Main

Balloon dilatation is finding increasing application as a palliative nonsurgical method for therapy of congenital heart diseases such as coarctation of the aorta(1, 2), pulmonary stenosis(3–5), and tetralogy of Fallot(6–9) in children, infants, and neonates. In particular, balloon angioplasty appears to reduce the probability of developing paradoxical hypertensive crisis(10) which is often observed after surgical relief of aortic coarctation(11, 12). The development of such hypertensive crisis is associated with a sudden elevation in circulating catecholamines, particularly norepinephrine, as a result of surgical stress(1, 13–15). Accordingly, balloon angioplasty of aortic coarctation appears to successfully prevent the events triggering an increased spillover of this hormone, implying therefore that this procedure may lead to an immediate restoration of normal sympathetic activity.

The influence of balloon dilatation on the sympathetic nervous system in pulmonary stenosis is unknown. Unlike coarctation of the aorta, patients with pulmonary stenosis do not experience clinically significant increase in arterial pressure after balloon dilatation. The increase in systolic pressure in the pulmonary artery and in peak pressure in these patients is generally attributed to improved flow through the valve. It is also well established that balloon dilatation usually leads to the relief of the valvular gradient and a drop in systolic gradient(4, 5), probably due to the changes in pulmonary blood flow and the valve area. These changes are usually observed soon after the balloon valvuloplasty. We hypothesized that such drastic acute alterations in the hemodynamic factors are likely to challenge the sympathetic system to adapt to the changed conditions in a similarly subtle fashion. In this study, we therefore compared the α- and β-adrenoceptor density and responsiveness, as well as the plasma catecholamine and cAMP levels in infants and children before and 10 min after balloon dilatation of pulmonary stenosis, to evaluate the potential alterations in sympathetic activity associated with this procedure and their clinical implication for the postprocedural care of such patients.

METHODS

Study population. Thirty-one infants and children ranging in age between 8 mo to 15 y (mean age 58.1 ± 17.4 mo) admitted for balloon dilatation of pulmonary stenosis were included in the study. Fourteen (45%) of the patients were male and 17 (55%) were female. All patients were premedicated with a mixture of meperidine, promethazine, and chlorpromazine. Ketamine (0.5 mg/kg) was occasionally administered, if required. The size of the angioplasty balloon catheter was selected to be 1.2-1.4 times that of the pulmonary artery annulus. Before advancing the catheter across the stenosis, the heart rate and the systemic pressure were recorded via an 18-inch sheath positioned in the right femoral artery. After obtaining the pressures and saturation in the right heart, the valvuloplasty catheter was positioned across the stenosis, and the balloon was inflated several times with diluted contrast medium to 4-5 atmospheres for a few seconds. Several hemodynamic variables including pulmonary artery pressure, right ventricular pressure, pullback peak to peak systolic gradient across the pulmonary valve, and systemic pressure were recorded before and 10 min after balloon dilatation. Fifteen consecutive children and infants ranging in age between 11 mo to 13 y(mean age 60.8 ± 12.3 mo) undergoing transcatheter occlusion for PDA with a Qp/Qs < 1.5 were elected as controls. These patients were considered to be functionally normal with regard to their sympathetic activity. The study was performed in accordance with the rules and regulations laid down by the Hospital's Ethics and Clinical Protocols Committees.

Binding studies. Blood samples were drawn in EDTA tubes between 0800 and 1000 h on the morning of the balloon dilatation in study patients or occlusion in PDA patients and 10 min after the respective procedures. The blood was layered over Leucoprep (Becton Dickinson, Heidelberg, Germany) and centrifuged at 1500 × g for 15 min. The supernatant was collected for catecholamine and cAMP determination; the fluffy layer was resuspended in Tris/EDTA buffer and centrifuged at 200 × g for 10 min. The resultant lymphocyte pellet was suspended in Tris buffer for the β-adrenoceptor studies. The supernatant was recentrifuged at 1800 × g to obtain the platelet pellet which was also suspended in Tris/EDTA buffer forα-adrenoceptor studies.

The α-adrenoceptor density and binding affinity were determined by specific binding of [3H]yohimbine (83 Ci/mmol, NEN Corp., Stevenage, UK) to platelets. Approximately 107-108 cells were suspended in Tris/EDTA buffer containing 50 mM Tris-HCl (pH 7.4), 5 mM EDTA, 120 mM NaCl, 10 mM MgCl2, and incubated with 0.5-8.0 nM [3H]yohimbine at room temperature for 30 min in a final volume of 300 μL. The reaction was terminated, and the cells subsequently washed twice with 5 mL of iced-cold Tris buffer through Whatman GF/B filters (Whatman Inc., Clifton, NJ) on a Brandel Harvester (Biomed Research and Development Laboratories, Inc., Gaithersburg, MD). The filters were dried, and the radioactivity was counted in 10 mL of Optifluor scintillation fluid (Packard, Meridian, CT). The difference between the platelet binding to [3H]yohimbine in the absence and presence of 50 μM phentolamine was considered specific binding.

The β-adrenoceptor activity was determined by specific binding of[125I]iodocyanopindolol (2200 Ci/mmol, NEN Corp.) to lymphocytes. Approximately 1 × 106 cells were suspended in Tris buffer (pH 7.4) containing 50 mM Tris-HCl, 120 mM NaCl, 10 mM MgCl2, and incubated with 10-160 [rho]M [125I]iodocyanopindolol at 37°C for 60 min in a final volume of 300 μL. The reaction was terminated, and the cells were washed as described above. The radioactivity was counted on an LKB gamma counter. The difference between the receptor binding to[125I]iodocyanopindolol in the absence and presence of 5.0 μM propranolol was considered specific binding. Assays were conducted in duplicate.

Plasma catecholamine and cAMP levels. The catecholamines norepinephrine, epinephrine, and dopamine were extracted from plasma with activated alumina according to the method of Anton and Sayre(16) and assessed by HPLC with electrochemical detector(Waters System, Miliford, MA). cAMP was determined by RIA (Amersham Corp., Amersham, UK). Drugs used were [3H]yohimbine,[125I]iodocyanopindolol (NEN), propranolol (Beck Pharmaceuticals, Macclesfield, UK), and phentolamine (Sigma Chemical Co., St. Louis, MO). All other reagents were of analytical grade.

Statistical analysis. Receptor density and ligand binding were calculated from Scatchard saturation binding isotherms using Receptorfit Saturation Two-site Program (Ligand Software, Cleveland, OH). Analysis of variance followed by Scheffe's test was performed using Statgraphics Software(Statistical Graphics Corp., Rockville, MD). The pre- and postangioplasty data were compared by the t test using Sigmaplot (Janel Scientific Software, San Rafael, CA) statistical package, taking the probability levels of less than 0.05 as indicating significant difference. Data are given as means ± SEM.

RESULTS

After balloon dilatation, the mean right ventricular systolic pressure decreased significantly from 110.1 ± 6.3 to 62.5 ± 6.9 mm Hg(p < 0.0000002), right ventricular diastolic pressure from 13.1± 1.3 to 9.8 ± 0.9 mm Hg (p < 0.05), peak-to-peak systolic gradient from 89.2 ± 7.4 to 13.3 ± 0.7 mm Hg(p < 0.000001), and the heart rate from 116 ± 6.5 to 107.5± 5.6 beats/min (p < 0.05). On the other hand, the systolic pulmonary pressure increased also significantly from 21.9 ± 0.9 to 25.8 ± 2.0 mm Hg (p < 0.05). A summary of these and other determined clinical parameters is given in Table 1.

Table 1 Hemodynamic and clinical data of 31 patients before and 10 min after balloon dilatation of pulmonary stenosis
Full size table

In the PDA patients, the mean Kd value for theα-adrenoceptor binding to [3H]yohimbine was 2.12 ± 0.32 nmol/L (n = 15) before occlusion, and 1.61 ± 0.15 nmol/L 10 min thereafter. The Bmax value was 3.57 ± 0.38 fmol/107 cells. It remained unchanged at 3.36 ± 0.47 fmol after occlusion. There was no significant difference in the α-adrenoceptor affinity and density of the PDA group before and immediately after occlusion. In the pulmonary stenosis population (n = 31), theKd was 3.13 ± 0.56 nmol/L before balloon dilatation and decreased to 2.18 ± 0.26 nmol/L 10 min after the valvuloplasty. The receptor density (Bmax) was 5.53 ± 0.64 fmol before, and 3.86 ± 0.24 fmol per 107 cells after, the balloon dilatation (Fig. 1). Analysis of variance showed that theα-adrenoceptor density before dilatation was significantly higher than that of the PDA as well as the postdilatation value. On the other hand, the postprocedure density was similar to that of the PDA group.

Figure 1
figure 1

The α-adrenoceptor density(Bmax) in 31 children with pulmonary stenosis before(PSA) and 10 min after (PSP) balloon valvuloplasty, compared with children undergoing transcatheter occlusion of PDA before(n = 15) and after (PDP, n = 7) the procedure.*p < 0.05 precompared with postdilatation.

Full size image

The Kd for the β-adrenoceptor binding to[125I]iodocyanopindolol in PDA patients was 8.09 ± 1.32 pmol/L before and 7.77 ± 1.76 pmol/L after occlusion. TheBmax also hardly changed from 64.8 ± 11.0 fmol to 71.2± 13.2 fmol/106 cells. In the pulmonary stenosis patients, theKd increased from 6.28 ± 0.62 pmol/L to 11.5 ± 2.7 pmol/L, and the density from 50.0 ± 5.5 fmol to 73.7 ± 13.4 fmol/106 cells 10 min after the dilatation (Fig. 2). Thus, in contrast to α-adrenoceptors, the β-adrenoceptor density of the study patients was lower than that of the control group prior to valvuloplasty, and increased to the level of the latter thereafter. The analysis of variance showed no significant difference among the densities of the study patients and the controls. However, the receptor density after dilatation was 47% (p < 0.07) higher than before the procedure, showing that the increase was nevertheless quite large.

Figure 2
figure 2

The β-adrenoceptor density (Bmax) in 31 children with pulmonary stenosis before (PSA) and 10 min after(PSP) balloon valvuloplasty, compared with children undergoing transcatheter occlusion of PDA before (n = 15) and after(PDP, n = 7) the procedure. *p < 0.07 pre- compared with postdilatation.

Full size image

Plasma norepinephrine level of the controls was 558.5 ± 104.5 pg/mL, epinephrine was 266.3 ± 63.1 pg/mL, dopamine was 103.6 ± 14.4 pg/mL, and cAMP was 15.5 ± 1.6 pg/mL. The plasma levels of these catecholamines and cAMP in the study patients are depicted in Table 2. The table shows that there was no significant difference between the pre- and post-dilatation plasma norepinephrine levels of the study group and those of the PDA group. On the other hand, epinephrine was only slightly elevated against the PDA levels before dilatation, but increased significantly (p < 0.05) by 73% after the procedure. Dopamine was 80% (p < 0.05) and cAMP 37% higher before dilatation, and remained significantly elevated at 70% (p < 0.05) and 23% above the PDA values thereafter.

Table 2 Plasma catecholamines and cAMP in 31 pulmonary stenosis patients before and 10 min after balloon angioplasty of pulmonary stenosis, compared with those of 15 PDA controls
Full size table

DISCUSSION

The present study investigated the influence of balloon valvuloplasty on the sympathetic function in children with pulmonary stenosis. As controls, we determined the α- and β-adrenoceptor density and their ligand binding affinity in children undergoing occlusion of PDA under the same conditions as the study population. The results show that neither the plateletα-adrenoceptor nor the lymphocyte β-adrenoceptor density and their ligand binding affinity were altered as a result of the occlusion in this group of patients. This finding supports the argument in favor of employing PDA patients with an insignificant left-to-right shunt as controls to evaluate alterations in the sympathetic function in children undergoing surgical or therapeutic manipulations for congenital disorders.

In children with pulmonary stenosis, the lymphocyte β-adrenoceptor density was reduced, and the α-adrenoceptor density significantly elevated. A decrease in β-adrenoceptor density is a well established phenomenon in heart failure associated with various forms of cardiomyopathies(17–20). It has also been previously suggested that, in heart failure, α-adrenoceptors may be mobilized to provide inotropic support to the failing heart(21–23). Heart failure is, however, not a common feature of pulmonary stenosis. Our findings suggest that the combination of an elevation in α-adrenoceptors and attenuation of theβ-adrenoceptors may explain the mode by which the sympathetic nervous system adapts itself to hemodynamic changes in pulmonary stenosis, with or without the manifestation of heart failure. Furthermore, we recently observed a similar increase in α-drenoceptors in patients with other congenital diseases(24) and in stenotic heart valvular disorders(25). A study by McGrath et al.(26) using neuronal markers also indicated that, in the infundibular tissue of patients with tetralogy of Fallot, α-adrenoceptor content was increased. Accordingly, the elevation in α-adrenoceptors may indeed constitute a compensatory mechanism for the decrease inβ-adrenoceptors, not only in pulmonary stenosis, but in congenital heart disorders in general.

The most important finding of the present study is the observation that both the elevation in the α-adrenoceptor density and the attenuation of the β-adrenoceptors were reversed immediately after balloon angioplasty. One possible mechanism for the decrease in the α-adrenoceptor density is receptor internalization. This phenomenon has been described forβ-adrenoceptors(27, 28), but there is hardly any documentation of a similar behavior by α-adrenoceptors in cardiovascular diseases. Conversely, the increase in β-adrenoceptor density after balloon valvuloplasty may be due to mobilization of spare receptors(29). The observation that both α- andβ-adrenoceptor levels tend to return to normal indicates that the underlying mechanism(s) controlling the two processes are probably interrelated. Furthermore, the time within which these alterations occurred suggests that they were acute responses of the sympathetic system to the effects of balloon valvuloplasty itself. This may occur as an adaptive mechanism by the sympathetic system to meet the requirements of the altered conditions triggered by the abrupt hemodynamic changes. It is noteworthy that the modifications in receptor activity were accompanied by significant alterations in some important hemodynamic variables, such as the systolic pulmonary pressure, right ventricular systolic and diastolic pressures, and the systolic gradient. These changes, which are essentially the product of improved blood flow through the pulmonary valve, are not likely to be directly responsible for the alterations in the receptor activity per se. Nonetheless, it appears plausible to suggest that several complex mechanism(s) may be responsible for the changes in the adrenoceptors, probably involving vagal responses to the sudden changes in the hemodynamic variables, at least indirectly, as a product of the balloon valvuloplasty. Thus, assuming that the sympathetic system may be chronically hyperstimulated in pulmonary stenosis to counteract the pressure resulting from the pulmonary obstruction, sudden relief from this pressure is likely to trigger the vagus to function in conformation with the new hemodynamic conditions. This may occur by transiently stimulating an increase in catecholamine turnover, as indicated by the rise in epinephrine in our patients. This, in turn, would lead to mobilization of spare β-adrenoceptors before a long-lasting adaptation is attained.

Similarly acute changes have been associated with a significant rise especially in plasma norepinephrine levels, possibly as a reflection of its spillover from sympathetic nerve terminals after surgical correction of aortic coarctation(13, 15). However, norepinephrine is less likely to be influenced in an abrupt fashion by a nonsurgical procedure such as balloon dilatation. Indeed, as observed in our patients, plasma norepinephrine levels were not influenced by either pulmonary stenosis or balloon valvuloplasty. On the other hand, epinephrine was only slightly elevated before valvuloplasty, but increased significantly thereafter. The trends established for norepinephrine and epinephrine in the present study are in general agreement with the observations of Lewis and Takahashi(30), showing no change in norepinephrine, but a significant elevation of epinephrine, although not significantly increased further by balloon angioplasty. They therefore provide additional evidence that norepinephrine is not influenced, whereas epinephrine may be increased in pulmonary stenosis and elevated further by balloon valvuloplasty. This rapid increase in circulating epinephrine soon after balloon dilatation is probably triggered by stress resulting from the angioplasty or valvuloplasty. This may explain, in part at least, the time course of the increase inβ-adrenoceptor activity. It is not likely, however, that stress alone can adequately explain all of the observed changes in both adrenoceptor subtypes.

The observation that plasma dopamine and cAMP may be elevated in congenital heart disorders is quite intriguing. Brodde(31) has suggested that dopamine may be influenced in cardiac disorders. However, there is hardly any further documentation in the literature to verify this notion. Further studies should enrich our understanding of its potential role in these diseases. The elevation in cAMP may, on the other hand, be a product of several complex events involved in the signaling pathways of either theα- or the β-adrenoceptor subfamilies. Thus, such an increase might be expected under conditions that will stimulate adenylate cyclase activity, or some other indirect mechanism(s) not associated with β-adrenoceptor signaling pathways. One such possibility is the stimulation of Gs-coupled synthesis of cAMP, either by increasing the signaling levels of the Gsα-protein at the expense of Giα protein(32–34), or presence of activating mutational changes in the Gsα proteins(35–37). Inasmuch as we actually observed a decrease in β-adrenoceptor activity in our patients, it is highly unlikely that the increase in cAMP in our study population is a result of the stimulation of the Gs via the β-adrenoceptor signaling pathway. Other possibilities include pathways that are not dependent on adenylate cyclase activity, such as the inhibition of membrane-associated cGMP-inhibited cAMP-phosphodiesterase activity(38), which should attract further attention.

Further studies may greatly enhance our understanding of the possible clinical implications of the rapid changes in both the sympathetic activity and circulating epinephrine, particularly with regard to postvalvuloplasty patient care. One interesting point in this regard is the observation that the patients who develop infundibular stenosis after balloon dilatation do respond well to β-adrenoceptor blockers(39, 40). There appears to be at least a causal relationship between the increase inβ-adrenoceptors and the events leading to the development of infundibular stenosis after balloon dilatation in such patients. These patients might be experiencing hyperactivity of these receptors as a result of the valvuloplasty, which would explain their response to β-adrenoceptor blockers. There are also a small number of patients who develop infundibular reaction in conjunction with other complications, such as low blood pressure. Recently, we found that the administration of an α-adrenoceptor blocker to be the ultimate remedy in a young patient who developed such complications, after all other attempts including β-blocker therapy had failed. Although the exact mechanism(s) remains to be clarified, it seems plausible to suggest that such a patient might have had very high levels of α-adrenoceptors which remained elevated even after the valvuloplasty. Put together, the present findings indicate that the changes in the adrenoceptors may be of clinical relevance, particularly with regard to the use of sympathetic agents in patients who develop complications after balloon valvuloplasty. Further studies should be directed at delineating the precise mechanism(s) involved in the changes in the receptor function.

In conclusion, the present study demonstrates that α-adrenoceptor density is significantly elevated in children with pulmonary stenosis and decreases significantly immediately after balloon valvuloplasty. Conversely,β-adrenoceptor density is attenuated in these patients and returns to normal levels after the procedure. These changes are associated with a significant drop in the systolic gradient across the pulmonary valve and a significant increase in plasma epinephrine levels. These observations suggest that balloon dilatation exerts acute effects on the level of sympathetic activity in patients with pulmonary stenosis.