CO(2)/H(+) chemoreception in the cat pre-Bötzinger complex in vivo.

We examined the effects of focal tissue acidosis in the pre-Bötzinger complex (pre-BötC; the proposed locus of respiratory rhythm generation) on phrenic nerve discharge in chloralose-anesthetized, vagotomized, paralyzed, mechanically ventilated cats. Focal tissue acidosis was produced by unilateral microinjection of 10-20 nl of the carbonic anhydrase inhibitors acetazolamide (AZ; 50 microM) or methazolamide (MZ; 50 microM). Microinjection of AZ and MZ into 14 sites in the pre-BötC reversibly increased the peak amplitude of integrated phrenic nerve discharge and, in some sites, produced augmented bursts (i.e., eupneic breath ending with a high-amplitude, short-duration burst). Microinjection of AZ and MZ into this region also reversibly increased the frequency of eupneic phrenic bursts in seven sites and produced premature bursts (i.e., doublets) in five sites. Phrenic nerve discharge increased within 5-15 min of microinjection of either agent; however, the time to the peak increase and the time to recovery were less with AZ than with MZ, consistent with the different pharmacological properties of AZ and MZ. In contrast to other CO(2)/H(+) brain stem respiratory chemosensitive sites demonstrated in vivo, which have only shown increases in amplitude of integrated phrenic nerve activity, focal tissue acidosis in the pre-BötC increases frequency of phrenic bursts and produces premature (i.e., doublet) bursts. These data indicate that the pre-BötC has the potential to play a role in the modulation of respiratory rhythm and pattern elicited by increased CO(2)/H(+) and lend additional support to the concept that the proposed locus for respiratory rhythm generation has intrinsic chemosensitivity.

Hypoxic excitation in neurons cultured from the rostral ventrolateral medulla of the neonatal rat.

Neurons within cardiorespiratory regions of the rostral ventrolateral medulla (RVLM) have been shown to be excited by local hypoxia. To determine the electrophysiological properties of these excitatory responses to hypoxia, we developed a primary dissociated cell culture system to examine the intrinsic response of RVLM neurons to hypoxia. Neonatal rat neurons plated on medullary astrocyte monolayers were studied using the whole cell perforated patch-clamp technique. Sodium cyanide (NaCN, 0.5-10 mM) was used, and membrane potential (V(m)), firing frequency, and input resistance were examined. In 11 of 19 neurons, NaCN produced a V(m) depolarization, an increase in firing frequency, and a decrease in input resistance, suggesting the opening of a cation channel. The hypoxic depolarization had a linear dose response and was dependent on baseline V(m), with a greater response at more hyperpolarized V(m). In 8 of 19 neurons, NaCN produced a V(m) hyperpolarization, decrease in firing frequency, and variable changes in input resistance. The V(m) hyperpolarization exhibited an all-or-none dose response and was independent of baseline V(m). These differential responses to NaCN were retained after synaptic blockade with low Ca(2+)-high Mg(2+) or TTX. Thus hypoxic excitation 1) is maintained in cell culture, 2) is an intrinsic response, and 3) is likely due to the increase in a cation current. These hypoxia-excited neurons are likely candidates to function as central oxygen sensors.

Pre-Bötzinger complex functions as a central hypoxia chemosensor for respiration in vivo.

Recently, we identified a region located in the pre-Bötzinger complex (pre-BötC; the proposed locus of respiratory rhythm generation) in which activation of ionotropic excitatory amino acid receptors using DL-homocysteic acid (DLH) elicits a variety of excitatory responses in the phrenic neurogram, ranging from tonic firing to a rapid series of high-amplitude, rapid rate of rise, short-duration inspiratory bursts that are indistinguishable from gasps produced by severe systemic hypoxia. Therefore we hypothesized that this unique region is chemosensitive to hypoxia. To test this hypothesis, we examined the response to unilateral microinjection of sodium cyanide (NaCN) into the pre-BötC in chloralose- or chloralose/urethan-anesthetized vagotomized, paralyzed, mechanically ventilated cats. In all experiments, sites in the pre-BötC were functionally identified using DLH (10 mM, 21 nl) as we have previously described. All sites were histologically confirmed to be in the pre-BötC after completion of the experiment. Unilateral microinjection of NaCN (1 mM, 21 nl) into the pre-BötC produced excitation of phrenic nerve discharge in 49 of the 81 sites examined. This augmentation of inspiratory output exhibited one of the following changes in cycle timing and/or pattern: 1) a series of high-amplitude, short-duration bursts in the phrenic neurogram (a discharge similar to a gasp), 2) a tonic excitation of phrenic neurogram output, 3) augmented bursts in the phrenic neurogram (i.e., eupneic breath ending with a gasplike burst), or 4) an increase in frequency of phrenic bursts accompanied by small increases or decreases in the amplitude of integrated phrenic nerve discharge. Our findings identify a locus in the brain stem in which focal hypoxia augments respiratory output. We propose that the respiratory rhythm generator in the pre-BötC has intrinsic hypoxic chemosensitivity that may play a role in hypoxia-induced gasping.

Patterns of phrenic motor output evoked by chemical stimulation of neurons located in the pre-Bötzinger complex in vivo.

The pre-Bötzinger complex (pre-BötC) has been proposed to be essential for respiratory rhythm generation from work in vitro. Much less, however, is known about its role in the generation and modulation of respiratory rhythm in vivo. Therefore we examined whether chemical stimulation of the in vivo pre-BötC manifests respiratory modulation consistent with a respiratory rhythm generator. In chloralose- or chloralose/urethan-anesthetized, vagotomized cats, we recorded phrenic nerve discharge and arterial blood pressure in response to chemical stimulation of neurons located in the pre-BötC with DL-homocysteic acid (DLH; 10 mM; 21 nl). In 115 of the 122 sites examined in the pre-BötC, unilateral microinjection of DLH produced an increase in phrenic nerve discharge that was characterized by one of the following changes in cycle timing and pattern: 1) a rapid series of high-amplitude, rapid rate of rise, short-duration bursts, 2) tonic excitation (with or without respiratory oscillations), 3) an integration of the first two types of responses (i.e., tonic excitation with high-amplitude, short-duration bursts superimposed), or 4) augmented bursts in the phrenic neurogram (i.e., eupneic breath ending with a high-amplitude, short-duration burst). In 107 of these sites, the phrenic neurogram response was accompanied by an increase or decrease (>/=10 mmHg) in arterial blood pressure. Thus increases in respiratory burst frequency and production of tonic discharge of inspiratory output, both of which have been seen in vitro, as well as modulation of burst pattern can be produced by local perturbations of excitatory amino acid neurotransmission in the pre-BötC in vivo. These findings are consistent with the proposed role of this region as the locus for respiratory rhythm generation.

Variable extrathoracic airflow obstruction and chronic laryngotracheitis in Gulf War veterans.

STUDY OBJECTIVES: To study the flow-volume loop for evidence of variable extrathoracic airflow obstruction in Persian Gulf War veterans. DESIGN: Retrospective case-control, single-center study. SETTING: The pulmonary division of an academic health-care center. SUBJECTS: A convenience sample of the Persian Gulf Registry. MEASUREMENTS AND INTERVENTIONS: (1) Midvital capacity ratio (ratio of maximum forced midexpiratory to maximum forced midinspiratory flow). This ratio is the criterion standard for the diagnosis of variable extrathoracic airflow obstruction. (2) Evaluation of the anatomy and function of the extrathoracic airway by fiberoptic bronchoscopy. (3) Further investigation into the airway abnormality by histologic evaluation of tracheal biopsy samples in Gulf War veterans only. RESULTS: Midvital capacity was > 1.0 in 32 of 37 Gulf War veterans compared with only 11 of 38 control subjects. The mean (+/-SD) value was 1.37+/-0.4 among Gulf War veterans and 0.88+/-0.3 among control subjects (p=0.0000005). FVC and its ratio to FEV1 were normal in all these subjects. Bronchoscopy showed inflamed larynx and trachea in all (n=17) Gulf War veterans. Histologic study showed chronic inflammation of the trachea in everyone (n=12) who had an adequate biopsy sample. CONCLUSION: Physicians should be made aware of the presence of chronic inflammation of the upper airways and inspiratory airflow limitation in a number of Gulf War veterans.

Autoregressive spectral analysis of phrenic neurogram during eupnea and gasping.

During hypoxic gasping, the phrenic neurogram (PN) has a steeper rate of rise, an augmented amplitude, and a shorter duration than is seen during eupnea. Because hypoxia reduces neuronal activity, we hypothesized that gasping would be characterized in the frequency domain by enhanced low-frequency power compared with eupnea. Autoregressive (AR) spectral analysis of the PN in chloralose-anesthetized, vagotomized, peripherally chemodenervated cats was performed during eupnea and hypoxic gasping. During eupnea, significant spectral peaks were seen at 41 +/- 2 and 93 +/- 2 (SE) Hz. In all cats, the 41-Hz spectral peak disappeared during hypoxic gasping and was replaced by a high-power, low-frequency peak at 26 +/- 1 Hz. No consistent change in the frequency or power of the high-frequency spectral peak was seen during gasping. To determine whether changes in the AR spectrum of the PN during gasping result from augmented respiratory output, we compared the AR spectra of the PN during gasping, hypercapnia (end-tidal CO2 fraction = 0.09), and carotid sinus nerve stimulation. Unlike during gasping, there was no shift in power toward lower frequencies during hypercapnia and carotid sinus nerve stimulation. We conclude that the spectral characteristics of gasping, loss of the medium-frequency peak and the appearance of low-frequency (< 30-Hz) power, are unique to this respiratory pattern.

Respiration and medullary blood flow during sinusoidal hypoxia in the peripherally chemodenervated cat.

The hypothesis that hypoxic respiratory depression is mediated by changes in medullary blood flow (MBF) was assessed in 18 anesthetized, paralyzed, vagotomized, peripherally chemodenervated, ventilated cats exposed to sinusoidal hypoxic hypoxia. In nine cats, the dynamic response of the central respiratory controller to hypoxia was studied by varying the cycle time of sinusoidal hypoxia (cycle time = 2.5, 4, 6, 10, and 15 min). Peak phrenic neurogram amplitude (PNA) followed sinusoidal oscillations in the hypoxic input [arterial O2 saturation (SaO2)] at all cycle times. The relationship between PNA and SaO2 was expressed as the transfer function of the system and was approximated as a first-order differential equation with a time constant of 78 +/- 1 s, a value consistent with a previous measurement of the time constant of the change in respiration following a change in brain blood flow. In a separate study, MBF was continuously measured during sinusoidal hypoxia (cycle time = 6 min; n = 9) with a laser-Doppler flow probe to directly assess the role of MBF in production of hypoxic respiratory depression. PNA and MBF followed SaO2 oscillations during sinusoidal hypoxia. Infusion of sodium nitroprusside (20 micrograms.kg-1.min-1 iv) increased MBF by 30-40% and abolished MBF oscillations during subsequent sinusoidal hypoxia but had no effect on PNA oscillations. We conclude that the increase in brain blood flow seen during sinusoidal hypoxia is not the primary cause of the accompanying central hypoxic respiratory depression.

Respiratory and sympathetic activity during recovery from hypoxic depression and gasping in cats.

In peripherally chemodenervated, vagotomized, chloralose-anesthetized cats, hypoxia can produce central cardiorespiratory depression or excitation depending on severity. We monitored phrenic and cervical sympathetic neurograms during either hypoxic depression or gasping and 30 min of subsequent isocapnic reoxygenation to determine whether the response of these outputs during hypoxia predicts their activity during recovery. Three levels of hypoxic response were produced in cats: 1) reduction of phrenic neurogram amplitude (PNA) by 30% [fractional inspired O2 (FIO2) = 14-18%)]; 2) production of phrenic apnea (FIO2 = 9-10%); and 3) hypoxic gasping (FIO2 = 6-8%). Recovery from the milder levels of hypoxia was characterized by transient (< 10 min) depression of PNA and inspiratory synchronous sympathetic activity. Respiratory frequency was unaffected or only transiently depressed. Tonic sympathetic activity was unaffected. During reoxygenation after gasping, both PNA and inspiratory synchronous sympathetic activity were initially increased by 80% over control levels and respiratory frequency was depressed. Tonic sympathetic activity increased during hypoxia but returned to control levels after a brief undershoot on reoxygenation. All variables returned to control levels within 15 min. Measurement of medullary extracellular K+ concentration ([K+]e) in a separate group of cats indicated that a significant increase in this variable was associated with hypoxic gasping but was not correlated with PNA augmentation during reoxygenation. We hypothesize that increased [K+]e coincident with gasping may trigger a postanoxic potentiation of respiratory premotor neurons similar to that described in hippocampus.

Effects of respiratory afferent stimulation on phrenic neurogram during hypoxic gasping in the cat.

Previous studies suggested that phrenic motor output is largely refractory to afferent stimuli during gasping. We tested this concept by electrically stimulating the carotid sinus nerve (CSN) or the superior laryngeal nerve (SLN) of anesthetized peripherally chemodenervated vagotomized ventilated cats during eupnea or gasping induced by hypoxia. During eupnea, phrenic neurogram amplitude (PNA) increased by 110% during 30 s of supramaximal CSN stimulation, but burst frequency did not change. Progressive hypoxia caused gasping after arterial O2 content was reduced by 75%. During gasping, CSN stimulation caused premature onset of gasp in 12 of 13 trials, shortened intergasp interval [6.3 +/- 0.9 vs. 14.8 +/- 2.5 (SE) s], and resulted in a small (20%) but significant increase in PNA. Intensity of SLN stimulation was adjusted to abolish phrenic activity during the 30-s eupneic stimulation period. During gasping, SLN stimulation of the same intensity tended to delay onset of the next gasp, increased intergasp interval (16.9 +/- 1.9 vs. 13.3 +/- 1.2 s), and reduced PNA by 32%. Thus the respiratory burst pattern formation circuitry responds to afferent stimuli during gasping, albeit in a manner different from the eupneic response. These observations suggest that gasping is the output of a modified eupneic burst pattern formation circuit.

Strength and endurance characteristics of the normal human genioglossus.

Since activity of the genioglossus muscle plays a primary role in maintaining upper airway patency during sleep, its strength and endurance characteristics are of potential importance. The purpose of this study was 2-fold. First, to define the strength and endurance characteristics of the normal human genioglossus. Second, we hypothesized that because the genioglossus has a high proportion of fast glycolytic muscle fibers, brief periods of increased activity would make it more susceptible to fatigue. In five normal male subjects strength of the tongue was evaluated by measuring maximal anterior force using a transducer (Fmax). In each subject tongue endurance was then tested at 100%, 80%, and 50% Fmax. To test the effect of a short-term increase in genioglossal activity on its endurance, an inspiratory flow-resistive load with mild hypercapnia was presented to the upper airway for 10 min, after which genioglossal endurance at 80% Fmax was repeated. On a separate day the effect of inspiratory loading plus hypercapnia on thoracic inspiratory muscle endurance was also tested. Our results showed that mean Fmax was 1,267 +/- 125 (SEM) g. Endurance time (Tlim) decreased progressively during 50%, 80% and 100% Fmax trials. Short-term activation of the genioglossus caused a reduction in Tlim at 80% Fmax to 51.4 +/- 4.8% of its value before loading (p < 0.05). Tlim for the inspiratory muscles, however, was unaffected. We conclude that, like other skeletal muscles, genioglossal endurance is reduced as the force of contraction increases. In addition, genioglossal endurance is significantly reduced by short-term activation insufficient to fatigue the thoracic inspiratory muscles.

Phrenic and sympathetic nerve responses to glutamergic blockade during normoxia and hypoxia.

Because hypoxia increases brain extracellular glutamate levels, we hypothesized that gasping and increased sympathetic activity during severe hypoxia result from glutamergic excitation. To test this hypothesis, we exposed anesthetized paralyzed vagotomized glomectomized cats to hypoxia before and after N-methyl-D-aspartate (NMDA) glutamergic blockade (MK-801, 1 mg/kg iv) or non-NMDA blockade (NBQX, 3 mg/kg iv) while monitoring phrenic neurogram (PN) and inspiratory-synchronous (ISSN) and tonic (TSN) activity in cervical sympathetic neurogram (SN). Before hypoxia, MK-801 caused apneusis and reduced PN and ISSN amplitude by 38 and 84%, respectively, but TSN activity was unaffected. During hypoxia, MK-801 had no effect on PN gasping or TSN activity but reduced ISSN amplitude during gasping. Before hypoxia, NBQX reduced PN and ISSN amplitude by 54 and 60%, respectively but did not affect inspiratory timing or TSN activity. Gasping activity in PN and ISSN and TSN activity during hypoxia were unaffected by NBQX. We conclude that 1) ionotropic glutamergic receptor activation is important for eupneic phrenic patterning but is not involved in genesis of gasping, 2) NMDA receptor activation is involved in integration of respiratory and sympathetic activity, and 3) changes in TSN activity are independent of ionotropic glutamergic receptor activation.

Respiratory muscle acidosis stimulates endogenous opioids during inspiratory loading.

Activation of endogenous opioid pathways during intense inspiratory flow-resistive loading (IRL) results in greater inhibition of EMG activity in the external oblique (EMGeo) relative to the diaphragm (EMGdi). Dichloroacetate (DCA) abolishes opioid-mediated inhibitory influences upon these muscles, suggesting a causal relationship between respiratory muscle lactic acidosis and activation of endogenous opioid pathways, during IRL. We tested the hypothesis that a more intense acidosis of the external oblique relative to the diaphragm may be the signal that determines the differential inhibitory opioid-mediated effect upon the respiratory muscles during IRL. Unanesthetized goats were exposed to IRL (50 cm H2O/1/s) for 120 min, before and after intravenous pretreatment with DCA (50 mg/kg) or saline. We measured peak phasic EMGdi and EMGeo, and respective muscle interstitial pH (pHdi, pHeo) using flexible pH probes. After 120 min IRL with saline, pHdi, and pHeo declined by -0.12 +/- 0.03 (mean +/- SEM) and -0.20 +/- 0.04 units, respectively (p < 0.05, pHdi versus pHeo). Naloxone (NLX), 0.3 mg/kg given intravenously at this time, increased EMGdi by 26.5 +/- 6.1%, but EMGeo by 81.9 +/- 13.3% (p < 0.05, EMGdi versus EMGeo). DCA blunted both the change in pHdi and pHeo during IRL (to -0.01 +/- 0.01 and -0.08 +/- 0.03 units, respectively) (p < 0.05, DCA versus saline) and the increase in EMGdi and EMGeo with NLX (to -1.0 +/- 2.6% and 5.7 +/- 5.8%, respectively) (p < 0.05, DCA versus saline).(ABSTRACT TRUNCATED AT 250 WORDS)

Modulation of respiratory responses to carotid sinus nerve stimulation by brain hypoxia.

This study examines the effect of progressive isocapnic CO hypoxemia on respiratory afterdischarge and the phrenic neurogram response to supramaximal carotid sinus nerve (CSN) stimulation. Twelve anesthetized, vagotomized, peripherally chemodenervated, ventilated cats with blood pressure controlled were studied. During isocapnic hypoxemia, the amplitude of the phrenic neurogram was progressively depressed. In contrast, the increase in peak phrenic amplitude produced by CSN stimulation was unchanged, suggesting that the central respiratory response to CSN stimulation is unaffected by progressive hypoxemia. The time constant of respiratory afterdischarge (tau) was calculated from best-fit plots of phrenic amplitude vs. time after cessation of CSN stimulation. Under control conditions the value of tau was 57.7 +/- 3 (SE) s (n = 12). During progressive isocapnic hypoxemia, tau decreased as a linear function of arterial O2 content (CaO2) such that a 40% reduction of CaO2 resulted in a 48% reduction in tau. This reduction of respiratory afterdischarge may contribute to the genesis of periodic breathing during hypoxia.

Triangularis sterni and phrenic nerve responses to progressive brain hypoxia.

Activity of the respiratory muscles that are not normally active during eupnea (genioglossal and abdominal) has been shown to be more vulnerable to hypoxic depression than inspiratory diaphragmatic activity. We hypothesized that respiratory muscles that are active at eupnea would be equally vulnerable to isocapnic progressive brain hypoxia (PBH). Phrenic (PHR) and triangularis sterni nerve (TSN) activity were recorded in anesthetized peripherally chemodenervated vagotomized ventilated cats. Hypercapnia [arterial PCO2 (PaCO2) = 57 +/- 3 (SE) Torr] produced parallel increases in peak PHR and TSN activity. PBH [0.5% CO-40% O2-59.5% N2, arterial O2 content (CaO2) reduced from 13.1 +/- 1.0 to 3.7 +/- 0.3 vol%] resulted in parallel decreases of peak PHR and TSN activity to neural apnea. PBH was continued until PHR gasping ensued (CaO2 = 2.9 +/- 0.2 vol%); TSN activity remained silent during gasping. After 6-12 min of recovery (95% O2-5% CO2; CaO2 = 7.8 +/- 0.8 vol%; PaCO2 = 55 +/- 2 Torr), peak PHR activity was increased to 110 +/- 18% (% of activity at 9% CO2) whereas peak TSN activity was augmented to 269 +/- 89%. The greater augmentation of TSN activity during the recovery period could not be explained solely by hypercapnia. In conclusion, we found that 1) TSN expiratory and PHR inspiratory activities are equally vulnerable to hypoxic depression and 2) recovery from severe hypoxia is characterized by a profound augmentation of TSN expiratory activity.

Central respiratory carbon dioxide chemosensitivity does not decrease during sleep.

The ventilatory response to CO2 decreases during slow-wave sleep (SWS) and rapid-eye-movement (REM) sleep compared with awake levels. However, it is not known to what extent this can be attributed to decreased sensitivity of the CO2 chemoreflex. Mechanical factors during sleep may decrease ventilatory output, or PCO2 at the central chemoreceptor may not increase to the same degree as PaCO2, particularly during REM sleep when brain blood flow (BBF) is increased. In 10 goats, we measured the ventilatory (VI), diaphragmatic electromyogram (EMGdI), and BBF responses to CO2 rebreathing during each sleep-wake state. delta VI/delta PaCO2 decreased from wakefulness to SWS (p less than 0.05) and REM sleep (p less than 0.05). In contrast, delta EMGdI/delta PaCO2 was decreased only during REM sleep (p less than 0.05). Concurrently, delta BBF/delta PaCO2 increased during REM sleep (p less than 0.05) compared with the awake state or SWS. A significant reciprocal correlation existed between delta EMGdI/delta PaCO2 and delta BBF/delta PaCO2 across sleep states (r = -0.786). When EMGdI was related to directly measured cerebral venous PCO2 (n = 4), a single linear function (r = 0.894) was found, independent of sleep-wake state. Similar results were obtained during quasi-steady-state hypercapnia. We conclude that central CO2 chemosensitivity is intact during sleep.

Dichloroacetate blocks endogenous opioid effects during inspiratory flow-resistive loading.

Inspiratory flow-resistive loading (IRL) in unanesthetized goats causes central elaboration of endogenous opioids, which is accompanied by inhibition of several respiratory muscles. The peripheral stimulus responsible for mediating this phenomenon is unknown. We hypothesized that lactic acid mediates release of endogenous opioids during IRL. Unanesthetized goats were pretreated with either saline or dichloroacetate (DCA; 50 mg/kg iv), a blocker of lactic acid formation, and subjected to IRL (50 cmH2O.l-1.s) for 120 min followed by naloxone (NLX; 0.3 mg/kg iv). Electromyographic activities of the diaphragm (EMGdi), external oblique (EMGeo), and external intercostal (EMGei) were measured and expressed as a percentage of activity at an end-tidal CO2 of 8%. DCA blocked the NLX-induced augmentation of all EMGs observed after 120 min of IRL as follows (means +/- SE): delta EMGdi from 20.8 +/- 5.6% (saline) to 1.2 +/- 2.7% (DCA), delta EMGeo from 116.6 +/- 30.9% (saline) to 5.3 +/- 11.4% (DCA), and delta EMGei from 43.8 +/- 11.3% (saline) to -4.5 +/- 5.6% (DCA) (all P less than 0.05, DCA vs. saline). We conclude that lactic acid produced by the contracting respiratory muscles is the stimulus responsible for endogenous opioid pathway activation during IRL.

Responses of diaphragm and external oblique muscles to flow-resistive loads during sleep.

Although it is generally agreed that rapid respiratory compensation for externally applied inspiratory loads is impaired or absent during sleep, the individual components of the "load-compensating reflex" may not be inhibited by sleep to the same degree. We studied the effect of inspiratory flow-resistive loading (18 cm H2O/L/s) for two consecutive breaths on inspiratory (diaphragm) and expiratory (external oblique) muscle activity, and respiratory timing, in six awake and sleeping goats. During the first loaded breath in the awake state, peak integrated diaphragmatic electromyogram activity (EMGdi) increased 16.7 +/- 3.9% (p less than 0.01), peak integrated external oblique EMG activity (EMGeo) increased 21.0 +/- 7.5% (p less than 0.001), and electrical inspiratory time (Ti) increased 18.1 +/- 2.1% (p less than 0.01). In contrast, loading did not significantly change peak EMGdi or EMGeo on the first or second breaths in any sleep state. However, Ti was significantly increased during loading in all sleep states (p less than 0.01) to a similar degree seen during wakefulness. Loading did not significantly alter electrical expiratory time. No significant differences were noted between the first and second loaded breaths. We conclude that the reflex increases in peak EMG of both inspiratory and expiratory muscles in response to inspiratory flow-resistive loading during the awake state are absent during all stages of sleep; however, one aspect of load compensation, prolongation of Ti, is preserved during sleep and aids in maintaining tidal volume.

Extracellular potassium homeostasis in the cat medulla during progressive brain hypoxia.

Brain extracellular potassium [( K+]ec) in the ventral respiratory group of the medulla and the phrenic neurogram were recorded in anesthetized vagotomized peripherally chemodenervated ventilated cats during progressive isocapnic carbon monoxide (CO) hypoxia. During hypoxia, the phrenic neurogram was progressively depressed and became silent when arterial O2 content (CaO2) was reduced by 62 +/- 3% (SE). Gasping was seen in the phrenic neurogram when CaO2 was reduced by 78 +/- 1%. Medullary [K+]ec, an indicator of energy production failure due to O2 insufficiency, was 3.2 +/- 0.4 mM before hypoxia and was statistically unchanged at the onset of phrenic apnea during CO hypoxia (4 +/- 0.7 mM). By the onset of gasping, [K+]ec had increased to 6.1 +/- 1 mM, a value that tended to be different from control (P less than 0.1). After initiation of gasping, the rate of rise of [K+]ec increased, and [K+]ec reached a maximum value of 14.3 +/- 2.7 mM before hypoxia was terminated. With reoxygenation, [K+]ec returned to control levels within 20 min. On the basis of these results, we have drawn two major conclusions. 1) Hypoxic depression to the point of phrenic apnea does not appear to be caused by medullary energy insufficiency as measured by loss of [K+]ec homeostasis. 2) The rapid rise in [K+]ec in the medulla that characterizes severe hypoxia is closely associated with the onset of gasping in the phrenic neurogram, suggesting that gasping may serve as a marker for loss of medullary ionic homeostasis and thus onset of medullary energy insufficiency during hypoxia.