PET imaging of regional 18F-FDG uptake and lung function after cigarette smoke inhalation.

Cigarette smoke is thought to promote local lung inflammation that leads to lung dysfunction. Lung neutrophilic inflammation is known to result in increased pulmonary uptake of (18)F-FDG. Using a sheep model of localized exposure to cigarette smoke, in this study we tested whether PET-imaged changes in regional intrapulmonary distribution of (18)F-FDG uptake are related to changes in regional lung function as assessed with the infused (13)NN-saline method. METHODS: Five anesthetized, mechanically ventilated sheep were exposed to unilateral inhalation of smoke from 10 tobacco cigarettes while the contralateral lung was ventilated with smoke-free gas. Two hours after the exposure, regional gas content was measured from a transmission scan; regional ventilation, perfusion, and shunt were measured from the kinetics of (13)NN-saline; and regional (18)F-FDG influx constant (K(i)) was calculated with the Patlak algorithm applied at a voxel-by-voxel level. RESULTS: K(i) was higher and more heterogeneous in the smoke-exposed lungs than in the control lungs (P < 0.05). Spatial heterogeneity of K(i) and impairment in regional lung function were quite variable among animals despite similar levels of smoke exposure. However, increases in mean K(i) correlated linearly with its spatial heterogeneity (Spearman correlation, r(s) = 0.94), and the highest levels of regional K(i) in smoke-exposed lungs and control lungs correlated with regional shunt fraction (r(s) = 0.78). Also, the heterogeneity of the ventilation-perfusion (V/Q) distribution of the smoke-exposed lungs was 10 times greater than that of the control lungs but correlated strongly with that of the control lungs (r = 0.998). CONCLUSION: Substantial interanimal variability and spatial heterogeneity in lung function and (18)F-FDG uptake seem to characterize the response to smoke exposure. The highest levels of local (18)F-FDG uptake were associated with differences in V/Q matching and shunt fraction among animals. The data also suggest that preexisting heterogeneity in V/Q could have been responsible for the large interanimal variability by affecting the heterogeneity and strength of the acute response to smoke inhalation.