Lung

, Volume 188, Issue 2, pp 107–113

Impulse Oscillometry and Respiratory Symptoms in World Trade Center Responders, 6 Years Post-9/11

Authors

    • Bureau of Occupational Health, Center for Environmental HealthNew York State Department of Health
  • Karen R. Cummings
    • Bureau of Occupational Health, Center for Environmental HealthNew York State Department of Health
Article

DOI: 10.1007/s00408-009-9206-y

Cite this article as:
Mauer, M.P. & Cummings, K.R. Lung (2010) 188: 107. doi:10.1007/s00408-009-9206-y

Abstract

This study evaluated whether impulse oscillometry (IOS) testing revealed signs of respiratory disease in New York State (NYS) World Trade Center (WTC) responders in comparison with unexposed NYS employees. It also compared self-reported respiratory symptoms between the two groups, 6 years post-9/11. For this evaluation participants completed a self-administered questionnaire regarding respiratory symptoms. IOS testing included measures of resistance and reactance to assess for peripheral versus central airway effects. Two hundred forty-eight subjects (99 exposed and 149 unexposed) were included in the final analysis. Since September 11, 2001, NYS responders were more likely to report new or worsening cough in the absence of a respiratory infection, cough consistent with chronic bronchitis, current respiratory symptoms, or lower respiratory symptoms in the last 12 months. Significant associations were found between IOS indices and gender, smoking history, and obesity. When comparing exposed and unexposed participants, there were no significant differences in the geometric means of the IOS indices. Responders who used a respirator with canister demonstrated significantly lower respiratory resistance at 5 and 20 Hz (R5 and R20). While this study has provided no evidence of an association between WTC exposure and peripheral airways disease in this cohort of responders, results do suggest that use of a respirator with canister may be protective for central airways in responders exposed to dust and smoke. This emphasizes the importance of stressing proper respirator use in planning responses to future disasters. Our control data also provide useful reference values for future IOS research.

Keywords

Impulse oscillometryRespiratory symptomsSeptember 11, 2001World Trade Center disaster

Introduction

Respiratory effects have been reported among World Trade Center (WTC) disaster responders and among residents and office workers near the WTC site [19]. Each of these groups was potentially exposed to a variety of materials and chemicals in the dust and smoke [1015]. New York State (NYS) employees who responded to the WTC disaster on or after September 11, 2001 (NYS responders) represent a cohort with potentially moderate exposures compared with first responders [5]. While some NYS responders were present in the first days of the event, many were on site during the months that followed. Ongoing exposures resulted from the resuspension of dust from recovery activities and intermittent smoke plumes from ongoing fires.

The precise nature of the respiratory effects from WTC exposure is still uncertain. A 2007 study concluded that impulse oscillometry (IOS) and frequency dependence of compliance abnormalities seen in symptomatic individuals with presumed WTC exposures likely reflect dysfunction in airways more distal to those evaluated by spirometry [16]. Some studies have suggested air trapping and small-airways disease as an explanation of pulmonary findings [1, 4, 17, 18]. Post-9/11 spirometry results among NYS responders did not demonstrate any significant abnormalities overall [5]. However, these personnel have reported respiratory symptoms associated with exposure [5], and these symptoms have been persistent in some.

A study of ironworkers exposed at the WTC site reported that routine spirometry underestimated the prevalence of lung function abnormalities in comparison to forced oscillation (FO) testing [3]. FO applies external pressure to the respiratory system to measure respiratory impedance. Impedance of larger central airways may be dissected from that of peripheral airways using multiple oscillation frequencies [19]. FO has been shown to be as sensitive as spirometry in detecting impairments of lung function due to smoking or exposure to occupational hazards [20]. It has been described as an ideal lung function test for epidemiological and field studies [20]. The test can detect changes in apparently normal subjects consistent with early airway disease prior to the onset of clear spirometric changes [21]. The impulse oscillation system (IOS) has been introduced as a user-friendly commercial version of the FO technique [22]. It differs from FO in that an impulse (rectangular wave form) rather than a pseudorandom noise signal is applied by a loudspeaker [22]. Studies have utilized IOS indices to measure airways obstruction and small-airway disease [16, 2327]. Other studies have suggested IOS is at least as sensitive as spirometry in measuring airways obstruction [3, 2830].

This study was conducted to evaluate whether IOS testing might reveal signs of respiratory disease in a group of NYS responders compared with a control group of similar but unexposed NYS employees. It also compared self-reported respiratory symptoms between the two groups, 6 years post-9/11. In addition, the study offered the opportunity to examine whether previous IOS findings among WTC workers could be reproduced [3]. Our findings might offer further insight into the nature of the respiratory symptoms widely reported among individuals with WTC exposures. An additional benefit was the ascertainment of mean IOS values for a randomly selected control population, which could be useful to future researchers using IOS.

Materials and Methods

Subjects

The NYS Department of Health (NYSDOH) Institutional Review Board approved this study. The study population consisted of exposed and unexposed cohorts. All subjects had participated in previous NYSDOH WTC studies. Exposed subjects were drawn from three exposure groups based on scores derived from an exposure assessment method developed for use in NYSDOH WTC studies, described elsewhere [31]. The three exposure groups included (1) those with a score above the mean for both smoke and dust exposure (high smoke and dust), (2) those with a score above the mean for smoke exposure (high smoke only), and (3) those with a score above the mean for dust exposure (high dust only). Unexposed subjects were NYS employees in similar agencies and job titles with no WTC exposures.

Data Collection

Subjects were evaluated by the same registered nurse at one of several regional test locations between June and October 2007. Each participant completed a questionnaire. Information was obtained regarding respiratory and psychological symptoms, medical history, medication use, smoking history, and other possible confounding factors. Respiratory symptoms were grouped into upper respiratory (nose irritation/burning, runny nose, nasal congestion, and throat irritation) and lower respiratory (dry cough, cough producing mucus, wheezing or whistling in chest, chest tightness, and shortness of breath with normal activity or at rest) symptoms. Information on WTC exposures was obtained from data collected previously. Prior to IOS testing, the nurse also asked each participant a series of questions regarding factors that could affect test results, such as if they were currently experiencing a cold or respiratory infection or seasonal allergy symptoms.

Impulse Oscillometry (IOS)

Respiratory function was measured using an impulse oscillometry system (Masterscreen IOS, VIASYS Respiratory Care, Yorba Linda, CA). IOS testing was conducted by a registered nurse, trained and certified in the operation of the Masterscreen system. The instrument was calibrated with a 3-L syringe at different flow rates at the beginning and midpoint of each test day. In addition, calibration was performed with a reference resistance device, per manufacturer, at each new test site. Ambient conditions of each test site were measured and entered into the system. These included temperature, relative humidity, barometric pressure, and altitude. Testing procedures followed European Respiratory Society recommendations [20]. Prior to testing, each individual’s weight and height were measured using a portable floor scale (Healthometer, Jarden Consumer Solutions, Boca Raton, FL) and a portable stadiometer (Seca Corporation, Hanover, MD). Participants were then seated with their head in a neutral position. A new bacterial filter was used on the apparatus for each participant. Participants were instructed to breathe quietly through the mouthpiece. At least three 30-s measurements were made during tidal breathing using a mouthpiece specially designed to stabilize tongue position and a nose clip, while the participant firmly supported his/her cheeks with the hands. Tests were repeated until acceptable coherence values at 5 and 20 Hz (Co5 ≥ 0.7, Co20 ≥ 0.9) were obtained, unless such values were unobtainable. The average of three acceptable tests was used for the analysis. Subjects unable to complete three trials with coherence at these levels were excluded from analysis. Four parameters were evaluated: (1) respiratory resistance at 5 Hz (R5), which reflects total airways resistance; (2) respiratory resistance at 20 Hz (R20), which reflects central airways resistance; (3) distal capacitive reactance (X5), which reflects peripheral airways function; and (4) a frequency-integrated measure of reactance at low frequencies (AX), which is complementary to frequency dependence of resistance and which reflects peripheral airways function [3, 19, 21].

Statistical Analysis

All statistical analyses were performed using SAS v9.1 (SAS Institute, Cary, NC). Continuous variables are expressed as mean ± SD and were analyzed by the two-sample t test. Categorical data were analyzed using the χ2 test or Fisher’s exact test. Relative risk (RR) was calculated with 95% confidence intervals (CI). Multiple regression analysis was performed using generalized linear models. Gender, history of smoking, and obesity were included in all of the models to control for possible confounding effects. Obesity was defined as a body mass index (BMI) ≥ 30. Because of skewed distribution, IOS measurements were transformed to the natural log scale prior to modeling. Negative numbers (X5) underwent a monotonic transformation prior to the logarithmic transformation. Data were back-transformed to the original scale and presented as the geometric mean. Data were considered statistically significant at p < 0.05.

Results

Two hundred fifty-nine subjects (105 exposed and 154 unexposed) underwent evaluations. Eleven individuals were excluded due to inability to complete three acceptable IOS tests. Thus, 248 subjects (99 exposed and 149 unexposed) were included in the final analysis. Characteristics of exposed and unexposed participants are summarized in Table 1. Roughly one quarter of exposed and unexposed participants reported a current cold or respiratory infection or symptoms of seasonal allergies (p = 0.24).
Table 1

Characteristics of NYS WTC responders and controls (NYS WTC IOS Study, 2007)

 

WTC respondera (n = 99)

Controlsa (n = 149)

p value

Gender

  

<0.0001

 Male

94 (95.0)

104 (69.8)

 

 Female

5 (5.0)

45 (30.2)

 

Race

  

0.01

 White

84 (86.6)

142 (96.0)

 

 Other

13 (13.4)

6 (4.0)

 

Hispanic

  

<0.01

 Yes

11 (12.1)

4 (2.8)

 

 No

80 (87.9)

141 (97.2)

 

Obese (BMI ≥ 30)

  

0.12

 Yes

34 (35.8)

39 (26.4)

 

 No

61 (64.2)

109 (73.7)

 

History of smoking

  

0.42

 Yes

43 (43.4)

57 (38.3)

 

 No

56 (56.6)

92 (61.7)

 

History of asthma

  

0.95

 Yes

13 (13.1)

20 (13.4)

 

 No

86 (86.9)

129 (86.6)

 

History of allergiesb

  

0.21

 Yes

34 (34.3)

63 (42.3)

 

 No

65 (65.7)

86 (57.7)

 

Age

42 ± 8

47 ± 9

<0.0001

BMI

29.3 ± 4.3

27.7 ± 5.0

<0.01

Height (in.)

70.3 ± 2.8

68.4 ± 3.7

<0.0001

Weight (lbs)

206.8 ± 35.4

185.0 ± 40.5

<0.0001

NYS New York State, WTC World Trade Center, IOS impulse oscillometry, BMI body mass index

aNumbers and percentages are shown for categorical variables, and means with standard deviations are shown for continuous variables

bSeasonal or perennial allergies

NYS responders were three times as likely as unexposed participants to report a new or worsening cough in the absence of a respiratory infection since September 11, 2001; they were twice as likely to report a cough consistent with chronic bronchitis (cough productive of mucus or phlegm for at least three consecutive months of the year for at least two consecutive years), since 9/11 (Table 2). WTC responders were also significantly more likely to report a current respiratory symptom or a lower respiratory symptom in the last 12 months (Table 3).
Table 2

Relative risks and 95% confidence intervals for cough symptoms among NYS WTC responders and controls (NYS WTC IOS Study, 2007)

 

WTC responders

Controls

RR

CI

 

N

%

N

%

New or worsening cough since 9/11/2001a

45

53.6

15

10.3

3.27

2.39, 4.47

Duration of cough >8 weeks

19

51.4

5

38.5

1.14

0.82, 1.59

Cough consistent with chronic bronchitis

22

24.7

8

5.5

2.24

1.68, 3.00

NYS New York State, WTC World Trade Center, IOS impulse oscillometry, RR  relative risk, CI confidence interval

aIn the absence of a cold or respiratory infection

Table 3

Relative risks and 95% confidence intervals for lower and upper respiratory symptomsa among NYS WTC responders and controls (NYS WTC IOS Study, 2007)

 

WTC responders

Controls

RR

CI

 

N

%

N

%

Lower respiratory symptoms: current

37

37.4

23

15.4

1.87

1.41, 2.49

Lower respiratory symptoms: last 12 months

69

69.7

80

53.7

1.53

1.08, 2.16

Upper respiratory symptoms: current

32

32.3

27

18.1

1.53

1.13, 2.07

Upper respiratory symptoms: last 12 months

67

67.7

96

64.4

1.09

0.79, 1.52

NYS New York State, WTC World Trade Center, IOS impulse oscillometry, RR  relative risk, CI confidence interval

aIn the absence of a cold or respiratory infection

Significant associations were found between the IOS indices and gender, smoking history, and obesity. These factors were controlled for in the analyses. Table 4 demonstrates the effect these three factors had on each of the IOS indices for members of the control population who did not report a history of asthma, emphysema, or chronic obstructive pulmonary disease (COPD) and who did not report a current cold, respiratory infection, or symptoms of seasonal allergies at the time of the IOS testing (n = 93). Female gender was associated with significantly higher R5 and R20 levels. A history of smoking was associated with significantly higher R5, R20, and AX levels. Obesity was associated with significantly higher R5 and AX levels and significantly lower X5 levels.
Table 4

IOS results for control populationa (n = 93) (NYS WTC IOS Study, 2007)b

 

R5

R20

X5

AX

Male (no risk factors)

3.02

2.84

−0.89

1.79

Female (no risk factors)

3.68

3.75

−0.90

1.79

Male + obesity

3.87

3.16

−2.01

4.38

Female + obesity

4.72

4.17

−2.02

4.38

Male + smoking history

3.61

3.16

−1.30

3.32

Female + smoking history

4.41

4.18

−1.32

3.32

Male + obesity + smoking history

4.63

3.51

−2.33

8.12

Female + obesity + smoking history

5.65

4.65

−2.35

8.11

NYS New York State, WTC World Trade Center, IOS impulse oscillometry

aPopulation was not exposed to WTC disaster site and did not report a history of asthma, emphysema, or chronic obstructive pulmonary disease (COPD), nor did they report a current cold, respiratory infection, or symptoms of seasonal allergies

bData were modeled using linear regression and presented as geometric mean

When controlling for gender, smoking history, and obesity, the geometric mean of R5 for exposed and unexposed participants was 3.5 and 3.6 (cmH2O/L/s), respectively. For unexposed participants, the geometric mean of R20 was 3.3, the geometric mean of X5 was –1.3, and the geometric mean of AX was 3.1. When comparing exposed and unexposed participants, there was no significant difference in the geometric mean of R5 (p = 0.2997), R20 (p = 0.5278), X5 (p = 0.1088), or AX (p = 0.6049). There were also no significant differences in IOS results between symptomatic and asymptomatic participants (data not shown).

WTC responders were classified into three exposure categories: high smoke and dust (n = 52), high smoke only (n = 25), and high dust only (n = 22). IOS results were not significantly different for these groups compared with the unexposed group (data not shown). Thirty (30.3%) responders arrived at the WTC site on 9/11. Arrival on 9/11 was not associated with significant differences in IOS results (Table 5). Twenty-seven (27.3%) responders used a respirator with canister. Characteristics of exposed participants, by use of respirator with canister, are presented in Table 6. Those who used a respirator with canister demonstrated significantly lower R5 and R20 results and a marginally higher X5 result compared with responders not using this type of respirator (n = 72) (Table 5).
Table 5

IOS results by arrival at the WTC site on September 11 and use of respirator with canister among NYS WTC responders (NYS WTC IOS Study, 2007)

 

Geometric mean

p valuea

R5

 Arrived on 9/11

 

0.7237

  Yes

3.42

 

  No

3.36

 

 Used respirator with canister

 

0.0151

  Yes

3.16

 

  No

3.63

 

R20

 Arrived on 9/11

 

0.4691

  Yes

3.10

 

  No

2.99

 

 Used respirator with canister

 

0.0121

  Yes

2.85

 

  No

3.25

 

X5

 Arrived on 9/11

 

0.9399

  Yes

−1.11

 

  No

−1.11

 

 Used respirator with canister

 

0.0476

  Yes

−1.03

 

  No

−1.20

 

AX

 Arrived on 9/11

 

0.9029

  Yes

2.78

 

  No

2.83

 

 Used respirator with canister

 

0.1100

  Yes

2.44

 

  No

3.23

 

NYS New York State, WTC World Trade Center, IOS impulse oscillometry

aGeneralized linear models included potential confounders of gender, ever smoked, and obesity

Table 6

Characteristics of NYS WTC responders by use of respirator with canister (NYS WTC IOS Study, 2007)

 

Respiratora (n = 27)

No respiratora (n = 72)

p value

Gender

  

0.51

 Male

25 (92.6)

69 (95.8)

 

 Female

2 (7.4)

3 (4.2)

 

Race

  

0.95

 White

23 (85.2)

61 (87.1)

 

 Other

4 (14.8)

9 (12.9)

 

Hispanic

  

0.42

 Yes

2 (7.7)

9 (13.9)

 

 No

24 (92.3)

56 (86.1)

 

Obese (BMI ≥ 30)

  

0.32

 Yes

11 (44.0)

23 (32.9)

 

 No

14 (56.0)

47 (67.0)

 

History of smoking

  

0.74

 Yes

11 (40.7)

32 (44.4)

 

 No

16 (59.3)

40 (55.6)

 

History of asthma

  

0.76

 Yes

4 (14.8)

9 (12.5)

 

 No

23 (85.2)

63 (87.5)

 

History of allergiesb

  

0.19

 Yes

12 (44.4)

22 (30.6)

 

 No

15 (55.6)

50 (69.4)

 

Age

42 ± 9

42 ± 8

0.84

BMI

29.4 ± 4.6

29.2 ± 4.2

0.85

Height (in.)

70.7 ± 3.0

70.2 ± 2.7

0.44

Weight (lbs)

208.8 ± 35.8

206 ± 35.5

0.74

NYS New York State, WTC World Trade Center, IOS impulse oscillometry, BMI body mass index

aNumbers and percentages are shown for categorical variables, and means with standard deviations are shown for continuous variables

bSeasonal or perennial allergies

Discussion

This study was conducted to evaluate respiratory function, as determined by IOS, among a cohort of WTC responders with moderate overall exposure in comparison to a control group, 6 years post-9/11. NYS responders continue to demonstrate an increased risk of reporting lower and upper respiratory symptoms, including cough consistent with chronic bronchitis, suggesting persistence of symptoms among some responders. We hypothesized that persistence of respiratory symptoms might be due to peripheral airways disease. Prior studies of individuals with WTC exposures reported abnormalities on IOS that were not revealed with spirometry [3, 16]. Abnormalities likely reflected distal airway dysfunction [16]. However, our results did not reveal any association between IOS indices and WTC exposure or reported symptoms. Our findings did demonstrate an association between history of using a respirator with canister and lower R5 and R20 values, which confirms similar findings previously reported for another WTC cohort [3].

The precise nature of the pulmonary effects experienced by those with WTC exposures has not yet been fully elucidated. WTC responders have demonstrated declines in forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) with preservation of the FEV1/FVC ratio [18]. In one WTC responder cohort, the most common spirometric abnormality was a low FVC [4]. These findings could be interpreted to represent restrictive impairments [4, 18]. However, these findings may result from air trapping, causing a “pseudorestrictive” pattern [4, 17]. Nineteen percent of WTC workers and volunteers demonstrated a partial reversal of low FVC following bronchodilator, which can be seen with low FVC due to air trapping [4]. A 2002 report indicated that among responders who underwent high-resolution computed tomography (HRCT), half demonstrated air trapping [1]. In 2007, air trapping was reported in 25 of 29 WTC workers with lower respiratory symptoms who underwent HRCT. There was a significant correlation between air trapping and duration of dust exposure. The authors concluded that air trapping from small-airways disease might be responsible for some of the respiratory abnormalities in WTC workers [18].

We were unable to demonstrate that peripheral airways disease resulted from WTC exposure. It is possible that such disease did not manifest itself due to the generally moderate levels of exposure our cohort experienced [5]. Testing was conducted 6 years post-9/11. Had testing been conducted closer to the time of exposure, respiratory effects may have been seen with IOS. Because testing was conducted during the summer and early fall, as opposed to winter, participants who have reported persistent symptoms such as those seen in chronic bronchitis may have been less likely to be experiencing active symptoms at the time of their test. In one WTC cohort, 78.5% of responders experienced upper-airway disease, while 48.9% experienced lower-airway disease, suggesting that upper respiratory symptoms are a major component of the symptom complex in WTC responders [32]. Thus, it is possible that some symptoms characterized in our study as lower respiratory, such as cough, have in fact been caused by upper respiratory effects.

Our results concur with those reported by Skloot et al. [3], who demonstrated significantly lower R5 and R20 values among responders who reported having worn a respirator with canister during their WTC response activities. This suggests that using a respirator with canister was protective in terms of effects on the large airways. The higher respiratory resistance noted for those who did not use such respirators is consistent with the concept that effective respiratory protection would favor large airways [3]. This finding emphasizes the critical importance of appropriate respirator use during response activities involving exposure to dust and smoke. Proper fit-testing, respirator choice, and respirator use must be assured in future disasters to help mitigate potential long-term respiratory health impacts.

This study presents IOS data for a randomly selected control group with no history of asthma, emphysema, or COPD, who did not report a current cold, respiratory infection, or symptoms of seasonal allergies at the time of the IOS testing. Our analysis, confirming a result first reported by Skloot et al. [3], found that AX was significantly elevated in control participants with a history of smoking. In addition, smoking was associated with an elevated R5 and R20. These two IOS indices were also impacted by female gender, and most IOS indices were impacted by obesity. The control group data reported here will provide useful reference values for future researchers using IOS.

Due to logistical considerations and safety concerns, we did not conduct pre- and post-bronchodilator IOS testing. Testing was conducted at various locations and available resources for responding to medical emergencies were limited at some locations. Thus, administration of medication was not included in the study protocol. It is possible that pre- and post-bronchodilator testing might have revealed subtle differences in IOS indices.

This study has provided no evidence of an association between WTC exposure and peripheral airways disease in this cohort of responders. Results, however, do suggest that the use of a respirator with canister may be protective for central airways in responders exposed to dust and smoke. This emphasizes the importance of stressing proper respirator use in planning responses to future disasters. Our data also provide useful reference values for future IOS research.

Acknowledgments

The authors thank Janine Palome, RN, BSN, with the NYSDOH, for her valuable contributions to data collection for this study. They are also grateful to Rebecca Hoen, MS, also with the NYSDOH, for her assistance in data review. This work was supported by the Centers for Disease Control and Prevention (CDC) [Cooperative Agreement Number U1Q/CCU221059]. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of CDC.

Copyright information

© Springer Science+Business Media, LLC 2009