ORIGINAL RESEARCH Validation of Inhalation Provocation Test in Chronic Bird-Related Hypersensitivity Pneumonitis and New Prediction Score Masahiro Ishizuka, Yasunari Miyazaki, Tomoya Tateishi, Toshiharu Tsutsui, Kimitake Tsuchiya, and Naohiko Inase Tokyo Medical and Dental University, Department of Respiratory Medicine, Tokyo, Japan
Abstract Rationale: Hypersensitivity pneumonitis (HP) is an immunologically mediated lung disease induced by the inhalation of any of a wide variety of antigens. For example, bird-related HP (BRHP) results from the inhalation of avian antigens. The clinical features of chronic HP, including imaging and histological findings, are similar to those of idiopathic pulmonary fibrosis. Despite its status as the “gold standard,” the inhalation provocation test (IPT) is rarely performed, because the methods and the criteria are not standardized. In 2000, we reported the utility of IPT for pigeon dropping extracts. Objectives: The purpose of the current study was to validate the utility and safety of the test, and to differentiate chronic HP from other interstitial lung diseases. Methods: A total of 28 patients with chronic BRHP and 19 control subjects were evaluated in this retrospective study. We validated the previous criteria and proposed new criteria using prediction scores.
Measurements and Main Results: In the IPT using pigeon dropping extracts, the previous criteria showed a sensitivity of 78.6% and a specificity of 94.7% in this retrospective study. The increases in the peripheral white blood cell count and C-reactive protein levels are good indicators of a positive response to the inhalation challenge. We propose the use of the IPT prediction score (ΔWBC [%] 1 2 3 ΔP[A 2 a]O2 [mm Hg], where WBC is white blood cell) and the prediction rule, which showed high sensitivity and specificity values of 92.9 and 94.7%, respectively. Two (1.5%) out of a total of 130 subjects who underwent the tests required treatment after the challenge. Conclusions: The IPT is a useful and safe tool for the diagnosis of chronic HP. The IPT prediction score that we have proposed has high sensitivity and specificity in the diagnosis of chronic BRHP. Keywords: inhalation provocation test; chronic hypersensitivity pneumonitis; validation; prediction score
(Received in original form August 2, 2014; accepted in final form December 30, 2014 ) Supported by a grant to the Diffuse Lung Diseases Research Group from the Ministry of Health, Labor and Welfare, Japan (N.I.) and by a grant by the Japan Society for the Promotion of Science 235,591,140 (Y.M.). Author Contributions: M.I. and Y.M. contributed to study design, data collection, data analysis, statistical analysis, and manuscript preparation; T. Tateishi contributed to data collection and data analysis; T. Tsutsui contributed to data collection; K.T. contributed to study design and data collection; N.I. contributed to study design and manuscript preparation. Correspondence and requests for reprints should be addressed to Yasunari Miyazaki, M.D., Health Administration Center, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Ann Am Thorac Soc Vol 12, No 2, pp 167–173, Feb 2015 Copyright © 2015 by the American Thoracic Society DOI: 10.1513/AnnalsATS.201408-350OC Internet address: www.atsjournals.org
Hypersensitivity pneumonitis (HP) is an immunologically mediated lung disease induced by the inhalation of any of a wide variety of antigens (1). For example, birdrelated HP (BRHP) is one of the common HPs resulting from exposure to avian antigens, such as those found in pigeon droppings and feathers (2). The presence of IgG class–specific antibodies in most cases of HP suggests that a type III
hypersensitivity mechanism may be responsible for the pathogenesis. In addition, a type IV hypersensitivity mechanism mediated by T cells is currently known to be involved (3). There appear to be three types of HP based on the clinical features: acute, subacute, and chronic (1, 4, 5). The symptoms of acute HP occur 4–6 hours after exposure to the etiologic antigen and
consist of the abrupt onset of a flu-like syndrome characterized by fever, chills, malaise, and myalgias (5). The respiratory symptoms include dyspnea, tightness of the chest, and a nonproductive cough. Acute BRHP reproduces the following symptoms after exposure to avian antigens: IgG and/or IgA antibodies specific to pigeon dropping extracts (PDEs) in the sera or bronchoalveolar lavage (BAL) fluids; the
Ishizuka, Miyazaki, Tateishi, et al.: Inhalation Provocation Test in Chronic Bird-Related HP
167
ORIGINAL RESEARCH proliferation of either BAL or peripheral lymphocytes within the pigeon sera; lymphocytosis in the BAL fluids; and alveolitis or granulomatous lesions in the lungs (6, 7). Data from European countries indicate that HP represents 4–15% of all interstitial lung diseases (8), with an incidence rate of 0.9 cases per 100,000 person-years (9). The most common forms of HP are farmer’s lung and pigeon breeder’s lung, but the proportion of individuals who are exposed to these antigens and will develop HP is unknown. An epidemiologic study of acute HP in Japan demonstrated that summertype HP was the most prevalent form of the disease (10). An epidemiologic survey of chronic HP in Japan revealed that BRHP was the most prevalent HP, representing over 60% of all chronic HP cases (222 cases); summer-type HP accounted for only 15% (11). After prolonged exposure to small amounts of an avian antigen when the patient could have been considered asymptomatic, chronic BRHP may develop due to the fluctuating acute episodes, including mild exertional dyspnea and cough (recurrent type) or the absence of a history of acute episodes (insidious type) (5, 12). In both recurrent and insidious chronic HP, patients develop symptoms of increasing dyspnea upon exertion, as well as fatigue, anorexia, cough, and weight loss after interstitial inflammation and fibrosis (4). The clinical features of chronic BRHP, including the imaging and histological findings, are similar to those of idiopathic pulmonary fibrosis (IPF). Because of the difficulty inherent in diagnosing chronic HP, we have observed an increasing number of cases of chronic HP that had previously been misdiagnosed as IPF. In clinical practice, avian contact is the only clue as to a potential diagnosis of BRHP (13). Ram´ırez-Venegas and colleagues (14) reported the utility of the inhalation provocation test (IPT) using pigeon serum for the diagnosis of chronic BRHP. Morell and colleagues (15) proposed that the IPT be considered the “gold standard” among noninvasive BRHP diagnostic examinations, focusing on the decreased FVC and/or the diffusing capacity of the lungs for carbon monoxide (DLCO). In spite of its status as the “gold standard,” the IPT is rarely performed, because the methods and the criteria are not 168
standardized. The duration of the nebulization and the concentrations of extract used are not standardized, and both parameters vary across studies. Munoz and colleagues (16) discussed how the test should be performed and which parameters should be assessed to ensure a correct interpretation. Possible HP may sometimes be mistaken for other interstitial lung diseases. In 2000, we reported the utility of the IPT using PDE and proposed the diagnostic criteria for chronic BRHP (17). The purpose of the current study was to evaluate the validity of the criteria in the diagnosis of chronic BRHP and the safety of this test. We also propose a new prediction score (PS) based on the analysis of the new data presented in this study.
Methods Informed Consent
Informed consent was obtained from all subjects, and this study was approved by institutional review boards at the Tokyo Medical and Dental University. Criteria for Chronic HP
We used the Yoshizawa’s criteria (18) for chronic HP for screening chronic BRHP. The criteria were three or more of the following (including either 1 or 2, either 3 or 4, and either 5 or 6): (1) reproduction of symptoms of HP by an environmental provocation or laboratory-controlled inhalation of the avian antigen; (2) positive titer of antibodies and/or lymphocyte proliferation upon exposure to an avian antigen; (3) evidence of pulmonary fibrosis with or without granulomas detected in histology; (4) a finding of honeycombing on computed tomography (CT) scans; (5) progressive deterioration of a restrictive impairment of pulmonary function over 1 year; and (6) existence of a 6-month or longer duration of respiratory symptoms related to the disease. Retrospective Study Subjects
A retrospective review of the medical records of patients who were recruited after sequential hospitalizations between April 2000 and March 2011 was undertaken at the Tokyo Medical and Dental University Hospital in Japan. Among the patients who underwent the IPT, we first separated out the subjects who did not fulfill Yoshizawa’s
criteria. Then, to select the core group of BRHP-positive patients to validate the IPT, we classified the patients whose lung tissues from surgical lung biopsies were compatible with the histopathological features of chronic HPs, including centrilobular fibrosis and bridging fibrosis; these patients were defined as the BRHP group. Among patients who did not meet the criteria for inclusion in the BRHP group, those who were diagnosed with other interstitial lung disease via a separate etiology, such as HPs by other antigens or collagen vascular diseases, were assigned to the control group. Finally, the remaining subjects were assigned to a group with “no definite diagnosis.” Inhalation Challenge
Patients and control subjects inhaled 2 ml of PDE through a hand nebulizer. The maximal duration of exposure to this antigen was approximately 10 minutes (17). Clinical symptoms, including cough, dyspnea, chills, and body temperature (BT), were evaluated immediately before and hourly during the 24 hours after the challenge. White blood cell (WBC) counts, C-reactive protein (CRP), and arterial blood gas were measured immediately before and 6 and 24 hours after the challenge. Pulmonary function tests, including measurement of the vital capacity (VC) and the DLCO, were performed immediately before and 24 hours after the inhalation challenge. Chest radiographs were taken at similar intervals. Criteria for the IPT
We previously reported monitoring parameters to determine the criteria for the inhalation provocation challenge in 2000 (17). Subjects who fulfilled two or more of the following were considered positive: (1) increased radiologic abnormalities; (2) increase in the alveolar–arterial oxygen pressure difference (P[A 2 a]O2) by more than 10 mm Hg and/or a decrease of DLCO by more than 20%; (3) decreased VC by more than 15%; (4) increase in the peripheral WBC count by more than 30%; (5) increased CRP by more than 1.0 mg/dl; (6) increased BT by more than 1.08 C and/or the development of systemic manifestations, including chills and general fatigue; and (7) development of respiratory symptoms (cough and dyspnea).
AnnalsATS Volume 12 Number 2 | February 2015
ORIGINAL RESEARCH Evaluation of High-Resolution CT
We evaluated the degree of fibrotic change in both the BRHP group and the control group using the Kazerooni fibrosis score in high-resolution CT scans (19). The details of the fibrosis score are explained in the online supplement. The high-resolution CT scans were reviewed independently by two experienced respiratory physicians (Y.M. and M.I.).
April 2000 – March 2011 Hospitalizations for interstitial lung disease (ILD) n=755
April 2000 – March 2011 Inhalation provocation test n=130
Out of chronic HP criteria n=9
Statistical Analysis
Data are expressed as the means (6SEM). The results were compared using the Mann-Whitney U test or unpaired t test. Comparisons between the groups were performed using the x2 test or Fisher’s exact test for categorical variables. All statistical comparisons between the chronic BRHP group and the control group were two sided and were performed at the 0.05 significance level. These statistical analyses were performed using GraphPad PRISM (GraphPad software, San Diego, CA). We incorporated the variables found to be significant at the 0.05 level in a stepwise logistic regression model to reduce the list of diagnostic criteria using SPSS Statistics (IBM, New York, NY). We proposed the IPT-PS using the ratio of the regression coefficients.
Results Patient Characteristics in Retrospective Study Samples
Between 2000 and 2011, 755 patients were hospitalized due to interstitial lung diseases. Among them, 130 patients underwent the IPT for the diagnosis of BRHP. We separated out nine patients who did not fulfill Yoshizawa’s criteria. Among the remaining subjects, 28 patients were classified into the BRHP group. Among those who did not meet the criteria for inclusion in the BRHP group, 19 patients who were diagnosed with other interstitial lung diseases via a separate etiology (including 6 HPs associated with other antigens, 12 collagen vascular diseases, and 1 drug-induced interstitial pneumonia) were assigned to the control group (Figure 1). The remaining 74 subjects were assigned to a group with “no definite diagnosis.” A total of 16 of the 28 patients in the BRHP group and 11 of the 19 subjects in the control group were smokers (Table 1); 11 of the 28 patients in the BRHP group raised budgerigars or bred pigeons. Among the
BRHP group Diagnosed by surgical lung biopsy n=28
Control group Diagnosed as other ILD n=19
No definite diagnosis n=74
Figure 1. Diagram illustrating each step in the patient classification process according to the Yoshizawa’s criteria of chronic hypersensitivity pneumonitis (HP) and the histopathological evaluation by surgical lung biopsies. BRHP = bird-related HP.
remaining patients with BRHP, 11 patients used feather quilts, and six patients were most likely exposed to avian antigens in the surrounding environment. In the BRHP group, five patients were classified into the recurrent type, whereas the remaining 23 patients were classified into the insidious type. A total of 21 of the 28 patients in the BRHP group and all 19 of the subjects in the control group had antibodies to PDE in their sera and/or BAL fluids; 21 of the 28 patients in the BRHP group and 10 of the 19 subjects in the control group had positive proliferative responses of the peripheral or bronchoalveolar lymphocytes in response to the pigeon sera. All subjects in the BRHP group and the control group presented honeycombing, traction bronchiectasis, or intralobular septal
thickening. In addition, all patients in the BRHP group had compatible histopathological features in surgical lung biopsies. We evaluated the extent of fibrosis using the Kazerooni fibrosis score in Table 1 (19). The means of the scores in the BRHP group and the control group were 1.97 (60.15; mean [6SEM]) and 2.10 (60.19), respectively (P = 0.565). The results of the tests for pulmonary function are shown in Table 1, and there were no significant differences between the two groups. Clinical and Laboratory Data after the Inhalation Challenge
After the inhalation challenge, 9 (32%) of the 28 patients in the BRHP group showed radiologic changes, but none of the 19 subjects in the control group showed similar
Table 1. Subject profile in the study
Age, yr, median (25th–75th percentiles) Sex, male/female, n Smoking history, ever/never, n Specific antibody, positive/negative, n LST, positive/negative, n VC, % predicted, mean 6 SEM FEV1, % predicted, mean 6 SEM DLCO, % predicted, mean 6 SEM PaO2, mm Hg, mean 6 SEM PaCO2, mm Hg, mean 6 SEM Fibrosis score, mean 6 SEM
BRHP (n = 28)
Control (n = 19)
P Value
64 (58–68) 18/10 16/12 21/7 21/6 80.4 6 3.7 75.5 6 3.4 58.6 6 4.6 81.3 6 2.1 43.1 6 0.7 1.97 6 0.15
66 (58–70) 10/9 11/8 19/0 10/9 77.6 6 4.0 72.6 6 3.6 53.9 6 4.7 81.6 6 2.7 40.7 6 1.2 2.10 6 0.19
0.362 0.547 1.000 0.032 0.111 0.617 0.567 0.504 0.925 0.064 0.565
Definition of abbreviations: BRHP = bird-related hypersensitivity pneumonitis; DLCO = lung diffusing capacity for carbon monoxide; LST = lymphocyte stimulating test; VC = vital capacity.
Ishizuka, Miyazaki, Tateishi, et al.: Inhalation Provocation Test in Chronic Bird-Related HP
169
ORIGINAL RESEARCH changes (P = 0.007). A total of 12 (43%) of the 28 patients in the BRHP group and 3 (16%) of the remaining 19 subjects developed respiratory symptoms (cough and dyspnea) after the inhalation challenge (P = 0.209). We evaluated the increase in the peripheral WBC count with the following measures: receiver operating characteristic (ROC) curve analysis at 6 hours (area under the curve [AUC] = 0.941, P , 0.001) and 24 hours (AUC = 0.682, P = 0.036) after the challenge; the increase of CRP by the ROC curve at 6 hours (AUC = 0.586, P = 0.324) and 24 hours (AUC = 0.878, P , 0.001) after the challenge; the increase of P(A 2 a)O2 by the ROC curve at 6 hours (AUC = 0.844, P , 0.001) and 24 hours (AUC = 0.588, P = 0.308) after the challenge; the decrease of VC (AUC = 0.503, P = 0.974) by the ROC curve at 24 hours after the challenge; the decrease of DLCO (AUC = 0.519, P = 0.832) by the ROC curve at 24 hours the after challenge; and the increase of BT by the ROC curve (AUC = 0.753, P = 0.004) for the 24 hours after the challenge (Table 2).
Table 2. The areas under the curves and P Values by the receiver operating characteristic curves in 6 and 24 hours in the study Variables 6 h after D%WBC DCRP DP(A 2 a)O2 D%VC* D%DLCO* 24 h after D%WBC DCRP DP(A 2 a)O2 D%VC* D%DLCO* DBT†
AUC
P Value
0.941 0.586 0.844 — —
,0.001 0.324 ,0.001 — —
0.682 0.878 0.588 0.503 0.519 0.753
0.036 ,0.001 0.308 0.974 0.832 0.004
Definition of abbreviations: AUC = area under the curve; BT = body temperature; CRP = C-reactive protein; DP(A 2 a)O2 = alveolar–arterial oxygen pressure difference; DLCO = lung diffusing capacity for carbon monoxide; VC = vital capacity; WBC = white blood cell. *VC and DLCO were measured only immediately before and 24 hours after the inhalation challenge. † BT was measured immediately before and hourly during the 24 hours after the challenge, and we evaluated the difference between the maximum and initial values.
170
Safety of the Inhalation Challenge
Most subjects recovered in 1 or 2 days after the challenge. Two (1.5%) of the 130 total subjects who underwent the IPT required treatment for changes related to the challenge. During the IPT, the first case increased the ground-glass opacity bilaterally, and the P(A 2 a)O2 increased from 44.7 to 115 mm Hg. The second case also increased the ground-glass opacity bilaterally, and the P(A 2 a)O2 increased from 28.4 to 131 mm Hg. These patients received oxygen and steroid pulse therapies without intubation or an intensive care unit stay. Both patients recovered to the baseline oxygenation levels. The first patient had respiratory failure (PaO2 , 60 mm Hg) and a high CRP level before the IPT, whereas the second patient had a high level of antibodies to PDE. Validation of the Previous Criteria for the Inhalation Provocation Challenge
A total of 22 (78.6%) of the 28 patients in the BRHP group in this study were classified as positive or most likely positive compared with 100% (all 11 patients) in our previous report (17). A total of 18 (94.7%) of the 19 subjects in the control group in this study were classified as negative compared with 100% (all 6 subjects) in our previous report (17). In the current study, the model had a sensitivity of 78.6% and a specificity of 94.7%. IPT-PS
We validated the utility of the IPT based on the previous criteria, but this interpretation was hardly sufficient to improve our diagnosis of chronic BRHP. We incorporated the variables that were found to be significant at the 0.05 level into a stepwise logistic regression model to reduce the monitoring parameters in the diagnostic criteria. The item of radiologic change was excluded from this analysis,
because no patients in the control group showed an increase in radiologic abnormalities after the challenge. The item of CRP was also excluded from this analysis, because there were intermediate correlations between the CRP and the peripheral WBC count (r = 0.599) and between CRP and P(A 2 a)O2 (r = 0.461). We thus propose the IPT-PS, which uses the ratios of the regression coefficients (Table 3), as follows: IPT-PS = ΔWBC (%) 1 2 3 ΔP(A 2 a)O2 (mm Hg). We showed the ROC curve for the prediction rule by the IPT-PS (AUC = 0.966, P , 0.001; Figure 2). The cut-off value of the IPT-PS was set to 35 (Figure 3), and the prediction rule showed high sensitivity and specificity values of 92.9% (26 of the 28 patients) and 94.7% (18 of the 19 subjects), respectively. A total of 60% of the subjects in the “no definitive diagnosis” group had IPT-PS values that were higher than the cut off. We were able to diagnose the 74 subjects as having BRHP by Yoshizawa’s criteria (18); however, other interstitial pneumonias can be included by the criteria, because the immunological criterion in Yoshizawa’s criteria did not exhibit high sensitivity, although it did present high specificity, according to our previous study (20). We compared the measurements between a higher IPT-PS group and a lower IPT-PS group. There was a significant difference between these two groups only in the values of IgG antibodies to PDE in the BAL fluids.
Discussion Here, we have validated the previous criteria for the diagnosis of BRHP. The sensitivity and specificity of the previous criteria were 78.6 and 94.7% in the current study compared with 100 and 100%, respectively, in our previous report (17). We confirmed
Table 3. Predictors of the inhalation provocation test Variables
Coefficient
Increase in peripheral WBC count, % Increase in P(A 2 a)O2, mm Hg Increase in BT, 8 C Development of respiratory symptoms Intercept
0.18 0.39 2.15 22.54 26.98
OR (95% CI) 1.20 1.48 11.7 0.08
(1.05–1.37) (1.07–2.04) (0.4–184.4) (0.001–4.5) —
P Value 0.01 0.02 0.17 0.22 —
Definition of abbreviations: BT = body temperature; CI = confidence interval; OR = odds ratio; P(A 2 a)O2 = alveolar–arterial oxygen pressure difference; WBC = white blood cell.
AnnalsATS Volume 12 Number 2 | February 2015
ORIGINAL RESEARCH
100
Sensitivity%
80 60 40 20 0 0
20 40 60 80 100% - Specificity%
100
Figure 2. Receiver operating characteristic curve for the prediction rule by the inhalation provocation test prediction score in the study.
that the criteria maintained high specificity, which is useful for the diagnosis of chronic BRHP. In the current study, we indicated that the changes in the peripheral WBC count, CRP, P(A 2 a)O2, and BT were useful monitoring parameters for the test. We proposed a new PS, the IPT-PS, which included the peripheral WBC count and P(A 2 a)O2 and had a higher sensitivity (92.9%) than the previous criteria (78.6%). We certified that the proposed scale was more clinically useful for diagnosis than the previous criteria. We evaluated the changes in the peripheral WBC count, CRP, P(A 2 a)O2, BT, VC, and DLCO by the ROC curve. The peripheral WBC count, CRP, P(A 2 a)O2, and BT showed a significant difference, but the VC and DLCO showed no significant difference. Morell and colleagues (15) reported that the FVC and DLCO are good indicators of a positive response to the inhalation challenge. In their report, the FVC and DLCO were recorded at baseline, 20 minutes after the inhalation, and every
Number of positive criteria
200 150 IPT-PS
hour for the next 8 hours. In the current study, there were no significant differences between the BRHP group and the control group in VC and DLCO, because these parameters were only recorded at baseline and 24 hours after the challenge. Ram´ırezVenegas and colleagues (14) reported that fever was the major symptom observed in all of the patients with a positive response to the inhalation challenge, and usually preceded all other clinical symptoms. The current study demonstrates that not only fever, but also peripheral leukocytosis, an increase of CRP, and an increase of P(A 2 a)O2, most likely act as good indicators of a positive response to the inhalation challenge. The IPT-PS proposed herein requires only two monitoring parameters: the peripheral WBC count and P(A 2 a)O2. In chronic HP, there is no diagnostic method that has as high a sensitivity as IPT. The diagnosis of BRHP is generally based on the presence of consistent clinical symptoms, which provide confirmation that the patient has been exposed and is sensitized to a causal agent (including high levels of antibodies to avian antigens), consistent chest X-rays or CT imaging findings, evidence of increased lymphocytes in BAL fluids, and demonstration of consistent histologic features (17). However, identifying the causal agent is not simple, even when the antibodies to the avian antigen are measured. The National Heart, Lung, and Blood Institute/Office of Rare Diseases workshop (21) and the Hypersensitivity Pneumonitis Study Group (22) agree that HP is caused by the inhalation of an antigen to which the patient is sensitized and hyperresponsive, and that sensitization and exposure alone in
100 50 35 0 –50 BRHP
Control
7 6 5 4 3 2 1 0 BRHP
Control
Figure 3. The scatter plot of inhalation provocation test prediction scores (IPT-PS) in the bird-related hypersensitivity pneumonitis (BRHP) group and the control group (left) compared with the number of positive parameters for the previous criteria (right) in the study. The cut-off value for IPT-PS was set at 35, which is given by the receiver operating characteristic curve.
the absence of symptoms do not define the disease, regardless of the presence of specific IgG antibodies. Costabel and colleagues (3) demonstrated that 30–60% of healthy farmers produced precipitating antibodies to the antigens to which they were exposed, whereas 10–15% of patients do not develop serum precipitins. Therefore, a negative finding does not rule out the presence of disease. In our previous report (20), the immunological components of Yoshizawa’s criteria, including antibodies to PDE and a proliferative response of lymphocytes to the pigeon sera, had low sensitivities (z40%) and high specificities (z80–90%). The subjects in the “no definite diagnosis” group in the current study include “suspected BRHP,” but not definite BRHP, mostly because the immunological criterion has a low AUC (z0.6) in ROC analysis (20). Therefore, the IPT-PS for BRHP has a higher diagnostic capability than Yoshizawa’s criteria. The clinical features of chronic HP, including imaging and histological findings, are similar to those of IPF. Centrilobular nodules, lobular air trapping, and basilar sparing are useful findings to distinguish chronic HP from IPF for some patients (23). However, we often find that patients with chronic HP present a usual interstitial pneumonia (UIP) pattern, particularly in cases of insidious-type chronic HP (24), and the distinction from IPF is difficult. In our previous report, there was no difference in the extent of the ground-glass opacities, consolidation, honeycombing, or emphysema in patients with chronic HP presenting a UIP pattern and IPF. In this respect, the IPT-PS is more useful than radiological findings for distinguishing chronic HP presenting a UIP pattern from IPF. Ram´ırez-Venegas and colleagues (14), Morell and colleagues (15), and Ohtani and colleagues (17) reported the utility and safety of the IPT for the diagnosis of chronic BRHP. In the current study, we further demonstrated the validity and safety of the IPT for individuals with chronic BRHP at the antigen concentration used. Two (1.5%) of the 130 subjects who underwent the IPT required steroid pulse therapy for the changes caused by the challenges. One patient had respiratory failure and had a high CRP level before the IPT, and the other patient had a high titer of antibodies against the avian antigen. Munoz and colleagues (16) recommended
Ishizuka, Miyazaki, Tateishi, et al.: Inhalation Provocation Test in Chronic Bird-Related HP
171
ORIGINAL RESEARCH that a low FVC and DLCO with respiratory insufficiency (PaO2 , 60 mm Hg) should constitute contraindications. In fact, the PaO2 of the patient with respiratory failure was the lowest of all the subjects. In addition, a high CRP value (2.4 mg/dl) was recorded at admission, and the challenge was conducted a day after admission. The challenge was also conducted a day after admission for the patient with a high titer of antibodies against the avian antigen, and we considered this to be adequate antigen avoidance, because a time period of 2 weeks of hospitalization was needed before the challenge. The limitations of the present study include the retrospective analysis and the time period during which the respiratory function tests were performed. One weakness of the study was the restriction
due to the choice of antigen, even though avian antigens represent the most common etiology in chronic HP. The utility of the IPT with avian antigens in the diagnosis of BRHP has been reported in four studies (14, 15, 17, 25). Furthermore, fungi (26, 27) and bacteria (28) were reported as antigens of HP, and have been used for diagnosis in IPTs. BRHP is reported to account for 34–68% of HP in the world according to the four studies (29–32). Avian antigens were among the major antigens in chronic HP; however, more than one-third of the patients had other causative antigens. Therefore, we are now working on developing IPTs for other antigens. To prioritize the development and use of less-invasive tests, we conducted environmental investigations and immunological tests first. Then, we
References 1 Fink JN. Hypersensitivity pneumonitis. J Allergy Clin Immunol 1984;74: 1–10. 2 Todd A, Coan RM, Allen A. Pigeon breeders’ lung: pigeon intestinal mucin, an antigen distinct from pigeon IgA. Clin Exp Immunol 1991; 85:453–458. 3 Costabel U, Bonella F, Guzman J. Chronic hypersensitivity pneumonitis. Clin Chest Med 2012;33:151–163. 4 Kaltreider HB. Hypersensitivity pneumonitis. West J Med 1993;159: 570–578. 5 Salvaggio JE, Millhollon BW. Allergic alveolitis: new insights into old mysteries. Respir Med 1993;87:495–501. 6 Selman M, Teran ´ L, Mendoza A, Camarena A, Mart´ınez-Cordero E, Lezama M, Rubio HM. Increase of HLA-DR7 in pigeon breeder’s lung in a Mexican population. Clin Immunol Immunopathol 1987;44: 63–70. 7 Coleman A, Colby TV. Histologic diagnosis of extrinsic allergic alveolitis. Am J Surg Pathol 1988;12:514–518. 8 Thomeer MJ, Costabe U, Rizzato G, Poletti V, Demedts M. Comparison of registries of interstitial lung diseases in three European countries. Eur Respir J Suppl 2001;32:114s–118s. 9 Solaymani-Dodaran M, West J, Smith C, Hubbard R. Extrinsic allergic alveolitis: incidence and mortality in the general population. QJM 2007;100:233–237. 10 Ando M, Konishi K, Yoneda R, Tamura M. Difference in the phenotypes of bronchoalveolar lavage lymphocytes in patients with summer-type hypersensitivity pneumonitis, farmer’s lung, ventilation pneumonitis, and bird fancier’s lung: report of a nationwide epidemiologic study in Japan. J Allergy Clin Immunol 1991;87:1002–1009. 11 Okamoto T, Miyazaki Y, Ogura T, Chida K, Kohno N, Kohno S, Taniguchi H, Akagawa S, Mochizuki Y, Yamauchi K, et al. Nationwide epidemiological survey of chronic hypersensitivity pneumonitis in Japan. Respir Investig 2013;51:191–199. 12 Ohtani Y, Saiki S, Sumi Y, Inase N, Miyake S, Costabel U, Yoshizawa Y. Clinical features of recurrent and insidious chronic bird fancier’s lung. Ann Allergy Asthma Immunol 2003;90:604–610. 13 Perez-Padilla ´ R, Salas J, Chapela R, Sanchez ´ M, Carrillo G, Perez ´ R, Sansores R, Gaxiola M, Selman M. Mortality in Mexican patients with chronic pigeon breeder’s lung compared with those with usual interstitial pneumonia. Am Rev Respir Dis 1993;148:49–53. 14 Ram´ırez-Venegas A, Sansores RH, Perez-Padilla ´ R, Carrillo G, Selman M. Utility of a provocation test for diagnosis of chronic pigeon breeder’s disease. Am J Respir Crit Care Med 1998;158:862–869.
172
performed the IPT for suspicious antigens one at a time using the test described previously here. The described procedure should be implemented for other causative antigens, and the criteria should be verified using the IPT-PS. In summary, this test could become a valuable addition to the current diagnostic criteria, and it might be useful for differentiating chronic HP from interstitial lung diseases, including IPF, as a noninvasive examination due to its utility and safety. n Author disclosures are available with the text of this article at www.atsjournals.org. Acknowledgment: The authors thank Makoto Tomita and Masako Akiyama for their assistance with aspects of the data analysis.
15 Morell F, Roger A, Reyes L, Cruz MJ, Murio C, Muñoz X. Bird fancier’s lung: a series of 86 patients. Medicine (Baltimore) 2008;87:110–130. 16 Munoz X, Morell F, Cruz MJ. The use of specific inhalation challenge in hypersensitivity pneumonitis. Curr Opin Allergy Clin Immunol 2013; 13:151–158. 17 Ohtani Y, Kojima K, Sumi Y, Sawada M, Inase N, Miyake S, Yoshizawa Y. Inhalation provocation tests in chronic bird fancier’s lung. Chest 2000;118:1382–1389. 18 Yoshizawa Y, Ohtani Y, Hayakawa H, Sato A, Suga M, Ando M. Chronic hypersensitivity pneumonitis in Japan: a nationwide epidemiologic survey. J Allergy Clin Immunol 1999;103:315–320. 19 Kazerooni EA, Martinez FJ, Flint A, Jamadar DA, Gross BH, Spizarny DL, Cascade PN, Whyte RI, Lynch JP III, Toews G. Thin-section CT obtained at 10-mm increments versus limited three-level thin-section CT for idiopathic pulmonary fibrosis: correlation with pathologic scoring. AJR Am J Roentgenol 1997;169:977–983. 20 Suhara K, Miyazaki Y, Okamoto T, Yasui M, Tsuchiya K, Inase N. Utility of immunological tests for bird-related hypersensitivity pneumonitis. Respir Investig 2015;53:13–21. 21 Fink JN, Ortega HG, Reynolds HY, Cormier YF, Fan LL, Franks TJ, Kreiss K, Kunkel S, Lynch D, Quirce S, et al. Needs and opportunities for research in hypersensitivity pneumonitis. Am J Respir Crit Care Med 2005;171:792–798. 22 Lacasse Y, Selman M, Costabel U, Dalphin JC, Ando M, Morell F, Erkinjuntti-Pekkanen R, Muller N, Colby TV, Schuyler M, et al.; HP Study Group. Clinical diagnosis of hypersensitivity pneumonitis. Am J Respir Crit Care Med 2003;168:952–958. 23 Silva CI, Muller ¨ NL, Lynch DA, Curran-Everett D, Brown KK, Lee KS, Chung MP, Churg A. Chronic hypersensitivity pneumonitis: differentiation from idiopathic pulmonary fibrosis and nonspecific interstitial pneumonia by using thin-section CT. Radiology 2008;246:288–297. 24 Tateishi T, Ohtani Y, Takemura T, Akashi T, Miyazaki Y, Inase N, Yoshizawa Y. Serial high-resolution computed tomography findings of acute and chronic hypersensitivity pneumonitis induced by avian antigen. J Comput Assist Tomogr 2011;35:272–279. 25 Hendrick DJ, Marshall R, Faux JA, Krall JM. Positive “alveolar” responses to antigen inhalation provocation tests: their validity and recognition. Thorax 1980;35:415–427. 26 Morell F, Cruz MJ, Gomez ´ FP, Rodriguez-Jerez F, Xaubet A, Muñoz X. Chacinero’s lung—hypersensitivity pneumonitis due to dry sausage dust. Scand J Work Environ Health 2011;37:349–356. 27 Mitsui C, Taniguchi M, Fukutomi Y, Saito A, Kawakami Y, Mori A, Akiyama K. Non occupational chronic hypersensitivity pneumonitis due to Aspergillus fumigatus on leaky walls. Allergol Int 2012;61: 501–502.
AnnalsATS Volume 12 Number 2 | February 2015
ORIGINAL RESEARCH 28 Williams JV. Inhalation and skin tests with extracts of hay and fungi in patients with farmer’s lung. Thorax 1963;18:182–196. 29 Xaubet A, Ancochea J, Morell F, Rodriguez-Arias JM, Villena V, Blanquer R, Montero C, Sueiro A, Disdier C, Vendrell M; Spanish Group on Interstitial Lung Diseases, SEPAR. Report on the incidence of interstitial lung diseases in Spain. Sarcoidosis Vasc Diffuse Lung Dis 2004;21:64–70. 30 Selman M, Lacasse Y, Pardo A, Cormier Y. Hypersensitivity pneumonitis caused by fungi. Proc Am Thorac Soc 2010;7:229–236.
31 Roelandt M, Demedts M, Callebaut W, Coolen D, Slabbynck H, Bockaert J, Kips J, Brie J, Ulburghs M, De Boeck K, et al. Epidemiology of interstitial lung disease (ILD) in Flanders: registration by pneumologists in 1992–1994. Working group on ILD, VRGT. Vereniging voor Respiratoire Gezondheidszorg en Tuberculosebestrijding. Acta Clin Belg 1995;50:260–268. 32 Hanak V, Golbin JM, Ryu JH. Causes and presenting features in 85 consecutive patients with hypersensitivity pneumonitis. Mayo Clin Proc 2007;82:812–816.
Ishizuka, Miyazaki, Tateishi, et al.: Inhalation Provocation Test in Chronic Bird-Related HP
173
Abstract Rationale: Hypersensitivity pneumonitis (HP) is an immunologically mediated lung disease induced by the inhalation of any of a wide variety of antigens. For example, bird-related HP (BRHP) results from the inhalation of avian antigens. The clinical features of chronic HP, including imaging and histological findings, are similar to those of idiopathic pulmonary fibrosis. Despite its status as the “gold standard,” the inhalation provocation test (IPT) is rarely performed, because the methods and the criteria are not standardized. In 2000, we reported the utility of IPT for pigeon dropping extracts. Objectives: The purpose of the current study was to validate the utility and safety of the test, and to differentiate chronic HP from other interstitial lung diseases. Methods: A total of 28 patients with chronic BRHP and 19 control subjects were evaluated in this retrospective study. We validated the previous criteria and proposed new criteria using prediction scores.
Measurements and Main Results: In the IPT using pigeon dropping extracts, the previous criteria showed a sensitivity of 78.6% and a specificity of 94.7% in this retrospective study. The increases in the peripheral white blood cell count and C-reactive protein levels are good indicators of a positive response to the inhalation challenge. We propose the use of the IPT prediction score (ΔWBC [%] 1 2 3 ΔP[A 2 a]O2 [mm Hg], where WBC is white blood cell) and the prediction rule, which showed high sensitivity and specificity values of 92.9 and 94.7%, respectively. Two (1.5%) out of a total of 130 subjects who underwent the tests required treatment after the challenge. Conclusions: The IPT is a useful and safe tool for the diagnosis of chronic HP. The IPT prediction score that we have proposed has high sensitivity and specificity in the diagnosis of chronic BRHP. Keywords: inhalation provocation test; chronic hypersensitivity pneumonitis; validation; prediction score
(Received in original form August 2, 2014; accepted in final form December 30, 2014 ) Supported by a grant to the Diffuse Lung Diseases Research Group from the Ministry of Health, Labor and Welfare, Japan (N.I.) and by a grant by the Japan Society for the Promotion of Science 235,591,140 (Y.M.). Author Contributions: M.I. and Y.M. contributed to study design, data collection, data analysis, statistical analysis, and manuscript preparation; T. Tateishi contributed to data collection and data analysis; T. Tsutsui contributed to data collection; K.T. contributed to study design and data collection; N.I. contributed to study design and manuscript preparation. Correspondence and requests for reprints should be addressed to Yasunari Miyazaki, M.D., Health Administration Center, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Ann Am Thorac Soc Vol 12, No 2, pp 167–173, Feb 2015 Copyright © 2015 by the American Thoracic Society DOI: 10.1513/AnnalsATS.201408-350OC Internet address: www.atsjournals.org
Hypersensitivity pneumonitis (HP) is an immunologically mediated lung disease induced by the inhalation of any of a wide variety of antigens (1). For example, birdrelated HP (BRHP) is one of the common HPs resulting from exposure to avian antigens, such as those found in pigeon droppings and feathers (2). The presence of IgG class–specific antibodies in most cases of HP suggests that a type III
hypersensitivity mechanism may be responsible for the pathogenesis. In addition, a type IV hypersensitivity mechanism mediated by T cells is currently known to be involved (3). There appear to be three types of HP based on the clinical features: acute, subacute, and chronic (1, 4, 5). The symptoms of acute HP occur 4–6 hours after exposure to the etiologic antigen and
consist of the abrupt onset of a flu-like syndrome characterized by fever, chills, malaise, and myalgias (5). The respiratory symptoms include dyspnea, tightness of the chest, and a nonproductive cough. Acute BRHP reproduces the following symptoms after exposure to avian antigens: IgG and/or IgA antibodies specific to pigeon dropping extracts (PDEs) in the sera or bronchoalveolar lavage (BAL) fluids; the
Ishizuka, Miyazaki, Tateishi, et al.: Inhalation Provocation Test in Chronic Bird-Related HP
167
ORIGINAL RESEARCH proliferation of either BAL or peripheral lymphocytes within the pigeon sera; lymphocytosis in the BAL fluids; and alveolitis or granulomatous lesions in the lungs (6, 7). Data from European countries indicate that HP represents 4–15% of all interstitial lung diseases (8), with an incidence rate of 0.9 cases per 100,000 person-years (9). The most common forms of HP are farmer’s lung and pigeon breeder’s lung, but the proportion of individuals who are exposed to these antigens and will develop HP is unknown. An epidemiologic study of acute HP in Japan demonstrated that summertype HP was the most prevalent form of the disease (10). An epidemiologic survey of chronic HP in Japan revealed that BRHP was the most prevalent HP, representing over 60% of all chronic HP cases (222 cases); summer-type HP accounted for only 15% (11). After prolonged exposure to small amounts of an avian antigen when the patient could have been considered asymptomatic, chronic BRHP may develop due to the fluctuating acute episodes, including mild exertional dyspnea and cough (recurrent type) or the absence of a history of acute episodes (insidious type) (5, 12). In both recurrent and insidious chronic HP, patients develop symptoms of increasing dyspnea upon exertion, as well as fatigue, anorexia, cough, and weight loss after interstitial inflammation and fibrosis (4). The clinical features of chronic BRHP, including the imaging and histological findings, are similar to those of idiopathic pulmonary fibrosis (IPF). Because of the difficulty inherent in diagnosing chronic HP, we have observed an increasing number of cases of chronic HP that had previously been misdiagnosed as IPF. In clinical practice, avian contact is the only clue as to a potential diagnosis of BRHP (13). Ram´ırez-Venegas and colleagues (14) reported the utility of the inhalation provocation test (IPT) using pigeon serum for the diagnosis of chronic BRHP. Morell and colleagues (15) proposed that the IPT be considered the “gold standard” among noninvasive BRHP diagnostic examinations, focusing on the decreased FVC and/or the diffusing capacity of the lungs for carbon monoxide (DLCO). In spite of its status as the “gold standard,” the IPT is rarely performed, because the methods and the criteria are not 168
standardized. The duration of the nebulization and the concentrations of extract used are not standardized, and both parameters vary across studies. Munoz and colleagues (16) discussed how the test should be performed and which parameters should be assessed to ensure a correct interpretation. Possible HP may sometimes be mistaken for other interstitial lung diseases. In 2000, we reported the utility of the IPT using PDE and proposed the diagnostic criteria for chronic BRHP (17). The purpose of the current study was to evaluate the validity of the criteria in the diagnosis of chronic BRHP and the safety of this test. We also propose a new prediction score (PS) based on the analysis of the new data presented in this study.
Methods Informed Consent
Informed consent was obtained from all subjects, and this study was approved by institutional review boards at the Tokyo Medical and Dental University. Criteria for Chronic HP
We used the Yoshizawa’s criteria (18) for chronic HP for screening chronic BRHP. The criteria were three or more of the following (including either 1 or 2, either 3 or 4, and either 5 or 6): (1) reproduction of symptoms of HP by an environmental provocation or laboratory-controlled inhalation of the avian antigen; (2) positive titer of antibodies and/or lymphocyte proliferation upon exposure to an avian antigen; (3) evidence of pulmonary fibrosis with or without granulomas detected in histology; (4) a finding of honeycombing on computed tomography (CT) scans; (5) progressive deterioration of a restrictive impairment of pulmonary function over 1 year; and (6) existence of a 6-month or longer duration of respiratory symptoms related to the disease. Retrospective Study Subjects
A retrospective review of the medical records of patients who were recruited after sequential hospitalizations between April 2000 and March 2011 was undertaken at the Tokyo Medical and Dental University Hospital in Japan. Among the patients who underwent the IPT, we first separated out the subjects who did not fulfill Yoshizawa’s
criteria. Then, to select the core group of BRHP-positive patients to validate the IPT, we classified the patients whose lung tissues from surgical lung biopsies were compatible with the histopathological features of chronic HPs, including centrilobular fibrosis and bridging fibrosis; these patients were defined as the BRHP group. Among patients who did not meet the criteria for inclusion in the BRHP group, those who were diagnosed with other interstitial lung disease via a separate etiology, such as HPs by other antigens or collagen vascular diseases, were assigned to the control group. Finally, the remaining subjects were assigned to a group with “no definite diagnosis.” Inhalation Challenge
Patients and control subjects inhaled 2 ml of PDE through a hand nebulizer. The maximal duration of exposure to this antigen was approximately 10 minutes (17). Clinical symptoms, including cough, dyspnea, chills, and body temperature (BT), were evaluated immediately before and hourly during the 24 hours after the challenge. White blood cell (WBC) counts, C-reactive protein (CRP), and arterial blood gas were measured immediately before and 6 and 24 hours after the challenge. Pulmonary function tests, including measurement of the vital capacity (VC) and the DLCO, were performed immediately before and 24 hours after the inhalation challenge. Chest radiographs were taken at similar intervals. Criteria for the IPT
We previously reported monitoring parameters to determine the criteria for the inhalation provocation challenge in 2000 (17). Subjects who fulfilled two or more of the following were considered positive: (1) increased radiologic abnormalities; (2) increase in the alveolar–arterial oxygen pressure difference (P[A 2 a]O2) by more than 10 mm Hg and/or a decrease of DLCO by more than 20%; (3) decreased VC by more than 15%; (4) increase in the peripheral WBC count by more than 30%; (5) increased CRP by more than 1.0 mg/dl; (6) increased BT by more than 1.08 C and/or the development of systemic manifestations, including chills and general fatigue; and (7) development of respiratory symptoms (cough and dyspnea).
AnnalsATS Volume 12 Number 2 | February 2015
ORIGINAL RESEARCH Evaluation of High-Resolution CT
We evaluated the degree of fibrotic change in both the BRHP group and the control group using the Kazerooni fibrosis score in high-resolution CT scans (19). The details of the fibrosis score are explained in the online supplement. The high-resolution CT scans were reviewed independently by two experienced respiratory physicians (Y.M. and M.I.).
April 2000 – March 2011 Hospitalizations for interstitial lung disease (ILD) n=755
April 2000 – March 2011 Inhalation provocation test n=130
Out of chronic HP criteria n=9
Statistical Analysis
Data are expressed as the means (6SEM). The results were compared using the Mann-Whitney U test or unpaired t test. Comparisons between the groups were performed using the x2 test or Fisher’s exact test for categorical variables. All statistical comparisons between the chronic BRHP group and the control group were two sided and were performed at the 0.05 significance level. These statistical analyses were performed using GraphPad PRISM (GraphPad software, San Diego, CA). We incorporated the variables found to be significant at the 0.05 level in a stepwise logistic regression model to reduce the list of diagnostic criteria using SPSS Statistics (IBM, New York, NY). We proposed the IPT-PS using the ratio of the regression coefficients.
Results Patient Characteristics in Retrospective Study Samples
Between 2000 and 2011, 755 patients were hospitalized due to interstitial lung diseases. Among them, 130 patients underwent the IPT for the diagnosis of BRHP. We separated out nine patients who did not fulfill Yoshizawa’s criteria. Among the remaining subjects, 28 patients were classified into the BRHP group. Among those who did not meet the criteria for inclusion in the BRHP group, 19 patients who were diagnosed with other interstitial lung diseases via a separate etiology (including 6 HPs associated with other antigens, 12 collagen vascular diseases, and 1 drug-induced interstitial pneumonia) were assigned to the control group (Figure 1). The remaining 74 subjects were assigned to a group with “no definite diagnosis.” A total of 16 of the 28 patients in the BRHP group and 11 of the 19 subjects in the control group were smokers (Table 1); 11 of the 28 patients in the BRHP group raised budgerigars or bred pigeons. Among the
BRHP group Diagnosed by surgical lung biopsy n=28
Control group Diagnosed as other ILD n=19
No definite diagnosis n=74
Figure 1. Diagram illustrating each step in the patient classification process according to the Yoshizawa’s criteria of chronic hypersensitivity pneumonitis (HP) and the histopathological evaluation by surgical lung biopsies. BRHP = bird-related HP.
remaining patients with BRHP, 11 patients used feather quilts, and six patients were most likely exposed to avian antigens in the surrounding environment. In the BRHP group, five patients were classified into the recurrent type, whereas the remaining 23 patients were classified into the insidious type. A total of 21 of the 28 patients in the BRHP group and all 19 of the subjects in the control group had antibodies to PDE in their sera and/or BAL fluids; 21 of the 28 patients in the BRHP group and 10 of the 19 subjects in the control group had positive proliferative responses of the peripheral or bronchoalveolar lymphocytes in response to the pigeon sera. All subjects in the BRHP group and the control group presented honeycombing, traction bronchiectasis, or intralobular septal
thickening. In addition, all patients in the BRHP group had compatible histopathological features in surgical lung biopsies. We evaluated the extent of fibrosis using the Kazerooni fibrosis score in Table 1 (19). The means of the scores in the BRHP group and the control group were 1.97 (60.15; mean [6SEM]) and 2.10 (60.19), respectively (P = 0.565). The results of the tests for pulmonary function are shown in Table 1, and there were no significant differences between the two groups. Clinical and Laboratory Data after the Inhalation Challenge
After the inhalation challenge, 9 (32%) of the 28 patients in the BRHP group showed radiologic changes, but none of the 19 subjects in the control group showed similar
Table 1. Subject profile in the study
Age, yr, median (25th–75th percentiles) Sex, male/female, n Smoking history, ever/never, n Specific antibody, positive/negative, n LST, positive/negative, n VC, % predicted, mean 6 SEM FEV1, % predicted, mean 6 SEM DLCO, % predicted, mean 6 SEM PaO2, mm Hg, mean 6 SEM PaCO2, mm Hg, mean 6 SEM Fibrosis score, mean 6 SEM
BRHP (n = 28)
Control (n = 19)
P Value
64 (58–68) 18/10 16/12 21/7 21/6 80.4 6 3.7 75.5 6 3.4 58.6 6 4.6 81.3 6 2.1 43.1 6 0.7 1.97 6 0.15
66 (58–70) 10/9 11/8 19/0 10/9 77.6 6 4.0 72.6 6 3.6 53.9 6 4.7 81.6 6 2.7 40.7 6 1.2 2.10 6 0.19
0.362 0.547 1.000 0.032 0.111 0.617 0.567 0.504 0.925 0.064 0.565
Definition of abbreviations: BRHP = bird-related hypersensitivity pneumonitis; DLCO = lung diffusing capacity for carbon monoxide; LST = lymphocyte stimulating test; VC = vital capacity.
Ishizuka, Miyazaki, Tateishi, et al.: Inhalation Provocation Test in Chronic Bird-Related HP
169
ORIGINAL RESEARCH changes (P = 0.007). A total of 12 (43%) of the 28 patients in the BRHP group and 3 (16%) of the remaining 19 subjects developed respiratory symptoms (cough and dyspnea) after the inhalation challenge (P = 0.209). We evaluated the increase in the peripheral WBC count with the following measures: receiver operating characteristic (ROC) curve analysis at 6 hours (area under the curve [AUC] = 0.941, P , 0.001) and 24 hours (AUC = 0.682, P = 0.036) after the challenge; the increase of CRP by the ROC curve at 6 hours (AUC = 0.586, P = 0.324) and 24 hours (AUC = 0.878, P , 0.001) after the challenge; the increase of P(A 2 a)O2 by the ROC curve at 6 hours (AUC = 0.844, P , 0.001) and 24 hours (AUC = 0.588, P = 0.308) after the challenge; the decrease of VC (AUC = 0.503, P = 0.974) by the ROC curve at 24 hours after the challenge; the decrease of DLCO (AUC = 0.519, P = 0.832) by the ROC curve at 24 hours the after challenge; and the increase of BT by the ROC curve (AUC = 0.753, P = 0.004) for the 24 hours after the challenge (Table 2).
Table 2. The areas under the curves and P Values by the receiver operating characteristic curves in 6 and 24 hours in the study Variables 6 h after D%WBC DCRP DP(A 2 a)O2 D%VC* D%DLCO* 24 h after D%WBC DCRP DP(A 2 a)O2 D%VC* D%DLCO* DBT†
AUC
P Value
0.941 0.586 0.844 — —
,0.001 0.324 ,0.001 — —
0.682 0.878 0.588 0.503 0.519 0.753
0.036 ,0.001 0.308 0.974 0.832 0.004
Definition of abbreviations: AUC = area under the curve; BT = body temperature; CRP = C-reactive protein; DP(A 2 a)O2 = alveolar–arterial oxygen pressure difference; DLCO = lung diffusing capacity for carbon monoxide; VC = vital capacity; WBC = white blood cell. *VC and DLCO were measured only immediately before and 24 hours after the inhalation challenge. † BT was measured immediately before and hourly during the 24 hours after the challenge, and we evaluated the difference between the maximum and initial values.
170
Safety of the Inhalation Challenge
Most subjects recovered in 1 or 2 days after the challenge. Two (1.5%) of the 130 total subjects who underwent the IPT required treatment for changes related to the challenge. During the IPT, the first case increased the ground-glass opacity bilaterally, and the P(A 2 a)O2 increased from 44.7 to 115 mm Hg. The second case also increased the ground-glass opacity bilaterally, and the P(A 2 a)O2 increased from 28.4 to 131 mm Hg. These patients received oxygen and steroid pulse therapies without intubation or an intensive care unit stay. Both patients recovered to the baseline oxygenation levels. The first patient had respiratory failure (PaO2 , 60 mm Hg) and a high CRP level before the IPT, whereas the second patient had a high level of antibodies to PDE. Validation of the Previous Criteria for the Inhalation Provocation Challenge
A total of 22 (78.6%) of the 28 patients in the BRHP group in this study were classified as positive or most likely positive compared with 100% (all 11 patients) in our previous report (17). A total of 18 (94.7%) of the 19 subjects in the control group in this study were classified as negative compared with 100% (all 6 subjects) in our previous report (17). In the current study, the model had a sensitivity of 78.6% and a specificity of 94.7%. IPT-PS
We validated the utility of the IPT based on the previous criteria, but this interpretation was hardly sufficient to improve our diagnosis of chronic BRHP. We incorporated the variables that were found to be significant at the 0.05 level into a stepwise logistic regression model to reduce the monitoring parameters in the diagnostic criteria. The item of radiologic change was excluded from this analysis,
because no patients in the control group showed an increase in radiologic abnormalities after the challenge. The item of CRP was also excluded from this analysis, because there were intermediate correlations between the CRP and the peripheral WBC count (r = 0.599) and between CRP and P(A 2 a)O2 (r = 0.461). We thus propose the IPT-PS, which uses the ratios of the regression coefficients (Table 3), as follows: IPT-PS = ΔWBC (%) 1 2 3 ΔP(A 2 a)O2 (mm Hg). We showed the ROC curve for the prediction rule by the IPT-PS (AUC = 0.966, P , 0.001; Figure 2). The cut-off value of the IPT-PS was set to 35 (Figure 3), and the prediction rule showed high sensitivity and specificity values of 92.9% (26 of the 28 patients) and 94.7% (18 of the 19 subjects), respectively. A total of 60% of the subjects in the “no definitive diagnosis” group had IPT-PS values that were higher than the cut off. We were able to diagnose the 74 subjects as having BRHP by Yoshizawa’s criteria (18); however, other interstitial pneumonias can be included by the criteria, because the immunological criterion in Yoshizawa’s criteria did not exhibit high sensitivity, although it did present high specificity, according to our previous study (20). We compared the measurements between a higher IPT-PS group and a lower IPT-PS group. There was a significant difference between these two groups only in the values of IgG antibodies to PDE in the BAL fluids.
Discussion Here, we have validated the previous criteria for the diagnosis of BRHP. The sensitivity and specificity of the previous criteria were 78.6 and 94.7% in the current study compared with 100 and 100%, respectively, in our previous report (17). We confirmed
Table 3. Predictors of the inhalation provocation test Variables
Coefficient
Increase in peripheral WBC count, % Increase in P(A 2 a)O2, mm Hg Increase in BT, 8 C Development of respiratory symptoms Intercept
0.18 0.39 2.15 22.54 26.98
OR (95% CI) 1.20 1.48 11.7 0.08
(1.05–1.37) (1.07–2.04) (0.4–184.4) (0.001–4.5) —
P Value 0.01 0.02 0.17 0.22 —
Definition of abbreviations: BT = body temperature; CI = confidence interval; OR = odds ratio; P(A 2 a)O2 = alveolar–arterial oxygen pressure difference; WBC = white blood cell.
AnnalsATS Volume 12 Number 2 | February 2015
ORIGINAL RESEARCH
100
Sensitivity%
80 60 40 20 0 0
20 40 60 80 100% - Specificity%
100
Figure 2. Receiver operating characteristic curve for the prediction rule by the inhalation provocation test prediction score in the study.
that the criteria maintained high specificity, which is useful for the diagnosis of chronic BRHP. In the current study, we indicated that the changes in the peripheral WBC count, CRP, P(A 2 a)O2, and BT were useful monitoring parameters for the test. We proposed a new PS, the IPT-PS, which included the peripheral WBC count and P(A 2 a)O2 and had a higher sensitivity (92.9%) than the previous criteria (78.6%). We certified that the proposed scale was more clinically useful for diagnosis than the previous criteria. We evaluated the changes in the peripheral WBC count, CRP, P(A 2 a)O2, BT, VC, and DLCO by the ROC curve. The peripheral WBC count, CRP, P(A 2 a)O2, and BT showed a significant difference, but the VC and DLCO showed no significant difference. Morell and colleagues (15) reported that the FVC and DLCO are good indicators of a positive response to the inhalation challenge. In their report, the FVC and DLCO were recorded at baseline, 20 minutes after the inhalation, and every
Number of positive criteria
200 150 IPT-PS
hour for the next 8 hours. In the current study, there were no significant differences between the BRHP group and the control group in VC and DLCO, because these parameters were only recorded at baseline and 24 hours after the challenge. Ram´ırezVenegas and colleagues (14) reported that fever was the major symptom observed in all of the patients with a positive response to the inhalation challenge, and usually preceded all other clinical symptoms. The current study demonstrates that not only fever, but also peripheral leukocytosis, an increase of CRP, and an increase of P(A 2 a)O2, most likely act as good indicators of a positive response to the inhalation challenge. The IPT-PS proposed herein requires only two monitoring parameters: the peripheral WBC count and P(A 2 a)O2. In chronic HP, there is no diagnostic method that has as high a sensitivity as IPT. The diagnosis of BRHP is generally based on the presence of consistent clinical symptoms, which provide confirmation that the patient has been exposed and is sensitized to a causal agent (including high levels of antibodies to avian antigens), consistent chest X-rays or CT imaging findings, evidence of increased lymphocytes in BAL fluids, and demonstration of consistent histologic features (17). However, identifying the causal agent is not simple, even when the antibodies to the avian antigen are measured. The National Heart, Lung, and Blood Institute/Office of Rare Diseases workshop (21) and the Hypersensitivity Pneumonitis Study Group (22) agree that HP is caused by the inhalation of an antigen to which the patient is sensitized and hyperresponsive, and that sensitization and exposure alone in
100 50 35 0 –50 BRHP
Control
7 6 5 4 3 2 1 0 BRHP
Control
Figure 3. The scatter plot of inhalation provocation test prediction scores (IPT-PS) in the bird-related hypersensitivity pneumonitis (BRHP) group and the control group (left) compared with the number of positive parameters for the previous criteria (right) in the study. The cut-off value for IPT-PS was set at 35, which is given by the receiver operating characteristic curve.
the absence of symptoms do not define the disease, regardless of the presence of specific IgG antibodies. Costabel and colleagues (3) demonstrated that 30–60% of healthy farmers produced precipitating antibodies to the antigens to which they were exposed, whereas 10–15% of patients do not develop serum precipitins. Therefore, a negative finding does not rule out the presence of disease. In our previous report (20), the immunological components of Yoshizawa’s criteria, including antibodies to PDE and a proliferative response of lymphocytes to the pigeon sera, had low sensitivities (z40%) and high specificities (z80–90%). The subjects in the “no definite diagnosis” group in the current study include “suspected BRHP,” but not definite BRHP, mostly because the immunological criterion has a low AUC (z0.6) in ROC analysis (20). Therefore, the IPT-PS for BRHP has a higher diagnostic capability than Yoshizawa’s criteria. The clinical features of chronic HP, including imaging and histological findings, are similar to those of IPF. Centrilobular nodules, lobular air trapping, and basilar sparing are useful findings to distinguish chronic HP from IPF for some patients (23). However, we often find that patients with chronic HP present a usual interstitial pneumonia (UIP) pattern, particularly in cases of insidious-type chronic HP (24), and the distinction from IPF is difficult. In our previous report, there was no difference in the extent of the ground-glass opacities, consolidation, honeycombing, or emphysema in patients with chronic HP presenting a UIP pattern and IPF. In this respect, the IPT-PS is more useful than radiological findings for distinguishing chronic HP presenting a UIP pattern from IPF. Ram´ırez-Venegas and colleagues (14), Morell and colleagues (15), and Ohtani and colleagues (17) reported the utility and safety of the IPT for the diagnosis of chronic BRHP. In the current study, we further demonstrated the validity and safety of the IPT for individuals with chronic BRHP at the antigen concentration used. Two (1.5%) of the 130 subjects who underwent the IPT required steroid pulse therapy for the changes caused by the challenges. One patient had respiratory failure and had a high CRP level before the IPT, and the other patient had a high titer of antibodies against the avian antigen. Munoz and colleagues (16) recommended
Ishizuka, Miyazaki, Tateishi, et al.: Inhalation Provocation Test in Chronic Bird-Related HP
171
ORIGINAL RESEARCH that a low FVC and DLCO with respiratory insufficiency (PaO2 , 60 mm Hg) should constitute contraindications. In fact, the PaO2 of the patient with respiratory failure was the lowest of all the subjects. In addition, a high CRP value (2.4 mg/dl) was recorded at admission, and the challenge was conducted a day after admission. The challenge was also conducted a day after admission for the patient with a high titer of antibodies against the avian antigen, and we considered this to be adequate antigen avoidance, because a time period of 2 weeks of hospitalization was needed before the challenge. The limitations of the present study include the retrospective analysis and the time period during which the respiratory function tests were performed. One weakness of the study was the restriction
due to the choice of antigen, even though avian antigens represent the most common etiology in chronic HP. The utility of the IPT with avian antigens in the diagnosis of BRHP has been reported in four studies (14, 15, 17, 25). Furthermore, fungi (26, 27) and bacteria (28) were reported as antigens of HP, and have been used for diagnosis in IPTs. BRHP is reported to account for 34–68% of HP in the world according to the four studies (29–32). Avian antigens were among the major antigens in chronic HP; however, more than one-third of the patients had other causative antigens. Therefore, we are now working on developing IPTs for other antigens. To prioritize the development and use of less-invasive tests, we conducted environmental investigations and immunological tests first. Then, we
References 1 Fink JN. Hypersensitivity pneumonitis. J Allergy Clin Immunol 1984;74: 1–10. 2 Todd A, Coan RM, Allen A. Pigeon breeders’ lung: pigeon intestinal mucin, an antigen distinct from pigeon IgA. Clin Exp Immunol 1991; 85:453–458. 3 Costabel U, Bonella F, Guzman J. Chronic hypersensitivity pneumonitis. Clin Chest Med 2012;33:151–163. 4 Kaltreider HB. Hypersensitivity pneumonitis. West J Med 1993;159: 570–578. 5 Salvaggio JE, Millhollon BW. Allergic alveolitis: new insights into old mysteries. Respir Med 1993;87:495–501. 6 Selman M, Teran ´ L, Mendoza A, Camarena A, Mart´ınez-Cordero E, Lezama M, Rubio HM. Increase of HLA-DR7 in pigeon breeder’s lung in a Mexican population. Clin Immunol Immunopathol 1987;44: 63–70. 7 Coleman A, Colby TV. Histologic diagnosis of extrinsic allergic alveolitis. Am J Surg Pathol 1988;12:514–518. 8 Thomeer MJ, Costabe U, Rizzato G, Poletti V, Demedts M. Comparison of registries of interstitial lung diseases in three European countries. Eur Respir J Suppl 2001;32:114s–118s. 9 Solaymani-Dodaran M, West J, Smith C, Hubbard R. Extrinsic allergic alveolitis: incidence and mortality in the general population. QJM 2007;100:233–237. 10 Ando M, Konishi K, Yoneda R, Tamura M. Difference in the phenotypes of bronchoalveolar lavage lymphocytes in patients with summer-type hypersensitivity pneumonitis, farmer’s lung, ventilation pneumonitis, and bird fancier’s lung: report of a nationwide epidemiologic study in Japan. J Allergy Clin Immunol 1991;87:1002–1009. 11 Okamoto T, Miyazaki Y, Ogura T, Chida K, Kohno N, Kohno S, Taniguchi H, Akagawa S, Mochizuki Y, Yamauchi K, et al. Nationwide epidemiological survey of chronic hypersensitivity pneumonitis in Japan. Respir Investig 2013;51:191–199. 12 Ohtani Y, Saiki S, Sumi Y, Inase N, Miyake S, Costabel U, Yoshizawa Y. Clinical features of recurrent and insidious chronic bird fancier’s lung. Ann Allergy Asthma Immunol 2003;90:604–610. 13 Perez-Padilla ´ R, Salas J, Chapela R, Sanchez ´ M, Carrillo G, Perez ´ R, Sansores R, Gaxiola M, Selman M. Mortality in Mexican patients with chronic pigeon breeder’s lung compared with those with usual interstitial pneumonia. Am Rev Respir Dis 1993;148:49–53. 14 Ram´ırez-Venegas A, Sansores RH, Perez-Padilla ´ R, Carrillo G, Selman M. Utility of a provocation test for diagnosis of chronic pigeon breeder’s disease. Am J Respir Crit Care Med 1998;158:862–869.
172
performed the IPT for suspicious antigens one at a time using the test described previously here. The described procedure should be implemented for other causative antigens, and the criteria should be verified using the IPT-PS. In summary, this test could become a valuable addition to the current diagnostic criteria, and it might be useful for differentiating chronic HP from interstitial lung diseases, including IPF, as a noninvasive examination due to its utility and safety. n Author disclosures are available with the text of this article at www.atsjournals.org. Acknowledgment: The authors thank Makoto Tomita and Masako Akiyama for their assistance with aspects of the data analysis.
15 Morell F, Roger A, Reyes L, Cruz MJ, Murio C, Muñoz X. Bird fancier’s lung: a series of 86 patients. Medicine (Baltimore) 2008;87:110–130. 16 Munoz X, Morell F, Cruz MJ. The use of specific inhalation challenge in hypersensitivity pneumonitis. Curr Opin Allergy Clin Immunol 2013; 13:151–158. 17 Ohtani Y, Kojima K, Sumi Y, Sawada M, Inase N, Miyake S, Yoshizawa Y. Inhalation provocation tests in chronic bird fancier’s lung. Chest 2000;118:1382–1389. 18 Yoshizawa Y, Ohtani Y, Hayakawa H, Sato A, Suga M, Ando M. Chronic hypersensitivity pneumonitis in Japan: a nationwide epidemiologic survey. J Allergy Clin Immunol 1999;103:315–320. 19 Kazerooni EA, Martinez FJ, Flint A, Jamadar DA, Gross BH, Spizarny DL, Cascade PN, Whyte RI, Lynch JP III, Toews G. Thin-section CT obtained at 10-mm increments versus limited three-level thin-section CT for idiopathic pulmonary fibrosis: correlation with pathologic scoring. AJR Am J Roentgenol 1997;169:977–983. 20 Suhara K, Miyazaki Y, Okamoto T, Yasui M, Tsuchiya K, Inase N. Utility of immunological tests for bird-related hypersensitivity pneumonitis. Respir Investig 2015;53:13–21. 21 Fink JN, Ortega HG, Reynolds HY, Cormier YF, Fan LL, Franks TJ, Kreiss K, Kunkel S, Lynch D, Quirce S, et al. Needs and opportunities for research in hypersensitivity pneumonitis. Am J Respir Crit Care Med 2005;171:792–798. 22 Lacasse Y, Selman M, Costabel U, Dalphin JC, Ando M, Morell F, Erkinjuntti-Pekkanen R, Muller N, Colby TV, Schuyler M, et al.; HP Study Group. Clinical diagnosis of hypersensitivity pneumonitis. Am J Respir Crit Care Med 2003;168:952–958. 23 Silva CI, Muller ¨ NL, Lynch DA, Curran-Everett D, Brown KK, Lee KS, Chung MP, Churg A. Chronic hypersensitivity pneumonitis: differentiation from idiopathic pulmonary fibrosis and nonspecific interstitial pneumonia by using thin-section CT. Radiology 2008;246:288–297. 24 Tateishi T, Ohtani Y, Takemura T, Akashi T, Miyazaki Y, Inase N, Yoshizawa Y. Serial high-resolution computed tomography findings of acute and chronic hypersensitivity pneumonitis induced by avian antigen. J Comput Assist Tomogr 2011;35:272–279. 25 Hendrick DJ, Marshall R, Faux JA, Krall JM. Positive “alveolar” responses to antigen inhalation provocation tests: their validity and recognition. Thorax 1980;35:415–427. 26 Morell F, Cruz MJ, Gomez ´ FP, Rodriguez-Jerez F, Xaubet A, Muñoz X. Chacinero’s lung—hypersensitivity pneumonitis due to dry sausage dust. Scand J Work Environ Health 2011;37:349–356. 27 Mitsui C, Taniguchi M, Fukutomi Y, Saito A, Kawakami Y, Mori A, Akiyama K. Non occupational chronic hypersensitivity pneumonitis due to Aspergillus fumigatus on leaky walls. Allergol Int 2012;61: 501–502.
AnnalsATS Volume 12 Number 2 | February 2015
ORIGINAL RESEARCH 28 Williams JV. Inhalation and skin tests with extracts of hay and fungi in patients with farmer’s lung. Thorax 1963;18:182–196. 29 Xaubet A, Ancochea J, Morell F, Rodriguez-Arias JM, Villena V, Blanquer R, Montero C, Sueiro A, Disdier C, Vendrell M; Spanish Group on Interstitial Lung Diseases, SEPAR. Report on the incidence of interstitial lung diseases in Spain. Sarcoidosis Vasc Diffuse Lung Dis 2004;21:64–70. 30 Selman M, Lacasse Y, Pardo A, Cormier Y. Hypersensitivity pneumonitis caused by fungi. Proc Am Thorac Soc 2010;7:229–236.
31 Roelandt M, Demedts M, Callebaut W, Coolen D, Slabbynck H, Bockaert J, Kips J, Brie J, Ulburghs M, De Boeck K, et al. Epidemiology of interstitial lung disease (ILD) in Flanders: registration by pneumologists in 1992–1994. Working group on ILD, VRGT. Vereniging voor Respiratoire Gezondheidszorg en Tuberculosebestrijding. Acta Clin Belg 1995;50:260–268. 32 Hanak V, Golbin JM, Ryu JH. Causes and presenting features in 85 consecutive patients with hypersensitivity pneumonitis. Mayo Clin Proc 2007;82:812–816.
Ishizuka, Miyazaki, Tateishi, et al.: Inhalation Provocation Test in Chronic Bird-Related HP
173
