First results of quantitative ASL perfusion measurements on a whole-body PET/MR system

Ziel/Aim:

In the beginning of 2011, a whole-body PET/MR system (Philips Ingenuity TF) became fully operational for patient investigations at our institute. With both imaging modalities, non-invasive quantitative perfusion measurements in the brain are possible. With PET, perfusion can be measured using [O-15]H2O which is considered as the gold standard but requires arterial blood sampling. Alternatively, perfusion values can be derived from arterial spin labeling (ASL) measurements in MR which has the advantage of being completely non-invasive and free of ionizing radiation. The combination of both imaging devices in a hybrid system allows a direct validation of ASL derived perfusion values against the PET measurements. Here, we present our first results of ASL perfusion measurements using the Ingenuity TF PET/MR.

Methodik/Methods:

We used multiphase EPISTAR ASL with Look-Locker sampling allowing a multiple inversion time image acquisition every 300 ms starting at 50 ms after the labeling. Crusher gradients were employed to remove signal from vascular blood. Two sequences with flip angles of 30° and 10° were performed to allow real flip-angle and T1 estimation. We performed two separated measurements on a healthy volunteer acquiring 4 slices in the brain with an in-plane pixel size of 3 mm by 3 mm and a slice thickness of 6 mm. The 3-parameter model proposed by Gunther et al. (1) was used to generate a parametric perfusion image and to calculate the cerebral blood flow (CBF).

Ergebnisse/Results:

The perfusion-weighted images showed reasonable signal-to-noise ratio and good contrast between gray and white matter. Gray matter structures were clearly delineated. Preliminary quantitative results are as follows: mean (slice average) CBF was 25 ml/100 g/min, gray matter CBF was 40 ml/100 g/min and white matter CBF was 12 ml/100 g/min. The deduced CBF value for gray matter fit the one given in (2) but are significantly lower than published results obtained with [O-15]H2O.

Schlussfolgerungen/Conclusions:

Our first measurements showed good image quality with the possibility of recognizing gray matter structures across the whole imaged area. The CBF quantification yielded values in partial agreement with the literature. Further work on the ASL quantification model needs to be done before validating the ASL results against [O-15]H2O PET measurements can be performed.

Literatur/References:

(1) Gunther, M., Bock, M., Schad, L. R., Magn. Reson. Med., 46:974-984, 2001.
(2) Petersen, E. T., Mouridsen, K., Golay, X., NeuroImage, 49:104-13, 2010