Minimizing the Inhaled Dose of NO With Breath-by-Breath Delivery of Spikes of Concentrated Gas
Background—Pulmonary vasodilatation with a 100 ppm concentration of NO given as a short burst of a few milliliters at the beginning of each breath (NOmin) was compared with conventionally inhaled NO, in which a full breath of 40 ppm of NO was inhaled (NOCD).
Methods and Results—NOmin was studied in 16 patients with severe pulmonary hypertension and in 16 isolated porcine lungs with experimentally induced pulmonary hypertension. We compared volumes of 8 to 38 mL of 100 ppm NO in N2 injected at the beginning of each breath with conventional inhalation of 40 ppm NO in air. NOCD and NOmin were studied in 4 pigs after inhibition of NO synthase with NG-nitro-l-arginine methyl ester (1 to 2 mg/kg IV) had raised the pulmonary vascular resistance index (PVRI) from 4.4±0.8 to 10.0±1.6 mm Hg · L−1 · min−1 · kg−1. A similar comparison was made in 7 isolated porcine lungs after the thromboxane analogue U46619 (10 pmol · L−1 · min−1) increased the mean PVRI from 4.6±0.8 to 12.2±1.3 mm Hg · L−1 · min−1 · kg−1. Patients’ mean PVRI was reduced from 29.2±3.7 to 24.0±3.1 with NOmin and 24.5±3.3 mm Hg · L−1 · min−1 · m−2 (mean±SEM) with NOCD. In isolated porcine lungs, there was the same reduction of PVRI for NOmin and NOCD between 12.7% and 34.8%.
Conclusions—A small volume of NO inhaled at the beginning of the breath was equally effective as NOCD but reduced the dose of NO per breath by 40-fold, which ranged from 1.2×10−8 (0.4 μg) to 1.6×10−7 mol/L (4.8 μg) compared with 5.3×10−7 (16 μg) to 1.2×10−6 mol/L (36 μg) per breath with NOCD.
Inhaled NO is a selective pulmonary vasodilator,1 2 and in acute respiratory distress syndrome (ARDS), it improves gas exchange.3 Current methods of administration of NO are complicated and not without hazard.4
NO is stored in concentrations of 100 to 10 000 ppm, with N2. Safe and effective concentrations of inhaled NO range from 1 to 40 ppm,5 which are achieved by mixing NO with respiratory gases. The flow rate of the mixed gases is carefully matched to the rate of ventilation, preventing buildup of NO2.6 While straightforward for ventilated patients, ambulatory patients who vary their rate of ventilation renders delivery difficult. A new method of delivery is needed for them because inhaled NO can be used to treat pulmonary hypertension.7
Higher doses of inhaled NO are needed to reduce pulmonary hypertension than to improve gas exchange in ARDS.8 Conventional delivery (NOCD), which distributes inhaled NO throughout the lungs, can, however, worsen gas exchange in patients with chronic obstructive pulmonary disease (COPD).9 10 Selective delivery of NO to fast-ventilated regions of the lungs could reduce this problem.
An alternative to NOCD is to inject a small volume of NO, added at the start of each breath. This has been learned from measurement of the gas transfer of carbon monoxide (TLCO) and allows NO to reach the resistance pulmonary arteries.11
Experimentally Induced Pulmonary Hypertension
Sixteen pathogen-free pigs weighing 36 to 60 kg were studied as previously described.12 13 Midazolam (0.3 mg/kg IM) and droperidol (0.5 mg/kg IM) premedication was followed by anesthesia with sodium pentobarbital (25 mg/kg IV). They were ventilated (Manley Ventilator, Blease Medical) with 40% O2 and 60% N2, a tidal volume of 10 to 12 mL/kg, and a maximum airway pressure of 10 mm Hg.
In Vivo Porcine Studies
In supine animals, pulmonary artery pressure (PAP), pulmonary wedge pressure (PWP), and right atrial pressure (RAP) in 9 animals, together with right carotid artery pressure (SAP), were measured. Euthanasia was effected immediately at the end with intravenous (10 mL of 1 mol/L) potassium chloride.
Isolated, Perfused Porcine Lung Studies
Isolated, salt-perfused lungs were prepared as previously described13 through a median sternotomy in 7 pigs. Euthanasia was undertaken by exsanguination. Pulmonary blood flow (Q) was measured with a Doppler flow probe and meter (model 16SB185 and model T101D, Transonic Systems Inc). The RAP, PAP, and left atrial pressure (LAP) were recorded.
The isolated lung was perfused with autologous blood and circulated to the pulmonary artery with a roller pump (model 5001R, Watson Marlow, Manchester). A 150-mL reservoir was interposed between the pump and the pulmonary artery. Perfusion rate of the lungs was slowly increased from 10 to 100 mL−1 · min−1 · kg−1. The lungs were ventilated at a tidal volume of 10 to 12 mL/kg with 20% O2, 5% CO2, and 75% N2.
Methods of NO Delivery
For the NOmin, a device14 delivered a range of volumes of a mixture of 100 ppm NO/N2 at the beginning of each breath. With a solenoid valve (M8B-3E2C-6DC, Valeader Engineering Ltd), NO/N2 was released at a flow rate of 12 L/min. The volume of gas and hence the dose delivered depended on the duration of opening times, which ranged from 10 to 1000 ms. This was calibrated before each study with a water spirometer. The valve was manual or automatic and synchronized with the start of inhalation, an automatic function triggered from the airway pressure transducer.
With NOCD, NO was delivered in a gas mixture of 79% N2 and 21% O2 (5% CO2 was added for the isolated lungs) at a concentration of 40 ppm diluted (Pneumopac Ltd) from a mixture of 10 000 ppm NO/N2.15
Endothelial NO synthase was inhibited with NG-nitro-l-arginine methyl ester (L-NAME) (1 to 2 mg/kg IV; Sigma Chemical Company Ltd). A stable rise of pulmonary vascular resistance index (PVRI) and systemic vascular resistance index (SVRI) occurred 20 to 30 minutes after injection. In the isolated lung studies, to achieve a stable elevation of PVRI after 10 to 15 minutes, the thromboxane analogue U46619 (10 pmol · L−1 · min−1, Sigma Chemical Company Ltd) was infused into the isolated lung perfusate. The experiments lasted 120 minutes.
A dose response to increasing concentrations of NOCD was undertaken in 4 animals. Concentrations of 10, 40, and 80 ppm of NO were tested in random order after the PVRI had been increased by L-NAME. Minimal dose of 100 ppm NO/N2 was tested in 4 pigs after L-NAME infusion. The valve opening time was set at 10, 20, 40, 80, 160, 320, and 1000 ms in random order. The measurements of PVRI were made over a period of 5 minutes.
In 7 isolated lungs, the PVRI after U46619 was reduced by NOCD of 40 ppm NO and NOmin with valve opening times set at 10, 80, 160, and 320 ms.
Clinical Study of Severe Pulmonary Hypertension
Sixteen patients with severe pulmonary hypertension were studied. The mean age of the patients was 44.2 (SD±13.9) years; 6 had thromboembolic pulmonary hypertension, 8 had unexplained pulmonary hypertension, 1 had pulmonary veno-occlusive disease, and 1 had pulmonary hypertension associated with sarcoidosis. All gave written consent, and the study was approved by the local hospital ethics committee. A diagnostic right heart catheter allowed measurement of mean RAP, mean PAP, and mean PWP.
Methods of NO Delivery
For the NOmin, the solenoid switch was operated manually at the start of inhalation. The patient inhaled from a 1-L reservoir bag, through a tightly fitting face mask (Nasal CPAP mask, Puritan-Bennet Corp) fitted with a pneumotachograph and differential manometer giving a record of the breathing pattern (PK Morgan, Maidstone). The reservoir was replenished with 79% N2 and 21% O2 (Pneumopac Ltd). For the NOCD, 40 ppm NO was delivered in a gas mixture in 79% N2 and 21% O2 to a 1-L reservoir bag from which the patient breathed with the tightly fitted mask.
The maximal fall in PVR with the vasodilator intravenous prostacyclin (Epoprostenol, PGI2)16 was compared with NOCD. For the NOmin, the duration of the spike of NO/N2 (100 ppm) was varied until a minimum dose caused a fall in PVRI of ≥10%.
Calculations and Statistical Analysis
The PVR was calculated by dividing the pressure difference across the lungs [PAP−LAP or PWP]/pulmonary blood flow (Q), and SVR [SAP−RAP]/Q. The PVRI and SVRI were standardized to body weight in the experimental studies (mm Hg · L−1 · min−1 · kg−1) and in clinical studies to body surface area (mm Hg · L−1 · min−1 · m−2).
The mean values for PVRI were calculated with standard errors (SEM). ANOVA and Fisher’s test for multiple comparison were undertaken to compare treatments. Paired Student’s t tests were performed to compare baseline and post–L-NAME or post-U46619 PVRI values.
The 10 ppm NOCD was a less effective vasodilator than 40 and 80 ppm of NOCD (Figure 1⇓) in 5 animals after L-NAME increased the PVRI by 6.5±1.1 mm Hg · mL−1 · min−1 · kg−1.
In 4 animals, L-NAME caused a rise of PVRI from 4.3±0.8 to 10.0±1.6 mm Hg · mL−1 · min−1 · kg−1. The NOmin at volumes of 8 to 38 mL produced the same fall in PVRI as did 115 mL (Table 1⇓), but the SVRI was unaffected by NO.
The thromboxane analogue U46619 (10 pmol · L−1 · min−1) caused an increase in PVRI from 4.6±0.8 to 12.2±1.3 mm Hg · mL−1 · min−1 kg−1. The NOCD caused a fall of PVRI to 66.5%, and similar reductions occurred with NOmin at volumes of 3 to 38 mL of 100 ppm NO in N2 (Figure 2⇓).
The mean PVRI was 29.2±3.7 (mean±SE) mm Hg · L−1 · min−1 · m−2, which fell to 24.0±3.1 with NOmin (8 to 38 mL of 100 ppm) and to 24.5±3.3 with 40 ppm of NOCD (Table 2⇓). The average tidal volume varied from 320 to 730 mL. PGI2 caused a comparable fall of mean PVRI to 20.0±2.6 mm Hg · L−1 · min−1 · m−2. Unlike inhaled NO, PGI2 also caused a fall in SVRI (Table 2⇓). The volumes of gas delivered by the solenoid valve ranged from 3 to 115 mL (Table 1⇑).
A small volume of NO/N2 of up to 38 mL, delivered at the beginning of the breath, was as effective a vasodilator as a 40-fold higher dose of NOCD. Conventional delivery required 5.3×10−7 to 1.2×10−6 mol/L per breath to achieve comparable pulmonary vasodilatation as 1.2×10−8 to 1.6×10−7 mol/L of the spike.
Pulmonary hypertension is reduced by inhaled NO acting on the precapillary arteries,12 located anatomically within the pulmonary acini. Inhaled NO must therefore reach the alveolar region at a sufficient concentration. We can learn from the measurement of gas diffusion (TLCO) with carbon monoxide. Being similar to CO, NO is also used to measure diffusion.17 18 19 The use of a small volume of CO added at the start of the inhalation provides an equivalent measure of TLCO.11 The rate of diffusion of CO (or NO) from this inhaled bolus is slower, for example, 4×10−4 · mL−1 · min−1 (400 ppb/min at 37.0°C)20 than the rate of convective flow of the inhaled air into the lungs, for example, 7500 mL/min.21 Little change in the concentration of NO (or CO) in the bolus is expected to occur until the alveoli are reached.
The practical advantage of NOmin is that the dose of inhaled NO depends on the frequency of breathing, for example, increasing with exercise. With NOCD it is not possible to match the flow rate of the gas mixture of NO to the patient’s rate of ventilation unless the patient is supported by a mechanical ventilator. The concentration of NO in the spike never will exceed that in the cylinder. It is not necessary to mix with respiratory gases or to monitor the concentrations of inhaled NO. Furthermore, by comparison with NOCD, slow diffusion from the spike reduces the oxidation of NO to NO2,6 therefore careful monitoring for a buildup of NO24 is not needed.
The delivery of a spike of NO greatly reduces the gas waste compared with continuous inhalation, in which 50% of the gas is lost during expiration. We estimate that for a 10-mL spike of NO/N2 to be delivered, in each breath only 216 L of NO/N2 would be needed each day. A small container the size of a hip flask could be easily carried by the patient, considerably reducing the daily volume of gas needed to treat primary pulmonary hypertension.7
The lowest effective vasodilatory concentration of inhaled NO remains to be determined. The NOCD systems give concentrations of NO in the inspirate of 1 to 120 ppm.22 To relax pulmonary arteries in vitro requires 2 ppm of NO gas.23 Most of the inhaled NO combines with oxyhemoglobin of red blood cells to form methamoglobin.24 This indicates a considerable redundancy in the NOCD dose.
Finally, by inhaling NO as a bolus at the beginning of inspiration, it is distributed only to fast-ventilated regions of the lungs.25 Slow-filling lung units will receive less NO than with NOCD. This should avoid the worsening gas exchange in COPD seen with NOCD,9 10 as with inhaled bronchodilators.25 This could extend treatment with inhaled NO to patients with COPD.
In summary, we describe a means of reducing the effective dose per breath of NO by 40-fold when compared with conventional delivery and that lessens potential toxicity26 as well as offering a safe means of treating ambulatory patients. It should overcome the problems of worsening VA/Q inhomogeneity, to allow inhaled NO to be used for COPD.
This work was supported by the HC Roscoe award of the BMA and the British Heart Foundation grant PG93/94043.
- Received June 9, 1998.
- Revision received July 22, 1998.
- Accepted July 30, 1998.
- Copyright © 1998 by American Heart Association
Katayama Y, Higenbottam TW, DeAtauri MJD, Cremona G, Akamine S, Barbera JA, Rodriguez-Roisin R. The effect of inhaled nitric oxide on gas exchange in patients with chronic obstructive pulmonary disease and severe pulmonary hypertension. Thorax. 1997;52:120–124.
Wagner WW, Latham LP, Brinkman PD, Filley GF. Pulmonary gas transport time: larynx to alveolus. Science. 1969;163:1210–1211.
Cremona G, Higenbottam T, Takao M, Bower EA, Hall LW. Nature and site of action of endogenous nitric oxide in vasculature of isolated pig lungs. Am J Physiol. 1997;82:23–31.
Higenbottam TW. Nitric oxide treatment. International patent. 1993, applied for and British patent 94/20,504.4, granted 1997.
Higenbottam TW, Spiegelhalter D, Scott JP, Fuster V, Dinh-Xuan AT, Caine N, Wallwork J. The value of prostacyclin (epoprostenol) and heart-lung transplantation as treatments for severe pulmonary hypertension. Br Heart J. 1993;70:366–370.
Borland CDR, Higenbottam TW. A simultaneous single breath measurement of pulmonary diffusing capacity with nitric oxide and carbon monoxide. Eur Respir J. 1989;2:56–63.
Stewart A, Allot PR, Cowles AL, Mapleson WW. Solubility coefficients for inhaled anaesthetics for water, oil, and biological media. Br J Anaesth. 1973;45:282–293.
Cussler EL. Values for diffusion coefficients. In: Diffusion. Mass Transfer in Fluid System. Cambridge, UK: Cambridge University Press; 1994:105–145.
Lugliana R, Whipp BJ, Seard C, Wasserman K. Effect of bilateral carotid-body resection on the ventilatory control at rest and during exercise in man. N Engl J Med. 1971;285:1105–1111.
Stenqvist O, Kjelltoft B, Lundin S. Evaluation of a new system for ventilatory administration of nitric oxide. Acta Anaesthesiol Scand. 1993;687–691.
Demoncheaux EAG, Smith APL, Davies M, Higenbottam TW. Is nitrite an important physiological circulating nitric oxide (NO) donor? J Physiol. 1996;491:101P.
Wennmalm A, Benthin G, Edlund A, Jungersten L, Kieler-Jensen N, Lundin S, Westfelt UN, Petersson AS, Waagstein F. Metabolism and excretion of nitric oxide in humans: an experimental and clinical study. Circ Res. 1993;73:1121–1127.
Warren JB, Higenbottam TW. Caution with use of inhaled nitric oxide. Lancet. 1996;348:629–630.NO is conventionally inhaled mixed with respiratory gases at a fixed concentration throughout the breath. A bolus of 8 to 38 mL of NO/N2 given at the onset of inspiration produces equivalent vasodilatation but with a 40-fold reduction of dose.