        |
|
For Presentation at The Air & Waste Management Association's 92nd Annual Meeting & Exhibition, June 20-24, 1999, St. Louis, Missouri New
Hypotheses for MTBE Combustion Products Also by Peter M. Joseph: ABSTRACT A 1997 AWMA paper (97-TA34.02) presented evidence linking explosive increases in asthma and other conditions to the introduction of gasoline oxygenated with methyl tertiary butyl ether (MTBE) in Philadelphia and other places. In the winter of 92-93 the state of Alaska withdrew from the MTBE program because of public health problems; their data suggested that MTBE exhaust could be toxic to some people. Over the last two years the epidemiologic reports are getting stronger. Both New York City and Philadelphia have established an "Asthma Task Force" to study the crisis. I presented new preliminary data to the Transportation Research Board (TRB) meeting in August 1998; the abstract is available at http://www.angelfire.com/va/pkleeman/peterjoseph.html. In this paper I present updated data from the Philadelphia Department of Health that, for the first time, includes neurological effects. Many environmental experts dismiss the idea that MTBE exhaust is causing these epidemics because existing speciation studies do not show any one toxic component to be so large as to be troublesome. In this paper I argue that production of methyl nitrite (MeONO) from MTBE combustion could be the cause of the observed problems. The proposed mechanism is the production of methoxy radicals from pyrolysis of MTBE at low to moderate temperatures, followed by combination with nitric oxide produced in the high temperature combustion region. MeONO has been observed in methanol-diesel combustion. It is shown that MeONO is sufficiently stable at temperatures < 550o K that it should survive pyrolysis in the exhaust pipe. An important property is that MeONO is easily photolyzed by sunlight. This lends strong support to anecdotal reports that many people experience symptoms only at night or on cloudy days (Arch. Env. Health 1995;50:395-396) INTRODUCTION While methyl tertiary butyl ether (MTBE) has been used as an octane enhancer in the USA since 1979, the quantities used were later substantially increased in certain areas where it is used as an oxygenate in concentrations of 11-15% by volume. Ever since the winter oxygenated gasoline (WOG) program was initiated (1989-90) many citizens have complained of uncomfortable symptoms and illnesses that they associate with its use. When the reformulated gasoline (RFG) program was introduced even further in January 1995, again more citizen complaints arose. In some areas, groups of people calling themselves "Oxybusters" have mounted powerful political protest movements against the substance. The first protests were encountered during the winter of 1992-93 in Fairbanks, AK, as well as in Missoula, MT. Subsequent protest groups have been active in Maine, Connecticut, New Hampshire, New Jersey, Pennsylvania, Texas, and especially California. In New Jersey, the Oxybusters group held a public rally on July 13, 1995 at which time approximately 15,000 protest petition signatures were presented to the governor.(1) In Maine and Connecticut, citizens protesting alleged health problems triggered public hearings by the legislature(2). In California, the Oxybusters collected over 100,000 protest signatures against the substance, which led to the State requiring a formal study of its potential health effects.(3) Despite several reviews (4), there is still considerable confusion and even skepticism about the problem. By and large, most reviews have concentrated on the hypothesis that MTBE itself is the relevant toxin, and have found it difficult to reconcile the known toxicology of MTBE with the reported symptoms. For example, the Interagency Assessment of Oxygenated Fuels said "The anecdotal reports of acute health symptoms among some individuals can not yet be explained or dismissed".(5) In a previous AWMA paper(6), I argued that a more likely hypothesis is that some of the symptoms are being caused by an unsuspected combustion byproduct of MTBE, rather than by MTBE itself. For example, studies by the Alaska State Health Department, which led to that state's withdrawal from the MTBE-WOG program, found that symptoms were at least as common while driving as while pumping gasoline.(7) However, MTBE concentrations in the vicinity of gasoline pumps are on the order of several parts per million, whereas concentrations in the ambient air are at least two orders of magnitude lower.(8) This suggests that the symptoms are due to an exhaust product other than MTBE. Furthermore, there are published reports of people experiencing unpleasant symptoms while driving in traffic(9). In my previous paper(6), I showed statistical data from the Philadelphia area that supports the claim that there have been large increases in certain disease conditions, especially asthma, following the introduction of MTBE-oxygenated gasoline in November 1992. I coined the phrase MTBE Derived Toxin (MDT) to represent the unknown toxin, and suggested that tertiary butyl formate (TBF) might be an exhaust byproduct. This was motivated by the fact that TBF has been shown to be the primary product of photo oxidation of MTBE. However, subsequent studies have shown no detectable concentrations of TBF in the exhaust from MTBE(10). In retrospect, this is not surprising since TBF is known to rapidly pyrolyze into isobutene and formic acid.(11) Recent studies do identify formic acid as a component of MTBE exhaust(12,13) but the concentrations are so low (less than 10% of the formaldehyde levels) that it appears unlikely that formic acid is playing a major role in the problem. What are needed are new measurements of formic acid in the ambient air in cities where MTBE-RFG is used. In the following sections I present new data on increases in purely neurological conditions in Philadelphia, as well as argue that the unknown MDT may be the indirect result of the production of methoxy radicals from MTBE pyrolysis. I especially emphasize the possible importance of methyl nitrite (CH3ONO) and hydro-peroxy radicals (HO2). DATA The Philadelphia Department of Health operates eight public health clinics. These clinics have a computerized data base, including diagnoses, for each patient visit since March 1993. While there were some changes in the diagnosis reporting forms used for children, there have been no changes in the reporting forms for adults from 1993 to 1996. For this reason, only results for adults are presented. During this period of time the total number of visits increased by 19%. The data base begins March 1993. For this reason, a "year" was defined as starting in March of a given calendar year and terminating at the end of February of the following year.
In the 1997 paper(6), similar data were presented for symptoms more characteristic of respiratory irritants, including asthma, chronic bronchitis, otitis, conjunctivitis, chronic sinusitis, and winter allergic rhinitis. These data covered the period 1993-95. Subsequent data on those conditions show that all have continued to increase with the single exception of chronic bronchitis. (These data will not be reproduced here.) It is argued that some people with previously diagnosed chronic bronchitis may have been rediagnosed with asthma. Conventional medical opinion holds that adults do not normally develop asthma de novo unless there is some significant irritant or toxic exposure. THE UTAH EFFECT A group under the direction of Delbert Eatough(14) has reported measurements of the ratio of SO4 (sulfate particles) to SO2 (gas phase) in Utah County, Utah, during two winters when oxygenated fuel was used. During the winter of 95-96, they found that the sulfate/SO2 ratio roughly doubled during the period of WOG. At that time, 28% of the oxygenate used was MTBE(15). The increase was unusual in that it occurred only during nighttime hours, in contradistinction to the better understood photo-oxidation process that occurs during daylight. The obvious implication is that some oxidizing agent is being directly or indirectly produced by MTBE combustion. Du et al(14) hypothesized that the result was due to aldehyde production from the oxygenates, but provided no concurrent measurements of aldehyde concentrations at the time. However, when the experiment was repeated during February 1997, no increase in sulfate was observed.(16) During this time, only ethanol was used as the oxygenate.(15) EXHAUST HYPOTHESES Current exhaust speciation data indicate that formaldehyde, methanol, isobutylene, and formic acid are significantly increased when MTBE is added to the fuel.(17,10,13,12) However, to explain the observed increases in disease, as well as anecdotal reports of symptoms, requires a substance whose percentage increase in the ambient air is very large, say, at least 100%. None of the identified products of MTBE exhaust, except for MTBE itself, have been shown to do that. For example, formaldehyde emissions increases are typically(18,17,19) found to be about 30-50%, whereas winter time increases in ambient formaldehyde are much less, on the order of 10%(20). It is quite likely that methanol will show a large ambient increase, but to date such data are not available. Methoxy radicals, CH3O (abbreviated MeO), are known to be an important intermediate state in combustion of many fuels at very high temperature.(21). However, at the temperatures characteristic of the main combustion chamber of the internal combustion engine (> 2000 degrees K), the MeO are quickly pyrolyzed(22): (1) CH3O -> CH2O + H The lifetime of such pyrolysis is about 10-11 seconds at 2000 K, while the combustion process itself takes about 5-10 milliseconds (ms) before the exhaust cycle begins. Thus one would not expect a significant contribution to MeO in the exhaust stream. However, it is generally recognized that many toxic exhaust products of cars are due to a small fraction of fuel that escapes the high temperature region, due to close contact with the walls of the engine. Such fuel molecules could undergo chemistry more characteristic of a lower temperature range, from 300-600 degrees K. They would then escape into the exhaust pipe and rapidly cool due to expansion. Ethers are virtually unique in the current fuel mix in the possibility of pyrolysis into organic radical components at modest temperatures. For example, one might predict that MTBE would pyrolyze as: (2) CH3-O-C4H9 -> CH3O + C4H9 that is, into methoxy and t-butyl free radicals. However, a careful study of the pyrolysis of MTBE by Choo et al(23) failed to detect any such radicals at all. Working at very low pressures to eliminate multimolecular collisions, they found that only methanol and isobutylene were produced: (3) CH3-O-C4H9 -> CH3OH + C4H8. This requires the breaking of the C-O-C bond, as well as the simultaneous extraction of a hydrogen atom from the tert-butyl moiety, a process called a "four center" reaction. However, recent data from Koshland et al(12) suggest that when MTBE is mixed with gasoline, processes other than reaction (3) take place. Working in a laboratory test reactor at a fixed temperature, they found a major difference between the pure MTBE and the MTBE-RFG situations. From pure MTBE combustion, they identified methanol as in reaction (3). However, when MTBE was mixed with RFG, they found no methanol but instead a roughly equimolar production of formaldehyde. This could be interpreted to mean that the presence of the other components in gasoline altered the chemistry of the pyrolysis process. If methoxy radicals were produced from MTBE, they would probably not have survived under the conditions of that experiment. That is because the temperature was held at a high value for a much longer time than would normally occur in the combustion chamber of an engine, and also that there was no analogous low temperature region corresponding to the exhaust pipe. Obviously, under the conditions of the experiment little nitric oxide would have been produced to react with the MeO radicals. This would explain the increase in formaldehyde by pyrolysis of the MeO as per reaction (1). If one accepts the hypothesis that MeO radicals are produced in the exhaust stream, one must consider the likely consequences. The three major effects I would consider are pyrolysis, survival to the ambient air, and combination with other free radicals in the exhaust stream. The pyrolysis lifetime of MeO, defined as reaction (1), increases rapidly as the temperature decreases.(22) The lifetime is equal to 1 second at a temperature of 418 degrees K, which suggests that at low temperatures some MeO might survive passage through the exhaust pipe. An alternative possibility is that methyl nitrite could be formed by reacting in the exhaust pipe with the NO, which is produced in the main combustion chamber, according to (4) CH3O + NO -> CH3-O-N-O
If some MeO radicals do survive into the ambient air, they will immediately (lifetime < 1 ms) react with atmospheric oxygen to give(27): (5) CH3O + O2 -> CH2O + HO2. The hydroperoxy radicals thus formed could perhaps explain the mysterious oxidation effect seen in Utah(14). The survival of free MeO radicals into the air could actually be aided by the formation of MeONO. The reason is that some MeONO will pyrolyze into CH3O and NO (i.e., the inverse of reaction (4)). That is, the MeONO could serve as a source of transport of MeO from the high temperature region down to the cooler parts of the exhaust stream. Perhaps some radicals would be regenerated by the action of the catalytic converter. DISCUSSION There are several reasons why methyl nitrite is an attractive hypothesis to explain the persistent citizen complaints of symptoms from MTBE. It is known to be an extremely toxic compound. The LC50 for an acute exposure of four hours to rats is only 170 ppm(28); this is approximately 100 times smaller than the corresponding LC50 for benzene, for example.(29) Another important property is that, as an organic nitrite, MeONO is rapidly photolyzed by sunlight. In much conventional thinking about urban air pollution, it is tacitly assumed that the toxic compounds, such as ozone, are produced by photo oxidation. Little attention has been given to possible nighttime toxins, and public health authorities often recommend that people with asthma and other respiratory problems stay indoors in the daytime during ozone alert days, and venture out only at night. This is curious advice, since it is known that asthma attacks occur far more often at night that during daylight.(30) In the present case, it is important to note that many citizens complaining of health effects from MTBE find that they get worse at night, or on cloudy days when the sun is not shining(31). Thus the existence of a highly toxic compound which is produced by MTBE combustion and photolyzed by sunlight perfectly fits the reported complaints. One should ask if there is any evidence for the production of MeONO from combustion of other fuels. There is considerable literature that suggests that MeONO is produced as a combustion product of methanol. However, there are convincing data that such measurements are artifactual, due to methanol in the exhaust sample reacting with NO2 to form the MeONO.(32) This is especially likely if the exhaust products are stored in a bag for a period of several hours. However, one experiment which did not use a storage technique, and which did not see any MeONO in methanol mixed with gasoline, did see MeONO from methanol mixed with diesel fuel.(33) This is again evidence that the chemistry of combustion has many subtle mysteries and surprises, and that the total chemical environment in the combustion process can have profound effects. For this reason I suggest that the Choo results(23) should not be interpreted as ruling out the proposed hypotheses. The lack of MeONO from methanol in gasoline actually supports the observed facts rather well. It suggests that alkoxy radicals are not formed from the pyrolysis of alcohols as they can be from ethers. This would explain the lack of effect seen in Utah during 1997. There is also considerable anecdotal evidence that people who suffer badly from the MDT in the ambient air where they live experience dramatic relief when they travel to areas where only ethanol is used as the oxygenate. In a few cases, people have literally moved from an MTBE region to an ethanol region solely to escape the MDT from MTBE!(34) There are still some curious contradictions in existing speciation studies of MTBE combustion. For example, whereas Koshland et al(12) found no methanol in the exhaust from MTBE-RFG, Stump et al(17) found that methanol was a consistently observed combustion product of MTBE gasoline in actual car engines. This again points to the importance of the complex, multi chamber, aspect of real car engines. Further evidence for a major role of MTBE in nitrogen combustion chemistry comes from two speciation studies of oxygenated fuels by the EPA. Of interest is the effect of oxygenate on the production of nitrogen oxides (NOx) at extremely cold temperatures. Knapp et al(35) found that when ethanol was used for the oxygenate there was a tendency for an increase in NOx; not an unexpected result. However, when MTBE was used at the coldest temperatures, the NOx decreased.(36) Since both substances raise the octane of the gasoline, it seems unlikely that this result could be explained in terms of changes in ignition timing or flame propagation. In view of the hypotheses in this paper, one can surmise that the use of MTBE caused methyl nitrite to be formed at the expense of NOx, and that this would explain the observed decrease. This result is also consistent with the failure to observe MeONO from methanol-gasoline fuels (i.e., the alcohols do not serve as a source of alkoxy radicals at low temperatures). The clinical data presented here suggest that some kinds of air pollution can have affects quite different from what is commonly assumed. It is of course known that various chemicals (gas phase or particulate) can be respiratory irritants, and that these can act via direct attack on the respiratory epithelium. However, the three symptoms documented here do not relate to any reaction with respiratory epithelium, but rather with the nervous system. This is not unexpected for MeONO, since there are analogous compounds, namely isobutyl and amyl nitrite, which are drugs of abuse due to neurological effects.(37,38) There is considerable evidence that the nervous system plays an important role in asthma; indeed, asthma could be defined as a dysfunction of the autonomic nervous system. Furthermore, there is an established effect called "neurogenic inflammation", in which certain nerve cells can incite an inflammatory reaction due to neurological stimulation.(39) This is a very plausible explanation for the well known existence of psychosomatic disorders. In particular, emotional stress is known to play a major role in inducing asthmatic attacks. This should not be interpreted as somehow diminishing the significance of such attacks, but rather as supporting the role of the nervous system in respiratory function. It is perhaps relevant that the specific nerve cells in the respiratory epithelium that respond to airborne toxins have receptors that are lipophilic.(40) This again suggests that a non-ionic substance, such as MeONO, could be responsible for initiating the symptoms experienced, rather than highly ionic substances such as the strong acids. If the ideas in this paper regarding MTBE are correct, then an obvious conclusion would be that any ether-based fuel would produce similar organic nitrites. Dimethyl ether (DME), in particular, would produce the same MeO radicals and MeONO toxic emissions. It is interesting to note that automobile fuel in New Zealand is largely based on dissolved DME, and that might explain the extraordinarily high asthma rates found now in that country(41). CONCLUSIONS Data were presented showing large increases in visits to Philadelphia clinics for insomnia, cardiac dysrhythmia, and malaise since March 1993. These increases may be related to previously reported increases in various respiratory and inflammatory symptoms such as asthma, sinusitis, allergy, etc. It is argued from the timing, as well as from extensive anecdotal reports, that the problems are caused by an unrecognized combustion product of MTBE. Further evidence for this theory comes from experiments in Utah which demonstrated a doubling of SO4/SO2 ratio during the winter oxygenated fuel season of 1995-96. The hypothesis is proposed that MTBE combustion will produce methoxy radicals, and that these will combine with NO to produce methyl nitrite. Methyl nitrite is an attractive hypothesis because it is very toxic and is known to produce some of the observed symptoms. Since analogous organic nitrites are drugs of abuse due to neurological effects, it is plausible that the nitrites act via the nervous system rather than as classical respiratory irritants. The fact that methyl nitrite is photolyzed by sunlight perfectly fits the anecdotal reports of people who experience symptoms only at night or on cloudy days. ACKNOWLEDGMENTS The author is indebted to Mr. Warner Tillack of the Philadelphia Department of Health for providing the statistical data in this paper. He is also indebted to several chemists for helpful discussions, including Felix Wehrli, Carlton Howard, Philip Taylor, and Barry Dellinger. Of course, these scientists are not responsible for any errors or misunderstandings that may be in this paper! For More Information, Contact: Peter
M. Joseph, Ph.D. Barry
Grossman Ivo
Granata REFERENCES 1.T.B.Bates, "Fueling a Controversy", Asbury Park Press, Jul 21, 1995 page 1, Asbury Park, NJ 2.D.Jones, "Rise in asthma cases linked to fuel additive", Hartford Courant, Feb 5, 1999 page A7, Hartford CT 3.W.Carlsen, "Plenty of interest groups lobby both sides, Unusual alliances form, Oxybusters take arguements to radio airwaves", Chronicle, Sep 16, 1997 page A9, San Francisco 4.A. Committee, "Toxicological and Performance Aspects of Oxygenated Motor Vehicle Fuels," Nat Acad Sci, Washington, DC, 1996 pp. 75-115. 5.A.Committee, "Interagency Assessment of Oxygenated Fuels," OSTP, Washington, 1997, p. 4-31. 6.P.M.Joseph, "Changes in disease rates in Philadelphia following the introduction of oxygenated gasoline. Paper 97-TA34.02," Air and Waste Management Assoc, Pittsburgh, PA, 1997 pp. 1-15. 7.M.Beller, M.Schloss, J.Middaugh, Evaluation of Health Effects from Exposure to Oxygenated Fuel in Fairbanks, Alaska. State of Alaska Bulletin 26, 1-1 (1992). 8.A.Committee, "Interagency Assessment of Oxygenated Fuels," OSTP, Washington, 1997, p. 6-6. 9.A.Carella, "Research Fuels Debate. Sufferers unswayed by reports dismissing dangers of gas additive", The Advocate, Nov 22, 1996 page A1, Stamford, CT 10.T.E.Kleindienst, D.F.Smith, F.M.Black, S.B.Tejada, "Atmospheric Chemistry studies of exhaust from vehicles operating with reformulated fuel," AWMA, Pittsburgh PA, 1997 pp. 1-18. 11.E.Gordon, S.J.W.Price, A.F.TrotmanDickerson, The pyrolysis of tert. Butyl Formate. J. Chem. Soc. 1957, 2813-2815 (1957). 12.C.P.Koshland, R.F.Sawyer, D.Lucas, P.Franklin, "Evaluation of automotive MTBE combustion byproducts in California reformulated gasoline," U.C.Davis, Davis, CA, 1998 pp. 1-48. 13.J.M.Davis, U.S. E.P.A., 1998, private communication. 14.Y.Du, Y.Ren, N.F.Mangelson, D.J.Eatough, "Enhanced formation of sulfate and nitrate associated with the use of oxygenated fuels. Paper 97-WP68B.05," Air and Waste Management Assoc., Pittsburgh, PA, 1997 pp. 1-14. 15.B.Allen, Utah Dept. Env. Quality, 1998, private communication. 16.D.J.Eatough, Provo,UT, 1998, private communication. 17.F.D.Stump, K.T.Knapp, W.D.Ray, P.D.Siudak, R.F.Snow, Influence of oxygenated fuels on the emissions from three pre-1985 light-duty passenger vehicles. J. AWMA 44, 781-786 (1994). 18.S.K.Hoekman, Speciated Measurements and Calculated Reactivities of Vehicle Exhaust Emissions from Conventional and Reformulated Gasolines. Environ. Sci. Technology 26, 1206-1216 (1992). 19.T.W.Kirchstetter, B.C.Singer, R.A.Harley, G.R.Kendall, W.Chan, Impact of Oxygenated Gasoline Use on California Light-Duty Vehicle Emissions. Environ. Sci. Technology 30, 661-670 (1996). 20.L.G.Anderson, J.A.Lanning, E.Wilkes, "Effects of using oxygenated fuels on carbon monoxide, formaldehyde, and acetaldehyde concentrations in Denver. Paper 97-RP139.05," Air and Waste Management Assoc., Pittsburgh, PA, 1997 pp. 1-13. 21.I.Glassman, "Combustion. Third Edition," Academic Press, San Diego, 1996 pp. 93-94. 22.L.Batt, J.P.Burrows, G.N.Robinson, On the isomerization of the methoxy radical: Relevance to atmospheric chemistry and combustion. Chem. Phys. Lett. 78, 467-470 (1981). 23.K.Y.Choo, D.M.Golden, S.W.Benson, Very Low-pressure pyrolysis (VLPP) of t-butyl methyl ether. Int. J. Chem. Kin. 6, 631-641 (1974). 24.L.Batt, The gas-phase decomposition of alkoxy radicals. Int. J. Chem. Kin. 11, 977-993 (1979). 25.L.Batt, Reactions of alkoxy and alkyl peroxy radicals. Int. Rev. Phys. Chem. 6, 53-90 (1987). 26.Y.He, W.A.Sanders, M.C.Lin, Thermal decomposition of methyl nitrite: Kinetic modeling of detailed product measurements by gas-liquid chromotography and Fourier transform infrared spectroscopy. J. Phys. Chem. 92, 5474-5481 (1988). 27.B.J.Finlayson-Pitts, J.N.Pitts, "Atmospheric Chemistry: Fundamentals and experimental techniques," John Wiley, New York, 1986 pp. 425-426. 28.D.R.Klonne, C.E.Ulrich, J.Weissman, A.K.Morgan, Acute inhalation toxicity of aliphatic (C1-C5) nitrites in rats. Fundam. Appl. Toxicol. 8, 101-106 (1987). 29.H.F.Smyth, C.P.Carpenter, C.S.Weil, Range finding toxicity data: List VI. J. Ind. Hyg. Assoc. 23, 95-107 (1962). 30.R.J.Martin, "Nocturnal Asthma: Mechanisms and treatment," Futura Pub, Mount Kisco NY, 1993 pp. 71-75. 31.P.M.Joseph, Letter: Atmospheric Byproducts of MTBE as a Source of Community-wide Illness. Arch. Env. Health 50, 395-396 (1995). 32.B.J.Finlayson, J.N.Pitts, A.C.Lloyd, Comment on "A study of the stability of methanol-fueled vehicle emissions in Tedlar bags". Envir. Sci. Tech. 26, 1668-1669 (1992). 33.A.Jonsson, B.M.Bertilsson, Formation of methyl nitrite in engines fueled with gasoline/methanol and methanol/diesel. Environ. Sci. Technol. 16, 106-110 (1982). 34.R.Malecki, Salt Lake City, 1995, private communication. 35.K.T.Knapp, F.D.Stump, S.B.Tejada, The effect of ethanol fuel on the emissions of vehicles over a wide range of temperatures. J. AWMA 48, 646-653 (1998). 36.K.T.Knapp, C.B.Prakash, "MTBE effect on vehicle emissions at cold temperatures," ISAF, Sun City,SA, 1996 pp. 786-797. 37.H.W.Haverkos, J.Dougherty, Health hazards of nitrite inhalants. Am. J. Med. 84, 479-482 (1988). 38.E.M.Dax, W.R.Lange, J.H.Jaffe, Allergic reactions to amyl nitrite inhalation. Am. J. Med. 86, 732-732 (1989). 39.W.J.Meggs, RADS and RUDS - The toxic induction of asthma and rhinitis. Clinical Toxicology 32, 487-501 (1994). 40.G.D.Nielsen, Mechanisms of Activation of the Sensory Irritant Receptor by Airborne Chemicals. Crit. Revs. Toxicology 21, 183-208 (1991). 41.R.Beasley, Worldwide variation in prevalence of symptoms of asthma, allergic rhinoconjunctivitis, and atopic eczema:ISSAC. Lancet 351, 1225-1232 (1998). Note: The
information on this website is presented for educational purposes only. |
