The TERESA project represents a novel approach for investigating the toxicity of PM derived from coal combustion. Past toxicological studies of coal combustion emissions have evaluated the health effects of primary emissions such as coal fly ash. Given that coal-fired power plants in the US have adopted controls to reduce the emission of primary particles, the relevance of toxicological studies using primary emission PM is unclear. However, even with such controls, stack emissions of SO2 and NOx still contribute substantially to ambient PM in the form of secondary particles. The health effects of these secondary particles, in combination with any residual primary PM, are largely unknown. The goal of the current study was to evaluate the effects of aged stack emissions from coal-fired power plants on electrocardiographic changes in a rat model of acute myocardial infarction.
We carried out these experiments at two power plants (Power Plants 2 and 3). At Power Plant 2, we found that exposure to a scenario reflecting an aged plume with unneutralized acidity and secondary organic aerosol was associated with increased frequency of ventricular arrhythmias, decreased respiratory expiratory time and decreased end-inspiratory pause. We did not observe statistically significant changes in heart rate, heart rate variability, or ECG intervals. As discussed below, the experiments from Power Plant 3 were uninterpretable.
We had chosen a priori
to carry out these experiments using the exposure scenario that demonstrated the most substantial change in cardiac in vivo
chemiluminescence in normal animals. This decision was based on studies showing important changes in this measure following exposure to concentrated ambient particles (Gurgueira et al., 2002
) that may be abrogated with the anti-oxidant N-acetyl cysteine (Rhoden et al., 2004
). Preliminary analyses of data from Power Plant 2 suggested similar increases in cardiac chemiluminescence in normal animals exposed under the POS and PONS scenarios and we chose to use the POS scenario for the MI experiments. In the final analysis, the change in heart chemiluminescence at Power Plant 2 was more pronounced for the PONS scenario than for the POS scenario. On the other hand, across all 3 power plants, exposure to the POS scenario increased heart chemiluminescence more than did exposure to the PONS scenario. Therefore, in retrospect, either the POS or PONS exposure scenarios would have been an acceptable choice for the MI studies.
Our choice to use a complex exposure scenario rather than a scenario involving only primary particles was based on the fact that emissions from coal-fired power plants interact with natural and anthropogenic organics in the atmosphere. Although the final products in the atmosphere are not strictly from the power plant, it can be argued that if power plant emissions were not present in that form, different reactions might take place in the environment. Because the reactions that we studied do in fact take place, we believe their use in a study of health effects is a strength of this project.
The experiment with MI animals at Power Plant 3 failed due to unplanned power plant shutdowns, unexpected variations in infarct location in our animal model, and technical difficulties obtaining ECG signals of sufficient quality. Plant shutdowns were uncommon in this series, and usually had minor impacts in experiments. The shutdowns at Power Plant 3 had greater impact because they occurred after the surgical preparation of the MI animals but prior to their exposure. Because exposure needed to occur within 24 hrs of the surgery, the period of greatest myocardial ectopic vulnerability (Wellenius et al., 2002
; Wellenius et al., 2004
), delaying exposures was not feasible. Unfortunately, extending our planned stay at Power Plant 3 or returning to the power plant at a later date to carry out additional experiments was not possible.
The predominant site of myocardial infarction (left ventricle at Power Plant 2, right ventricle in Power Plant 3) is important in this animal model, as evidenced by the much higher rate of ventricular ectopy among the sham-exposed animals at Power Plant 2 versus Power Plant 3. The difference in infarct distribution is not explained by the technical factors mentioned above. One possible explanation is that Sprague-Dawley is an outbred strain of rats, and variations in coronary artery distribution can occur. That it occurred in a series of animals all received at the same time may suggest that this variation could well have been a feature of this cohort of animals, many of which could have been siblings. Coupled with an unexpectedly high rate of technical failure in ECG recording equipment, these problems resulted in the data set from Power Plant 3 being small and highly unbalanced. This not only limited the analyses that we could perform on these data, but also casts serious doubts on the assumptions underlying the statistical models applied. Specifically, it is unlikely that data were missing at random or that the two exposure groups were successfully randomized. For these reasons, we do not believe that valid conclusions can be drawn from the data from Power Plant 3..
Of note, this rat model of acute vulnerability to arrhythmia after MI requires survival thoracic surgery on an animal that has developed a large, transmural myocardial infarction during the surgery. Infarcts of the size typically observed in this model are associated with a high peri-operative mortality rate. The challenges of successfully producing and using this model in a mobile laboratory in a field setting are substantial. Given these challenges, it is disappointing, but not entirely unexpected, that we would experience more complications at some sites than others.
The results from Power Plant 2 may suggest that in this susceptible animal model, emissions from coal-fired power plants that are photochemically aged in the presence of pinene, a naturally occurring pollutant, can increase ventricular ectopy. If causal, this effect may be mediated by changes in autonomic function, as suggested by a number of toxicologic and epidemiologic studies of ambient particles (Godleski et al., 2000
; Gold et al., 2000
; Devlin et al., 2003
). However, in this study we did not observe any substantial or statistically significant changes in heart rate or heart rate variability. This could be either because the POS scenario had no effect on autonomic nervous system function or because such changes were difficult to observe in this animal model which already has reduced heart rate variability. Alternatively, the increased arrhythmia frequency could reflect increased oxidative stress. In normal animals, the POS exposure at Power Plant 2 led to increased cardiac oxidative stress as measured by in vivo
cardiac chemiluminescence, although this difference did not reach statistical significance (Lemos et al., 2010
We have previously used this animal model to evaluate the effects of concentrated ambient particles (CAPs) and found that CAPs exposure was associated with a non-significant 64.2%, (95% CI: –17.7, 227.6%; p
= 0.16) increase in arrhythmia frequency during a 1 hr exposure (Wellenius et al., 2004
). We have also shown that a 1 hr inhalation exposure to residual oil fly ash significantly increased arrhythmia frequency among animals with preexisting arrhythmias (Wellenius et al., 2002
). In the current study, at Power Plant 2 the frequency of ventricular premature beats peaked during the 4th
exposure hour where it was almost three fold higher among animals exposed to POS scenario as compared to filtered air controls. Because of differences in the duration of exposure and experimental design, direct comparison of these results to those of previous studies is not possible.
At Power Plant 2, the respiratory effects of POS exposure were quite different in normal animals as compared to animals with MI. Specifically, in normal animals POS exposure led to statistically significant decreases in tidal volume, pause, and Penh (Diaz et al., 2010
). In contrast, in the current study in animals with MI, POS exposure was associated with decreased expiratory time and end-inspiratory pause. Whether these divergent results are due to differences in the animal models or observed differences in the POS exposure atmosphere is not estimable from the existing data. Alternatively, given the small magnitude of the effects and the large number of respiratory outcomes examined, the changes observed in respiratory parameters in the MI animals may reflect chance findings.
This study has several potential limitations that warrant discussion. First, for unknown reasons we observed unexpectedly high levels of iron, chromium, and nickel at Power Plant 2 on three of the four exposure days. The source of these trace elements is unclear, but assessment of these particles by single particle analyses using scanning electron microscopy and energy dispersive X-ray analyses suggested they were derived from the emissions of the plant rather than contamination by corrosion of the sampling line from stack (Kang et al., 2010
). How (or if) the elevated concentrations may have affected the measured outcomes is unknown, although there is recent evidence to suggest that some of these elements may play a role in cardiovascular effects (Chen and Lippmann, 2009
). However, the results were not materially different when we excluded from analyses the 2 days with the highest levels of these metals. A second limitation is that the duration of exposure was limited to 5 h. Thus, it is not known whether a longer exposure would produce similar results. Third, to reduce biologic variability, only mature, male, Sprague-Dawley rats were studied. Thus, it is unknown how the effect of exposure to a POS scenario might vary by gender or age. Fourth, even at Power Plant 2 there were some differences in infarct size and location across the exposure groups. It is unclear how these differences might have affected the results. Fifth, there are important differences between the rat and human heart, including differences in the degree of collateral blood flow, ventricular mass, and electrical properties (Janse et al., 1998
) which make extrapolation of these findings to human populations difficult.. Finally, our conclusions are based on results from 4 exposure days at a single power plant. Thus, we were unable to evaluate how variations in power plant and coal characteristics may affect these results.
The overall goal of the TERESA project is to investigate the adverse health effects of specific emission sources and components by examining the relative toxicity of coal combustion and mobile sources (gasoline and/or diesel engine) emissions and their oxidative products. The first phase of the project was to evaluate the health effects of emissions from coal combustion. The second phase of the project evaluating the health effects of emissions from mobile sources is currently in progress. It will be important to compare the relative toxicity of these two important sources of ambient fine PM.