| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 1998;14:494-502
© 1998 Elsevier Science NL
a Department of Cardiothoracic and Vascular Surgery, Hannover Medical School, Hannover, Germany
b Department of Anaesthesiology, Hannover Medical School, Hannover, Germany
Received 23 March 1998; accepted 22 July 1998.
Corresponding author. Thorax-, Herz- und Gefäßchirurgie, OE6210, Medizinische Hochschule Hannover, 30623 Hannover, Germany. Tel.: +49 511 5326580; fax: +49 511 5325404; e-mail: krieg@thg.mh-hannover.de
 |
Abstract |
|---|
 Top Abstract Introduction
Material and methods
Results
Discussion
Appendix A. Conference...References |
|---|
Key Words: Left ventricular function • Nitric oxide • Prostaglandin E1 • Inhalation of Drugs • Pulmonary hypertension
|
Introduction |
|---|
|
TopAbstractIntroduction
Material and methods
Results
Discussion
Appendix A. Conference...References |
|---|
Currently available therapeutic measures include i.v. drug therapy, e.g. diuretics, digitalis, afterload-reducing therapy and prostaglandins, aside from mechanical support [3]. Recent work has shown that inhaled nitric oxide (NO) [4] [5], inhaled prostaglandin E1 (PGE1) [4] and their combination [4] reduce pulmonary hypertension but their influence on cardiac contractility is less well defined. The intravenous infusion of PGE1 has been shown to improve the hemodynamic state in patients with end-stage chronic heart failure [6] but the cardiac effects of PGE1 application by inhalation have not been described previously. On the other hand, while NO has shown to reduce PHT in patients with congestive heart failure and in patients awaiting heart transplantation [7] [8] the reduction in pulmonary vascular resistance (PVR) is usually accompanied by a rise in pulmonary capillary wedge pressure and reduction of cardiac index [7] [8] which has even led to pulmonary edema [9]. Despite evidence of a negative inotropic effect of NO on humans [8] [10] and animal preparations [11] its inotropic effect is a topic of ongoing debate [12] [13] [14].
This study, therefore, investigated left ventricular contractility as measured by the load-independent `Preload Recruitable Stroke Work-relation' (PRSW) in PHT treated with NO, PGE1 and the combination of both.
|
Material and methods |
|---|
|
TopAbstractIntroduction
Material and methods
Results
Discussion
Appendix A. Conference...References |
|---|
Experimental preparation
Under continuously administered general anesthesia with intravenous thiopental-sodium (5 mg/kg per h) and fentanyl (10 µg/kg per h) in combination with intermittent pancuronium relaxation (0.05 mg/kg) and while on continuous mechanical ventilation (Siemens-Elema AB, Solna, Sweden), the hearts of 24 healthy 7–8 months old pigs 39.1±0.9 kg (mean±SEM) were exposed through a median sternotomy. Mean arterial pressure, monitored with a micromanometer-tipped catheter (model MPC-500, Millar Instruments, Houston, TX) in the aortic root, was maintained at 55–75 mmHg, and blood temperature was kept at 38°C. Micromanometer-tipped catheters were placed in the pulmonary trunc, left atrium, left ventricle and right ventricle (
Fig. 1 ).
|
The superior and inferior vena cava were prepared with circumferential occluders to enable variation of left and right ventricular filling pressures and volumes. The pericardium was left open. For induction of PHT, glass beads (0.5–1.0 g/kg) were applied into the pulmonary trunc.
After termination of each study the animal was sacrificed using 20 ml intracardial T61 (Hoechst, Unterschleissheim/Munich, Germany), a combination of embutramide, mebenzonium-jodide and tetracaine-hydrochloride, 3 min after an intravenous bolus of fentanyl and thiopental had been given. Then the heart was excised and proper position of the transducers was confirmed. Left ventricular wall volume (Vwall) was measured by saline displacement after excising the atria, right ventricular free wall, aortic and mitral valves, and chordae tendineae.
Experimental design and realization
The 24 pigs were randomly divided into two groups of 12 pigs each. One group received verum substances, the other control substances. After induction of PHT the application of verum substances: inhaled nitric oxide (NO) 50 ppm; inhaled nebulized prostaglandin E1 (PGE1) 20 µg/ml; and their combination and control substances: air; ethanole-saline (4 ml waterfree ethanole/100 ml saline, the solutant in the PGE1 solution); and their combination was performed according to the randomized study protocol. NO (1260 ppm) diluted in N2 (Linde AG, Höllriegelskreuth, Germany) was blended to the inspiratory limb of the respirator via its low pressure inlet. The proper inspiratory and expiratory NO concentration (50 ppm) were controlled by electrochemical NO-meters, PAC II NO (Drägerwerk AG, Lübeck, Germany). PGE1 (Minprog®, Upjohn GmbH, Heppenheim, Germany) was diluted in normal saline (final concentration: 20 µg/ml) and added as an aerosol to the inspired gas by a Servo Nebulizer 945® (Siemens-Elema AB, Solna, Sweden) to achieve an effective PGE1 application of 2.66 µg/min. Except for the application of NO and PGE1, which in the control group were replaced by air-application and by the solutant of PGE1, respectively, both groups received the same treatment with the same randomized application schedule. Application of verum and control-substances were performed for 15 min each before data-recording. The order of application was randomized according to a latin square.
Data acquisition
The ultrasonic dimension transducers, being coupled to a sonomicrometer (Triton Technology, San Diego, CA) and the micromanometer-tipped catheters, which had been balanced and calibrated simultaneously to a mercury manometer immediately before each study, were connected to a multi-channel-12-bit A/D-converter (Scientific Solutions Inc., Solon, OH). Digitized data were recorded realtime on a 80386 40 MHz PC with 200 Hz sample rate to be analyzed later, offline.
For each intervention, physiologic data were digitized during vena caval occlusion for 15 s.
Data analysis
Left ventricular chamber volume (V) was calculated by fitting the epicardial base-to-apex (a), anteroposterior (b), and septal-free wall (c) left ventricular dimensions (
Fig. 1) to a cylindric ellipsoidal volume model, representing the `epicardial' ventricular surface, and subtracting ventricular wall volume (Vwall)
[16] (Eq. (1)).
 | (1) |
Left ventricular transmural pressures (P) were measured simultaneously using the above mentioned intracavitary micromanometer, introduced through the left ventricular free wall.
Left ventricular stroke work (SW) was calculated as the integral of left ventricular transmural pressure (P) with respect to the calculated chamber volume (V) over each cardiac cycle (Eq. (2)):
 | (2) |
The relation between left ventricular stroke work and end-diastolic volume (EDV) during vena caval occlusion was determined by linear regression analysis, as described previously
[15] and data were fitted to the formula (Eq. (3)):
 | (3) |
Comparison of cardiac contractility as measured by the PRSW-relation and expressed by changes in slope (with constant x-intercept) were made between verum and control groups for each verum-/control-substance. To control for inter-individual differences, e.g. in transducer placement, first changes of parameters within each of the 24 hearts between initially induced PHT and after application of the verum-/control-substances were computed. These differences were analyzed by MANOVA (slope and x-intercept being the dependant variables with the null hypothesis that the slope and x-intercept means are the same for the two verum- and control-group categories) using SPSS software (SPSS, Chicago, IL) and corrected for heart-rate by including heart-rate as the covariate. The level at which statistical significance was accepted was P<0.05. Unless otherwise stated, data are expressed as mean±SEM.
|
Results |
|---|
|
TopAbstractIntroduction
Material and methods
Results
Discussion
Appendix A. Conference...References |
|---|
Concerning left ventricular contractility and comparing the verum- and the control-group (Table 1) baseline PRSW-values showed no statistically significant differences (P=0.51). PRSW-slope (measuring contractility, PRSW-x-intercept did not change significantly) showed a significant reduction in the verum group under sole application of NO 50 ppm by 14.6±4.4% ( Fig. 2 Fig. 3 ). There were no significant differences in PRSW between verum and control group under sole application of PGE1 20 µg/ml nor under the combination of NO and PGE1. (PRSW-change±SEM in percent of initial PRSW after induction of PHT was -14.6%±4.4% versus 1.6%±4.4% for NO versus Control (P=0.004), -8.8%±4.6% versus 1%±3.3% (not significant, P=0.18) for PGE1 versus Control and -5.7%±4.4% versus 2.5%±4.2% for NO+PGE1 versus Control (not significant, P=0.33, respectively).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroduction
Material and methods
Results
Discussion
Appendix A. Conference...References |
|---|
Concerning the reduction of pulmonary pressure in hypoxic PHT, NO seems to be more potent than PGE1, but the combination of both did not reduce the potential of NO to lower PHT [4]. The effects in this study with embolic PHT were similar to those with hypoxic PHT. Again NO as well as NO plus PGE1 had the same potential to reduce the experimental embolic PHT (by 25%). Sole application of PGE1 had a slightly less effect in reducing embolic PHT, but still reduced it significantly by 16%.
Concerning left ventricular contractility sole application of NO 50 ppm showed a significant reduction, while the combined application of NO and PGE1 as well as sole application of PGE1 did not.
In clinical studies a negative inotropic effect of bicoronary infusion of substance P (which releases NO from the endothelium) [17] and an improvement of the positive inotropic response to the ß-adrenergic stimulation with dobutamin of NG-monomethyl-L-arginine (a NO synthase inhibitor) [10] are discussed. Further evidence of a negative inotropic effect of NO is demonstrated by a marked increase in ventricular preload with little benefit to pulmonary pressures in patients with impaired cardiac reserve receiving inhaled NO [8] as well as in animal preparations [11].
Other studies could not find a significant reduction of LV contractility following inhalation of NO [13] [14] [18]. Inhalation of 20 ppm NO delivered for 10 min to humans with normal LV function showed no significant inotropic effect [13].
A second study evaluated left and right ventricular contractility in dogs with monocrotaline induced pulmonary hypertension after experimental orthotopic cardiac transplantation. Starting sequential administration of NO (40 ppm for 10 min and then 80 ppm for 10 min) 60 min after weaning from cardiopulmonary bypass did not show significant changes [14].
In a third study applying 20, 40 and 80 ppm of inhaled NO in a porcine model of PHT (induced by a intravenous thromboxane A2 analog) no statistically significant changes on LV contractile function were detected [18].
While all three studies also used load-independent parameters of contractility, it is important to note, that in the first [13] hearts with normal LV function were investigated. In the second study [14] increasing doses of NO were sequentially applied without randomization to a heart that was only 60 min after weaning from cardiopulmonary bypass after cardioplegic arrest. A negative inotropic effect of the increasing doses of inhaled NO might therefore be masked by a positive inotropic effect of recovery from cardioplegic arrest. To overcome these problems in our study the substances were applied in randomized order and further the comparison was made between a verum- and a control-group.
In the third study [18] an intravenous thromboxane A2 analog was applied for induction of PHT and the conductance catheter technique was used to assess left ventricular function. On one hand methodological differences in measuring left ventricular contractility might have contributed to the different results. While the conductance catheter technique yielded SEMs of the left ventricular volume of around 9 ml, our SEMs of left ventricular volume were around 2 ml (with about equal sample sizes: 10 pigs vs. 12 pigs). Therefore, a component of measuring left ventricular volume less precise with the conductance catheter technique can not be excluded. On the other hand the combined application of NO and thromboxane A2 may well have led to a zero net effect on the myocardial contractility, similar to the antagonistic effects on the pulmonary circulation [18]. In our study no antagonists of NO were used. Rather, PHT was induced mechanically in an equal fashion in the verum- and control-group.
An effect of ventricular interaction in our study secondary to a different reduction of PHT (25% by NO and the combination of NO and PGE1 vs. 16% by sole PGE1) is unlikely due to the fact that NO and the combination reduced PHT to equal pulmonary pressures and only the case of sole NO application led to a significant reduction of contractility. Furthermore, performing the same MANOVA for the changes of the septal-free-wall dimension (c in Fig. 1) in steady state conditions (i.e. without vena caval occlusions) shows no significant difference between verum- and control-group for changes of the end-diastolic length under each of the three conditions (NO, PGE1 and combination). If different outflow impedances of the right ventricle and concomitant higher right ventricular end-diastolic volumes had caused an end-diastolic septal shift towards the left ventricle, differences in the end-diastolic septal-free-wall dimension should have occurred. Therefore, a diastolic ventricular interaction can be excluded. Analyzing steady state conditions for end-systolic changes of the septal-free-wall dimension, a significant difference between the verum- and the control-group can be demonstrated under sole NO and the combination (more increase in verum than in control; an increase in the end-systolic septal-free-wall distance demonstrates less contraction in this dimension). A contribution of systolic ventricular interaction under steady state conditions can not be excluded in these cases and could be explained by the better reduction of PHT compared to sole application of PGE1 (no significant difference to control with less reduction of PHT by PGE1; note that in control also no reduction of PHT occurred). Nevertheless, this effect under steady state conditions has to be viewed separately and is likely to be excluded by the applied methodological approach of measuring contractility. One of the advantages of the vena caval occlusion technique as a method for assessing left ventricular performance is that this technique produces a rapid reduction in right ventricular volume and pressure, which precedes the reduction in left ventricular volume and pressure, thereby reducing the interactive contribution of the right ventricle to left ventricular function [16]. If the demonstrated negative inotropic effect of NO was due to ventricular interaction, the combination of NO and PGE1 with the same reduction of PHT would have shown a significant negative inotropic effect, as well.
NO has shown to reduce PHT in patients with congestive heart failure and in candidates awaiting heart transplantation [7] [8]. The reduction in pulmonary vascular resistance (PVR) is usually accompanied by a rise in pulmonary capillary wedge pressure and reduction of cardiac index [7] [8]. In some instances it has even lead to pulmonary edema in stable severe heart failure [9].
As opposed to the normal LV a variety of molecular alterations with subsequent functional consequences occur in heart failure and cardiac hypertrophy [19] [20] [21]. Of these the following may be interpreted to be adaptive in character [19]:
The molecular changes of the contractile proteins and their regulatory proteins with fundamental alterations of the cross-bridge mechanics.
The downregulation of beta-adrenoceptors with subsequent functional reduction of the contractile reserve.
On the other hand, the reduction of the sarcoplasmic reticulum Ca2+-ATPase with a related reversal of the force-frequency relationship may be causally related to the heart failure syndrome [19].
Recent studies show a cGMP-induced reduction in myofilament response to Ca2+ by NO [22] [23] which may add to the above mentioned reduction of the sarcoplasmic reticulum Ca2+-ATPase in the failing heart and thereby potentiate the negative inotropic effect of NO. This may explain why the negative inotropic effects are mainly observed in the failing heart and not in the heart with normal function.
On the other hand myocardial relaxation is enhanced by the cGMP-induced reduction in myofilament response to Ca2+ by NO [23]. This support in myocardial relaxation seems to be helpful in conditions with primarily diastolic dysfunction such as (brief) hypoxia/re-oxygenation states [23]. This effect may well have been operative in the above mentioned study of experimental orthotopic cardiac transplantation with monocrotaline induced pulmonary hypertension not showing significant changes in left and right ventricular contractility [14].
Intravenous application of PGE1 has been used experimentally and clinically for the treatment of PHT [24]. Despite the fact that PGE1 is significantly inactivated by a single passage through the lungs, it did not show pulmonary selectivity when administered intravenously [24]. Inhalation of PGE1 aerosol on the other hand reduced PHT in a dose-related manner [4]. Only a PGE1-solutant (for aerosolization) of 20 µg/ml revealed pulmonary selectivity, whereas 5 µg/ml showed no effect on PHT and 80 µg/ml an additional decrease in systemic pressure [4]. For this study, therefore, a PGE1-solutant for aerosolization of 20 µg/ml was used for the assessment of the inotropic effects of inhaled aerosolized PGE1.
Intravenous application of PGE1 has been demonstrated to exert no or a small species-dependent positive inotropic effect on cardiac contractility [6]. To our knowledge, no data concerning LV function were available previously for inhaled application of PGE1. In this study, sole application of inhaled PGE1 did not result in a statistically significant change of LV-contractility in the pig. Furthermore, an increase in cardiac output (however, at least in part induced by a reduction of outflow impedance) was found in humans when administering PGE1 intravenously [6]. Therefore, inhaled application of PGE1 in humans is not expected to have negative inotropic side effects when used therapeutically.
Interestingly, a reversion of the negative inotropic effect of NO could be demonstrated by combining NO with PGE1, but the molecular mechanism of interaction remains to be revealed.
A number of different concentrations of NO (generally up to 80 ppm) had been administered clinically in the past with a tendency to lower concentrations (10 ppm or less), recently [5] [7] [8]. For our study a concentration of 50 ppm was chosen, because an inotropic effect was expected to be detected better in the upper range of clinically used concentrations.
The 14.6% reduction of LV contractility by inhaled NO may be explained despite the short half-life of the molecule NO (between 0.1 and 5 s). The active agent may still reach the endocardial and endothelial coronary sites in the form of S-nitrosohemoglobin, which acts as a souped-up version of the NO molecule, enhancing its physiologic effects [25]. In normal myocardial muscle endogenous endothelium derived NO can attenuate mitochondrial respiration. In the failing heart there seems to be a defect in the ability of coronary blood vessels to produce NO and to attenuate mitochondrial respiration. Here a loss of the regulatory role of NO in optimizing the O2 utilization and a 54% higher basal O2 consumption rate than that of normal muscle were found [12]. Furthermore, recent studies suggest that the same conformational change that, upon reaching oxygen-poor tissues, reduces the affinity of hemoglobin for oxygen also releases S-nitrosothiol (SNO) [25]. Therefore the increased mitochondrial respiration in the failing heart may trigger an increased release of SNO which may then exert its negative inotropic effects via the above mentioned mechanisms (i.e. cGMP-induced reduction in myofilament response to Ca2+ by nitric oxide [22] [23] which may add to the reduction of the sarcoplasmic reticulum Ca2+-ATPase in the failing heart). Again, this may explain why the negative inotropic effects are mainly observed in the failing heart and less in the heart with normal function.
Lacking more detailed information concerning the interaction of NO and PGE1, at present only a descriptive recommendation can be given from the present study:
If NO is not available, the sole application of nebulized PGE1 20 µg/ml appears to be safe with respect to left ventricular contractility in the setting of PHT. Combined application of NO 50 ppm and PGE1 20 µg/ml is favorable opposed to sole application of NO 50 ppm with respect to LV contractility for the treatment of PHT. The combination of NO and PGE1 for the treatment of pulmonary hypertension should be considered for clinical application in situations where a combination of pulmonary hypertension and decreased left ventricular function is present.
Inhaled PGE1-application should be used for short-term treatment only until a risk of toxicity of the ethanole-saline preparation of PGE1 can be excluded for higher-dose and/or long-term inhalation.
|
Footnotes |
|---|
|
Appendix A. Conference discussion |
|---|
|
TopAbstractIntroduction
Material and methods
Results
Discussion
Appendix A. Conference...References |
|---|
Dr Krieg: We took those measurements in the inhaled as well as in the exhaled air. We were first looking if there were any differences. We could not detect any significant differences between inhaled and exhaled air. But these are the points where we did those measurements.
Dr Ciulli: Was it a pulse delivery system or was it continuous?
Dr Krieg: Yes, it was a pulse delivery system.
Dr M. Hein (Kiel, Germany): Did you compare PGE1 infusion versus no inhalation, and did you test different dosages of NO concentrations, because for the therapy of pulmonary hypertension in the clinic concentrations between 5 and, at maximum, NO 20 ppm were used.
Dr Krieg: We didn't do any i.v. infusions with prostaglandin E1 because of the known side effects of increasing V/Q mismatch. Therefore we were concentrating on the inhaled application. Regarding your second question, we did look at different nitric oxide concentrations. Basically we were first looking at a dose response curve but finally decided to do the study with an NO concentration of 50 ppm because at the time of the study it wasn't that clear that lower doses of inhaled nitric oxide would be as effective as the higher ones.
Dr Hein: Ten ppm NO might not have a negative inotropic effect?
Dr Krieg: It may not. We don't know. At least we didn't measure it.
Dr G. Szabo (Heidelberg, Germany): The first question is do you measure nitric oxide in blood? It would be very important because we know that nitric oxide has a very short half-time in blood and it is questionable if it is inhaled that it has a significant effect on the left ventricle. Second, you do not show any parameters of the changes of afterload/preload after your intervention. You showed only the preload recruitable stroke work as a single index. Is it possible that nitric oxide and prostaglandin have different effects in amount on the changes of afterload and preload and your results may be explained in the different quantitative effects on nitric oxide and prostaglandin on afterload and preload conditions?
Dr Krieg: First, we did not measure nitric oxide in the blood since the half-time is just a few seconds and it is expected to be bound to hemoglobin very quickly. So basically all you can measure are nitrates and nitrites, but nitric oxide itself is very difficult to measure in the blood, so we didn't do that. Your second question concerning differences in preload, we chose the preload recruitable stroke work relation as a parameter of contractility because it is preload-independent. That is, because it is measured over a range of different preloads, so to speak, and only by measuring the stroke work at different preload levels can you generate this parameter. Therefore it is preload-independent by the way it is measured, and it has been shown that it is afterload-insensitive.
Dr Szabo: I ask this question because it's true that under certain circumstances preload recruitable stroke work is load-independent, but it is also known that the intrinsic myocardiac contractility may be changed due to altered loading conditions during a longer period (within 10 min). It means that if you apply your therapy, you can change preload conditions in the different groups and the hearts may adapt with a change of intrinsic contractility. In this case your change of contractility reflects not to the applied drug but to the differently altered loads.
Dr Krieg: Yes, that is true, but to take care of that problem we did our study in a randomized fashion; all the pigs received the same treatment, getting nitric oxide, prostaglandin, the combination, and the control substances, respectively. This was switched around according to a Latin square randomization schedule. So the time effect was taken out by this randomization process.
Dr C.M. Peniston (Toronto, Canada): First of all, did your treatments actually reduce the pressure in the pulmonary arteries? Secondly, do you think that somehow there could be an interaction between the effects of your drugs on the right ventricle which might affect the contractility of the left ventricle?
Dr Krieg: First, yes, pulmonary artery pressure was reduced by all three measurements; nitric oxide, prostaglandin, and the combination reduced it. The microsphere beads induced pulmonary hypertension. After the injection of microspheres, pulmonary pressure rose to about 26 from 15 mmHg, while after giving those substances we had a markedly reduced pulmonary pressure, a drop for about 5 to 10 points, almost back to normal. Regarding your second question concerning the right ventricular contractility. In a different experiment we also measured right ventricular contractility and could not detect any changes in contractility on the right side, which is a little bit confusing at first, but after thinking it over for a while, we hypothesized that maybe the process of how the contractility is decreased works over the endocardium. If NO would excert its negative inotropic effect through the coronaries, you would expect that both sides, left and right, would be depressed in their contractility. It may be the case that the way from the lungs to the coronaries is too long but the way from the lungs to the endocardium is still short enough for nitric oxide to have some effect on contractility.
|
References |
|---|
|
TopAbstractIntroduction
Material and methods
Results
Discussion
Appendix A. Conference...References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
