A standardized method of generating time-to-peak perfusion maps in dynamic-susceptibility contrast-enhanced MR imaging.

BACKGROUND AND PURPOSE

Perfusion MR imaging, performed as dynamic-susceptibility contrast-enhanced MR imaging, is sensitive to hemodynamic risks for patients with cerebrovascular disease. We sought to define a quantitative parameter for perfusion MR imaging, which shows brain areas at hemodynamic risk and enables direct comparison of different perfusion MR imaging examinations.

METHODS

A new standardization procedure for the time-to-peak (TTP) parameter, standardized time to peak (stdTTP), was introduced. The stdTTP automatically calculates a time offset correlated to the earliest enhancing voxels in a section and rescales all TTP values accordingly. Because of a close relation between this offset and stdTTP of early enhancing voxels in central vascular territories (CVTs), stdTTP provides an estimate of the bolus run time between CVTs and related border zones (BZs). The stdTTP in CVTs and BZs was measured in 11 patients without hemodynamic impairment by using high temporal resolution dynamic-susceptibility contrast-enhanced perfusion MR imaging.

RESULTS

An excellent comparability of different dynamic susceptibility contrast-enhanced MR imaging studies was found. The stdTTP in CVTs was 0.4 +/- 0.5 s (minimum, 0 s; maximum, 1.3 s) for the anterior, 0.5 +/- 0.3 s (minimum, 0 s; maximum, 1.0 s) for the middle, and 1.4 +/- 0.5 s (minimum, 0.4 s; maximum, 2.4 s) for the posterior cerebral artery. In the anterior BZ, stdTTP was 2.3 +/- 0.4 s (minimum, 1.6 s; maximum, 3.2 s), and in the posterior BZ, stdTTP was 2.8 +/- 0.4 s (minimum, 2.0 s; maximum, 3.4 s).

CONCLUSION

The results suggest a limit for stdTTP of approximately 3.5 s in the anterior and posterior BZs. The stdTTP could serve as a quantitative measure for the hemodynamic risk assessment of patients with cerebrovascular disease. Because stdTTP can be directly derived from the measured curves, the hemodynamic situation of a patient can be judged with a minimum of computational effort.