Human response to controlled levels of combinations of sulfur dioxide and inert dust

Human response to controlled 1---'7. Under controlled cDnditions in an environmental chamber 16 -healthy young volunteers were eJeposed to combinations of sulfur dioxide (S02) (2.6 or 13 mg/m 3) and inert plastic dust (2 or 10 mg/m;l) or of S09 (13 mg/m:i) and dust (1'0 mg/m 3) coated with vanadium. During periods in clean air and during eXiposures of 5-h duration nasal mucus flow rate, nasal airflow resistance, forced vital capacity, and subjective discomfort were measured. Reductions in nasal mucus flow rate, forced expiratory flow (FEF25-75 %), and discomfort were related principally to the S02 concentration. The combined effects of 802 and dust were, at the most, additive, and there was no indica tion of potentiation effects. No effect could be attributed to the coating of the dust with vanadium. we have reported studies of the effects of 80 2 5 and 25 ppm) and of inert (at 10 and 25 mg/m 3) on the nasal airway, on forced vital ca pacity, and on subjective discomfort

The effect -of combinations of gases and particles on man is a topic of great theoretical and practical interest. Knowledge about the .effects of pollutants is mainly based on exposure studies wilth a single pollutant without any admixtures. This situation is however very far from the real-life situation in outdoor air or indoors in the work environment, where many pollutants are present simultaneously. Therefore, investig.ations of the interacting effects of pollutants are important.
Two important pollutants of the outdoor air and the work environment are sulfur dioxide (80 2 ) and dust. Earlier Reprint requests to: Dr Ib Andersen, Arbejdsmiljeinstituttet, Baunegaardsvej 73, DK-2900 Hellerup, Denmark.
we have reported studies of the effects of 80 2 (at 1, 5 and 25 ppm) and of inert dust (at 2, 10 and 25 mg/m 3 ) on the human nasal airway, on forced vital capacity, and on subjective discomfort (4,7). The purpose of the present study was to evaluate, with the same experimental technique, the effects of exposures to 1 or 5 ppm of 80 2 (2.6 and 13 mg/m 3 , respectively) combined with 2 or 10 mg/m 3 of an inept plastic dust. These conditions were not meant to be representative of any particular environment, but were instead meant to produce a high absorpti'on of 80 2 and a high deposition of dust in the nose. The highest concentrations of 80 2 and dust used were equal to the timeweighted averages (TWAs) widely accepted as saf.e for workers over .a 40-h week (2). In one extra series the dust was coated with vanadium, a catalyst, which like carbon might increase the conversion of 80 2 to su:lfate (1, '9).

Materials and methods
The study took place in an environmental chamber (3) at the Institute of Hygiene, Aarhus, Denmark. The subjects were 16 healthy university students, 8 female and 8 male, with an average age of 22 a ,and age range of 19 to 28 a. None of them smoked heavily. Ten were nonsmokers, and six had smoked from 2 ,to 20 cigarettes per day for 1 to 9 a. Previous studies have detected no nasal function difference attributable to smoking -not a surprising fact since smokers deliberately draw the smoke thf{)ugh the mouth into the lungs. All had apparently healthy upper airways and were habitually nasal breathers. Nobody had a history of chronk or recent acute respiratory disease. The subjects were stud'ied in groups of four (two males and two females), each group undergoing the five different exposures on five consecutive days. The schedule for each group is shown in table 1. A fixed schedule was used in the hope of avoiding the confusing carry-over effects demonstrated earlier for 80 2 (4).
Each day began in the chamber with a control period with clean air at 23°C and 50 % relative humidity (RH), and the measurements ,each day during that period, together with a set of measurements performed prior to the first day (day 0), were taken as control values. Air supplied to the chamber was filtered through absolute and charcoal filters, and, during the control periods, no significant number of particles could be demonstrated in the unoccupied chamber with either an optical particle detector or a ,condensation nuclei counter. Air temperature was maintained at 23 ± 0.5°C and RH at 50 ± 50/0. The temperature of the chamber walls and air was identical. Air velocity was 10 ± 3 cm/ The dust employed in the study was a finely divided, fully polymerized plastic impregnated with carbon black (Rank Xerox toner 6R90005). The dust was chosen because the particles had an AED (aerodynamic equivalent diameter) of 2.2 to 15.3 pm, 55 0J0 of the particles by number and 27 0J0 by weight having an AED of 1.9 to 8.9 pm. A complete description of the dust, its generation as an aerosol, and its measurement has been given in a previous paper (7). Dust concentration and particle size were monitored every half hour by both gravimetric analysis and the optical microscopic counting of samples taken with a Konimeter® (C.Zeiss, VEB, Jena, German Democratic Republic). The dust used on day 5 was coated with vanadium, 0.58 mg/g of dust. The vanadium coating was performed by evaporation from a divanadium oxide solution in 50 0J0 ethyl alcohol, in which the particles were suspended. No significant changes in aerodynamic diameter resulted, and, by microscopy, vanadium crystals were never observed. The vanadium concentration was determined by atomic absorption spectrophotometry.
The first measurement each day was performed in clean air, whereas the second and the third measurements were made after 2-3 and 4-5 h of exposure, respectively, to the pollutants. Each day the schedule of events was the same, the only variable was ,the level of exposure to 802 and dust during the exposure periods. The concentrations were kept constant within ± 10 0J0 during the whole exposure period after an initial increase, which took about 45 min.
During each series of measurements the procedure was as follows: We measured nasal mucociliary flow by external detection of the motion of a resin particl,e (diameter 0.056 mm) labeled with 2 ,uCi of TC 99rn placed under direct vision on the superior surface of the inf.erior turbinate, 'a point in the mainline of inspiratory airflow. The technique has been described fully in our previous papers (4,7). On the completion of that test we measured nasal airfolow resistance through an oronasal plastic mask attached to a pneumotachometer. Air flow and oral pressure were disp1ayed on a storage oscilloscope. From this curve we calculated an area proportional to the effective cross-sectional area of the nasal airway (4). Thereafter, we recorded the forced expiratory vital capacity (FVC) with a single breath instrument (Vitalograp2l®, V,ita10graph Ltd, Bucking-halII1, England), from the record of wh'ich we later measured the forced expiratory volume in 1 s (FEVl.o) and calculated the forced expiratory flow (FEF 25 _ 75 %). Finally we obtained samples of pharyngeal air for dust analysis by dmwing chamber air through the nose and out the mouth at a flow rate of ahout 20 lImin while the subject performed a Valsalva maneuver. All of these procedures have been described fUlly in previous papers (4,7).
Throughout each day the subjects were asked to adjust the pointer on a voting machine expressing degree of airway irritation on a scale of 0 (for complete comfort) to 100 (for severe discomfort). The position of each individual's pointer was concealed from fellow subjects, and it was recorded continuously. At the end of each day the subjects were also closely questioned as to the degree and nature of discomfort they had experienced. All procedures were in accord with the Helsinki declaration Fluids were permitted ad' libitum; lunch was served during the third hour; and subjects were allowed to walk about, play .:ards, or read as they wished. No smoking was allowed.
For the physiological measurements an analysis of variance was used (13). Due to a "within subject variation" and a variation between measurements performed on different days, the analysis was based on differences obtained by sUbtracting the values measured during the control condition from the values measured during the exposure .peri{)d of the same day.
The results of the discomfort voting data were analyzed with the nonparametric Friedmann's test. as the assumptions for performing a parametric analysis of vari-ance were not fuHmed (11). The level of significance was always 5 0/0.

Nasal mucus flow
Results on nasal mucus flow are available from 14 subjects only, as two subjects prov·ed to have a mucostasis on the first day of exposure and during the rest of the study. To reduce the local radiation on the epithelium, we removed the tagged bead after 10 min, and theref.ore no mucus flow measurements were obtained. In the prestudy investigation (day 0) these two subjects had normal mucus flow rates. There was no apparent reason for the mucostasis. The mucus membrane appeared normal, and there were no symptoms or signs indicating a common cold.
The is also shown. The results of the measurements in clean air on day 0 and day 1 did not differ. During the pollutant exposures the mucus flow rate in the anterior third of the nose always decr·eased with time during exposure. In the middle third 0.£ the nose (slit 3 to 4) no effecls of dust concentration or of the vanadium coating of the dust were found. Irrespective of the dust concentration, the flow rates were reduced in subjects exposed to 80 2 (p < 0.01 for 80 2 2.6 mg/m 3 and p < 0.001 for 80 2 13 mg/m 3 ). The effecl of the 80 2 concentration of 13 mg/m 3 was greater than that of 2.6 mg/m 3 (p~0.03). In the posterior third of the nose no exposure effects appeared. The responses of the male and female subjects and of the smokers and nonsmokers did not differ.

Airway resistance
The results for airway resistance were based on measurements from 16 subjects. The airflow resistance has been expressed as the cross-sectional nasal area. Irrespective of the dust concentration, exposure The forced expiratory test showed no significant changes in VC 'and FEV1.0 at any exposure. The FEF 25 _ 75 % was significantly reduced during exposure to high 80 2 concentrations on days 2 and 4. There was no dirfference in the response of smokers and nonsmokers. The measurements of dust particles in the air after nasal transit in relation to the same measurement in chamber air revealed an eff.ectiveness of the nasal filter similar to that reported in our study of inert dust alone (7). 80 2 was not measured in pharyngeal air during this study. The nasal absorption of 80 2 has been described in a previous paper (4).
The averag,e -subjective discomf.ort votes, as indicated on the voting apparatus, are shown in fig 2. In the control periods the voting never exceeded two units on a scale from 0 to 100. When the pollutants were added to the air, the discomfort increased in almost direct proportion to the concentration of the pollutants. The responses on day 1 and 3 were very similar, as was also true for days 2, 4, and 5.
The discomfort votings during exposure to any combination of 80 2 and dust were all significantly increased when compared to votings during clean air exposures (p < 0.01). There was no difference be-   (12), and Burton et al (8) studied the effects of sodium ·chloride aerosol and S02 on airway resistance during 15-3{) min of exposure, but none of them were able to demonstrate in man the potentiation found due to this combination in experiments with guinea pigs (1). Ulmer (14) studied 16 subjects exposed 8 hid for 4 d to a combination of S02 (17 mg/m 3 ) and coal dust (8 mg/m"l The addition of coal dust did not influence the increase of airway resistance caused by S02' No previous study exists of the effects of combinations of S02 and dust on human nasal airways. Our present study was carried out with exactly ,the same techniques as in two previous studies, one on the effects of SO( 4) and the other on inert plastic dust (7). In fig 3 and 4 the results of the present study and the two previous ones are compared. The measurements of the three groups had different means even under Discussion tween the votings during exposure to the two different dust concentrations and with or without a vanadium coating. The discomfort was greater during high than during low levels of exposure to S02' The average voting never exceeded 20 on a scale of 0 to 100, where the scale from 0 to 33 was marked "slight discomfort. " We also asked about the symptoms experienced in the chamber at the end of each day, and we repeated the questions the following morning. The questions covered eye, nose, mouth, throat and 10wer airway, irritation, sneezing and coughing, headache, and dizziness. After these specific questions the subjects were asked if they had experienced symptoms other than those mentioned. The main complaint was irritation of the nose and throat, most often expressed as a feeling of dryness. The number of subjects wilth these complaints on days 1 through 5 was 5, 11, 5, 8, and 9, respectively. It appears that this irritation effect was the most pronounced the days on which the highest S02 concentration was used. The f.ollowing morning all the symptoms had disappeared. the control conditions, and therefore the measurements for each group, under any condition, have been expressed as the percentage of the measurements made under the control conditions in clean air on the same day. To enable a comparison between the three experiments, an analysis of variance similar to that ,used in the present study was performed on the results from the previous two exposure experiments. The calculations were based on the data obtained at concentrations similar to those in the combination study, and therefore the results from the highest S02 and dust levels (65 and 25 mg/m:l, respectively) were omitted.
In the upper part of fig 3 the nasal mucus flow rate measurements from the middle third of the nose are shown. There was a tendency towards a decreased mucus flow rate in subjects exposed to S02 only (0.05 < P < 0.1), whereas there was no effect of exposure to dust only. In the present experiment with combined exposure to both pollutants, the S02 exposure caused a significant decrease in the mucus flow rate (p < 0.01), more pronounced at the high than at the low concentration. This result indicates that the depression of nasal mucus flow caused by S02 might be increased by an addition of plastic dust to the S02 air mixture. It is apparent from fig 3 that this decrease in mucus flow is a pure additive effect without any sign of potentia1tion. With respect to nasal mucus flow it can be stated that the depressing effect of the studied pollutants was greater in the nose anteriorly than posteriorly, that S02 had a depressing effect whereas the plastic dust with or without a vanadium coating had no effect, and that no potentiation occurred during combined exposures to both pollutants.
The middle part of fig 3 shows the cross-sectional flow area variations. The analysis of variance showed that during exposure to S02 alone the subjects had a smaller cross-sectional area than in clean air (p < 0.05), There was no effect due either to the S02 concentration or to the length of the exposure period. In contrast, during exposure to dust only the subjects showed an increase in the cross-sectional nasal area (p < 0.01), and even under these circumstances the length or concentration of the exposure did not influence the effect. 6 In the combined exposure study subjects exposed to the low S02 concentration had a decrease, and subjects exposed to the high S02 concentration had an increase in cross-sectional area. This result differs from that of the exposure to S02 alone and indicates that dust is able to neutralize the constricting effect of high S02 concentrations.
The lower part of fig 3 shows the FEV1,0' The analysis of variance showed that the air flow was reduced during S02 exposure (p < 0..05) and during exposure to the low dust concentration (p < 0.01), whereas no significant difference appeared between the -control condition and the high dust concentration. These results from the dust only experiment are mainly due to two subjects (no 1 & 3), who had an unusually high flow rate during the control condition on the first day. If these measurements are excluded, there are no significant differences due to dust exposure. During the combined exposure experiment no significant changes were found under any condition. As the clear flow-decreasing effect of SOi was not found in the combination experiment, it may be concluded that the concomitant dust inhalation changes the response to the inha'lation of SOi' No statistically significant differences were observed in the discomfort voting (fig 4) of the subjects exposed only to S02 or dust at the low concentration. During the exposure to dust in a concentration of 10 mg/m'!, the discomfort was significantly higher (p < 0.05) than under the control conditions. In contrast, a significant increase in discomfort voting (p < 0.05) was registered for all combinations of S02 and dust. From these results we conclude that combined exposure to dust and S02 produces an additive effect with respect to experienced discomfort.
From the very few studies in the literature and from our own investigations, it is also apparent that the presence of an inert dust -adds to but does not potentiate the effects of S02 on the upper airways during exposures of 5-h duration. The dust particles used were of a size which would not penetrate into the deep lung area optimaUy. The effects of the dust itself and of dust coated with vanadium were indistinguishable.
The subjective discomfort occurring with both dust and S02 exposur·e bears a strong resemblance to the complaints generally attributed to dry ambient air. In our studies of exposure to dry clean air there was a remarkable absence of subjective discomfort (5,6). The feeling of dryness experienced by the subjects in this study, therefore, was probably due to a stimulation of the trigeminal nerve endings in the nose elicited by the air pollutants.