Sildenafil Modulates Hemodynamics and Pulmonary Gas Exchange AXEL KLEINSASSER, ALEXANDER LOECKINGER, CHRISTOPH HOERMANN, FRIEDRICH PUEHRINGER, NORBERT MUTZ, GEORG BARTSCH, and KARL H. LINDNER Department of Anesthesiology and Critical Care Medicine and Department of Urology, Leopold-Franzens-University of Innsbruck, Innsbruck, Austria

The effects of sildenafil (Viagra) on hemodynamics and pulmonary gas exchange remain uncertain. The aim of this study was to investigate what effect sildenafil had on gas exchange. A total of 24 anesthetized pigs were randomly assigned into four groups of six animals each: Group Low received 25 mg of sildenafil, which is equivalent to half the recommended dose for humans; group Normal received 50 mg; group High received 100 mg; and one group served as control. Inert gas and hemodynamic measurements were performed to define dose-dependent effects of sildenafil on cardiac and pulmonary function. Measurements were taken 30, 60, and 90 min after the administration of sildenafil via gastric tube. All doses of sildenafil caused significant increases in intrapulmonary shunt flow (maximum amplitude, 4.4 ⫾ 0.3 to 11.9 ⫾ 0.5%; mean ⫾ SEM), which was reflected by marked decreases in PaO2. Sildenafil elicited some significant increases in cardiac index (CI) (high dose, 142 ⫾ 10 to 196 ⫾ 13 ml kg⫺1, mean ⫾ SEM). Mean arterial pressure was significantly depressed after the high dose of sildenafil. Pulmonary artery pressure was decreased after high-dose sildenafil (maximum amplitude, 16 ⫾ 1.6 to 14 ⫾ 1.8, mean ⫾ SEM). No significant differences between the three treatment groups were found. Sildenafil represents and orally active substance with phosphodiesterase V inhibitory and cardiac output-increasing actions. PaO2 decrease after 50 and 100 mg of sildenafil was observed in the presence of significant rises in pulmonary shunt flow and CI.

After the introduction of the oral phosphodiesterase (PDE) V inhibitor sildenafil (Viagra), this drug has rapidly found widespread use. In the United States, sildenafil has already been prescribed more than 6 million times (representing 50 million tablets), and 130 deaths of patients who had been prescribed Viagra were reported to the U.S. Food and Drug Administration by November 1998 (1). Sildenafil inhibits PDE V, the enzyme that specifically hydrolyzes guanosine 3⬘,5⬘-cyclic monophosphate (cGMP). By a complex mechanism, increased intracellular cGMP leads to the hyperpolarization of smooth-muscle membranes and subsequent vascular relaxation. Moreover, cGMP is involved in the regulation of myocardial L-type Ca2⫹ channel current (ICa): In guinea pig, frog, and human cardiomyocytes, cGMP can also stimulate ICa via inhibition of cGMP-inhibited cyclic adenosine monophosphate (cAMP) PDE (PDE III) (2), thus augmenting cardiac output. PDE isozyme types I, III, IV, and V are present in the human pulmonary artery (3). Hanson and coworkers have reported that cGMP exercises considerable influence on pulmonary vascular resistance (RLva), and in that there seems to be a direct correlation between the activity of PDE V and RLva (4). Pulmonary vasodilatation induces intrapulmonary right-to-left shunt flow,

(Received in original form March 7, 2000 and in revised form July 14, 2000) Correspondence and requests for reprints should be addressed to Axel Kleinsasser, M.D., Department of Anesthesiology and Critical Care Medicine, LeopoldFranzens-University of Innsbruck, Anichstrasse 35, 6020 Innsbruck, Austria. E-mail: [email protected] Am J Respir Crit Care Med Vol 163. pp 339–343, 2001 Internet address: www.atsjournals.org

which in turn leads to a decrease in arterial partial pressure of oxygen (PaO2) (5, 6). Sildenafil is a selective PDE V inhibitor (7), and therefore pulmonary vascular dilatation and a subsequent increase in intrapulmonary shunt flow are likely to occur. The extent of sildenafil-associated pulmonary vasodilatation and the influence of this substance on hemodynamics remain unclear at present. We examined the effects of sildenafil on hemodynamics and pulmonary gas exchange, using the multiple inert gas elimination technique (MIGET) in a porcine model. Our hypothesis was that the administration of sildenafil would result in a dose-dependent fall in RLva, a marked increase in ventilation– · · perfusion ( VA/Q) heterogeneity with a subsequent deterioration of pulmonary gas exchange.

METHODS After approval by the Austrian Federal Animal Investigational Committee, a total of 24 domestic pigs of either sex, 12 to 16 wk old and weighing 38.8 ⫾ 2 kg (mean ⫾ standard deviation) were studied. Animals were managed according to institutional guidelines. Anesthesia was induced intramuscularly with ketamine (15 mg/kg) and maintained by continuous infusion of propofol (10 mg/ml) at a rate of 1 ml/ kg per hour. High-dose piritramide (0.8 mg/kg) was administered to achieve an adequate depth of anesthesia. This anesthetic regimen prevented adrenergic activation and non-steady-state anesthesia. Subsequently, all animals were placed in the supine position. A collid plasma expander (3% gelatin solution, 4 ml/kg per hour) was administered continuously throughout the study period. The tracheas of the animals were intubated with endotracheal tubes (7.0-mm i.d.). Lungs were ventilated in time-cycled volume-controlled mode (Servo 900 D; Siemens, Elema, Sweden) at a respiratory rate of 15 breaths/min at fraction of inspired oxygen FIO2 ⫽ 0.21. A positive end-expiratory pressure (PEEP) of 5 cm H2O was applied in all animals. The minute volume was adjusted to maintain an arterial PCO2 between 30 and 35 mm Hg. Resulting pressures were 24 ⫾ 1.2 cm H2O peak inspiratory pressure and 10 ⫾ 0.43 cm H2O mean inspiratory pressure, respectively. Two inspiratory holds of 20 s each were performed before all measurements to prevent nonspecific atelectasis. Body temperature was maintained between 38 and 39⬚ C by the use of an electric heating blanket.

Protocol The animals were randomly assigned into four groups: Group Low received 25 mg of sildenafil, Group Normal received 50 mg, and Group High received 100 mg. A control group of six animals was anesthetized according to the identical regimen as for the treatment groups but received no sildenafil. Measurements were carried out at four points in time. After a period of at least 60 min stabilization to establish steady state conditions, baseline measurements were taken. Sildenafil was then administered via gastric tube and further measurements were taken 30, 60, and 90 min later. Each set of measurements included heart rate; systolic, diastolic, and mean arterial blood pressures; and central venous pressure, mean pulmonary arterial pressure (Ppa), and pulmonary capillary wedge pressure (Ppc,we). Respiratory measurements were performed as described below. Inert gas as well as arterial and mixed venous blood gas samples were taken at the same time.

340

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE

Animal Instrumentation

RESULTS

Venous catheters were inserted percutaneously into auricular veins for inert gas infusion and continuous infusion of propofol. A no. 7F thermistor-tipped Swan-Ganz catheter (Baxter Edwards, Irvine, CA) was inserted into the right jugular vein and advanced into a main pulmonary artery by use of direct-pressure monitoring. This allowed measurement of cardiac output, pulmonary arterial pressure, and mixed venous blood sampling. The left femoral artery was cannulated for measurement of systemic pressure and arterial blood gas sampling.

Hemodynamic Measurements

Hemodynamic Measurements Mean arterial pressure (Pa), central venous pressure (Pcv), Ppa, and Ppc, we were measured with an ICU monitor (Servomed; Hellige GmbH, Frieburg, Germany) with standard pressure transducers (model 1290A; Hewlett-Packard, Böblingen, Germany) that had been zeroed to the level of the right atrium. Cardiac output was measured with a Baxter-Vigilance Monitor (Edwards Critical Care Division, Irvine, CA) and a no. 7F Swan-Ganz catheter. The mean of six determinations of cardiac output was recorded.

VOL 163

2001

Results are shown in Table 1. The administration of low, normal, and high doses of sildenafil resulted in significant increases in cardiac index (CI) and stroke volume index (SVI). The greatest extent of these changes was observed 90 min after drug application via gastric tube. The changes in heart rate did not reach significance. CI rose 30, 60, and 90 min after all doses of sildenafil (p ⬍ 0.01). SVI was significantly augmented in all measurements after drug application. Pa was depressed 60 and 90 min after high-dose sildenafil relative to the control group. Trends toward an increase in Pa were observed in the control group as well as in the low- and normal-dose groups. However, these differences were insignificant. Central venous pressure (Pcv) and RLva did not undergo significant change in comparison with control. Ppa was depressed 30 and 60 min after low-dose sildenafil and in all measurements postbaseline after high-dose sildenafil.

Blood Gas and Metabolic Measurements Arterial and mixed venous samples (2 ml each) were collected simultaneously and immediately analyzed for oxygen and carbon dioxide partial pressure, pH, hemoglobin concentration, and hematocrit using AVL 995 S and AVL 912 CO-Oxylite (AVL AG, Schaffhausen, Switzerland). All values were corrected to body temperature.

TABLE 1 HEMODYNAMIC VARIABLES Baseline

30 min

60 min

90 min

⫺1

Inert Gas Measurements · · VA/Q distributions were determined using the MIGET as previously described (8, 9). A mixture of six inert gases, including sulfur hexafluoride, ethane, cyclopropane, halothane, diethyl ether, and acetone dissolved in saline, was infused via an auricular vein at a rate of 3 ml/kg per hour. This infusion was started at least 1 h before the first set of measurements. Blood samples, 10 ml, were collected in duplicate into heparinized glass syringes from the pulmonary artery and the left femoral artery. Mixed expired gas samples of 30 ml were collected from a specially designed heated mixing chamber into preheated gas-tight glass syringes. All samples were kept at a temperature of 39⬚ C and then analyzed. Gas extraction was performed as described by Wagner and coworkers (9). Concentrations of inert gases were measured by gas chromatography · · (HP-5890, Series II; Hewlett-Packard). VA/Q distributions were determined from inert gas data by use of the 50-compartment model of Wagner . and . colleagues (8) and Evans and . Wagner. (10). Inert gas shunt flow (QS/QT), log standard deviation of Q (log SDQ), .log standard deviation · · · · · of VA (log SDVA), mean VA/Q ratios of VA and Q distributions, and rafrom this 50tio of dead space to tidal volume (VD/VT) were calculated . · compartment model. Distributions of VA and Q are presented as fol· · lows: (1) blood flow to unventilated lung units, shunt flow (VA/Q ⬍ · · 0.005); (2) blood flow to poorly ventilated lung units (low VA/Q, ⬎ 0.005 through 0.1); (3) blood flow to normally ventilated lung units (normal · · VA/Q, ⬎ 0.1 through 10); (4) ventilation of poorly perfused lung units · · (5) ventilation of unperfused lung (high VA/Q, ⬎ 10 through 100); and . · units, alveolar dead space (VA,D/Q) ⬎ 100). The residual sum of squares was used as an indicator of fit of the experimental data to this 50-compartment model (6). After obtaining · · the VA/Q distribution, each set of data was examined for possible pulmonary oxygen diffusion limitation, as inert gas exchange is unaffected by this. Mixed PaO2 was calculated from compartmental endcapillary values and compared with the measured value. Respiratory Measurements Airway pressures, expiratory tidal volume, expiratory minute volume, and respiratory rate values were recorded by the built-in detectors of the ventilator.

Statistical Analyses The data were examined by two-factor analysis of variance for repeated measurements. Significant post hoc differences were further analyzed by the Newman–Keuls test; p values less than 0.05 in intergroup comparison were considered significant. Results are expressed as means ⫾ standard error of the mean (SEM).

HR, beats min Control Sildenafil 25 mg 50 mg 100 mg CI, ml min⫺1 kg⫺1 Control Sildenafil 25 mg 50 mg 100 mg SVI, ml kg⫺1 beat⫺1 Control Sildenafil 25 mg 50 mg 100 mg RP, dyn ⭈ s cm⫺5 Control Sildenafil 25 mg 50 mg 100 mg Pa, mm Hg Control Sildenafil 25 mg 50 mg 100 mg Ppa , mm Hg Control Sildenafil 25 mg 50 mg 100 mg Ppc,we, mm Hg Control Sildenafil 25 mg 50 mg 100 mg

84 ⫾ 2

90 ⫾ 3

92 ⫾ 3

94 ⫾ 3

88 ⫾ 5 89 ⫾ 4 89 ⫾ 3

97 ⫾ 6 94 ⫾ 4 97 ⫾ 1

95 ⫾ 5 96 ⫾ 3 103 ⫾ 4

99 ⫾ 6 104 ⫾ 3 104 ⫾ 3

138 ⫾ 17

130 ⫾ 16

128 ⫾ 13

130 ⫾ 32

164 ⫾ 5 143 ⫾ 14 142 ⫾ 10

163 ⫾ 9† 162 ⫾ 17† 174 ⫾ 12†

175 ⫾ 6† 174 ⫾ 12† 190 ⫾ 15†

186 ⫾ 10† 198 ⫾ 8† 196 ⫾ 13†

1.64 ⫾ 0.24

1.44 ⫾ 0.19

1.38 ⫾ 0.11

1.38 ⫾ 0.36

1.77 ⫾ 0.03 1.59 ⫾ 0.08 1.59 ⫾ 0.08

1.67 ⫾ 0.05‡ 1.68 ⫾ 0.11† 1.79 ⫾ 0.10†

1.85 ⫾ 0.05† 1.79 ⫾ 0.08† 1.83 ⫾ 0.12†

1.89 ⫾ 0.04† 1.92 ⫾ 0.06† 1.87 ⫾ 0.11†

165 ⫾ 20

18 ⫾ 26

233 ⫾ 29

202 ⫾ 81

147 ⫾ 10 116 ⫾ 15 137 ⫾ 22

136 ⫾ 11 102 ⫾ 15 102 ⫾ 8

133 ⫾ 13 101 ⫾ 15 95 ⫾ 10

132 ⫾ 9 98 ⫾ 12 114 ⫾ 11

82 ⫾ 1

86 ⫾ 2

91 ⫾ 2

93 ⫾ 3

81 ⫾ 3 76 ⫾ 2 75 ⫾ 1

78 ⫾ 6 80 ⫾ 3 73 ⫾ 3

87 ⫾ 6 88 ⫾ 6 76 ⫾ 4‡

89 ⫾ 7 95 ⫾ 5 77 ⫾ 4‡

18 ⫾ 0.2

19 ⫾ 0.9

21 ⫾ 0.6

20 ⫾ 0.8

18 ⫾ 1.3 19 ⫾ 1.4 16 ⫾ 1.6

16 ⫾ 1.3† 18 ⫾ 1.2 14 ⫾ 1.8†

16 ⫾ 1.6† 20 ⫾ 1.3 15 ⫾ 2.1†

17 ⫾ 1.7 21 ⫾ 1.3 15 ⫾ 1.5†

7 ⫾ 0.5

8 ⫾ 0.5

7 ⫾ 0.6

8 ⫾ 0.8

8 ⫾ 1.4 9 ⫾ 1.6 7 ⫾ 1.1

6 ⫾ 0.9 8 ⫾ 0.9 6 ⫾ 1.3

6 ⫾ 0.9 9 ⫾ 0.6 7 ⫾ 1.3

6 ⫾ 1.0 9 ⫾ 0.6 6 ⫾ 0.9

Definition of abbreviations: CI ⫽ cardiac index; HR ⫽ heart rate; Pa ⫽ mean arterial pressure; Ppa ⫽ mean pulmonary arterial pressure; Ppc,we ⫽ pulmonary capillary wedge pressure; RP ⫽ pulmonary vascular resistance; SVI ⫽ stroke volume index. * Data represent means ⫾ SEM. † p ⬍ 0.01 in comparison with control. ‡ p ⬍ 0.05 in comparison with control.

341

Kleinsasser, Loeckinger, Hoermann, et al.: Sildenafil, Hemodynamics, and Lung

Figure 1. Inert gas measurements. (A) Shunt perfusion as percentage of cardiac output 30, 60, and 90 min after administration. Values represent means ⫾ SEM; *p ⬍ 0.05, † p ⬍ 0.01 in comparison with control. BL ⫽ baseline. (B) Low · · VA/Q perfusion as percentage of cardiac output 30, 60, and 90 min after administration. Values represent · · means ⫾ SEM. BL ⫽ baseline. (C) Normal VA/Q perfusion as percentage of cardiac output 30, 60, and 90 min after administration. Values represent means ⫾ SEM; *p ⬍ 0.05, † p ⬍ 0.01 in comparison · ·with control. BL ·⫽ baseline. · (D) Shunt, low VA/Q , and normal VA/Q perfusion after administration of 50 mg of sildenafil. The y axis represents blood flow as percentage of cardiac· output to lung units with zero, low, or · normal VA/Q ratio. Values represent means ⫾ SEM; † p ⬍ 0.01 in comparison with baseline value. BL ⫽ baseline.

Inert Gas Measurements

Normal-, low-, and high-dose sildenafil significantly increased · · blood flow to units with a VA/Q ratio or zero (intrapulmonary shunt); significant reductions in blood flow to units with a nor· · mal VA/Q ratio were hence found in all groups (Figure 1). The · · amount of blood flow to units with a low VA/Q ratio did not change significantly (Figure 1). Alveolar dead space ventila· · tion (VA/Q ⬎ 100) remained unchanged in all four groups (Ta· · · · ble 2). Log SD VA as well as mean VA/Q of VA did not change · · significantly in any of the four groups (Table 2). Mean VA/Q of · · Q and log SDQ decreases reflected the redistribution of blood flow as described earlier, although these changes were not significant. No measured–predicted PaO2 difference could be detected in comparing treatment groups with control. An indica· · tion of acceptable quality of the VA/Q distribution is a residual sum of squares (RSS) of 5.3 or less in half the experimental runs (50th percentile) or 10.6 or smaller in 90% of the experimental runs (90th percentile) (6). In our experiment, 88.9% of the RSS was less than 5.3 and 100% was less than 10.6.

TABLE 2 INERT GAS EXCHANGE* Baseline

30 min

60 min

90 min

45 ⫾ 1.4

43 ⫾ 0.5

43 ⫾ 0.7

43 ⫾ 0.9

45 ⫾ 1.7 46 ⫾ 2.3 46 ⫾ 3.4

47 ⫾ 2.2 42 ⫾ 0.8 46 ⫾ 2.3

47 ⫾ 2.7 42 ⫾ 1.2 45 ⫾ 1.6

40 ⫾ 7.7 45 ⫾ 1.2 46 ⫾ 2.2

0.77 ⫾ 0.03

0.71 ⫾ 0.03

0.68 ⫾ 0.04

0.75 ⫾ 0.03

0.92 ⫾ 0.11 0.87 ⫾ 0.06 0.92 ⫾ 0.08

0.72 ⫾ 0.07 0.77 ⫾ 0.04 0.71 ⫾ 0.04

0.72 ⫾ 0.06 0.78 ⫾ 0.08 0.72 ⫾ 0.05

0.71 ⫾ 0.04 0.77 ⫾ 0.06 0.74 ⫾ 0.03

0.66 ⫾ 0.05

0.68 ⫾ 0.04

0.66 ⫾ 0.04

0.66 ⫾ 0.06

0.70 ⫾ 0.06 0.88 ⫾ 0.13 0.67 ⫾ 0.14

0.90 ⫾ 0.08 0.82 ⫾ 0.14 0.77 ⫾ 0.14

0.71 ⫾ 0.08 0.70 ⫾ 0.09 0.66 ⫾ 0.07

0.68 ⫾ 0.04 0.67 ⫾ 0.08 0.66 ⫾ 0.08

1.13 ⫾ 0.01

1.11 ⫾ 0.01

1.05 ⫾ 0.05

1.15 ⫾ 0.09

1.42 ⫾ 0.20 1.49 ⫾ 0.25 1.26 ⫾ 0.14

1.46 ⫾ 0.22 1.35 ⫾ 0.22 1.19 ⫾ 0.13

1.16 ⫾ 0.09 3.70 ⫾ 2.34 1.17 ⫾ 0.09

1.10 ⫾ 0.08 1.21 ⫾ 0.14 1.11 ⫾ 0.09

0.58 ⫾ 0.02

0.61 ⫾ 0.04

0.58 ⫾ 0.01

0.58 ⫾ 0.03

0.61 ⫾ 0.04 0.59 ⫾ 0.02 0.63 ⫾ 0.08

0.68 ⫾ 0.05 0.56 ⫾ 0.04 0.56 ⫾ 0.03

0.60 ⫾ 0.04 0.54 ⫾ 0.05 0.67 ⫾ 0.10

0.59 ⫾ 0.03 0.58 ⫾ 0.05 0.55 ⫾ 0.03

Respiratory Measurements

· · VD/VT,% Control Sildenafil 25 mg 50 mg 100 mg · · · Mean VA/Q of Q Control Sildenafil 25 mg 50 mg 100 mg · Log SDQ Control Sildenafil 25 mg 50 mg 100 mg · · · Mean VA/Q of VA Control Sildenafil 25 mg 50 mg 100 mg · Log SDVA Control Sildenafil 25 mg 50 mg 100 mg

Mean airway pressure was 10 cm H2O (mean), and peak airway pressure was 24 cm H2O (mean). No significant differences could be observed. Respiratory system compliance remained stable in all animals.

· Definition of abbreviations: log SD Q ⫽ log standard deviation of the distribution of · · · perfusion; log SDVA ⫽ standard deviation of the distribution of ventilation; mean VA/Q · · · · V of Q ⫽ mean of the distribution of perfusion; mean VA/Q of ⫽ mean of the distribu· · tion of ventilation; VD/VT ⫽ alveolar dead space ventilation. * Data represent means ⫾ SEM.

Blood Gas Data

Arterial blood gas data and calculations are presented in Table 3, and mixed venous blood gas data are presented in Table 4. PaO2 was significantly decreased 30, 60, and 90 min after administration of normal- and high-dose sildenafil in comparison with control, but not after 25 mg. Alveolar-arterial oxygen pressure difference [P(A–a)O2], oxygen delivery, and uptake were increased in all animals treated with sildenafil in comparison with control. There were no significant changes in arterial and mixed venous partial pressures of carbon dioxide, mixed venous partial pressure of oxygen, or in arterial pH (pHa) and mixed venous pH (pH v).

342

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE TABLE 3

TABLE 4

ARTERIAL BLOOD GAS DATA AND OXYGEN CALCULATIONS*

MIXED VENOUS BLOOD GAS DATA*

Baseline PaO2, mm Hg Control Sildenafil 25 mg 50 mg 100 mg PaCO2, mm Hg Control Sildenafil 25 mg 50 mg 100 mg SaO2, % Control Sildenafil 25 mg 50 mg 100 mg pHa Control Sildenafil 25 mg 50 mg 100 mg P(A–a)O2, mm Hg Control Sildenafil 25 mg 50 mg 100 mg DO2, ml O2 min⫺1 Control Sildenafil 25 mg 50 mg 100 mg · VO2, ml O2 min⫺1 Control Sildenafil 25 mg 50 mg 100 mg

30 min

60 min

90 min

91 ⫾ 1.8

89 ⫾ 0.9

89 ⫾ 1.1

90 ⫾ 0.9

90 ⫾ 1.1 93 ⫾ 1.5 94 ⫾ 1.1

85 ⫾ 1.1 81 ⫾ 2.4† 82 ⫾ 3.5‡

86 ⫾ 1.9 80 ⫾ 1.6† 82 ⫾ 3.2‡

84 ⫾ 2.0 78 ⫾ 2.2† 80 ⫾ 2.9†

32 ⫾ 0.4

33 ⫾ 0.7

33 ⫾ 0.2

32 ⫾ 0.8

30 ⫾ 1.0 31 ⫾ 1.9 31 ⫾ 0.6

30 ⫾ 1.2 32 ⫾ 1.5 30 ⫾ 0.6

31 ⫾ 1.0 32 ⫾ 1.0 30 ⫾ 0.6

30 ⫾ 1.0 33 ⫾ 1.4 30 ⫾ 0.9

97 ⫾ 0.3

96 ⫾ 0.7

96 ⫾ 0.8

96 ⫾ 0.6

96 ⫾ 0.4 96 ⫾ 1.0 96 ⫾ 0.6

96 ⫾ 0.4 95 ⫾ 1.0 94 ⫾ 0.7

96 ⫾ 0.2 94 ⫾ 0.9 95 ⫾ 0.6

96 ⫾ 0.4 93 ⫾ 1.3 94 ⫾ 1.0

7.53 ⫾ 0.01

7.52 ⫾ 0.01

7.52 ⫾ 0.01

7.53 ⫾ 0.01

7.54 ⫾ 0.01 7.52 ⫾ 0.02 7.53 ⫾ 0.01

7.54 ⫾ 0.01 7.50 ⫾ 0.02 7.54 ⫾ 0.01

7.53 ⫾ 0.01 7.49 ⫾ 0.02 7.54 ⫾ 0.01

7.53 ⫾ 0.01 7.47 ⫾ 0.02 7.53 ⫾ 0.01

13 ⫾ 1

12 ⫾ 1

12 ⫾ 1

13 ⫾ 1

14 ⫾ 1 13 ⫾ 2 12 ⫾ 1

22 ⫾ 1† 24 ⫾ 2† 25 ⫾ 3†

21 ⫾ 1† 24 ⫾ 2† 26 ⫾ 3†

23 ⫾ 2† 24 ⫾ 2† 27 ⫾ 3†

523 ⫾ 25

453 ⫾ 23

460 ⫾ 12

463 ⫾ 33

583 ⫾ 39 542 ⫾ 34 518 ⫾ 25

596 ⫾ 40 578 ⫾ 51† 604 ⫾ 23†

625 ⫾ 29 602 ⫾ 50† 614 ⫾ 12†

632 ⫾ 27 664 ⫾ 33† 649 ⫾ 33†

69 ⫾ 3

60 ⫾ 3

61 ⫾ 2

61 ⫾ 4

77 ⫾ 5 72 ⫾ 4 69 ⫾ 4

79 ⫾ 5 77 ⫾ 6† 80 ⫾ 3†

83 ⫾ 4 80 ⫾ 6† 81 ⫾ 5†

84 ⫾ 3† 88 ⫾ 4† 86 ⫾ 5†

†

†

†

†

P vO2, mm Hg Control Sildenafil 25 mg 50 mg 100 mg P vCO2, mm Hg Control Sildenafil 25 mg 50 mg 100 mg S vO2, % Control Sildenafil 25 mg 50 mg 100 mg pH v Control Sildenafil 25 mg 50 mg 100 mg

VOL 163

2001

Baseline

30 min

60 min

90 min

42 ⫾ 2.3

41 ⫾ 1.8

42 ⫾ 2.7

41 ⫾ 2.6

45 ⫾ 0.8 45 ⫾ 0.7 45 ⫾ 0.7

46 ⫾ 0.9 46 ⫾ 1.1 46 ⫾ 1.0

46 ⫾ 1.1 46 ⫾ 0.5 45 ⫾ 1.1

45 ⫾ 0.4 45 ⫾ 0.7 45 ⫾ 1.0

38 ⫾ 0.7

39 ⫾ 0.7

38 ⫾ 0.4

37 ⫾ 0.7

37 ⫾ 1.2 39 ⫾ 1.6 36 ⫾ 1.35

36 ⫾ 1.65 39 ⫾ 2.2 35 ⫾ 0.6

35 ⫾ 1.0 39 ⫾ 1.9 35 ⫾ 0.8

35 ⫾ 0.9 39 ⫾ 2.1 35 ⫾ 1.1

84 ⫾ 0.3

82 ⫾ 0.6

82 ⫾ 0.7

83 ⫾ 0.5

83 ⫾ 0.4 83 ⫾ 0.8 83 ⫾ 0.5

83 ⫾ 0.4 82 ⫾ 0.9 82 ⫾ 0.6

83 ⫾ 0.2 81 ⫾ 0.8 83 ⫾ 0.5

83 ⫾ 0.3 80 ⫾ 1.2 82 ⫾ 0.9

7.47 ⫾ 0.01

7.47 ⫾ 0.01

7.47 ⫾ 0.01

7.48 ⫾ 0.01

7.48 ⫾ 0.1 7.44 ⫾ 0.02 7.49 ⫾ 0.01

7.48 ⫾ 0.02 7.44 ⫾ 0.03 7.49 ⫾ 0.01

7.48 ⫾ 0.01 7.43 ⫾ 0.03 7.49 ⫾ 0.01

7.49 ⫾ 0.01 7.42 ⫾ 0.03 7.49 ⫾ 0.01

Definition of abbreviations: pH v ⫽ mixed venous pH; P vO2 ⫽ mixed venous pressure of oxygen; P vCO2 ⫽ mixed venous pressure of carbon dioxide; S vO2 ⫽ mixed venous oxyhemoglobin saturation. * Data represent means ⫾ SEM.

†

Definition of abbreviations: DO2 ⫽ O2 delivery; P(A–a)O2 ⫽ alveolar–arterial oxygen pressure difference; PaCO2 ⫽ arterial partial pressure of carbon dioxide; PaO2 ⫽ arterial partial pressure of oxygen; pHa ⫽ arterial pH; SaO2 ⫽ arterial oxyhemoglobin satura· tion; VO2 ⫽ O2 uptake. * Values represent means ⫾ SEM. † p ⬍ 0.01. ‡ p ⬍ 0.05.

Dose Dependency

No significant differences between the three treatment groups could be detected in any of the parameters measured. However, there was a trend for normal- and high-dose sildenafil to cause the greatest extent of change with respect to the depression of PaO2 and hemodynamic stimulation (Tables 1 and 2).

DISCUSSION The main findings of our study were that 50 and 100 mg of sildenafil depressed PaO2 and that all applied doses of sildenafil stimulated hemodynamics in anesthetized, ventilated pigs. Increased perfusion of a lung with unchanged ventilation may have caused this inadequate gas exchange. In the presence of the cardiac depressant effects of propofol (11), all doses of sildenafil elicited significant increases in CI (p ⬍ 0.01). We propose two possible explanations for the CI-increasing action of sildenafil. On the one hand, sildenafil might possess consid-

erable affinity for PDEs V and III. On the other hand, an intracellular cGMP-dependent inhibition of PDE III resulting in increased myocardial cAMP (2, 12) and consequently in increased cardiac performance might be a potential explanation. The inhibition of AMP breakdown by GMP has already been shown in a platelet cell model by Maurice and Haslam in 1990 (13). Increased myocardial AMP is consistent with the increases in CI and heart rate observed in our experiment. Although the pattern of ventilation was maintained, significant depression of arterial PO2 was noted in the presence of increased cardiac output. A positive correlation between shunt flow and cardiac output was first suggested by Kelman and coworkers (14). Rennotte and associates found that the inotropic effect of dopamine and dobutamine is linked to a redistri· · bution of blood flow to lung units with a VA/Q ratio of zero (intrapulmonary shunt flow) (15). In addition Prielipp and colleagues have shown that the inotropic amrinone (a PDE III inhibitor) significantly increases pulmonary shunt flow in extubated patients 24 h after aortocoronary bypass graft (16). However, it cannot be stated to what extent the changes in gas exchange can be attributed to presumed vasodilatatory effects on the one hand, or to the substantial increase in CI on the other hand. The falls in Ppa observed after low- and high-dose sildenafil are consistent with the presence of PDE isozyme types I, III, IV, and V in the human pulmonary artery (3). Calculation of RLva showed no significant differences in comparing treatment groups with control. The latter was presumably due to the distension of the pulmonary blood vessels in the presence of an elevated cardiac output (Table 1). Although no significant difference between the treatment groups could be detected, normal- and high-dose sildenafil caused the highest response with respect to depression of PaO2 and increase in CI. The dose applied did not have an effect on onset time or similar other response. On the basis of the results of this study, sildenafil-associ-

Kleinsasser, Loeckinger, Hoermann, et al.: Sildenafil, Hemodynamics, and Lung

ated modulation of hemodynamics and pulmonary gas exchange may be associated with specific risks. First, in patients with chronic obstructive pulmonary disease there is already a large amount of blood flow going to lung units with very low · · · · VA/Q ratios. Further increase in VA/Q heterogeneity will worsen hypoxemia. Second, the marked decrease in PaO2 associated with an increase in cardiac output may be deleterious in patients with coronary heart disease. Some limitations of this study should be mentioned. In the present study, the drug was administered via gastric tube. Because the amount of enteral resorption is variable, the plasma concentrations of sildenafil remained uncertain even though all animals had been fasted overnight and all animals were of similar age and weight. However, sildenafil is available only as a tablet, so our experiment resembled oral administration in humans. Also, differences between anesthetized pigs and awake humans must be remembered, as pigs do not have a collateral ventilation but a very reactive pulmonary vascular system and therefore more easily develop intrapulmonary right-to-left shunt. Sildenafil represents a novel orally effective substance with cardiotonic actions. Arterial PO2 decrease has been associated with significant rises in pulmonary shunt flow and CI. References 1. U.S. Food and Drug Administration. Postmarketing safety of sildenafil citrate. Summary of reports of deaths in sildenafil citrate users from late March through mid-November 1998. Washington DC: U.S. Food and Drug Administration (Center for Drug Evaluation and Research); 1998. 2. Fischmeister R, Mery PF. Regulation of cardiac calcium current by cGMP/NO route. CR Seances Soc Biol Fil 1996;190:181–206. 3. Rabe KF, Tenor H, Dent G, Schudt C, Nakashima M, Magnussen H. Identification of PDE isozymes in human pulmonary artery and effect of selective PDE inhibitors. Am J Physiol 1994;266:536–543. 4. Hanson KA, Burns F, Rybalkin SD, Miller JW, Beavo J, Clarke WR. Developmental changes in lung cGMP phosphodiesterase-5 activity, protein and message. Am J Respir Crit Care Med 1998;158:279–288.

343

5. West JB. Ventilation–perfusion inequality and overall gas exchange in computer models of the lung. Respir Physiol 1969;7:88–110. 6. Roca J, Wagner PD. Contribution of multiple inert gas elimination technique to pulmonary medicine: 1. Principles and information content of the multiple inert gas elimination technique. Thorax 1994;49:815–824. 7. Boolell M, Allen MJ, Ballard SA, Gepi-Attee S, Muirhead GJ, Naylor AM, Osterloh IH, Gingell C. Sildenafil: an orally active type 5 cyclic GMP-specific phosphodiesterase inhibitor for the treatment of penile erectile dysfunction. Int J Impot Res 1996;8:47–52. 8. Wagner PD, Saltzman JA, West JB. Measurement of continuous distributions of ventilation–perfusion ratios: theory. J Appl Physiol 1974;36: 588–599. 9. Wagner PD, Naumann PF, Laravuso RB. Simultaneous measurement of eight foreign gases in blood by gas chromatography. J Appl Physiol 1974; 36:600–605. 10. Evans JW, Wagner PD. Limits on VA/Q distributions from analysis of experimental inert gas elimination. J Appl Physiol 1977;42:889–898. 11. Sherry KM, Sartain J, Bell JH, Wilkinson GA. Comparison of the use of a propofol infusion in cardiac surgical patients with normal and low cardiac output states. J Cardiothorac Vasc Anesth 1995;9:368–372. 12. Mery PF, Pavoine C, Belhassen L, Pecker F, Fischmeister R. Nitric oxide regulates cardiac Ca⫹⫹ current. Involvement of cGMP-inhibited and cGMP-stimulated phosphodiesterase through guanylyl cyclase activation. J Biol Chem 1993;268:26286–26295. 13. Maurice DH, Haslam RJ. Molecular basis of the synergistic inhibition of platelet function by nitrovasodilators and activators of adenylate cyclase: inhibition of cyclic AMP breakdown by cyclic GMP. Mol Pharmacol 1990;37:671–681. 14. Kelman GR, Nunn JF, Prys-Roberts C, Greenbaum R. The influence of cardiac output on arterial oxygenation: a theoretical study. Br J Anaesth 1967;39:450–458. 15. Rennotte MT, Reynaert M, Clerbaux T, Willems E, Roeseleer J, Veriter C, Rodenstein D, Frans A. Effects of two inotropic drugs, dopamine and dobutamine, on pulmonary gas exchange in artificially ventilated patients. Intensive Care Med 1989;15:160–165. 16. Prielipp RC, Butterworth JF IV, Zaloga GP, Robertie PG, Royster RL. Effects of amrinone on cardiac index, venous oxygen saturation and venous admixture in patients recovering from cardiac surgery. Chest 1991;99:820–825.