INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The cysteinyl leukotrienes (CysLTs) (LTC₄, LTCD₄, and LTE₄) are known to be potent inducers of bronchoconstriction in healthy individuals and patients with asthma (1, 2). Their involvement as key mediators in the pathogenesis of airway narrowing in bronchial asthma has recently been established by the introduction of antileukotriene drugs for the treatment of asthma (3). Bronchoprovocation with LTD₄ in asthmatics has been shown to increase eosinophils in induced sputum after maximal (4) but not submaximal bronchoconstriction (5), raising the possibility that CysLTs indeed are also mediators of airway inflammation, a hallmark of bronchial asthma. This is further suggested by potent effects of CysLTs on blood flow and airway microvascular leakage, as well as on mucus secretion, in human airways in vitro (1).

Another characteristic feature of both acute and chronic asthma is ventilation-perfusion (A/) imbalance (6). Moreover, we have convincingly demonstrated that challenge with platelet-activating factor (PAF) in both healthy individuals (7) and asthmatics (8) causes A/ inequalities with a pattern that is similar to that observed in naturally occurring acute asthma (8). Remarkably, systemic, cellular and lung function effects caused by PAF were completely blocked by a short-acting -adrenergic agonist, but not by an anticholinergic, thereby suggesting that PAF-induced A/ defects could be more related to abnormal airway vascular permeability than to bronchoconstriction per se (9). Similar abnormal A/ patterns have been also shown in adult asthmatics immediately after exercise, methacholine, histamine, and allergen challenges (6, 10, 11). Altogether these studies provide evidence that many endogenously released inflammatory mediators have the potential to contribute to the development of pulmonary gas exchange disturbances characteristically observed in asthma. Notwithstanding, it has not been investigated if CysLTs have effects on alveolar ventilation to pulmonary perfusion matching in patients with asthma.

The present study was therefore conducted to evaluate whether inhalation of LTD ₄ was able to provoke pulmonary gas exchange abnormalities in patients with stable, mild, intermittent asthma. This is in fact the first bronchoprovocation study with LTD₄ in which A/ mismatching is used primarily as a marker of peripheral airway obstruction. To further understand the proinflammatory effects of CysLTs, it was also investigated if LTD ₄-induced bronchoconstriction was associated with immediate effects on the differential cell counts in induced sputum, as previous studies only have assessed the effects of CysLTs at later phases after challenge (4, 5, 12).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Thirteen atopic, nonsmoking, patients with stable mild intermittent asthma were recruited for the study (Table 1), which was approved by our Ethical Research Committee. The inclusion criteria were as follows: age older than 18 yr and younger than 45 yr; no respiratory infection or exacerbation of asthma within the preceding 6 wk; FEV₁  70% predicted and  1.5 L after discontinuing bronchodilators for 12 h and inhaled steroids for 24 h, and positive methacholine bronchial challenge (provocative dose causing a 20% reduction in FEV₁ [PD₂₀] < 1.9 µmol) on their first visit; no previous treatment with oral steroids; and absence of any systemic or cardiopulmonary disease other than asthma.

Study Design

Patients attended the laboratory on three separate visits from 8:00 A.M. On the first visit, a complete clinical, functional, and allergic evaluation was performed. On the second visit, the patient attended for baseline sputum induction. On the third occasion, the patient attended for LTD₄ bronchoprovocation test. A set of duplicate measurements was performed before the inhalation challenge, immediately after challenge, and then 15 and 45 min after inhalation of LTD₄.

Measurements

Forced spirometry (predicted values in reference 13) and total respiratory system resistance (Rrs) were performed in each patient. Blood samples were collected anaerobically through a catheter inserted into the radial artery for arterial blood gases. Oxygen uptake (O₂) and CO₂ production (CO₂) were calculated from mixed expired O₂ and CO₂ concentrations. Minute ventilation (E) and respiratory rate (RR) were measured. The alveolar-arterial PO₂ gradient [(A-a)PO₂] was calculated according to the alveolar gas equation using the measured respiratory exchange ratio. We used the multiple inert gas elimination technique (MIGET) to estimate the distributions of A/ ratios without sampling mixed venous inert gases in the customary manner (14), under steady-state conditions (15). With this approach cardiac output () needs to be directly measured by dye dilution technique (14). An electrocardiogram, heart rate (HR), systemic arterial pressure (Psa), and arterial oxygen saturation were continuously recorded. Urinary LTE₄ (uLTE₄) was measured and expressed in relation to urinary creatinine (16).

Sputum Induction and Processing

This procedure was performed 75 min after challenge according to a validated method (17). The repeatability of the sputum differential cell counts examined for all cell types in our laboratory showed a good reproducibility (18). Eosinophil cationic protein (ECP) and interleukin-8 (IL-8) in sputum supernatants were also assessed.

LTD₄ Inhalation Challenge

Bronchoprovocations with LTD₄ were performed according to a previously validated method (19) using a dosimeter-controlled jet nebulizer and a protocol permitting the administration of increasing doses ranging from 3 pmol to 30 nmol. The challenge was conducted until at least a 20% decrease in FEV₁ was produced.

Statistical Analysis

Results are expressed as mean ± SE. PD₂₀ was derived by linear interpolation from the log-cumulated dose-response curve and its geometric mean was calculated on log-transformed raw data. The effects of LTD₄ challenge were assessed using one-way repeated analysis of variance (ANOVA). Wilcoxon's test and Pearson's or Spearman's rank tests were also used. Statistical significance was set at p  0.05.

For a fuller account of the METHODS, see the online data supplement.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Baseline Findings

Before challenge, patients had normal FEV₁, mild decreases in FEF_25-75%, and modest increases in Rrs (Tables 1 and 2; Figure 1). Arterial blood gases were normal and A/ relationships were mostly within the normal range (20), with narrowly unimodal distributions of pulmonary blood flow (Log SDQ) and alveolar ventilation (Log SDV) (Table 2; Figure 1). Intrapulmonary shunt (percentage of blood flow to units with A/ ratios < 0.005) and areas with low (blood flow to units with low A/ ratios [ < 0.1, excluding shunt]) or high (ventilation to units with high A/ ratios [10 >-< 100]) A/ ratios were conspicuously absent.

View larger version (18K): [in this window] [in a new window]   Figure 1.   Individual time courses of Rrs (in cm H2O · L1·s), PaO2 (in mm Hg), and dispersion of pulmonary blood flow (Log SDQ) (dimensionless) values after LTD4 challenge. Measurements correspond to baseline; immediately after bronchoprovocation (within 5 min); and 15 and 45 min after LTD4 inhalation (bars denote mean values; arrows indicate LTD4 challenge).

LTD ₄ Challenge

The challenge produced a significant bronchoconstriction, as shown by a marked FEV₁ reduction (by 32 ± 3%; range, 22 to 56%), a decreased FEF_25-75% (by 50 ± 4%), and an increased Rrs (by 106 ± 12%) (Table 2; Figures 1 and 2). The nadir of the reduced airflow rates always occurred between 5 and 10 min after the inhalation of the last dose of LTD₄. Compared with methacholine, LTD₄ was approximately 450 times more potent on a molar basis, with geometric mean PD₂₀ values being 0.87 nmol (range, 0.23 to 0.75 nmol) for LTD₄ and 355.6 nmol (range, 195 to 1,545 nmol) for methacholine (Table 1).

View larger version (13K): [in this window] [in a new window]   Figure 2.   Mean (± SE) time courses of changes in Rrs (open triangles), A/ mismatching as assessed by the dispersion of blood flow (Log SDQ) ( closed triangles), FEV1 (open circles), and PaO2 (closed circles), all expressed as percentage of change from baseline (before challenge). Note that the pattern of improvement of all parameters is steadily toward baseline values and runs in parallel (bold arrow: LTD4 challenge).

This intense bronchoconstriction was associated with a moderate to severe decline in Pa_O2 (by 25 ± 4 mm Hg; range, 84 to 56 mm Hg) and a substantial increase in (A-a)P_O2 (p < 0.05 each) at the nadir of LTD₄ challenge (Table 2; Figure 1). Arterial deoxygenation was essentially caused by moderate to severe A/ worsening, as reflected by increases in Log SDQ (range, 0.63 to 0.83) and Log SDV (range, 0.71 to 1.05) (normal values,  0.60 to 0.65 [20]) (p < 0.05 each). Similarly, an overall index of A/ heterogeneity (DISP R E*) (root mean square difference among measured retentions [R] and excretions [E] of the inert gases [except acetone] corrected for the dead space) markedly deteriorated (range, 6.2 to 10.3; normal values,  3.0 [21]; p < 0.05). Broadened unimodal A/ patterns were observed in each participant. Areas with low or high A/ ratios were never present. The mean A/ ratio of the pulmonary blood flow (mean ) distribution decreased (p < 0.05) whereas that of alveolar ventilation (mean ) was unchanged. There were no changes in intrapulmonary shunt or dead space (percentage of ventilation to units with A/ ratios > 100) after challenge. All the other variables (E, RR, Pa_CO2, pH, O₂, HR, Psa, and ), within normal limits at baseline, remained stable after challenge.

Baseline FEV₁ (r = 0.74, p < 0.05), Pa_O2 (r = 0.70, p < 0.01), (A-a)P_O2 (r = 0.78, p < 0.001), Log SDQ (r = 0.78, p < 0.01), Log SDV (r = 0.89, p < 0.001), and DISP RE* (r = 0.70, p < 0.05) values were significantly correlated with their respective differences after challenge. Thus, the better the basal lung function variables, the smaller their LTD₄-induced disturbances. Differences in DISP RE* and in Log SDQ and Log SDV before and immediately after LTD₄ bronchoprovocation were also closely correlated (r = 0.71, p < 0.05; and r = 0.97, p < 0.001, respectively). By contrast, there were no correlations between spirometric or lung mechanic parameters and the pulmonary gas exchange descriptors.

Time Course after Challenge

At 15 min, FEV₁, FEF_25-75%, Rrs, and the respiratory [Pa_O2 and (A-a)P_O2] and inert (Log SDQ, Log SDV, and DISP RE*) gas exchange descriptors were still abnormal (p < 0.05 each), including a decreased mean (p < 0.05) (Table 2; Figures 1 and 2). At 45 min, except for a mild, residual, increase in Rrs (by 28 ± 6%, p < 0.05), all the other previously altered indices returned to normal or showed a trend toward baseline (Figure 2); mean was reduced (p < 0.05). The discrepancy between Pa_O2 and (A-a)P_O2, still slightly abnormal in relation to baseline values, and all the inert gas indices may likely be explained by a type 2 error owing to the relatively small number of patients included. It is of note that the time course and pattern of improvement/recovery after challenge of the most relevant variables, namely FEV₁, Rrs, Pa_O2, and Log SDQ, ran in parallel throughout the whole period of study.

Induced Sputum

The cellular analysis of induced sputum yielded a mild increase of eosinophils (p < 0.05) after LTD₄ challenge (Table 3). At baseline, there was a significant correlation between the number of eosinophils and the sputum concentrations of ECP (r = 0.78, p < 0.02). uLTE₄ increased (from 1,404 ± 397 to 2,634 ± 566 pg · mg creatinine¹, p < 0.005) after challenge.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Major Findings

There were three novel findings in our study. First, while confirming the exquisite potency of LTD₄ as a bronchoconstrictor in this subset of patients with asthma, with an average molar dose ratio to methacholine of approximately 450, the study disclosed that airway narrowing was associated with pronounced arterial deoxygenation resulting from moderate to severe A/ imbalance. Second, the pattern of improvement of spirometry, lung mechanics, and pulmonary gas exchange indices ran in parallel throughout the study. Third, the LTD ₄ challenge caused mild eosinophilia in induced sputum at 75 min after the challenge. Moreover, the study confirmed that urinary elimination of LTE₄ is increased after the inhalation of LTD₄ (19), a finding consistent with the rapid elimination of CysLTs by this route (1).

Gas Exchange Defects

Immediately after LTD₄ bronchoprovocation, A/ relationships worsened moderately to severely, resulting in a commensurate decrease in Pa _O2, in comparison with the relatively intense bronchoconstriction. It may be likely that the degree of severity of A/ worsening was offset, at least in part, by a simultaneous enhancement of hypoxic pulmonary vasoconstriction caused by the inhaled mediator. The latter mechanism should improve A/ balance, other things being equal. Pulmonary gas exchange defects were characterized by the development of broadened unimodal A/ patterns in each patient, just as is often seen in stable patients with moderate to severe persistent asthma (6, 10) or after several types of challenges (10, 11). Notwithstanding, these A/ findings differed from those extremely severe, including the characteristic bimodal blood flow profile, highly prevalent in patients with life-threatening acute asthma who may (22) or may not need (6) ventilatory support, but seldom seen in chronic asthmatics regularly treated with inhaled steroids (10). Areas with low A/ ratios were conspicuously absent and both basal intrapulmonary shunt and dead space remained unchanged after challenge. The similarities between the degree of pulmonary gas exchange disturbances herein observed with those shown after other agents, i.e., histamine (10, 11) or PAF (6), may indicate that the mechanisms of bronchoconstriction in all these challenges are similarly distributed in both central and distal airways. Although it is generally held that gas exchange impairment in asthma is mainly related to small airway obstruction (6), experimental studies have shown that the A/ matching becomes much more disrupted with larger airway obstruction (23). Methacholine-induced A/ mismatch in asthma patients may (24) or may not (25) be correlated with alveolar ventilation maldistribution.

Intrapulmonary shunt did not increase after challenge, hence suggesting an active hypoxic pulmonary vasoconstriction, a very efficient collateral ventilation, or incomplete airway occlusion (6). An active hypoxic pulmonary vasoconstrictor response at the peak of the challenge may have attenuated, at least in part, the degree of A/ deterioration initially caused by LTD ₄ bronchoconstriction. In principle, any enhancement of hypoxic vasoconstriction should ameliorate the alveolar ventilation to pulmonary perfusion balance (26). Maximal response to nebulized LTC₄ in monkeys included marked increases in transpulmonary pressure due to a reduced pulmonary compliance, slight changes in pulmonary resistance, severe arterial hypoxemia, mild increments of systemic arterial pressure, and severe pulmonary hypertension (27). Although pulmonary hemodynamics was not measured, systemic hemodynamics after LTD₄ challenge was stable. In this context, therefore, it is likely that pulmonary vascular pressures remained unchanged. Pulmonary hemodynamics after PAF challenge in both normals and asthmatics resulting in similar gas exchange abnormalities remained unaltered (6). The response of the pulmonary circulation to the effects of CysLTs may be complex. For instance, at least in vitro, the direct contractile effect of CysLTs on the human pulmonary artery involves potent feedback regulation by locally released prostacyclin (28).

Gas Exchange-Bronchoconstriction Relationships

The patterns of improvement of airflow obstruction and pulmonary gas exchange disturbances after challenge ran in parallel. At the end of the study there was still a mild, residual, increased Rrs and a decreased FEV₁ along with discrete hypoxemia and increased (A-a)P_O2. This close time course in both maximal airflow rates and gas exchange parameters is akin to that shown by patients with acute asthma discharged home, but at variance with that seen by those who are hospitalized (6, 10). Acute asthmatic inpatients show a marked delay in the recovery of gas exchange abnormalities in relation to that of spirometric alterations (6, 10). This gives support to the hypothesis that airflow rates may reflect predominantly bronchoconstriction of larger airways whereas gas exchange defects would represent a more silent manifestation of the disease, presumably more related to uneven inflammatory peripheral airway narrowing (6, 10). It can be postulated that LTD₄ inhalation induces less intense widespread airway narrowing and, as its effect tends to vanish, improves more readily, just as outpatients with acute asthma have less airway inflammation that, with adequate therapy, ameliorates more efficiently than in those who are hospitalized. All in all, it is most likely that LTD₄-induced abnormal pulmonary gas exchange in our study was predominantly reflecting widespread bronchoconstriction.

Induced Sputum

Along with all these functional abnormalities there was a mild but significant increase in eosinophils in induced sputum 75 min after LTD₄ inhalation. This is the first evidence that inhaled LTD₄ has a comparatively rapid effect on airway migration of eosinophils in asthmatics, just as previous studies have documented the same finding at a later stage after challenge (4, 12). Although the mechanisms involved remain elusive, there are indications that LTD₄ challenge may provoke cellular chemoattraction within the airways directly (29), or by secondary mediator release (30), or enhanced permeability of the bronchial postcapillary venules and extravasation (31). LTD₄ in rats facilitates the rolling and adhesion of circulating leukocytes through upregulation of adhesion molecules (32), thereby allowing for eosinophil migration through the vascular wall (33).

Our observations may thus support the suggestion by Diamant and coworkers (4) that inhaled LTD₄ results in more marked bronchoconstriction-induced sputum eosinophilia in the airways of asthmatics 4 h postchallenge. By contrast, Mulder and coworkers (5) did not observe eosinophils in sputum over a period of 24 h. Differences in patients' characteristics, inhalational procedures, or time points of measurements might explain part of these discrepancies. With our current data it is tempting to speculate, however, that LTD₄ challenge abruptly provoked spirometric and mechanic disturbances, gas exchange defects, and cellular changes, similar to those reported in spontaneous mild to moderate acute asthma. Notwithstanding, maximal bronchoconstriction, also invoked as an alternative, single mechanism of sputum eosinophilia after LTD₄ (4, 5), could not be disregarded.

Summary

We have demonstrated that bronchoprovocation with LTD₄ resulting in prominent airflow obstruction provoked moderate to severe pulmonary gas exchange defects along with sputum eosinophilia, hence suggesting that potent counterpulmonary vascular regulatory mechanisms remain. Furthermore, the pattern of improvement of pulmonary gas exchange disturbances paralleled both spirometric and lung mechanic alterations. These findings likely suggest that the endogenous release of leukotrienes contributes in concerted action with other mediators to the pathobiology of pulmonary gas exchange abnormalities commonly shown in patients with acute severe asthma. Furthermore, in view of the current therapeutic efficacy of LT receptor antagonists (3), a similar pathogenic mechanism, possibly less prominent, may occur in stable chronic asthma to further illustrate the frequent coexistence of underlying mild A/ inequalities (34). Besides, our current laboratory-induced LTD ₄ model of gas exchange defects in patients with mild asthma may be useful to investigate the efficacy of new agents to optimize therapeutic strategies in the setting of critically ill patients with asthma. In conclusion, LTD₄ challenge provokes both functional and cellular changes similar to those reported in spontaneous acute asthma. Accordingly, it may be tempting to suggest that future interventions, by for example LT receptor antagonists (3), could be warranted in the emergency room set-up (35), where there is evidence of increased uLTE₄ excretion (36).