- Email: [email protected]
Stimulus-response curves of the lung inflation cardio-depressor reflex
259
Respiration Physiology (1984) 57, 259-268 Elsevier
STIMULUS-RESPONSE CURVES OF THE LUNG INFLATION CARDIO-DEPRESSOR REFLEX*
SHARON S. CASSIDY** Pauline and Adolph Weinberger Laboratory for Cardiopulmonary Research, Department of Internal Medicine, Southwestern Medical School, University of Texas Health Science Center at Dallas, TX 75235, U.S.A.
Abstract. In order to determine the relationship between extent of lung expansion and reflex depression of cardiovascular function in dogs, we used a preparation in which the left lung, isolated in situ, was subjected to a series of inflations ranging between 5 and 55 era H20 (60 and 800 ml) before and following left cervical vagotomy. The threshold level of left lung inflation that would cause bradycardia and hypotension was 15 cm H20 transpulmonary pressure (63 ml) when the preceding level of inflation pressure was lower, and 10 cm H20 (102 ral) when the preceding level was higher. Increasing inflation pressure and volume above threshold produced a graded fall in heart rate and blood pressure until maximum expansion was reached at 40 era H20 (778 ml). Maximum expansion caused a tran~ent 45~o fall in heart rate and 30~ fall in blood pressure. Division of the ipsilateral (left) cervical vagosympathetic trunk eliminated these responses to unilateral lung inflation conftrming the predominant, if not exclusive, afferent pathways. These data suggest that the lungs, as a function of the degree of expansion, impart a control over the neural regulation of the cardiovascular system. Blood pressure Bradycardia Cardiac frequency
Cardiovascular reflexes Dog Hypotension
Inflation Lung inflation Vagus nerve
Lung expansion with large lung volumes has been shown repeatedly to produce hypotension or relaxation of peripheral vascular smooth muscle and bradycardia (Anrep et al., 1936; Salisbury et al., 1959; Daly et al., 1967; Glick et al., 1969; Hainsworth, 1974; Mancia and Donald, 1975; Lloyd, 1978; Cassidy et al., 1979). Daly et al. (1967) and Lloyd (1978) demonstrated a direct relationship between Accepted for publication 25 May 1984 * A preliminary report was presented to the American Physiological Society, 16 October, 1979, in New Orleans, LA (Cassidy and Johnson, 1979). ** Address for reprint requests: Sharon S. Cassidy, M.D., Research Associate Professor of Medicine, University of Texas Health Science Center, 5323 Harry Hines Boulevard, Dallas, TX 75235, U.S.A. 0034-5687/84/$03.00 © 1984 Elsevier Science Publishers B.V.
260
s.s. CASSIDY
the level of lung inflation pressures and the reflex fall in systemic vascular resistance using 10 cm H20 increments in inflating pressure between 10 and 30 cm H20, and Hainsworth (1974) found a proportional bradycardia in open chest dogs with inflation pressures between 10 and 40 cm H20. Previously, Kaufman et al. (1982) have reported that lung afferent fibers discharge proportional to the degree of lung expansion with different thresholds and sensitivities depending on the fiber type. The purpose of the present study is to determine more precisely the sensitivity of the cardiovascular lung inflation reflex. To quantitate blood pressure and heart rate responses that are caused only by reflexes originating in the lung, we used a canine preparation in which the left airway and left pulmonary artery were isolated, and in which the thorax was opened widely (Cassidy et al., 1979); thus, cardiac output traversed the right lung, and inflation of the left lung could be accomplished without directly altering pulmonary blood flow or cardiac mechanics.
Methods
Experimental animal preparation. Seven mongrel dogs of either sex weighing 16-25 kg (17.6 + 2.8 SD) were anesthetized with morphine (1-2 mg/kg, i.m.) and ~-chloralose (50-100 mg/kg, i.v.). They were intubated through a tracheostomy with a double lumen endotracheal tube constructed with an inflatable cuff on the distal end that enabled separation of the right and left mainstem bronchi. Initially, both lungs were ventilated with a tidal volume of 15 ml/kg at a rate to maintain Paco2 at 36-40 mm Hg. The heart and lungs were exposed through a midline sternotomy, and the left pulmonary artery was ligated, directing all pulmonary blood flow to the right lung. At this time, all ventilation also was directed to the right lung, and expiratory pressure was elevated to 2 cm H20 to prevent atelectasis. Systemic arterial pressure was measured via a fluid-filled catheter placed in a femoral artery, and heart rate was monitored with a lead II electrocardiogram. Right and left bronchial pressures were monitored at the proximal ends of the double lumen endotracheal tube. The volume of left lung inflation was measured by integrating the left airway flow signal using a screen pneumotachograph and differential pressure transducer. All signals were recorded on a polygraph recorder. Experimentalprotocols. The left lung was inflated to the desired pressure with a gas mixture of 5 ~ CO2 in a balance of air. The lung was inflated for one min at five-rain intervals. Between inflations, it was deflated to 2 cm H20 and was not ventilated. Beginning with an initial inflation pressure of 5 cm H20, inflation pressure was increased serially by 5 cm H20 until an inflation pressure of 60 cm H20 was reached. Subsequently, pressures were decreased serially by 5 cm H20 decrements. Next, the left cervical vagosympathetic trunk was severed, and the serial inflations were repeated. Maximum fall in systemic arterial blood pressures
261
LUNG INFLATION REFLEX
AIRWAY PRESSURE
(cm HL,O)
20
o r
SYSTEMIC ARTERIAL PRESSURE (mmHg)
0
......
:
I
t0 S
I
Fig. 1. A polygraph recording representing airway pressure and systemic arterial blood pressure before and during left lung inflation. Either right or left airway pressure is displayed depending on which direction a stopcock was turned. Right airway pressure was recorded up to point (A) at which time the stopcock was switched to display left airway pressure in order to visualize the inflation of the left lung to 30 cm H20 at point (B). Inflation was continued for 1 min although the recording of airway pressure was switched back to the right airway after approximately 10 see of inflation, point (C), in order to ascertain whether right airway pressure had been altered by the left lung inflation. Lung inflation at (B) was associated with an abrupt fall in blood pressure and heart rate.
was obtained by inspection of the tracing, fig. 1, and heart rate was counted for a period of 10 see coinciding with the maximum fall in systemic arterial pressure. The airway pressure signals were closely examined with each inflation to ascertain that right airway pressure did not increase which indicated that separation of the airways was complete. Femoral arterial blood was obtained anaerobically at the beginning, middle, and end of each experiment for analysis of pH, Pao_~ and Paco:. Supplemental oxygen was administered for Pao2 less than 60 mm Hg. At the end of the experimental protocol, the dogs were killed by exsanguination. The lungs were removed and allowed to passively drain and were dissected free of extrapulmonary bronchi and vessels. Each lung was weighed separately, dried in an oven at 85 °C, and reweighed daily until the weight was stable.
Calculations. Results are reported as means + SD unless otherwise indicated. Student's two-tailed t-test for paired data was used to test the significance of changes that occurred with inflation, and to analyze the significance of differences in responses occurring after left vagotomy.
Results
Inflation pressure/volume relationship of the left lung (fig. 2). On the increasing inflation pressure limb of the inflation pressure-volume curve (fig. 2), a pressure of 15 cm H20 was required before significant volume change occurred. Only
262
s.s. CASSIDY 800
i
60C
z 2oo
%
to
do do
I
70
LEFTLUNGINFLATION PRESSURE (CMH20)
Fig. 2. Pressure-volume relationships for the left lung before (~-------A) and after (A------A) ipsilateral (left) vagosympatheetomy.Arrows indicate the direction of the pressure--volumehistory of the stepwise inflations. Each symbol represents the mean of seven animals. 10 cm H20 was required to elevate volume significantly on the decreasing inflation pressure limb. Maximum inflation volumes o f the left lung were the same before vagotomy, 778 +.66 ml, compared to afterwards, 777 + 111 ml. This occurred at inflation pressures o f 40 cm H20 when previous inflations had been as large or larger. At the steep portion of the inflation pressure-volume curve, we observed that inflation volumes were two to three times greater on the descending limb as compared to the ascending limb. Ipsilateral (left) vagotomy did not alter the descending limb of the inflation pressure/volume curve.
Relationship between lung expansion and reflex cardiovascular responses. During left lung inflation, heart rate and systemic arterial pressure began to fall within 2-4 sec (fig. 1), reaching a maximum response within 8-10 sec, slowly returning toward the preinflation level, but the extent of this return was not quantitated. Heart rate. The maximum heart rate responses to the series of lung inflations before and after left vagotomy are illustrated in fig. 3. The inflation threshold causing a significant reflex bradycardia was 15 cm 1-I20 (63 ml) during the incremental series of lung inflations, and 10 crn 1-120 (102 ml) during the decremental series. Inflations above these threshold levels caused graded responses until maximum lung expansion was attained. Further increasing inflation pressure did not cause a further drop in heart rate response if volume could not be expanded. Minimal hysteresis o f inflation pressures vs heart rate responses was observed, although the limb of incremental pressures tended to give smaller heart rate responses than did inflations on the decremental pressure limb. Plotting inflation volume rather than pressure not only eliminated the hysteresis in the inflation vs heart rate responses, but actually reversed it.
263
LUNG INFLATION REFLEX
10 0 ,-t0 -20 "r
.-.q -30 t~
-50 -60 I
0
I
I
I
I
I
I
I
I
I
10 20 30 40 50 60 0 200 400 600 800 Inflation Pressure (cm H20) Inflation Volume (ml)
Fig. 3. Heart rate responses to unilateral (left) lung inflation before ( ) and after (............. ) ipsilateral vagosympathectomy. Arrows indicate the direction of the volume history of the stepwise left lung inflations. Each point represents the average from seven dogs; vertical bars represent the SE.
Vagotomy eliminated the bradycardia responses entirely and exposed a transient mild tachycardia at the higher inflations occurring at the same time interval. Blood pressure. Systemic arterial pressure response to lung inflation also was proportional to the inflation pressure (fig. 4). The average threshold inflation to cause a significant depressor response was the same as for heart rate. The maximum fall in blood pressure occurred at lung inflations within 100 ml of maximum volume. In spite of the striking degree of pressure-volume hysteresis, blood pressure responses were not significantly different with respect to inflation pressure history. Similar to the heart rate responses, the hysteresis of the inflation volume/blood pressure response was reversed, i.e. incremental inflations gave larger responses than did the inflations of decreasing pressures. Left vagotomy reduced the blood pressure responses to lung inflation by twothirds to three-quarters of the pre-vagotomy responses. These responses remaining post vagotomy were observed only at maximum lung expansion. Arterial blood gas tensions and pH were within the normal range (Pao2 83 + 17 torr; Paco2, 41 + 6 torr; pH, 7.40 + 0.04) and did not change significantly as a consequence of the 30+ inflations or with ipsilateral vagotomy. Wet weight/dry weight measurements of the left lung (5.18 + 0.22) were increased compared to the right (4.27 + 0.10; P < 0.05).
264
S.S. CASSIDY t0 "
o ....
D.
-10
_
o
•-
-20
._u
~-30
(n - 4 0 .E -50 ¢J I
I
I
I
I0
20
30
40
/
50
I
I
I
I
I
60
200
400
600
800
Inflotion Pressure (cmH20)
Inflotion Volume (ml)
Fig. 4. Systemic arterial pressure responses to unilateral (left) lung inflation before ( ) and after (............ .) ipsilateral vagosympathectomy. Refer to legend of fig. 3 for explanation of symbols.
Discussion
When cardiovascular reflex responses to lung expansion are measured, a preparation is necessary in which the mechanical consequences of expansion are eliminated, allowing only indirect (reflex or humoral) responses to occur. The reflex nature of the responses must be established or confirmed by demonstrating the essential afferent or efferent pathways. This preparation differs from total cardiopulmonary bypass used by others to study cardiovascular reflex effects of lung inflation in that cardiac output was intact in this unilateral lung preparation, whereas circulation is provided at a fixed rate during cardiopulmonary bypass. Utilizing this unilaterally isolated left lung, we previously demonstrated that inflation to 30 cm H20 substantiallY reduced heart rate, blood pressure, cardiac output and stroke volume which return toward control, but in general do not reach, control values by 15 min (Cassidy et al., 1979). The present experiments were performed to quantitate the lung expansion at which these reflex responses to lung inflation were initiated and to establish the relationship between inflation and reflex responses throughout the entire range of lung expansion. This method of lung expansion, a l-rain inflation executed by exposing the airway to a square-wave of 5% CO2 in air at the desired pressure, was not meant to mimic any physiological event. Rather, this method of lung expansion was used in order to single out the potential cardiovascular reflex responses in much the same way as barosensitive segments of the circulation have been isolated and stretched with a square wave to study potential reflex responses initiated by elevated pressures in a specific vascular region. Before examining the reflex responses to lung inflation, it was necessary to know the pressure-volume relationship of the lung in this preparation.
LUNG INFLATION REFLEX
265
Pressure-volume curves of the left lung (fig. 2) Once the lung had been expanded repeatedly to pressures greater than or equal to 30 crn H20, additional inflations to 30 cm H20 pressure consistently expanded the left lung to approximately 90% of its total capacity. Ipsilateral vagotomy did not alter the relationship between inflation pressure and volume when the steps of inflation pressure were decreasing; therefore it seemed reasonable to compare the cardiovascular responses to lung inflation that were obtained before vagotomy with those obtained after vagotomy at comparable pressures or volumes using the descending limbs of the inflation pressures. The differences in the pressurevolume relationships during the incremental inflation pressures after vagotomy compared with corresponding inflations before vagotomy were thought to be the consequence of the previous inflation pressure history of the post-vagotomy inflations, but a change in compliance induced by the vagotomy cannot be excluded. Hysteresis in the pressure-volume relationship was an expected finding. It is worthy of mention that with inflations between 15 and 30 cm H20 pressure, lung volume at a given pressure might vary threefold depending on the volume history. Lung volume did not expand greatly at either extreme of inflation pressures, (increasing by approximately 100 ml between 0 and 10 cm H20 inflation pressure and increasing less than 100 ml above 30 cm H20 inflation pressure). Presumably, increased surface tension was responsible for the stiffness of the lung at very low lung volumes, and reaching the limits of elasticity was responsible for the stiffness at large lung volumes. Reflex cardiovascular responses to lung expansion Within the limits of lung expansion, heart rate and blood pressure responses were directly related to inflation pressures since there was little or no hysteresis to the inflation pressure/cardiovascular response relationships (figs. 3 and 4). However, at either extreme of lung expansion when there was little change in volume as inflation pressures were increased, there was correspondingly little change in the heart rate or blood pressure responses. Thus, transmission of pressure changes to responsive elements in the lung parenchyma that could initiate the reflex presumably was limited by surface forces at low lung volumes and by collagen at high lung volumes. In the mid range, these dements that initiate the reflex are responsive to either pressure or volume changes. Heart rate. Previously reported data regarding the effects of increasing lung volume on heart rate have shown diverse responses. Anrep et al. (1936) demonstrated a transient tachycardia at levels of inflation 9 cm 1-120 and less; however, with 18 cm H20 inflating pressure, the transient tachycardia was followed by a sustained bradycardia. Re.cently, Vatner and Rutherford (1981) demonstrated that stimulation of the Carotid bodies with hypoxia often produced tachycardia which was secondary to the hyperpnea that resulted. By controlling ventilation during hypoxic stimulation of the carotid bodies, the tachycardia that they reported was
266
s.S. CASSIDY
eliminated. The origin in the lungs of the reflex tachycardia response to hypoxic stimulation of the carotid bodies was confirmed when division of the pulmonary branches of the vagi eliminated or attenuated the response. These studies are important to illustrate that one mechanism which increases lung volume, hyperpnea, seems to produce reflex tachycardia. Alternatively, a sustained increase in lung volume such as was used by Glick et al. (1969) and in our present and previous experiments (Cassidy et al., 1979) resulted in bradycardia. Since heart rate change was determined in the present experiments by examining the tracing for the lowest rate that occurred during the inflation and counting beats for I0 sec (fig. 1), this precluded our identifying increases in heart rate lasting only 3 to 4 beats which may have preceded the onset of the bradycardia. Hainsworth (1974) offered valuable insight into the issue of the variable heart rate response to lung inflation by examining heart rate response to lung inflation under a variety of circumstances. In closed chest animals with both lungs perfused by the heart, he showed that inflation of one lung, which produced changes in intrathoracic pressure and shifts in thoracic blood volume between the lungs, resulted in tachycardia. Whereas, inflation of one lung in an open chest animal in whom blood flow had been diverted previously to the opposite lung produced bradycardia that was proportional to inflation pressure between 10 and 40 cm 1-120. From all the above cited data, we conclude that the reflex chronotropic response to sustained expansion of the lung is negative and is a direct function of lung inflation pressure within the constraints of induced changes in lung volume. B l o o d pressure. Primary determinants that would result in lowering blood pressure must be derived from previous experiments. A graded fall in systemic vascular resistance during lung inflation in dogs with cardiopulmonary bypass has been demonstrated clearly by Daly et al. (1967) and by Lloyd (1978), both of whom point out that there may be variations in the responses in the vascular beds of different organs. Hainsworth (1974) demonstrated that variability in responses of a specific organ may occur also since the hindlimb vessels both vasoconstricted and dilated with inflation. We previously demonstrated that lung expansion to 30 cm H20 inflation pressure produced a sustained reduction in stroke volume and cardiac output (Cassidy et ai., 1979). Thus, the fall in blood pressure in the present study does not define a specific mechanism, but serves to illustrate the magnitude of reflex systemic hypotension that will result from the vasodilation and fall in cardiac output with lung expansion. As with the heart response to lung inflation, the hypotensive response is sensitive to inflation pressure except at the extremes of expansion. This indicates that blood pressure will be depressed reflexly by sustained lung volume expansion within physiological constraints of lung distensibility. This concept is supported by the studies of Mancia and Donald (1975) who demonstrated that a component of tonic inhibition of the vasomotor center occurred at resting lung volume and had originated in the lungs.
LUNG INFLATION REFLEX
267
Systemic chemoreceptors and baroreceptors were exposed to the systemic circulatory milieu in our studies. Yet, there are no reasons to suspect that transient changes had occurred in blood gases or chemistry great enough to have initiated the responses that were observed. This is so because ventilation to the right lung, which served as the gas exchanger for the entire cardiac output, was not appreciably altered by the step inflation procedure of the left lung. This was documented by the right airway pressure not having changed as a consequence of left lung inflation (fig. 1). Since the arterial baroreceptors were not perfused separately at constant pressures, they undoubtedly buffered the maximum bradycardia and hypotension that developed with inflation.
Neural afferent pathways The negative chronotropic response to lung inflation was entirely eliminated by left cervical vagosympathectomy indicating that the bradycardia was reflex in nature initiated by afferent receptors within the distribution of the ipsilateral vagal nerve. Using a similar preparation, Kaufman et al. (1982) in our laboratories demonstrated that bronchial and pulmonary C-fibers were stimulated in proportion to the extent of lung expansion with thresholds comparable to that initiating the reflex responses, indicating that the reflex responses of lung inflation would be consistent with a C-fiber reflex. However, myelinated fibers are also stimulated in proportion to the degree of lung expansion. Whereas, ipsilateral vagotomy always eliminated the bradycardia response t o lung inflation, a hypotensive response to inflation, especially at large inflations, remained to varying degrees after ipsilateral vagotomy. This was observed also by Daly et ai. (1967) who found the hypotensive response to lung inflation that remained post vagotomy could be eliminated either by crushing the nerlees at the hilum of each lobe or by destroying the stellate ganglia. This suggested that some afferent fibers responsible for the reflex hypotension may course via the thoracic nerves and the stellate ganglia. The small hypotensive responses to very large lung inflations that remained after ipsilateral vagotomy may have been either mechanical or reflex in nature; nevertheless, the majority (two-thirds) of the hypotensive response to lung inflation is a reflex mediated via the ipsilateral vagal nerve.
Significance of the lung inflation depressor reflex It is clear from the literature cited above that mechanical stretching of the lung is capable of generating cardiovascular reflexes that range from tachycardia and vasoconstriction to profound bradycardia, hypotension and vasodilation. While the specific source of reflex responses to increasing rates of inspiration and tidal volume variation is still unresolved, it seems certain that a sustained elevation of lung volume is capable of producing a graded inhibition of the heart rate and blood pressure. The physiological function of this potent cardioinhibitory reflex
268
~s,s. CASSIDY
remains a mystery. Perhaps this reflex contributes to cardiovascular responses to hyperpnea of exercise and to arising from supine posture. From a pathophysiological viewpoint this reflex may produce undesired cardiovascular inhibition in pathological conditions which raise lung volume such as acute asthma with air trapping or during ventilation with positive end-expired pressure in which functional residual capacity of the lung is elevated. The reflex responses to unilateral lung inflation clearly establish that substantial reflex lowering of heart rate and blood pressure can be generated by lung expansion throughout a range of lung volumes that would be expected to occur physiologically.
Acknowledgements The author gratefully acknowledges the technical assistance of Barbara Smith and Taiwan Huang and the help of S.M. Self in preparation of the manuscript. Grant-in-aid support was provided by the American Heart Association, Texas Affiliate and by National Institutes of Health HL 22589. Portions of this work were completed during theauthor's tenure as an Established Investigator of the American Heart Association.
References Anrep, G.V., W. Pascual and R. Rossler (1936). Respiratory variations of the heart rate. I. The reflex mechanisms of the respiratory arrhythmia. Proc. R. Soc. Lond. Set. B l 19:191-217. Cassidy, S.S. and R.L. Johnson, Jr. 0979). Pressure-volume (P-V) characteristics of the reflex cardiovascular (CV) response to lung inflation in dogs. The Physiologist 22: 18. Cassidy, S.S., W.L. Eschenbacher and R.L. Johnson, Jr. (1979). Reflex cardiovascular depression during unilateral lung hyperinflation in the dog. J. Clin. Invest. 64: 620-626. Duly, M.D., J.L. Hazzledine and A. Ungar (1967). The reflex effects of alterations in lung volume on systemic vascular resistance in the dog. J. Physiol. (London) 188: 331-351. Glick, G., A.S. Wechsler and S.E. Epstein (1969). Reflex cardiovascular depression produced by stimulation of pulmonary stretch receptors in the dog. J. Clin. Invest. 48: 467-473. Hainsworth, R. (1974). Circulatory responses from lung inflation in anesthetized dogs. Am. J. Physiol. 226: 247-255. Kaufman, M.P., G.A. Iwamoto, J.H. Ashton and S.S. Cassidy (1982). Responses to inflation of vagal afferents with endings in the dog lung. Circ. Res. 51 : 525-531. Lloyd, T.C., Jr. (1978). Reflex effects of lung inflation and inhalation of halothane, ether and ammonia. J. Appl. Physiol. 45: 212-218. Mancia, G. and D.E. Donald (1975). Demonstration that the atria, ventricles, and lungs each are responsible for a tonic inhibition of the vasomotor center in the dog. Circ. Res. 36: 310-318. Salisbury, P.F., P.M. GaUetti, R.J. Lewin and D.A. Rieben (1959). Stretch reflexes from the dog's lung to the systemic circulation. Circ. Res. 7: 62-67. Vatner, S.F. and J.D. Rutherford (1981). Interaction of carotid chemoreceptor and pulmonary inflation reflexes in circulatory regulation in conscious dogs. Fed. Proc. 40:2188-2193.
Respiration Physiology (1984) 57, 259-268 Elsevier
STIMULUS-RESPONSE CURVES OF THE LUNG INFLATION CARDIO-DEPRESSOR REFLEX*
SHARON S. CASSIDY** Pauline and Adolph Weinberger Laboratory for Cardiopulmonary Research, Department of Internal Medicine, Southwestern Medical School, University of Texas Health Science Center at Dallas, TX 75235, U.S.A.
Abstract. In order to determine the relationship between extent of lung expansion and reflex depression of cardiovascular function in dogs, we used a preparation in which the left lung, isolated in situ, was subjected to a series of inflations ranging between 5 and 55 era H20 (60 and 800 ml) before and following left cervical vagotomy. The threshold level of left lung inflation that would cause bradycardia and hypotension was 15 cm H20 transpulmonary pressure (63 ml) when the preceding level of inflation pressure was lower, and 10 cm H20 (102 ral) when the preceding level was higher. Increasing inflation pressure and volume above threshold produced a graded fall in heart rate and blood pressure until maximum expansion was reached at 40 era H20 (778 ml). Maximum expansion caused a tran~ent 45~o fall in heart rate and 30~ fall in blood pressure. Division of the ipsilateral (left) cervical vagosympathetic trunk eliminated these responses to unilateral lung inflation conftrming the predominant, if not exclusive, afferent pathways. These data suggest that the lungs, as a function of the degree of expansion, impart a control over the neural regulation of the cardiovascular system. Blood pressure Bradycardia Cardiac frequency
Cardiovascular reflexes Dog Hypotension
Inflation Lung inflation Vagus nerve
Lung expansion with large lung volumes has been shown repeatedly to produce hypotension or relaxation of peripheral vascular smooth muscle and bradycardia (Anrep et al., 1936; Salisbury et al., 1959; Daly et al., 1967; Glick et al., 1969; Hainsworth, 1974; Mancia and Donald, 1975; Lloyd, 1978; Cassidy et al., 1979). Daly et al. (1967) and Lloyd (1978) demonstrated a direct relationship between Accepted for publication 25 May 1984 * A preliminary report was presented to the American Physiological Society, 16 October, 1979, in New Orleans, LA (Cassidy and Johnson, 1979). ** Address for reprint requests: Sharon S. Cassidy, M.D., Research Associate Professor of Medicine, University of Texas Health Science Center, 5323 Harry Hines Boulevard, Dallas, TX 75235, U.S.A. 0034-5687/84/$03.00 © 1984 Elsevier Science Publishers B.V.
260
s.s. CASSIDY
the level of lung inflation pressures and the reflex fall in systemic vascular resistance using 10 cm H20 increments in inflating pressure between 10 and 30 cm H20, and Hainsworth (1974) found a proportional bradycardia in open chest dogs with inflation pressures between 10 and 40 cm H20. Previously, Kaufman et al. (1982) have reported that lung afferent fibers discharge proportional to the degree of lung expansion with different thresholds and sensitivities depending on the fiber type. The purpose of the present study is to determine more precisely the sensitivity of the cardiovascular lung inflation reflex. To quantitate blood pressure and heart rate responses that are caused only by reflexes originating in the lung, we used a canine preparation in which the left airway and left pulmonary artery were isolated, and in which the thorax was opened widely (Cassidy et al., 1979); thus, cardiac output traversed the right lung, and inflation of the left lung could be accomplished without directly altering pulmonary blood flow or cardiac mechanics.
Methods
Experimental animal preparation. Seven mongrel dogs of either sex weighing 16-25 kg (17.6 + 2.8 SD) were anesthetized with morphine (1-2 mg/kg, i.m.) and ~-chloralose (50-100 mg/kg, i.v.). They were intubated through a tracheostomy with a double lumen endotracheal tube constructed with an inflatable cuff on the distal end that enabled separation of the right and left mainstem bronchi. Initially, both lungs were ventilated with a tidal volume of 15 ml/kg at a rate to maintain Paco2 at 36-40 mm Hg. The heart and lungs were exposed through a midline sternotomy, and the left pulmonary artery was ligated, directing all pulmonary blood flow to the right lung. At this time, all ventilation also was directed to the right lung, and expiratory pressure was elevated to 2 cm H20 to prevent atelectasis. Systemic arterial pressure was measured via a fluid-filled catheter placed in a femoral artery, and heart rate was monitored with a lead II electrocardiogram. Right and left bronchial pressures were monitored at the proximal ends of the double lumen endotracheal tube. The volume of left lung inflation was measured by integrating the left airway flow signal using a screen pneumotachograph and differential pressure transducer. All signals were recorded on a polygraph recorder. Experimentalprotocols. The left lung was inflated to the desired pressure with a gas mixture of 5 ~ CO2 in a balance of air. The lung was inflated for one min at five-rain intervals. Between inflations, it was deflated to 2 cm H20 and was not ventilated. Beginning with an initial inflation pressure of 5 cm H20, inflation pressure was increased serially by 5 cm H20 until an inflation pressure of 60 cm H20 was reached. Subsequently, pressures were decreased serially by 5 cm H20 decrements. Next, the left cervical vagosympathetic trunk was severed, and the serial inflations were repeated. Maximum fall in systemic arterial blood pressures
261
LUNG INFLATION REFLEX
AIRWAY PRESSURE
(cm HL,O)
20
o r
SYSTEMIC ARTERIAL PRESSURE (mmHg)
0
......
:
I
t0 S
I
Fig. 1. A polygraph recording representing airway pressure and systemic arterial blood pressure before and during left lung inflation. Either right or left airway pressure is displayed depending on which direction a stopcock was turned. Right airway pressure was recorded up to point (A) at which time the stopcock was switched to display left airway pressure in order to visualize the inflation of the left lung to 30 cm H20 at point (B). Inflation was continued for 1 min although the recording of airway pressure was switched back to the right airway after approximately 10 see of inflation, point (C), in order to ascertain whether right airway pressure had been altered by the left lung inflation. Lung inflation at (B) was associated with an abrupt fall in blood pressure and heart rate.
was obtained by inspection of the tracing, fig. 1, and heart rate was counted for a period of 10 see coinciding with the maximum fall in systemic arterial pressure. The airway pressure signals were closely examined with each inflation to ascertain that right airway pressure did not increase which indicated that separation of the airways was complete. Femoral arterial blood was obtained anaerobically at the beginning, middle, and end of each experiment for analysis of pH, Pao_~ and Paco:. Supplemental oxygen was administered for Pao2 less than 60 mm Hg. At the end of the experimental protocol, the dogs were killed by exsanguination. The lungs were removed and allowed to passively drain and were dissected free of extrapulmonary bronchi and vessels. Each lung was weighed separately, dried in an oven at 85 °C, and reweighed daily until the weight was stable.
Calculations. Results are reported as means + SD unless otherwise indicated. Student's two-tailed t-test for paired data was used to test the significance of changes that occurred with inflation, and to analyze the significance of differences in responses occurring after left vagotomy.
Results
Inflation pressure/volume relationship of the left lung (fig. 2). On the increasing inflation pressure limb of the inflation pressure-volume curve (fig. 2), a pressure of 15 cm H20 was required before significant volume change occurred. Only
262
s.s. CASSIDY 800
i
60C
z 2oo
%
to
do do
I
70
LEFTLUNGINFLATION PRESSURE (CMH20)
Fig. 2. Pressure-volume relationships for the left lung before (~-------A) and after (A------A) ipsilateral (left) vagosympatheetomy.Arrows indicate the direction of the pressure--volumehistory of the stepwise inflations. Each symbol represents the mean of seven animals. 10 cm H20 was required to elevate volume significantly on the decreasing inflation pressure limb. Maximum inflation volumes o f the left lung were the same before vagotomy, 778 +.66 ml, compared to afterwards, 777 + 111 ml. This occurred at inflation pressures o f 40 cm H20 when previous inflations had been as large or larger. At the steep portion of the inflation pressure-volume curve, we observed that inflation volumes were two to three times greater on the descending limb as compared to the ascending limb. Ipsilateral (left) vagotomy did not alter the descending limb of the inflation pressure/volume curve.
Relationship between lung expansion and reflex cardiovascular responses. During left lung inflation, heart rate and systemic arterial pressure began to fall within 2-4 sec (fig. 1), reaching a maximum response within 8-10 sec, slowly returning toward the preinflation level, but the extent of this return was not quantitated. Heart rate. The maximum heart rate responses to the series of lung inflations before and after left vagotomy are illustrated in fig. 3. The inflation threshold causing a significant reflex bradycardia was 15 cm 1-I20 (63 ml) during the incremental series of lung inflations, and 10 crn 1-120 (102 ml) during the decremental series. Inflations above these threshold levels caused graded responses until maximum lung expansion was attained. Further increasing inflation pressure did not cause a further drop in heart rate response if volume could not be expanded. Minimal hysteresis o f inflation pressures vs heart rate responses was observed, although the limb of incremental pressures tended to give smaller heart rate responses than did inflations on the decremental pressure limb. Plotting inflation volume rather than pressure not only eliminated the hysteresis in the inflation vs heart rate responses, but actually reversed it.
263
LUNG INFLATION REFLEX
10 0 ,-t0 -20 "r
.-.q -30 t~
-50 -60 I
0
I
I
I
I
I
I
I
I
I
10 20 30 40 50 60 0 200 400 600 800 Inflation Pressure (cm H20) Inflation Volume (ml)
Fig. 3. Heart rate responses to unilateral (left) lung inflation before ( ) and after (............. ) ipsilateral vagosympathectomy. Arrows indicate the direction of the volume history of the stepwise left lung inflations. Each point represents the average from seven dogs; vertical bars represent the SE.
Vagotomy eliminated the bradycardia responses entirely and exposed a transient mild tachycardia at the higher inflations occurring at the same time interval. Blood pressure. Systemic arterial pressure response to lung inflation also was proportional to the inflation pressure (fig. 4). The average threshold inflation to cause a significant depressor response was the same as for heart rate. The maximum fall in blood pressure occurred at lung inflations within 100 ml of maximum volume. In spite of the striking degree of pressure-volume hysteresis, blood pressure responses were not significantly different with respect to inflation pressure history. Similar to the heart rate responses, the hysteresis of the inflation volume/blood pressure response was reversed, i.e. incremental inflations gave larger responses than did the inflations of decreasing pressures. Left vagotomy reduced the blood pressure responses to lung inflation by twothirds to three-quarters of the pre-vagotomy responses. These responses remaining post vagotomy were observed only at maximum lung expansion. Arterial blood gas tensions and pH were within the normal range (Pao2 83 + 17 torr; Paco2, 41 + 6 torr; pH, 7.40 + 0.04) and did not change significantly as a consequence of the 30+ inflations or with ipsilateral vagotomy. Wet weight/dry weight measurements of the left lung (5.18 + 0.22) were increased compared to the right (4.27 + 0.10; P < 0.05).
264
S.S. CASSIDY t0 "
o ....
D.
-10
_
o
•-
-20
._u
~-30
(n - 4 0 .E -50 ¢J I
I
I
I
I0
20
30
40
/
50
I
I
I
I
I
60
200
400
600
800
Inflotion Pressure (cmH20)
Inflotion Volume (ml)
Fig. 4. Systemic arterial pressure responses to unilateral (left) lung inflation before ( ) and after (............ .) ipsilateral vagosympathectomy. Refer to legend of fig. 3 for explanation of symbols.
Discussion
When cardiovascular reflex responses to lung expansion are measured, a preparation is necessary in which the mechanical consequences of expansion are eliminated, allowing only indirect (reflex or humoral) responses to occur. The reflex nature of the responses must be established or confirmed by demonstrating the essential afferent or efferent pathways. This preparation differs from total cardiopulmonary bypass used by others to study cardiovascular reflex effects of lung inflation in that cardiac output was intact in this unilateral lung preparation, whereas circulation is provided at a fixed rate during cardiopulmonary bypass. Utilizing this unilaterally isolated left lung, we previously demonstrated that inflation to 30 cm H20 substantiallY reduced heart rate, blood pressure, cardiac output and stroke volume which return toward control, but in general do not reach, control values by 15 min (Cassidy et al., 1979). The present experiments were performed to quantitate the lung expansion at which these reflex responses to lung inflation were initiated and to establish the relationship between inflation and reflex responses throughout the entire range of lung expansion. This method of lung expansion, a l-rain inflation executed by exposing the airway to a square-wave of 5% CO2 in air at the desired pressure, was not meant to mimic any physiological event. Rather, this method of lung expansion was used in order to single out the potential cardiovascular reflex responses in much the same way as barosensitive segments of the circulation have been isolated and stretched with a square wave to study potential reflex responses initiated by elevated pressures in a specific vascular region. Before examining the reflex responses to lung inflation, it was necessary to know the pressure-volume relationship of the lung in this preparation.
LUNG INFLATION REFLEX
265
Pressure-volume curves of the left lung (fig. 2) Once the lung had been expanded repeatedly to pressures greater than or equal to 30 crn H20, additional inflations to 30 cm H20 pressure consistently expanded the left lung to approximately 90% of its total capacity. Ipsilateral vagotomy did not alter the relationship between inflation pressure and volume when the steps of inflation pressure were decreasing; therefore it seemed reasonable to compare the cardiovascular responses to lung inflation that were obtained before vagotomy with those obtained after vagotomy at comparable pressures or volumes using the descending limbs of the inflation pressures. The differences in the pressurevolume relationships during the incremental inflation pressures after vagotomy compared with corresponding inflations before vagotomy were thought to be the consequence of the previous inflation pressure history of the post-vagotomy inflations, but a change in compliance induced by the vagotomy cannot be excluded. Hysteresis in the pressure-volume relationship was an expected finding. It is worthy of mention that with inflations between 15 and 30 cm H20 pressure, lung volume at a given pressure might vary threefold depending on the volume history. Lung volume did not expand greatly at either extreme of inflation pressures, (increasing by approximately 100 ml between 0 and 10 cm H20 inflation pressure and increasing less than 100 ml above 30 cm H20 inflation pressure). Presumably, increased surface tension was responsible for the stiffness of the lung at very low lung volumes, and reaching the limits of elasticity was responsible for the stiffness at large lung volumes. Reflex cardiovascular responses to lung expansion Within the limits of lung expansion, heart rate and blood pressure responses were directly related to inflation pressures since there was little or no hysteresis to the inflation pressure/cardiovascular response relationships (figs. 3 and 4). However, at either extreme of lung expansion when there was little change in volume as inflation pressures were increased, there was correspondingly little change in the heart rate or blood pressure responses. Thus, transmission of pressure changes to responsive elements in the lung parenchyma that could initiate the reflex presumably was limited by surface forces at low lung volumes and by collagen at high lung volumes. In the mid range, these dements that initiate the reflex are responsive to either pressure or volume changes. Heart rate. Previously reported data regarding the effects of increasing lung volume on heart rate have shown diverse responses. Anrep et al. (1936) demonstrated a transient tachycardia at levels of inflation 9 cm 1-120 and less; however, with 18 cm H20 inflating pressure, the transient tachycardia was followed by a sustained bradycardia. Re.cently, Vatner and Rutherford (1981) demonstrated that stimulation of the Carotid bodies with hypoxia often produced tachycardia which was secondary to the hyperpnea that resulted. By controlling ventilation during hypoxic stimulation of the carotid bodies, the tachycardia that they reported was
266
s.S. CASSIDY
eliminated. The origin in the lungs of the reflex tachycardia response to hypoxic stimulation of the carotid bodies was confirmed when division of the pulmonary branches of the vagi eliminated or attenuated the response. These studies are important to illustrate that one mechanism which increases lung volume, hyperpnea, seems to produce reflex tachycardia. Alternatively, a sustained increase in lung volume such as was used by Glick et al. (1969) and in our present and previous experiments (Cassidy et al., 1979) resulted in bradycardia. Since heart rate change was determined in the present experiments by examining the tracing for the lowest rate that occurred during the inflation and counting beats for I0 sec (fig. 1), this precluded our identifying increases in heart rate lasting only 3 to 4 beats which may have preceded the onset of the bradycardia. Hainsworth (1974) offered valuable insight into the issue of the variable heart rate response to lung inflation by examining heart rate response to lung inflation under a variety of circumstances. In closed chest animals with both lungs perfused by the heart, he showed that inflation of one lung, which produced changes in intrathoracic pressure and shifts in thoracic blood volume between the lungs, resulted in tachycardia. Whereas, inflation of one lung in an open chest animal in whom blood flow had been diverted previously to the opposite lung produced bradycardia that was proportional to inflation pressure between 10 and 40 cm 1-120. From all the above cited data, we conclude that the reflex chronotropic response to sustained expansion of the lung is negative and is a direct function of lung inflation pressure within the constraints of induced changes in lung volume. B l o o d pressure. Primary determinants that would result in lowering blood pressure must be derived from previous experiments. A graded fall in systemic vascular resistance during lung inflation in dogs with cardiopulmonary bypass has been demonstrated clearly by Daly et al. (1967) and by Lloyd (1978), both of whom point out that there may be variations in the responses in the vascular beds of different organs. Hainsworth (1974) demonstrated that variability in responses of a specific organ may occur also since the hindlimb vessels both vasoconstricted and dilated with inflation. We previously demonstrated that lung expansion to 30 cm H20 inflation pressure produced a sustained reduction in stroke volume and cardiac output (Cassidy et ai., 1979). Thus, the fall in blood pressure in the present study does not define a specific mechanism, but serves to illustrate the magnitude of reflex systemic hypotension that will result from the vasodilation and fall in cardiac output with lung expansion. As with the heart response to lung inflation, the hypotensive response is sensitive to inflation pressure except at the extremes of expansion. This indicates that blood pressure will be depressed reflexly by sustained lung volume expansion within physiological constraints of lung distensibility. This concept is supported by the studies of Mancia and Donald (1975) who demonstrated that a component of tonic inhibition of the vasomotor center occurred at resting lung volume and had originated in the lungs.
LUNG INFLATION REFLEX
267
Systemic chemoreceptors and baroreceptors were exposed to the systemic circulatory milieu in our studies. Yet, there are no reasons to suspect that transient changes had occurred in blood gases or chemistry great enough to have initiated the responses that were observed. This is so because ventilation to the right lung, which served as the gas exchanger for the entire cardiac output, was not appreciably altered by the step inflation procedure of the left lung. This was documented by the right airway pressure not having changed as a consequence of left lung inflation (fig. 1). Since the arterial baroreceptors were not perfused separately at constant pressures, they undoubtedly buffered the maximum bradycardia and hypotension that developed with inflation.
Neural afferent pathways The negative chronotropic response to lung inflation was entirely eliminated by left cervical vagosympathectomy indicating that the bradycardia was reflex in nature initiated by afferent receptors within the distribution of the ipsilateral vagal nerve. Using a similar preparation, Kaufman et al. (1982) in our laboratories demonstrated that bronchial and pulmonary C-fibers were stimulated in proportion to the extent of lung expansion with thresholds comparable to that initiating the reflex responses, indicating that the reflex responses of lung inflation would be consistent with a C-fiber reflex. However, myelinated fibers are also stimulated in proportion to the degree of lung expansion. Whereas, ipsilateral vagotomy always eliminated the bradycardia response t o lung inflation, a hypotensive response to inflation, especially at large inflations, remained to varying degrees after ipsilateral vagotomy. This was observed also by Daly et ai. (1967) who found the hypotensive response to lung inflation that remained post vagotomy could be eliminated either by crushing the nerlees at the hilum of each lobe or by destroying the stellate ganglia. This suggested that some afferent fibers responsible for the reflex hypotension may course via the thoracic nerves and the stellate ganglia. The small hypotensive responses to very large lung inflations that remained after ipsilateral vagotomy may have been either mechanical or reflex in nature; nevertheless, the majority (two-thirds) of the hypotensive response to lung inflation is a reflex mediated via the ipsilateral vagal nerve.
Significance of the lung inflation depressor reflex It is clear from the literature cited above that mechanical stretching of the lung is capable of generating cardiovascular reflexes that range from tachycardia and vasoconstriction to profound bradycardia, hypotension and vasodilation. While the specific source of reflex responses to increasing rates of inspiration and tidal volume variation is still unresolved, it seems certain that a sustained elevation of lung volume is capable of producing a graded inhibition of the heart rate and blood pressure. The physiological function of this potent cardioinhibitory reflex
268
~s,s. CASSIDY
remains a mystery. Perhaps this reflex contributes to cardiovascular responses to hyperpnea of exercise and to arising from supine posture. From a pathophysiological viewpoint this reflex may produce undesired cardiovascular inhibition in pathological conditions which raise lung volume such as acute asthma with air trapping or during ventilation with positive end-expired pressure in which functional residual capacity of the lung is elevated. The reflex responses to unilateral lung inflation clearly establish that substantial reflex lowering of heart rate and blood pressure can be generated by lung expansion throughout a range of lung volumes that would be expected to occur physiologically.
Acknowledgements The author gratefully acknowledges the technical assistance of Barbara Smith and Taiwan Huang and the help of S.M. Self in preparation of the manuscript. Grant-in-aid support was provided by the American Heart Association, Texas Affiliate and by National Institutes of Health HL 22589. Portions of this work were completed during theauthor's tenure as an Established Investigator of the American Heart Association.
References Anrep, G.V., W. Pascual and R. Rossler (1936). Respiratory variations of the heart rate. I. The reflex mechanisms of the respiratory arrhythmia. Proc. R. Soc. Lond. Set. B l 19:191-217. Cassidy, S.S. and R.L. Johnson, Jr. 0979). Pressure-volume (P-V) characteristics of the reflex cardiovascular (CV) response to lung inflation in dogs. The Physiologist 22: 18. Cassidy, S.S., W.L. Eschenbacher and R.L. Johnson, Jr. (1979). Reflex cardiovascular depression during unilateral lung hyperinflation in the dog. J. Clin. Invest. 64: 620-626. Duly, M.D., J.L. Hazzledine and A. Ungar (1967). The reflex effects of alterations in lung volume on systemic vascular resistance in the dog. J. Physiol. (London) 188: 331-351. Glick, G., A.S. Wechsler and S.E. Epstein (1969). Reflex cardiovascular depression produced by stimulation of pulmonary stretch receptors in the dog. J. Clin. Invest. 48: 467-473. Hainsworth, R. (1974). Circulatory responses from lung inflation in anesthetized dogs. Am. J. Physiol. 226: 247-255. Kaufman, M.P., G.A. Iwamoto, J.H. Ashton and S.S. Cassidy (1982). Responses to inflation of vagal afferents with endings in the dog lung. Circ. Res. 51 : 525-531. Lloyd, T.C., Jr. (1978). Reflex effects of lung inflation and inhalation of halothane, ether and ammonia. J. Appl. Physiol. 45: 212-218. Mancia, G. and D.E. Donald (1975). Demonstration that the atria, ventricles, and lungs each are responsible for a tonic inhibition of the vasomotor center in the dog. Circ. Res. 36: 310-318. Salisbury, P.F., P.M. GaUetti, R.J. Lewin and D.A. Rieben (1959). Stretch reflexes from the dog's lung to the systemic circulation. Circ. Res. 7: 62-67. Vatner, S.F. and J.D. Rutherford (1981). Interaction of carotid chemoreceptor and pulmonary inflation reflexes in circulatory regulation in conscious dogs. Fed. Proc. 40:2188-2193.