INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION APPENDIX REFERENCES

THE RESPIRATORY TRACT PRODUCES nitric oxide (NO) in the nasal pharynx, the conducting airways, and the alveoli (10, 13, 24). Recently developed methods permit determination of the rate of NO production from these three locations (10, 24, 30). Representative values in normal subjects are 0.4 µl/min for the nose and nasal pharynx (10, 11), 0.08 µl/min from the conducting airways (24, 30), and 0.2 µl/min from the alveoli (24). The most popular method to estimate NO production by the conducting airways is based on measurement of expired NO concentration at constant expiratory flow rates from 50 to 200 ml/min while the subject exhales against a positive pressure of 5-20 cmH₂O (26). The positive pressure excludes contamination of the expired samples from NO in the nasal pharynx. At these relatively slow expiratory flow rates, the expired NO concentrations mainly reflect the NO production by the conducting airways, with little contribution from the alveoli. Airway inflammation from bronchial asthma can cause dramatic increases of conducting airway NO (26). Measurements of expired NO have attracted interest as a means to monitor airway inflammation and investigate the mechanisms controlling NO production by the conducting airways (26).

Determination of NO production by the alveoli (A_NO) is more complex. It requires measurements of expired NO concentration collected while the subject expires against a positive pressure of 5-20 cmH₂O at a number of different constant expiratory flow rates (24, 29, 30). These measurements permit extrapolation to the NO concentration present at an infinitely fast expiratory flow rate that is free of any contribution from the conducting airways to the expired air. This extrapolated value represents the alveolar NO partial pressure (PA_NO) present during a steady state without entry of NO into the alveoli from the conducting airways. PA_NO multiplied by the pulmonary NO diffusing capacity (DL_NO) equals A_NO (15) or

(1)

To date, A_NO has only been measured in normal human subjects (24), but it is suspected to be elevated in some diseases, such as cirrhosis of the liver (19), while decreased in other diseases, such as primary pulmonary hypertension (8).

Measurements of A_NO require a measurement of DL_NO (Eq. 1). In humans, several groups have measured DL_NO with a modification of the carbon monoxide (CO) breath-holding, single-breath method (Dsb_CO) (2, 3, 12, 18). Short breath-holding times of 3-8 s were required because the rapid clearance of NO from the alveoli prevented accurate measurements of the small expired NO signal after 10 s of breath holding. The shorter breath-holding periods make it difficult to determine accurately the breath-holding time interval. In the past decade, rapidly responding, highly sensitive chemiluminescent NO analyzers have become available that permit continuous breath-by-breath measurements of NO concentration during the ventilatory cycle.

The purpose of this report is to see if these analyzers can be applied to make accurate measurements of DL_NO suitable for calculating A_NO. We describe a single exhalation method and rebreathing method that do not require a concurrent measurement of an inert, relatively insoluble gas, such as helium or methane. Compared with the breath-holding techniques, advantages include the use of lower concentrations of NO in the order of 5-10 ppm and the elimination of errors from estimating the breath-holding time interval. These methods are also compared with Dsb_CO obtained in the same subjects.

Glossary

CV	Coefficient of variation
DexCO	Pulmonary CO diffusing capacity measured during a single, constant maximum exhalation
DexNO	Pulmonary NO diffusing capacity measured during a single, constant maximum exhalation
DL	Pulmonary diffusing capacity; test gas not specified
DLCO	Pulmonary CO diffusing capacity; method of measurement not specified
DLNO	Pulmonary NO diffusing capacity
DrbCO	Pulmonary CO diffusing capacity measured during rebreathing CO
DrbNO	Pulmonary NO diffusing capacity measured during rebreathing NO
DsbCO	Breath-holding, single-breath CO diffusing capacity
DsbNO	Breath-holding, single-breath NO diffusing capacity
PANO	Partial pressure of NO in alveoli
PNOex	Expired concentration of NO
PNOex	Minimal partial pressure of NO that can be present during measurements of DexNO
PNOrb	Minimal partial pressure of NO that can be present during measurements of DrbNO
VA	Alveolar volume
ANO	NO production by the alveoli

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION APPENDIX REFERENCES

Measurement of NO. Details of the method to measure NO have been previously described (11, 24). A rapidly responding chemiluminescent NO analyzer (model 270B, Sievers, Boulder, CO), operating at a sample rate of 250 ml/min, continuously measured exhaled levels of NO at the mouthpiece. Response time of the analyzer was <200 ms for a signal of 90% full scale with a lag time of 0.8 s. The analyzer was adjusted to provide 40 measurements of the NO concentration per second that could be averaged over any time interval. The NO analyzer was calibrated daily by serial dilution of a gas containing 229 parts per billion (ppb) of NO. To obtain reference gas samples free of NO, air from a gas cylinder containing <2 ppb of NO (Scott Specialty Gases, Plumsteadville, PA) was passed through a filter packed with potassium permanganate (Purafil, Thermoenvironmental Instruments, Franklin, MA).

6

6

Single, rapid, maximal exhalation at constant flow rate for measuring DL_NO. After resting for 5 min in the sitting position, each subject first exhaled to residual volume (RV) through the mouthpiece of the apparatus attached to a four-way valve connected to room air (Fig. 1). The valve was turned 90°, and the subject rapidly inhaled 5-10 ppm of NO from the bag-in-box to total lung capacity, breath held for 2-3 s, and then exhaled into the spirometer at an expiratory flow rate of 460 ± 80 (SD) ml/s (range: 370-560 ml/s) to RV. For most of the measurements, the constant expiratory flow rate was facilitated by narrowing the expiratory line to the spirometer with a cork penetrated by an open tube with a cross-sectional area of 12 mm². The subject was instructed to maintain an expiratory pressure of +5 cmH₂O observed on the water manometer attached at the mouthpiece. The concentration of NO at the mouthpiece and changes in lung volume were recorded (Fig. 2). DL_NO measured during a single, constant maximum exhalation (Dex_NO) was calculated by using a modification of the method described by Cotton and coworkers (6). At any instant, the amount of NO leaving the alveolar volume (VA) equals the amount of NO diffusing into the blood or

(2)

where Dex_NO is recorded in ml · min¹ · mmHg¹, PNO_ex is the partial pressure of NO at any instant in the expired gas measured at the mouth in mmHg, VA is expressed in ml STPD, PB is barometric pressure, and PNO_ex is the minimal partial pressure of NO that can be present during the measurement whose subtraction corrects for NO production by the alveoli. Integration gives

(3)

where PNO_ex1 and PNO_ex2 are the initial and final values of PNO_ex during the time interval t₂  t₁ recorded in seconds during the exhalation. The computer calculated Dex_NO every 0.025 s and then averaged this value with the previous 19 values, so that a value of Dex_NO was recorded as a 0.5-s moving average plotted every 0.025 s (Fig. 3). A single value for Dex_NO for each exhalation was then calculated from the mean of these values for Dex_NO, which was obtained after the initial data were discarded that was measured during exhaling of a volume equal to four times the total dead space, which was calculated as the sum of the subject's estimated anatomic dead space and the instrument's dead space of 100 ml. The subject's anatomic dead space was assumed to equal the subject's ideal body weight in pounds and was expressed in milliliters (4). The final 15% of the exhaled vital capacity was also discarded.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Diagram of apparatus used to measure nitric oxide (NO) diffusing capacity with single, constant exhalations (DexNO) and rebreathing (DrbNO). For measuring DexNO, the subject inhaled 5-10 parts/million (ppm) of NO from the bag-in-box, breath held for 2-3 s, and then exhaled through the orifice of the cork while maintaining +5 cm of pressure in the manometer. The cork and manometer result in a constant expiratory flow rate of ~0.5 l/s. To measure DrbNO, the cork and manometer were removed, and subjects rebreathed 5-10 ppm of NO placed in the bag, emptying the bag at the end of each inspiration. For details see text.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Recording of NO signal and change in lung volume during the measurement of DexNO. The subject rapidly inhaled 3.3 liters of NO-enriched air, breath held for 3 s, and then exhaled at a constant flow rate of 370 ml/s. DexNO was calculated from the data collected after discarding an initial volume equal to 4 times the subject's and instrument's dead space (4 × DS) and the final 15% of the expired vital capacity (15% VC).

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Computer-generated values of DexNO vs. exhaled lung volume calculated from the data shown in Fig. 2. The mean value for DexNO was obtained from the average of recorded values after discarding an initial volume equal to the instrument's and subject's dead space (4 × DS) and the final 15% VC. To determine the effects of lung volume on DexNO, the value recorded during the first 20% of the measurement was compared with the last 20%.

Estimation of the back pressure or PNO_ex. The subjects repeated the same maneuver used to measure Dex_NO with the bag-in-box filled with room air containing <15 ppb of NO. PNO_ex was calculated from the mean value of the expired NO concentration that was recorded after the initial NO signal was discarded that was obtained while the subject exhaled a volume equal to four times the total dead space calculated as described above, as well as the final 10% of the expired volume. For six of the subjects, measurements of PNO_ex were made at several different constant expiratory flow rates above and below the flow rate used to measure Dex_NO. This allowed extrapolation to a value of PNO_ex that matched the expiratory flow rate present during the measurement of Dex_NO.

Rebreathing method for measuring DL_NO. The rebreathing maneuver was performed by first filling the bag-in-box with 5-10 ppm of NO in air at a gas volume of ~1 liter less than the subject's vital capacity. The subject inserted the mouthpiece and exhaled into the room to RV. The valve at the mouthpiece was turned to the bag, and the subject inspired the contents of the bag and then rebreathed 1.6 ± 0.3 liters (STPD) at a constant rebreathing rate of 23 ± 4 breaths/min, emptying the bag with each breath for 20-30 s (Fig. 4, top and middle). The computer transformed the NO signal to a plot of the natural logarithm of the partial pressure of NO at the mouth less the minimal partial pressure of NO present during rebreathing (PNO_rb) vs. time (Fig. 4, bottom), where PNO_rb is calculated as described below. After one to three breaths, there is a linear decrease in the logarithm of this NO signal for the subsequent six to eight breaths, and these data were used to calculate rebreathing DL_NO Drb_NO. Data collected after 20 s of rebreathing were usually discarded because of the low signal-to-noise ratio at NO concentrations <30 mmHg × 10⁶ (Fig. 4, bottom). The operator measured the bag concentration (P_bag) recorded at the end of inspiration and the end-expiratory concentration (PNO_ex) of these breaths. The computer calculated the slope of two parallel lines fitted to P_bag  PNO_rb and PNO_ex  PNO_rb (k_NO) expressed as a fractional change in concentration per minute. From two parallel lines drawn through P_bag  PNO_rb and PNO_ex  PNO_rb, the ratio of P_bag  PNO_rb to PNO_ex  PNO_rb at any instant (H) was calculated (Fig. 4). Drb_NO was then calculated with an equation similar to equations derived by Hook and Meyer (14) and Meyer and coworkers (21) that were used for calculating rebreathing measurements of oxygen, CO, and NO diffusing capacity (see APPENDIX for derivation)

(4)

where V_tot is the total gas volume in ml (STPD) in the rebreathing circuit, which consists of the subject's RV, the gas volume of the initial inspiration, and the instrument's dead space of 100 ml. _eff is effective ventilation in ml/min during rebreathing and was calculated as the mean volume of the rebreathing breaths less the instrument's dead space of 100 ml and the subject's dead space, which is estimated to be 25% of the rebreathing volume (23) multiplied by the rebreathing rate in breaths/min.

View larger version (40K): [in this window] [in a new window]   Fig. 4.   Measurement of DrbNO. Top: NO concentration at the mouthpiece during rebreathing of air enriched with 6,000 ppb of NO. Middle: changes in lung volume during rebreathing. The subject inspired 4.6 liters and then rebreathed 1.6 liters for 34 s. Bottom: the oscillating signal is the computer-processed NO concentration recorded as the difference between NO concentration at the mouth (PNO) less the minimal partial pressure of NO present during rebreathing (PNOrb). This signal is plotted as the logarithm of PNO  PNOrb vs. time. After the first 2 breaths were discarded, the computer fitted 2 straight, parallel, dashed lines to the inspired concentration from the breathing bag (top dashed line) and the end-expiratory concentration (bottom dashed line). The slope of these lines (kNO) and the ratio of the concentration of the 2 dashed lines at any instant (H) were used to calculate DrbNO with Eq. 4. In this subject, H can be calculated at time 0 and equaled 9.0 × 103  3.2 × 103, or 2.8.

Estimation of the PNO_rb. PNO_rb during rebreathing was measured by having the subject perform the same rebreathing maneuver for 15-20 s with the bag-in-box filled with room air. After 5-10 s of rebreathing, the NO concentration reaches a constant value equal to PNO_rb (see Fig. 4 in Ref. 11). The value for PNO_rb was measured from the mean NO concentration recorded at the mouth during the last 3 s of rebreathing after discarding the final exhalation during rebreathing (11). In seven of the subjects, the rebreathing maneuver to measure PNO_rb was performed at ventilatory rates above and below the value present for measuring Drb_NO. These measurements permitted extrapolation to a value for PNO_rb that matched the ventilatory rate present during the measurement of Drb_NO.

Effects of alveolar oxygen tension on Dex_NO. Five of the subjects performed measurements of Dex_NO at end-expiratory oxygen tensions varying between 42 and 570 Torr. End-expiratory O₂ was measured with an oxygen sensor (AG-17 O₂ sensor, Ceramatec, Salt Lake City, UT; and TED200-TX microprocessor, Teledyne, City of Industry, CA) installed in the expiratory line of the apparatus (Fig. 1). To vary the end-expiratory partial pressure of O₂, the bag-in-box was filled with O₂ concentrations varying from 1 to 99% before the NO was added just before the measurement of Dex_NO. To obtain end-expired O₂ tensions in excess of 500 Torr, the subject performed three slow vital capacity maneuvers while inhaling 100% oxygen before inhaling 5-10 ppm NO in 99% oxygen. Dex_NO was calculated as described above.

Effects of VA on Dex_NO and Drb_NO. Effects of changes in VA on Dex_NO were determined by comparing its value obtained during the first 20% of the expirate used to calculate Dex_NO to the final 20% (Fig. 3). Six of the subjects decreased their VA during measurements of Drb_NO by starting the initial inspiration from RV, with the rebreathing bag containing a volume reduced to 2-3 liters less than the subject's vital capacity.

Measurement of Dsb_CO. Dsb_CO was measured by the technique of Jones and Meade (16) using automated equipment (P. K. Morgan, Haverhill, MA). Dsb_CO was multiplied by 5 to provide an estimated value of NO diffusing capacity (5 × Dsb_CO). The factor 5 was chosen on the basis of published reports showing a ratio of NO diffusing capacity to CO diffusing capacity breathing air ranging from 4.3 to 5.3 (2, 3, 12, 18, 28, 31).

Measurement of A_NO. A_NO is the product of PA_NO and the diffusing capacity for NO (Eq. 1). PA_NO for seven of the subjects was measured with the technique described by Pietropaoli and coworkers (24), where subjects performed a series of exhalations at different constant expiratory flow rates after inspiring room air and breath holding for 10-15 s. The faster the exhalation, the less is the contribution of the NO produced by the conducting airways to the NO originating from the alveoli (PA_NO). The expired NO concentration in a series of expirations at different expiratory flow rates permits extrapolation to the concentration of NO at an infinite expiratory flow rate. This value was considered to be PA_NO free of contamination by NO from the conducting airways. A_NO was then calculated with Eq. 1.

Subjects. Measurements of Dex_NO, Drb_NO, and Dsb_CO were made in nine healthy, nonsmoking subjects ranging in age from 31 to 72 yr (mean 46 ± 18 yr). Height was 173 ± 9 cm, and weight was 73 ± 11 kg. Six were men, and three were women. All subjects were free of cardiopulmonary disease and respiratory symptoms. Spirometry showed values of 90% of predicted (7) for the forced expiratory volume in 1 s with a mean value of 106 ± 14% of the predicted forced expiratory volume in 1 s. The study was approved by the University of Rochester's Research Subjects Review Board.

Statistical methods. Results are given as means ± SD. In experiments in which subjects served as their own control, results were compared using a two-tailed paired t-test. Groups of subjects were compared with an unpaired t-test. A P value < 0.05 was required for statistical significance.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION APPENDIX REFERENCES

Values for Dex_NO, Drb_NO, and Dsb_CO. Figure 5 shows the individual values for Dex_NO, Drb_NO, and Dsb_CO multiplied by 5 in the nine subjects. Dex_NO equaled 125 ± 29 ml · min¹ · mmHg¹, Drb_NO equaled 122 ± 26 ml · min¹ · mmHg¹, and 5 × Dsb_CO equaled 135 ± 36 ml · min¹ · mmHg¹. The 3% difference between Dex_NO and Drb_NO reached statistical significance (P = 0.046). Dex_NO was 4.7 ± 0.6 times greater than Dsb_CO, and Drb_NO was 4.6 ± 0.6 times greater than Dsb_CO.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Values for DexNO, DrbNO, and 5 times the carbon monoxide breath-holding, single-breath diffusing capacity (5 × DsbCO) for 9 normal subjects. DLNO, pulmonary lung diffusing capacity of NO. Symbols are as follows: , subject MC (31 yr, male); , subject AP (33 yr, male); , subject AT (33 yr, male); , subject JB (34 yr, male); , subject PP (72 yr, male); , subject RH (68 yr, male); , subject CG (32 yr, female); , subject IP (32 yr, female); , subject SH (65 yr, female).

Intraday and interday variability of Dex_NO and Drb_NO. Table 1 shows interday and intraday variability of Dex_NO and Drb_NO expressed as the coefficient of variation (CV). Intraday CV in the nine subjects for Dex_NO was 4.4 ± 3.9% and for Drb_NO was 3.4 ± 1.9%. Interday CVs were slightly larger at 8.1 ± 4.9% for Dex_NO measured in seven of the subjects and 13.6 ± 10.1% for Drb_NO measured in six of the subjects.

Influence of alveolar oxygen tension on Dex_NO. Figure 6 shows values for Dex_NO at end-expiratory O₂ tensions that varied from 42 to 682 Torr in five of the subjects. Dex_NO showed no consistent change with different oxygen tensions. For example, the mean values for the five subjects obtained <100 Torr at 60 ± 10 Torr equaled 126 ± 30 ml · min¹ · mmHg¹ compared with 128 ± 31 ml · min¹ · mmHg¹ for all measurements at values >450 Torr obtained at 557 ± 68 Torr and did not differ significantly (P = 0.14).

View larger version (18K): [in this window] [in a new window]   Fig. 6.   DexNO at different end-expired PO2 in 5 subjects. DexNO obtained at PO2 < 100 Torr did not differ from values measured at PO2 > 450 Torr. See Fig. 5 legend for definition of symbols.

Effects of changes in VA on Dex_NO and Drb_NO. Figure 7A shows values for Dex_NO determined from the first 20% and the final 20% of the expirate used to determine PNO_ex in nine subjects. Dex_NO fell from 138 ± 35 to 100 ± 21 ml · min¹ · mmHg¹, or 28% (P = 0.001), with a change in VA from 6,578 ± 1,508 to 4,714 ± 1,112 ml BTPS. Figure 7B shows the changes for Drb_NO performed at VA values of 6,186 ± 791 and 4,682 ± 373 ml BTPS in six of the subjects. Drb_NO fell from 127 ± 30 to 107 ± 21 ml · min¹ · mmHg¹, or 19% (P = 0.050).

View larger version (18K): [in this window] [in a new window]   Fig. 7.   A: changes in DexNO at different lung volumes in 9 subjects. B: changes in DrbNO at different lung volumes in 6 of the subjects. M, male; F, female.

A_NO in humans. In seven of the subjects, A_NO was calculated from the product of PA_NO with Dex_NO, Drb_NO, and 5 × Dsb_CO (Fig. 8). PA_NO equaled 1.8 ± 0.7 mmHg × 10⁶. The respective values of A_NO were 0.21 ± 0.06, 0.20 ± 0.06, and 0.22 ± 0.06 µl/min. The small difference between A_NO calculated with Dex_NO and Drb_NO was significant (P = 0.009).

View larger version (16K): [in this window] [in a new window]   Fig. 8.   Alveolar NO production calculated using DexNO, DrbNO, and 5 × DsbCO. Symbols are as defined in Fig. 7A.

PNO_ex and PNO_rb or back pressure. PNO_ex equaled 3.8 ± 1.7 mmHg × 10⁶ and PNO_rb equaled 2.6 ± 0.7 mmHg × 10⁶. If PNO_ex and PNO_rb were assumed to equal zero, Dex_NO was underestimated by 2.3 ± 1.6% (P = 0.002) and Drb_NO by 1.7 ± 0.9% (P = 0.0002).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION APPENDIX REFERENCES

These data show that rapidly responding chemiluminescent NO analyzers, combined with simple pulmonary function equipment (Fig. 1), permit reproducible measurements of DL_NO in normal subjects using either single exhalations (Dex_NO) or rebreathing (Drb_NO). A separate measurement of the subject's RV with techniques such as body plethysmography or helium gas dilution is required. This measurement obviates the complexity of continuous analysis of an inert, insoluble gas such as helium with mass spectrometry, which was employed by previous investigators using single exhalations or rebreathing methods (6, 21, 28, 31). These are the first measurements of DL_NO that take into account the NO back pressure to diffusion resulting from NO production by the lungs.

Comparison of breath-holding, single-breath NO diffusing capacity and Dex_NO to measurements reported by others. Our values for Dex_NO of 125 ml · min¹ · mmHg¹ and Drb_NO of 122 ml · min¹ · mmHg¹ obtained in normal subjects are similar to measurements reported by others using breath holding [single-breath NO diffusing capacity (Dsb_NO)] and constant exhalations (Dex_NO). Borland and Higenbottam (3) in one series obtained values for Dsb_NO of 147 ml · min¹ · mmHg¹ in 13 normal subjects and 126 ml · min¹ · mmHg¹ in another group of 10 normal subjects (2). Guenard and coworkers (12), using 3-s breath holdings in normal subjects, obtained values for Dsb_NO of 136 ml · min¹ · mmHg¹ in 14 subjects and 140 ml · min¹ · mmHg¹ in another series of 11 subjects (18). Tsoukias and coworkers (28, 31) recently reported values for Dex_NO of 130 ml · min¹ · mmHg¹ at a VA of 5.4 liters (BTPS) in seven normal men aged 28 ± 4 (SD) yr. Values for Dsb_NO in the same subjects shown in a figure in their report (28) appear to be slightly lower at the same VA.

Influence of NO back pressure on measurements of NO diffusing capacity. Production of NO by mammalian airways and its presence in exhaled air were not recognized until 1991 (13). Therefore, previous measurements of NO diffusing capacity by breath holding in humans and rebreathing in animals did not take this NO back pressure into account. Unlike for CO, NO back pressure has two components. First is the production of NO by the pulmonary capillary endothelium and the alveolar-capillary membrane that results in a PA_NO in the order of 2 × 10⁶ mmHg (24, 30). Second, during exhalation this NO is added to the NO production by the conducting airways. This concentration from the conducting airways is highly dependent on expiratory flow rates (24, 26, 30), but typically at flow rates of 500 ml/s it equals ~4 × 10⁶ mmHg (24). Alveolar NO during the measurement of Dex_NO or Drb_NO falls rapidly, with a half-time in the order of 2-3 s; thus the background pressure can make an appreciable contribution to the expired NO concentration later in the maneuver. Our measurements of Dex_NO and Drb_NO performed with 5-10 ppm of NO would be underestimated by ~2% if the back pressure were ignored. If the inspired NO concentration were decreased threefold to 2-3 ppm, the estimated error from ignoring NO back pressure would increase to 5% for Dex_NO and 3% for Drb_NO. Initially, inhaling a fourfold higher concentration of NO of 40 ppm reduces the estimated error to 0.6% for Dex_NO and 0.4% for Drb_NO. Inhaling 80 ppm of NO reduced the error at 40 ppm in half. Therefore, inhaling 40-80 ppm of NO instead of the 5-10 ppm employed in our normal subjects would make the error from ignoring NO back pressure trivial. However, subjects with elevated levels of exhaled NO, such as observed in asthmatic subjects, may require measurements of NO back pressure to avoid underestimation of NO diffusing capacity, even with the initial inhalation of 40-80 ppm of NO.

Effect of changes in lung volume on Dex_NO, Drb_NO, and Dsb_NO. Borland and Higenbottam (3) measured the effects of decreasing lung volume on Dsb_NO in five of their subjects. A 44 ± 9% decrease in VA decreased Dsb_NO by 29 ± 11%, resulting in a slope of 0.63 ± 0.18 for the decrease in Dsb_NO vs. the decrease in VA. This value is similar to the slope for Dex_NO vs. VA of 0.95 ± 0.44 and Drb_NO vs. VA of 0.45 ± 0.45 observed in our subjects. Recently, Tsoukias and coworkers (28, 31) completed a sophisticated analysis of the effects of VA and sequential filling on Dex_NO. For the mean change of VA from 6.6 to 4.7 liters shown in Fig. 8 of their report (28), seven normal subjects decreased Dex_NO from 132 to 116 ml · min¹ · mmHg¹, resulting in a slope of 0.97. This value is in close agreement with the slope of 0.95 for VA vs. Dex_NO found in our nine subjects.

Cause of lack of effect of changes in alveolar oxygen tension on Dsb_NO and Dex_NO. Borland and coworkers (3) observed no change in Dsb_NO in five of their subjects when they varied alveolar oxygen concentration from 19 to 69%, whereas Dsb_CO showed the expected fall from 34 to 20 ml · min¹ · mmHg¹. We measured Dex_NO over a wider range with end-expired oxygen concentrations varying from 6 to 96% and also observed no significant change in Dex_NO. Lack of change in Dsb_NO and Dex_NO with the changes in oxygen tension has several explanations. First, unlike for CO, the very rapid rate of reaction of NO with oxyhemoglobin is similar to the reaction rate with reduced hemoglobin. The bimolecular rate constant for the combination of NO with oxyhemoglobin is reported to be 3.4 × 10⁷ M/s vs. 2.2 × 10⁷ M/s with reduced hemoglobin (17). Therefore, the rate of uptake of NO by red blood cells should not be markedly altered by changes in pulmonary capillary oxygen saturation and, if anything, should increase with higher oxygen saturations. Second, as pointed out by Morris and Gibson (22), the rate of reaction of NO with hemoglobin "is so high that effectively every molecule of NO which enters the reaction radius is captured by a heme group. The observed rate would then be a measure of diffusion to the site." Placing these concepts in the terminology developed by Roughton and Forster (25) for CO diffusion in the lungs, the uptake of NO by the pulmonary capillaries can be divided into its extra erythrocyte diffusion component and a reactive or intraerythrocyte component, V_c × _NO, where V_c is the pulmonary capillary blood volume and _NO is the rate of combination of NO with the hemoglobin in 1 ml of blood measured in vitro. According to Morris and Gibson (22), _NO has a finite value because of the time required for the advancing front of NO to travel within the blood cell to the combining site on the hemoglobin molecule. Because the chemical reaction of NO with hemoglobin is limited by diffusion to the combining site within the red blood cells but not by the chemical reaction with hemoglobin, the modest differences in the rate of combination of NO with O₂ hemoglobin compared with reduced hemoglobin should not alter _NO. As a result, DL_NO measured at different alveolar O₂ concentrations remains constant.

Effects of parallel heterogeneity of lung VA, ventilation, and DL on Drb_NO and Dex_NO. Meyer and coworkers (21) performed an elegant analysis of the effects of uneven distribution of VA, ventilation (E), and DL on measured values of Drb_NO and rebreathing CO DL (Drb_CO) in dogs. In their model containing two parallel compartments, uneven distribution of VA between the compartments with the same DL-to-E ratios caused overestimations of Drb_NO and Drb_CO, whereas uneven distribution of E and DL resulted in underestimations. Most striking was their finding that all patterns of uneven distribution caused approximately twice as large an error in Drb_NO compared with Drb_CO.

Selection of method to measure DL_NO for calculating A_NO. Three practical methods are now available to measure DL_NO. They are single-breath measurement (Dsb_NO) with breath holdings usually between 3 and 8 s, the single, constant exhalation method (Dex_NO), and the rebreathing method (Drb_NO). Dsb_NO has the advantage of off-line measurements of NO with analyzers with slow response times but requires analysis of an inert gas such as helium in the inspired gas and the expired alveolar sample. Accurate measurements of the short breath-holding time are critical, and, with longer breath holds, the resulting lower exhaled NO concentration is difficult to measure accurately. We found that Drb_NO required considerable training even in normal subjects to obtain constant rebreathing rates and sufficient inspiratory effort with each breath to completely empty the rebreathing bag. For the measurement of Dex_NO, the constant exhalation rate obtained by maintaining a mouth pressure of +5 cmH₂O against a fixed resistance in the expiratory circuit was easily achieved by normal subjects. We avoided higher expiratory pressures because they reduce CO diffusing capacity and, by inference, Dex_NO (27). Dex_NO in our hands has proven to be a practical method for calculating A_NO in ongoing studies evaluating the potential toxic effects of inhaled particulates on the lungs. Dex_NO has the added advantage of permitting measurements of NO diffusing capacity at different lung volumes (Fig. 3). This may contribute useful information for detecting uneven distribution of DL_NO within the lung. Whereas Dex_NO and Drb_NO do not require measurement of an inert insoluble gas during the maneuver, a separate measurement of RV is needed. All three methods have similar intraday and interday CV, except for Drb_NO, which had a greater interday CV than the other methods (Refs. 3, 12; Table 1).