| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institute for Cardiovascular Research, The School of Medicine, University of Leeds, Leed LS2 9JT, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
(Received 19 November 2003;
accepted after revision 16 January 2004;
first published online 23 January 2004)
Corresponding author M. J. Drinkhill: Institute for Cardiovascular Research, The School of Medicine, University of Leeds, Leeds LS2 9JT, UK. Email: cvsmjd{at}leeds.ac.uk
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
There are a number of differences between the previous experimental studies and the intact animal: (1) pulmonary pressure has a phasic character, varying with the cardiac cycle; (2) the chest is sealed; and (3) intrathoracic pressure is normally negative and varies throughout the respiratory cycle. We hypothesized that the reason for the abnormally high pressures required to induce reflex responses was due to the unphysiological nature of previous preparations. To test this hypothesis we devised experiments to examine the stimulus–response characteristics of pulmonary arterial afferents in three conditions: chest open, chest closed, and chest closed with phasic negative intrathoracic pressures.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Experiments had local ethical committee approval and were carried out in accordance with the Animals (Scientific Procedures) Act, 1986. At the end of the experimental procedures animals were killed by exsanguination while under deep anaesthesia.
Adult female beagle dogs were anaesthetized with 
-chloralose (100 mg kg-1I.V. in saline, Vickers Laboratories Ltd, Leeds, UK), administered through a cannula inserted into the right saphenous vein under under local anaesthesia (2% lignocaine hydrochloride) so that its tip passed into the inferior vena cava. A stable level of surgical anaesthesia was maintained throughout the duration of the experiments by a continuous infusion of chloralose (0.5–1.0 mg kg-1 min-1). Alfentanyl (30 µg kg-1I.V. Janssen-Cilag Ltd, High Wycombe, UK) was administered over 10 min prior to major surgical procedures. During surgery, alfentanyl was infused continuously at 2.5 µg kg-1 min-1, and this infusion was terminated 60 min prior to the start of the experimental protocol. At intervals throughout the experiment, the appropriate depth of anaesthesia was assessed from the stability of blood pressure and heart rate, the absence of a response to toe pinch or to a sharp tap on the surgical table.
A longitudinal mid-line incision was made in the neck, the trachea cannulated, and the animal was artificially ventilated with oxygen-enriched air (approximately 40%) by means of a Starling ‘Ideal’ pump, initially set at 17 ml kg-1 and 18 strokes min-1. Arterial blood pH, PCO2 and PO2 were monitored frequently using a pH/blood gas analyser (Instrumentation Laboratory, model IL 1610, Lexington MA, USA) and maintained within normal limits by adjustments of the stroke of the respiratory pump, the rate of oxygen inflow, and infusions of molar sodium bicarbonate solution.
The sternum was split along the midline and each side of the chest was divided between the 4th and 5th ribs. When the pleural cavity was opened a positive end-expiratory pressure was maintained at 3 cmH2O to prevent lung collapse. The pericardium was opened to expose the right atrium and ventricle for subsequent cannulation. The descending aorta was mobilized by tying and dividing the 2nd to 7th pairs of intercostal arteries. The inferior vena cava was mobilized immediately above the diaphragm and a loose thread placed round it. Both pulmonary arteries were dissected free to the points of their first branches and loose threads placed around them. Extreme care was taken to avoid damage to the nerves running over the lung roots.
The animal was then given heparin (500 i.u. kg-1I.V.; Leo Laboratories Ltd, Princes Risborough, UK) and the perfusion circuit was connected to the animal (Fig. 1). The circuit volume was 1 litre and this consisted of a heparinized mixture of equal parts of mammalian Ringer solution (g l-1: NaCl, 6.9; KCl, 0.35; CaCl2, 0.28; MgSO4, 0.14; NaHCO3, 2.09; KH2PO4, 0.16; glucose, 1.0) and dextran in dextrose solution (50 g l-1: dextran molecular weight 181 000).
|
The pulmonary pouch was then established. A modified balloon cannula (7 mm i.d.; Baxter International Inc., IL, USA) was passed into the left pulmonary artery via an incision in the right ventricle. The balloon was inflated with approximately 5 ml of saline and then withdrawn so that it occluded the pulmonary valve. The cannula was secured by means of a purse string suture, and the position of the balloon was always confirmed post mortem. The cannula was perfused with venous blood from a pressurized reservoir that received blood from the open reservoir via a heat exchanger. Another cannula (1.5 mm i.d) was inserted into the central end of a right lobar artery and this provided the outflow from the pulmonary arterial pouch, which drained back to the open reservoir. All of the remaining left and right lobar branches were firmly tied to prevent perfusion of the intrapulmonary blood vessels. A pulsatile pouch pressure was generated using an electronic timer switch and two solenoid valves (Bürkert timer unit 1078-2, solenoid unit 311, Bürkert Centromatic, Stroud, UK) to switch between high and low pressure sources.
A length of the left or right cervical vagus nerve was dissected free of the surrounding tissue including the carotid artery. A portion of the nerve was laid on a black Bakelite platform and fine strands were dissected from the nerve under a pool of warm paraffin oil (37°C) using watchmakers forceps and a binocular microscope (Carl Zeiss, Jena, Germany). Afferent activity from single or small multifibre preparations was recorded using bipolar silver electrodes. The output was amplified and filtered (Neurolog system, Digitimer Ltd, Welwyn Garden City, UK) and the action potentials were displayed on a digital storage oscilloscope (Model OS 1420, Gould Ltd, Hainault, UK). The activity was also monitored by an audioamplifier (Model D130 Digitimer Ltd).
In some experiments, all tubing connected to cannulae in the chest was externalized through separate incisions in the chest wall. Two large tubes were placed in the thoracic cavity and connected to an electronic timer switch and solenoid valve system that switched between atmospheric pressure and a negative pressure source to simulate respiratory changes. Intrathoracic pressure was measured by an air-filled catheter positioned inside the chest cavity. The chest was closed by strapping the ribs and sternum together using nylon straps, and tightly suturing the overlying muscle and skin.
Blood pressures were recorded using saline-filled nylon catheters attached to strain gauge transducers (Gould-Statham P23Gb, Oxnard, CA, USA), connected to the pulmonary arterial pouch and a femoral artery (systemic arterial perfusion pressure). Before each experiment the pressure transducers were calibrated against a mercury column. Signals of pressures and nerve activity were amplified (EMMA system, SE Laboratories, Feltham, UK) and recorded on a computerized data acquisition system (Fastdaq, Lectromed, Letchworth, UK), VHS tape (Racal V-store, Racal Recorders Ltd, Southampton, UK) and a direct-writing electrostatic recorder (Model ES 1000, Gould Electronics, France).
Throughout all of the experiments the temperature of the animals was monitored using an oesophageal thermistor probe. The temperature was maintained at 37–39°C by heat exchangers in the circuit and by heaters under the animal table.
Experimental protocol
Initial identification of vagal afferents consisted of a systematic search for a unit displaying pulsatile discharge and which responded to distension of the pulmonary pouch with an increase in activity. The increase in distending pressure was achieved by applying a pulsatile pressure (amplitude of approximately 10 mmHg and a frequency of 80 pulses min-1) to the pulmonary reservoir. Once a suitable unit had been identified, the response of activity to stepwise increases in distension pressure was studied. Each test commenced at the lowest attainable perfusion pressure, which acted as a control for the test. Stepwise increases in mean pulmonary pressure of approximately 5 mmHg were applied. At each step, the distending pressure was maintained until steady-state responses were seen (about 60 s). The steady-state data were collected for subsequent analysis. In order to preserve the viability of the preparation, individual pressure tests were determined as complete when afferent discharge failed to increase by around 10–15% of that at the previous pressure step, or once the distension pressure was around 50 mmHg.
In closed-chest preparations, tests were undertaken in the presence of either atmospheric pressure in the chest cavity or phasic negative intrathoracic pressure changes similar to those encountered during normal breathing (approximate amplitude of 10 mmHg and frequency of 18 cycles min-1). These procedures were randomized and bracketed.
At the end of each experiment, we attempted to determine the receptive field of the investigated vagal afferent fibre by careful exploration of the pulmonary pouch with a fine probe.
Analysis of results
Data were digitized at 1 kHz, except the neurogram (12 kHz), with signal processing software (Fastdaq, Lectromed) and analysed off-line. The pressure–baroreceptor activity relationship for individual tests was plotted using curve-fitting computer software (GraphPad version 4.0, GraphPad Software Inc., San Diego, CA, USA) to fit a sigmoid regression curve to the data points. For each increment, the mean afferent discharge was plotted against mean pulmonary artery distension pressure. Various measurements were made from each stimulus–response curve. Threshold pressures were taken as the pulmonary arterial pressure corresponding to 5% of the overall change in discharge. The maximum slope of the curve was determined from the first differential of the sigmoid function. The inflexion point was the pulmonary pressure corresponding to the peak differential of the sigmoid function.
Data are presented as means ±standard errors of the means unless otherwise stated. All values were tested for normality and parametric tests were used. Differences in variables within one condition were analysed by repeated measures ANOVA with a post hoc Dunnett's test. Comparisons were made of the threshold pressure, maximum slope and inflexion pressure for each condition. Between-condition comparisons were performed by Student's unpaired/paired t tests and by two-way repeated measures ANOVA with post hoc Bonferroni tests where appropriate. Statistical significance was accepted at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Vagal afferent responses in open-chest animals
Seven vagal afferent fibres in seven open-chest preparations were studied. All exhibited a regular spontaneous discharge synchronized to the pulsatile pressure distending the pulmonary arterial pouch, and all responded with an increase in discharge to a large step increase in pulmonary artery pressure. Examples of vagal afferent activity recordings in one animal are shown in Fig. 2.
|
|
|
Studies were undertaken on an additional six vagal afferents in five preparations that had the chest closed and resealed following connection to the perfusion circuit. An example of vagal afferent activity and recorded in one animal under different conditions of pulmonary arterial pressure and intrachoracic pressure is shown in Fig. 5.
|
|
Individual and group average values for the threshold pressures, inflexion points and maximal slopes of open- and closed-chest animals are presented in Tables 1 and 2. The values for the 5% threshold pressure were lower in closed-chest animals. Closure alone resulted in a reduction of approximately one-third and the threshold pressure was significantly lower in the presence of phasic negative intrathoracic pressure when compared with open-chest animals (P < 0.05, Student's unpaired t test). The inflexion pressures of the stimulus–response curves were also lower in the closed-chest animals. With phasic negative intrathoracic pressures the value for the inflexion pressure was approximately half of that for the experiments in the open-chest animals (P < 0.003, Student's unpaired t test). No significant effect was observed on the maximum slope of the stimulus–response curves, although the values tended to be lower in closed-chest animals. Comparisons of the group average values for the inflexion pressure and threshold pressures in each of the experimental conditions are presented in Fig. 7.
|
|
|
Of the receptors whose locations were determined, all were found in the main bifurcation, or in the right and left branches of the pulmonary artery in the vicinity of the bifurcation and the origins of the lobe branches.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The existence of sensory nerve endings in the pulmonary trunk and proximal parts of the left and right pulmonary arteries has been known for a long time (Coleridge et al. 1961; Verity et al. 1965). These endings were described at the media-adventitial junction and were similar to those seen in the carotid sinuses and aortic arch. Previous electrophysiological studies have clearly demonstrated pulsatile discharge in myelinated vagal afferents synchronized to normal pulmonary arterial pressure in cats (Bianconi & Green, 1959) and dogs (Coleridge & Kidd, 1960, 1961). In these studies bursts of activity normally occurred during the systolic rise of pulmonary artery pressure.
Although there have been previous electrophysiological studies of vagal afferent activity originating from pulmonary artery baroreceptors, the experiments described here are the first to examine responses over a range of precisely controlled pressures. It is difficult to localize a physiological stimulus to the pulmonary artery and its right and left main branches and early workers (Bianconi & Green, 1959; Coleridge & Kidd, 1960) brought about changes in pulmonary arterial pressures using a variety of methods; e.g. infusion of a volume expander, bleeding the animal, occlusion and release the inferior vena cava, or temporary obstruction of the pulmonary artery. None of these methods apply a stimulus that is possible to relate to physiological events.
A better method for studying responses from pulmonary arterial baroreceptors is to isolate the region vascularly and to determine the responses to its distension with blood at different pressures. In the present study, we used a pouch of the vascularly isolated pulmonary artery, formed by ligating the branches of the left and right arteries at each of the lung lobes. The functional anatomy of the main pulmonary arteries was preserved relatively intact and the pouch was perfused via a cannula in the left pulmonary artery and drained via a cannula in a lobar branch of the right pulmonary artery. Ligating the lobes not only prevented perfusion of intrapulmonary vessels but also crushed the sensory innervation of the lungs.
By recording the vagal afferent activity at each pulmonary arterial pouch pressure we were able to construct stimulus–response relationships over a wide pressure range. We defined the curve using an equation that assumes the curve is sigmoidal and that the lowest pressure tested is near threshold and that highest pressure is near saturation. We analysed the curves to determine the pulmonary pouch pressures corresponding to 5% of the overall response, which we took to be threshold. The maximal slope and the inflexion point of the relationship were determined from the first differential of the stimulus–response curve.
A major aspect of this investigation was to determine whether discharge from pulmonary baroreceptors could be evoked at physiological pulmonary artery pressures and then to define the normal stimulus–response relationship. Our results are qualitatively in agreement with earlier findings by Coleridge & Kidd (1961, 1963) that incremental distension of the vascularly isolated pulmonary artery is signalled by an increase in vagal afferent discharge at each step. In the present study, however, experiments on vagal afferents in open-chest animals resulted in a significant increase in baroreceptor discharge only when the mean distending pressure was 25 mmHg or higher. We also determined values for the threshold pressure and inflexion pressure of the pulmonary baroreceptor stimulus–response relationship that were higher than the normal pulmonary arterial pressure reported in conscious and anaesthetized dogs (Hamilton et al. 1939; Simmons et al. 1958).
There are, however, some considerations which may allow for significant stimulation of pulmonary baroreceptors at lower pulmonary pressures. In animals breathing normally, the chest is closed and the pressure in the chest cavity is negative and subject to phasic variations during the ventilatory cycle. During quiet breathing, intrathoracic pressure changes from about -8 mmHg during inspiration to -3 mmHg during expiration (Janicki et al. 1996). It is possible therefore that closure of the chest and application of negative intrathoracic pressures would affect the pressure gradients within the pulmonary arteries, and we hypothesized that this may increase the responsiveness of pulmonary baroreceptors. We tested this in a second experimental group that had the chests resealed following connection to the perfusion circuit. The results showed that post-operative closure of the chest wall alone, even without negative intrathoracic pressure, leads to lower values for the threshold pressure and the inflexion point of the stimulus response curve. Furthermore, when we applied a phasic change to the intrathoracic pressure of around -10 mmHg we observed a further reduction in the values for the threshold pulmonary arterial pouch pressure and in the inflexion point of the stimulus–response curve. Under conditions of phasic negative intrathoracic pressure changes, the stimulus–response curve was displaced to the left. This resulted in group average values for the threshold pressure and inflexion point that are approximately half those obtained in the open-chest group (Fig. 7). These lower values for threshold pressure and the inflexion point of the stimulus–response curve are consistent with the normal pulmonary arterial pressure reported in conscious and anaesthetized dogs.
We are unsure why we found a lower threshold pressure and a lower inflexion pressure in closed-chest animals even before a negative intrathoracic pressure changes were applied. One possibility may be that temperature differences between the open- and closed-chest state had an effect on baroreceptor activity. In an earlier study, Ledsome et al. (1981) observed that cooling of the pulmonary arterial pouch, by changing the temperature of the perfusate from 37 to 30°C, caused a decrease in the responses of systemic vascular resistance and respiratory frequency during pulmonary arterial distension between 20 and 80 mmHg. This led to the suggestion that discharge from pulmonary arterial baroreceptors, like that for a number of stretch receptors including aortic baroreceptors, is temperature sensitive (Ledsome et al. 1981). The effect of a temperature change in the pulmonary pouch on the stimulus–response characteristics of pulmonary arterial afferents was not, however, investigated in the current experiments.
The effect of phasic intrathoracic pressure changes on the stimulus–response characteristics pulmonary arterial afferents in closed-chest animals may be related to changes in the transmural pressure of the pulmonary arteries. Although not measured directly in these experiments, the transmural pressure will be equal to the pulmonary arterial pressure minus intrathoracic pressure. Since both pulmonary arterial and intrathoracic pressures varied by about 10 mmHg, the effective instantaneous pressure therefore would be the prevailing mean pulmonary arterial pressure plus up to 15 mmHg, depending on the phases of the pulmonary arterial and intrathoracic pressure cycles. This may result in altered baroreceptor stimulation, even in the absence of large change in the intraluminal pulmonary arterial pressure. This view is supported by the observation that vagal afferent activity was modified in the presence of a phasic intrathoracic pressure change so that the impulse frequency is greatest during the ‘inspiratory’ phase of the intrathoracic pressure cycle (Fig. 5).
To conclude, the importance of this study is that it is the first to compare the stimulus–response characteristics of pulmonary vagal afferents over a range of pressures in open- and closed-chest preparations. The major new finding is that addition of phasic negative intrathoracic pressure to a resealed chest lowers the threshold pressure and inflexion pressure of the pulmonary pressure–baroreceptor activity relationship. This new evidence implies that under conditions of normal breathing, pulmonary baroreceptors are responsive at the pressures normally prevailing in the pulmonary circulation. We hypothesize that this is probably due to the effect of the intrathoracic pressure cycle on transmural pulmonary arterial pressure.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Coleridge JCG & Kidd C (1960). Electrophysiological evidence of baroreceptors in the pulmonary artery of the dog. J Physiol 158, 197–205.
Coleridge JCG & Kidd C (1961). Relationship between pulmonary arterial pressure and impulse activity in pulmonary arterial baroreceptor fibres. J Physiol 158, 197–205.
Coleridge JCG & Kidd C (1963). Reflex effects of stimulating baroreceptors in the pulmonary artery. J Physiol 166, 197–210.
Coleridge JCG, Kidd C & Sharp JA (1961). The distribution, connexions and histology of baroreceptors in the pulmonary arteray, with some observations on the sensory innervation of the ductus arteriosus. J Physiol 156, 591–602.
Hamilton WF, Woodbury RA & Vogt E (1939). Differential pressures in the lesser circulation of the anaesthetized dog. Am J Physiol 125, 130–141.
Janicki JS, Sheriff DD, Robotham JL & Wise RA (1996). Cardiac output during exercise: contributions of the cardiac, circulatory and respiratory systems. In Handbook of Physiologyi, section 12, Exercise: Regulation and Integration of Multiple Systems, ed. Rowell LB &. Shepherd JT, pp. 649–704. American Physiological Society, Bethseda, MD, USA.
Ledsome JR & Kan WO (1977). Reflex changes in hindlimb and renal vascular resistance in response to distention of the isolated pulmonary arteries of the dog. Circ Res 40, 64–72.
Ledsome JR, Kan WO & Bolter CP (1981). Respiratory and cardiovascular responses to temperature changes in the perfused pulmonary arteries of the dog. Can J Physiol Pharm 59, 493–499.[Medline]
McMahon NC, Drinkhill MJ, Myers DS & Hainsworth R (2000). Reflex responses from the main pulmonary artery and bifurcation in anaesthetised dogs. Exp Physiol 85, 411–420.[Abstract]
Simmons DH, Hemmingway A & Ricchiuti N (1958). Acute circulatory effects of pneumothorax in dogs. J App Physiol 12, 255–261.
Verity MA, Hughes T & Bevan JA (1965). Innervation of the pulmonary artery bifurcation of the cat. Am J Anat 116, 75–90.[CrossRef][Medline]
|
Acknowledgements |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and phasic negative intrathoracic pressure changes (
) in the closed chest. Mean discharge was significantly greater versus baseline at distending pressures of 20 mmHg and above for atmospheric pressure, and 15 mmHg and above for phasic negative intrathoracic pressures. *P < 0.05 (two-way repeated measures ANOVA).