Several cases of systemic arteriovenous fistula diagnosed in the human fetus have been associated with the postnatal development of persistent pulmonary hypertension. The aim of this study was to determine the effects of a prenatally created systemic arteriovenous fistula on the structure and reactivity of the pulmonary circulation in the fetal lamb. A fistula between the jugular vein and carotid artery was created in fetal lambs at 119-124 days of gestation. At delivery (134-139 days), left pulmonary artery (LPA) pressure was increased in the fistula group (n = 12) compared with controls (n = 11, P < 0.01). The pulmonary vascular resistance was significantly higher in the fistula group (P < 0.05), whereas mean LPA blood flow was not statistically different between the two groups. Morphometric analysis of the pulmonary vascular bed revealed an increase in the number of peripheral muscular arteries, together with an increase in pulmonary arterial medial thickness in the fistula group. There was no difference in the relative number or size of intraacinar arteries. In vitro organ bath studies on pulmonary arterial rings showed impaired endothelium-dependent relaxation in the fistula group compared with controls. However, endothelial nitric oxide synthase protein expression was similar in both groups, whereas endothelium-independent relaxation to sodium nitroprusside was greater in the fistula group compared with controls. A systemic arteriovenous fistula leads to both structural and functional alteration of the pulmonary vasculature, which might lead to the development of persistent pulmonary hypertension after birth.
- pulmonary circulation
- persistent pulmonary hypertension of the newborn
in the fetus, the ductus arteriosus (DA) allows right ventricular (RV) output to be diverted from the pulmonary circulation to the descending aorta; therefore, <10% of the cardiac output reaches the pulmonary circulation. This diversion is facilitated by a high pulmonary vascular resistance. During the first minutes of extra uterine life, pulmonary vascular resistance dramatically falls. This fall is associated with a 10-fold increase of the pulmonary arterial flow, allowing pulmonary gas exchange (29). Failure of the process may lead to persistent pulmonary hypertension of the newborn (PPHN). This is characterized by elevated pulmonary arterial pressure, abnormal pulmonary vascular reactivity, and high pulmonary vascular resistance leading to severe hypoxemia. The timing of onset, within the first hours of life, and the presence of striking structural remodeling of the pulmonary vascular bed associated with this syndrome have led to the hypothesis that intrauterine events may lead to the development of PPHN in the postnatal period (25). Several prenatal models of fetal pulmonary hypertension have been described over the past 20 years. Chronic fetal hypoxia induced by maternal hypoxia, serial embolization of the placenta, or cord compression induces pulmonary hypertension in the fetal lamb (2, 11, 31). However, these models were associated with a high fetal mortality of up to 60%. Fetal lambs born 7-14 days after prenatal ligation or constriction of the DA have persistent pulmonary hypertension at birth with many of the physiological and structural hallmarks of human PPHN (4, 22). This model has been widely used to demonstrate the key role of nitric oxide (NO) in the regulation of the pulmonary vascular tone in the perinatal period (30, 36).
PPHN can be associated with a wide variety of perinatal disorders, including intrapartum asphyxia, meconium aspiration, sepsis, congenital heart defects, and congenital diaphragmatic hernia or can be idiopathic (28). We and other study groups have reported cases of systemic arteriovenous fistula diagnosed in the prenatal period that were associated with the development of PPHN in the postnatal period (9, 13, 15, 26). These observations have led us to hypothesize that a systemic arteriovenous fistula by perturbating the fetal hemodynamics may affect the pulmonary vascular circulation leading to PPHN at birth. The aim of our study was to create a prenatal model of arteriovenous fistula in the fetal lamb and to determine if such a lesion alters the fetal pulmonary vascular reactivity and structure.
MATERIALS AND METHODS
Twenty-nine fetal lambs were obtained from 22 pregnant Pre-Alp ewes and divided into two groups. Animals received care in accordance with the “Principles of Laboratory Animal Care” and the “Guide for the Care and Use of Laboratory Animals” (5). In the first group (fistula group, n = 18), a systemic arteriovenous fistula was created at 119-124 days of gestation (full term 145 days). The control group (n = 11) consisted of unoperated fetuses whose cotwins or cotriplets had been operated upon and of fetuses from unoperated ewes. Cotwins and cotriplets were considered to be control animals, because previous reports have shown that when the amniotic cavity of the second twin is not opened, hemodynamic and histological data were not different from fetuses from unoperated ewes (6, 7). The details of the surgical protocol have been previously described (16). In brief, pregnant ewes were fasted for 48 h before surgery. Ewes were sedated with intravenous injection of penthothal (10 mg/kg body wt) and anesthetized with 1% halothane. The fetal neck was exposed through a 5-cm uterine incision while the fetal head was kept in the amniotic cavity. The internal jugular vein and the carotid artery were exposed by gentle dissection and controlled with vessel loops. A jugular venotomy and a carotid arteriotomy, 10-12 mm in length, were performed with a no. 11 bladed knife. The anastomosis was then performed by a continuous suture technique with 7.0-Proline (Ethicon). The anastomosis was heparinized just before closure, and vessel loops were removed. The neck incision was sutured. Warm saline was infused into the amniotic cavity to replace the lost amniotic fluid, and the uterine incision was closed. Antibiotics (1 g of ampicillin and 50 mg of gentamicin sulfate) were administered to the ewe during surgery and repeated daily for 5 days.
Delivery and Physiological Measurements
Fourteen to fifteen days postoperatively (134-139 days), ewes underwent a second laparotomy under the same anesthetic procedures. The fetal chest was exposed through a uterine incision. A left thoracotomy exposed the heart and great vessels. Polyvinyl catheters were inserted into the atria, ventricles, aorta, and left pulmonary artery (LPA) by direct puncture after the placement of purse string sutures. An ultrasonic flow transducer size 6 (Transonic Systems, Ithaca, NY) was placed around the LPA to measure blood flow. All measurements were made after a 10-min recovery period. Mean pressures were obtained by electrical integration and recorded on a multichannel recorder (Sirem Siemens, Erlangen, Germany). Pulmonary vascular resistance was calculated as the difference between mean LPA and left atrial pressure divided by flow (4). Finally, 5-ml blood samples were obtained from the ascending aorta and from the LPA for blood gas and pH analysis carried out with a blood gas analyzer at 37°C (ABL 700; Radiometer, Copenhagen, Denmark). The patency of the fistula was then confirmed, and lambs were killed by quick exsanguination.
The lambs were weighed. The heart and lungs were removed en bloc immediately after death and weighed. The free wall of the RV was removed from the fetal heart by gentle dissection, and RV and left ventricle plus septum (LV + S) were weighed separately. To assess the presence of RV hypertrophy, we calculated the ratio of the weights of the RV to LV + S as previously described (4).
Lung tissue preparation. We prepared the left lung (fistula group n = 6, control group n = 6) for light microscopy by cannulating the LPA and the left main bronchus. The pulmonary artery was perfused with 10% formaldehyde at 40 mmHg for 10 min as previously described (4). This fixative pressure was also similar to the physiological pressures found in our controls. The same pressure was used for both groups so that measurements on the pulmonary arteries were comparable. The airway was then distended with 10% formaldehyde at 35 cmH2O pressure for 2 min, the left main bronchus was ligated, and the fully distended lung was immersed in fixative for 48 h. After fixation, the lung was cut into slices 3-5 mm thick. Four blocks from different slices of parenchymal tissue were selected for each lamb. Tissue blocks were processed through graded alcohols, embedded in paraffin, and sectioned at 4 μm. Tissue sections were stained with hematoxylin and eosin and also Miller's elastic van Gieson stain to demonstrate muscle, connective tissue, and elastin.
Structure of pulmonary arteries. In each lamb, at least 100 pulmonary arteries were analyzed through a ×40 objective. The external diameter (ED) was measured as the shortest distance between the outer elastic lamina in any transverse section. The structure of each vessel was noted (fully or partially muscular, nonmuscular). For the muscular vessels, the medial thickness was measured as the distance between the internal and the external elastic lamina. The percentage medial wall thickness (MT) was calculated: %MT = (2 × MT)/(ED × 100) (14). For analysis, vessels were separated into five groups according to ED.
Size of intraacinar arteries and arterial density. The diameter of arteries alongside alveolar ducts was measured via a ×40 objective. At least 12 arteries were measured in each lamb. The number of arteries and the number of alveolar profiles were counted in the same field through a ×20 objective. At least 10 fields were examined per lamb.
Measurement of the MT in the conduit arteries. To study the structure of the arteries used in the organ chambers studies, we took blocks of tissue from the midlung region of the lower lobe to include the conduit arteries. Five-micrometer sections were stained with Miller's elastic van Gieson stain. The ED conduit arteries were measured, and four measurements of the wall thickness at intervals around the wall were made. At the same points, the number of elastic laminae in the wall was counted. The MT, the percentage wall thickness, and the interelastic lamina distance were calculated.
Isometric Tension Study
Tissue preparation. Immediately after death, the right lung (fistula group n = 12, control group n = 11) was immersed in an oxygenated cold Krebs solution (in mM): 118 NaCl, 5.9 KCl, 1.2 MgSO4, 2.5 CaCl2, 1.2 NaH2PO4, 25.5 NaHCO3 and 5.5 glucose. Axial intralobar arteries were isolated from the lung by gentle dissection, carefully removed to minimize vascular compression or stretch, and cleaned under magnification with gentle removal of the surrounding adventitia. Third-generation pulmonary artery rings (1-1.5 mm diameter) were cut into 3-mm lengths, placed on horizontally oriented thin steel wires attached to a force displacement transducer (Sigma-Aldrich, St. Quentin-Fallavier, France), suspended in 20 ml of Krebs-Ringer in a glass-jacket muscle bath at 38°C, and continuously oxygenated with 21% O2, 5% CO2, and 74% N23). They were then allowed to equilibrate for 40 min and rinsed with fresh buffer three times during the equilibration period. The vessels were then contracted with cumulative concentrations of phenylephrine from 10-8 to 10-5 M.
Experimental design. To investigate the NO/cGMP pathway, we measured change in isometric force of phenylephrine-precontracted rings after the cumulative addition of one of the following pharmacological agents.
) Endothelium-dependent relaxation to acetylcholine (ACh, 10-8-10-4 M), adenosine diphosphate (ADP, 10-8-10-4 M), and calcium ionophore A-23187 (10-8-10-4 M) was assessed in pulmonary artery rings with endothelium to study the role of stimulated NO production. The relaxation to ACh was also determined in the absence and presence of the specific NO synthase (NOS) inhibitor 10-4 M Nω-nitro-l-arginine (l-NA).
) Endothelium-independent relaxation to the NO donor sodium nitroprusside (SNP, 10-8-10-4 M) was also assessed.
All pulmonary artery rings were preincubated with indomethacin (10-5 M) to exclude the involvement of endogenous cyclooxygenase metabolites. All pharmacological agents were purchased from Sigma-Aldrich. All vessels rings were exposed to only a single drug. For each experiment, n refers to the number of animals. Relaxation was expressed as percentage of maximal precontraction of the vessels to phenylephrine. To assess the integrity of the endothelium of the vessels studied, we fixed five rings from the fistula group and six from the control group in 10% formaldehyde for further light microscopy examination.
Endothelial NOS Western Blotting
Western blot experiments were performed as described previously (17). In brief, peripheral lung parenchyma were homogenized in lysis buffer (50 mM Tris · HCl pH 7.4, 0.1 mM EGTA, 1 μM EDTA, 1 μM leupeptin, 1 μM aprotinin, and 1 μM PMSF) with an Ultraturrax T25 (Janke and Kunkel; IKA Works, Cincinnati, OH). Samples were then centrifuged at 3,000 g for 15 min. Aliquots of supernatants containing 150 μg of total protein were denaturated by boiling in sample buffer for 5 min (0.5 M Tris · HCl, pH 6.8, 10% SDS, 10% glycerol, 5% 2β-mercaptoethanol, and 1.25% bromphenol blue), then loaded and separated by electrophoresis in each well of a 7.5% SDS-PAGE gel. We then transferred proteins to a polyvinylidene difluoride membrane (Bio-Rad Laboratories, Richmond, CA) and immunoblotted them with monoclonal anti-endothelial NOS (eNOS) antibodies (Transduction Laboratories, Lexington, KY) at a 1:1,000 dilution. Immunoreactive proteins were then visualized by chemiluminescence. The eNOS band was quantified by densitometric analysis. After detection, membranes were stained with red Ponceau's solution, and the total amount of proteins was also assessed by densitometric analysis.
Data were expressed as means ± SE. The hemodynamic data were analyzed by a t-test. Data that consisted of repeated measurements (structural analysis and isometric tension studies) were compared by a two-way analysis of variance for repeated measures. When significant differences were identified, a post hoc analysis with Fisher's protected least significant difference test was performed. A P value <0.05 was considered as significant.
Of the 18 fetuses in which a fistula was performed, 12 were alive at the time of the elective cesarean section. Another two fetuses that delivered vaginally within 48 h of the scheduled cesarean date died within the first minutes of life with generalized cyanosis and were excluded from the study. Four fetuses from this group were dead at the time of cesarean section and were also excluded from the study. Postmortem examination did not reveal any sign of hydrops or infection, but the macroscopic examination of the heart was in favor of RV hypertrophy. The fistula was patent in all but one case of the 12 live fetuses from the fistula group. This fetus was not excluded because surgical exploration gave evidence that the anastomosis had been previously patent with major dilatation of the jugular vein. There were 11 fetuses in the control group, four cotwin or cotriplet fetuses, and seven fetuses from four unoperated ewes. All fetuses were delivered between 134 and 139 days.
The mean fetal weight in the fistula group was 2,485 g (range: 1,930-3,015 g). Fetal lambs from the control group showed a similar distribution of weight, ranging from 2,000 to 2,930 g. Lung weights were not different in the two groups (fistula group, 123 ± 8 g; control group, 123 ± 7 g). Hemodynamic data are shown in Table 1. Mean LPA pressure was 48 ± 2 mmHg in the fistula group and was significantly higher than in the control group (40 ± 2 mmHg, P < 0.01, Student's t-test). Although there was no statistical difference between the two groups, mean aortic and both atrial pressures showed a trend to be increased in the fistula group. The ratio of pressure between the LPA and the aorta (LPA/aorta) was significantly greater in the fistula group (P < 0.05). There was no difference for O2 or CO2 saturation in the LPA or the aorta, and the pH was normal in both groups (Table 2). LPA blood flow measurements were performed on six lambs in the fistula group and five lambs in the control group. There was no difference between the two groups for LPA blood flow (fistula group, 131 ± 18 ml/min; control group, 146 ± 14 ml/min). In the fistula group, the pulmonary vascular resistance was 0.36 mmHg · ml-1 · min-1, significantly higher than in controls (0.26 mmHg · ml-1 · min-1, P < 0.05). The ratio of the weight of the RV to the LV+S was 0.53 ± 0.3 in the fistula group, significantly higher than in the control group (0.41 ± 0.3, P < 0.01).
Macroscopic appearance of lungs in both groups was normal. Structural examination of pulmonary arteries showed a greater extension of muscle into smaller arteries in the fistula group compared with controls. There was an increase in the percentage of vessels that had a fully or partially muscular wall in arteries <55 μm in diameter in the fistula group (Table 3). The percentage wall thickness of arteries up to 120 μm in diameter was greater in the fistula group than in controls (Table 4), with a significant increase in each range of arteries studied (P < 0.01) (Fig. 1). The size of arteries accompanying alveolar ducts was similar in both groups with ED = 21.7 ± 0.6 and 22.1 ± 0.3 μm, respectively, in the fistula and control groups. There was no difference in the number of alveoli or arteries per field between the two groups, and the number of arteries per 100 alveoli was 10.2 ± 0.6 and 10 ± 0.8, respectively, in the fistula and control groups.
Measurements of the axial arteries in the two groups showed no difference in the ED of vessels measured (fistula: 1,200 ± 92 vs. control 946 ± 87 μm), the absolute wall thickness (135 ± 15 vs. 110 ± 16 μm), the %MT (22.5 ± 2.1 vs. 22.9 ± 1.6%), or the interelastic lamina distance (11.1 ± 0.8 vs. 10.1 ± 0.6 μm).
Isometric Tension Study
The vessel rings were of similar size (1-1.5 mm in diameter) in both groups. The maximum contractile response to 10-5 M phenylephrine was not different between the two groups (fistula group, 938 ± 160 mg; control group, 1,188 ± 282 mg).
Investigation of the endothelium-dependent vasodilatation. Vasorelaxation to ACh (fistula n = 5, control n = 5). Although ACh induced a modest relaxation in both groups, the maximal relaxation to ACh 10-4 was 30 ± 3% in the control group. In the fistula group, maximum relaxation was 11 ± 7% at 10-6 M, and higher concentrations caused no relaxation and even led to a contractile response of 3.9 ± 8% at 10-4 M (Fig. 2A).
Vasorelaxation to ADP (fistula n = 4, control n = 5). ADP induced a more pronounced relaxation in both groups than ACh. The relaxation in the control group was significantly higher for ADP 10-4 than in the fistula group (P < 0.05) (Fig. 2B).
Vasorelaxation to A-23187 (fistula n = 5, control n = 5). A-23187, an endothelium-dependent dilator that does not require membrane receptor activation, induced a relaxation that was significantly greater in the control group for A-23187 10-7 and 10-6 (P < 0.01) than in the fistula group, which did not show relaxation (Fig. 2C).
To investigate the endogenous activity of NO in both groups, we also studied contraction to phenylephrine and relaxation to ACh in the presence of l-NA, a specific inhibitor of NOS (fistula n = 5, control n = 5). Pretreatment with l-NA caused a modest and comparable contraction in both groups of animals (fistula group, 137 ± 43 mg; control group, 122 ± 51 mg). Although the presence of l-NA reduced significantly the relaxation to ACh 10-6 and 10-4 in control animals, it did not modify the response to ACh in the fistula group (Fig. 3).
Investigation of the Endothelium-Independent Vasodilatation
Vasorelaxation to SNP (fistula n = 6, control n = 5). SNP induced a dose-dependent relaxation in pulmonary rings from both groups, reaching complete relaxation in response to SNP 10-4. At lower doses, relaxation was significantly greater in the fistula group (P < 0.05) as illustrated by the lower EC50 in the fistula group (5.9 ± 0.7 10-8 M) compared with EC50 in the control group (4.6 ± 0.7 10-7 M, P < 0.01) (Fig. 4).
Light microscopy examination of the five vessel rings from the fistula group and of the six vessel rings from the control group showed the integrity of the endothelium (Fig. 5).
eNOS Western Blotting
Western blot analysis revealed eNOS protein expression in peripheral lung parenchyma homogenates of control lambs and lambs with fistula. The molecular mass of eNOS protein was identical to that of eNOS expressed in rat aorta (Fig. 6A). Analysis of red Ponceau's staining revealed no difference in the amount of loaded proteins between the different experiments (data not shown). Quantification of the intensity of eNOS bands showed no difference between the two groups of animals (Fig. 6B).
Our data suggest that a systemic arteriovenous fistula created at 120 days' gestation not only increased pulmonary arterial blood pressure and resistance but also led to structural pulmonary vascular abnormalities and altered pulmonary vascular reactivity. These changes in the fetal arterial pulmonary vascular bed might be consistent with the development of PPHN after birth. We believe that this new model of fetal pulmonary hypertension, associated with a moderate rate of miscarriage, may be relevant for studying mechanisms that regulate vascular tone in the perinatal period.
The fetal hemodynamic changes induced by the fistula differ from those described in the DA occlusion model where the increase in arterial pulmonary pressure was higher with only minor changes of systemic pressures (4, 22). In the present study, systemic pressures tended to be higher in the lambs who had a prenatal systemic arteriovenous fistula, although the tendency did not reach significance. This might be the consequence of a fistula-induced cardiac output increase.
At birth, LPA blood flow values were not different between the two groups, suggesting that the fistula model is not associated with chronically elevated pulmonary blood flow. Previous studies on the DA compression model have also shown that permanent DA compression was not associated with increased chronic pulmonary blood flow (1, 7). Partial compression of the DA increases both pulmonary arterial pressure and blood flow (1). As a result, pulmonary vascular resistance progressively falls during the initial 30 min of compression (1). When the compression is maintained, pulmonary blood flow rapidly returns to the baseline value, while pressure and resistance remain high, suggesting the presence of an arterial pulmonary myogenic response. Under physiological conditions there is a balance between blood flow-induced vasodilatation and pressure-induced vasoconstriction (33). When the endothelium-dependent vasodilatation is impaired, the myogenic response is unmasked as demonstrated by Storme et al. (33). Because we observed no difference in pulmonary blood flow at birth, it is likely that the pulmonary vasculature reacted to normalize blood flow, while arterial pulmonary pressure remained high, leading to structural arterial remodeling. Our findings of striking structural alterations of the pulmonary vessels with an extension of muscle into smaller arteries than normal and an increase in the medial thickness in small pulmonary arteries were similar to the pulmonary vascular structural abnormalities found in neonates who died with PPHN (12, 18, 23, 24). The 50% increase in the medial thickness of pulmonary arteries in this study was comparable with the one reported by Abman et al. (4) in the model of prenatal DA ligation. With a similar technique of lung fixation they also found that this was associated with an abnormal extension of muscle into smaller arteries.
On the basis of in vitro studies of conduit pulmonary arteries we found that vasodilator responses to both ACh and ADP were reduced (Fig. 2), suggesting altered endothelium-dependent relaxation in the fistula group compared with control animals. As vasodilator response to the calcium ionophore A-23187 was also impaired (Fig. 2), reduced endothelium-dependent relaxation in the fistula group is likely to be due to postreceptor alterations of signaling pathways underlying endothelium-dependent relaxation mechanisms. As all vessel ring studies were performed in the presence of indomethacin, excluding any role of cyclooxygenase metabolite, NO and nonprostanoid endothelium-derived hyperpolarizing factors are likely to account for ACh-induced relaxation in control pulmonary arteries (19). The role of NO is further supported by the blunted responses to ACh we observed in the control group when arterial rings were pretreated with the NOS inhibitor l-NA (Fig. 3). By contrast, in pulmonary arterial rings from the fistula group, l-NA had no effect on the dose-response curve to ACh (Fig. 3). This could reflect impaired stimulated NO release in this group. As eNOS protein expression in lung tissues of animals with fistula did not differ from that of control animals, it is likely that the observed impaired endothelium-dependent relaxation to ACh, ADP, and the calcium ionophore A-23187 is related to posttranslational mechanisms altering eNOS function and activity. Alternatively, impaired soluble guanylate cyclase (sGC) responsiveness to endothelium-derived NO (34) or changes in ion channel expression or function downstream of the NO-cGMP signaling pathways may also contribute to the observed impairment of endothelium-dependent relaxation in animals with fistula.
Unlike Belik et al. (6), who found decreased force development by vascular smooth muscle from fetal sheep following ductus ligation, vasocontractile response to phenylephrine was the same in pulmonary arterial rings from animals with fistula and controls in this study. We did not measure myosin and actin content in pulmonary vessels, and we are therefore unable to speculate on the effect of fistula on the expression and activity of contractile proteins in pulmonary vascular smooth muscle. As l-NA increased maximal contraction to phenylephrine in both groups, it is likely that basal NO release (unlike stimulated NO synthesis) is preserved in both groups. The apparent discrepancy observed with l-NA, which, on the one hand, enhanced contraction in response to phenylephrine but, on the other hand, failed to reduce relaxation to ACh, might result from selective impairment of stimulated (but not basal) NO synthesis (35). Both in vivo and in vitro studies of intrapulmonary arteries in the DA constriction model have revealed impairment of endothelium-dependent vasodilation (20, 32, 33). Additional studies in their model have further demonstrated decreased pulmonary eNOS gene expression with a decrease in eNOS protein and eNOS mRNA contents (30, 36). In another model of PPHN, namely aortopulmonary shunt placement in the late-gestational fetal lamb, endothelium-dependent vasodilation was impaired in lambs of 4 mo of age (27). However, in this model, which is associated with high pulmonary blood flow, eNOS has been found to be upregulated (8).
Although fetal pulmonary arterial rings from both groups relaxed completely in response to the NO donor SNP, the potency of SNP differed between vessels from fistula and control animals, as illustrated by the lower EC50 observed in the fistula group. We did not normalize the response to SNP to the arterial rings' weight. However, we have demonstrated that the medial thickness of the conduit pulmonary arterial rings that were studied did not differ between the two groups, confirming that this increased sensitivity to SNP was not related to a structural change. The increased response to SNP in pulmonary arterial rings from the fistula group in the present study could result from increased sensitivity of molecular targets that might not belong to the sGC pathway. Alternatively, reduced production of stimulated endothelium-derived NO in the fistula group may have accounted for the increased relaxation to SNP, as hypersensitivity of sGC occurs when NO is lacking (10, 21).
This study demonstrates that the creation of a systemic arteriovenous fistula leads to both structural and functional alteration of the pulmonary vasculature that could be consistent with the development of a PPHN syndrome after birth. We therefore conclude that the newly described model of systemic arteriovenous fistula may be relevant to the study of mechanisms that modulate pulmonary vascular tone in the perinatal period as shown by altered responses to both endothelium-dependent and -independent vasodilators in experimental animals.
This work was supported by a grant from Fondation de l'Avenir, Paris, France.
Alison A. Hislop is supported by the British Heart Foundation.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked ”advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2003 the American Physiological Society