Pulmonary and sensory irritation of diphenylmethane-4,4′- and dicyclohexylmethane-4,4′-diisocyanate

TOXICOLOGY

AND.4PPLIEDPHARMACOLOGY

77,427-433

(1985)

Pulmonary and Sensory Irritation of Diphenylmethane-4,4’and Dicyclohexylmethane-4,4’-diisocyanate DIETRICH A. WEYEL AND RUSSELL B. SCHAFPPR Department

of Industrial Environmental Health Sciences, Graduate Health, University of Pittsburgh, Pittsburgh, Pennsylvania

Received

August

6, 1984; accepted

October

School 15261

of Public

8, 1984

Pulmonary and Sensory Irritation of Diphenylmethane-4,4’- and Dicyclohexylmethane-44,4’diisocyanate. WEYEL, D. A., AND SCHAFFER, R. B. (1985). Toxicol. Appl. Pharmacol. 77,427433. The use of isocyanates in industry has been increasing and, therefore. the potential for human exposure has also increased. Two such isocyanates are diphenylmethane+l’diisocyanate (MDI) and dicyclohexylmethane-4,4’diisocyanate (SMDI). Furthermore, there are only a few reports describing the toxicity of these diisocyanates. The pulmonary irritation of the aromatic isocyanate MD1 and the sensory and pulmonary irritation of the cycloaliphatic isocyanate SMDI were studied in an animal bioassay. Groups of male, Swiss-Webster mice were exposed to aerosol concentrations of MD1 varying from 17 to 67 mg/m3. The total exposure time for both isocyanates was 240 min, and the respiratory patterns and frequency of four mice were recorded during each exposure. Concentrations of MD1 and SMDI in the exposure chamber were determined gravimetrically. The mass median aerodynamic diameter (MMAD) and geometric standard deviation for the MD1 aerosol were 0.7 pm and 1.6 and for the SMDI aerosol were 0.9 pm and 1.5, respectively. The inhalation responses during the 4-hr exposures to aerosols of MD1 and SMDI were investigated, and the animal model was used to determine time-response and concentration-response relationships for all exposures. From these results it was determined that the level of effect was dependent on both the duration of exposure and the exposure concentration. Unlike many other isocyanates tested with this animal model, MD1 and SMDI acted primarily as pulmonary irritants, evoking little or no sensory irritation. The concentrations required to reduce the respiratory rate 50% (RD50) due to pulmonary irritation was 32 mg/m3 for MD1 and 40 mg/m3 for SMDI. Increases in lung weight were found in groups of animals killed 24 hr following all exposures to MD1 and SMDI. Using the animal model, which has been calibrated to human responses with nitrogen dioxide and other pulmonary irritants, the recommended TLV-TWAs for MD1 and SMDI in industry should be no higher than 0.3 and 0.4 mg/m’, respectively. 0 1985 Academic press, I~C.

The last decade has seen a tremendous increase in the use of many polyurethane materials. Due to their versatility this trend will undoubtedly continue. A major raw material used in the production of all polyurethanes is a group of compounds commonly known as the isocyanates, and these chemicals are characterized by reactive NC0 groups. Because of the demand for polyurethane products and the consequent need for the various types of isocyanates, it was projected that, by

1980, over 100,000 workers would have been exposed to isocyanates during some phase of their production (NIOSH, 1978). Additionally, many more people would be exposed to isocyanates during manufacturing processes of polyurethane products. The number of isocyanates which differ chemically is extensive, but only a few types comprise the isocyanates most commonly produced and manufactured. These include toluene diisocyanate (TDI), hexamethylene 427

0041-008X/85

$3.00

Copyright 0 1985 by Academic Press. Inc. All rights of reproduction in any form reserved.

428

WEYEL AND SCHAFFER

diisocyanate (HDI), diphenylmethane-4,4’-diisocyanate (MDI), and dicyclohexylmethane4p’diisocyanate (saturated MD1 or SMDI). MDI, whose structure appears in Fig. 1, is an aromatic, monomeric, difunctional isocyanate. MD1 is of major importance to a variety of industrial applications, including the manufacture of rigid polyurethane and polyisocyanurate foams, polyurethane plastics, elastomers, and surface coatings. SMDI is a liquid, cycloaliphatic, difunctional isocyanate. The structure of SMDI is also shown in Fig. 1. SMDI is used in the manufacture of high-performance specialty polymers such as light-fast and color-stable urethane coatings, elastomers, and foams. Principal applications are found in floor, textile, and bottle coatings, and urethane water dispersions (Mobay Chemical Corporation, 198 1). One method to analyze the toxicological effect of a chemical is through an inhalation study. From the earlier works of Alarie ( 1966) to more recent studies (Alarie, 1973; Sangha and Alarie, 1979; Weyel et al., 1982) it has been demonstrated that the inhalation of irritating chemicals will stimulate nerve endings in the respiratory tract, resulting in sensory irritation, and/or pulmonary irritation with associated typical breathing patterns. Sensory irritation is characterized by a pause during expiration, whereas pulmonary irritation shows a pause at the end of expiration and the following inspiration (Alarie, 198 1a). The animal model used in these studies

0;3;M+H2~N=C=0 MDI DIPHENYLMETHANE-4,4’-DIISOCYANATE

O=D=Mo

CH,~N=C=O

SMDI DICYCLOHEXYLMETHANE-4.4’.DIISCCYANATE

FIG. 1. Chemical structure of diphenylmethane-4,4’diisocyanate (MDI) and dicyclohexylmethane-4,4’diisocyanate.

has been shown to be an excellent predictor of airborne concentrations of many chemicals, including isocyanates, which will not provoke eye, nose, throat, or pulmonary irritation in exposed workers (Alarie, 198 1b). This report will summarize the findings of the effects of inhalation of MD1 and SMDI, using the same animal models. METHODS Animak. The mice used in this study were obtained from Hilltop Laboratories, (Scottdale. Pa.). They were outbred, specific-pathogen free, male, Swiss-Webster mice, weighing between 24 and 29 g. New groups of four mice each were used for each exposure. Isocyanates. The MD1 was obtained from Mobay Chemical Corporation, which markets this chemical under the tradename Mondur M. The chemical is a high purity grade (99.5% minimum) which appears as an odorless, fused solid of white to light yellow flakes. The molecular weight of the material is 250.26 with a specific gravity of 1.I9 g/cm’ at 25°C. It is soluble in acetone and, like most isocyanates, it is reactive with water (Mobay Chemical Corporation, 1979). The saturated MD1 (SMDI) was also obtained from Mobay Chemical Corporation and its tradename is Desmodur W. The chemical is high purity grade (99.3% minimum) and appears as a clear, colorless liquid which is odorless at or below the TLV. The molecular weight of the material is 262.35 with a specific gravity of 1.07 g/cm3 at 25°C. It is also soluble in acetone and reactive with water (Mobay Chemical Corporation. 198 1). Exposure chamber conditions. The exposure chamber used was composed of glass. has an internal volume of about 2.1 liters, and has been previously described in detail (Barrow et al.. 1977). There are four animal body plethysmographs on the chamber with accompanying pressure transducer connectors and sampling ports. The bodies of each of four mice were secured in the plethysmographs with the heads protruding through a rubber latex dam which was taped to the inside of the chamber. A rubber stopper was used to seal the end of the plethysmograph and also prevented the mice from escaping. A constant flow of 20 liters/min air, monitored by a flowmeter. was maintained through the chamber for all exposures. Measurement of response. The average respiratory rate of each group of four exposed mice was monitored before, during, and after each run as previously described (Kane and Alarie, 1977). Exposure time was, in all cases, 4 hr preceded by a period of 15 min of no exposure which served as control baseline. Aerosol generation and measurement. Weighed amounts of MD1 and SMDI. both of which have very

PULMONARY

IRRITATION

low vapor pressures (I X 10m4and I X lo-’ mm Hg at 25°C) were dissolved in reagent-grade, water-free acetone to make final w/v solutions ranging from 0.1 to 5.0% for MD1 and from 0.25 to 1.5% for SMDI. These solutions were used to generate the various concentrations of airborne isocyanate aerosols. The aerosols were generated by delivering the isocyanate/acetone solutions at a constant flow of 0.22 ml/min through a 30-ml syringe with a no. 20-gauge needle mounted on a syringe pump. The syringe pump was connected to a Pitt no. 1 glass aerosol generator via a section of PE-90 tubing from the syringe needle. The Pitt no. I generator has been previously described (Wang and Alarie, 1982). The chamber concentrations of MD1 and SMDI were determined gravimetrically by drawing known amounts of the chamber atmosphere at 2 liters/min through preweighed Teflon membrane filters (0.4 pm pore size; Millipore). These samples were taken at 30, 90, 150, and 2 10 min during each exposure. The particle size distributions of both the MD1 and SMDI aerosols were determined with an Andersen impactor. The MMAD for the MD1 aerosol was 0.7 pm with a geometric standard deviation of 1.6. The MMAD and geometric standard deviation for the SMDI aerosol were 0.9 pm and 1.5, respectively. Acetone concentrations in the chamber were determined with detector tubes (National Draeger CH 22901). The acetone concentration was 2800 to 3000 ppm. Similar results were previously reported by Weyel et al. (1982). Exposure groups. Series I. In this series of experiments groups of four mice were exposed to six concentrations of MD1 aerosol for 240 min. By plotting the timeresponse relationships obtained from these exposures, the extent of respiratory rate depression at each exposure concentration was determined. In addition, from the concentration-response relationship, the RD50 was obtained according to methods previously used for other isocyanates (Sangha and Alarie, 1979; Sangha et al., 198 1; Weyel et al., 1982). All animals were killed by cervical dislocation 24 hr postexposure. The thoracic cavity was dissected, and the lungs were removed. After separating the heart, trachea, and esophagus from the lungs, the lungs were blotted and the weight was obtained. The dry lung weight was obtained by air-drying the lungs at room temperature for a period of 24 hr and then reweighing. Series II. In this series of experiments, groups of four mice were exposed to five concentrations of SMDI aerosol for 240 min. All the procedures used to evaluate the effects of the SMDI exposures were identical to those used for Series I. Series III. In this series, animals were exposed via tracheal cannulation to MD1 aerosol. The exposure concentration of MD1 aerosol was that expected to result in a 50% decrease in respiration rate (RD50) in normal, control mice. Except for the cannulation, all other experimental conditions were identical to those of Series I.

OF MD1 AND SMDI

429

Series IV. In this series, animals were exposed via tracheal cannulation to SMDI aerosol. The concentration chosen for this exposure was also the concentration expected to elicit the RD50 in normal, control mice. Again, all other experimental conditions remained identical to those of Series I.

RESULTS The time-response relationship observed during inhalation of MD1 aerosol in Series I is shown in Fig. 2. The aerosol concentrations ranged from 7 to 59 mg/m3 for the 4-hr exposures. For the two lowest concentrations (7 and 10 mg/m3, respectively), an increase in respiratory rate above the control was observed for almost 3 hr of the exposure, followed by a gradual decline in respiratory rate during the last hour. In the course of exposure to the highest concentration of MD1 aerosol (59 mg/m3), there was only a slight increase in respiratory rate, of short duration, followed by a fairly rapid decline during the last 3 hr. For the remaining exposure concentrations the respiratory rate was initially elevated above the control for approximately 1 hr before beginning to gradually decline for the last 3 hr of exposure. A plateau in response was reached during the last 30 min of the exposure. Although not shown in Fig. 2, little or no recovery was observed during the 20 min following each exposure. The small initial increase in respiratory rate followed by a decrease in respiration has been a consistent finding in mice being exposed to pulmonary irritants (Alarie, 1981a), and was also observed for MDI. The results of Series II, shown in Fig. 3, depict the decline in respiratory rate due to inhalation of various concentrations of SMDI aerosol ranging from 17 to 67 mg/m3. All exposures were conducted for 4 hr. Results for these exposures differ from those of Series I (MD1 aerosols), in that no increase in the respiratory rate was observed at any time during SMDI inhalation. Again, these relationships were concentration related because the effect (decline in

430

WEYEL AND SCHAFFER

130 120-

0 0

20

40

60

60

100 120 140 TIME IMINUTE)

160

180

200

220

240

FIG. 2. Time-response relationship for various concentrations of MDI. Each point represents the average respiratory rate of four animals.

respiratory rate) increased with increases in the concentration and duration of the test. Although not as clearly discernible as the MD1 exposures, a plateau was reached during the last half-hour of each test. Also, no recovery was seen in the animals during the 20 min following each exposure. Concentration-response relationships for both MD1 and SMDI exposures are shown in Fig. 4. Each point represents the average

01

,.,.,./ 0

maximum decrease in respiratory rate of four mice at the different exposure concentrations. By evaluating the final 30 min of each exposure to MD1 and SMDI in the timeresponse relationships, the average maximum decrease in respiratory rate was determined. The concentration-response curves were then obtained by plotting the maximum percentage decrease in respiratory rate against the logarithm of the concentration. Curves were

,.,,,,,.,,/., 20

40

60

60

100 120 140 TIME (MINUTE)

,., 160

160

200

220

240

FIG. 3. Time-response relationship for various concentrations of SMDI. Each point represents the average of four animals.

PULMONARY

IRRITATION

DES-N

-1

CONCEN;ORAT,ON

(MC/M3)

100

FIG. 4. Concentration-response relationships for MD1 and SMDI. Each point represents the decrease in respiration rate after 4 hr of exposure as a percentage of the preexposure level and is the average of four mice. The concentration-response relationship for DES-N is shown for comparison (Weyel et al., 1982). Curves were fitted by linear least-squares analysis.

431

OF MD1 AND SMDI

hexamethylene diisocyanate (HDI) and toluene diisocyanate (TDI), were found to be potent sensory irritants (Sangha et al., 198 1). Instead, the actions of MD1 and SMDI were similar to those reported by Weyel et al. (1982) for an aliphatic triisocyanate (DESN), showing initially some sensory irritation, changing to a pattern of pulmonary irritation described earlier by Alarie (1981b). To confirm the pulmonary irritation properties of MD1 and SMDI, mice were exposed via tracheal cannula (Series III and IV), which will eliminate sensory irritation and pulmonary irritation can be studied alone. Indeed, a decrease in respiratory frequency was observed, indicating that MD1 and SMDI act as pulmonary irritants. Those results for MD1 and SMDI are also shown in Figs. 2 and 3, respectively. DISCUSSION

fitted by the method of least squares and, from the regression analysis, the RDSO’s were calculated. The relative potency of the chemicals can be compared on the basis of the RD50 values (Kane et al., 1980; Alarie, 198 lb). The RD50 values obtained from Fig. 4 for MD1 and SMDI were 32 and 40 mg/m3, respectively. The results of the lung weights taken during Series I and Series II are shown in Fig. 5. The lung weights increased with exposure concentrations following both MDI and SMDI inhalation. Lung weights were determined 24 hr after exposure, since Weyel et al. (1982) reported a maximum increase after this time period with another isocyanate. The lung weights reached an increase of 42% above normal at a concentration of 59 mg/ m3 MD1 aerosol and 3 1% above normal at a concentration of 67 mg/m3 SMDI aerosol. During the course of all exposures to MD1 and SMDI aerosols, it was noticed that the breathing pattern did not exhibit the characteristic pattern of sensory irritation (Alarie, 1966). This observation was a surprising result, because other diisocyanates, such as

The overall reaction to the inhalation of MD1 and SMDI tested in the animal model

o!I

100 CONCEN’&ATION

(MG/M3)

FIG. 5. Concentration-response relationships obtained for lung weight increases following 4 hr of exposure. Each point represents the average of four mice sacrificed 24 hr after exposure. The increase is expressed as a percentage of lung weights from a control group of eight mice of comparable body weight. The concentrationresponse relationship for DES-N is shown for comparison (Weyel et al., 1982). Curves were fitted by linear leastsquare analysis.

432

WEYEL

AND

was a decline in the respiratory rate. Furthermore, it was determined that the decrease in respiration rate was due mainly to the action of these isocyanates as pulmonary irritants. This was an unexpected result. Except for the trimer of HDI (DES-N) (Weyel et al., 1982) other isocyanates investigated with the same animal model failed to induce a breathing pattern indicative of pulmonary irritation but, rather, resulted in a strong response of sensory irritation. However, the long time required to reach a plateau response for MD1 and SMDI is typical of other isocyanates (Sangha and Alarie, 1979; Weyel et al., 1982). Confirmation of the action of MD1 and SMDI as pulmonary irritants was provided by the results of exposure during tracheal cannulation and the lung weights 24 hr postexposure. If MD1 and SMDI acted solely as sensory irritants, there should have been no response during exposure via tracheal cannulation because sensory irritation is eliminated by bypassing the trigeminal nerve. However, the response which was observed during exposures of cannulated mice was that of pronounced pulmonary irritation, indicated by the characteristic breathing pattern. Thus, stimulation of the lower respiratory tract receptors and not the trigeminal nerve was established. Lung weight increases 24 hr following exposure indicated the pulmonary irritating nature of both isocyanates. Lung weight increases were observed following all exposures and resulted from development of pulmonary edema. The possibility that acetone could have contributed to the response in animal model has been previously investigated by Weyel et al. (1982), who concluded that about 3000 ppm, acetone did not invoke sensory or pulmonary irritation. As is shown in Figs. 4 and 5, MD1 and SMDI are of similar potency when compared on the basis of their RDSOs or lung weight increases. Previous studies by Alarie (198 lb) with several chemicals have shown that the RD50

SCHAFFER

can be successfully used to predict safe concentrations of a chemical for industrial exposures. However, that animal model was based on the RD50 due to sensory irritation alone and therefore, cannot be used directly to evaluate pulmonary irritants. Rather, comparison of RDSOs and established exposure limits for pulmonary irritants can be used to rank MD1 and SMDI. Nitrogen dioxide (NO,) has been evaluated with the same model (Alarie, 198 la), and the RD50 for pulmonary irritation was reported to be 658 mg/m3 (349 ppm). The TLV for NO2 is 6 mg/m3 and protects against pulmonary irritation. Comparing the potency of MD1 and SMDI with that of NO* as a pulmonary irritant, one finds that MD1 and SMDI are about 20 and 16 times more potent, respectively. Thus, suggested exposure limits would be 0.3 mg/m3 for MD1 and 0.4 mg/m3 for SMDI. Such limits should prevent pulmonary irritation. It is interesting to compare the predicted values from this study with the TLVs published by the American Conference of Governmental Industrial Hygienists (ACGIH, 1980) which have been used by industry for many years. The TLV for MD1 is 0.2 mg/m3 and for SMDI 0.11 mg/m3; both are also ceiling limits (ACGIH, 1983). The TLV for MD1 of 0.2 mg/m3 is close to the value of 0.3 mg/m3 predicted from this study. What is interesting is that the TLV for MD1 was not based on extensive toxicological studies but rather by comparison with TDI, whose TLV has been lowered to 0.04 mg/m3 (0.005 ppm) (ACGIH, 1983). Because MD1 has not been evaluated for its potential to induce pulmonary hypersensitivity as has TDI (Karol et al., 1981), no change of the current TLV is proposed. However, SMDI has been evaluated for its sensitization potential (Karol and Magreni, 1982), and was found to be a potent skin sensitizer and exhibited no pulmonary sensitization, findings which were also reported from industrial exposures (Emmett, 1976; Israeli et al., 198 1). The same animal model found TDI to be a potent skin and pulmonary sensitizer.

PULMONARY

IRRITATIC tN OF MD1 AND SMDI

The TLV for SMDI of 0.11 mg/m3 was based only on comparisons with TDI and HDI. The findings of this study and that of Karol and Magreni (1982) suggest that the TLV was set too low, which was a prudent decision at a time when no published data were available. Because a low TLV for airborne concentrations will not prevent the possibility of skin sensitization, it is suggested that the TLV be raised from 0.11 mg/m3 to at least 0.2 mg/m3, which will still prevent pulmonary and sensory irritation. ACKNOWLEDGMENTS We thank National Draeger, Inc. (Pittsburgh, Pa.), for the donation of the acetone detector tubes which were used in this study. We thank Mobay Chemical Corporation (Pittsburgh, Pa.), for supplying us with MD1 and SMDI. Finally, we thank Dr. Yves Alarie and Dr. Meryl Karol, of the University of Pittsburgh, for their advice and suggestions, as well as Ms. Maryanne Stock for her technical assistance.

irritation response in mice to chlorine and hydrogen chloride. Arch. Environ. Health 32, 68-76. EMMETT, E. A. (1976). Allergic contact dermatitis in polyurethane plastic moulders. J. Occup. Med. 113, 571-585.

ISRAELI, R., SMIRNOV, V., AND SCULSKY, M. (1981). Vergiftungserscheinungen bei Dicyclohexyl-Methan-44’-Diisocyanat-Exposition. Int. Arch. Occup. Environ. Health

48, 179-184.

KANE, L. E., AND ALARIE, Y. (1977). Sensory irritation to formaldehyde and acrolein during single and repeated exposures in mice. Amer. Ind. Hyg. Assoc. J. 38, 509-522.

KANE, L. E., DOMBROSKI, M. D., AND ALARIE, Y. (1980). Evaluation of sensory irritation from some common industrial solvents. Amer. Ind. Hyg. Assoc. J. 41,451-455.

KAROL, M. H., HAUTH, B. A., REILLY, E. J., AND MAGRENI, C. M. (1981). Survey of industrial workers for antibodies to toluene diisocyanate. J. Occup. Med. 23, 741-747.

KAROL, M. H., AND MAGRENI, C. M. (1982). Extensive skin sensitization with minimal antibody production in guinea pigs as a result of exposure to dicyclohexylmethane 4’4diisocyanate. Toxicol. Appl. Pharmacol. 65, 291-301.

Mobay Chemical Corporation (1979). Product tion and Material Safety 0226 for Diphenylmethane

REFERENCES ALARIE, Y. (1966). Irritating properties of airborne materials to the upper respiratory tract. Arch. Environ. Health

13, 443-449.

ALARIE, Y. (1973). Sensory irritation by airborne chemicals. CRC Crif. Rev. Toxicol. 2, 299-363. ALARIE, Y. (198 la). Toxicological evaluation of airborne chemical irritants and allergens using respiratory reflex reactions. In Proceedings of the Inhalation Toxicology and Technology Symposium (B. K. J. Leong, ed.), pp. 207-231, Ann Arbor Science, Ann Arbor, Michigan. ALARIE, Y. ( 198 1b). Bioassay for evaluating the potency of airborne sensory irritants and predicting acceptable levels of exposure in man. Food Cosmef. Toxicol. 19, 623-626, American Conference of Governmental Industrial Hygienists (ACGIH). (1980). Documentation of the Threshold Limit Values, 4th ed. Cincinnati, Ohio. American Conference of Governmental Industrial Hygienists (ACGIH). (1983). Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment. Cincinnati, Ohio. BARROW, C. S., ALARIE, Y., WARRICK, J. C., AND STOCK, M. F. (1977). Comparison of the sensory

433

Data Sheet Diisocyanate

lnforma(MSDS) No. (MDI). Pitts-

burgh, Pennsylvania. Mobay Chemical Corporation (198 1). Product tion and Material 1196 for Methylene

Safety Data Sheet Bis-4-cyclohexylisocyanate

Pittsburgh, Pennsylvania. NIOSH (1978). Recommended

Standard

Jnforma(MSDS) No. (SMDI).

for Occupational

Pub. No. 78-215. U.S. Dept. of Health, Education and Welfare, Washington, DC. SANGHA, G. K., AND ALARIE, Y. (1979). Sensory irritation by toluene diisocyanate in single and repeated exposures. Toxicol. Appl. Pharmacol. 50, 533-547. SANGHA, G. K., MATIJAK, M., AND ALARIE, Y. (1981). Comparison of some mono- and diisocyanates as sensory irritants. Toxicol. Appl. Pharmacol. 57, 241Exposure

to Diisocyanates,

246.

WEYEL, D. A., RODNEY, B. S., AND ALARIE, Y. (1982). Sensory irritation, pulmonary irritation and acute lethality of a polymeric isocyanate and sensory irritation of 2.6 toluene diisocyanate. Toxicol. Appt. Pharmacol. 64, 423-430.

WONG, K. L., AND ALARIE, Y. (1982). A method for repeated evaluation of pulmonary performance in unanesthetized, unrestrained guinea pigs and its application to detect effects of sulfuric acid mist inhalation. Toxicol.

Appl. Pharmacol.

63, 72-90.