INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nocturnal asthma is characterized by an increased circadian rhythm in airway patency, with airway obstruction generally maximal at 4 h and minimal at 16 h (1, 2). The underlying mechanisms for this are only partly understood (3). One hypothesis for this rhythm in airway patency is that airway inflammation is increased at night (4), probably from a lack of counteracting mechanisms, such as nocturnal decreases in plasma cortisol and epinephrine (5). Biopsies of the central airways could not confirm a nocturnal increase in inflammatory cells (8, 9). However, bronchoalveolar lavage fluid (BALF) and transbronchial biopsies in one study group showed higher numbers of eosinophils in asthmatic subjects with increased airway obstruction at night than in asthmatic subjects without such obstruction (9, 10). This suggests that inflammatory changes inducing nocturnal airway obstruction take place in a compartment that is inaccessible for routine monitoring of inflammation.

Measurement of exhaled nitric oxide (NO) has been suggested as a simple, noninvasive way in which to investigate asthmatic airway inflammation (11, 12). It has been shown that the level of exhaled NO is greater in asthmatic individuals than in healthy controls, and that it decreases after administration of corticosteroids (13). Moreover, levels of exhaled NO are not changed by pharmacologically induced bronchoconstriction (14). Recently, it has been shown that measurements of exhaled NO are correlated with the expression by alveolar macrophages (AM) and eosinophils of inducible nitric oxide synthase (iNOS) in induced sputum (15, 16). Apparently, measurements of exhaled NO are suitable for detecting upregulation of iNOS and activation of inflammatory cells in peripheral airways and alveolar tissues.

Because increased airway inflammation has been shown to occur at night in peripheral airways of patients with increased nocturnal airway obstruction, we sought to measure exhaled NO in a study of nocturnal asthma. We hypothesized that subjects with nocturnal asthma would have a circadian rhythm in their production of NO that was inverse to the circadian rhythm in airway patency.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Nonsmoking asthmatic patients and healthy controls aged 18 to 45 yr were recruited at the outpatient clinic of our institution or by advertisements in local newspapers. All subjects gave written informed consent to participate in this study. The study protocol was approved by the university ethics committee.

Asthmatic subjects were selected on the basis of: (1) a history consistent with the diagnosis of asthma; (2) a positive history of atopy and a positive Phadiatop (Pharmacia, Woerden, The Netherlands) test result; (3) FEV₁ > 1.5 L and > 60% predicted; (4) a concentration of methacholine bromide that reduced FEV₁ by 20% (PC₂₀)  9.8 mg/ml; and (5) no use of corticosteroids and no respiratory infection within 1 mo before the study.

Healthy controls had: (1) no history of lung disease; (2) no history of atopy and a negative Phadiatop test result; (3) less than a 20% decrease in FEV₁ after inhalation of methacholine bromide at a concentration of 9.8 mg/ml; and (4) no respiratory infection within 1 mo before the study.

Study Design

After characterization of each patient, peak expiratory flow (PEF) was measured in the patient's home for seven subsequent days at 4 h and 16 h. Asthmatic subjects were grouped as subjects with a mean 4-h to 16-h variation in PEF  15% (nonnocturnal asthma) and those with a mean 4-h to 16-h variation in PEF > 15% (nocturnal asthma). Subjects were then admitted to the hospital during the next 24 h. Exhaled NO was measured at 12, 16, 20, and 24 h, and at 4, 8, and 12 h of the next day. Shortly after determination of exhaled NO, PEF, FEV₁, and blood pressure (RR) were measured. To achieve maximal uniformity, all measurements were made by one person (H.v.d.V.).

Lung Function

FEV₁ and percent improvement in FEV₁ after inhaling 400 µg salbutamol were measured with a pneumotachograph (Masterscreen; Jaeger, Wurzburg, Germany) according to standardized guidelines (17). Hyperresponsiveness was assessed with a water-sealed spirometer (Lode, Groningen, The Netherlands) using increasing doubling concentrations of 0.03 to 9.8 mg/ml methacholine bromide (Sigma Chemical Co., St. Louis, MO), and a 2-min tidal breathing method adapted from Cockcroft and coworkers (18). In-home measurements of PEF were made with a Personal Best peak-flow meter (Glaxo Wellcome, Zeist, The Netherlands) at 4 h and 16 h on the seven subsequent days when this was done. During these days, bronchodilators were withheld as much as possible. Variation in PEF from 4 h to 16 h was defined as: (PEF from 4 h to 16 h)/(mean of 4 h + 16 h values). The mean variation in PEF was calculated as the average of 7-d variation in 4-h to 16-h PEF.

Exhaled NO

Upon admission to the hospital, the participating subjects were put on a low-protein diet (< 20 g/d) at 8:30 A.M., 12:30 P.M., and 6:30 P.M., and avoided heavy exercise. They rested for 15 min before each test. Single-breath measurements of exhaled NO were made according to published guidelines (19), using a chemiluminescence analyzer (CLD 700 AL; Eco Physics, Basel, Switzerland). Constant flow rates during the exhalation maneuver were targeted at 10 L/min. The mean value of three NO measurements was used for analyses. The lower NO-detection limit of the analyzer was 1 ppb, with a resolution of ± 1 ppb. The sampling flow rate was 600 ml/min, and the response time, including lag and rise time, was < 7 s. Repeated measurements, made within 10 min of each other, showed coefficients of variation (CV) in healthy and asthmatic subjects of 19% and 10%, respectively.

Data Analysis

All analyses were done with the SPSS/PC 6.01 software package (SPSS Inc., Chicago, IL). Two-sided values of p < 0.05 were considered statistically significant. Parametric analysis was performed with Student's t test for differences between groups at the same time point, and with Student's paired t test for differences between 4 h and 16 h values within a group, after checking for normal distribution. The 12, 16, 20, and 24 h, and 4, 8, and 12 h measurements were analyzed in a longitudinal way through cosinor analysis (20). Correlations between NO and other variables in the total group of asthmatic subjects were investigated with Pearson's test and analyzed in a multiple regression model. In this model, NO concentration was entered as the dependent variable, and age, gender, height, blood pressure, prior use of inhaled corticosteroids, FEV₁, and PC₂₀ methacholine were entered as independent variables.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Clinical Characteristics

Six healthy controls, eight asthmatic subjects with PEF variation  15%, and six asthmatic subjects with PEF variation > 15% participated in the study (Table 1). Systolic and diastolic blood pressures were higher, but not significantly, in the asthmatic subjects with PEF variation > 15%. No cosinelike rhythm in blood pressure was found in any of the three groups during a 24 h measurement period. Mean FEV₁% predicted, post bronchodilator increase in FEV₁, and PC₂₀ methacholine were similar in the two asthmatic groups. In-home variation in PEF from 4 h to 16 h correlated well with the 4-h to 16-h PEF variation found in the hospital (r = 0.77, p < 0.001).

Twenty-four-hour FEV₁

FEV₁% predicted showed a significant cosinelike rhythm in the two asthmatic groups (Figure 1), in contrast to the healthy controls. The peak and the nadir of the cosine function were at 4 h, and at 16 h, respectively, in both groups. The amplitude of the cosine function was 5% in the asthmatic subjects with PEF variation  15%, and 8% in the asthmatic subjects with PEF variation > 15%.

View larger version (36K): [in this window] [in a new window]   Figure 1.   Mean ± SEM exhaled NO (upper panel ) and mean FEV1% predicted (lower panel ) values for six healthy controls, eight asthmatic subjects with 4-h to 16-h PEF variation  15%, and six asthmatic subjects with 4-h to 16-h PEF variation > 15%, at seven time points during 24 h. The two asthmatic groups have a significant circadian rhythm (cosinor analysis, p < 0.05) in FEV1. No circadian rhythm was found in exhaled NO in any of the three groups (cosinor analysis). *p < 0.05 versus 4 h.

FEV₁% predicted at 4 h was significantly lower than at 16 h in the asthmatic subjects with PEF variation > 15%; at 80 ± 25% versus 92 ± 31% (mean ± SD) (p < 0.05), as well as in the asthmatic subjects with PEF variation  15%; at 90 ± 16% versus 98 ± 12% (p < 0.05), and was borderline normal in the healthy controls; at 92 ± 12% versus 97 ± 11% (p = 0.07).

Exhaled Nitric Oxide

Differences between the three groups. Exhaled NO concentrations of the asthmatic subjects with a 4-h to 16-h PEF variation > 15% were higher than those of the asthmatic subjects with a 4-h to 16-h PEF variation  15% (p < 0.05), and also higher than those of the healthy controls at all time points (p < 0.01). Also, exhaled NO concentrations of the asthmatic subjects with a 4-h to 16-h PEF variation  15% were significantly higher than those of the healthy controls at all time points (p < 0.05). The NO concentrations at the seven time points investigated in each group were: 46.0 ± 18.7 ppb (mean ± SD) for the asthmatic subjects with PEF variation > 15%, 26.6 ± 12.6 ppb for the asthmatic subjects with PEF variation  15%, and 9.9 ± 2.1 ppb for the healthy controls.

Variation during 24 h. Exhaled NO concentrations did not fit a true cosinor function in any of the three groups, although for the asthmatic subjects with PEF variation > 15% the cosinor analysis had a value of p = 0.06, with an amplitude of 4 ppb (see also Figure 1). In the asthmatic subjects with PEF variation > 15%, the NO concentration at 4 h was higher than at 16 h; at 50 ± 20 ppb versus 42 ± 15 ppb (mean ± SD) (p = 0.029). There was no 4-h versus 16-h difference in the two other groups. Mean CVs of NO values at the seven time points at which NO was measured were 15% in asthmatic subjects with PEF variation > 15%, 20% in asthmatic subjects with PEF variation  15%, and 42% in healthy controls.

Relation to clinical variables. The level of exhaled NO was significantly and positively correlated with the 4-h to 16-h variation in PEF (Figure 2). Exhaled NO was not significantly correlated with any of the other clinical variables described in Table 1. In a multiple regression model, the variation in exhaled NO was not explained by age, gender, height, blood pressure, prior use of inhaled corticosteroids, FEV₁, or PC₂₀ methacholine. The 4-h to 16-h difference in NO was not correlated to the 4-h to 16-h difference in FEV₁ in any of the three study groups.

View larger version (12K): [in this window] [in a new window]   Figure 2.   Positive correlation between exhaled NO and 4-h to 16-h variation in PEF. Open squares: six healthy controls; open circles: eight asthmatic subjects with 4-h to 16-h PEF variation  15%; solid circles: six asthmatic subjects with 4-h to 16-h PEF variation > 15%.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study confirmed that asthmatic patients with increased variability in PEF as measured in the home setting have an increased circadian rhythm in airway patency as determined in the hospital, with the highest FEV₁ at 16 h and the lowest FEV₁ at 4 h (1, 2). Despite an isolated increase in exhaled NO at 4 h in subjects with large variations in PEF, we were unable to show an inverse cosinelike rhythm in exhaled NO; nor did we observe a significant correlation between exhaled NO levels and FEV₁ values. Subjects with nocturnal asthma exhaled significantly higher levels of NO than did subjects with nonnocturnal asthma throughout the day, suggesting that they had a higher overall level of airway inflammation. Except for the variation in 4-h to 16-h PEF, there was no relationship between NO levels on the one hand and clinical variables on the other.

The most important finding in this study were the significantly higher NO levels in subjects with nocturnal asthma than in those with nonnocturnal asthma. This difference was present at all seven time points at which NO was measured. In our opinion, this indicates that nocturnal asthma is associated with higher degrees of airway inflammation. However, the higher NO levels in subjects with nocturnal asthma were not associated with lower FEV₁, PEF, or PC₂₀ methacholine values. Apparently, such lung-function measurements do not reflect airway inflammation associated with high levels of NO production. We assume that they reflect other aspects of inflammation (e.g., the more chronic sequelae of inflammation). Because exhaled NO appears to be closely related to instability of the airways as measured through variability in PEF, we suggest that NO may help clinicians to monitor the control of asthma. The recognition of subjects with nocturnal asthma may especially be facilitated by this new technique.

Another finding in the study was that asthmatic subjects with marked variations in PEF showed higher levels of NO at 4 h at night than at 16 h. However, in contrast to the case with FEV₁, no true circadian rhythm was present in exhaled NO. The result of the cosine analysis was that the amplitude of the cosine function failed to reach statistical significance in all three study groups. This means that either a cosine function was absent, or that its amplitude was very small. In fact, the asthmatic subjects with PEF variation > 15% had an amplitude of 4 ppb (p = 0.06). In view of the differences between the three study groups in exhaled NO concentrations, this amplitude seems irrelevant. Moreover, the 4-h to 16-h change in NO was not correlated with the 4-h to 16-h change in FEV₁. Therefore, we assume that the small increase in NO found at 4 h, but not at the other time points, indicates an aspect of inflammation that is less important for the generation of increased nocturnal airway obstruction. Several inflammatory factors may explain the nocturnal increase in NO. First, the exaggerated nadirs in plasma cortisol at 24 h in subjects with nocturnal asthma (6) may lead to enhanced transcription of DNA encoding for inducible NOS. Second, it has been shown that the proinflammatory cytokines interleukin-1 (IL-1) and interferon- (IFN-), both of which are capable of stimulating NO production, are increased at 4 h (21, 22). Moreover, noninflammatory factors (3), such as decreased serum catecholamine levels (5), increased parasympathetic nervous tone (23), and supine position at night (2), may have led to increased nocturnal airway obstruction independent of airway inflammation and NO production. Whatever the cause, we conclude that the observed nocturnal increase in NO was too small to solely explain the increased nocturnal airway obstruction in subjects with nocturnal asthma.

In this study, we took all steps indicated to minimize factors that might interfere with a possible circadian rhythm in NO. Participating subjects were given a low-protein diet to prevent a postprandial increase in serum L-arginine (24). They had to rest for at least 15 min prior to NO testing, because exercise has been found to decrease NO values (25). Also, smoking and (intermittent) use of corticosteroids were excluded because of their downregulating effects on NO (11, 13, 26). Diastolic blood pressure was measured together with NO, because hypertension has been reported to be associated with low NO values (26). However, diastolic blood pressure did not show a circadian rhythm, and was not a variable contributing to the variance in NO. Additionally, NO measurements were made according to recently published guidelines (19), and special attention was given to ensuring uniform expiratory flow rates and pressures. Constant flow rates during the exhalation maneuver were targeted to 10 L/min. It is possible that a lower flow rate resulted in larger intergroup differences, as suggested by Silkoff and colleagues (27). However, the absolute values of exhaled NO concentrations are not easily related to inflammatory parameters. Therefore, any standardized flow rate for measuring exhaled NO would be appropriate. Following the recommendations of the European Respiratory Society (ERS) Task Force (19), we showed that repeated measurements of exhaled NO at short intervals (within 10 min) are highly reproducible.

This study recruited subjects with and without nocturnal asthma, based on 4-h and 16-h PEF measurements made at the patients' homes during a 1-wk period. In the hospital, both the mean FEV₁ and PEF decreased at night in subjects with nocturnal asthma and those with nonnocturnal asthma. As expected, FEV₁% predicted decreased more in the subjects with nocturnal asthma than in those with nonnocturnal asthma; by 8% versus 12%, respectively. However, some subjects classified as having nocturnal asthma did not show a significant decrease in PEF or FEV₁ at night. This frequently occurring problem in studies of nocturnal asthma may result from disturbed sleep, stress, or lack of exposure to house dust mite during hospitalization (28). We feel that 7 d of PEF measurement at home are of greater importance than 1-d of lung-function testing in the hospital, which is in accord with most recruitment strategies for studies of nocturnal asthma.

The results in Table 1, might suggest that the two asthmatic groups in our study were quite different in important baseline characteristics. Therefore, we analyzed these potentially confounding variables as for their effect on exhaled NO with a multiple regression model. Age, gender, blood pressure, inhaled corticosteroids taken prior to the study, FEV₁, and PC₂₀ methacholine did not contribute to the variation in NO. Therefore, an unequal distribution of these factors was not responsible for the significant differences in NO observed in the two asthmatic groups.

Our study is partly in line with another study of day-night variations in exhaled NO. Kharitinov and colleagues measured exhaled NO during a period of 1 d, at intervals of 4 h, in four healthy controls and seven mildly asthmatic subjects (29). As in our study, NO increased at 4 h at night, and by the same order of magnitude. In contrast to the finding in our study, the nocturnal increase in NO in the study by Kharitinov and colleagues was positively correlated with the nocturnal decrease in FEV₁, suggesting a pathophysiologic relationship. However, the subjects in Kharitinov and colleagues' study were not recruited based on circadian variations in airway obstruction, and none had ever had nocturnal symptoms. Patel and colleagues measured between-day PEF variation and exhaled NO concentrations in 23 asthmatic subjects and 18 healthy controls over a period of 12 wk (30). They found no correlation between changes in exhaled NO and between-day PEF variation. Wempe and associates showed that inhaled corticosteroids decreased both within-day and between-day PEF variation, whereas -mimetic agents influenced only within-day PEF variation (31). These studies indicate that within-day and between-day PEF variation are influenced by different pathophysiologic mechanisms. Before drawing definitive conclusions, it will be necessary to await further (interventional) studies of the relationship between NO and variability in PEF, especially in the field of nocturnal asthma.

Assuming that exhaled NO reflects airway inflammation (11, 12, 15, 16, 19, 32, 33), the present study has several theoretical, therapeutic, and practical implications. First, our data suggest that a circadian rhythm in airway inflammation is not the most prominent feature of nocturnal asthma. As we have suggested in other studies, we believe that nocturnal asthma merely reflects more severe airway inflammation (34). Therefore, effective therapy should maximally suppress airway inflammation throughout the day. Measurements of exhaled NO may support measurements of within-day PEF variability to identify asthmatic patients with high levels of airway inflammation, who are at risk for nocturnal asthma. Additionally, in the framework of standardizing protocols for measurement of NO, it is not necessary to measure NO at fixed time points; measurement at any time is good for showing higher NO levels in individuals at risk for nocturnal asthma.