Objective quantification of contrast enhancement of unruptured intracranial aneurysms: a high-resolution vessel wall imaging validation study

Abstract

Objective: High-resolution vessel wall imaging (HR-VWI) has emerged as a valuable tool in assessing unruptured intracranial aneurysms (UIAs). There is no standardized method to quantify contrast enhancement of the aneurysm wall. Contrast enhancement can be objectively measured as signal intensity (SI) or subjectively adjudicated. In this study, the authors compared the different methods to quantify wall enhancement of UIAs and determined the sensitivity and specificity of each method as a surrogate of aneurysm instability. They also compared SI quantification between scanners from different manufacturers.

Methods: The University of Iowa HR-VWI Project database was analyzed. This database compiles patients with UIAs who prospectively underwent HR-VWI using a 3T MRI scanner. The mean and maximal SI values of the aneurysm wall, pituitary stalk, and genu of the corpus callosum were used to compare 3 different measurement methods: 1) aneurysm enhancement ratio AER = (SIwall post - SIwall pre)/SIwall pre; 2) aneurysm-to-pituitary stalk contrast ratio CRstalk = SIwall post/SIstalk post; and 3) aneurysm enhancement index AEI = ([SIwall post/SIbrain post] - [SIwall pre/SIbrain pre])/(SIwall pre/SIbrain pre) (where "pre" indicates precontrast images and "post" indicates postcontrast images). Size ≥ 7 mm was used as a surrogate of aneurysm instability for receiver operating characteristic (ROC) curve analysis. To determine if the objective quantification of SI varies among scanners from different manufacturers, 9 UIAs underwent the same HR-VWI protocol using a 3T General Electric (GE) scanner and a 3T Siemens scanner. Three UIAs also underwent a third scanning procedure on a unit with a different magnet strength (7T GE).

Results: Eighty patients with 102 UIAs were included in the study. The mean age was 64.5 ± 12.2 years, and 64 (80%) patients were women. UIAs ≥ 7 mm had significantly higher SIs than smaller UIAs (< 7 mm): AER = 0.82 vs 0.49, p < 0.001; CRstalk = 0.84 vs 0.61, p < 0.001; and AEI = 0.81 vs 0.48, p < 0.001. ROC curves demonstrated optimal sensitivity of 81.5% for CRstalk ≥ 0.60, 75.9% for AEI ≥ 0.50, and 74.1% for AER ≥ 0.49. Intermanufacturer correlation between 3T GE and 3T Siemens MRI scanners for CRstalk using mean and maximal SI values was excellent (Pearson coefficients > 0.80, p < 0.001). A similar correlation was identified among the 3 UIAs that underwent 7T imaging.

Conclusions: CRstalk using maximal SI values was the most reliable objective method to quantify enhancement of UIAs on HR-VWI. The same ratios were obtained between different manufacturers and on scans obtained using magnets of different strengths.

Keywords: aneurysm; aneurysmal subarachnoid hemorrhage; circumferential enhancement; high-resolution vessel wall imaging; magnetic resonance imaging; validation; vascular disorders.

Figure 1.

Co-registration of CE-MRA, pre- and post-contrast T1-weighted sequences in a basilar tip UIA using (A) axial, (B) coronal and (C) sagittal projections.

Figure 2.

Post-contrast T1-weighted sagittal projection of a basilar tip aneurysm showing the ROI of aneurysm wall (yellow), sampling of the genu of the corpus callosum (red) and of the pituitary stalk (black) for normalization. In this case, the maximal CR_stalk would be calculated as follows: 446 / 700 = 0.63. AR: area; AV: average; SD: standard deviation.

Figure 3.

ROC curves for AER, CR_stalk and AEI using (A) maximal SI values and (B) mean SI values.