltage-dependent calcium channels, MgSO4-induced relaxation may involve a decrease in Ca2+ entry via these channels.

Inhaled nitric oxide attenuates bronchoconstriction in canine peripheral airways.

Inhaled nitric oxide has been proposed as a bronchodilator because it relaxes vascular and airway smooth muscle and attenuates cholinergic reflexes. Although inhaled nitric oxide has been shown to act as a bronchodilator in central airways, effects on peripheral airways are largely unknown. To determine whether nitric oxide produces direct relaxation of peripheral airways, we investigated the ability of nitric oxide to attenuate hypocapnia- and acetylcholine-induced constriction in the peripheral airways of anesthetized dogs. Peripheral airway resistance (RP) was measured using a wedged bronchoscope technique. RP was increased by either hypocapnia (0% CO2 through the bronchoscope for 3 min) or by aerosolized acetylcholine (30 to 60 micrograms/ml for 1 to 3 min), in the presence or absence of nitric oxide. Nitric oxide was delivered directly to the lung periphery in the absence of O2. Nitric oxide (14.5 to 250 ppm) attenuated responses to hypocapnia by 38 +/- 0 to 74 +/- 0% (n = 6) and to acetylcholine by 36 +/- 0 to 52 +/- 0% (n = 6). The ability of inhaled nitric oxide (< 100 ppm) to attenuate Rp responses to two different direct-acting stimuli suggests that nitric oxide acts as a bronchodilator in the lung periphery. The mechanism for this effect may involve relaxation of airway and/or vascular smooth muscle.

Direct effects of inhaled nitric oxide on canine peripheral airways.

The effects of nitric oxide on peripheral airways in vivo, and whether these effects occur via direct or indirect mechanisms, are unknown. We studied effects of inhaled nitric oxide on histamine-constricted canine peripheral airways in the presence or absence of atropine and an inhibitor of guanylyl cyclase, methylene blue. Peripheral resistance (Rp) was measured by using a wedged-bronchoscope technique in anesthetized dogs. A stable baseline Rp was established. Histamine was infused intravenously, and increasing concentrations of nitric oxide (50-500 ppm) were delivered through the bronchoscope. In separate experiments, histamine was infused intravenously in the presence or absence of atropine (0.2 mg/kg iv) or methylene blue (20 mg/min iv). When Rp stabilized, nitric oxide (500 ppm) was delivered. Nitric oxide partially reversed histamine-induced bronchoconstriction in a dose-dependent fashion (maximum of 42 +/- 3% reduction at 500 ppm; n = 5; P < 0.01) that did not differ in the presence or absence of atropine. Methylene blue blocked the effect of nitric oxide on histamine-induced constriction (n = 6; P = 0.45). These findings suggest that high concentrations of nitric oxide produce small but significant bronchodilation of peripheral airways through a mechanism independent of the cholinergic neural pathway. The mechanism of action appears to involve activation of guanylyl cyclase.

Intracellular pH regulates voltage-dependent Ca2+ channels in porcine tracheal smooth muscle cells.

Changes in CO2 or in pH modify airway smooth muscle contractility. To investigate the mechanisms involved, we compared K(+)-induced contractions in porcine bronchial rings exposed to different CO2 concentrations and directly measured the effects of changes in intracellular (pHi) or extracellular pH (pHo) on Ca2+ currents (ICa) through voltage-dependent Ca2+ channels (VDC) in porcine tracheal smooth muscle cells. Hypocapnia and hypercapnia caused leftward and rightward shifts, respectively, in the dose-response to K+ (P < 0.05) but did not change the maximum force obtained. Peak ICa (10 mM external Ca2+) elicited by depolarizing pulses from -80 mV was maximal [-265 +/- 12 pA (mean +/- SE), n = 19] at +10 mV. Intracellular acidification decreased the peak ICa at +10 mV from -261 +/- 20 pA to -177 +/- 12 pA (P < 0.05, n = 4), while intracellular alkalinization increased the peak ICa at +10 mV from -302 +/- 27 pA to -368 +/- 26 pA (P < 0.05, n = 4). Changes in pHo had little effect on ICa. There was no shift in the voltage-dependence of induced ICa with any change. We conclude that pHi, but not pHo, directly modulates the entry of Ca2+ into airway smooth muscle cells through VDC. This mechanism may contribute to regulation of airway tone by CO2.

The paired availability design: a proposal for evaluating epidural analgesia during labor.

The paired availability design (PAD) can reduce selection bias when it is not possible to randomize subjects. PAD consists of independent pairs of experimental and control groups. Within each pair, the intervention is the availability of treatment not its receipt. In the experimental group, the new treatment is made available to all subjects although some may not receive it. In the control group, the experimental treatment is generally not available to subjects although some may receive it in special circumstances. We present a statistic to test a null hypothesis that the receipt of intervention will increase response by a specified non-zero amount delta. We propose this design for use in a study of the effect of epidural analgesia on the rate of Caesarean section.

Differential effects of nitrovasodilators and nitric oxide on porcine tracheal and bronchial muscle in vitro.

Nitrovasodilators and nitric oxide relax airway smooth muscle. The mechanism by which nitrovasodilators are thought to act is by release of nitric oxide, but the importance of nitric oxide in nitrovasodilator-induced airway smooth muscle relaxation is unclear. The aim of this study was to compare the relaxing effects of nitric oxide itself with those of nitrovasodilators in porcine tracheal muscle and intrapulmonary airways and to investigate the mechanisms involved. Strips of porcine tracheal smooth muscle, rings of bronchi, and strips of bronchi from the same animal were suspended in organ chambers in modified Krebs Ringer solution (95% O2-5% CO2, 37 degrees C). Tissues were contracted with carbachol, and concentration-response curves to nitric oxide, sodium nitroprusside, and SIN-1 (an active metabolite of molsidomine) were obtained. All tissues relaxed to sodium nitroprusside, SIN-1, and nitric oxide. The relaxation to nitric oxide but not to SIN-1 or sodium nitroprusside was inhibited by methylene blue. Tissues pretreated with methylene blue that failed to relax to nitric oxide were, however, relaxed by sodium nitroprusside. These results demonstrate that nitrovasodilators relax airways by a mechanism other than by or in addition to the release of nitric oxide.

Interaction between halothane and the nonadrenergic, noncholinergic inhibitory system in porcine trachealis muscle.

BACKGROUND: Volatile anesthetics significantly affect cholinergic neural transmission in the airways and relax airway smooth muscle. Activation of the nonadrenergic, noncholinergic inhibitory neural pathway, which is thought to be mediated by nitric oxide, relaxes human and procine airways. The purpose of the current study was to determine in the isolated porcine trachealis muscle whether relaxation of airway smooth muscle by halothane is mediated in part by activation of the nonadrenergic, noncholinergic inhibitory system. METHODS: Isometric tension was measured in porcine trachealis muscle suspended in tissue baths in the presence of propranalol (10(-6) M). After stimulation of postsynaptic nicotinic cholinergic receptors with 1,1-dimethyl-4-phenyl-piper-azinium iodide (10(-4) M) to prevent contractile responses to subsequent electrical field stimulation, carbachol (3 x 10(-7) M) was added to increase tone. Nonadrenergic, noncholinergic relaxation responses to electrical field stimulation were then measured in the presence of inhibitors of nitric oxide synthase or L-arginine (the substrate for nitric oxide synthase), in the presence and absence halothane. RESULTS: Electrical field stimulation produced frequency-dependent relaxations that were attenuated by inhibitors of nitric oxide synthase (NG-nitro-L-arginine methyl ester [L-NAME] or NG-monomethyl-L-arginine, 10(-4) M). Pretreatment with L-arginine (10(-4) M) prevented the effect of L-NAME. Halothane (0.5% or 1.0%) neither enhanced nor attenuated nonadrenergic, noncholinergic relaxations in the presence of L-NAME, D-NAME, L-arginine, or D-arginine. CONCLUSIONS: Halothane, at concentrations < or = 1.0%, does not relax porcine airway smooth muscle in vitro by activating the nonadrenergic, noncholinergic inhibitory system.

Role of potassium channels in hypoxic relaxation of porcine bronchi in vitro.

To elucidate the mechanism of hypoxic relaxation of airway smooth muscle in vitro, we investigated the role of adenosine triphosphate-sensitive potassium (KATP) channels in this response. Second- and third-order porcine bronchial rings were suspended in 10-ml organ baths containing Krebs-Henseleit solution. To demonstrate the presence of KATP channels in this tissue, bronchial rings were contracted with carbachol (1 microM) in the presence of glibenclamide (100 microM), a KATP channel blocker, or the vehicle dimethyl sulfoxide (DMSO) (0.1 ml), and dose-response curves to levcromakalim (a KATP channel opener) or isoproterenol were constructed. In separate experiments, either glibenclamide or DMSO was added to the chamber and rings were contracted with carbachol (1 microM) in the presence of 95% O2-5% CO2. At the plateau, airways were relaxed with either isoproterenol (0.1 or 0.3 microM) or hypoxia (50, 28, or 0% O2, with constant 5% CO2). Glibenclamide, when compared with DMSO, shifted the dose-response curve to levcromakalim, but not to isoproterenol. Glibenclamide attenuated hypoxic relaxation in rings exposed to 50% O2 (from 35 +/- 4% to 23 +/- 3%, n = 6, P < 0.001) and increased the time to 63% relaxation in rings exposed to 50% O2 or to 28% O2. Responses in rings exposed to 0% O2 or to isoproterenol (0.1 or 0.3 microM) were not significantly altered. The ability of glibenclamide to attenuate the maximum response to 50% O2 and to increase the time to 63% relaxation during exposure to 50 or 28% O2 suggests that one component of hypoxic bronchodilation during moderate degrees of hypoxia is opening of KATP channels.

Corticosteroid withdrawal restores responses to calcium chelators and enhances cholinergic responsiveness.

To investigate the importance of the inflammatory response in acute peripheral airway constriction, we measured peripheral airway responses to calcium chelators and acetylcholine in anesthetized Basenji-Greyhound (BG) dogs before, during, and after chronic corticosteroid treatment. A wedged bronchoscope technique was used to measure peripheral airway resistance before and after aerosol challenge with 4% Na2EDTA or acetylcholine (10 micrograms/ml) in contralateral lungs. After the initial measurements, five BG dogs received long-term treatment with methylprednisolone (2 mg/kg/d, subcutaneously), and five dogs were not treated and served as controls. Four weeks of methylprednisolone treatment almost totally abolished responses to Na2EDTA, but responses to acetylcholine did not change significantly. After discontinuing corticosteroid therapy, responses to Na2EDTA returned to levels found before corticosteroid treatment; responses to acetylcholine were significantly enhanced. We conclude that chronic corticosteroid treatment reduces acute response to calcium chelators and that withdrawal of corticosteroid therapy is associated with enhanced cholinergic responsiveness.

Importance of prostanoids in calcium chelator-induced airway constriction.

To elucidate the mechanism of calcium chelator-induced airway constriction, we examined the relationship between prostanoids and histamine in bronchoalveolar lavage (BAL) fluid and the magnitude of airway constriction in peripheral airways of anesthetized Basenji-Greyhound dogs. A wedged bronchoscope technique was used to measure collateral system resistance (Rcs) before and after aerosol challenges. Sublobar segments were challenged either with acetylcholine or with Na2EDTA in the presence or absence of meclofenamate sodium (3 mg/kg iv) or methylprednisolone (2 mg.kg-1.day-1). After measurements of Rcs, BAL was performed, and the fluid was analyzed for prostanoids with the use of gas chromatography-mass spectrometry. Sublobar segments challenged with Na2EDTA showed increased concentrations of prostaglandin (PG) D2 but no increases in PGE2, PGF2 alpha, 9 alpha,11 beta-PGF2, 6-keto-PGF1 alpha, thromboxane B2, or histamine. There was a strong relationship (r = 0.84, P = 0.005) between changes in Rcs after Na2EDTA and concentrations of PGD2 in BAL fluid. Acetylcholine, which increased Rcs to a similar degree as Na2EDTA did, produced no significant increase in prostanoid concentrations. Changes in Rcs after Na2EDTA and concentrations of PGD2 were reduced in the presence of meclofenamate or methylprednisolone. These data support the idea that the mechanism of calcium chelator-induced bronchoconstriction involves the release of bronchoconstricting prostanoids.

Effects of halothane, propofol, and thiopental on peripheral airway reactivity.

BACKGROUND: General anesthetics modify airway responsiveness by several mechanisms, including direct effects on airway smooth muscle and reductions in neural reflex activity. Halothane has been shown to reduce responsiveness through both of these mechanisms. The airway effects of barbiturates are controversial, and the effects of propofol are unknown. METHODS: To compare the direct effects of halothane, thiopental, and propofol in vivo, canine peripheral airways were constricted with two stimuli, histamine and hypocapnia, which are thought to directly contract smooth muscle. The authors then investigated the role of ATP-sensitive potassium (KATP) channels as a mechanism for attenuating these responses. Basenji-Greyhound (BG) dogs were anesthetized with either halothane (1.5 MAC), thiopental (7.5 mg.kg-1 x min-1 intravenously) plus fentanyl (25 micrograms intravenously every 20-30 min), or propofol (0.6 mg.kg-1 x min-1 intravenously). A wedged bronchoscope technique was used to measure peripheral airway resistance (RP). After a stable baseline was obtained, dose-response curves to histamine (50, 100, or 200 micrograms intravenous bolus) or hypocapnia (0% CO2 for 2 min with 100, 200, or 400 ml/min collateral flow) were constructed. On separate occasions, the same sublobar segments were pretreated with glibenclamide (2 mg/ml aerosol), a KATP channel blocker, and dose-response curves to hypocapnia were repeated. RESULTS: Dose-response curves to histamine were similar during all three anesthetics. Halothane decreased airway responsiveness to hypocapnia, compared with either thiopental or propofol (P < 0.05). Pretreatment with glibenclamide abolished the effect of halothane on hypocapnia-induced airway constriction. CONCLUSIONS: These results indicate that propofol afforded no benefit over thiopental or halothane in reducing peripheral airway responsiveness. Furthermore, the beneficial effects of halothane in reducing responsiveness to hypocapnia appear to be mediated by the opening of KATP channels.