For noninvasive visualization of inflammatory processes by magnetic resonance imaging (MRI) up to now predominantly superparamagnetic iron oxide particles (SPIOs) are used, taking advantage of the high affinity of these species for the monocyte-macrophage system. Since SPIOs lead to a depletion of the MR signal, this approach cannot be applied to the lung, because the anatomical structure of the lung is such that already under control conditions ¹H MR images appear dark. In the present study we examined the feasibility to image pulmonary inflammation with a positive ¹⁹F contrast. For this purpose we used emulsified perfluorocarbons (PFCs) as contrast agent, which are biochemically inert and are phagocytized similar to SPIOs by the reticuloendothelial system.

Experiments were performed at a vertical Bruker 9.4 T NMR spectrometer equipped with an actively shielded 57-mm gradient set and a ¹H/¹⁹F 30-mm birdcage resonator. For induction of acute pneumonia mice were exposed to aerosolized lipopolysaccharide (LPS) in a custom-built cylindrical chamber connected to an air nebulizer. LPS was dissolved in 0.9% saline (100–500 mg/ml), and mice were allowed to inhale LPS for 30 min. After exposure to LPS, 200–500 ml of a perfluoro crown ether (15C5) emulsion (10%, particle size 130 nm) was injected into the tail vein of the mice. In order to monitor the time course of PFC accumulation within the thorax, anatomically corresponding ¹H and ¹⁹F MR images were acquired: FOV 3x3 cm², ¹H: Respiratory-triggered FLASH, slice thickness (ST) 1 mm, matrix 256x256, ¹⁹F: RARE (RARE factor 64), ST 2 mm, matrix 128x128.

One day after induction of pneumonia normal ¹H images gave no evidence at all for the presence of lung injury. However, at the same point in time ¹⁹F images unequivocally showed PFC infiltration into both pulmonary lobes Follow-up imaging after 2 and 4 days revealed alterations in ¹H images at the same location as the early ¹⁹F signals. The occurrence of these ¹H signals within the lung is most likely caused by the development of edema with progressive pneumonia, which could be predicted by ¹⁹F MRI after PFC administration. Histology confirmed the presence of infiltrating monocytes/macrophages already at day 1 after LPS exposition. Additional experiments with stepwise increasing dosage of LPS indicated that severity of pneumonia is reflected by the intensity of the PFC signal (integral) determined in ¹⁹F images.

The results of the present study show that intravenous application of emulsified PFCs after induction of pneumonia results in the accumulation of these particles in inflamed lung areas, most likely after phagocytosis by the monocyte-macrophage system. Thus, PFCs can serve as positive contrast agent for pulmonary inflammation, which is characterized by a high degree of specificity due to the lack of any ¹⁹F background. Since PFCs are not toxic, this approach may also be suitable for human applications.