INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Severe ₁-antitrypsin deficiency (₁-ATD) is an uncommon inherited disorder that predisposes, particularly in smokers, to the development of panacinar emphysema at an early age. Earlier reports (1) and our own experience suggest that asthma may occur in patients with well-established panacinar emphysema. Furthermore, patients with ₁-ATD may carry the diagnosis of asthma while progressive and irreversible panacinar emphysema develops. Conversely, asthma may be difficult to diagnose in patients with coexistent "fixed" airway obstruction. In order to better characterize the clinical spectrum of airway obstruction in subjects with ₁-ATD, we define asthma as episodic airway obstruction associated with wheezing, an increased serum IgE concentration, and atopy. The airway obstruction is wholly or partly reversible with bronchodilator therapy.

The development of asthma in patients with ₁-ATD may have additive long-term effects on the development of irreversible airway obstruction and emphysema. In this regard, both an increased serum IgE titer and atopy have been associated with the development of chronic obstructive lung disease (5).

In order to determine whether asthma is an associated feature of severe ₁-ATD, we compared the prevalence of the clinical characteristics of attacks of wheezing, bronchodilator responsiveness, an increased serum IgE concentration, and atopy in subjects with ₁-ATD with their prevalence in a group of patients with well-defined COPD but without ₁-ATD. We found that a significant proportion of individuals with ₁-ATD-related emphysema are asthmatic: a condition that may increase the severity and promote the accelerated development of chronic airway obstruction. In addition, our findings indicate that asthma is more common in this group than in individuals without ₁-ATD.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Population (Groups 1 and 3)

The study population consisted of 43 consecutively enrolled patients with ₁-ATD who had been identified from a population of subjects suspected of having ₁-ATD through their clinical and family history. All 43 patients were enrolled in a National Heart, Lung, and Blood Institute (NHLBI)-sponsored ₁-ATD registry to study the natural history of the homozygous PiZ condition (9). Of the 43 patients, 39 were index cases (Group 1) and five (Group 3) were referred because of a family history of ₁-ATD but were asymptomatic and had well-preserved lung function. The diagnosis was confirmed and the phenotype established by sending blood samples from each patient to a central phenotyping laboratory (9). Eligibility criteria for participation in the registry included age  18 yr, a serum concentration of ₁-AT of 11 µM or less (normal: 20 to 48 µM or 150 to 350 mg/dl), and the ability to give informed consent.

A complete medical history was obtained for each patient, which included a standard American Thoracic Society (ATS) symptom questionnaire (ATS-DLD 78) and past medical history. Standard posteroanterior (PA) and lateral chest roentgenograms were obtained for each patient and read for emphysema. The study was approved by the Institutional Review Board of St. Luke's-Roosevelt Hospital Center.

Control Group (Group 2)

All pulmonary function tests for patients referred to the pulmonary function laboratory in 1995 were reviewed by the authors (E.E. and H.K.). Patients with tests showing obstructive indices were selected as potential participants. Individuals under 18 yr of age, those receiving immunosuppressive therapy, and those with human immunodeficiency virus (HIV) infection or cancer were excluded. Subjects of African-American descent were also excluded from the study as controls because all of the study cases of ₁-ATD were either white or Hispanic. Fifty-four subjects were eligible for the study and 22 agreed to participate. They were interviewed and completed the ATS questionnaire. Skin testing for atopy was performed (see the subsequent discussion). In addition, serum was collected for the measurement of IgE, and was screened for antitrypsin activity (10). Subjects were paid for their participation in the study.

Pulmonary Function Testing

All subjects enrolled in the study underwent pulmonary function testing with a P. K. Morgan system (Benchmark, Gillingham, Kent, UK) rolling-seal spirometer that conformed to ATS standards. Patients were asked to refrain from taking their own bronchodilator medications or caffeine-containing beverages for 24 h prior to testing.

Patients were not tested within 4 wk of an acute respiratory infection. Spirometry procedures followed ATS guidelines and conformed to the ₁-ATD registry protocol (9). All patients completed up to eight maneuvers to generate three acceptable and reproducible spirometric tracings. Two puffs of bronchodilator (albuterol) were administered with a metered dose inhaler (MDI) and testing was repeated after 20 min. A bronchodilator response was defined as an increase of either 12% or 200 ml in FEV₁ or FVC, respectively. Lung volumes were obtained by helium dilution. Diffusion capacity for carbon monoxide (DL_CO) (transfer factor) was obtained by the single-breath method and adjusted for hemoglobin and alveolar volume to yield the diffusion coefficient (KCO). Lung-function data reported in this paper for subjects with ₁-ATD are values obtained during the initial visit.

Evaluation of Atopy

An atopy evaluation was performed by members of the Allergy and Immunology Division of St. Lukes-Roosevelt Hospital Center (D.M. and M.J.), and included examination for a history of rhinitis and ocular or systemic manifestations of allergy or asthma. Group 1 patients underwent intradermal skin testing for IgE-mediated hypersensitivity. Group 2 patients underwent scratch testing followed by intradermal testing if the scratch-testing result was negative. The subjects did not use antihistamines for at least 48 h before skin testing. Skin testing was done with five standard environmental allergen extracts: dust mite (Dermatophagoides farinae), ragweed mix, grass mix (timothy, june, orchard), tree mix (ash, beech, birch, hickory, oak, poplar), and the mold Alternaria tenuis. Atopy was defined as a positive intradermal test to at least one allergen with a 2+ or greater wheal-and-flare response with or without pseudopods (21 to 30 mm of erythema with a 5 to 10-mm wheal). Histamine reactivity was documented by an intradermal injection of 0.05 ml of a solution containing histamine at 1 mg/ml (Greer Laboratories, Lenoir, NC).

Serum IgE Measurement

Serum samples collected from patients with ₁-ATD at the time of skin testing were screened for total IgE with the Fluoro-Fast Test (3M Diagnostics Inc., Santa Clara, CA). IgE levels were not standardized for age or sex. During the study, the laboratory changed the assay method for total IgE to a FEIA-CAP system (Pharmacia Inc., Columbus, OH), and all Group 2 subjects therefore had serum IgE measurements made with the new method. Measurement of the serum IgE concentration in 28 unselected subjects from the allergy clinic showed excellent agreement between the methods (mean IgE: 609 ± 184 IU/ ml 3M method versus 647 ± 182 IU/ml Pharmacia method) with a correlation coefficient (r) of 0.98. An IgE level above 100 IU/ml was considered to indicate a high probability of atopy (8).

Criteria Used for the Diagnosis of Asthma

For each patient, the presence of any three of the following four equally weighted criteria was considered diagnostic of asthma: (1) a history of attacks of wheezing associated with shortness of breath; (2) spirometric evidence of a bronchodilator response; (3) the presence of atopy upon skin testing; (4) a total serum IgE above 100 IU/ml (8).

Statistics

Unless otherwise stated, group data are presented as mean ± SD. Comparisons between two independent data sets were done with group two-tailed t tests. Comparisons of the means of three or more groups were done with analysis of variance (ANOVA) and the extended Tukey's test for multiple comparisons. Comparisons between proportions of different groups were made with standard techniques to derive the Z statistic. IgE values were log-transformed (to assure a normal distribution) before comparison. A value of p  0.05 was considered to reject the null hypothesis and reflect a significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Population

Of the 43 patients in the ₁-ATD study group, data from those referred for a family history of the disease and with normal spirometry (n = 5) were analyzed as a separate group (Group 3). The results for the remaining 38 patients are presented as distinct from this (Group 1). In Group 1, the mean serum ₁-AT was 5.43 ± 1.73 µM (range: 0 to 8.09 µM). Three related patients were of Pi-null-variant phenotype. All others were of PiZ phenotype. The mean ₁-AT value for Group 3 was 5.08 ± 0.85 µM. There was no significant difference between Groups 1 and 3 in mean ₁-AT concentration.

Fifty-four subjects were eligible for participation as the study control group. Of these eligible subjects, 31 refused or could not be contacted, leaving 23 who agreed to participate (Group 2). Of this group, one subject showed a low serum total antitrypsin activity consistent with a heterozygous PiZ phenotype, and was therefore excluded. This left 22 subjects for entry into the study. The percentage of participants, referred for a clinical diagnosis of asthma was 18%, for COPD 50%, and for other diagnoses (usually dyspnea) 32%. These percentages were similar to those for individuals not participating (13% for asthma, 58% for COPD, and 29% for other diagnoses).

The baseline demographic data for these groups is shown in Table 1. The subjects with obstructive lung disease only (Group 2) were significantly older (p < 0.05) than those with ₁-ATD.

Symptom Questionnaire and Self-Reported Diagnosis

Results of the symptom questionnaire and referral diagnosis are given in Table 2. The clinical diagnosis was derived from the patient's self-reported diagnosis and/or the diagnosis stated by the referring physician on the pulmonary function request form. The diagnosis of COPD was broken down to emphysema and/or chronic bronchitis when this information was available. The results show that a history of allergy was common in Groups 1 and 2, being slightly but not significantly more prevalent in Group 1 with ₁-ATD (47% versus 32% respectively). The referral and/or self-reported diagnosis of "asthma" was made in 26% of the patients with ₁-ATD in Group 1, and in a similar proportion (22%) of patients in Group 2. The proportion of subjects in group 1 reporting attacks of wheezing was significantly greater than that in Group 2 (Z = 2.98, p < 0.05). It is notable that the prevalence of wheezing was much greater than the self-reported diagnosis of "asthma" in the subjects in all groups. In Group 3, without spirometric evidence of obstruction, wheezing was also commonly reported (60%), and one subject reported a diagnosis of asthma.

A significantly higher proportion of subjects with ₁-ATD (Group 1 and 3 combined) self-reported attacks of wheezing than did controls (Group 2) (p < 0.05, Z = 2.84).

Chest X-Ray

The chest X-rays (CXR) of Group 1 showed evidence of emphysema with hyperinflation, hyperlucency, or vascular pruning in 32 of the 38 patients (84%). Five patients with airway obstruction had a normal CXR, and one patient had findings consistent with bronchiectasis. The CXR was normal in the five subjects of Group 3. In Group 2, the CXR of 21 subjects was available for review. The previously noted features of emphysema were present in 54%, changes previously associated with COPD were present in 27%, one subject (5%) had evidence of pulmonary artery enlargement and cardiomegaly, and two subjects (9%) had a normal film.

Lung Function

Pulmonary function results for the groups are shown in Table 3. Because one patient in Group 1 was too impaired to undergo pre- and postbronchodilator testing, only postbronchodilator spirometry was performed. Prior spirometry had shown no bronchodilator effect, and this subject was therefore analyzed as non-bronchodilator-responsive. In Group 3, one patient had normal spirometry at baseline and refused to take a bronchodilator. For analysis, this patient was considered to be non-bronchodilator-responsive. Mean pre- and postbronchodilator absolute FEV₁ and FVC values in Groups 1 and 2 were not significantly different, but were significantly less than those in Group 3, in which pulmonary function was normal. FEV₁ %predicted was significantly greater in Group 2 than in Group 1 (p = 0.04). TLC %predicted was slightly higher in Groups 1 and 2 than in Group 3, but not significantly so. On the other hand, mean RV %predicted by helium dilution was significantly greater in Group 1 than in Group 2, suggesting a greater degree of air trapping. Mean RV %predicted in Group 3 was within the normal range for our laboratory. DL_CO %predicted (KCO) was similarly reduced in Groups 1 and 2 and lower than in Group 3, but not significantly so.

Twenty-one of the 38 patients (55%) in Group 1 who had airway obstruction showed evidence of reversibility after bronchodilator administration. The pattern of response was as follows: FEV₁ and FVC were significantly improved in 10 patients, FVC alone was improved in nine patients, and FEV₁ alone was improved in two patients. Mean prebronchodilator absolute and %predicted FEV₁ were significantly lower in the group of patients showing a bronchodilator response than in those who did not (0.93 ± 0.45 L versus 1.84 ± 0.82 L, and 24 ± 11% versus 52 ± 23%, respectively; p < 0.001). Of the 10 patients with a self-reported diagnosis of asthma, five showed a bronchodilator response on testing.

In Group 2, a similar proportion of patients (59%) showed reversibility after aerosolized bronchodilator administration as in Group 1. The pattern of response was also similar to that in Group 1, with five patients showing improvement in FEV₁ and FVC, seven improvement in FVC alone, and one in FEV₁ alone. All five subjects with a self-reported diagnosis of asthma showed bronchodilator responsiveness. FEV₁ %predicted was significantly lower in the subset of patients showing a bronchodilator response (39.9 ± 16.7% versus 58.2 ± 21.6%, p < 0.01). Absolute FEV₁ was also lower in this subset, but not significantly so (1.16 ± 0.49 L versus 1.74 ± 0.88 L). None of the subjects in Group 3 showed a bronchodilator response.

Skin Testing

Skin testing was performed in 35 patients in Group 1. Three patients refused to give permission for testing. A reaction  2+ to one or more intradermal antigens, indicating atopy, was present in 17 patients (48%). One subject gave a history of a positive skin test response to dust, mold, grass, and cat dander 10 yr earlier, yet was negative on retesting. Addition of this patient gave a prevalence of atopy in this group of 51%. The most common reactions were to dust mite (n = 11). Positive reactions to more than one allergen were seen in 12 patients.

In Group 2, one subject gave a history of a severe reaction to prior skin testing, and testing of this subject was therefore not repeated. For analysis, this subject was considered atopic. Six of the 22 subjects in Group 2 (27%) showed a reaction to one or more intradermal antigens. Positive reactions to more than one antigen were seen in four subjects. The most common reactions were to dust (n = 4), mold (n = 5), and grass (n = 4). In Group 3, two of the five subjects (40%) gave a positive response to skin testing.

Comparison of the proportion of patients showing atopy in Groups 1 and 2 did not reveal a significant difference. If the subject in Group 1 who gave a history of atopy but was negative on retesting was included in the analysis as atopic, the difference was significant. In this case, p < 0.05 (Z = 1.8) for a one-tailed comparison.

Serum IgE

In Group 1, serum IgE levels were measured in 35 patients. More than one measurement was performed at different times in 21 patients, yielding a coefficient of variation (CV) of 40%. Mean IgE for the group was 101 ± 248 IU/ml (range: 2 to 1,440 IU/ml). Exclusion of one outlier value of 1,446 IU/ml gave a group mean of 62 ± 86 IU/ml. The IgE level was greater than 100 IU/ml in eight patients (23%). Comparison of the mean IgE level in current smokers, ex-smokers, and nonsmokers showed no significant difference.

In Group 2, the mean serum IgE level was 112 ± 170 IU/ml (range: 2 to 746 IU/ml). Exclusion of one outlier value of 746 IU/ml gave a group mean of 82 ± 97 IU/ml. An IgE exceeding 100 IU/ml was seen in six patients (26%). In Group 3, mean serum IgE was 61 ± 48 IU/ml, with one subject having an IgE above 100 IU/ml. Comparison of the mean total IgE in the three groups and between patients with and without ₁-ATD showed no significant difference.

Relationship Between Reversible Airway Obstruction, IgE, Atopy, and Clinical Diagnosis of Asthma

Table 4 shows the relationship between symptoms and signs associated with asthma and reversible airway obstruction in the three study groups. Complete data were available for 35 subjects in Group 1. Overall, there was no difference between patients with and without bronchodilator responsiveness in Groups 1 and 2 in self-reported and/or clinical diagnoses of asthma and allergy. Among patients with reversible airway obstruction, atopy was significantly more common in those in Group 1 than in those in Group 2 (52% versus 23%) (Z = 1.78, p < 0.05 for a one-tailed comparison).

In Group 1, log IgE was significantly greater in subjects showing a bronchodilator response than in those not showing such a response (p = 0.05). The log IgE data from 57 subjects in Groups 1, 2, and 3 are shown in Figure 1. The mean group value for subjects showing atopy was significantly greater than for those not showing it (1.87 ± 0.62 versus 1.35 ± 0.62 log IU/ ml; p = 0.027). Geometric means for these groups were 74 IU/ ml and 22 IU/ml, respectively.

View larger version (11K): [in this window] [in a new window]   Figure 1.   Individual and mean ± SD log IgE values shown for the group with one or more positive reactions to intradermal testing (skin-test positive) and the group with no reaction (skin-test negative). The results indicate that mean log IgE is significantly greater in the group showing skin test evidence of atopy (1.87 ± 0.62 versus 1.35 ± 0.62, p < 0.05).

Estimation of the Prevalence of Asthma in the Study Groups

The diagnosis of asthma was made in an individual if three of the four criteria defined in METHODS were present. Complete data were available for 35 subjects in Group 1. Three or more criteria for asthma were present in 22% (eight of 35) patients in Group 1, as compared with 4.5% (one of 22) patients in Group 2. By analysis for differences in proportion, this finding was significant for a one-tailed analysis (p < 0.05, Z = 1.92). Absence of all criteria occurred in 6% of patients in Group 1 compared with 26% in Group 2. No subject in Group 3 had more than two criteria for asthma. When the definition of an increased IgE was changed from 100 IU/ml to 74 IU/ml, the group geometric mean for patients with atopy, it made no appreciable difference in the number of diagnostic criteria assigned to each subject.

The proportion of patients with asthma in Group 1 was about the same as the proportion of those who self-reported this diagnosis in this group (21% versus 26%). In contrast, the proportion of those with asthma in Group 2 was lower than those who self-reported the diagnosis (5% versus 22%).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Severe (PiZ) ₁-ATD is a genetic disorder that predisposes one to the early development of emphysema and has a prevalence of one in 3,000 to 5,000 of the United States population (9). We have observed that patients with ₁-ATD-related emphysema may also suffer from environmentally triggered episodes of reversible airway obstruction and wheezing. In the report by Makino and colleagues (2) of 11 patients presenting with dyspnea, five were diagnosed as asthmatic. In a British Thoracic Association survey of 166 patients with the PiZ phenotype (1), 11% were considered to have asthma. Within this subset, 57% gave a history of atopy.

The present study compares the prevalence of asthma in a group of patients with ₁-ATD-related emphysema to that in a group of patients with COPD and normal ₁-AT phenotype. In this comparison we use the term "asthma" to signify the presence of three or more of the following; episodic airway obstruction associated with wheezing, spirometric evidence of a bronchodilator response, skin-test reactivity to intradermal antigens (atopy), and an increased serum IgE (8, 15). We used atopy and an increased IgE level as clinical markers for asthma in patients with chronic airway obstruction and to identify those who might be susceptible to allergen-mediated bronchospasm.

A comparison group of patients with COPD but without ₁-ATD (Group 2) was studied to control for the effects of airway obstruction on the symptoms and signs of asthma. In Group 2, 72% of the patients were referred for airway obstruction that was not clinically asthma. The referral diagnosis was similar for those patients accepted for study and for those not participating. Thus, our control cohort was representative of individuals with airflow obstruction referred to our laboratory for pulmonary function testing.

As expected, the subjects in Group 2 were significantly older than those in Group 1, with ₁-ATD, but had a similar degree of airflow obstruction. The diffusion coefficient (KCO) was equally reduced in both groups to an extent consistent with the presence of emphysema. RV was greater in Group 1, with ₁-ATD but both Groups 1 and 2 showed significant air-trapping. In both groups the chest X-ray commonly showed findings of emphysema that confirmed the nature of the underlying pulmonary disease.

Wheezing is a characteristic symptom in asthma. However, wheezing is also a common symptom in COPD, and is associated with cigarette smoking, airway hyperreactivity, and altered autonomic function (11). In addition, a loss of structural airway support may lead to airway narrowing and wheezing. Attacks of wheezing may, however, be a more discriminatory symptom for the clinical diagnosis of asthma, since it may indicate a triggering event. It is notable that 60% of our Group 1 patients reported this symptom, compared with 23% of those in Group 2. The difference was statistically significant. Of interest was that three of the five subjects with ₁-ATD but without spirometric evidence of airway obstruction reported a history of wheezing that suggested the presence of episodic bronchospasm before the onset of fixed airway obstruction.

A common criterion for the diagnosis of asthma is a spirometric response to a bronchodilator. However, this also occurs in patients with COPD, and is probably related to a low baseline FEV₁ (12, 13). In this regard, we found FEV₁ to be significantly lower in the subset of patients in Groups 1 and 2 who showed a bronchodilator response. Thus, in COPD, the clinical diagnosis of asthma cannot be made on the basis only of wheezing and spirometric criteria. Accordingly, we used atopy and an increased total serum IgE (14, 15) as additional criteria for asthma in our patient groups. Both are strongly associated with asthma, but lack specificity. For example, epidemiologic studies show that atopy exists in up to 40% of persons without asthma as compared with 72% of those with asthma (8). We found that between 48% and 51% of our Group 1 patients, with ₁-ATD, were atopic, but we may have underestimated the prevalence of atopy by limiting the number of skin-test antigens. In addition, atopy may wane over time, as was the case in one of our subjects.

An increased serum IgE level is also not specific for asthma. It is associated with cigarette smoking (5, 6), and may be a marker of airway inflammation. We report that mean serum IgE was increased in both Groups 1 and 2. However, it is unlikely that smoking was the causative factor in this, since the proportion of current smokers in the two groups was small (11% and 22%, respectively). In addition, there was no significant difference between smokers and nonsmokers in mean total serum IgE concentrations. Therefore, airway inflammation is a more likely cause of the increased mean serum IgE in our subjects with ₁-ATD and those with COPD.

Our study indicates that asthma occurs in a subset of individuals with ₁-ATD, in whom it may be triggered by inhaled aeroallergens. First, a greater proportion of our subjects in Group 1, with ₁-ATD (47%), self-reported allergic phenomena than did those in Group 2 (32%). Second, a greater percentage of subjects in Group 1, with ₁-ATD, were atopic than in Group 2 (48% versus 27%), and significantly so when the subject with prior atopy was included. Third, although there was no difference in serum IgE level between Group 1 and 2 for all subjects, those with atopy had a significantly higher mean IgE than those without atopy (Figure 1). Fourth, subjects in Group 1 with ₁-ATD who experienced improvement with bronchodilator administration showed a significantly greater mean serum IgE level than those in this group who did not respond. This was not so for Group 2 (Table 4). Fifth, on the basis of the criteria of atopy, increased serum IgE, episodic wheezing, and bronchodilator responsiveness, significantly more patients in Group 1 than in Group 2 showed clinical characteristics of asthma (21% versus 5%, respectively).

Over the long term, asthma may have an adverse impact on lung function in persons with ₁-ATD. Chronic bronchial-wall inflammation could result in structural remodeling that leads to irreversible narrowing of airways. In this regard, Villar and coworkers reported that atopy and bronchial responsiveness in elderly former and current smokers predisposes to an accelerated decline in FEV₁ (7). Gottleib and associates (16) showed an adverse effect of atopy on FEV₁ decline in normal subjects. Also, an increased total serum IgE contributes to an accelerated decline in FEV₁ in subjects with asthma and in those with COPD but no asthma (5, 18). Similarly, atopy is a marker for accelerated loss of lung function in middle-aged and older men without overt lung disease (16). An inverse relationship between total serum IgE and a longitudinal decline in FEV₁/FVC ratio that is independent of smoking has also been reported (18).

The asthmatic state could be expected to have an adverse impact in ₁-ATD if a lack of bronchial-wall protease inhibition induces airway hyperreactivity. Several clinical studies support this concept. For example, heterozygous PiZ deficiency is associated with more severe asthma (3, 4). In addition, Sigsgaard and colleagues (19) reported that cotton-mill workers with low ₁-AT serum levels and a family history of allergy were more likely to develop byssinosis, characterized by airway obstruction. Once asthma develops in patients with ₁-ATD, the antiprotease and antiinflammatory effectiveness of any ₁-AT present in the airways may be functionally reduced, as Gaillard and associates have reported in asthmatic subjects with a normal serum Pi phenotype (20).

Increased inflammatory mediator release as a result of lack of ₁-AT inhibition may lead to the development of chronic airway hyperreactivity. For example, Hubbard and colleagues (21) reported that alveolar macrophages from patients with ₁-ATD spontaneously release increased amounts of leukotriene B4 (LTB₄), a potent chemotactic agent for neutrophils. A major stimulus for the release of LTB₄ is free neutrophil elastase, which has been found in the lower respiratory tract of patients with ₁-ATD (22). Forteza and coworkers (23) showed that in sheep challenged with Ascaris antigen, pretreatment with aerosolized ₁-AT blocked the development of airway hyperreactivity and the increase in bronchoalveolar lavage fluid (BALF) concentration of tissue kallikrein. Because ₁-AT inhibits airway inflammation induced by neutrophil elastase, its lack may predispose to the development of chronic airway hyperreactivity. Although we did not directly measure airway hyperreactivity through bronchial challenge in our study groups, it is a common occurrence in smokers with mild COPD (17). The high prevalence of bronchodilator responsiveness and wheezing in our subjects is consistent with bronchial hyperreactivity in most of them.

In summary, we report that a significant proportion of patients with severe ₁-ATD and advanced emphysema show clinical features of asthma, and that asthma appears to be more common in patients with this condition than in those with COPD and a normal Pi phenotype. The presence of atopy and an increased serum IgE level indicates that allergic mechanisms could contribute to the development of chronic airway obstruction. We suggest that because individuals with ₁-ATD lack a major antiprotease defense against airway inflammation, they are more susceptible to allergen-mediated asthma and consequent progressive airway obstruction. The presence of atopy represents a genetic predisposition to this process. Such patients may be candidates for measures aimed at reducing the impact of environmental aeroallergens (24). In addition, increasing the concentration of ₁-AT in these patient's airways might ameliorate the effect that environmental factors have on them.