Archives ol

Toxicology

Arch Toxicol (1988) 61 : 229-236

9 Springer-Verlag 1988

The effects of n-butanol vapour on respiratory rate and tidal volume Uffe Kristiansen, Anne Marie Vinggaard, and Gunnar Damglrd Nielsen Danish National Institute of Occupational Health, Baunegaardsvej 73, DK-2900 Hellerup, Denmark

Abstract. Exposure to n-butanol vapour gave rise to a sensory irritation response which was measured by the reflexively induced decrease in respiratory rate in mice according to the American standard method (E981-84). The response reached maximum within the 1st min of exposure. In this period the expected threshold response (RD-0) and the concentration expected to depress the respiratory rate by 50% (RD-50) were extrapolated to be 233 ppm and 11696 ppm, respectively. The response followed the dynamics of a bimolecular reaction between butanol and the sensory irritant receptor. For concentrations below 3000 ppm, the response faded due to desensitization. However, concentrations above 3000 ppm gave rise to a new decrease in respiratory rate due to activation of lung receptors. Two types of lung receptors, probably J-receptors and stretch receptors, were involved. The sensory irritation response measured by the standard method gave a threshold response which was comparable to that found by electrophysiological experiments in rats. The irritation response in man as well as the maximum allowable concentration in the working environment were adequately predicted from the sensory irritation response in mice.

Key words: Receptors - Trigeminus - Vagus - Sensory irritation - Lungs - Mice

sponse is termed "pulmonary irritation" and is believed to be mediated via the vagal J-receptors. In humans sensory irritation gives rise to an effect varying from slight irritation to a burning and painful sensation (Alarie 1973). Pulmonary irritation gives rise to discomfort, and to feelings of tightness in the chest and dyspnoea (for references see Nielsen and Bakbo 1985 a). Both types of reactions are commonly found in the working environment. Thus, about 40% of the threshold limit values (TLVs) from the American Conference of Governmental Industrial Hygienists (ACGIH) have been set due to irritation of the upper respiratory tract (Alarie 1981a). In general, these standards are the basis for the Scandinavian standards and are the maximum allowable concentrations in the industrial environment. In this investigation the effect on the trigeminal nerve endings is determined by the reflexively-induced decrease in the respiratory rate in mice (ASTM 1984). The effects on the lower respiratory tract and the effect on the central nervous system (CNS) were also determined by the decrease in respiratory rate (for discussion of the method see Alarie 1981b, and Nielsen et al. 1985b). The differentiation between effects on the lungs and on the CNS was obtained by comparing effects of the doses absorbed by inhalation and the effects following intraperitoneal injection.

Introduction

Materials and methods

Butanol is a widely used organic solvent (/kstrand et al. 1976). Among other things, it is used within human physiology as a reference substance in the determination of odour intensities (e.g. Dravnieks 1977). The odour sensation is mediated partly by the olfactory nerve and partly as an irritation sensation by the trigeminal nerve (Cain 1974, 1976). Little is, however, known about the irritating effect on the eyes and on the upper respiratory tract. These effects are mediated via receptors in the trigeminal nerve endings (Ulrich et al. 1972; Alarie 1973) and are termed "sensory irritation". For the closely related alcohol n-propanol, Kristiansen et al. (1986) found that the sensory irritation response was very short lasting due to desensitization of the receptor. On the other hand, a long-lasting response from the lower respiratory tract was found. This re-

Animals. Ssc: CF-I male mice, mean weight 25.2 g and S.D. 2.7 g, supplied by Statens Seruminstitut, Denmark, were used. A group of four mice was used for each experiment. For exposure of mice via tracheal cannula, a cannula was inserted under pentobarbital anesthesia and the mice were allowed to recover prior to exposure (Nielsen and Alarie 1982; Nielsen and Bakbo 1985b). Butanol dissolved in 0.9% NaCl was used for intraperitoneal (i.p.) injections. The injected volume varied between 0.1 and 0.4 ml.

Offprint requests to: G. D. Nielsen

Chemical. n-Butanol (99.5% pure) from Merck, Darmstadt in Germany. Generation of test atmospheres. Butanol was evaporated, diluted with room air and led to a 2.3 1 exposure chamber (ASTM 1984). The airflow rate in the chamber was between 20 and 40 l/min. The nominal concentration was

230 calculated from the delivery rate of evaporated chemical and the airflow in the exposure chamber. The chamber concentrations, range 435-9250ppm, were monitored continuously by infrared spectroscopy (Nielsen and Alarie 1982; Nielsen and Bakbo 1985b). The differences between the nominal and monitored exposure concentration were normally less than 10%. Air concentrations are given in ppm, ml gas per m a gas-air-mixture (for calculations see Kristiansen et al. 1987).

Exposure conditions and measurement of the response. Briefly, each animal was placed in a body plethysmograph attached to the exposure chamber so that the head of the animal protruded into the chamber. The respiratory rate and the relative tidal volume (taken as % of the pre-exposure volume) of each animal were obtained by attaching a pressure transducer to each plethysmograph. The absolute tidal volume was obtained only for a group of control animals. The calibrations were carried out by injecting known amounts of air in the plethysmographs. The absolute tidal volume for animals exposed to butanol was estimated from the absolute tidal volume of the control animals and the ratio between the volume in the exposure period and the pre-exposure volume. The respiratory minute volume in exposed animals was obtained as the product of the actual respiratory frequency and the estimated tidal volume. The airborne exposure was for 30 min, followed by a 20-min recovery period during which the respiratory pattern of each animal, the respiratory rate and the relative tidal volume of each group were monitored continuously. Details are given in ASTM (1984) and Nielsen and Bakbo (1985b). In the i.p.-injected group, the pre-exposure respiratory rate as well as the relative tidal volume were obtained prior to the injection. Then the animals were given the injections and the parameters were followed for 50 min.

Evaluation of sensory irritation, pulmonary irritation, and anesthesia. Sensory irritation of the upper respiratory tract causes a characteristic pause before exhalation and thereby a decrease in respiratory rate. This bradypnoea occurs reflexively from stimulation of the trigeminal nerve endings in the nasal mucosa and causes a characteristic respiratory pattern (Alarie 1981b). Frequencies were recorded as mean values of l-min periods. The level of response for each exposure was taken as the percentage change in the average respiratory rate of each group of mice from their pre-exposure level (control level). If exposures are carried out in animals via tracheal cannula, the bypass of the nasal trigeminal nerve endings excludes the development of sensory irritation of the upper respiratory tract. Stimulation of pulmonary receptors by airborne irritants in mice results in a reflexive decrease in respiratory rate with a pattern different from that caused by trigeminal stimulation. The decrease in respiratory rate is due to a pause between the end of expiration and the beginning of the following inspiration resulting in a net decrease in respiratory rate (Alarie 1981 b). This pattern of response occurs not only with pulmon/~ry irritants but also with asphyxiants such as CO, HCN, and general anesthetics, due to depression of the central nervous system (Alarie 1981b; Nielsen et al. 1985b). By observation of escape attempts of the mice in every experiment, it was determined if anesthesia was present.

Estimation of the absorbed dose. The absorption fraction (i.e. the ratio between the absorbed dose and the inhaled dose) was determined at a concentration about 7600 ppm. Four mice were inserted in a whole glass desiccator which was connected to the infrared analyzer via a closed recirculating system of Teflon tubing and a metal bellows pump with an injection port. The entire volume of the system was 12 1. Butanol was injected into the system and evaporated. After equilibration the absorbed dose was calculated from the disappearance rate in the period 10-20 min. The test was repeated with the same group of mice on 3 different days. Afterwards the mice were killed and the adsorption to external surfaces was determined. The absorption fraction was corrected for disappearance from the test system and for adsorption to the external surface of the animals. The inhaled dose was calculated from the exposure concentration and the expected ventilation, which was obtained from linear regression between the log concentration and the ventilation in the concentration range 2500-9200 ppm for normal mice. The absorption fraction was assumed to be constant and estimates of the absorbed doses at all other concentrations were calculated from the product of the inhaled dose and the absorption fraction. Concentration-response relationships and statistical procedures. The concentration-response relationships were investigated by least-square linear regression analyses. The 95% confidence intervals (given in brackets) and the statistical tests were calculated according to Hald (1960) and Colquhoun (1971). The dynamic constants were derived by a direct computer-fit (Ross 1980) of the data to the Michaelis-Menten model. Unit of the maximum response (Emax) is "% decrease from control" and unit of the apparent dissociation constant (KD) is " p p m ' . P values less than 0.05 were regarded as significant. Results

Respiratory rate, tidal volume and respiratory minute volume of control animals At first it was investigated if the respiratory rate, the tidal volume and the respiratory minute volume varied with the body weight, range 20-30 g, of the mice. The regressions were not significant and thus the mean respiratory rate (261 _+39 resp./min), the mean tidal volume (0.301 +_0.080 ml), and the mean respiratory minute volume (78 +_20 ml) were applied for representing the parameters in the mentioned weight range. The plus/minus values are the standard deviations. The mean values were obtained from 16 mice equally distributed on the four weight ranges: 20.2-21.6, 22.6-24.3, 26.6-27.0 and 28.5-30.9 g, respectively.

Time-response relationships, concentration-response tionships and dynamic constants

rela-

The characteristic sensory irritation patterns appeared immediately after onset of the exposure. The patterns were most conspicuous within the 1st min and were then followed by a fast fading of the responses. The patterns were, however, occasionally seen during the entire exposure period. In normal (non-cannulated) mice the sensory irritation response gave rise to a maximum decrease in the re-

231

a. ._1 0n-

10o

~"

\

F

....0 PjPM..,._.../- - -

\

4

/"

J

L

I

" ~ t,.. / ~ ....,

.

}Z 0

ro LL 0 Z~ I

li , % ~ . ? " - -

80

if,,,//

'~-

t't/I,,

.... . . 2 5 ~ . e ~ M

\_

i,.,,,l"/,>~."~..~:.~.'

~

.... ----.

" " ~ ,.

,, \--

\,-"""--,"

~

_,." ,'"

60

LU i-

P7

~7~o--'~V~--" -v.1--

n'* 40 nO I< n* 0. r uJ 13:

I

20

normal I

0

0

,

I

10

,

20

I

,

30

I E X P O S U R E ON

i

b.

I

,

40

50

EXPOSURE OFF

T I M E IN M I N U T E S

...i

o~C

100

I-Z

o

0 14. 0 I

Ill I-< CC >nO I-<: Q. O9 UJ

80

:i

5400 PP

-

~

~

- 17" '~"~'-'~"~"~r

li

60

/i'

i'""/'"~" PP"

!I

40

I

\

,

,250 P.~._-..

ix _ii

'-"

I

_..-

20

0

~

cannulated

,

l 10

0

,

I 20

,

~EXPOSURE ON

l 30

L

I 40

50

'I'EXPOSURE OFF T I M E IN M I N U T E S

C. -J 0

rt' I-z

1.6 mg/mou e 100

o

0 LL 0 I

Fig. 1. A The time-response relationship for butanol inhalation in normal (noncannulated) mice. B The time-response relationship for cannulated mice exposed by inhalation. In C is given the time-response relationship for i.p.-injected mice, dose in mg per mouse. The lag time is due to the time needed for injecting and inserting the animals in the plethysmographs. In all the figures each curve represents the average respiratory rate of four mice. For clarity, results are shown for a few exposure levels only

4.1 mg/mouse 8o

0 I.< nQ. 09 UJ n-

,-.-.... ~"...~"--

""

--,/---

-,.12.1m g l m o u ~ _ / ' ~ ' \ 60

20.2 mg~mouse

LU I->_

~

/

"~

/1~.J" ~,

.~.....~.~ ..~

40

i.p. i n j e c t i o n 20

I

O o i

I

10

,

I

,

20

INJECTION T I M E IN M I N U T E S

I

30

,

I

40

50

232 @O-

Inhe]etlon meen 0-1 mln o noPme] 9 cannuleted

Emex: 57,7 :E 12.4 K= :2,208 4- 796

9

Inhalation 11-20

mean

SO-

..~.~

_

100

c

mln

80

0 n

Inhalation l.p. injection

mean

o

normal

9

cannulated

0-30

mln

40-. ~ 20-

Go

80 0~ 5000

100(30

GO

/ o

o /

/

oo / / /

/

c~ o o

o

o

9

20 o

o

o O~o

9

oo

ooo . . . . . . . . . . . . . . . .

10 2

i l

103

. . . . . . . . . . . . . . . .

i L

. . . . . . . . . . . . . . . .

104 log

i l

I0s conc,

0

/

I

40

01

: :::;:::~ : ::'.::::'~ : :;:;;'.:~ . 102 103 104.

(ppm)

log

conc.

(ppm)

~ o I0 ~

1 01 log

absorbed

10 2 does

[mg)

Fig. 2. A The relation between the concentration of butanol vapour and the decrease in the respiratory rate within the 1st min in normal (non-cannulated) and in cannulated mice. The sensory irritation response from the upper respiratory tract is dominant and the insert shows that the response can be described by a Michaelis-Menten equation. The maximum response, the apparent dissociation constant and the SDs are given as Emax, K D and the ___values. B The relation between the concentration of butanol vapour and the mean decrease in the respiratory rate during the period 11-20 rain. At concentrations above 3089 ppm the response is seen mainly to be due to effects below the upper respiratory tract. C The relation between the mean decrease (0-30 min) in the respiratory rate for normal mice and the total dose absorbed by inhalation at the six highest concentrations. The mean decrease (5-30 min) after i.p. injection is also shown spiratory rate (Fig. 1 a) in the 1st min. This decrease was found not to be influenced by effects from the lower respiratory tract or from the C N S at concentrations below 3552 ppm, which is the extrapolated threshold concentration, RD-0 ( 0 - 1 rain), for cannulated mice (Fig. 2a). The relationship between the decrease in the respiratory rate (y) in normal mice and the logarithm o f the exposure concentration (log x) constituted a straight line (y = 29.4 log x - 69.6) for concentrations below 3552 ppm. This relationship gives the extrapolated RD-0 ( 0 - 1 min): 233 (143-380) p p m and the extrapolated concentration expected to depress the respiratory rate by 50%, RD-50 ( 0 - 1 min): 11696 (5510-24825) ppm. The insert in Fig. 2 a shows that the direct relation between the concentration a n d the sensory irritation response follows that o f a bent curve which can be described by a Michaelis-Menten equation. The extrapolated value o f the m a x i m u m response and the dissociation constant are obtained from concentrations below 3552 p p m , a n d were 57.7_+ 12.4 and 2208+796, respectively. Calculations showed that no threshold was needed, when describing the curve by this model. After the 1st rain two different patterns o f the time-response relationship were f o u n d (Fig. 1 a). F o r concentrations below about 3000 p p m , the responses faded to low levels which were a p p r o x i m a t e l y stable in the exposure period from 11 to 30 min. A b o v e this concentration a second decrease in the respiratory rate was seen. This response must be due to an effect b e y o n d the u p p e r respiratory tract, as the same response was found in c a n n u l a t e d animals. The response had stabilized within 10 min after onset o f the exposure (Fig. I a a n d b). The p e r i o d from 11 to 20 min after onset o f the exposure is therefore selected to

represent the effects after the sensory irritation response has faded. Figure 2b shows the relationship between the logarithm o f the exposure concentration a n d the response for n o r m a l and cannulated mice. F o r normal mice it was not possible, due to the distribution o f the data, to fit any particular model to the d a t a for concentrations below 3089 p p m , which is the RD-0 ( 1 1 - 2 0 rain) in cannulated mice. This value is an estimate o f the threshold for the effects from the lower respiratory tract as well as from the CNS. However, the control experiments, in which the mice were exposed to air only (these experiments cannot be shown in the logarithmic transformation) indicate that the dose-response relationship follows a function which has a response close to zero at a zero exposure concentration. F o r the cannulated animals a straight line [Response (% decrease from control) = 147 x log conc. (ppm) - 513] could be fitted to the mean decrease o f the respiratory rate (Fig. 2b). F o r the six concentrations above 3089 p p m the responses o f the normal mice followed the equation: Response (% decrease from control) = 73.9 • log conc. (ppm) - 241. The slopes o f the two lines were not found statistically different, although the p-value was close to 0.05. In the limited concentration range investigated, the responses o f n o r m a l and o f cannulated mice follow the same general trend (Fig. 2b). This suggests that at high concentrations the same mechanism is responsible for the decrease in the respiratory rate, whether the animals respire n o r m a l l y or through a tracheal cannula.

Differentiation between effects on the lungs and effects on the C N S Conspicuous effects on the C N S such as anaesthesia and asphyxia are definitely excluded, as the mice showed es-

233

cape behaviour (body movements) at all concentrations. The differentiation between the two mechanisms was further investigated by comparing the decrease in respiratory rate after i.p. injection and after absorption of inhaled doses (Fig. 2 c). For the inhalation curve the decrease in the respiratory rate is taken as the mean during the entire 30-min exposure period. For the i.p. injections it took about 5 min to inject and to insert the animals in the plethysmographs (Fig. lc) and thus the responses are recorded in the period 5 - 3 0 min of the 50-min periods. A rapid and complete absorption after i.p. injection has been assumed, as butanol belongs to the low molecular weight alcohols. Further, it has not been possible to correct the curves for metabolism. In the inhalation experiments the absorbed dose is calculated as the product of the absorption fraction and the inhaled dose. At an exposure level of about 7600 ppm the mean absorption fraction in mice was found to be 0.35 +0.12 which is close to the fraction, range: 0.36-0.48, found in humans at a 100-200 ppm exposure level (Astrand et al. 1976). At the 50 ppm level the absorption fraction in dogs was about 0.55 (DiVincenzo and Hamilton 1979). Due to the agreement with the data in the literature, the mean value has been used at all concentrations. The relation between the estimates of the total absorbed dose (0-30 min) and the respiratory frequency is given in Fig. 2c. As the responses after inhalation are more conspicuous than those after i.p. injection, they cannot be explained by an effect exclusively from the CNS and thus it would seem that lung receptors are involved in the response. This is also substantiated from Fig. 1 c, where it is seen that no abrupt change in the respiratory rate occurs for i.p.-injected animals, contrary to what is seen when ceasing the airborne exposure for cannulated animals (Fig. I b).

Different types of lung receptors are involved Figure 3b shows that i.p.-injected butanol depresses the tidal volume. The same tendency is seen in normal mice (Fig. 3a). In cannulated mice, however, high concentrations give rise to an increase in the tidal volume (Fig. 3 a). The frequency response in normal mice and in cannulated mice (Fig. 2b) was very close to equal, which suggests that this response is mediated via the same type of lung recepa

meen 120

tors. On the other hand the respiratory frequency and the tidal volume did not show any coherence, and thus these responses are most likely mediated by different lung receptors. The frequency response may to a minor extent be influenced by the volume response. This may explain the greater response of cannulated mice compared to normal mice (Fig. 1) as well as the tendency to a steeper curve for cannulated mice (Fig. 2 b).

Discussion

Sensory irritation The butanol-induced response in mice has previously been investigated by Kane et al. (1980) and by de Ceaurriz et al. (1981). The RD-50 (0-10 rain) was reported to be 4784 ppm and 1268 ppm, respectively. In this investigation the RD-50 (0-1 min) is found by extrapolation to be 11696 ppm, a value which is not influenced by effects from the lower respiratory tract or the CNS. Butanol inhalation in rats also affected the trigeminal nerve. The threshold for the response found by these electrophysiological investigations varied between 164 and 499 ppm (Silver and Moulton 1982). The thresholds in rats are of the same size as the threshold, 233 ppm, found in this investigation. The butanol-induced activation of the sensory irritant receptor followed the dynamics applying to a reversible bimolecular reaction if presupposing that the effect is directly proportional to the receptor occupancy (O'Flaherty 1981). This has also been found for n-propanol (Kristiansen et al. 1986) and for acrolein and formaldehyde (Kane and Alarie 1978). In our investigation the dynamic constants (Ema x and KD) were determined too, as such values can be used for calculating the irritating effect of mixtures (Kane and Alarie 1978). After the initial response, a conspicuous desensitization takes place. The same has been found for n-propanol (Kristiansen et al. 1986), where the desensitization was suggested to be due to an effect at the receptor. The same is expected for butanol as the short period (about 2 - 3 min), before the response has faded, did not allow any considerable absorption with a resulting systemic effect. 0-30

min

o

normal

9

cannul ated o o~ o OOOOOO

o

0

80

o

100

9

0 o

o

0

80

o

o

t 0 E

0-30 mln injection

o oo

q-

120

-.

100 C 0 0

mean i.p.

b 9

oo

60

60 0 0

Fig. 3. A The relation between the concentration of butanol vapour and the relative tidal volume. B The relation between the i.p.-injected dose and the relative tidal volume. As the responses by i.p. injection have not been obtained for the first 5 min, the mean during the period 0 - 3 0 min has been estimated from the last 25 rain of the period

"~ 4o >

40

% u

20

20

I---

102

:

: : :::::', 103 ]og

:

: : :,,,,I 104

conc.

(ppm)

. . . . . . . .

0 10 ~

,

. . . . . . . .

101 lo 9

i

10 z dose

(mg)

234

Evaluation of pulmonary irritation and CNS effect Effects beyond the upper respiratory tract seen for concentrations above 3089 ppm may be due to effects on the lungs or effects on the CNS. Body movements were observed at all concentrations, which indicate that no conspicuous CNS depression is involved. Further comparison between butanol absorbed by inhalation and i.p.-injected butanol (Fig. 2c) shows that inhalation gives rise to a more powerful effect. Effects from the lung receptors are therefore involved. This is further substantiated by the differences in the time-response relationships, where the responses in cannulated animals (Fig. 1 b) were maximum shortly after the onset of the exposure, were stable during the exposure period and disappeared quickly after cessation of the exposure. A response due to absorption would be expected to develop gradually as the absorption proceeds. The disappearance of the response would also be expected to take place slowly as seen from the responses of injected butanol (Fig. 1 c). The decrease in respiratory rate after i.p. injection of butanol is in all probability mediated directly via the CNS and not via the lung receptors, as the concentration in the exhaled air is far too low to activate the lung receptors. Thus, the butanol concentration in the aqueous compartment can be calculated according to McCreery and Hunt (1978). The Henry's law constant can be estimated from the saturated concentration of butanol in water at 37~ (Sch~ifer and Lax 1962), the vapour pressure (37~ and 760 mm Hg) of pure butanol (Nielsen and Bakbo 1985a) and the relative partial pressure of the saturated concentration of butanol in water (Moelwyn-Hughes 1964). The alveolar concentration of butanol can then be estimated from Henry's law. For the highest i.p.-injected dose, 20.2 mg/mouse equal to 798 mg/kg, the alveolar concentration is calculated to be about 130 ppm. This is in agreement with the finding of DiVincenzo and Hamilton (1979) who found no exhalation of butanol after intravenous injection in dogs. However a CNS effect has been demonstrated in rats as ataxia was observed after 400 mg/kg i.p. (McCreery and Hunt 1978), and a decreased performance was found in a tilted plane test after 1210 m g / k g per os (Wallgren 1960).

Evaluation of the influence of metabolism Comparison between an inhaled dose and an i.p.-injected dose has as a prerequisite that the "first-pass effect" is negligible. For rats it has been shown that butanol is metabolized via alcohol dehydrogenase (ADH) and that the elimination follows Michaelis-Menten kinetics with a Michaelis constant of 0.86mM and a maximum velocity of 0.077 mmol/min (Auty and Branch 1976). Accepting that mice and rats have equal concentrations of ADH in the liver and correcting for difference in body weights, the maximum velocity in mice is expected to be 0.65 mg/min x mouse. Following i.p. injection, a stable level was reached within 5 min, except for the highest concentration (Fig. 1 c), suggesting that the "first-pass metabolism" has finished. Thus the maximum amount metabolized due to the "first-pass effect" would be about 3.3 mg. This would give rise only to a slight displacement leftwards of the curve in Fig. 2c. Furthermore, for the inhalation-curve (not corrected for metabolism) the ventilation for the whole 30 min-period has been selected for calculating the

absorbed dose, which then becomes an over-estimate, as the effect on the lungs is seen already after 5 rain exposure (Fig. 1 b). Thus, ignoring metabolism is not expected to affect the conclusions, although the curves would be slightly displaced.

Location of the pulmonary receptors The effects on the respiratory rate are not seriously affected by excluding the scrubbing effect of the nose (Fig. 2 b); the mucous layer absorbs butanol from the inhaled air and liberates the butanol to the exhaled air (see Astrand et al. 1976). This may suggest that the surface area of the upper respiratory tract is small compared to the surface area between the upper respiratory tract and the receptors, i.e. the receptors are probably situated deep in the lungs. On the other hand, excluding the nose gives rise to opposite effects on the tidal volume in normal and in cannulated mice (Fig. 3 a). The same has been found for n-propanol (Kristiansen et al. 1986), where the effect on the respiratory rate was explained by an effect on the J-receptors in the alveolar wall and the effect on the tidal volume was explained by a block of the pulmonary stretch receptors. The stretch receptors are concentrated in the trachea and the larger airways (Widdicombe 1982). Thus, excluding the scrubbing effect of the nose can be expected to increase the concentration of butanol at these receptors which, when paralysed (cessation of the Hering-Breuer reflex), gives rise to an increase of the tidal volume.

Relation to threshold limit value and workplace experiences Multiplication of the RD-50 due to sensory irritation in mice with the factor 0.03 has been shown to give a value corresponding to a low degree of sensory irritation in humans and to be the maximum which should be accepted in industrial situations (Alarie 1981 a). The factor can be applied for full agonists. For partial agonists, which have a low slope of the concentration-response curve, a better approach may be to use the threshold response for sensory irritation (RD-0) and multiply by 0.2 (Nielsen et al. 1985 a), which for n-butanol gives 47 ppm. The result can be compared with the Danish threshold limit value, year 1986, which is 50 ppm and equivalent to the A C G I H value. The sensory irritation findings in humans are summarized in Table 1. At short exposure periods conspicuous effects are seen at low concentrations, whereas low effects may be seen at high concentrations. This is most likely explained by desensitization, which is observed in this investigation. Pulmonary irritation due to activation of the J-receptors is responsible for a sensation of dyspnea (Paintal 1973; Widdicombe 1982), a symptom which has frequently been reported from workplaces (e.g. Hedenstierna et al. 1983; Kilburn et al. 1985; Levy et al. 1985). The activation of lung receptors as a result of occupational exposure has only been investigated to a very limited extent. Thus the relationship between the effect in man (e.g. the TLV value) and the pulmonary irritation response in mice has not been firmly established. However, Nielsen et al. (1985b) multiplied the threshold response for pulmonary irritation in mice by 1/15 to obtain an estimate of pulmonary irritation in humans. This suggests that pulmonary irritation of butanol might be expected above 200 ppm. Sensory irrita-

235 Table 1. Sensory irritation of butanol in humans Concentration (ppm)

Duration of exposure

Exposure conditions

Effect

Reference

5-14

Daily

Working environment

None

Tabershaw et al. (1944)

Less than 25

3 - 5 min

Exposure chamber

Highest concentration which the majority of subjects estimated satisfactory for 8-h exposure

Nelson et al. (1943)

25

3 - 5 min

Exposure chamber

Irritation of nose and throat in the majority of the subjects

Nelson et al. (1943)

20 - 65

Daily

Working environment

Eye inflammation by 5 of 30 workers

Tabershaw et al. (1944)

50

3 - 5 min

Exposure chamber

Irritation of eyes in the majority of the subjects

Nelson et al. (1943)

About 100

Considerable time

Working environment

Members of the industrial hygiene environmental department did not find the conditions uncomfortable after a short acclimatization. The members feel 100 ppm as a reasonable working condition

Sterner et al. (1949)

About 100

Years

Working environment

Complaints of irritation or disagreeable odour have been rare

Sterner et al. (1949)

100 and 200

2h

Exposure chamber

None of the subjects were troubled by the exposure, neither at rest nor during exercise

]kstrand et al. (1976)

About 200

Years

Working environment

Transient corneal inflammation with associated burning sensation, lacrimation, and photophobia was encountered in a few workmen

Sterner et a1.(1949)

tion thus is believed to be the most important of the two effects, which is in agreement with reports o f sensory irritation in humans (Table 1). Butanol has a conspicuous, but fast-fading effect on the trigeminal nerve and at higher concentrations also an effect on the vagus nerve. Therefore these effects should always be considered when using butanol as a standard for o d o u r intensity measurements.

Acknowledgements. We thank Assistant Professor E. Sonnich Thomsen, The Royal Danish School of Pharmacy, for valuable discussions. The investigation was supported by a grant from the Danish Rockwool Fund. References Alarie Y (1973) Sensory irritation by airborne chemicals. CRC Crit Rev Toxicol 2:299-363 Alarie Y (1981 a) Dose-response analysis in animal studies: prediction of human responses. Environ Health Perspect 42: 9-13 Alarie Y (1981b) Toxicological evaluation of airborne chemical irritants and allergens using respiratory reflex reactions. In: Leong KJ (ed) Proceedings of the inhalation toxicology and technology symposium. Ann Arbor Science, Collingwood, pp 207- 231 ASTM (1984) Standard test method for estimating sensory irritancy of airborne chemicals. Designation: E 981-84. American Society for Testing and Materials, Philadelphia, pp 1- 16 Auty RM, Branch RA (1976) The elimination of ethyl, n-propyl, n-butyl and iso-amyl alcohols by the isolated perfused rat liver. J Pharmacol Exp Ther 197:669-674 Cain WS (1974) Contribution of the trigeminal nerve to perceived odor magnitude. Ann NY Acad Sci 237:28-34

Cain WS (1976) Olfaction and the common chemical sense: Some psychophysical contrasts. Sens Processes 1:57-67 Colquhoun D (1971) Lectures in biostatistics. Clarendon Press, Oxford De Ceaurriz JC, Micillino JC, Bonnet P, Guenier JP(1981) Sensory irritation caused by various industrial airborne chemicals. Toxicol Lett 9:137-143 DiVincenzo GD, Hamilton ML (1979) Fate of n-butanol in rats after oral administration and its uptake by dogs after inhalation or skin application. Toxicol Appl Pharmacol 48: 317-325 Dravnieks A (1977) Correlation of odor intensities and vapor pressures with structural properties of odorants. In: American Chemical Society Symp Ser, Vol 57, ISS Flavor Quality: Objective Measurement, Symp, pp 11-28 Hald A (1960) Statistical theory with engineering applications. John Wiley & Sons, New York Hedenstierna G, Alexandersson R, Wimander K, Ros+n G (1983) Exposure to terpenes: effects on pulmonary function. Int Arch Occup Environ Health 51:191-198 Kane LE, Alarie Y (1978) Evaluation of sensory irritation from acrolein-formaldehyde mixtures. Am Ind Hyg Assoc J 39: 270-274 Kane LE, Dombroske R, Alarie Y (1980) Evaluation of sensory irritation from some common industrial solvents. Am Ind Hyg Assoc J 41 : 451-455 Kilburn KH, Seidman BC, Warshaw R (1985) Neurobehavioral and respiratory symptoms of formaldehyde and xylene exposure in histology technicians. Arch Environ Health 40: 229-233 Kristiansen U, Hansen L, Nielsen GD (1986) Sensory irritation and pulmonary irritation of cumene and n-propanol: mechanisms of receptor activation and desensitization. Acta Pharmacol Toxicol 59:60-72 Kristiansen U, Hansen L, Nielsen GD (1987) Determination of concentrations in air of two-component mixtures by single-

236 beam infrared spectrophotometry. Arch Pharm Chem Sci Ed 15:16-23 Levy BS, Davis F, Johnson B (1985) Respiratory symptoms among glass bottle makers exposed to stannic chloride solution and other potentially hazardous substances. J Occup Med 27:277-282 McCreery MJ, Hunt WA (1978) Physico-chemical correlates of alcohol intoxication. Neuropharmacology 17:451-461 Moelwyn-Hughes EA (1964) Physical chemistry. MacMillan, New York, p 781 Nelson KW, Ege JF, Ross M, Woodman LE, Silverman L (1943) Sensory response to certain industrial solvent vapors. J Ind Hyg Toxicol 25:282-285 Nielsen GD, Alarie Y (1982) Sensory irritation, pulmonary irritation, and respiratory stimulation by airborne benzene and alkylbenzenes: prediction of safe industrial exposure levels and correlation with their thermodynamic properties. Toxicol Appl Pharmacol 65:459-477 Nielsen GD, Bakbo JC (1985a) Exposure limits for irritants. In: International Symposium on Occupational Exposure Limits. American Conference of Governmental Industrial Hygienists, Cincinnati. Ann Am Conf Ind Hyg 12:119-133 Nielsen GD, Bakbo JC (1985b) Sensory irritating effects of allyl halides and a role for hydrogen bonding as a likely feature at the receptor site. Acta Pharmacol Toxicol 57: 106-116 Nielsen GD, Olsen J, Bakbo JC, Hoist E (1985a) Propyl ether. I. Interaction with the sensory irritant receptor. Acta Pharmacol Toxicol 56: 158-164 Nielsen GD, Olsen J, Bakbo JC, Hoist E (1985b) Propyl ether. II. Pulmonary irritation and anaestesia. Acta Pharmacol Toxicoi 56:165-175 O'Flaherty EJ (1981) Toxicants and drugs: kinetics and dynamics. John Wiley, New York

Paintal AS (1973) Vagal sensory receptors and their reflex effects. Physiol Rev 53:159-227 Ross GJS (1980) MLP. Maximum likelihood programs. Rothamsted Experimental Station, Harpenden Schfifer K, Lax E (1962) Eigenschaften der Materie in ihren Aggregatzust/inden. 2. Teil. Bandteil b. L6sungsgleichgewichten I. In: Landolt-B6rnstein. Zahlenwerte und Funktionen aus Physik-Chemie-Astronomie-Geophysik und Technik. Springer-Verlag, Berlin, 3-406-3-407 Silver WL, Moulton DG (1982) Chemosensitivity of rat nasal trigeminal receptors. Physiol Behav 28:927-931 Sterner JH, Crouch HC, Brockmyre HF, Cusack M (1949) A tenyear study of butyl alcohol exposure. Am Ind Hyg Assoc Q 10:53-59 Tabershaw IR, Fahy JP, Skinner JB (1944) Industrial exposure to butanol. J Ind Hyg Toxicol 26:328-330 Ulrich CE, Haddock MP, Alarie Y (1972) Airborne chemical irritants. Arch Environ Health 24:37-42 Wallgren H (1960) Relative intoxicating effects on rats of ethyl, propyl and butyl alcohols. Acta Pharmacol Toxicol 16: 217-222 Widdicombe JG (1982) Pulmonary and respiratory tract receptors. J Exp Biol 100:41-57 Astrand I, Ovrum P, Lindqvist T, Hultengren M (1976) Exposure to butyl alcohol. Uptake and distribution in man. Scand J Work Environ Health 3: 165-175

Received March 16, 1987/Accepted September 9, 1987