Samenvatting
Introduction:
In breast tumors, a prebolus first pass perfusion method was recently proposed for improved perfusion quantification. This method enables simultaneous accomplishment of an unsaturated AIF curve with the initial prebolus and a tissue curve with adequate SNR with the subsequent high dose. However, an important concern in the prebolus technique in breast tumors
remains the selection of a suitable contrast dose protocol. In this study we investigated the effect of dose reduction in the context of prebolus DCE MRI.
Materials & Methods:
In vivo perfusion measurements were performed in 23 women with breast tumors on a 1.5 T scanner. The routine MR mammography protocol was applied first. In the slice where the tumor enhanced maximally, 10 minutes later prebolus protocol was applied. 1ml of Gd-DTPA solution at 2ml/s was injected at the beginning of a dynamic axial single slice
inversion-prepared (IR prepared) TFE acquisition (800 dynamics at a temporal resolution of 0.3s). At the 400th dynamic, a high dose of either 10ml (8 patients, 5 malignant and 3 benign) or 20ml (15 patients, 11 malignant and 4 benign) of contrast agent was injected at 2ml/s and a further 400 dynamics were acquired. ROIs were placed manually over the central part of the aorta and the breast lesion with highest enhancement. From the prebolus curve, AIFs were reconstructed by time-shifting and adding the prebolus response curve until the relevant high dose equivalent is reached. The RE time course from the
tumor ROI was then deconvolved. Finally, parametric values were calculated from the impulse response function.
Results:
Comparison of the tumor RE curves with 10ml and 20ml in two separate patients with same histopathological diagnosis shows a less noisy time intensity curve with the 20ml bolus. The median values of TBF are presented separately for the malignant and benign tumors for both injection protocols, indicating that the median TBF for both malignant and benign tumors are in a matching range. On the other hand, the 1-10 ml protocol achieves a significant difference in the median TBF between benign and malignant tumors (P<0.05), while the 1-20 ml one does not (P = 0.102).
Conclusion:
In spite of the improved SNR with the 1-20ml injection protocol, the tumor differentiation was only significant with the 1-10ml dose regime. The median TBF values with both contrast regimes of the present study match the previously published PET based perfusion values very well.
In breast tumors, a prebolus first pass perfusion method was recently proposed for improved perfusion quantification. This method enables simultaneous accomplishment of an unsaturated AIF curve with the initial prebolus and a tissue curve with adequate SNR with the subsequent high dose. However, an important concern in the prebolus technique in breast tumors
remains the selection of a suitable contrast dose protocol. In this study we investigated the effect of dose reduction in the context of prebolus DCE MRI.
Materials & Methods:
In vivo perfusion measurements were performed in 23 women with breast tumors on a 1.5 T scanner. The routine MR mammography protocol was applied first. In the slice where the tumor enhanced maximally, 10 minutes later prebolus protocol was applied. 1ml of Gd-DTPA solution at 2ml/s was injected at the beginning of a dynamic axial single slice
inversion-prepared (IR prepared) TFE acquisition (800 dynamics at a temporal resolution of 0.3s). At the 400th dynamic, a high dose of either 10ml (8 patients, 5 malignant and 3 benign) or 20ml (15 patients, 11 malignant and 4 benign) of contrast agent was injected at 2ml/s and a further 400 dynamics were acquired. ROIs were placed manually over the central part of the aorta and the breast lesion with highest enhancement. From the prebolus curve, AIFs were reconstructed by time-shifting and adding the prebolus response curve until the relevant high dose equivalent is reached. The RE time course from the
tumor ROI was then deconvolved. Finally, parametric values were calculated from the impulse response function.
Results:
Comparison of the tumor RE curves with 10ml and 20ml in two separate patients with same histopathological diagnosis shows a less noisy time intensity curve with the 20ml bolus. The median values of TBF are presented separately for the malignant and benign tumors for both injection protocols, indicating that the median TBF for both malignant and benign tumors are in a matching range. On the other hand, the 1-10 ml protocol achieves a significant difference in the median TBF between benign and malignant tumors (P<0.05), while the 1-20 ml one does not (P = 0.102).
Conclusion:
In spite of the improved SNR with the 1-20ml injection protocol, the tumor differentiation was only significant with the 1-10ml dose regime. The median TBF values with both contrast regimes of the present study match the previously published PET based perfusion values very well.