Eckehard A. E. Stuth

Alteration of the critical arteriovenous oxygen saturation relationship by sustained afterload reduction after the Norwood procedure

OBJECTIVES Hemodynamic vulnerability after the Norwood procedure for hypoplastic left heart syndrome results from impaired myocardial function, and critical inefficiency of parallel circulation. Traditional management strategies have attempted to optimize circulatory efficiency by using arterial oxygen saturation (SaO(2)) as an index of pulmonary/systemic flow balance, attempting to maintain SaO(2) within a theoretically optimal critical range of 75% to 80%. This optimal range of SaO(2) has not been verified clinically, and strategies targeting SaO(2) may be limited by the fact that SaO(2) is a poor predictor of systemic oxygen delivery. We have previously reported higher venous saturation (SvO(2)), lower arteriovenous oxygen content difference, lower systemic vascular resistance, lower pulmonary/systemic flow ratio, and improved survival with the perioperative use of phenoxybenzamine and continuous monitoring of SvO(2). In this investigation, we tested the hypothesis that intense afterload reduction with phenoxybenzamine would modify the SvO(2)-SaO(2) relationship by preventing deterioration of systemic oxygen delivery at high SaO(2). METHODS Seventy-one consecutive neonates undergoing the Norwood procedure with and without phenoxybenzamine were studied. Perioperative hemodynamic management targeted SvO(2) greater than 50%. Hemodynamic data were prospectively acquired for 48 hours postoperatively and analyzed to assess the effect of phenoxybenzamine on the relationship between SaO(2) and SvO(2) and other hemodynamic indices. Sixty-two patients received phenoxybenzamine 0.25 mg/kg on cardiopulmonary bypass; 9 who did not served as controls. RESULTS In control patients, SvO(2) peaked at an SaO(2) of 77%, with reduced SvO(2) at SaO(2) > 85% and SaO(2) < 70% (P <.01), while arteriovenous oxygen content difference increased with SaO(2) greater than 80% (P <.001). In patients receiving phenoxybenzamine, the SvO(2) increased linearly with SaO(2) greater than 65% (P <.001), and arteriovenous oxygen content difference was constant at all SaO(2) (P = ns). The SvO(2) was higher, and the arteriovenous oxygen content difference lower, across the whole SaO(2) range with phenoxybenzamine (P <.0001). CONCLUSIONS A critical range of SaO(2) for optimizing systemic oxygen delivery was confirmed in control patients, and was effectively eliminated by phenoxybenzamine, specifically by eliminating the systemic hypoperfusion associated with high SaO(2). This effect allows higher SaO(2) to be included in a rational hemodynamic strategy to improve systemic oxygen delivery in the early postoperative management of patients receiving intense afterload reduction with phenoxybenzamine. The predictability of SvO(2) from SaO(2) is low in both groups, emphasizing the importance of SvO(2) measurement in these patients.

Modulation of the synaptic drive to respiratory premotor and motor neurons.

The characteristics of GABAergic inhibitory modulation of respiratory bulbospinal neuronal activity and short-term potentiation (STP) of phrenic motoneuronal activity were studied. Extracellular unit recording and picoejection techniques in anesthetized dogs showed that both the spontaneous rhythmic and reflexly induced discharge patterns of inspiratory (I) and expiratory (E) premotor neurons were proportionately amplified by the localized application of picomole amounts of bicuculline (Bic), a competitive GABAA antagonist. Intracellular recording and paired-pulse stimulation techniques in anesthetized rats demonstrated an STP of phrenic motor output that appears to be mediated by NMDA receptors and is associated with facilitation of EPSPs and prolonged depolarization of individual phrenic motoneurons. We speculate that both GABAergic gain modulation of premotor neuronal activity and NMDA-mediated STP of phrenic activity may be neural substrates which are involved with the optimization of respiratory and non-respiratory behaviors, via adaptive and/or differential control of breathing.

Effects of Sevoflurane on excitatory neurotransmission to medullary expiratory neurons and on phrenic nerve activity in a decerebrate dog model

Background Sevoflurane is a new volatile anesthetic with a pronounced respiratory depressant effect. Synaptic neurotransmission in canine expiratory bulbospinal neurons is mainly mediated by excitatory N-methyl-d-aspartatic acid (NMDA) receptor input and modulated by inhibitory &ggr;-aminobutyric acid type A (GABAA) receptors. The authors investigated the effect of sevoflurane on these mechanisms in decerebrate dogs. Methods Studies were performed in decerebrate, vagotomized, paralyzed and mechanically ventilated dogs during hypercapnic hyperoxia. The effect of 1 minimum alveolar concentration (MAC; 2.4%) sevoflurane on extracellularly recorded neuronal activity was measured during localized picoejection of the glutamate agonist NMDA and the GABAA receptor blocker bicuculline in a two-part protocol. First, complete blockade of the GABAAergic mechanism by bicuculline allowed differentiation between the effects of sevoflurane on overall GABAAergic inhibition and on overall glutamatergic excitation. In a second step, the neuronal response to exogenous NMDA was used to estimate sevoflurane’s effect on postsynaptic glutamatergic neurotransmission. Results One minimum alveolar concentration sevoflurane depressed the spontaneous activity of 16 expiratory neurons by 36.7 ± 22.4% (mean ± SD). Overall glutamatergic excitation was depressed 19.5 ± 16.2%, and GABAAergic inhibition was enhanced 18.7 ± 20.6%. However, the postsynaptic response to exogenous NMDA was not significantly altered. In addition, 1 MAC sevoflurane depressed peak phrenic nerve activity by 61.8 ± 17.7%. Conclusions In the authors’in vivo expiratory neuronal model, the depressive effect of sevoflurane on synaptic neurotransmission was caused by a reduction of presynaptic glutamatergic excitation and an enhancement of GABAAergic inhibition. The effects on expiratory neuronal activity were similar to halothane, but sevoflurane caused a stronger depression of phrenic nerve activity than halothane.

NMDA receptor-mediated transmission of carotid body chemoreceptor input to expiratory bulbospinal neurones in dogs.

1. This study tested the hypothesis that excitatory amino acid receptors mediate the excitatory response of expiratory bulbospinal neurones to carotid body chemoreceptor inputs. 2. Studies were carried out in thiopental sodium anaesthetized, paralysed, ventilated, vagotomized dogs. 3. Brisk, short‐duration chemoreceptor activation was produced by bilateral bolus injections of CO2‐saturated saline (PCO2 > 700 mmHg) into the autoperfused carotid arteries. A pressurized‐reservoir‐solenoid valve system was used to deliver the CO2 bolus injections just prior to the onset of the neural expiratory phase, as determined from the phrenic neurogram, about once per minute. 4. Multibarrelled micropipettes were used to record neuronal unit activity and deliver neurotransmitter agents. Net responses of expiratory bulbospinal neurones to peripheral chemoreceptor activation were determined by subtracting the mean discharge frequencies (Fn) during three control expiratory cycles from the Fn during administration of a CO2 test bolus. The role of excitatory amino acid receptors in mediating this response was determined by comparing the baseline and bolus expiratory neuronal Fn before, during and after the pressure microejection of the NMDA receptor antagonist 2‐amino‐5‐phosphonovalerate (AP5) or the non‐NMDA receptor antagonist 2,3‐dihydroxy‐6‐nitro‐7‐sulphamoyl‐ benzo(f)quinoxaline (NBQX). Ejection rates of AP5 and NBQX were measured by monitoring the movement of the pipette meniscus. 5. AP5 reduced Fn during both the control and bolus cycles, as well as reducing the change in Fn between control and bolus cycles. NBQX had no effect on either baseline or bolus responses. 6. AP5 did not prevent excitation of expiratory bulbospinal neurones by AMPA. Coadministration of AMPA with AP5 prevented the AP5‐mediated decrease in Fn but not the dose‐dependent reduction in the CO2 bolus response. 7. Taken together, these data strongly suggest that the carotid chemoreceptor‐mediated excitation of expiratory bulbospinal neurones is dependent on NMDA but not non‐NMDA glutamate receptors.

Clinically Relevant Infusion Rates of μ-Opioid Agonist Remifentanil Cause Bradypnea in Decerebrate Dogs but not Via Direct Effects in the pre-Bötzinger Complex Region

Systemic administration of mu-opioids at clinical doses for analgesia typically slows respiratory rate. Mu-opioid receptors (MORs) on pre-Bötzinger Complex (pre-BötC) respiratory neurons, the putative kernel of respiratory rhythmogenesis, are potential targets. The purpose of this study was to determine the contribution of pre-BötC MORs to the bradypnea produced in vivo by intravenous administration of clinically relevant infusion rates of remifentanil (remi), a short-acting, potent mu-opioid analgesic. In decerebrate dogs, multibarrel micropipettes were used to record pre-BötC neuronal activity and to eject the opioid antagonist naloxone (NAL, 0.5 mM), the glutamate agonist D-homocysteic acid (DLH, 20 mM), or the MOR agonist [D-Ala(2), N-Me-Phe(4), gly-ol(5)]-enkephalin (DAMGO, 100 microM). Inspiratory and expiratory durations (T(I) and T(E)) and peak phrenic nerve activity (PPA) were measured from the phrenic neurogram. The pre-BötC was functionally identified by its rate altering response (typically tachypnea) to DLH microinjection. During intravenous remi-induced bradypnea (approximately 60% decrease in central breathing frequency, f(B)), bilateral injections of NAL in the pre-BötC did not change T(I), T(E), f(B), and PPA. Also, NAL picoejected onto single pre-BötC neurons depressed by intravenous remi had no effect on their discharge. In contrast, approximately 60 microg/kg of intravenous NAL rapidly reversed all remi-induced effects. In a separate group of dogs, microinjections of DAMGO in the pre-BötC increased f(B) by 44%, while subsequent intravenous remi infusion more than offset this DAMGO induced tachypnea. These results indicate that mu-opioids at plasma concentrations that cause profound analgesia produce their bradypneic effect via MORs located outside the pre-BötC region.

Dose-dependent Effects of Halothane on the Carbon Dioxide Responses of Expiratory and Inspiratory Bulbospinal Neurons and the Phrenic Nerve Activities in Dogs

Background:Expiratory bulbospinal and inspiratory bulbospinal neurons in the ventral respiratory group provide drive for thoracoabdominal expiratory and phrenic and thoracic inspiratory motor neurons. Potent inhalational agents such as halothane may have differential effects on inspiratory and expiratory neurons, but detailed studies comparing neurons at a homologous level are lacking. Methods:The dose-dependent effects of anesthesia with 1.0- 2.5 minimum alveolar concentration halothane on the CO2 responses of single expiratory and inspiratory bulbospinal neurons of the ventral respiratory group and on phrenic neural activities were studied in nonpremedicated, anesthetized, paralyzed, vagotomlzed dogs. Hyperventllation with O2 and the addition of CO2-O2 mixtures were used to produce low, medium, and high steady-state levels of central CO2 drive. Results:Peak neuron discharge frequency decreased progressively with increasing halothane dose at all levels of CO2 drive for both types of neurons. The sensitivities of inspiratory and expiratory bulbospinal neuronal activities to halothane were not significantly different from one another, whereas the sensitivity to halothane of the peak phrenic activity was markedly greater than those of the neurons. Increasing halothane dose caused a downward, predominantly parallel shift of the CO2 response curves. Phrenic nerve activity also showed a decrease in slope of the CO2 response. Conclusions:The activities of respiratory premotor neurons are less depressed by Increasing doses of halothane than is phrenic nerve activity. The greater depression of phrenic activity may result from additional anesthetic actions on the efferent motor pathways, resulting in decreased descending synaptic inputs to phrenic motor neurons.

Dose-dependent effects of isoflurane on the CO2 responses of expiratory medullary neurons and the phrenic nerve activities in dogs.

Expiratory bulbospinal neurons (EBS) neurons in the region of the nucleus retroambigualis provide a major source of drive for thoracic and abdominal expiratory motoneurons. These studies examined the dose-dependent effects of isoflurane anesthesia from 0.5-2.5 minimum alveolar concentration (MAC) on the CO2 responses of extracellularly recorded single EBS neurons and phrenic neural activities in unpremedicated, anesthetized, paralyzed, vagotomized dogs. Hyperventilation with O2 and the addition of CO2-O2 mixtures were used to produce three steady-state levels of central CO2 drive: low, medium, and high with corresponding mean PaCO2 values of 29, 45, and 67 mmHg. Plots of peak expiratory neuron discharge frequency (Fn) versus steady-state levels of isoflurane dose showed a progressive decrease in Fn for both high and medium drive levels, whereas for low drive most neurons fired tonically at all MAC levels with little change in Fn. Peak phrenic activity showed a similar but more pronounced decline with increased MAC for all levels of CO2 drive. Increasing isoflurane dose caused a downward, predominantly parallel shift of the CO2 response curves for both expiratory neuronal and phrenic neural activities. Phrenic activity ceased at 2-2.5 MAC isoflurane for all CO2 drive levels, whereas EBS neurons continued to fire tonically at the same depth at 30-50% of Fn at 1 MAC for the high level. These studies suggest that phrenic (inspiratory) activity is more sensitive than EBS neurons to isoflurane and that increases in CO2 drive can only partially offset the depressant effect of isoflurane on central respiratory activities.

Pontine μ-opioid receptors mediate bradypnea caused by intravenous remifentanil infusions at clinically relevant concentrations in dogs

Life-threatening side effects such as profound bradypnea or apnea and variable upper airway obstruction limit the use of opioids for analgesia. It is yet unclear which sites containing μ-opioid receptors (μORs) within the intact in vivo mammalian respiratory control network are responsible. The purpose of this study was 1) to define the pontine region in which μOR agonists produce bradypnea and 2) to determine whether antagonism of those μORs reverses bradypnea produced by intravenous remifentanil (remi; 0.1-1.0 μg·kg(-1)·min(-1)). The effects of microinjections of agonist [D-Ala(2),N-Me-Phe(4),Gly-ol(5)]-enkephalin (DAMGO; 100 μM) and antagonist naloxone (NAL; 100 μM) into the dorsal rostral pons on the phrenic neurogram were studied in a decerebrate, vagotomized, ventilated, paralyzed canine preparation during hyperoxia. A 1-mm grid pattern of microinjections was used. The DAMGO-sensitive region extended from 5 to 7 mm lateral of midline and from 0 to 2 mm caudal of the inferior colliculus at a depth of 3-4 mm. During remi-induced bradypnea (~72% reduction in fictive breathing rate) NAL microinjections (~500 nl each) within the region defined by the DAMGO protocol were able to reverse bradypnea by 47% (SD 48.0%) per microinjection, with 13 of 84 microinjections producing complete reversal. Histological examination of fluorescent microsphere injections shows that the sensitive region corresponds to the parabrachial/Kölliker-Fuse complex.

Negative-pressure pulmonary edema in a child with hiccups during induction

An 8-yr-old, 20-kg girl was scheduled to undergo dental restorations and extractions as an outpatient. The patient had a history of seizures and multiple ischemic strokes secondary to Moyamoya disease. She had undergone three previous operations during general anesthesia, without anesthesia-related difficulties. At the preoperative visit, she was aphasic and drooling. Significant left lower extremity weakness and diffuse fine motor deficits were also present, but the patient was ambulatory. The patient received 10 mg midazolam via G-tube as premedication. After application of electrocardiogram leads and pulse oximetry, inhalational induction with 70% nitrous oxide in oxygen and halothane was begun. Within seconds after the start of induction, vigorous hiccups, which were accompanied by tracheal tugging, developed in the patient. Pulse oximetry showed a stable arterial saturation of 99-10076 throughout induction, and capnography did not show impairment of gas exchange. Therefore, inhalational induction was continued, with the child breathing spontaneously. The strong hiccups never ceased during approximately 10 min of induction. Finally, intravenous access was obtained and 20 mg rocuronium was administered intravenously. As neuromuscular blockade occurred, spontaneous ventilation and the hiccups stopped, and positive pres-

Effects of halothane on synaptic neurotransmission to medullary expiratory neurons in the ventral respiratory group of dogs.

BACKGROUND The activity of canine expiratory neurons is primarily dependent on N-methyl-D-aspartic acid (NMDA)-receptor mediated excitatory chemodrive inputs and a powerful inhibitory gain modulatory mechanism mediated via gamma-aminobutyric acidA (GABA(A)) receptors. We examined whether the depressant effect of halothane on expiratory neuronal activity is primarily caused by a reduction in glutamatergic excitation or a potentiation of the inhibitory mechanism. METHODS Experiments were performed in halothane-anesthetized, vagotomized, paralyzed, and mechanically ventilated dogs during hypercapnic hyperoxia. The effect of a halothane dose increase from one minimum alveolar concentration (MAC) to 2 MAC on extracellularly recorded expiratory neuronal activity was studied before and during complete GABA(A) receptor blockade by localized picoejection of bicuculline close to the neuron. Complete blockade of the inhibitory mechanism allowed differentiation between the effects of halothane on overall NMDA-mediated excitation and on GABA(A)-mediated inhibition. RESULTS The spontaneous activity of 12 expiratory neurons was significantly depressed (18.1%) by the 1-MAC halothane dose increase. Overall glutamatergic excitation was depressed 38.3+/-12.3% (mean +/- SD) by the 1-MAC halothane increase. The prevailing GABA(A)ergic attenuation of neuronal output decreased significantly from 49.5+/-10 to 32.0+/-10.4%. Thus overall inhibition was reduced by halothane by 33.5+/-17.2%. CONCLUSIONS These results suggest that the depressive effect of a 1-MAC halothane dose increase on expiratory neuronal activity in our in vivo preparation with an intact neural network was mainly caused by a reduction of synaptic excitatory mechanisms and not an enhancement of synaptic inhibitory mechanisms.