Increased Delay Between Gadolinium Chelate Administration and T1-Weighted Magnetic Resonance Imaging Acquisition Increases Contrast-Enhancing Tumor Volumes and T1 Intensities in Brain Tumor Patients

Abstract

The aim of this study was to evaluate the impact of delayed T1-weighted (T1-w) MRI acquisition after gadolinium chelate administration on brain tumor volumes and T1-w intensities. Fifty-five patients with histologically confirmed, contrast-enhancing intra-axial brain tumors were analyzed in this prospective test-retest study. Patients underwent 2 consecutive 3 T MRI scans (separated by a 1-minute break) during routine follow-up with contrast-enhanced T1 (ceT1-w), T2, and FLAIR acquisition. Macrocyclic gadolinium chelate-based contrast agent was only administered before the first ceT1-w acquisition; median latency to ceT1-w acquisition was 6.72 minutes (IQR, 6.53-6.92) in the first and 16.27 minutes (IQR, 15.49-17.26) in the second scan. Changes in tumor volumes and relative ceT1-w intensities between the 2 acquisitions were quantitatively assessed following semiautomated tumor segmentation (separately for contrast-enhancement [CE], necrosis [NEC], and nonenhancing [NE] tumor). Semiautomatically segmented CE tumor volumes were significantly larger in the second acquisition (median +32% [1.2 cm]; IQR, 16%-62%; P < 0.01), which corresponded to a 10% increase in CE tumor diameter (+0.3 cm). Contrarily, NEC and NE tumor volumes were significantly smaller (median -24% [IQR, -36% to -54%], P < 0.01 for NEC and -2% [IQR, -1% to -3%], P = 0.02 for NE tumor). Bland-Altman plots confirmed a proportional bias toward higher CE and lower NEC volumes for the second ceT1-w acquisition. Relative ceT1-w intensities for both early- (regions already enhancing in the first scan) and late-enhancing (newly enhancing regions in the second scan) tumor were significantly increased in the second acquisition (by 5.8% and 27.3% [P < 0.01, respectively]). Linear-mixed effects modeling confirmed that the increase in CE volumes and CE intensities is a function of the interval between contrast agent injection and ceT1-w acquisition (P < 0.01 each). Our study indicates that the maximum extent of CE tumor volumes and intensities may increase beyond the time frame of 4 to 8 minutes after contrast agent injection and potentially affects the diagnosis of progressive or recurrent disease because late-enhancing recurrent disease might not be unequivocally detected on standard follow-up MRI.