|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Cardiology (Drs. Ciarka, Vachièry, Gujic, Stoupel, Velez-Roa, and van de Borne, and Ms. Houssière), Erasme University Hospital, Brussels, Belgium; the Department of Pathophysiology (Dr. Naeije), Faculty of Medicine, Free University of Brussels, Brussels, Belgium.
Correspondence to: Agnieszka Ciarka, MD, Department of Cardiology, Erasme Hospital, 808, Lennik Rd, 1070 Brussels, Belgium; e-mail: Agnieszka.Ciarka{at}ulb.ac.be
Abstract
Background: We have reported previously that the sympathetic nervous system is activated in patients with pulmonary arterial hypertension (PAH), and that this is only partly explained by a decrease in arterial oxygenation. Possible causes for increased muscle sympathetic nerve activity (MSNA) in patients with PAH include right atrial distension and decreased cardiac output. Both may be improved by atrial septostomy, but this intervention also further decreases arterial oxygenation. In the present study, we wanted to investigate the effect of atrial septostomy on MSNA in patients with PAH.
Methods: We recorded BP, heart rate (HR), arterial O2 saturation (SaO2), and MSNA before and after atrial septostomy in PAH patients (mean [± SE] age, 48 ± 5 years) and in closely matched control subjects. Measurements were also performed after septostomy, while SaO2 was brought to the preprocedure level by supplemental O2 therapy.
Results: Compared to the control subjects (n = 10), the PAH patients (n = 11) had a lower mean BP (75 ± 2 vs 96 ± 3 mm Hg, respectively; p < 0.001), lower mean SaO2 (92 ± 1% vs 97 ± 0%, respectively; p < 0.001), increased mean HR (84 ± 4 vs 68 ± 3 beats/min; p < 0.01), and markedly increased mean MSNA (76 ± 5 vs 29 ± 2 bursts per minute; p < 0.001). Atrial septostomy decreased mean SaO2 (to 85 ± 2%; p < 0.001) and mean MSNA (to 69 ± 4 bursts per minute; p < 0.01), but did not affect HR or BP. Therapy with supplemental O2 did not affect MSNA, BP, or HR. The decrease in MSNA was correlated to the decrease in right atrial pressure (r = 0.62; p < 0.05).
Conclusions: Atrial septostomy in PAH patients decreases sympathetic hyperactivity despite an associated decrease in arterial oxygenation, and this appears to be related to decreased right atrial distension.
Key Words: atrial septostomy muscle sympathetic nerve activity pulmonary arterial hypertension
Heart failure is associated with activation of the sympathetic nervous system.12 While activation is initially beneficial due to maintenance of tissue perfusion pressure and cardiac output, sympathetic overactivation aggravates heart failure by increasing the metabolic demand of the failing myocardium, decreasing myocardial oxygen supply, increasing sodium and water retention, decreasing skeletal muscle strength, decreasing the arrhythmogenic threshold, and inducing myocardial as well as systemic vascular remodeling.34 Accordingly, in heart failure patients, sympathetic overactivity is associated with a poor prognosis, and ß-adrenergic blocking agents improve the clinical state and survival.5
In 2004, we reported6 that sympathetic nervous system activity, as assessed by peroneal nerve microneurography, is markedly increased in patients with pulmonary arterial hypertension (PAH), suggesting that the neurohumoral derangements reported in heart failure patients also occur in the particular situation of right ventricular dysfunction due to increased afterload. In that study,6 one fourth of the increased muscle sympathetic nerve activity (MSNA) was corrected by breathing supplemental O2, suggesting minimal chemoreflex participation, in contrast with the lack of any effect of supplemental O2 in patients with congestive heart failure.7
The mechanisms of sympathetic overactivity in heart failure patients are complex, and are believed to involve a decrease in cardiac output, atrial and ventricular wall stress, pulmonary arterial distension, decreased baroreflex sensitivity, increased muscle metaboreflex and chemoreflex sensitivity, and positive interactions with the endothelin and renin-angiotensin-aldosterone systems.1234589 Some of these derangements may be corrected by an atrial septostomy in patients with advanced PAH and refractory right ventricular failure.10 The procedure has been reported to improve clinical state, exercise capacity, and survival, as well as to decrease right atrial pressure (RAP) and to improve cardiac output, at the price, however, of a decrease in arterial oxygenation.10
We hypothesized that septostomy would decrease sympathetic nervous system overactivity in PAH patients if the beneficial effects of the improved cardiac hemodynamics could compensate for the shunt-induced, hypoxemia-associated chemoreflex activation. We, therefore, measured MSNA before and after septostomy in patients with advanced PAH and correlated the results to changes in right ventricular filling pressures, cardiac output, and arterial O2 saturation (SaO2), with and without the correction of hypoxemia with supplemental O2. The results suggest that increased right ventricular filling pressures contribute to sympathetic overactivity in PAH patients.
Materials and Methods
Patients
Eleven patients with PAH (5 men and 6 women; mean [± SE] age, 48 ± 5 years; body mass index [BMI], 22 ± 1 kg/m2) gave informed consent to participate in the study, which was approved by the Ethics Committee of Erasme Hospital. The patients had been scheduled to undergo atrial septostomy because of clinical deterioration despite optimal medical therapy. In two patients, atrial septostomy was being performed for the second time due to suspected atrial septal defect closure (the delays from the first atrial septostomy were 2 and 4 years). PAH was idiopathic in six patients, familial in two patients, associated with the previous intake of fenfluramine in two patients, and associated with a corrected ventricular septal defect in one patient. Specific therapies included beraprost (n = 2), epoprostenol (n = 2), treprostinil (n = 5), calcium channel blockers (n = 3), sildenafil (n = 1), and endothelin receptor blockers (ie, bosentan [n = 5], ambrisentan [n = 1], sitaxsentan [n = 2]). Conventional therapies included diuretics (ie, bumetanide [n = 4], furosemide [n = 6], spironolactone [n = 8], and hydrochlorothiazide [n = 1]), and anticoagulants. Therapy with anticoagulants was withdrawn 48 h before patients underwent atrial septostomy. All medications were kept unchanged during the study, except for therapy with diuretics, which was stopped or reduced in nine of the patients after they underwent atrial septostomy. All of the patients presented with advanced PAH, with New York Heart Association functional classes IV (n = 6) or III (n = 5), a mean 6-min walk distance of 376 ± 29 m, a mean Borg dyspnea score of 5.7 ± 0.5, and clinical signs of right heart failure, including turgescent jugular veins (n = 10), marked hepatomegaly (n = 6), and edema of the lower extremities (n = 4). Two patients presented with repetitive episodes of syncope.
Control Subjects
Ten healthy persons matched for age (mean age, 48 ± 2 years), BMI (mean BMI, 23 ± 1 kg/m2), and gender (5 were women) served as control subjects. All of them had normal clinical examination findings. None was receiving any medication.
Atrial Septostomy and Invasive Hemodynamic Determinations
A Swan-Ganz pulmonary artery catheter was inserted in nine patients via a femoral vein for the measurement of RAP, pulmonary artery pressure (PAP), pulmonary artery occlusion pressure, and mixed venous O2 saturation. SaO2 and mixed venous blood O2 saturation were measured by oximetry before and after patients underwent atrial septostomy. In two patients, it was impossible to place a Swan-Ganz catheter beyond the right atrium. In one of these patients, a Cournand catheter was placed to determine PAP. Left atrial pressure (LAP) was measured before and after the procedure using a pigtail fluid-filled catheter positioned across the interatrial septum. In one patient, we did not measure LAP before the atrial septostomy. In this patient, we used pulmonary artery occlusion pressure as a surrogate measure of LAP.
Atrial septostomy was performed as previously reported.11 A balloon catheter was passed across the septum through the sheath on a guidewire. The sheath was withdrawn to the right atrium, and the balloon was inflated until the waist was abolished under fluoroscopic control. Serial measurements were made of SaO2 and PAP. To obtain the measurements, the balloon was withdrawn into the sheath. The procedure was repeated with increasing balloon sizes until a septal defect was created ensuring a 10% fall in SaO2. The maximum sizes of the balloons used were as follows: 20 mm, two patients; 18 mm, two patients; 15 mm, two patients; 14 mm, three patients; and 12 mm, two patients. Cardiac output was estimated by the Fick method, based on an estimated O2 uptake, according to the equations of LaFarge and Miettinen.12
Microneurography
Microneurographic investigations were performed the day before and the day after atrial septostomy. The recordings were made in a quiet investigation room, after 15 min of rest in the supine position. Mean arterial BP was measured every minute using an oscillometric sphygmomanometer (Physiocontrol Colin BP-880; Colin Corp.; Komaki City, Japan). MSNA was recorded as previously reported (Nerve Traffic Analysis System; University of Iowa; Iowa City, IA).6 MSNA was recorded continuously in the patients and the control subjects by obtaining multiunit recordings of postganglionic sympathetic activity, measured from the nerve fascicle in the peroneal nerve, posterior to the fibular head. Electric activity in the nerve fascicle was measured with the use of tungsten microelectrodes (shaft diameter, 200 µm [tapering to an noninsulated tip of 1 to 5 µm]). A subcutaneous reference electrode was inserted 2 to 3 cm away from the recording electrode, which was inserted into the nerve fascicle. The neural signals were amplified, filtered, rectified, and integrated to obtain a mean voltage display of sympathetic nerve activity.7 The ECG and pulse oximetry O2 saturation (N100C; Nellcor; Pleasanton, CA) were recorded continuously.
All measurements were obtained the day before patients underwent atrial septostomy while they were breathing room air. Measurements obtained the day after atrial septostomy were performed first when breathing room air, then in nine patients during 15 min of breathing supplemental oxygen to increase O2 saturation to the preintervention level. The last 5 min of oxygen breathing were incorporated into the analysis.
Plasma Neurohormones
Venous blood was sampled the day before the atrial septostomy (n = 8) and the day after atrial septostomy (n = 7) for plasma norepinephrine, aldosterone, and renin measurements. Norepinephrine was measured by reversed phase, ion-pairing, high-performance liquid chromatography with electrochemical detection after a two-step extraction procedure involving cation exchange and alumina adsorption. Plasma renin level was measured by a radioimmunometric assay kit (N.V. Schering S.A.; Diegem, Belgium). Aldosterone was measured by radioimmunoassay (Diagnostic Products Corporation; Los Angeles, CA).
Statistical Analysis
Measurements of MSNA, heart rate (HR), mean BP (MBP), and SaO2 were averaged during 5 min of room air breathing before and after atrial septostomy, and also during 5 min while patients achieved pre-atrial septostomy O2 saturation levels. All of the results are expressed as the mean ± SE. Comparisons between patients and control subjects were performed using an unpaired Student t test. Measurements before and after atrial septostomy were compared using a paired Student t test. Comparisons between measurements obtained before and after atrial septostomy and again with patients receiving supplemental O2 therapy were performed using a repeated measures analysis of variance, with a posteriori Fisher exact tests. Correlations between a decrease in MSNA, a decrease in RAP, and an increase in cardiac output were analyzed using linear regression tests.
Results
Comparison of Baseline Variables Between PAH Patients and Control Subjects the Day Before Atrial Septostomy
The control subjects (n = 10) were closely matched to the PAH patients (n = 11) for age (mean age, 48 ± 2 vs 48 ± 5 years, respectively; p = 0.96), gender (five men in each group), and BMI (23 ± 1 vs 22 ± 1 kg/m2, respectively; p = 0.32). Compared to the control subjects, the PAH patients had a lower MBP (75 ± 2 vs 96 ± 3 mm Hg, respectively; p < 0.001) and SaO2 (92 ± 1% vs 97 ± 0%, respectively; p < 0.001), and a higher HR (84 ± 4 vs 68 ± 3 beats/min, respectively; p < 0.01) and MSNA (76 ± 5 vs 29 ± 2 bursts per minute, respectively; p < 0.001).
Hemodynamic Effects of Atrial Septostomy in PAH Patients in the Cardiac Catheterization Laboratory
Atrial septostomy was associated with a decrease in RAP and in SaO2 saturation, and an increase in LAP and cardiac output (Tables 1, 2
).
|
|
Effects of Preprocedural O2 Saturation Restoration During Recordings Obtained the Day After Atrial Septostomy vs the Day Before Atrial Septostomy
Supplemental O2 after atrial septostomy, to restore O2 saturation to preatrial septostomy values (n = 9), had no effect on MSNA, HR, or BP (Table 3
, Fig 1
).
|
|
Circulating Neurohormones
Atrial septostomy did not change the mean plasma levels of norepinephrine (0.532 ± 0.138 vs 0.447 ± 0.115 ng/mL, respectively; p = 0.66 [n = 6]; normal value, < 0.80 ng/mL), aldosterone (403 ± 121 vs 347 ± 104 pg/mL, respectively; p = 0.74 [n = 7]; normal value, < 160 pg/mL), or renin (182 ± 80 vs 197 ± 88 ng/L, respectively; p = 0.68 [n = 7]; normal value, < 16 ng/L).
Correlations Between MSNA and Hemodynamic Variables
The decrease in MSNA was directly related to the decrease in RAP (Fig 2
). There was no significant correlation between the change in MSNA and change in cardiac output.
|
The present results show that atrial septostomy in patients with advanced PAH is associated with a decrease in sympathetic nervous system overactivity, which appears to be related to decreased right atrial distension.
Heart failure is associated with neurohumoral activation.12345 We have shown that neurohumoral activation also occurs when heart failure is limited to the right ventricle and is caused by an increase in afterload in PAH patients.6 Thus, it is likely that the deleterious effects of neurohumoral activation, which include salt and water retention, altered chemoreflexes and muscle metaboreflexes, increased rest and exercise ventilation, cardiac and arterial remodeling, and arrhythmias,1234589 also contribute to PAH symptomatology.
The mechanisms of sympathetic nervous system activation in heart failure remain incompletely understood. In healthy subjects, the autonomic control of the cardiovascular system is mainly under the influence of sympathoinhibitory input from arterial and cardiopulmonary baroreceptors rather than sympathoexcitatory input from chemoreceptors and metaboreceptors.13 In patients with heart failure, the situation is reversed, with loss of restraining input from baroreceptors2 and increased excitatory inputs from chemoreceptors9 and metaboreceptors.8 In addition, there is evidence that increased cardiac filling pressures contribute to the activation of the sympathetic nervous system.14
Almost a century ago, Bainbridge15 observed that an increase in venous pressure caused by rapid volume infusion increased HR in dogs, and that this response was abolished by cardiac denervation. He thought that this response was neural in origin and that the receptors were located in the right atrium.15 The Bainbridge reflex explains the forearm vasodilatation reported in patients with severe congestive heart failure in response to upright tilt16 or lower body negative pressure.17 These observations are in keeping with the direct correlations between cardiac filling pressures and MSNA reported in heart failure patients.118 As already mentioned by Bainbridge,15 the observation of a relationship between RAP and measures of sympathetic activity requires the avoidance of arterial baroreflex interference. In healthy subjects, a decrease in filling pressures is associated with a decrease in cardiac output and BP, which triggers a baroreflex-mediated sympathoexcitation.17 A reduction in cardiac filling pressures by the application of lower body negative pressure has been reported to decrease cardiac norepinephrine spillover as a measure of cardiac sympathetic activation in patients with chronic heart failure, while the opposite was observed in healthy subjects.19
In the present study, atrial septostomy was followed by a decrease in MSNA but no change in HR. This is in keeping with the results of studies in patients with advanced heart failure, in whom the application of lower body negative pressure to decrease cardiac filling pressures decreased sympathetic activity but did not affect HR.19 Previous studies20 have reported a poor correlation between HR responses and indexes of sympathetic activity such as plasma norepinephrine or MSNA. In the present study, the absence of a decrease in HR with decreased RAP could be related to sinus node dysfunction with altered chronotropic responses, which have been reported in patients with chronic heart failure21 and also in patients with PAH.22
Cardiopulmonary baroreceptor unloading by changes in body position increases peripheral chemoreceptor sensitivity in healthy subjects.23 However, SaO2 reduction after cardiopulmonary baroreceptor unloading by atrial septostomy did not further increase the heightened sympathetic activity of the patients. This suggests important differences in MSNA regulatory mechanisms between PAH patients and control subjects. In addition, our previous study6 reported that the suppression of peripheral chemoreflex drive by 100% oxygen breathing decreased MSNA slightly among PAH patients. In the present study, oxygen flow was titrated in order to achieve an SaO2 of 92%, which corresponded to the level of SaO2 observed before the procedure. Thus, we did not suppress the resting peripheral chemoreflex drive after atrial septostomy, and this may explain the lack of sympathetic activity inhibition in response to oxygen supplementation in our PAH patients.
Plasma norepinephrine, aldosterone, and renin activity, which were measured in only seven of our patients, were not affected by atrial septostomy. This negative result may be explained by small sample size in the study, and by the fact that hormonal plasma levels are only indirect and insensitive estimates of production rates. We previously reported normal circulating catecholamine levels in the presence of markedly increased MSNA.6
Atrial septostomy was initially used to treat children with congenital cardiac defects24 and introduced in 1983 as a palliative therapy in patients with refractory idiopathic PAH.25 The rationale of this procedure is based on the notions that right heart failure is of poor prognosis in pulmonary hypertension,26 that pulmonary hypertensive animals with an interatrial communication have an improved exercise capacity,27 and that patients with severe pulmonary hypertension and a patent foramen ovale28 or Eisenmenger syndrome29 have a better quality of life and survival than patients with the same severity of pulmonary hypertension but no right-to-left shunting. The current worldwide experience of atrial septostomy in approximately 100 reported patients indicates that the procedure is associated with improved exercise capacity, quality of life, and even survival, although early mortality rates remain high, particularly in the most severely ill patients with extremely high RAPs.30313233
The hemodynamic effects of atrial septostomy include a decrease in RAP, usually by 2 to 3 mm Hg, a proportional increase in LAP, a decrease in SaO2, and an increase in cardiac output of up to 30 to 35%.11 Similar changes were observed in our patients in terms of the increase in cardiac output and the decrease in RAP. It is also of interest that in our patient group arterial BP remained unchanged, as expected,11 allowing for the observation of an isolated decrease in RAP, as the clinical counterpart of an isolated increase in RAP in the experiments of Bainbridge.15
Sympathetic overactivity in patients with advanced PAH appears to be related to right atrial distention. However, the Bainbridge reflex cannot entirely explain sympathetic overactivity because the decrease in RAP after atrial septostomy decreased MSNA by only about 10%. Accordingly, MSNA still remained more than twice as high in PAH patients after the procedure, as in control subjects. Limited cardiac output and other mechanisms such as abnormal metaboreflex and baroreflex functioning, and renin-angiotensin-aldosterone system activation could contribute to persistent MSNA elevation after atrial septostomy.
Therapy with diuretics was stopped shortly after atrial septostomy in the majority of patients because they presented with satisfactory diuresis and rapid peripheral edema clearance. This can be explained by the lesser activation of the renin-angiotensin-aldosterone system and better renal perfusion; both phenomena related to the improved cardiac output after atrial septostomy. In addition, there is evidence that a selective reduction in filling pressures lowers cardiac norepinephrine spillover in patients with congestive heart failure.19 Taken together, it seems unlikely that diuretic discontinuation after atrial septostomy can explain the reduction in MSNA that we observed.
The clinical benefit of atrial septostomy is generally believed to be related to increased O2 delivery to the tissues, because of a proportionally more important increase in cardiac output than right-to-left shunt-related decrease in arterial O2 content.1531 However, there are no reports of increased maximum O2 uptake related to an increase in maximum O2 delivery in PAH patients after atrial septostomy. The present observations indicate that at least part of the clinical improvement in PAH patients after atrial septostomy may be related to a lesser degree of sympathetic nervous system activation.
Acknowledgements
We are indebted to Marie Thérèse Gautiers help with data collection.
Footnotes
Abbreviations: BMI = body mass index; HR = heart rate; LAP = left atrial pressure; MBP = mean BP; MSNA = muscle sympathetic nerve activity; PAH = pulmonary arterial hypertension; PAP = pulmonary artery pressure; RAP = right atrial pressure; SaO2 = arterial oxygen saturation
This study was supported by AstraZeneca (Dr. Ciarka), Sankyo (Ms. Houssière), Foundation for Cardiac Surgery (Drs. Ciarka, Naeije, and van de Borne), the Erasme Foundation (Dr. Velez-Roa), the National Fund for Research-Belgium (Drs. Naeije and van de Borne), the Emile Saucez-René Van Poucke Foundation (Dr. van de Borne), and the Foundation David et Alice Van Buuren (Dr. van de Borne).
The authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.
Received for publication December 6, 2006. Accepted for publication March 5, 2007.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |