and in the presence of other CNS active gases, such as hydrogen cyanide (HCN) and carbon dioxide (CO2). In the absence of reliable definitive data at low-to-moderate CO concentrations and in order to provide a safety margin for neuropsychological and neurophysiologic effects and potential cardiovascular effects, governmental groups have generally chosen to rely on the studies reporting effects at low levels when setting standards for critical tasks (Wong 1994).
Recommendation Information provided to the committee by the Army indicates that exposure scenarios in armored vehicles deployed in battle are expected to generate COHb levels less than 10% in vehicle personnel (M. Bazar and T. Kluchinsky, CHPPM, personal commun., April 14, 2008).
In view of the lack of reliable and relevant neuropsychological data and the inconsistencies in existing data at levels in this range, the committee recommends that the Army consider controlled human experiments using scenarios and exposure concentrations of CO relevant to combat conditions. Such experiments should be designed to seek a dose-response relationship at CO and COHb levels at less than 10% and with neuropsychological end points, such as visual and reaction-time decrements relevant to real-life scenarios in enclosed armored vehicles. It would not be feasible to measure these same end points in a meaningful way in exposure experiments using animals (also see discussion in Chapter 3). Also, the lack of consistent neuropsychological and neurophysiologic data precludes the use of computational models to estimate expected changes in human performance with increased COHb.
Controlled human experiments could be done best with vehicle simulators in an atmospheric chamber (or less desirably with face-delivery systems) where scenarios, CO concentrations, temperature, and humidity could be individually controlled. The Army will also need to control for “background” CO exposure from smoking and other sources. If these studies were done under controlled conditions, the Army would need to decide whether to model a scenario of rapid CO buildup or slow CO buildup, because the physiologic effects, and subsequent behavioral responses, might be different for these two scenarios. The small number of subjects who can practically be tested will likely be a limiting factor in these experiments, but given the current state of existing data, such controlled experimentation is the only conclusive way to assess the qualitative and quantitative risks of performance degradation resulting from invehicle CO release. Alternatively, experiments could be done in a test vehicle with standardized measures of performance under battlefield conditions. Such experiments, however, are unlikely to give definitive results due to multiple uncontrollable variables, such as heat, workload, and stress.
Attention should be paid to assessment of armored-vehicle crews for medical conditions that might adversely affect any performance degradation from CO or other hypoxia-producing conditions (for example, high altitude). Such conditions would also include diseases affecting pulmonary efficiency (for example, asthma and chronic obstructive pulmonary disease), cardiac diseases that can increase the risk of cardiac dysrhythmias or dysfunction, and blood diseases that can affect oxygen transport (for example, anemia and hemoglobinopathies). The committee acknowledges that the Army probably currently screens for some or all of these conditions in its crews.