"Low-Level" Carbon Monoxide Administration May Carry Risk/From the Authors

To the Editor:

We were concerned by the article of Dr. Mayr and colleagues, which describes exposures of healthy subjects to inhaled carbon monoxide (CO) (1). In this study, conducted in Austria, 13 volunteers were injected with intravenous LPS and asked to breathe 500 ppm CO for 1 hour. The authors sought to determine whether CO has antiinflammatory effects in human endotoxemia as previously claimed in mice (2). The authors acknowledged CO's known toxicity, but claimed the exposures were safe because blood carboxyhemoglobin (HbCO) levels were at 10% or less. They reasoned that since smokers experience similar blood concentrations, such levels are "innocuous."

In 2000, Emanuel and coworkers published seven requirements for clinical research to be ethical (3). This study appears to violate at least one of them, that of a favorable risk-benefit ratio. This requirement necessitates that risk be minimized and be proportionate to the benefits to the subject and society. We believe that the CO exposures used by Mayr and coworkers should not have been assumed to be low risk. A simple comparison of blood levels to those of a different but potentially toxic exposure does not prove safety. Chronic CO exposure may have different effects in smokers than acute exposures in nonsmokers, and endotoxemia is not present in apparently healthy smokers. Finally, such "low-level" CO exposures in normal volunteers have demonstrated toxicity: neuropsychologic impairment in subjects breathing up to 100 ppm CO for 1.5 to 2.5 hours (4). In the United States, the National Institute for Occupational Safety and Health has set a ceiling limit for maximum CO exposure at 200 ppm, an 8-hour workday limit of 50 ppm, and a maximum HbCO level of 5% (5, 6).

The authors noted that the antiinflammatory effect they sought was seen in prior murine studies both in vitro and in vivo. Thus, at a minimum, it would seem appropriate to demonstrate the effect at a reasonable CO concentration in vitro before exposing human volunteers to a recognized toxin. If it could not be demonstrated in human cells in vitro, there would have been little or no reason to proceed to human studies.

Despite the fact that the authors' local ethics committee approved the study, it appears not to meet accepted standards for ethical research. If questionable studies such as this are published, other investigators may be encouraged to pursue similar work and cause needless harm to research participants.

Conflict of Interest Statement: None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

NEIL B. HAMPSON

Virginia Mason Medical Center

Seattle, Washington

LINDELL K. WEAVER

LDS Hospital

Salt Lake City, Utah

CLAUDE A. PIANTADOSI

Duke University Medical Center

Durham, North Carolina

References

1. Mayr FB, Spiel A, Leitner J, Marsik C, Germann P, Ullrich R, Wagner O, Jilma B. Effects of carbon monoxide inhalation during experimental endotoxemia in humans. Am J Respir Crit Care Med 2005;171:354-360.

2. Otterbein LE, Bach FH, Alam J, Soares M, Tao Lu H, Wysk M, Davis RJ, Flavell RA, Choi AM. Carbon monoxide has anti-inflammatory effects involving the mitogen-activated protein kinase pathway. Nat Med 2000;6:422-428.

3. Emanuel EJ, Wendler D, Grady C. What makes clinical research ethical? JAMA 2000;283:2701-2711.

4. Amitai Y, Zlotogorski Z, Golan-Katzav V, Wexler A, Gross D. Neuropsychological impairment from acute low-level exposure to carbon monoxide. Arch Neurol 1998;55:845-848.

5. National Institute for Occupational Safety and Health. Recommendations for occupational safety and health: compendium of policy documents and statements. Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Institute for Occupational Safety and Health; 1992. DHHS (NIOSH) Publication No. 92-100.

6. National Institute for Occupational Safety and Health. Registry of toxic effects of chemical substances: carbon monoxide. Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Institute for Occupational Safety and Health; 1993.

From the Authors:

Drs. Hampson, Weaver, and Piantadosi have accused us of performing a "questionable" study (1) that "did not meet accepted standards for ethical research" and caused "needless harm" to research participants. Although we are convinced that critical assessment by colleagues is invaluable, we feel that this harsh criticism is not warranted.

Our study was the first conducted in humans with the goal to reproduce the antiinflammatory effects of carbon monoxide (CO) inhalation observed in a murine endotoxemia model (2). Phase I studies are always troublesome with respect to potential individual benefits, since participants are healthy and do not directly benefit from treatment. Thus, one has to assess whether potential societal benefits in terms of knowledge justify the excess risk to individual subjects.

As discussed in our article (1), CO inhalation with similar concentrations has been performed in healthy volunteers previously. Our carboxyhemoglobin (HbCO) cut-off level was determined after extensive review of the literature, which showed that short-term exposure to CO concentrations of 500 ppm had no "appreciable effect" and exposure to concentrations as high as 1,000 to 1,200 ppm were not detrimental (3). Furthermore, we addressed potential adverse effects of low-dose CO inhalation in great detail and strongly focused on safety issues. To demonstrate our awareness, we conducted a placebo-controlled four-way, crossover, dose-escalation pilot trial (without endotoxemia) to evaluate safety and tolerability of CO inhalation in humans with serial HbCO measurements performed every 15 minutes and continuous monitoring of vital parameters (i.e., ECG, heart rate, blood pressure, and pulse oximetry). Although experts in the field (including one reviewer) advised us to prolong CO treatment or increase CO concentrations, we rejected these suggestions because of potential safety concerns and lack of up-to-date published data. We are also aware that clinical severity of CO poisoning often poorly correlates with HbCO values and has to be evaluated in context with duration of exposure, the concentration of CO inhaled, and degree of activity during exposure (4). For example, HbCO levels are higher in volunteers who have an elevated heart rate. Our volunteers had mean heart rates of 67 beats per minute and were confined to bed rest.

As we also discussed in our article, impairment of cognitive and psychomotor abilities in acute CO intoxication at HbCO levels as low as 5% is well described. However, there is no evidence that mild CO intoxication leads to permanent neuropsychologic sequelae as assessed by neuropsychologic testing. For example, moderate to severe CO intoxication (median HbCO levels of 23% at admission, i.e., threefold our peak HbCO levels) had no permanent neuropsychologic sequelae in subjects without risk factors for cognitive disturbances (similar to our healthy volunteers) (5).

Recommended exposure limits for CO exposure at workplaces are heterogeneous. For example, the National Institute for Occupational Safety and Health recommended exposure limit for CO is 35 ppm for an 8-hour time-weighted average exposure, with a ceiling limit of 200 ppm for short-term exposure, and is designed to protect workers from health effects associated with HbCO concentrations in excess of 5% (6). These recommendations differ from those of the American Conference of Governmental Industrial Hygienists, which again are different from the standards recommended by the Occupational Safety and Health Administration and the National Research Council (7, 8). Although our volunteers reached maximum HbCO values slightly higher than 5% (1), these values dropped below 5% approximately 2 hours after the end of inhalation. All participants were observed until 8 hours after the end of inhalation before being dismissed from our study ward.

Hampson and colleagues also maintain that the antiinflammatory effects of CO should have been tested in vitro before "exposing human volunteers to a recognized toxin." Before exposing our volunteers to CO, we were able to demonstrate that CO had antiinflammatory effects in human peripheral blood monocytes in vitro (M.B., unpublished data). In short, monocytes were cultured and pretreated with 250 ppm CO or air for 3 hours. Thereafter, cells were incubated with 1

In conclusion, we do not think that the harsh criticism of Drs. Hampson, Weaver, and Piantadosi is justified. Data from a rodent model (2) as well as results from human and murine in vitro experiments provided a sound rationale to investigate putative antiinflammatory effects of CO in humans.

Conflict of Interest Statement: None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

FLORIAN MAYR

MARTIN BILBAN

BERND JILMA

Medical University of Vienna

Vienna, Austria

References

1. Mayr FB, Spiel A, Leitner J, Marsik C, Germann P, Ullrich R, Wagner O, Jilma B. Effects of carbon monoxide inhalation during experimental endotoxemia in humans. Am J Respir Crit Care Med 2005;171:354-360.

2. Otterbein LE, Bach FH, Alam J, Soares M, Tao Lu H, Wysk M, Davis RJ, Flavell RA, Choi AM. Carbon monoxide has anti-inflammatory effects involving the mitogen-activated protein kinase pathway. Nat Med 2000;6:422-428.

3. Henderson Y, Haggard HW. Noxious gases. New York: Reinhold Publishing; 1943.

4. Forbes WH, Sargent F, Roughton FJW. The rate of carbon monoxide uptake by normal men. Am J Physiol 1945;143:594-608.

5. Deschamps D, Geraud C, Julien H, Baud FJ, Dally S. Memory one month after acute carbon monoxide intoxication: a prospective study. Occup Environ Med 2003;60:212-216.

6. National Institute for Occupational Safety and Health. NIOSH recommendations for occupational safety and health standards: compendium of policy documents and statements. Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Institute for Occupational Safety and Health; 1992. DHHS (NIOSH) Publication No. 92-100.

7. American Conference of Governmental Industrial Hygienists. Threshold limit values and biological exposure indices for 1995-1996. Cincinnati, OH: American Conference of Governmental Industrial Hygienists; 1995.

8. National Research Council. Emergency and continuous exposure guidance levels for selected contaminants. Washington, DC: National Research Council, National Academy Press; 1985.

© 2005 American Thoracic Society Provided by ProQuest LLC. All Rights Reserved.