Moving Toward a Consensus DSC-MRI Protocol: Validation of a Low-Flip Angle Single-Dose Option as a Reference Standard...

Area of Science:

• Radiology
• Medical Imaging
• Neuroimaging

Background:

• Dynamic susceptibility contrast-MR imaging (DSC-MR) with a 60° flip angle, preload, and postprocessing leakage correction is a standard method for assessing relative cerebral blood volume (rCBV).
• Simulations suggest that a lower flip angle (30°) without preload could yield equivalent rCBV measurements.
• This study aimed to validate this hypothesis in a clinical setting.

Purpose of the Study:

• To compare the accuracy and reliability of DSC-MR imaging with a 30° flip angle and no preload, with leakage correction (30°/P-/C+), against the standard 60° flip angle with preload and leakage correction (60°/P+/C+).
• To evaluate alternative DSC-MR imaging protocols (30°/P-/C- and 60°/P+/C-) for rCBV quantification.
• To assess the temporal signal-to-noise ratio (SNR) for different imaging protocols.

Main Methods:

• Eighty-four patients with brain lesions were enrolled; 43 met inclusion criteria.
• DSC-MR imaging was performed twice per patient at 3T using a single dose of gadobutrol.
• Protocols included: 30° flip angle with no preload (P-) and with preload (P+), and 60° flip angle with preload (P+). Contrast leakage correction (C+ or C-) was applied.
• Standardized relative CBV (srCBV) and normalized relative CBV (nCBV) maps were generated.
• Comparisons were made using Lin's concordance correlation coefficient and Bland-Altman analysis.

Main Results:

• The 30°/P-/C+ protocol demonstrated the closest mean srCBV to the reference 60°/P+/C+ (P = .61) with the highest Lin's CCC (0.96) and lowest Bland-Altman bias (μ = 1.89).
• The 30°/P-/C- and 60°/P+/C- protocols showed significant differences and higher bias compared to the reference standard.
• nCBV and srCBV were statistically equivalent at the 10% level between 30°/P-/C+ and 60°/P+/C+.
• No significant difference in contrast-to-noise ratios was observed across protocols.
• Temporal SNR was not significantly different between 30°/P- and 60°/P+ protocols (P = .06).

Conclusions:

• DSC-MR imaging with a low flip angle (30°) and no preload, combined with leakage correction, provides accurate and equivalent tumor relative cerebral blood volume (rCBV) measurements compared to the standard higher-dose reference method.
• This simplified protocol offers a potentially more efficient and attractive single-dose alternative for DSC-MR imaging.
• The findings support the clinical utility of optimized DSC-MR imaging parameters for reliable rCBV assessment.