INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nitric oxide (NO) is formed in biological systems from L-arginine and oxygen by a family of enzymes, the nitric oxide synthases (NOSs). Of the three subclasses of NOS, types I and III can be distinguished by their constitutive activity (cNOS), whereas type II is an inducible isoform (iNOS) (1). Human airways have been shown to contain all three NOS isoforms (4). Type II NOS, the inducible form, has been localized to the airway and alveolar epithelium, the vascular endothelium, and alveolar macrophages, whereas type I is present in human airway nerves and type III is found in lung epithelial cells and the vascular endothelium.

NO has been detected in the expired gas of normal individuals; higher levels of NO have been identified in the expired gas of patients with asthma (8). Glucocorticoids have been shown to inhibit type II NOS expression in vitro (13) and to lower expired NO levels in asthmatic individuals experiencing an acute exacerbation (12). Furthermore, treatment with glucocorticoids lowers expired NO in patients with stable asthma (14). These findings have led to the speculation by several investigative groups that expired NO may be either a marker for, or mediator of, airway inflammation (12, 15, 16). In this regard, a study has reported that expired NO rises after inhalational allergen challenge (16).

Isocapnic cold air hyperventilation (ISH) has also been used as an experimental model of exercise-induced asthma. The pathogenesis of bronchoconstriction with this technique has not been clearly defined. However, it is hypothesized that heat and water fluxes during hyperventilation result in bronchoconstriction in part related to release of inflammatory mediators including histamine and leukotrienes (17). To investigate whether levels of expired NO are altered in a cotemporal fashion with the airflow obstruction that occurs after ISH, we compared mixed expired NO in a group of individuals with asthma after allergen challenge, after ISH challenge, and after repeated spirometry alone. Interestingly, we found that levels of expired NO decrease after repeated spirometry. Furthermore, our data indicate that ISH, as well as allergen, is associated with relative increases in mixed expired levels of NO compared with levels detected after spirometry alone.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Nonsmoking individuals with atopic asthma were recruited by advertisement. All subjects had a history compatible with asthma (18), an FEV₁  70% predicted, and a positive skin test to common allergens. In addition, subjects were required to demonstrate a bronchoconstrictor response to ISH, defined as a  12% fall in FEV₁ following a 3-min period of maximal voluntary ventilation (MVV) with cold, dry air as described below. Subjects were not using any asthma medications with the exception of inhaled  agonists, which were withheld for 8 h before all testing; caffeine was withheld for 12 h before any testing and corticosteroids (oral and inhaled) could not have been used for at least 6 wk. Healthy nonsmokers without a history of asthma were recruited from the hospital staff. The study protocol was approved by the Human Research Committee at the Brigham and Women's Hospital (Boston, MA). All subjects gave their written, informed consent.

Effect of Ambient NO on Expired NO

Ambient NO levels in our laboratory varied from 2 to 140 ppb over the duration of our study. To assess the effect of inhaled ambient NO on measured expired NO, mixed expired NO was collected from three subjects without asthma inhaling varying concentration of NO. Subjects were seated and wore noseclips while breathing for 5 min through a one-way valve from a reservoir bag containing medical-grade air with varying concentrations of NO. At the end of the 5-min period, subjects inhaled to TLC from the reservoir bag and then slowly exhaled into a Mylar bag until they reached FRC. Immediately after exhaling into the first bag, the subjects then inhaled ambient air (2-14 ppb during these experiments) to TLC, and once again inhaled to FRC into a second Mylar bag for NO analysis.

Main Protocol

The main study protocol entailed four sessionsa screening session and three challenge sessions (isocapnic cold air hyperventilation, allergen bronchoprovocation, and sham challenge with serial spirometry) as described in detail subsequently. Screening included a medical history and asthma questionnaire, a physical examination, allergen skin testing, spirometry, and screening ISH challenge. Baseline single-breath expired NO concentration was also measured. The three challenges were administered in random order to exclude an ordinal effect. There was at least a 1-wk washout period between allergen or ISH and any subsequent challenge.

Challenge sessions took place between 7:30 and 9:00 A.M. Subjects withheld medications and caffeine as noted previously. Before challenge, baseline spirometry and mixed expired NO were measured. If the baseline FEV₁ was above or below 10% of the baseline established during the screening session, testing was deferred. Following challenge, spirometry and NO collection were repeated every 15 min for the first 2 h and then hourly for the next 6 h.

Spirometry. Spirometry was performed using a Collins Survey III spirometer (Warren E. Collins, Braintree, MA). Baseline forced expiratory maneuvers were repeated until three consecutive maneuvers with  5% difference between the highest and the lowest FEV₁ were achieved. Subsequently, the best of two technically adequate maneuvers with  5% difference was recorded. All spirometry took place in the sitting position.

Skin testing. Skin testing with 10 common allergens was performed; the largest reaction was determined by measuring the wheal diameters. The allergen producing the largest wheal was selected for further testing with 10 serial 3-fold dilutions of allergen using an aqueous solution of 0.09% NaCl, 0.03% albumin, 0.4% phenol. Fifteen minutes after skin prick, the lowest dilution producing a wheal the diameter of which was 3 mm greater than diluent was determined.

Isocapnic cold air hyperventilation challenge. On the day of ISH challenge subjects inhaled cold ( 15° C), dry (0% humidity) air containing  7 ppb NO and 5% CO₂ for 3-min periods. Initially, the ventilatory rate was 20 breaths/min with tidal volume (VT) equal to FEV₁. At each subsequent level, the E was set to be approximately 50% greater than the preceding level. After each level of E, a single FEV₁ measurement was obtained at 0.5 min; two technically adequate FEV₁ maneuvers were then obtained at 1.5 and 3.5 min, and every 2 min thereafter until there was a < 1% drop (or an improvement) in FEV₁. The lowest FEV₁ occurring 1.5 min or later was recorded as the fall for that level of ventilation. Ventilation escalation continued until the FEV₁ fell to  80% of prechallenge baseline, or to maximal voluntary ventilation. Exhaled air for NO measurements was collected at 3.5 min after each level of E, just before spirometry. After the FEV₁ fell to  80% of prechallenge baseline, FEV₁ and NO measurements were made at 15-min intervals for 2 h, and then hourly for the next 6 h.

Allergen bronchoprovocation testing. On the day of allergen bronchoprovocation, subjects inhaled five tidal breaths of the lowest aqueous dilution of antigen that produced a wheal the diameter of which was 3 mm greater than diluent, as described previously. The antigen was administered from a nebulizer (model 646; DiVilbiss Health Care, Somerset, PA) attached to a 20-psi compressed air source and a calibrated dosimeter (S+M Instruments, Doylestown, PA). The FEV₁ was measured 10 and 11 min after antigen inhalation. If the FEV₁ was within 20% of the starting FEV₁, the next higher threefold concentration of antigen was administered until a 20% decrease in FEV₁ (the better of two technically adequate maneuvers) occurred. After inhalation of the antigen concentration that induced a 20% fall in FEV₁ spirometry and NO collection were repeated every 15 min for the first 2 h and then hourly for the next 6 h.

Sham challenge. On the day of sham challenge, baseline spirometry was performed and NO was collected. After 1 h, NO and FEV₁ measurements were made at 15-min intervals for 2 h, and then hourly for the next 6 h.

NO collection. Exhaled gas was collected before spirometry at each time point. During collection, subjects were seated and wore noseclips to exclude air entrained from the nasopharynx. After taking two full breaths from a Douglas bag filled with low-NO air (0-5 ppb) and 5% CO₂ subjects performed a 10-s breath hold at TLC followed by a slow vital capacity maneuver, over 10 s through a mouthpiece into a Mylar (NO-impermeable) bag. The bag contained a resistor in its neck that forced the subject to develop positive pressure in the oropharynx while collecting the sample. Three bags were collected for baseline measurements and two bags were collected at all other time points. The bags were sealed and analyzed for mixed expired NO levels (to the nearest 0.1 ppb) by chemiluminescence (model 42S; Thermo Environmental Instruments, Franklin, MA) as described previously (12). The average NO level at each time was reported.

CO₂-Free Sham Challenge

Six of the 10 subjects returned to the laboratory in order to address whether the decline in NO following sham challenge was related to 5% CO₂ in the low-NO air used as the inhaled gas in the sham challenge. On this occasion, we performed a sham challenge using an inspired gas that had low concentrations of NO ( 5 ppb) and was CO₂ free. Baseline spirometry and NO were collected as described previously for the sham challenge.

Spirometry-Free Sham Challenge

Four of the 10 subjects returned to the laboratory to assess the effect of spirometry on exhaled NO. On this visit we performed a sham challenge using 5% CO₂ as the inspired gas and did not perform spirometry; subjects provided serial exhaled gas samples only, as described previously.

Statistics

Data for expired NO during the challenge protocols are reported as percentage of baseline expired NO values with the standard error of the mean. Data for baseline expired NO are reported in parts per billion, also with the standard error of the mean. Challenge data were analyzed using two-way analysis of variance (ANOVA) for repeated measures (Systat v 5.0; Systat, Evanston, IL). Once a statistically significant difference between challenges was recognized, data from individual time points were compared by Student's t test. Correlations were made using Pearson's correlation coefficient.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Ten subjects underwent allergen, ISH, and sham challenges. All patients who underwent an initial challenge completed all three arms of the main protocol. This group included six males and four females, ages 19-39 yr; baseline FEV₁ and allergen data are presented in Table 1. Two subjects demonstrated a late-phase response to inhaled allergen as defined as a fall in FEV₁ of  15% from baseline 4-8 h following allergen challenge.

Effect of Ambient NO on Expired NO

To assess the effect of ambient NO on expired NO measurements, we measured mixed expired NO immediately after 5 min of breathing air with supplemental NO levels (0-200 ppb) in three normal subjects. Immediately after this collection, we measured expired NO after the first inhalation of low-NO air (2-14 ppb). As the amount of supplemental NO in the inspired gas increased, expired NO increased (Figure 1). Expired NO returned to baseline after one breath of low-NO air. The difference between measured expired NO levels after inhalation of NO-supplemented air compared with NO levels after inhalation of 0-ppb NO air was significant when the supplemented concentration reached 40 ppb (7.3 ± 0.87 ppb after 40 ppb, versus 4.6 ± 1.0 ppb after 0 ppb; p < 0.05).

View larger version (17K): [in this window] [in a new window]   Figure 1.   Effect of 5 min of inhalation of supplemented NO air, and subsequent effect of a single breath of low-NO air, on mixed expired NO levels (n = 3). NO was measured directly after 5 min of NO-supplemented air (solid bars) and once again after a single breath of low-NO room air (open bars) that followed immediately after the NO-supplemented air. *p < 0.05 versus breathing 0 ppb NO.

ISH, Allergen, and Sham Challenges

Bronchoprovocation with allergen and ISH produced equal initial falls in FEV₁ (25.1 ± 1% fall in FEV₁ after allergen versus 26.3 ± 3.6% after ISH, p = 0.77). Serial spirometry alone (sham challenge) produced a progressive drop in expired NO (p < 0.01) (Figure 2). Compared with the prechallenge baseline this reduction was significant after 15 min (23 ± 9.5% from prechallenge baseline, p < 0.001) and was maximal at 30 min (36.6 ± 8.5% from prechallenge baseline, p < 0.0001). ISH challenge blunted the decline in mixed expired NO seen with spirometry, so that there was no significant change in mixed expired NO compared with baseline after ISH challenge (p = 0.7) (Figure 2). After ISH challenge expired NO increased compared with levels after sham challenge (two-way ANOVA, p = 0.065). Allergen challenge also prevented the decline in mixed expired NO seen with sham challenge (two-way ANOVA, p < 0.01; Figure 2). The difference between mixed expired NO levels compared with baseline measured after allergen challenge and those measured after sham challenge reached a maximum at 8 h postchallenge (6.1 ± 8.2% versus 36.3 ± 8.8%, p < 0.01).

View larger version (17K): [in this window] [in a new window]   Figure 2.   Effect of challenge on exhaled NO (n = 10). Sham challenge was accompanied by significant reductions in NO compared with baseline. No significant fall in exhaled NO was observed after allergen or isocapnic cold air hyperventilation challenges. +p < 0.05 versus sham; *p < 0.05 versus baseline; **p < 0.01 versus baseline.

Only two subjects experienced a late-phase response to inhaled allergen as defined as a fall in FEV₁ of  15% from baseline 4-8 h after allergen challenge. However, for the entire study group (n = 10), the maximal late (4-8 h) fall in FEV₁ was significantly correlated with baseline NO (Pearson correlation coefficient r = 0.80, p < 0.01; Figure 3).

View larger version (11K): [in this window] [in a new window]   Figure 3.   Maximal late fall (4-8 h) in FEV1 after allergen challenge versus baseline exhaled NO. r = 0.80; p < 0.01.

CO₂-Free and Spirometry-Free Sham Challenges

To exclude the possibility that 5% CO₂ in the low-NO air used as the inspired gas during the main protocol caused a reduction in expired NO in the sham challenge, six subjects underwent an additional sham challenge (serial spirometry and NO collection only) using prior breaths of CO₂-free, low-NO air (0-5 ppb NO). These data were compared with the data previously obtained for these six subjects using 5% CO₂ as the inspired gas (Figure 4). CO₂-free air did not change the fall in expired NO seen after sham challenge with 5% CO₂ in these patients (two-way ANOVA, p = 0.78). In fact, at 10 of 13 time points measured, expired NO levels were lower using CO₂-free air than when measured with 5% CO₂ as the inhaled gas. The maximum fall in expired NO after sham challenge while breathing 5% CO₂ was 20.0 ± 3.9% and the maximum fall after sham challenge with 0% CO₂ was 18.8 ± 2.6% (p = 0.88).

View larger version (16K): [in this window] [in a new window]   Figure 4.   CO2-free sham challenge using prior breaths containing low NO, 0% CO2 plotted versus the original sham challenge data using prior breaths of air containing low NO, 5% CO2 for 6 of the 10 original subjects. Horizontal bar and broken lines define baseline ± 5%, respectively.

Four of our original 10 patients returned for a fifth challenge using 5% CO₂ as the inspired gas. On this occasion no spirometry was performed and no fall in NO was observed over the course of the challenge (one-way ANOVA, p = 0.26). Specifically, the maximum fall was 13.1% and at all other time points, expired NO was at least 92.8% of baseline (Figure 5).

View larger version (11K): [in this window] [in a new window]   Figure 5.   Serial NO collection using prior breaths containing low NO, 5% CO2 without spirometry (n = 4). Horizontal bar and broken lines define baseline ± 5%, respectively.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

NO is elevated in patients with asthma (8) and has previously been reported to rise following allergen inhalation in those patients who develop a late fall in FEV₁ (16). We set out to establish whether the bronchoconstrictor responses to other bronchoprovocative stimuli produce a similar change in mixed expired NO. We first established that inspired (ambient) NO must be controlled because fluctuations in inspired NO within the range found in our laboratory affect these expired measurements. Controlling levels of inspired NO, we compared mixed expired NO following bronchoprovocation with expired NO following serial spirometry alone. Surprisingly, we found that NO measured in the mixed expired gas of asthmatic subjects falls following repeated spirometry. This fall occurs after relatively few spirometric efforts and persists for at least 8 h after repeated spirometric maneuvers. Furthermore, we established that bronchoprovocation with either cold air hyperventilation or allergen prevents the fall in expired NO that occurs after repeated spirometry, and therefore by inference both are associated with further increased NO production.

Our data clearly demonstrate that repeated spirometry alone, without a bronchoprovocation stimulus, reduces the level of NO in mixed expired air as measured by our methods. We originally thought that this effect was possibly related to the use of 5% CO₂ in the inspired air, as Stromberg and co-workers have shown that inspired CO₂ can reduce expired NO in a rabbit model (19). However, our data from six of our subjects inspiring 0% CO₂ still demonstrate a fall in mixed expired NO of equal magnitude. We speculate that this decrease is related to repeated spirometric maneuvers, because we saw no such drop in expired NO levels when four of our patients returned to the laboratory and we measured mixed expired NO serially over 8 h but did not perform spirometry.

Although this phenomenon has not previously been reported, to our knowledge ours is the first systematic investigation of the effect of repeated spirometry alone on mixed expired NO. Our data are consistent with those reported by Persson and Gustafsson (20), which demonstrate that in mechanically ventilated guinea pigs, after an initial rise in response to allergen challenge, expired NO fell by 57% from baseline. These authors speculated that this fall represented an inhibitory effect of allergic mediators on NO production. Although the etiology of this fall with repeated FVC maneuvers is unclear, we speculate that the mechanical stress of repeated spirometry reduces NO production possibly through an effect on bronchodilating nonadrenergic, noncholinergic (NANC) nerves within the airways. These nerves have been shown to use NO as a neurotransmitter (21, 22). Data reported by De Sanctis and colleagues suggest that neural NO production is an important contributor to total expired NO. In mice with a deletion of constitutively active neural nitric oxide synthase (type I NOS), baseline mixed expired NO is reduced by 40% compared with wild-type animals (23). Although the precise cause of the drop in expired NO that we observed is yet to be determined, potentially the stress of repeated spirometry could suppress the formation of NO via this mechanism.

Our data with bronchoprovocative stimuli demonstrate that challenge with inhaled allergen or ISH increases levels of expired NO compared with levels seen following serial spirometry. We found that mixed expired NO increased, compared with sham, 30 min after challenge. These findings contrast with those of Kharitonov and co-workers (16), who reported that peak expired NO rose only between 8 to 21 h after antigen challenge. Our findings may be related to the fact that we compared the levels postbronchoprovocation with the levels after sham challenge which consisted of repeated spirometry alone. This allowed us to detect the fall in expired NO with repeated spirometry, and thus the relative rise in NO after bronchoprovocation. The early rise would not have been detected without this sham challenge as a control.

Kharitonov and colleagues speculated that the rise in NO seen after allergen challenge indicated type II NOS induction by late phase-related cytokines (16). We detected an early rise in NO. Our finding is consistent with two reports of a rapid rise in expired NO 10 min after allergen challenge in guinea pigs (20, 24). It is unlikely that this early rise involves type II NOS induction, which would require protein synthesis. The early rise in NO after ISH and allergen bronchoprovocation could be mediated via modulation of type II NOS by cofactors such as biopterin or local arginine concentrations (3). It is possible that later-inducible production accounts for the continued elevation of NO observed by Kharitonov and coworkers after allergen challenge (16).

In contrast to expired NO in allergen-induced asthma, little is known about expired NO in cold air- or exercise-induced asthma. Our data demonstrate that after ISH, similar to the response after antigen, expired NO levels rise compared with levels after repeated spirometry alone. Although the role of inflammatory mediators in ISH-induced airway narrowing has been controversial (25), the relative rise in mixed expired NO we have documented following ISH suggests that such an inflammatory process may be taking place. Alternatively, perhaps NO is playing a role in mediating the reactive hyperemia that has been postulated to account for part of the bronchoconstriction that occurs after exercise and ISH (26, 27). NO has been shown to be a potent vasodilator in the bronchial circulation (28, 29) and has been shown to magnify airway neurogenic plasma exudation (30, 31). This effect may be integral to the allergic response as well.

We have also demonstrated that the magnitude of the late fall in FEV₁ in response to inhaled allergen is significantly correlated with baseline NO levels. It is unclear whether NO is causative in this regard or merely reflective of a concomitant process, such as airway inflammation, contributing to bronchoconstriction after allergen inhalation. There is evidence to suggest that NO levels are elevated in conditions associated with airway inflammation. For example, upper respiratory infection is associated both with elevated expired NO (32) and an increased likelihood of significant late declines in airflow in response to inhaled allergen (33, 34). Alternatively, NO could potentially contribute to the appearance of late declines in FEV₁ by facilitating inflammatory cell influx to the airway through its known role in increasing vascular permeability (30, 31).

In summary, we have found that ambient levels of inspired NO must be controlled in order to discern the effects of intervention on mixed expired NO. When inspired NO levels are controlled, in individuals with asthma, repeated spirometry is associated with a reduction in mixed expired NO. Although the mechanism producing the fall after repeated spirometry is unclear, both ISH and allergen challenge produce a relative increase in expired NO. The reduction in NO with repeated spirometry alone suggests that mechanical stimuli in the asthmatic lung may affect NO homeostasis. It is unclear whether the increased NO levels after these stimuli may serve as a marker of airway inflammation or whether NO plays a pathophysiological role in the response. Studies with inhibitors of noninducible NOS may allow us to define further the role of NO in this regard.