Effect of the addition of 123 I-iomazenil single-photon emission computed tomography to brain perfusion single-photon emission computed tomography on the detection accuracy of misery perfusion in adult patients with ischemic moyamoya disease.

OBJECTIVE: The present prospective study aimed to determine whether the addition of 123 I-iomazenil (IMZ) single-photon emission computed tomography (SPECT) to brain perfusion SPECT could improve the detection accuracy of misery perfusion on positron emission tomography (PET) in adult patients with ischemic moyamoya disease (MMD).

METHODS: Oxygen extraction fraction (OEF) and brain perfusion were assessed using 15 O gas PET and N-isopropyl-p-[123 I]-iodoamphetamine (IMP) SPECT, respectively, in 137 patients. IMZ SPECT was also performed. Regions of interest (ROIs) were automatically placed in the five middle cerebral artery (MCA) territories ipsilateral to the symptomatic cerebral hemisphere and in the contralateral posterior cerebral artery territory using a three-dimensional stereotaxic ROI template. The radioactive count of the MCA ROI to the contralateral posterior cerebral artery ROI was calculated on IMP SPECT (relative SPECT-IMP uptake) and IMZ SPECT (relative SPECT-IMZ uptake). The relative SPECT-IMZ uptake to the relative SPECT-IMP uptake was also calculated (relative SPECT-IMZ/IMP uptake). Of the five MCA ROIs in the symptomatic cerebral hemisphere in each patient, the ROI with the highest PET-OEF value (one ROI per patient) was selected for analysis.

RESULTS: Significant correlations were observed between the PET-OEF and relative SPECT-IMP uptake (correlation coefficient, - 0.683) and relative SPECT-IMZ/IMP uptake (correlation coefficient, 0.875). The area under the receiver operating characteristic curve for detecting misery perfusion (PET-OEF > 51.3%) was significantly greater for the relative SPECT-IMZ/IMP uptake than for the relative SPECT-IMP uptake (difference between areas, 0.080; p = 0.0004). The sensitivity, specificity, and positive- and negative-predictive values for the relative SPECT-IMZ/IMP uptake for detecting misery perfusion were 100%, 92%, 81%, and 100%, respectively. The specificity and positive-predictive value were significantly greater for the relative SPECT-IMZ/IMP uptake than for the relative SPECT-IMP uptake.

CONCLUSIONS: The addition of IMZ SPECT to brain perfusion SPECT improves the detection accuracy of misery perfusion on PET in adult patients with ischemic MMD.