Acute cardiovascular effects of controlled exposure to dilute Petrodiesel and biodiesel exhaust in healthy volunteers: a crossover study

Abstract

Background: Air pollution derived from combustion is associated with considerable cardiorespiratory morbidity and mortality in addition to environmental effects. Replacing petrodiesel with biodiesel may have ecological benefits, but impacts on human health remain unquantified. The objective was to compare acute cardiovascular effects of blended and pure biodiesel exhaust exposure against known adverse effects of petrodiesel exhaust (PDE) exposure in human subjects. In two randomized controlled double-blind crossover studies, healthy volunteers were exposed to PDE or biodiesel exhaust for one hour. In study one, 16 subjects were exposed, on separate occasions, to PDE and 30% rapeseed methyl ester biodiesel blend (RME30) exhaust, aiming at PM10 300 μg/m3. In study two, 19 male subjects were separately exposed to PDE and exhaust from a 100% RME fuel (RME100) using similar engine load and exhaust dilution. Generated exhaust was analyzed for physicochemical composition and oxidative potential. Following exposure, vascular endothelial function was assessed using forearm venous occlusion plethysmography and ex vivo thrombus formation was assessed using a Badimon chamber model of acute arterial injury. Biomarkers of inflammation, platelet activation and fibrinolysis were measured in the blood.

Results: In study 1, PDE and RME30 exposures were at comparable PM levels (314 ± 27 μg/m3; (PM10 ± SD) and 309 ± 30 μg/m3 respectively), whereas in study 2, the PDE exposure concentrations remained similar (310 ± 34 μg/m3), but RME100 levels were lower in PM (165 ± 16 μg/m3) and PAHs, but higher in particle number concentration. Compared to PDE, PM from RME had less oxidative potential. Forearm infusion of the vasodilators acetylcholine, bradykinin, sodium nitroprusside and verapamil resulted in dose-dependent increases in blood flow after all exposures. Vasodilatation and ex vivo thrombus formation were similar following exposure to exhaust from petrodiesel and the two biodiesel formulations (RME30 and RME100). There were no significant differences in blood biomarkers or exhaled nitric oxide levels between exposures.

Conclusions: Despite differences in PM composition and particle reactivity, controlled exposure to biodiesel exhaust was associated with similar cardiovascular effects to PDE. We suggest that the potential adverse health effects of biodiesel fuel emissions should be taken into account when evaluating future fuel policies.

Trial registration: ClinicalTrials.gov, NCT01337882 /NCT01883466. Date of first enrollment March 11, 2011, registered April 19, 2011, i.e. retrospectively registered.

Keywords: Air pollution; Biodiesel; Cardiovascular system; Diesel; Endothelial function; Particulate matter; Thrombosis; Vascular function; Vasomotor dysfunction.

Conflict of interest statement

The authors report no competing financial or other interests.

Figures

Fig. 1

Particle size distribution during PDE and RME100 exhaust exposure. Mean with standard deviation. Exposures were kept constant and measurements (n = 3) were spread out over the exposure series

Fig. 2

Oxidative potential and metal content of exhaust PM derived from petrol diesel and RME blended fuel combustion. Panel a shows superoxide free radical generation using electron paramagnetic resonance (EPR) with the spin-trap Tempone-H (1 mM). All particulates were suspended at an equivalent concentration of 0.1 mg/mL in physiological saline. Pyrogallol (0.1 mM) is used as a positive control to spontaneously generate superoxide. RME100 generated significantly less superoxide than petrodiesel (*p < 0.05) or RME30 (†p < 0.05) (unpaired t-tests, n = 6–10). Panel b represents ascorbate- and glutathione-dependent oxidative potentials (OPAA and OPGSH, respectively) for the PM < 0.2 μm and PM0.2–0.5 μm fractions are illustrated, with the data expressed per μg of extracted PM (n = 3, separate filters, per fraction and fuel type). A total aggregated OP (OPTOT) is also illustrated reflecting the sum of the OPAA and OPGSH measures. Data are illustrated as means with standard deviation, with comparison between groups performed on the sum of the OP for the two fractions combined using the students t-test (P < 0.05): ‘a’ petrol diesel vs, RME30; ‘b’ RME30 vs RME100; no significant differences were observed between petrol diesel and RME100. Panel c represents the concentration of a selection of the measured metals in both PM fractions derived from each fuel type (n = 3). Zn = zinc, Cr = Chromium, V = Vanadium, Mn = Manganese, Cu = Copper, Mo = Molybdenum, Ni = Nickel, Fe = Iron. Asterisks represent significant differences (p < 0.05) in concentration relative to petrodiesel. No significant differences were noted between the fuel types in the PM < 0.2 μm fraction and metal concentrations did not differ between the two fractions under each condition

Fig. 3

Forearm blood flow during intrabrachial infusion of vasoactive drugs 4–6 h post exposure, mL/100 mL tissue. Mean with 95% CI. The graph shows response to incremental doses of acetylcholine, bradykinin, sodium nitroprusside and verapamil following respective exposure. All vasodilators caused an increase in blood flow (p < 0.05 for all) that was similar between exhaust exposures (p > 0.05 for all). P values in the graph for respective vasodilator response in the infused arm following exposure to RME30 exhaust compared to PDE (n = 15) and RME100 exhaust compared to PDE (n = 18), 2-way ANOVA

Fig. 4

Ex vivo thrombosis formation in a model of acute arterial injury. Individual data points for mean thrombus area for each individual, line for group mean with 95% CI. There were no significant differences between RME30 and PDE (study one) or RME100 and PDE (study two) RME30 vs. PDE p = 0.37, n = 13 and RME100 vs. PDE p = 0.48, n = 19. P values from paired Student’s t-test

Fig. 5

Mean plasma t-PA antigen concentrations after exposure to petrodiesel vs. RME 30 (upper left panel) and petrodiesel vs. RME100 (lower left panel) were not significantly different between exposures (p = 0.73 and p = 0.22, respectively). Stimulated release of t-PA plasma antigen concentrations following incremental doses of bradykinin infusions, expressed as nanogram per 100 mL tissue per minute, mean with 95% CI (upper and lower right panels). Bradykinin infusions caused dose dependent increases in t-PA antigen concentrations that were significant at p < 0.01 level for all exposures; petrodiesel and RME 30 (upper right panel) as well as petrodiesel and RME 100 (lower right panel) by 2-way ANOVA. The bradykinin stimulated increases in t-PA did not differ between petrodiesel or RME exposures (p = 0.43 and p = 0.35, respectively).