INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Dexfenfluramine is the active enantiomer of a substituted phenethylamine, fenfluramine, and is used as a weight-reducing agent in obese patients (1). Its appetite-suppressant effect is thought to be dependent upon inhibition of the neuronal reuptake of 5-hydroxytryptamine (5-HT), which is accompanied by release of 5-HT from intraneuronal stores (2). Most of the 5-HT in the blood is contained in a reserve pool within the platelets. Blood platelets share with neurons a high-affinity uptake mechanism for 5-HT, which involves a carrier of similar structure in both cell types (3). Endothelial cells also remove 5-HT from plasma via the same mechanism (4). Moreover, inhibition of 5-HT uptake by fenfluramine is not confined to nerve cells but also occurs in platelets (5) and lung tissue (6). In animals chronically treated with dexfenfluramine, 5-HT concentrations have been shown to decrease in whole blood (reflecting platelet 5-HT depletion) and to increase in plasma (7).

Several clinical studies suggest an association between the development of primary pulmonary hypertension and the intake of dexfenfluramine. Brenot and colleagues reported that 15 of 73 retrospectively studied patients with primary pulmonary hypertension had previously taken either dexfenfluramine, the racemic compound fenfluramine, or both (8). More recently, an epidemiologic study provided further support for an increased risk of primary pulmonary hypertension in subjects taking appetite suppressants, mainly fenfluramine derivatives (9). The effect of dexfenfluramine on platelets somewhat resembles the genetic platelet storage pool disease observed in fawn hooded rats (10). This strain has elevated plasma 5-HT concentrations and develops pulmonary hypertension when exposed to Denver altitude (an environment with a mild reduction in oxygen tension), whereas control Sprague-Dawley rats exhibit normal pulmonary pressures at Denver altitude (11). Moreover, primary pulmonary hypertension has been reported in patients with increased plasma 5-HT levels and a similar platelet storage disease (12). The possibility that 5-HT may be linked to the development of pulmonary hypertension is further suggested by a recent study in which we observed that a sustained increase in plasma 5-HT levels achieved via a continuous infusion of 5-HT aggravated the development of pulmonary hypertension in response to a 2-wk exposure to hypoxia (13).

In the present study, we sought to determine whether dexfenfluramine worsened the progression of pulmonary hypertension in rats exposed to various levels of chronic hypoxia. To this end, we first investigated in isolated lungs from normoxic rats the direct vasoactive effects of dexfenfluramine and the effect of acute in vitro dexfenfluramine pretreatment on the response to 5-HT. We then used conscious normoxic rats to evaluate the effect of dexfenfluramine on the acute pulmonary pressor response to hypoxia. Finally, we studied the effects on hemodynamics and structural pulmonary vasculature changes of continuous dexfenfluramine treatment during chronic exposure to moderate or severe hypoxia. We also investigated the effect of chronic dexfenfluramine intake on the development of hypoxic pulmonary hypertension in rats receiving a continuous infusion of 5-HT.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pharmacological Studies in Isolated Lungs from Normoxic Rats

Rats anesthetized with sodium pentobarbital (40 mg intraperitoneally) were subjected to tracheal cannulation and ventilated with a warmed normoxic gas mixture (95% air and 5% CO₂) at 60 breaths/ min with an inspiratory pressure of 9 cm H₂O and an expiratory pressure of 2.5 cm H₂O. A median sternotomy was performed, and 100 IU heparin was administered through the right ventricle. The heart and lungs were removed and suspended in a humidified chamber at 37° C. The lungs were perfused through a pulmonary artery cannula using a peristaltic pump at a constant flow of 0.05 ml/(g body weight)/min. The recirculated perfusate was a warmed (38° C) physiological salt solution with the following composition (millimolar): 116 NaCl, 4.7 KCl, 19 NaHCO₃, 0.83 MgSO₄, 1.8 CaCl₂, 2 H₂O, 1.04 NaH₂PO₄, 5.5 glucose, and phenol red Na (0.11 g/L) and ficoll (4 g/100 ml, type 70; Sigma Chemical Co., St. Louis, MO). Meclofenamate (3.2 µM) was included in the perfusate at the start of experiments. The effluent perfusate was drained from the left ventricular cannula into a reservoir. Mean perfusion pressure was measured from a side port of the pulmonary artery line (P23 XL transducer; Gould Electronics, Ballainvilliers, France), and the pulmonary venous pressure was assumed to be zero. Each lung preparation was used for only one of the following studies.

Effect of dexfenfluramine on lungs preconstricted with U-46619. The endoperoxide analog U-46619 was diluted in a 20-ml volume of physiological salt solution and infused into the pulmonary artery line at a constant rate of 50 pmol/min, using an infusion pump (Vial-Medical, Grenoble, France), and starting after a 30-min equilibration period. Pulmonary arterial pressure (Ppa) increased steadily in response to U-46619, without reaching a plateau. After at least 10 min of U-46619 infusion, when the Ppa increase reached 8 to 9 mm Hg, dexfenfluramine was added to the perfusate reservoir (n = 4). Solutions containing increasing concentrations of dexfenfluramine were injected as 50-µl boluses at 3-min intervals to obtain final concentrations in the recirculating perfusate that increased from 10⁹ to 10⁵ M. These concentrations were selected based on pharmacokinetic studies in humans (14).

Effect of acute dexfenfluramine pretreatment on vasodilator responses to 5-HT in lungs preconstricted with U-46619. After at least 10 min of U-46619 infusion, and when the Ppa increase reached 8-9 mm Hg, 5-HT was added to the perfusate reservoir. Solutions containing increasing concentrations of 5-HT were injected as 50-µl boluses at 3-min intervals to obtain final concentrations in the recirculating perfusate that increased from 10¹³ to 10⁸ M. The vasodilator response to 5-HT was examined in lungs pretreated with dexfenfluramine (10⁵ M) or its vehicle, which was added to the perfusate reservoir at the end of the equilibration period (n = 5 for each).

Effect of dexfenfluramine on baseline tone of lungs. After 30 min of equilibration, dexfenfluramine was administered at 10-min intervals in the perfusate reservoir as 50-µl boluses to obtain final concentrations in the recirculating perfusate that increased from 10⁹ to 10⁴ M (n = 5).

Effect of acute pretreatment with dexfenfluramine on vasoconstrictor response of lungs to 5-HT. In another series of experiments, the effect of 5-HT on baseline tone was examined. After 30 min of equilibration, 5-HT was administered at 10-min intervals into the perfusate reservoir as 50-µl boluses to obtain final concentrations in the recirculating perfusate that increased from 10⁷ to 10⁴ M. The pressor response to 5-HT was examined in lungs pretreated with dexfenfluramine (10⁵ M, n = 4) or its vehicle (n = 5), added into the perfusate reservoir 10 min before the dose-response curves to 5-HT were started.

Effect of Pretreatment with Dexfenfluramine on the Hemodynamic Response to Acute Hypoxia

On the day before the hemodynamic study, normoxic rats were anesthetized with an intramuscular injection of ketamine (20 mg) and xylazine (1 mg). After exposure of the right jugular vein, a polyvinyl catheter was inserted and manipulated through the right ventricle into the pulmonary artery. A second polyvinyl catheter was also inserted into the right jugular vein to allow injection of indocyanine green dye and dexfenfluramine. A polyethylene catheter was placed in the right carotid artery, and a second polyethylene catheter was placed in the left jugular vein for returning blood to the rat during cardiac output measurements. The catheters were sealed and tunneled under the skin to the back of the neck, where they were exteriorized, secured, and protected in a small plastic container.

Measurements were taken on the day after surgery, while the rat was awake. Pulmonary and systemic arterial pressures were measured using Gould P 23 ID transducers (Gould Electronics) coupled to pressure modules and a Gould TA 550 multichannel recorder. Cardiac output (CO) was measured using a dye dilution technique, as follows: A 50-µg bolus of indocyanine (1 mg/ml) was injected into the jugular vein. Blood (1 ml/min) was drawn from the carotid artery through a densitometer cuvette (Waters Instruments, Rochester, MN) and returned to the rat through the left jugular vein. Cardiac output was calculated from the dye dilution curve area after exponential extrapolation of the downslope. Calibration was performed at the end of each study, using a known concentration of green dye (5 µg/ml rat blood).

On the day of the experiments, animals were placed in a clear plastic box flushed with compressed air or a hypoxic gas mixture (10% O₂, 90% N₂). Hemodynamic variables were recorded first while the animal was breathing air, then after 10 min of exposure to hypoxia, and finally 20 min after returning to normoxia. Five similar successive hypoxic challenges were repeated in each rat. In preliminary experiments, the pulmonary arterial pressor response to hypoxia increased across the first three challenges, then reached a plateau. Therefore, dexfenfluramine (n = 4) or its vehicle (n = 5) was administered as an intravenous bolus in increasing doses (10, 100, and 500 µg) 5 min before the beginning of each of the last three hypoxic challenges.

Chronic Hypoxia

Male Wistar rats weighing 250-300 g at the start of the experiments were exposed to chronic hypoxia (15 or 10% O₂) in a ventilated chamber (500-L volume; Flufrance, Cachan, France), as previously described (15). To create the hypoxic environment, the chamber was flushed with a mixture of room air and nitrogen, and the gas was recirculated. The chamber environment was monitored using an oxygen analyzer (Servomex OA150; Crowborough, UK). Carbon dioxide was removed using soda lime granules, and excess humidity prevented by cooling the recirculation circuit. The chamber temperature was kept at 22-24° C. The chamber was opened every other day for 1 h to clean the cages and to replenish food and water stores. Normoxic rats were kept in the same room, with the same light-dark cycle. Rat food and tap water were provided ad libitum.

After 15 d of hypoxia, the rats were anesthetized with ketamine (20 mg) and xylazine (1 mg). Immediately after insertion of the catheters, Ppa and systemic arterial pressure (Psa) were measured. Blood was then sampled for hematocrit determination and 5-HT measurements. Finally, after an intraperitoneal injection of sodium pentobarbital (20 mg/kg), the thorax was opened and the heart was quickly removed, dissected, and weighed. The ratio of right ventricular free wall weight to the sum of septum plus left ventricular free wall weight (fresh tissue) was used as an index of right ventricular hypertrophy.

The lungs were fixed in the distended state by infusing 4% aqueous buffered formalin into the trachea at a pressure of 25 cm H₂O. The entire specimen was placed in a bath of the same fixative for 1 wk. A midsagittal slice of the right lung including the apical, azygous, and diaphragmatic lobes was processed for paraffin embedding. Sections 5 µm thick were cut for light microscopy and stained with hematoxylin phloxin saffron and orcein-picroindigo-carmine. In each rat, a total of 35-65 intra-acinar vessels accompanying either alveolar ducts or alveoli were examined. Their type was identified as muscular, partially muscular, or nonmuscular. Muscular arteries had a complete layer of smooth muscle cells bound by two orcein-stained elastic lamina. Smooth muscle cells were identified as elongated cells that were stained red by phloxin and had square-ended nuclei. They were seen in only part of the arterial circumference of partially muscular arteries and were absent from nonmuscular arteries.

Effect of chronic treatment with dexfenfluramine during exposure to chronic hypoxia. The effects of chronic treatment with dexfenfluramine from Days 1 to 15 of continuous exposure to 10 or 15% O₂ were investigated. The rats were randomly divided into four groups: one group exposed to 10% O₂ and treated with dexfenfluramine (n = 10), one exposed to 10% O₂ and treated with the vehicle (n = 10), one exposed to 15% O₂ and treated with dexfenfluramine (n = 6), and one exposed to 15% O₂ and treated with the vehicle (n = 8). Dexfenfluramine (2 mg/[kg body weight]/d) or its vehicle was administered by gastric gavage once a day, beginning on Day 1 of the hypoxic period. This dose of dexfenfluramine was selected based on studies reporting that it inhibited food intake in various animal models (16).

Effect of concomitant treatment with dexfenfluramine and 5-HT during chronic hypoxia. In a previous study, we observed that continuous 5-HT infusion during exposure to 10% O₂ aggravated the development of pulmonary hypertension. To examine the effect of chronic treatment with dexfenfluramine on pulmonary hypertension potentiation by 5-HT, we studied four additional groups of rats during a 2-wk exposure to 10% O₂: one group received 5-HT as a continuous infusion and dexfenfluramine (2 mg/kg/d) once a day by gavage (n = 10), one received 5-HT as a continuous infusion and the vehicle by gavage (n = 8), one received dexfenfluramine alone by gavage (n = 6), and one received the vehicle alone by gavage (n = 8). Miniosmotic infusion pumps (Alzet model 2002; Alza Scientific Products, Palo Alto, CA) containing 5-HT were implanted into the left jugular vein on the day before exposure to hypoxia. The pumps delivered a volume of 0.5 µL/h, which was equivalent to 5 nmol/h of 5-HT.

Blood and Plasma 5-Hydroxytryptamine Concentrations

5-Hydroxytryptamine was measured in whole blood and platelet-poor plasma after ethanol-acetone extraction by means of a specific radioenzymatic method. Samples were analyzed using ¹⁴C[5-HT]binoxalate as the internal standard (17).

Drugs

5-Hydroxytryptamine (Sigma) was dissolved in saline. The endoperoxide analog U-46619 (Sigma) was dissolved in ethanol, stored as a stock solution at 30° C, and diluted in saline as required. In the isolated lung experiments, reported concentrations are the final concentrations in the recirculating perfusate. Dexfenfluramine hydrochloride (batch number 45788) obtained from the Institut de Recherches Internationales Servier was dissolved in distilled water for acute administration in the isolated lung experiments, and suspended in 20% arabic gum for the chronic gavage experiments.

Statistical Analysis

All results are expressed as means ± SEM.

To compare the effect of pretreatment with dexfenfluramine or its vehicle on pressure changes caused by 5-HT in isolated rat lungs, two-way analysis of variance (ANOVA) with repeated measurements was performed, testing for a pretreatment effect, the vasopressor agent dose, and interaction. When the interaction was significant, Mann-Whitney's nonparametric test was used to compare dexfenfluramine with its vehicle at each dose.

One-way repeated measures ANOVA was used to evaluate the hemodynamic effect of dexfenfluramine during acute hypoxic challenge. When the ANOVA indicated significant differences between dexfenfluramine and its vehicle, multiple comparisons were performed using the Fisher test.

One-way ANOVA was performed to compare hemodynamic values in rats chronically treated with dexfenfluramine or its vehicle during continuous exposure to 10 or 15% oxygen. When the ANOVA indicated significant differences between the groups, these were compared using Scheffe's method. Between-group comparisons of ratios of right ventricle weight to left ventricle plus septum weight and of hematocrits were performed using similar statistical methods after arcsine transformation of individual values.

To compare the degree of muscularization of pulmonary vessels among the various groups of animals, pulmonary vessels were ordinally classified as nonmuscular, partially muscular, and muscular. Comparison of muscularization was performed separately at the alveolar duct and wall levels using a nonparametric Kruskal-Wallis test. When a significant difference was observed, multiple pairwise comparisons were performed using Scheffe's method.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experiments on Isolated Lungs from Normoxic Rats

In lungs preconstricted by infusion of U-46619 (50 pmol/min), dexfenfluramine (10⁹ to 10⁵ M) did not elicit any vasodilation or vasoconstriction (data not shown). By contrast, 5-HT (10¹³ to 10⁸ M) produced concentration-dependent vasodilation (Figure 1), an effect that remained unchanged after acute pretreatment with dexfenfluramine (10⁵ M). Similarly, pretreatment with dexfenfluramine did not affect the vasodilator response to ionophore (10⁹ to 10⁷ M, data not shown).

View larger version (12K): [in this window] [in a new window]   Figure 1.   Vasodilator (left panel ) and vasoconstrictor (right panel ) responses to 5-HT in lungs from normoxic rats in the presence of dexfenfluramine. Dexfenfluramine (105 M) or its vehicle was added to the perfusate reservoir at the end of the incubation period. Values are means ± SEM of n = 5 for each curve, except the vasoconstrictor response curve to 5-HT in the presence of dexfenfluramine (n = 4). There were no significant differences between the dexfenfluramine and vehicle curves.

Under conditions of baseline tone, dexfenfluramine up to a final concentration of 10⁴ M did not change baseline Ppa values (data not shown). A dose-dependent increase in Ppa was seen with 5-HT (10⁷ to 10⁴ M) and was unaffected by acute pretreatment with (10⁵ M) dexfenfluramine. Similarly, pretreatment with dexfenfluramine did not affect the vasoconstrictor response to endothelin-1 (10¹⁰ to 10⁹ mol, data not shown).

Effect of Dexfenfluramine on the Hemodynamic Response to Acute Hypoxia

In rats subjected to 10% O₂ challenges, hypoxia was associated with an increase in Ppa (Figure 2). However, both Psa and CO remained unchanged. The pulmonary arterial pressor response to hypoxia increased across the first three hypoxic challenges, then reached a plateau (Ppa was 3.5 and 3.7 mm Hg during the fourth and fifth hypoxic challenges, respectively). Dexfenfluramine given in increasing doses (10, 100, and 500 µg intravenously as a 200-µl bolus) during normoxic periods did not change the hemodynamic variables. However, dexfenfluramine produced dose-dependent potentiation of the pulmonary arterial pressor response to hypoxia. As compared with that obtained after vehicle pretreatment, the pulmonary arterial pressor response during the fifth hypoxic challenge was increased twofold after administration of a 500-µg bolus of dexfenfluramine (p < 0.05).

View larger version (11K): [in this window] [in a new window]   Figure 2.   Influence of dexfenfluramine on the pulmonary artery pressure increase in response to acute hypoxia (10% O2) in conscious rats. Rats were exposed to five successive hypoxic challenges (H1 through H5) separated by normoxic periods (N1 through N5). Dexfenfluramine (n = 4) or its vehicle (n = 5) was administered as indicated by the arrows, 5 min before the hypoxic period. *p < 0.05 for the comparison with the pressor response to hypoxia after vehicle pretreatment.

Effect of Continuous Dexfenfluramine on the Development of Hypoxic Pulmonary Hypertension

Compared with the rats that remained in normoxia, rats exposed to 15% O₂ for 15 d had an increased ratio of right ventricle weight to left ventricle plus septum weight (p < 0.01, Figure 3), whereas Ppa and the degree of distal pulmonary artery muscularization (Table 1) remained similar in the two groups. In the rats exposed to 15% O₂, all these parameters were similar between dexfenfluramine- and vehicle-treated rats, as was the hematocrit (49 ± 0.4 and 50 ± 0.5%, respectively).

View larger version (26K): [in this window] [in a new window]   Figure 3.   Pulmonary artery pressure (left panel ) and ratio of right ventricle/left ventricle + septum weight [RV/(LV + S)] (right panel ) in rats exposed to hypoxia (15% or 10% O2) for 2 wk and chronically treated with dexfenfluramine or its vehicle. Values in normoxic rats are also given. Results are expressed as means ± SEM. At neither level of hypoxia was any statistically significant difference found between dexfenfluramine and its vehicle.

Compared with 15% O₂ (Table 1), severe hypoxia (10% O₂) caused an increase in the hematocrit, Ppa (p < 0.01), and ratio of right ventricle weight to left ventricle plus septum weight (p < 0.05), as well as in the degree of pulmonary artery muscularization at both the alveolar duct and alveolar wall levels (p < 0.001). However, Ppa, the ratio of right ventricle weight to left ventricle plus septum weight, and the degree of distal pulmonary artery muscularization were similar in dexfenfluramine- and vehicle-treated rats. After exposure to 10% O₂, the hematocrit was also similar in dexfenfluramine- and vehicle-treated rats (55 ± 0.1 and 55 ± 0.7%, respectively).

Effect of Continuous Dexfenfluramine on Blood and Plasma 5-HT Concentrations

As previously observed (13), the level of hypoxia did not affect blood or plasma 5-HT concentrations, which remained unchanged in rats treated with the vehicle during exposure to 15% or 10% O₂ (Table 2). Chronic dexfenfluramine caused a decrease in blood and an increase in plasma 5-HT levels (p < 0.001 compared with vehicle treatment); these changes were similar when rats were exposed to 15% or 10% O₂.

Effect of Concomitant Dexfenfluramine and 5-HT During Chronic Hypoxia on Development of Hypoxic Pulmonary Hypertension

As in our earlier study, continuous 5-HT infusion during a 2-wk exposure to 10% O₂ was associated with greater pulmonary vascular remodeling. The degree of distal pulmonary artery muscularization was greater at both the alveolar duct and alveolar wall levels in rats chronically treated with 5-HT by infusion and with the vehicle by gavage than in the other groups exposed to a similar level of hypoxia (p < 0.01, Table 3 and Figure 4). Concomitant treatment with dexfenfluramine during the 5-HT infusion prevented the increase in distal pulmonary artery muscularization caused by the 5-HT infusion. Although there was also a trend toward a higher ratio of right ventricle weight to left ventricle plus septum weight in the groups treated with 5-HT by infusion and the vehicle by gavage when compared with others exposed to a similar level of hypoxia, this difference did not reach statistical significance (Figure 4). Neither Ppa nor hematocrit differed among the various groups exposed to 10% O₂.

View larger version (38K): [in this window] [in a new window]   Figure 4.   Ratio of right ventricle/left ventricle + septum weight [RV/(LV + S)] (left panel ) and percentage of muscular and partially muscular arteries at the alveolar duct level (right panel ) in rats exposed to 10% O2 for 2 wk and chronically treated with dexfenfluramine or its vehicle during a continuous 5-HT infusion. Values in the other two groups treated only by dexfenfluramine or its vehicle via gavage are also given. Results are expressed as means ± SEM. *p < 0.01 for the comparison between dexfenfluramine and vehicle gavage during 5-HT infusion.

Effect of Concomitant Dexfenfluramine and 5-HT During Chronic Hypoxia on Blood and Plasma 5-HT Concentrations

Chronic 5-HT infusion during continuous exposure to 10% O₂ caused increases in both blood and plasma 5-HT concentrations. During 5-HT infusion, blood 5-HT levels did not differ in rats chronically treated with dexfenfluramine or its vehicle (5.4 ± 1.2 versus 3.28 ± 0.30 µm, respectively), whereas plasma 5-HT concentrations were higher in dexfenfluramine- than in vehicle-treated rats (28.08 ± 8.99 versus 9.39 ± 2.2 nm, respectively; p < 0.001).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The chronic dexfenfluramine regimen used in this study during a 2-wk exposure to mild or severe hypoxia did not affect the development of pulmonary hypertension in rats. This result contrasts with the finding that dexfenfluramine potentiated acute hypoxic vasoconstriction in vivo when it was administered for a short time. Rats treated with dexfenfluramine demonstrated the same degree of pulmonary hypertension, right ventricular hypertrophy, and structural distal pulmonary artery remodeling as vehicle-treated rats exposed to a similar level of hypoxia. Moreover, 5-HT's effect on pulmonary vascular remodeling in response to chronic exposure to 10% O₂ was prevented by concomitant dexfenfluramine treatment (2 mg/kg/d).

A contributing role of increased 5-HT plasma levels has been suggested in various forms of pulmonary hypertension. We recently observed that 5-HT potentiated in vivo acute pressor responses to hypoxia; it also promoted the development of pulmonary hypertension when infused continuously during a 2-wk exposure to 10% O₂ (13). Rats treated with 5-HT demonstrated a greater degree of pulmonary hypertension, right ventricular hypertrophy, and structural distal pulmonary artery remodeling than saline-treated rats exposed to a similar level of hypoxia. Fawn hooded rats, a strain of rats with serotonin platelet storage disease and increased plasma 5-HT levels, developed pulmonary hypertension when they were raised at Denver altitude (an environment with a mild reduction in oxygen tension), whereas control Sprague-Dawley rats exhibited normal pulmonary artery pressure at Denver altitude (11). Moreover, exposure of fawn hooded rats to a moderately O₂-enriched environment prevented the development of pulmonary hypertension at Denver altitude. We therefore hypothesized that dexfenfluramine, which impairs 5-HT uptake by platelets, may worsen chronic hypoxic pulmonary hypertension and may also increase the susceptibility of the pulmonary circulation to hypoxia. To investigate this hypothesis, we examined the vasoactive effects of dexfenfluramine on isolated lungs from normoxic rats, as well as the effects of dexfenfluramine pretreatment on in vivo pulmonary vascular responses of rats to acute and chronic hypoxia.

In isolated lungs from normoxic rats, dexfenfluramine up to a concentration of 10⁴ M had no vasoactive effects. It did not alter baseline tone, and neither did it produce any vasodilator effects when the lungs were preconstricted with the endoperoxide analog U-46619. These results do not differ noticeably from those of previous studies. Only suprapharmacologic concentrations of dexfenfluramine have been reported to increase the tone of isolated lungs or pulmonary arteries from humans or animals (18, 19). In our study, acute pretreatment of isolated lungs from normoxic rats with dexfenfluramine similarly failed to alter vasodilation in response to 5-HT or to modify the vasoconstrictive effect of 5-HT observed with higher concentrations under conditions of baseline tone. We previously demonstrated that 5-HT elicited vasodilation in preconstricted lungs from normoxic rats, an effect mediated through nitric oxide release because it was completely abolished after pretreatment with L-arginine analogs (13). Taken together, these results suggest that acute pretreatment with dexfenfluramine did not interfere with activation of the 5-HT receptors involved in either vasoconstriction or endothelium-dependent vasodilation.

In conscious normoxic rats, administration of dexfenfluramine in doses that did not affect baseline pulmonary arterial pressure potentiated the acute pressor response to 10% O₂. These results are consistent with the reported effect of dexfenfluramine on the acute hypoxic pressor response in dogs (20). In this latter study, dexfenfluramine given intravenously restored pulmonary vasoconstriction in animals with a small or absent pressor response to hypoxia, whereas it did not affect pulmonary vascular tone during hyperoxia or hypoxia in animals with a normal hypoxic response. Patch-clamp experiments have recently demonstrated that dexfenfluramine, as well as other anorexiant drugs (fenfluramine and aminorex), shared with hypoxia the ability to produce dose-dependent inhibition of the whole-cell potassium current in rat pulmonary vessel smooth muscle cells (19). These drugs may potentiate hypoxic pulmonary vasoconstriction by facilitating the depolarizing effect of hypoxia on smooth muscle and thereby increasing the entry of Ca²⁺ through voltage-gated Ca²⁺ channels. The potentiation of the in vivo acute hypoxic pressor response associated with acute dexfenfluramine treatment may also be mediated by 5-HT release. We recently observed that 5-HT in doses that did not affect baseline pulmonary arterial pressures potentiated the acute pressor response to 10% O₂. This effect persisted despite pretreatment with 5-HT₁ and 5-HT₂ receptor antagonists, in keeping with findings from a study by Naeije and colleagues of dexfenfluramine in dogs (20). Therefore, a reasonable hypothesis is that potentiation of in vivo hypoxic pulmonary vasoconstriction resulted from an indirect mechanism rather than from a direct action of 5-HT on pulmonary vessels.

Despite its potentiating effect on acute hypoxic pulmonary vasoconstriction, dexfenfluramine given chronically during continuous exposure to hypoxia failed to increase right ventricular hypertrophy or vascular remodeling. These results are in accordance with those of Laude and colleagues showing that chronic fenfluramine in rats did not potentiate the development of pulmonary hypertension (21). Thus, chronic dexfenfluramine did not replicate the pulmonary vascular abnormalities observed in fawn hooded rats, which develop severe pulmonary hypertension when exposed to a modest degree of hypoxia. This was true despite evidence of effective blockade of 5-HT uptake by platelets. As previously observed in normoxic rats (7), continuous dexfenfluramine treatment was associated in our study with a significant decrease in whole blood and an increase in plasma 5-HT concentrations in rats chronically exposed to either 10 or 15% O₂. Our results with dexfenfluramine also contrast with the worsening effect of continuous 5-HT infusion on the development of pulmonary hypertension during exposure to severe hypoxia (13). Interestingly, not only did chronic dexfenfluramine treatment fail to increase pulmonary vascular remodeling in response to a 2-wk exposure to 10% O₂, it also prevented the potentiating effect of 5-HT as a continuous infusion. After a 2-wk exposure to 10% O₂, pulmonary vessel muscularization at both the alveolar duct and alveolar wall levels was less marked in rats given both 5-HT and dexfenfluramine than in those given only 5-HT.

Previous studies have shown that 5-HT has comitogenic properties on bovine and rat pulmonary vascular smooth muscle cells in culture. A synergistic mitogenic response was demonstrated when smooth muscle cells were incubated with 5-HT in the presence of various growth factors such as platelet-derived growth factor (PDGF) and epidermal growth factor (EGF) (22, 23). This effect may require either binding of 5-HT to cell membrane receptors or active transport of 5-HT into the cell. The mitogenic effect of 5-HT on rat pulmonary vascular smooth muscles was attenuated by the 5-HT₂ receptor antagonist ketanserin (22). In contrast, the stimulating effect of 5-HT on [³H]thymidine incorporation by bovine pulmonary artery smooth muscle cells was abolished after inhibition of 5-HT transport into the cell, and the effect remained unaltered in the presence of 5-HT receptor antagonists (24, 25). Chronic dexfenfluramine treatment was associated with decreased levels of 5-HT and its metabolite 5-hydroxyindoleacetic acid in lung tissue (7), suggesting that dexfenfluramine may affect the cellular uptake of 5-HT by pulmonary vessels. A hypothesis that fits these data is that the protection afforded by dexfenfluramine against 5-HT-induced potentiation of pulmonary vascular remodeling may be related to inhibition of 5-HT transport into smooth muscle cells from pulmonary vessels. However, we found substantial evidence that a significant amount of 5-HT was stored in platelets during 5-HT infusion despite dexfenfluramine treatment: 5-HT infusion was associated with marked increases in both blood and plasma 5-HT levels, and concomitant dexfenfluramine treatment caused a further increase in plasma 5-HT levels but did not reduce blood 5-HT levels. Dexfenfluramine has to bind to the 5-HT carrier to exert its inhibitory effect on 5-HT transport (26). Conceivably, the sustained increase in plasma 5-HT levels produced by the infusion may have resulted in competitive displacement of dexfenfluramine. Alternatively, the high plasma levels of 5-HT may have caused 5-HT to be stored in platelets via a transport mechanism not affected by dexfenfluramine. If such a mechanism exists, dexfenfluramine may significantly reduce 5-HT uptake by pulmonary smooth muscle cells in the absence of evidence of platelet 5-HT depletion. Similarly, the protective effect of dexfenfluramine against 5-HT-induced potentiation of hypoxic pulmonary vascular remodeling may not be related to inhibition of 5-HT transport into smooth muscle cells but may rather involve another mechanism, such as an antagonistic effect of dexfenfluramine on 5-HT receptors. However, the finding in our study that pretreatment with dexfenfluramine did not alter the vasoactive effects of 5-HT in isolated lungs militates against this hypothesis.

Our finding that continuous dexfenfluramine treatment did not potentiate the development of hypoxic pulmonary hypertension in rats does not rule out a causal relationship between dexfenfluramine intake and primary pulmonary hypertension in humans. A longer period of treatment may be needed to observe such an effect in the animals. Because humans metabolize dexfenfluramine at a lower rate than rats (27), sensitivity to dexfenfluramine may differ between these two species. Sensitivity to dexfenfluramine may also differ across individuals and depend on susceptibility factors.