Send letter to journal:
Re: A Novel Hypothesis to Explain Associations of Carbon Monoxide and Nitrogen Dioxide with Deaths from

“A Novel Hypothesis to Explain Associations of Carbon Monoxide and Nitrogen Dioxide with Deaths from Respiratory Disease”

Dear Editor:

I am writing to comment on the article[1] entitled “Atmospheric pollutants and mortalities in English local authority areas” by E. G. Knox. The paper analyzed a large body of epidemiologic data linking various forms of air pollution to the death rates from several diseases, including pneumonia, COPD, and lung cancer.

I want to suggest that in some cases the author unknowingly mis-classified the exposure, and that the associations with carbon monoxide (CO) and nitrogen dioxide (NO₂) may be more plausibly linked to an unsuspected exhaust component of engines, namely, methyl nitrite (MN).

The paper does not give any information on the actual ambient concentrations of any of the pollutants, it presents only the results of a correlation analysis between death rates and pollutant concentrations. That is unfortunate, because at least for CO and NO₂ there are many other studies in England that indicate that the levels of those two pollutants are far below the levels known to have any harmful health effects. Carslaw[2] published a recent study of CO concentration near Marylebone Road, which is a major arterial road located in central London, and found that mean CO concentrations dropped from 2.1mg/m3 (approximately 2 ppm) in 1998 down to 0.8 mg/m3 in 2005. Another study[3] found that, depending on wind speed, the CO concentration in Manchester varied from 0.4 to 1.12 ppm. The same study found that NOx concentrations varied from 24 to 79 ppb in Manchester, from 34 to 55 ppb in Edinburgh, and from 7 to 31 ppb in Birmingham.

The toxicology of CO is very well known to alter oxygen saturation in red blood cells, but only at concentrations that are far above the those observed in England during the time period 1996-2004. Exposure to CO at the current criterion concentration (9 ppm) produces[4] carboxyhemoglobin (CoHg ) of only about 1.4-2.0%, which is clinically insignificant (and only slightly above the endogenous levels of CoHg). Thom[5] has demonstrated that CO can have more subtle biological effects that he thinks can be significant in cases of frank CO poisoning. However, his in vivo studies with rats involved CO concentrations of 1000 ppm and above. In one animal study[5] he found no significant effects at concentrations below 100 ppm.

Regarding attributing deaths from lung diseases to ambient CO, a remarkable development is the recent interest in using CO as a therapeutic agent for lung disease. One recent review[6] cited 159 references, with many studies indicating that the gas can produce a variety of useful therapeutic effects, including anti-inflammatory, anti-proliferative, and vasodilation. Brief animal exposures of 250 ppm of CO protected against ventilator-induced lung injury and reduced the asthmatic response to allergens[7]. It also reduced mortality in an animal model of pneumonia[8]. One clinical trial[9] of patients with COPD using CO exposure of 125 ppm demonstrated a reduction in sputum eosinophils and in response to methacholine challenge.

Hence I find it less than plausible that the very low ambient levels of CO in England could be the direct cause of so many deaths from respiratory disease.

I want to present an alternative hypothesis to explain these pollutant-mortality correlation results. I have argued[10] [11] that the use of methyl ether (such as MTBE or TAME) in gasoline creates MN in the exhaust. Since CO is primarily an engine exhaust product, increased CO would be expected to be strongly correlated with MN exhaust. Dr. Knox stated: “Engine exhaust accounts for the greater part of carbon monoxide, ... and nitrogen oxides...” Furthermore, MN can easily be confounded with NO₂, as I have argued previously (Occ & Env Med 2008, in press). One review[12] stated: ”The overall results suggest that outdoor NO₂“. was serving as a marker for more causal airborne agents rather than a direct effect of NO₂“. A more detailed analysis of this confounding effect, based on many previously published studies of the epidemiology of NO₂, is currently in preparation. Another review[11] indicated that the alkyl nitrites are known to be harmful to both the respiratory and immune systems, and MN has been shown[13] to be mutagenic by the Ames test.

While I feel that this MN hypothesis is extremely plausible (and especially more plausible than that the effects are directly due to CO and NO₂), until such time as MN is positively identified its role in environmental epidemiology remains tentative. We need a careful study of the presence of MN in engine exhaust using the techniques of gas-phase analytical chemistry. However, I (personally) have neither the equipment nor expertise for such studies. I can only urge that those with such expertise investigate this issue.

Sincerely

Professor Peter M. Joseph, Ph.D.
School of Medicine
University of Pennsylvania
Philadelphia, PA, USA
email = joseph@rad.upenn.edu

I have no competing interests.

References

1. Knox EG. Atmospheric pollutants and mortalities in English local authority areas. J Epidemiol Community Health 2008;62(5):442-7.

2. Carslaw DC, Beevers SD, Tate JE. Modelling and assessing trends in traffic-related emissions using a generalised additive modelling approach. Atmospheric Environment 2007;41(26):5289-5299.

3. Longley ID, Inglis DWF, Gallagher MW, Williams PI, Allan JD, Coe H. Using NOx and CO monitoring data to indicate fine aerosol number concentrations and emission factors in three UK conurbations. Atmospheric Environment 2005;39(28):5157-5169.

4. Utell MJ, Warren J, Sawyer RF. Public health risks from motor vehicle emissions. Annu Rev Public Health 1994;15:157-78.

5. Thom SR, Fisher D, Xu YA, Garner S, Ischiropoulos H. Role of nitric oxide-derived oxidants in vascular injury from carbon monoxide in the rat. Am J Physiol 1999;276(3 Pt 2):H984-92.

6. Hoetzel A, Schmidt R. Kohlenmonoxid--Gift oder potenzielles Therapeutikum? Anaesthesist 2006;55(10):1068-79.

7. Ryter SW, Choi AM. Therapeutic applications of carbon monoxide in lung disease. Curr Opin Pharmacol 2006;6(3):257-62.

8. Hoetzel A, Dolinay T, Schmidt R, Choi AM, Ryter SW. Carbon monoxide in sepsis. Antioxid Redox Signal 2007;9(11):2013-26.

9. Bathoorn E, Slebos DJ, Postma DS, Koeter GH, van Oosterhout AJ, van der Toorn M, et al. Anti-inflammatory effects of inhaled carbon monoxide in patients with COPD: a pilot study. Eur Respir J 2007;30(6):1131-7.

10. Evidence for methyl nitrite as an exhaust component from engines with certain fuels. Annual Meeting Proceedings CD-ROM Air & Waste Management Association, 99th, New Orleans, LA, June 20-23, 2006; 2006; New Orleans, LA.

11. Joseph PM. Paradoxical ozone associations could be due to methyl nitrite from combustion of methyl ethers or esters in engine fuels. Environ Int 2007;33(8):1090-106.

12. Delfino RJ. Epidemiologic evidence for asthma and exposure to air toxics: linkages between occupational, indoor, and community air pollution research. Environ Health Perspect 2002;110 Suppl 4:573-89.