Is the Kaiser score more effective than the Breast Imaging Reporting and Data System (BI-RADS) for detecting breast cancer on magnetic resonance imaging (MRI)?
For a new meta-analysis, recently published in the European Journal of Radiology, researchers reviewed data from 29 studies that included a total of 7,918 patients (8,451 breast lesions).
For malignant breast lesion detection, the Kaiser score offered 95 percent sensitivity and a 94 percent area under the receiver operating characteristic curve (AUC) in comparison to a 97 percent sensitivity rate and an 89 percent AUC for BI-RADS, according to the meta-analysis authors.
However, the researchers found that the Kaiser score offered significantly higher specificity (70 percent) in contrast to BI-RADS (46 percent).
“ … (The Kaiser score) demonstrated superior specificity compared to BI-RADS, potentially leading to a reduction in unnecessary biopsies while maintaining high diagnostic accuracy across various lesion types,” wrote lead study author Saeed Mohammadzadeh, M.D., who is affiliated with the Department of Radiology at Tehran University of Medical Sciences and the Imam Khomeini Hospital in Tehran, Iran, and colleagues.
In comparison to the BI-RADS system, the meta-analysis authors found the Kaiser score offered comparable specificity for mass lesions (69 percent vs. 68 percent) but slightly higher sensitivity (96 percent vs. 93 percent) and AUC (96 percent vs. 91 percent).
“The (Kaiser score) displayed high diagnostic efficacy in identifying malignant breast lesions, maintaining constant accuracy for both mass-like and non-mass-like lesions. This indicates that (the Kaiser score) could serve as an effective scoring system for assessing a broad range of breast MRI data, irrespective of their morphological characteristics,” added Mohammadzadeh and colleagues.
Three Key Takeaways
1. Higher specificity with the Kaiser score. The Kaiser score demonstrated significantly higher specificity (70 percent) compared to BI-RADS (46 percent), potentially reducing unnecessary biopsies while maintaining high diagnostic accuracy.
2. Comparable or better sensitivity and AUC. While BI-RADS had slightly higher sensitivity (97 percent vs. 95 percent), the Kaiser score showed a better area under the curve (AUC) (94 percent vs. 89 percent), indicating strong overall diagnostic performance.
3. Improved reproducibility for non-experienced users. Unlike BI-RADS, which integrates clinical judgment, the Kaiser score is purely imaging-based, making it advantageous for non-experienced users by enhancing diagnostic accuracy and report consistency.
While the BI-RADS system incorporates contextual factors and clinical judgment, the meta-analysis authors pointed out that the Kaiser score is based solely on imaging.
“This method is particularly advantageous for non-experienced users, enhancing diagnostic accuracy and the reproducibility of reports,” maintained Mohammadzadeh and colleagues.
(Editor’s note: For related content, see “Is the Kaiser Score More Effective than BI-RADS for Assessing Contrast-Enhanced Mammography and MRI?,” “Enhancing Lesions on Breast MRI: Can an Updated Kaiser Scoring Model Improve Detection?” and “Surveillance Breast MRI Associated with Lower Risks of Advanced Second Breast Cancers.”)
In regard to limitations with the meta-analysis, the authors acknowledged that 28 of the 29 reviewed studies were retrospective. They also conceded that some studies only involved assessment of the largest lesion in the breast without considerations of multiple tumor involvement.
