Upper Respiratory Tract Environmental Tobacco Smoke Sensitivity1-4

REBECCA BASCOM, THOMAS KULLE, ANNE KAGEY-SOBOTKA, and DAVID PROUD Introduction

Environmental tobacco smoke (ETS), defined as the tobacco smoke that nonsmokers inhale, is one of the most common forms of indoor air pollution. Investigators have focused on irritation of the eyeand of the upper respiratory tract that begins upon exposure to ETS containing carbon monoxide (CO) concentrations as low as 1.5to 3 ppm (1-4). Over 15 yr ago clinicians recognized that some individuals were sensitive to tobacco smoke, reporting symptoms of rhinorrhea, congestion, or sneezingafter smoke exposure (5-7). However,in 1986 the Surgeon General's Report on the Health Consequences of Involuntary Smoking could cite only subjective data to support the existence of upper respiratory sensitivity to ETS (8). The presence of increased lower respiratory responsiveness in subjects is also controversial, sinceconflicting data exist (9-15). It has been suggested that a specific IgE-mediated allergy to tobacco smoke may exist to explain ETS rhinitis, and some clinicians have noted that allergic subjects seem to be clinically more sensitivethan nonallergic subjects (5, 16, 17). However, only a small proportion of ETS-S subjects have positive skin tests to extracts of tobacco leaf. On the other hand, combustion products of tobacco are very different from the unburnt leaf and contain many potential antigens (18). Therefore, investigators have recently urged a reexamination of "tobacco smoke allergy" (11). Techniques of controlled inhalation challenge combined with measurements of symptoms, physiologic changes, and the performance of nasal lavage have enabled the elucidation of the pathogenesis of rhinitis induced by immunologic as wellas non-immunologic mechanisms (19-21). In studies of antigen- or coldinduced rhinitis, a historically symptomatic group has been selected for systematic study and compared with a historically asymptomatic control group. We used a similar approach in the present study to address the problem of sensitiv1304

SUMMARY Some patients report rhinitis symptoms after exposure to environmental tobacco smoke (ETS), but ObJective assessments of this response hava been lacking. Furthermore, the mechanism ofthls response Is unknown. We aasessed the frequency of ETS-related symptoms by administering a questionnaire to rr healthy nonsmoking young adults who were participating In an unrelated stUdy. Of the subJects 34% (26 of n) reported one or more rhinitis symptoms (congestion, rhinorrhea, or sneezing) following ETS exposure. We then exposed 10 historically ETS-sensltlve (ETS-S) and 11 historically ETS-nonsensltlve (ETS-NS) subjects to 15 min of clean air followed by 15 min of sidestream tobacco smoke (CO concentration of 45 parts per million). At selected time points during these procedures we recorded symptoms, posterior nassl resistance, and spirometry and performed nasal lavages. ETS-S but not ETS-NS SUbJectsreported significant (p < 0.01) Increases In nasal eongestlon, headache, chest discomfort or tightness, and cough following exposure to sldestream tobacco smoke. Rhinorrhea symptoms were greater and more prolonged In ETS-S subJects compared to ETS-NS subJects. Significant (p < 0.01) Increases In perception of odor and In eye, nose, and throat Irritation occurred in both study groups, but ETS-S subJects reported significantly more nose and throat Irritation. No significant changes In posterior nasal resistance occurred In the ETS-NS group but a significant Increase occurred In the ETS-S SUbJects, with the resistance rising from 3.8 ± 0.5 cm H 20/LIs (mean ± SE) preexposure to a peak of8.0 ± 2.7 cm H 20/Us 20 min after completion of the smoke expOsure (p < 0.001). Pulmonary function changes were of a small magnitUde « 7%) In both groups. No significant Increases occurred In the levels of histamine, albumin, kinin, or TAMEesterase activity In nasal lavage fluids. These data provide obJective evidence of an Increased responsiveness to ETS In historically sensitive subJects but suggest that an allergic, IgE-medlated mechanism Is unlikely sInce no histamine wes detected despite the symptomatic and physiologic response. AM REV RESPIR DIS 1991; 143:1304-1311

ity to ETS. Wechose a relativelyhigh concentration of sidestream tobacco smoke, similar to that measured in bars or taverns

with subjects with no history of ETS sensitivity (ETS-NS).

(22).

(Received in original form February 26, 1990 and in revised form January 3, 1991)

In initiating the study of ETS rhinitis, it appeared desirable to study the phenomenon in individuals who did not perceive themselves as ill from ETS or other chemicals. The perception of illness could relate to the severity of the response but could equally result from a psychological interpretation of the meaning of the response. We chose to avoid studying a group of individuals who had an illness or illness behavior related to ETS. However, it wasunknown how common ETS-related rhinitis symptoms were among healthy young adults and, therefore, whether recruiting from this group would be productive. The aims of this study were to determine the frequency of rhinitis symptoms among healthy nonsmokers and then to characterize the symptomatic, physiologic, and inflammatory upper respiratory response to controlled challenge with tobacco smoke in subjects with a history of ETS sensitivity (ETS-S) compared

1 From the Environmental Research Facility, Division of Pulmonary and Critical Care Medicine, Department of Medicine, University of Maryland School of Medicine, and the Division of Clinical Immunology, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland. 2 Supported in part by the American Lung Association of Maryland, the Special Research Initiative Support Program of the Designated Research Initiative Funds, University of Maryland, and Grant Nos. HL 40945, HL 32272, and AI 08270 from the National Institutes of Health. 3 Correspondence and requests for reprints should be addressed to Rebecca Bascom, M.D., M.P.H., University of Maryland Environmental Research Facility, MSTF-8oo, 10 South Pine Street, Baltimore, MD 21201. 4 See NAPS document 04869 for 13 pp of supplementary material. Order from NAPS clo Microfiche Publications, P.O. Box 3513, Grand Central Station, New York, NY 10163-3513. Remit with your order $7.75 (US. funds on a US. bank only) for photocopies or $4.00 for microfiche. Outside the US. and Canada, add $4.50 postage ($1.50 for microfiche). Purchase orders accepted but add $15 handling charge plus any applicable postage.

1305

ENVIRONMENTAL TOBACCO SMOKE SENSITIVITY

Methods

Nomenclature In this report ETS is defined as the components of cigarette smoke that nonsmoking individuals encounter in the environment. ETS includes both sidestream tobacco smoke (STS) and smoke that has been filtered by the lungs of the smoker. STS is the smoke issuing from the burning end of a cigarette. Among the factors that affect the properties of STS in ETS are the sizeof the room, ventilation rates, and the number of smokers in the room (8). Differences exist, but the components ofETS and STS are similar, so STS may be used as a reasonable surrogate for ETS for challenge studies (23). We define ETS sensitivity as hist?rical clinical upper respiratory hyperresponsiveness to ETS. The term was chosen because it is commonly used, but we do not intend to imply an allergic etiology since the pathophysiology of ETS sensitivity has not been determined. Survey: Prevalence of Historical ETS-related Symptoms At the time the authors were considering initiating studies of ETS rhinitis, the University of Maryland Environmental Research Facility (UMERF) was conducting the third year of a 3-yr study that assessed the effect of exposure to N0 2 on viral infectivity. Leaflets were distributed and notices were posted at University of Maryland at Baltimore, a campus comprised of professional schools of medicine, dentistry, nursing, pharmacy, and social work. Notices were also posted at the University of Maryland at Baltimore County, a nearby campus. The advertisements asked for healthy young adults who werewilling to participate in a study at the Environmental Research Facility. An unusually large screening effort was required because only individuals who were seronegative to influenza A would qualify for participation. A total of 77 healthy nonsmoking adults aged 18 to 45 presented for screening for possible participation in the N0 2-virus study. Potential subjects were informed that the criteria for participation in the study were that they be free of respiratory illness and be seronegative to the influenza virus. At the screening visit subjects completed the usual UMERF history questionnaire, which included questions about illnesses, allergies, and respiratory symptoms. New, minor additions to the standard questionnaire were two questions. The first asked about the occurrence of annoyance (none, mild, moderate, or severe) on exposure to ETS. The second asked about the occurrence of ETS-related eye irritation, nasa~ irritation, nasal congestion, sneezing, rhinorrhea, postnasal drip, chest tightness, and wheezing on exposure to ETS. If subjects answered yesto a specific symptom, they were asked to record the time from ETS exposure to onset of symptoms, and the duration of the symptom. The responses to these questions were not a criteria for participation in the N0 2-virus study. There was no reason for the technical staff or the study participants

to expect that the response to this question would in any way affect a subject's ability to participate in the study for which they were being screened. The questions relating to ETS responsiveness were tabulated and expressed'as the percentage of subjects who checked yes for each individual symptom. We also calculated the percentage of subjects with any rhinitis symptom (nasal congestion, sneezing, rhinorrhea, or postnasal drip).

Controlled STS Exposure Subject selection and characterization. Subjects who were screened for participation in the N0 2-virus study were the primary pool of subjects recruited for the second phase of the study. Individuals who reported the most and the fewestsymptoms werecontacted first. Months had elapsed since the first questionnaire, so a second questionnaire was administered. Prospective subjects were asked to rate symptoms resulting from historical ETS exposure on a 0 to 5 scale (0 = symptom absent; 1 = mild; 2 = mild to moderate; 3 = moderate; 4 = moderate to severe; and 5 = severe). For each subject we calculated a "historical ETS rhinitis index" and a "historical ETS irritant index." The historical rhinitis index was the sum of symptoms of rhinorrhea, nasal congestion, and sneezing. The historical irritant index was the sum of symptoms of nose and eye irritation. Individuals were classified as ETS-S if they had a historical rhinitis index of 3 or greater and were classified as ETS-NS if they had a historical rhinitis index of 1 or less. A total of 21subjects participated in the controlled challenge study (table 1): 10subjects were historically ETS-S and 11 subjects were historically ETS-NS. The atopic status of all subjects was determined by skin prick tests to a panel of common environmental allergens, including trees, grasses, ragweed, mold mixtures 1 and 2, and house dust mite (Greer Laboratories). Subjects were labeled as atopic if they had one or more positive skin tests (~ 2+). The grading scale we used for skin prick tests is the one used by the Johns Hopkins Clinical Immunology Division. A score ~ 2+ was defined as the presence of a wheal that is greater than or equal to half the diameter of the subject's histamine wheal. Sidestream tobacco smoke generation. An

environmentally controlled research exposure chamber was used to limit exposure to STS and to provide constant conditions of temperature and relative humidity. The environmental chamber air supply consists of ambient air passed through high-efficiency particulate absolute (HEPA) filters to permit cleanliness approaching class 100 « 100particles/foot" > 0.5 11m) and activated carbon filters to remove gaseous pollutants. The environmental chamber consists of two rooms: a 41-m3 clean air room and a 22.2-m3 exposure room. A ventilation rate of 3.0 mvmin in the exposure room enabled a complete air change every 7.5 min. All exposure room air was exhausted to the outside without recirculation. STSwasgenerated inside the exposure room by using a smoking machine consisting of a glass manifold holding 12cigarettes attached to a vacuum source outside the exposure chamber. Intermittent puffs were generated by occluding a Y connector. All mainstream smoke was exhausted externally. The 12cigarettes werereplaced approximately every5 min during the study. The concentration of carbon monoxide (CO) was monitored continuously using an Ecolyzer Model 2100CO monitor (Energetics Science Inc., New York, NY), and a target concentration of 45 parts per million (ppm) CO was achieved by controlling the number of cigarettes puffed. Particle concentrations were determined using a Climet Model 226-210 multichannel particle analyzer (0.3to > 1011m; Climet Instruments Inc., Redwood, CAl. The Climet is an optical device that counts particles based on light scattering. At high particle densities two small particles may simultaneously fall within the light beam and can be enumerated as a single larger particle. Dilution studies indicate that coincidence accounts for most particles reported as larger than 2 11m (unpublished data).

Experimental Methods Symptom questionnaire. A symptom questionnaire was used to assess the intensity of perceived odor and the presence and severity of eyelrritation, headache, nasal congestion rhinorrhea, nose and throat irritation, chest tightness or discomfort, and cough. Subjects classified their response as 0 = none, 1 = mild, 2 = mild to moderate, 3 = moderate, 4 = moderate to severe, and 5 = severe. Posterior nasal resistance. Nasal resistance

TABLE 1 SUBJECT CHARACTERISTICS ETS-S

ETS-NS

Age, median (range) Sex, M:F

23 (20-24) 3:7

25 (20-30) 4:7

H/wk E1;S exposu re

5.1 ± 2.0

6.5 ± 2.6

Historical symptom score Rhinitis Irritation

4.5 ± 0.6 7.1 ± 0.9

0.2 ± 0.1" 4.5 ± 0.6"

70

27t

Atopic (.. 1 positive skin test), % •p

< 0.001,

t p < 0.05,

ETS-S versus ETS-NS. ETS-S versus ETS-NS.

1306

BASCOM. KUI.I.E. KAGEY·S08OTKA, AND PROUD

(f) was measured using posterior rhinometry (24, 10 0 ~ 25). A full-face airline respirator was used that U enabled an airtight fit but no deformation 75 Q) .....-, of the nasal airways. Flow was measured by .0 modifying the breathing tube assembly with 50 ~ a heated pneumotachygraph. The exhaust (f) valve of the respirator was replaced with a 25 \4-inch Tygon~ (Norton Performance Plastics, Akron, Ohio) pressure tap that extended through the subject's tightly closed lips into the oropharyngeal region. The differential pressure between the oropharyngeal catheter and the pressure exterior to the nose was measured with a differential pressure transducer. The pressure transducers (Validyne Model MP45) were connected to a Validyne Model MCI-3 power supply demodulator, a Tektron- Fig. 1. Prevalence of historical environtnentaltobacco ix 564 storage oscilloscope, and an IBM-PC smoke-related symptoms reported in a questionnaire completed by 77 healthy nonsmoking young adults. equipped with a Data Translation DT-28OI AID input board. The subject peformed four to six consecutive panting maneuvers through the nose point between treatments (clean air versus around functional residual capacity, and the STS) and between the ETS-S and ETS-NS groups. Tukey's test for comparisons between results were displayed on the storage oscilloscope. The digitized pressure-flow signals were means was used to compare group means for analyzed by dividing the driving pressure (SP) each time point (31). The symptom responsat a flow rate of approximately 0.5 Lis for es were standardized by subtracting the t = both inspiration and expiration by the corre- ovalue, and the nasal resistance response was sponding flow to give nasal resistance in em analyzed both by computing a ratio and by H 20/Lis. Three sets of panting maneuvers subtracting the t = 0 value. Symptom scores were performed, and inspiratory and expira- on the 2 exposure days were compared with tory resistance were then averaged to deter- a correlation coefficient. The spirometric mine mean nasal resistance. measurements were standardized by computSpirometry. Spirometric measurements ing the ratio of each subject's pulmonary funcwere made using a lO-LStead-Wells spirometion measurement to the corresponding baseter interfaced with an Eagle II microproces- line (t = 0) measurement. For comparison sor (Warren E. Collins, Braintree, MA). The of the concentration of mediators in the inispecifications for calibration and operation tial lavage the Mann-Whitney U test was used met the American Thoracic Society criteria with the StatVje~ 512+ computerized statistical package (BrainPower, lnc., Calabasas, (26). A minimum ofthree technically accept ~ able forced expiratory maneuvers were CAl. A two-tailed level of 0.05 was considobtained. ered significant for all comparisons. Nasal lavage and mediator analysis. SubExperimental Protocol jects tilted the head back 45° and closed the palate, and 5 ml 0.9% sodium chloride All subjects were studied on 2 days separated by at least I wk. On I study day measurewarmed to 37° C was instilled in each nostril. After 10 s subjects flexed the neck and ex- ment of nasal resistance and spirometry was pelled the lavage fluid into a basin. The la- performed, and on the other study day nasal vage fluid was immediately decanted and lavage was performed. Symptoms were replaced on ice until further processing at the corded on both study days. The 2 study days conclusion of the experiment. After centrifuwere needed because the performance of nagation (4 0 C, 15 min, 3000 rpm) the sol phase sal lavage could interfere with physiologic measurements, was pipetted and divided for mediator analysis. Samples for histamine were stored in 1.6070 After a 0.5-h equilibration period in the perchloric acid at - 20° C and then assayed clean room, baseline measurements were obusing a spectrofluorometric assay (27). Sam- tained. Subjects remained in the clean room ples for TAME-esterase activity were stored for a I5-min clean air (Air) exposure, and at - 20 C and assayed via a previously pubmeasurements were repeated. SUbjects then Iished radiochemical assay (28). Kinin sam- entered the exposure room, were instructed pies (0.5 ml sample and 0.125ml 0.2 M EDT;,).) to breathe through the nose, and sat quietly and albumin samples were frozen at - 80° C for the I5-min STS (Smoke) exposure. Subuntil analysis using previously published jects then exited the exposure room and radioimmunoassays (29, 30). The lower limit returned to the clean room where symptoms of detection was I nglml for histamine, 1000 and physiologic measurements were obtained cpm for TAME-esterase activity, I nglml for immediately and 20 min later (20' post). On albumin, and 20 pg/ml for kinins, the nasal lavage study day four nasal lavages Statisticalanalysis. All values are expressed were obtained at baseline to reduce the rnedias mean ± SEM. A two-way analysis of vari- ators to a low and stable level. Following the ance (ANOVA)for repeated measures was em- I5-min air exposure three samples were obployed to compare responses at each time tained at lO-min intervals, and following the 0

0.3 0.5 0.7 1.0 Z.O 3.0 5.0 10

Particle size (urn) Fig. 2. Reproducibility ot the sideslream tobacco smoke during challenge studies. Shown are the concentration and particle size distribution ot the sidestream tobacco smoke on 6 of the 9 separate stUdy days as measured by a Climet 226-210 particle size analyzer. Particles ranging from 0.3 to 10 11m are enumerated. Actual particle size is overestimated at this concentration of tobacco smoke because of coincidence of particles at the light source of the Clime\.

I5-min STSexposure an additional three samples were obtained at la-min intervals. Results

Prevalence of Historical ETS-related Symptoms The prevalence of historical ETS-related symptoms as reported by 77 healthy nonsmoking young adults who completed a screening questionnaire is shown in figure 1. The subjects were 25 ± 0.7 yr old (range 18to 44, median 24);32 were males and 45 were females. No history at all of irritant or rhinitis symptoms was reported by 17ltJo of subjects. Eye irritation was the most common symptom (820,10), and one or more symptoms of rhinitis were reported by 340,10 of the subjects. The degree ofETS-related annoyance reported by the subjects ranged from severe in 100,10, to moderate in 380,10, to mild in 390,10, and to none in 130,10.

Controlled Sidestream Tobacco Smoke Exposure Subject characteristics. A total of 21 subjects were exposed to STS in the environmental chamber. The 10 ETS-S and the 11 ETS-NS subjects were similar with respect to age, sex, and hours per week of ETS exposure (table 1). The historical ETS-rhinitis index and the historical ETS-irritation index were significantly greater for the ETS-S subjects. The ETS-S subjects were more likely to be atopic by skin prick test (70 versus 270,10, p < 0.05). Exposure characterization. The temperature (22.2 ± 0.5" C) and relative humidity (65 ± 2%) were consistent on all study days. The particle concentration and size distribution were similar on all

1307

ENVIRONMENTAL TOBACCO SMOKE SENSITIVITY

il'SlL:: WSL6: WSLrt WSlL! 8

*

4

'" 3

*

2 1

"

~ ~

'" 0

Pre

Air

*

'*

Smoke 20' posl

~

4

~

1

~ ~

'" 0

I

'" 3

I ~ .-At Air

Smoke 20' post

a.,

~

p~

84 ~ 3 02 ~ 1

..

'" 0

....

**

Air

Smoke 20' post

WSL;. ~~

~

;n

...

2

*

1 0

-+Pre

Ar

$n'll::N2O'post

~

2 ~ 1 '" 0

*

Smoke 20' post

++

**

Pre

Air

..

-+-

Smoke 20' posl

NOSE-THROAT IRRITATION

8WS~' 4 * '"*

'" 3

~~

2 1

'" 0

/1'Pre

Air

a:

5

u

4

o

**

'1"*

++

CHEST TIGHTNESS

~

Air

:

RHINORRHEA

Pre

,

I

HEADACHE

84

w

*

NASAL CONGESTION

Pre

•

EYE IRRITATION

ODOR

Fig. 4. Individual subject symptoms following sldsstream tobacco smoke exposure. Circles show the score for each individual immediately following tobacco smoke exposure: 0 = symptoms absent; 1 = mild symptoms; 2 = mild to moderate symptoms; 3 = moderate symptoms; 4 = moderate to severe symptoms; 5 = severe symptoms. Closed circles = ETS-S subjects; open circles = ETS-NS subjects.

r..n L

o

I0...

L

>-

3

• •

. ooo.to o•

2 ftdl . 1 •°

r..n

00 •

0

00

'

° °

• •

I

•

. e::> o,. o • 00 " 00

~

•

•

00

o ·~

*

Smoke 20' post

COUGH

!WS~ ~ . 0-

*

2

~ 1 ~ 0

-+Pre

Ail Smoke20' poll

Fig. 3. Symptomatic response to sidestream tobacco smoke exposure. Values represent the mean ± SEM for 10 ETS-S subjects (solid lines) and 11 ETSnonsensitive subjects (broken lines). Pre is the basel ine symptom score; Air is the symptom score immediately after the 15-min clean air exposure; Smoke is the symptom score immediately after the 15-min exposure to sidestream tobacco smoke at 45 ppm CO; 20' post is the symptom score 20 min after completion of the smoke exposure. Asterisks indicate p < 0.05 versus baseline; double asterisks indicate p < 0.Q1 versus baseline; daggers indicate p < 0.05 ETS-sensitive subjects versus ETS-nonsensitive subjects; double daggers indicate p < 0.Q1 ETS-S subjects versus ETS-NS subjects.

study days (figure 2). At the concentration of STS used in this study coincidence of particles occurs in the light beam of the Climet, causing the enumeration of two small particles as one large particle and apparent increased variability of the higher particle sizes. The carbon monoxide concentration ranged from 44 ± 3 to 48 ± 3 ppm on different study days. Symptoms. The symptom scores for the ETS-S subjects and the ETS-NS subjects are shown in figure 3. Odor and eye irritation increased significantly for each group (p < om, smoke and 20' post versus baseline), and the magnitude of the eye irritation was significantly greater for the ETS-S group (p < 0.05, ETS-S versus ETS-NS, 20' post). Symptoms of nasal congestion and headache rose significantly for the ETS-S group (p < om, smoke versus baseline) but did not rise for the ETS-NS group. The magnitude of the headache symptom was significantly greater for the ETS-S group (p < 0.01, smoke and 20' post, ETS-S versus ETS-NS). Symptoms of rhinorrhea and nose and throat irritation rose in both

groups immediately after the smoke exposure (p < 0.01, smoke versus baseline). Rhinorrhea was significantly greater in the ETS-S group 20 min postsmoke (p < 0.01, 20' post smoke, ETS-S versus ETS-NS), but nose and throat irritation was significantly greater in the ETS-S group immediately postsmoke (p < 0.01, smoke, ETS-S versus ETS-NS). Lower respiratory symptoms of cough and chest tightness increased significantly only in the ETS-S group and were significantly greater in the ETS-S compared with the ETS-NS group (p < 0.01, smoke, ETS-S versus ETS-NS). The individual symptom scores immediately following tobacco smoke on 1 study day are shown in figure 4. Symptoms were recorded on both study days. The mean values were closely correlated on the 2 study days (r = 0.89). The correlation coefficient for individual scores on the 2 days was as follows: odor (r = 0.91), eye irritation (r = 0.88), nasal congestion (r = 0.69), headache (r = 0.79), rhinorrhea (r = 0.66), nose and throat irritation (r = 0.74), chest tightness (r = 0.75), and cough (r = 0.79). Nasalresistance. The baseline nasal resistance was 3.8 ± 0.6 em H 20/L/s in the ETS-S group and 3.7 ± 0.3 em H 20/L/s in the ETS-NS group. The effect of STS exposure on nasal resistance, expressed as the increase in nasal resistance above baseline preexposure values, is shown in figure 5. The nasal resistance in the ETS-S group rose 0.4 ± 0.2 em H 20/L/s post-clean air, 2.4 ± 0.6 cm H 20/LIs postsmoke (p = NS), and 4.2 ± 2.2 em H 20/L/s 20 min postsmoke (p < 0.01, air versus 20 min postsmoke). The nasal resistance in the ETS-NS group did not change significantly. The nasal

resistance between the two groups was significantly different 20 min postsmoke (p < 0.01). Individual nasal resistance responses are shown in table 2. The nasal resistance response was also calculated as the ratio of the postexposure and t = 0 values. The nasal re7

** 6

w

o

Z