Selective Pulmonary Vasodilation by Intravenous Infusion of an Ultrashort Half-life Nucleophile/Nitric Oxide Adduct 

These investigations were approved by the Subcommittee for Research Animal Studies at Massachusetts General Hospital (Boston, MA).

Eight Suffolk lambs weighing 25–30 kg each were anesthetized by inhalation of halothane in oxygen. Their tracheas were intubated and their lungs mechanically ventilated at 15 breaths/min and a tidal volume of 15 ml/kg with a large animal ventilator (Harvard Apparatus, Natick, MA). A femoral artery was cannulated with a polyvinyl chloride catheter (2-mm internal diameter) advanced 20 cm into the aorta for continuous arterial pressure monitoring and arterial blood sampling. A tracheotomy was performed, and a 7.0-mm (internal diameter) cuffed tracheotomy tube (Portex, Keene, NH) was inserted. A three-port thermodilution pulmonary artery catheter (right ventricular ejection fraction/volumetric thermodilution catheter, model 93A-434H-7.5F; Baxter, Irvine, CA) was placed via the right external jugular vein through an 8.5-French introducer (Cordis, Miami, FL). The lambs were housed in a Babraham cage, had free access to food and water, and allowed 2 h to recover from anesthesia. Animals meeting the following criteria were excluded: a peripheral leukocyte count < 4,000 or > 12,000 per cubic millimeter, a mean pulmonary arterial pressure (PAP) > 20 mmHg, or a core temperature > 40.1 [degree sign] Celsius.

Systemic arterial pressure (SAP), PAP, and central venous pressure were measured continuously, and pulmonary capillary wedge pressure (PCWP) was measured intermittently with calibrated pressure transducers (Cobe Laboratories, Lakewood, CO) zeroed at the midchest level. After amplification of pressure signals (model 7700; Hewlett Packard, Palo Alto, CA), the values were recorded (Western Graphtec, Inc., Irvine, CA). Mean measurements were obtained at end expiration. Cardiac output was measured by thermodilution as the average of three determinations after injection of 5 ml of Ringer's lactate at 0 [degree sign] Celsius. Pulmonary vascular resistance (PVR) and systemic vascular resistance (SVR) were computed with standard formulas.

A two-way nonrebreathing valve (Hans Rudolph, Inc., Kansas City, MO) was attached to the tracheotomy to separate inspired from expired gas. The lambs breathed 100% oxygen administered through a 5-l rubber reservoir bag. The expired concentration of NO was measured continuously with a chemiluminescence NO/NO^Xanalyzer [10](CLD 700 AL; Eco Physics, Durnten, Switzerland) at the exhalation port of the nonrebreathing valve. Before analysis, the exhaled gas was passed through a water trap to remove any moisture. Separate breathing circuits and valves were used during the administration of NO gas and for the measurement of exhaled NO during the SNP or PROLI/NO infusion to avoid any NO release by residual tubing contamination. Nitric oxide gas was introduced into the inspiratory limb of the breathing circuit immediately before the 5-l reservoir bag. The inspired concentration of NO was measured continuously by the same NO/NO^Xanalyzer at the inhalation port of the nonrebreathing valve. Exhaled gas was scavenged and discarded by continuous aspiration.

After baseline measurements were taken, an intravenous infusion of the potent pulmonary vasoconstrictor 9,11-dideoxy-9 alpha,11 alpha-methanoepoxy prostaglandin F²alpha (U46619; Cayman Chemical Company, Ann Arbor, MI) was administered at a rate of 0.4–0.8 micro gram [center dot] kg sup -1 [center dot] min sup -1 to increase the mean PAP to 30 mmHg. Nitric oxide gas (10, 20, 40, and 80 parts per million by volume [ppm]) was inhaled for 6 min followed by 6-min NO-free intervals. All parameters returned to baseline values within the 6 min. Intravenous PROLI/NO infusions (0.75, 1.5, 3, 6, and 12 micro gram [center dot] kg sup -1 [center dot] min sup -1) and SNP infusions (0.25, 0.5, 1, 2, and 4 micro gram [center dot] kg sup -1 [center dot] min sup -1) were administered for 15 min, allowing 15-min intervals between doses because all hemodynamic parameters returned to baseline values within these 15 min. SNP and PROLI/NO were administered via the third port of the Swan Ganz catheter. Each experiment lasted approximately 6 h. The order of administration and doses of the vasoactive drugs (SNP, PROLI/NO, and NO gas) were randomized. Cardiac output was measured every 15 min. Arterial blood samples for the measurement of methemoglobin concentrations were obtained at baseline and at the end of each drug administration in three sheep. Plasma thiocyanate concentrations were obtained at baseline and after each dose of SNP was administered in three sheep.

Ten milligrams of the stable endoperoxide analogue of thromboxane, U46619, were dissolved in 50 ml of lactated Ringer's solution just before administration. Twenty-five milligrams of PROLI/NO complex, prepared as described previously, [9]were dissolved in 50 ml of saline containing 0.1-M NaOH (pH 13). Ten milligrams of SNP (Elkins-Sinn Inc., Cherry Hill, NJ) were dissolved in 50 ml of lactated Ringer's solution just before administration. The solution was shielded from light by wrapping the syringe and line with aluminum foil. Nitric oxide was obtained from Airco (Murray Hill, NJ) as a mixture of 800 ppm NO in nitrogen. Less than 1% of the stock NO gas was present as NO². Nitric oxide was mixed with O²in a 5-l reservoir bag before reaching the inhalation port of the two-way valve.

Values for the hemodynamic variables at the end of each period are reported as mean +/- SE. The effects of each vasoactive agent (PROLI/NO, NO, SNP) on these variables were compared with the average of the baseline value before and after each drug administration. Differences among treatments were analyzed with a repeated measures analysis of variance. To compare the selectivity of these drugs, a linear regression analysis of the change in PAP (or resistance) versus the change in SAP (or resistance) was performed. Paired Student's t tests were performed to compare exhaled NO concentrations during PROLI/NO and SNP infusions (all data pooled) and Delta SAP/Delta PAP slopes and Delta SVR/Delta PVR slopes with Bonferroni correction for three groups (SNP, PROLI/NO, and NO). Differences were considered significant at P < 0.05.

During infusion of U46619, PAP, PVR, SAP, SVR, and PCWP increased whereas cardiac output decreased (Table 1). Exhaled NO concentrations at baseline remained unchanged during infusion of U46619 (3.8 +/- 0.6 vs. 3.6 +/- 0.7 parts per billion by volume [ppb], respectively; n = 4).

At all dose levels, inhalation of NO produced a prompt and stable reduction of pulmonary hypertension in a dose-dependent manner (Table 1). The onset of pulmonary vasodilation occurred within seconds after beginning NO inhalation, and the vasodilator effect was maximal within 3 min. The previous level of pulmonary vasoconstriction returned within 3–6 min after terminating NO inhalation. Inhalation of NO produced highly selective pulmonary vasodilation, as mean SVR or SAP was unchanged at all the doses we tested (Figure 1and Table 1). Pulmonary capillary wedge pressure was not altered during inhalation of NO (Table 1). Methemoglobin concentrations remained < 1.5% at all levels of NO administration (n = 3).

PROLI/NO infusion selectively decreased the pulmonary vascular resistance in a dose-dependent manner at infusion rates up to 6 micro gram [center dot] kg sup -1 [center dot] min sup -1 (Figure 1and Table 1). The Delta SVR/Delta PVR slope induced by the four lowest doses of PROLI/NO infusion was similar to the inhaled NO slope (0.16 +/- 0.53 vs. -0.10 +/- 0.28 [not significant];Figure 1(B)). At 12 micro gram [center dot] kg sup -1 [center dot] min sup -1, PROLI/NO significantly decreased SAP, PAP, and PCWP (Figure 1and Table 1). The previous level of pulmonary vasoconstriction returned within 3–6 min of termination of the PROLI/NO infusion. Exhaled NO concentrations increased in a dose-dependent manner during infusions of PROLI/NO (Figure 2). Methemoglobin concentrations remained < 1.5% at all administered doses of PROLI/NO (n = 3).

The administration of SNP induced nonselective vasodilation of the pulmonary and systemic vasculature. Any decrease of the PAP or PVR was associated with a concomitant decrease of SAP or SVR at all concentrations of SNP (Figure 1(A and B)). For SNP, the slope of the relation Delta SVR/Delta PVR differed from the slopes of inhaled NO and intravenous PROLI/NO (4.09 +/- 0.55 vs. -0.10 +/- 0.28 and 0.16 +/- 0.53, respectively; P < 0.001 for both). The increases in exhaled NO concentrations were low compared with those produced by infusion of PROLI/NO and were significant only at a 4-micro gram [center dot] kg sup -1 [center dot] min sup -1 dose (Figure 2). The PCWP decreased significantly at the 1-, 2-, and 4-micro gram [center dot] kg sup -1 [center dot] min sup -1 infusion rate for SNP (Table 1). Methemoglobin concentrations remained < 1.5 mg/ml at all levels of SNP infusion (n = 3). Thiocyanate concentrations remained low (< 2.5 mg/dl) at all levels of SNP infusion (n = 3; toxic range > 12 mg/dl).

This study demonstrates that PROLI/NO, because of its short half-life, can release NO and produce vasodilation within the pulmonary vasculature and lose its vasodilatory activity before reaching and affecting the systemic circulation. In the lamb, this compound appeared to be a selective pulmonary vasodilator at doses up to 6 micro gram [center dot] kg sup -1 [center dot] min sup -1. This infused dose has a similar potency to that of 40 ppm inhaled NO in our acute pulmonary hypertension model. As expected, NO gas was a selective pulmonary vasodilator at all levels. Infused SNP was not selective and induced systemic vasodilation in proportion to the degree of pulmonary vasodilation. Exhaled NO concentrations measured during the PROLI/NO infusion were far higher than those observed during the SNP infusion. Because of the response time (> 5 s) of the chemiluminescence device used in this study, the expired NO does not represent the concentration of NO in the alveolar space but rather an average of the latter and the anatomic dead space free of NO gas.

We observed that inhaled NO selectively vasodilated the pulmonary circulation at all concentrations inhaled in this study. This confirms previous studies that reported the high pulmonary vasodilator selectivity of this drug in the awake lamb [3]and humans. [11]Inspired NO gas mediates pulmonary vasodilation by activating pulmonary smooth muscle-soluble guanylyl cyclase and increasing intracellular cGMP concentrations. [12]Inhaled NO diffuses into the intravascular space and is rapidly inactivated by oxyhemoglobin, becoming incapable of causing systemic vasodilation. [3,13] 

Sodium nitroprusside nonselectively vasodilated both the pulmonary and systemic vasculature in our ovine pulmonary hypertension model. Previous studies showed that infusion of nitroprusside or nitroglycerin can also reduce the pulmonary vasoconstriction produced by infusion of U46119 in dogs, but such infusions produced marked concomitant systemic arterial vasodilation and hypotension. [14]Because of the longer biologic half-life ([nearly =] 2 min) of SNP in blood, [15]there was no pulmonary selectivity of infused SNP at any dose tested.

PROLI/NO produced selective pulmonary vasodilation when administered intravenously at doses up to 6 micro gram [center dot] kg sup -1 [center dot] min sup -1. Intravenous administration is possible because of the stability of this compound in alkaline solution. The extensive release of NO within the lung is due to the very short half-life ([nearly =] 2 s) of this compound under physiologic conditions. This is supported by the far greater exhaled NO concentrations measured during PROLI/NO infusion compared with those seen during infusion of SNP. At higher infusion rates (12 micro gram [center dot] kg sup -1 [center dot] min sup -1), PROLI/NO induced a marked systemic vasodilation. The increased cardiac output accompanying systemic vasodilation reduces the pulmonary and systemic transit time of PROLI/NO, further increasing the exposure of the systemic vascular bed to the drug. This process augments systemic vasodilation. The reduced PCWP accompanying systemic vasodilation reduces the pulmonary blood volume and also contributes to an increased pulmonary transit time. The highest dose of infused PROLI/NO that we found to be selective in our study (6 micro gram [center dot] kg sup -1 [center dot] min sup -1) was as potent as inhaling 40 ppm NO in reducing mean PAP from 31 to [nearly =] 22 mmHg. The complete reversal of the U46619-induced pulmonary hypertension produced by inhaling 80 ppm NO was not achieved by infusing PROLI/NO, however, without also inducing systemic vasodilation.

In this study, we have shown that in lambs that were given PROLI/NO intravenously the drug can selectively vasodilate the pulmonary vasculature by promptly releasing NO in the lung. The maximum selective effect obtained with this drug was equivalent to inhaling 40 ppm NO.

The authors thank Melahat Kavosi, B.S., for technical assistance, Dr. Allan Zaslavski, Department of Statistics, Harvard University, for statistical help, Dr. Joseph Saavedra for providing the PROLI/NO, and Dr. Keith Davies for kinetic measurements.