Original contributionCarboxyhemoglobin measurement by hospitals: Implications for the diagnosis of carbon monoxide poisoning
Introduction
Carbon monoxide (CO) poisoning is common in the United States, accounting for an estimated 40,000 Emergency Department visits for diagnosed cases annually (1). Because the signs and symptoms of CO poisoning are nonspecific, it is likely that many more cases are unsuspected, attributed to other etiologies, and therefore go undiagnosed.
When CO poisoning is suspected, measurement of blood carboxyhemoglobin (COHb) is typically performed. An elevated COHb level (greater than 2% for nonsmokers and greater than 9% for smokers) documents exposure to exogenous CO and supports the diagnosis (2). COHb is measured in hospital laboratories by multi-wavelength CO-oximetry. Not all hospitals have CO-oximeters due to the expense of the equipment. In that case, blood samples are typically sent to an outside laboratory for COHb measurement. Alternate methods that are sometimes used to demonstrate CO exposure include a qualitative colorimetric screening test performed on blood or measurement of exhaled carbon monoxide (3).
Lack of ability to measure COHb in a hospital has the potential to result in failure to diagnose cases of CO poisoning or contribute to delay in diagnosis in the case where a blood sample is sent elsewhere. To assess the capability to measure COHb in a region of the United States with a high incidence of CO poisoning (4), we surveyed all acute care hospitals in the states of Alaska, Idaho, Montana, and Washington. Results were then compared with the list of hospitals referring patients with acute CO poisoning to a major regional center for hyperbaric oxygen (HBO2) therapy.
Section snippets
Methods
The clinical laboratories of all acute care hospitals in the states of Alaska (n = 20), Idaho (n = 37), Montana (n = 50) and Washington (n = 97) were surveyed by telephone in January 2005. The laboratory supervisor was asked the following questions:
Can your laboratory measure carboxyhemoglobin levels?
If yes:
- 1
What type of blood sample do you use (arterial, venous, either)?
- 2
What method do you use?
- 3
From the time blood is drawn in your Emergency Department, how long does it take to get the result?If
Results
Of the 204 acute care hospitals in the four-state region, 90 (44%) have the capability to measure carboxyhemoglobin, all with laboratory CO-oximetry. The frequency ranges from a low of 32% in both Idaho and Montana to a high of 54% in Washington. The rest of the hospitals send blood samples to outside laboratories for COHb measurement. None use the qualitative colorimetric blood assay for detection of COHb. The population of the town in which a hospital is located correlates significantly with
Discussion
This study demonstrates that less than one-half of the hospitals in the four-state region surveyed have the capability to measure carboxyhemoglobin. This has great potential significance in light of the fact that CO poisoning is common in the region, with a combined death rate more than twice that of the other 46 states (4). The fact that treatment with hyperbaric oxygen has been proven to dramatically reduce the incidence of long-term cognitive sequelae from CO poisoning further underscores
Acknowledgment
The Edward H. Morgan Chair in Pulmonary and Critical Care Medicine, Virginia Mason Medical Center, Seattle, provided financial support for this study.
References (13)
Emergency department visits for carbon monoxide poisoning
J Emerg Med
(1998)- et al.
Selection criteria utilized for hyperbaric oxygen treatment of carbon monoxide poisoning
J Emerg Med
(1995) Pulse oximetry in severe carbon monoxide poisoning
Chest
(1998)- et al.
Blood carbon monoxide levels in persons 3–74 years of ageUnited States, 1976–80
Adv Data
(1982) - et al.
Breath analysis to detect recent exposure to carbon monoxide
Postgrad Med J
(2002) - et al.
Unintentional carbon monoxide-related deaths in the United States, 1979 through 1988
JAMA
(1991)