The upper esophageal sphincter in the cat: the role of central innervation assessed by transient vagal blockade

Studies were performed on four cats to assess the role of extrinsic innervation via the cervical nerve trunks in the control of upper esophageal sphincter function. Transient vagal nerve blockade was accomplished by cooling the cervical vagosympathetic nerve trunks previously isolated in skin loops on each side of the neck. Upper esophageal sphincter pressure was measured using a multilumen oval manometry tube and a rapid pull-through technique. The upper esophageal sphincter response to cervical intraesophageal balloon distention and acid perfusion was assessed. The feline upper esophageal sphincter has a distinct asymmetric pressure profile, whereby anterior pressure greater than posterior pressure greater than left pressure greater than right pressure. Bilateral vagal nerve blockade lowered the mean upper esophageal sphincter pressure from 18.5 +/- 1.5 to 12.0 +/- 2.8 mmHg (1 mmHg = 133.3 Pa) (p less than 0.001), with a significant reduction in pressure in all four quadrants. Intraesophageal balloon distention and acid perfusion both produced a significant increase in upper esophageal sphincter pressure. Bilateral vagal nerve blockade completely abolished the response of the upper esophageal sphincter to balloon distention and acid perfusion. We conclude that normal upper esophageal sphincter tone in the cat is partially mediated by excitatory neural input via the cervical nerve trunks, presumably via the recurrent laryngeal nerves; and cervical intraesophageal balloon distention and acid perfusion produce reflex contraction of the upper esophageal sphincter, which is dependent on neural pathways via the cervical vagal nerve trunks, but the relative contribution of afferent and efferent pathways remains unknown.     

Changes in tracheal cross-sectional area during Mueller and Valsalva maneuvers in humans

Pressure-area behavior of the excised trachea is well documented, but little is known of tracheal compliance in vivo. Extratracheal tissue pressures are not directly measurable, but transmural pressure for the intrathoracic trachea is inferred from intra-airway and pleural pressure differences. Extramural pressure of the cervical trachea is assumed to be atmospheric. The difference in transmural pressure between the intra- and extrathoracic tracheal segments should be exaggerated during Mueller and Valsalva maneuvers. We used the acoustic reflection technique to measure tracheal areas above and below the thoracic inlet during these isovolume-pressure maneuvers. We found that 10 cmH2O positive pressure increased tracheal area in the extrathoracic segment by 34 +/- 16% (mean +/- SD) and in the intrathoracic segment by 35 +/- 15%. There was a reduction in area of 27 +/- 16 and 24 +/- 14%, respectively, for the extra- and intrathoracic segments with 10 cmH2O negative pressure. We conclude that the effective transmural pressure gradients do not vary significantly between intra- and extrathoracic tracheal segments.     

Optimal electrode placement for noninvasive electrical stimulation of human abdominal muscles.

Abdominal muscles are the most important expiratory muscles for coughing. Spinal cord-injured patients have respiratory complications because of abdominal muscle weakness and paralysis and impaired ability to cough. We aimed to determine the optimal positioning of stimulating electrodes on the trunk for the noninvasive electrical activation of the abdominal muscles. In six healthy subjects, we compared twitch pressures produced by a single electrical pulse through surface electrodes placed either posterolaterally or anteriorly on the trunk with twitch pressures produced by magnetic stimulation of nerve roots at the T(10) level. A gastroesophageal catheter measured gastric pressure (Pga) and esophageal pressure (Pes). Twitches were recorded at increasing stimulus intensities at functional residual capacity (FRC) in the seated posture. The maximal intensity used was also delivered at total lung capacity (TLC). At FRC, twitch pressures were greatest with electrical stimulation posterolaterally and magnetic stimulation at T(10) and smallest at the anterior site (Pga, 30 +/- 3 and 33 +/- 6 cm H(2)O vs. 12 +/- 3 cm H(2)O; Pes 8 +/- 2 and 11 +/- 3 cm H(2)O vs. 5 +/- 1 cm H(2)O; means +/- SE). At TLC, twitch pressures were larger. The values for posterolateral electrical stimulation were comparable to those evoked by thoracic magnetic stimulation. The posterolateral stimulation site is the optimal site for generating gastric and esophageal twitch pressures with electrical stimulation.     

Factors affecting the accuracy of esophageal balloon measurement of pleural pressure in dogs.

Simultaneous measurement of esophageal and tracheal pressures during an occluded inspiratory effort was used to assess the accuracy of the esophageal balloon for measuring pleural pressure in dogs. Esophageal balloons were inserted in five mongrel dogs, and an occlusion test was performed with the balloon tip 5, 10, 15, 20, and 25 cm above the esophageal sphincter; at lung volumes of functional residual capacity (FRC) and FRC + 600 ml; and in supine and right- and left-side lying postures. The protocol was repeated in paralyzed animals. This time the occlusion test was performed by injecting air into a plethysmograph to change the body surface pressure, simulating pressure changes produced by respiratory efforts in spontaneously breathing animals. In 47% of the tests in spontaneously breathing dogs, the slope of esophageal vs. tracheal pressure varied greater than 10% from unity. After paralysis the slope did not vary greater than 5% from unity under any circumstance. These data indicate that the poorer performance of the occlusion test in nonparalyzed dogs is due to active tension in the walls of the esophagus and stress induced in the intrathoracic soft tissues by the descent of the diaphragm during a breathing effort.     

In vivo estimation of tracheal distensibility and hysteresis in normal adults

We used the acoustic reflection technique to measure the cross-sectional area of tracheal and bronchial airway segments of eight healthy adults. We measured airway area during a slow continuous expiration from total lung capacity (TLC) to residual volume (RV) and during inspiration back to TLC. Lung volume and esophageal pressure were monitored continuously during this quasi-static, double vital capacity maneuver. We found that 1) the area of tracheal and bronchial segments increases with increasing lung volume and transpulmonary pressure, 2) the trachea and bronchi exhibit a variable degree of hysteresis, which may be greater or less than that of the lung parenchyma, 3) extrathoracic and intrathoracic tracheal segments behaved as if they were subjected to similar transmural pressure and had similar elastic properties, and 4) specific compliance (means +/- SE) for the intrathoracic and bronchial segments, calculated with the assumption that transmural pressure is equal to the transpulmonary pressure, was significantly (P less than 0.05) smaller for the intrathoracic segment than for the bronchial segment: (2.1 +/- 2.0) X 10(-3) cmH2O-1 vs. (9.1 +/- 2.1) X 10(-3) cmH2O-1. Direct measurements of airway area using acoustic reflections are in good agreement with previous estimates of airway distensibility in vivo, obtained by radiography or endoscopy.     

A comparison of changes in esophageal pressure and regional juxtacardiac pressures

The relationship between esophageal pressure and juxtacardiac pressures was studied during positive end-expiratory pressure (PEEP) ventilation applied to both lungs or selectively to one lung. The experiments were performed in eight anesthetized dogs with balloon catheters in the esophagus and in the left and right pericardial and overlying pleural cavities and with an open-ended liquid-filled catheter in the pleural cavity. Bilateral PEEP (10, 20, and 30 cmH2O) caused progressive and similar increments in left and right pleural pressure. Selective PEEP, however, increased ipsilateral pleural balloon pressure more than contralateral pressure. The increase in ipsilateral pleural balloon pressure markedly exceeded the increase in esophageal pressure. There was a small increase in pleural open-ended catheter pressure that approximated the increase in esophageal pressure. During selective PEEP, pericardial balloon pressure remained uniform because of a decrease in ipsilateral pericardial transmural pressure. In conclusion, selective PEEP caused nonuniform increments in regional pleural balloon pressure. Left and right pericardial balloon pressure, however, increased uniformly with selective PEEP because of reduced ipsilateral pericardial transmural pressure. The esophageal balloon did not reflect the marked regional increments in pleural balloon pressure with selective PEEP and consistently underestimated the changes in pleural balloon pressure with general PEEP.     

Increased venous pressure causes increased thoracic duct pressure in awake sheep.

The lymph from most organs drains through the thoracic duct and into veins in the neck. We hypothesized that increases in neck vein pressure (Pnv) are reflected through the thoracic duct to the lung lymphatic-thoracic duct junction. To test this, we cannulated the lung lymphatics in the direction of flow in four sheep. We advanced each cannula until it entered the thoracic duct. Thus the pressure at the tip of the lymphatic cannula (Px) was the pressure at the outflow of the lung lymphatics. We also placed a balloon into the superior vena cava. One to two days later, we measured Px in the awake sheep as we inflated the balloon and increased Pnv in steps to 25-45 cmH2O. We found no significant differences in Px and Pnv. Furthermore, Px closely followed Pnv after each step increase in Pnv. These results support our hypothesis that increases in Pnv cause increases in the outflow pressure to lung lymphatics.     

Role of the chest wall in detection of added elastic loads

Changes in respiratory mechanical loads are readily detected by humans. Although it is widely believed that respiratory muscle afferents serve as the primary source of information for load detection, there is, in fact, no convincing evidence to support this belief. We developed a shell that encloses the body, excluding the head and neck. A special loading apparatus altered pressure in proportion to respired volume (elastic load) in one of three ways: 1) at the mouth only (T), producing a conventional load in which respiratory muscles are loaded and airway and intrathoracic pressures are made negative in proportion to volume, 2) both at the mouth and in the shell (AW), where the same pattern of airway and intrathoracic pressure occurs but the muscles are not loaded because Prs (i.e., mouth pressure minus pressure in the shell is unchanged, and 3) positive pressure in proportion to volume at the shell only, loading the chest wall but causing no change in airway or thoracic pressures (CW). The threshold for detection (delta E50) with the three types of application was determined in seven normal subjects: 2.16 +/- 0.22, 2.65 +/- 0.54, and 6.21 +/- 0.85 (SE) cmH2O/l for T, AW, and CW, respectively. Therefore the active chest wall, including muscles, is a much less potent source of information than structures affected by the negative airway and intrathoracic pressure. The latter account for the very low threshold for load detection.     

Measurement of bronchial arterial blood flow and bronchovascular resistance in sheep

We studied the bronchial arterial blood flow (Qbr) and bronchial vascular resistance (BVR) in sheep prepared with carotid-bronchial artery shunt. Nine adult sheep were anesthetized, and through a left thoracotomy a heparinized Teflon-tipped Silastic catheter was introduced into the bronchial artery. The other end of the catheter was brought out through the chest wall and through a neck incision was introduced into the carotid artery. A reservoir filled with warm heparinized blood was connected to this shunt. The height of blood column in the reservoir was kept constant at 150 cm by adding more blood. Qbr was measured, after interrupting the carotid-bronchial artery flow, by the changes in the reservoir volume. The bronchial arterial back pressure (Pbr) was measured through the shunt when both carotid-bronchial artery and reservoir Qbr had been temporarily interrupted. The mean Qbr was 34.1 +/- 2.9 (SE) ml/min, Pbr = 17.5 +/- 3.3 cmH2O, BVR = 3.9 +/- 0.5 cmH2O X ml-1 X min, mean pulmonary arterial pressure = 21.5 +/- 3.6 cmH2O, and pulmonary capillary wedge pressure (Ppcw) = 14.3 +/- 3.7 cmH2O. We further studied the effect of increased left atrial pressure on these parameters by inflating a balloon in the left atrium. The left atrial balloon inflation increased Ppcw to 25.3 +/- 3.1 cmH2O, Qbr decreased to 21.8 +/- 2.4 ml/min (P less than 0.05), and BVR increased to 5.5 +/- 1.0 cmH2O.ml-1.min (P less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)     

Gastroesophageal dynamics during immersion in water to the neck.

This investigation was undertaken to study the effect of hydrostatic pressure on gastroesophageal dynamics during immersion in thermoneutral water to the neck. In 5 healthy male subjects (normal end-expiratory), gastric pressure (PG), esophageal pressure (PE), location and pressure of distal esophageal sphincter (des), location of respiratory inversion point (RIP), and gastroesophageal pH gradient were measured standing in air (A), standing in water to the neck (B), and standing in air with abdominal compression (C). The pressure was measured with a Honeywell esophageal catheter (model 31) with built-in pressure transducer. A Beckman stomach pH electrode (no. 39042) was positioned adjacent to the pressure transducer. PG increased from 4.6 +/- 0.6 (SE) mmHg in A to nearly 20 mmHg in B and C, while PE increased from -6.0 +/- 0.8 mmHg in A to -0.8 +/- 1.0 and -3.4 +/- 0.9 mmHg in B and C, respectively. However, PDES was always 11-15 mmHg higher than PG. The superior limit of DES was displaced cephalad by indicating a stretching of DES and a shortening of the esophagus. Qualitatively similar findings were obtained in C. In all experiments, the esophageal pH remained above 6, and no alteration in the amplitude of primary peristaltic waves was seen. It is concluded that a head-out immersion with increased gastroesophageal pressure gradient predisposes to gastric reflux in the absence of a competent DES mechanism.     

Functional capacities of lungs and thorax in beagles after prolonged residence at 3,100 m

Functional capacities of the lungs and thorax in beagles taken to high altitude as adults for 33 mo or in beagles raised from puppies at high altitude were compared with functional capacities in corresponding sets of beagles kept simultaneously at sea level. Comparisons were made after reacclimatization to sea level. Lung volumes, airway pressures, esophageal pressures, CO diffusing capacities (DLCO), pulmonary blood flow, and lung tissue volume (Vt) were measured by a rebreathing technique at inspired volumes ranging from 15 to 90 ml/kg. In beagles raised from puppies we measured anatomical distribution of intrathoracic air and tissue using X-ray computed tomography at transpulmonary pressures of 20 cm H2O. Lung and thoracic distensibility, DLCO, and Vt were not different between beagles that had been kept at high altitude for 33 mo as adults and control subjects kept simultaneously at sea level. Lung distensibility, DLCO, and Vt were significantly greater in beagles raised at high altitude than control subjects raised simultaneously at sea level. Thoracic distensibility was not increased in beagles raised at high altitude; the larger lung volume was accommodated by a lower diaphragm, not a larger rib cage.     

Simulated breath-hold diving to 20 meters: cardiac performance in humans

Cardiac performance was assessed in six subjects breath-hold diving to 20 m in a hyperbaric chamber, while nonsubmersed or submersed in a thermoneutral environment. Cardiac index and systolic time intervals were obtained with impedance cardiography and intrathoracic pressure with an esophageal balloon. Breath holding at large lung volume (80% vital capacity) decreased cardiac index, probably by increasing intrathoracic pressure and thereby impeding venous return. During diving, cardiac index increased (compared with breath holding at the surface) by 35.1% in the nonsubmersed and by 29.5% in the submersed condition. This increase was attributed to a fall in intrathoracic pressure. Combination of the opposite effects of breath holding and diving to 20 m left cardiac performance unchanged during the dives (relative to the surface control). A larger intrathoracic blood redistribution probably explains a smaller reduction in intrathoracic pressure observed during submersed compared with nonsubmersed diving. Submersed breath-hold diving may entail a smaller risk of thoracic squeeze (lesser intrathoracic pressure drop) but a greater risk of overloading the central circulation (larger intrathoracic blood pooling) than simulated nonsubmersed diving.     

Volume-related and volume-independent effects of posture on esophageal and transpulmonary pressures in healthy subjects.

Ventilator management decisions in acute lung injury could be better informed with knowledge of the patient's transpulmonary pressure, which can be estimated using measurements of esophageal pressure. Esophageal manometry is seldom used for this, however, in part because of a presumed postural artifact in the supine position. Here, we characterize the magnitude and variability of postural effects on esophageal pressure in healthy subjects to better assess its significance in patients with acute lung injury. We measured the posture-related changes in relaxation volume and total lung capacity in 10 healthy subjects in four postures: upright, supine, prone, and left lateral decubitus. Then, in the same subjects, we measured static pressure-volume characteristics of the lung over a wide range of lung volumes in each posture by using an esophageal balloon catheter. Transpulmonary pressure during relaxation (PLrel) averaged 3.7 (SD 2.0) cmH2O upright and -3.3 (SD 3.2) cmH2O supine. Approximately 58% of the decrease in PLrel between the upright and supine postures was due to a corresponding decrease in relaxation volume. The remaining 2.9-cmH2O difference is consistent with reported values of a presumed postural artifact. Relaxation volumes and pressures in prone and lateral postures were intermediate. To correct estimated transpulmonary pressure for the effect of lying supine, we suggest adding 3 cmH2O (95% confidence interval: -1 to +7 cmH2O). We conclude that postural differences in estimated transpulmonary pressure at a given lung volume are small compared with the substantial range of PLrel in patients with acute lung injury.     

Airway closure in humans does not result in overestimation of plethysmographic lung volume

Exercise Physiol. 52: 638-641, 1982) have shown in dogs that airway closure may induce rib cage deformation and nonhomogeneous alveolar pressure swings, and they have suggested that this could lead to thoracic gas volume (TGV) overestimation by body plethysmography. However, in humans the rib cage is less easy to distort than in dogs. In four healthy volunteers we measured TGV by plethysmography before (B) and during (D) the occlusion of the middle and lower right lobes by a balloon (attached to a double-lumen catheter) positioned in the intermediate right bronchus. Subjects were trained to perform panting maneuvers preferentially with intercostals and accessory muscles or the diaphragm. Five to eleven TGV measurements were made in each subject with each panting pattern B and D occlusion. Balloon inflation resulted in no change in TGV whether low [13.3 +/- 3.4 (SD) cmH2O] or high (46.8 +/- 8.4 cmH2O) transdiaphragmatic pressures (Pdi) were used: TGV 4.0 +/- 0.4 (B) vs. 4.0 +/- 0.4 liters (D) and 4.3 +/- 0.4 (B) vs. 4.3 +/- 0.4 liters (D) for low and high Pdi, respectively. Thus, in trained subjects performing maneuvers aimed to distort the rib cage, no pressure difference was observed between the occluded and the nonoccluded lung during panting against the closed shutter. We conclude that it is unlikely that the mechanism proposed by Brown et al. might explain errors in lung volume measurements by body plethysmography in humans.     

Pulmonary hypertension induced with an intra-arterial balloon: an alternate mechanism

The Laks catheter is a triple-lumen balloon catheter used to distend the canine main pulmonary artery while recording right ventricular pressure and the arterial pressure distal to the balloon. A rise in arterial pressure reported to occur during distension has been attributed to vasoconstriction rather than passive obstruction by the balloon. We tested this in six anesthetized dogs by inflating the Laks catheter-balloon while recording pressure distal to the balloon from the Laks catheter as well as from additional catheters in right and left pulmonary arteries placed retrogradely through lobar branches following thoracotomy. We found that balloon inflation increased pressures in the arterial port of the Laks catheter and in the left pulmonary artery catheter but reduced it in the right pulmonary artery. Tightening a snare around the right pulmonary artery had the same effects on pressures. Similar results were obtained while cardiac output was controlled by left ventricular bypass perfusion in four dogs. We conclude that the Laks catheter-balloon obstructs flow to the right lung and that the arterial pressure rise recorded in it during balloon inflation cannot be distinguished from that caused by occlusion of the right pulmonary artery.     

Atypical localization of myenteric neurons in the opossum lower esophageal sphincter

This study investigated sphincter-body differences in neuronal density and morphometry between the esophageal sphincter and body with a view to determining whether previously reported differences are authentic. The anatomical limits of the opossum lower esophageal sphincter were correlated with its physiological behavior by manometric demarcation. Following this, peeled whole mounts and paraffin and cryosections were used to study the morphology and morphometry of the esophageal myenteric plexus. Thirty animals were used and seven quantitated. The plexus of the esophageal body was located as usual in a plane between the longitudinal and circular muscle, which coincided with the plane of cleavage when these muscle layers were peeled apart for studying the plexus in whole mounts. In contrast, the plexus was located in several planes in the lower esophageal sphincter, which had no cleavage plane. Therefore, peeling the sphincter removes neurons and yields falsely low counts, making peel preparations of this region unsuitable for neuronal quantitation. In paraffin sections, the neuron density in the esophageal body 7 cm above the sphincter was 6,353 +/- 850/cm2, but decreased significantly to 2,254 +/- 353/cm2 at the 1-cm segment. In the lower esophageal sphincter, the neuronal count increased again to 8,530 +/- 1,606/cm2. Flash-frozen cryosections, which produced neuronal morphology similar to the in vivo condition, showed that there was no difference in neuronal size between esophageal body and sphincter. These studies show that atypical myenteric plexus localization causes spuriously low neuronal counts reported in the lower esophageal sphincter and that reported neuronal size differences are technique-dependent.     

Mechanical properties of porcine intralobar pulmonary arteries

The isobaric and isovolumetric properties of intrapulmonary arteries were evaluated by placing a highly compliant balloon inside arterial segments. The passive pressure-volume (P-V) curve was obtained by changing volume (0.004 ml/s) and measuring pressure. The isobaric active volume change (delta V) or isovolumetric active pressure change (delta P) generated by submaximal histamine was measured at four different transmural pressures (Ptm's) reached by balloon inflation. The maximal delta P = 11.2 +/- 0.6 cmH2O (mean +/- SE) was achieved at 30.8 +/- 1.2 cmH2O Ptm and maximal delta V = 0.20 +/- 0.02 ml at 16.7 +/- 1.7 cmH2O Ptm. The P-V relationships were similar when volume was increased after either isobaric or isovolumetric contraction. The calculated length-tension (L-T) relationship showed that the active tension curve was relatively flat and that the passive tension at the optimal length was 149 +/- 11% of maximal active tension. These data show that 1) a large elastic component operates in parallel with the smooth muscle in intralobar pulmonary arteries, and 2) the change in resistance associated with vascular expansion of the proximal arteries is independent of the type of contraction that occurs in the more distal arterial segments.     

Diaphragmatic electromyography using a multiple electrode array

We have developed a new technique for diaphragmatic electromyography using an array of seven sequential electrode pairs at 1.0-cm spacing on an esophageal catheter. This array provides information about the spatial distribution of the electrical field generated by the diaphragm and reveals a sharply peaked variation of electrical potential with distance along the esophagus. The rectified and integrated information from each of the seven pairs is summed to give an approximation to the total electrical activity over the span of the array, providing a signal that is relatively insensitive to the position of the array over approximately 4 cm of catheter movement and removes the requirement for balloon stabilization of the catheter. With our array, we have confirmed the artifact in the evoked compound muscle action potential that seems to be related to diaphragmatic shape as reported by others who used supramaximal phrenic nerve stimulation, but the magnitude of this artifact (compared with the functional residual capacity level) was modest near functional residual capacity, averaging 12 +/- 14% (SD) for lung volumes 1.0 l above and -4 +/- 15% for lung volumes 1.0 l below functional residual capacity along the rib cage-abdomen relaxation line.     

Effect of hypoproteinemia on lung fluid balance in awake newborn lambs

To study the influence of plasma protein concentration on fluid balance in the newborn lung, we measured pulmonary arterial and left atrial pressures, lung lymph flow, and concentrations of protein in lymph and plasma of eight lambs, 2-3 wk old, before and after we reduced their plasma protein concentration from 5.8 +/- 0.3 to 3.6 +/- 0.6 g/dl. Each lamb underwent two studies, interrupted by a 3-day period in which we drained protein-rich systemic lymph through a thoracic duct fistula and replaced fluid losses with feedings of a protein-free solution of electrolytes and glucose. Each study consisted of a 2-h control period followed by 4 h of increased lung microvascular pressure produced by inflation of a balloon in the left atrium. Body weight and vascular pressures did not differ significantly during the two studies, but lung lymph flow increased from 2.6 +/- 0.1 ml/h during normoproteinemia to 4.1 +/- 0.1 ml/h during hypoproteinemia. During development of hypoproteinemia, the average difference in protein osmotic pressure between plasma and lymph decreased by 1.6 +/- 2 Torr at normal left atrial pressure and by 4.9 +/- 2.2 Torr at elevated left atrial pressure. When applied to the Starling equation governing microvascular fluid balance, these changes in liquid driving pressure were sufficient to account for the observed increases in lung fluid filtration; reduction of plasma protein concentration did not cause a statistically significant change in calculated filtration coefficient. Protein loss did not influence net protein clearance from the lungs nor did it accentuate the increase in lymph flow associated with left atrial pressure elevation.(ABSTRACT TRUNCATED AT 250 WORDS)     

Intravascular pressure profiles in elephant seals: hypotheses on the caval sphincter, extradural vein and venous return to the heart

In order to evaluate hemodynamics in the complex vascular system of phocid seals, intravascular pressure profiles were measured during periods of rest-associated apnea in young elephant seals (Mirounga angustirostris). There were no significant differences between apneic and eupneic mean arterial pressures. During apnea, venous pressure profiles (pulmonary artery, thoracic portion of the vena cava (thoracic vena cava), extradural vein, and hepatic sinus) demonstrated only minor, transient fluctuations. During eupnea, all venous pressure profiles were dominated by respiratory fluctuations. During inspiration, pressures in the thoracic vena cava and extradural vein decreased -9 to -21 mm Hg, and -9 to -17 mm Hg, respectively. In contrast, hepatic sinus pressure increased 2-6 mm Hg during inspiration. Nearly constant hepatic sinus and intrathoracic vascular pressure profiles during the breath-hold period are consistent with incomplete constriction of the caval sphincter during these rest-associated apneas. During eupnea, negative inspiratory intravascular pressures in the chest ("the respiratory pump") should augment venous return via both the venae cavae and the extradural vein. It is hypothesized that, in addition to the venae cavae, the prominent para-caval venous system of phocid seals (i.e., the extradural vein) is necessary to allow adequate venous return for maintenance of high cardiac outputs and blood pressure during eupnea.