Sensitivity of MRI for Breast Cancer Documented
A meta-analysis from a research team in the Netherlands recently evaluated rates of detection utilizing magnetic resonance imaging (MRI) among women suspected of having breast cancer. The findings were published in the journal Radiology.
The study completed at the University of Medical Center Utrecht examined 44 different studies, pulled from a pool of 251 published between 1985 and 2005. The number of patients participating in the various trials ranged from 14 to 821 women. Cancer prevalence rates ranged between 23 and 84 percent.
Studies considered included those which assessed patients suspected of having breast cancer, rather than a broad screening population. Patients underwent MRI assessment of breast tissue and biopsy.
The purpose of the meta-analysis was to determine the rate of sensitivity and specificity for breast cancer detection using MRI. Findings from the 44 different studies were pooled to estimate these rates with a larger sample.
The pooled, weighted data yielded an estimate of sensitivity of 0.90 and specificity was estimated to be 0.72. Comparing the different studies, the meta-analysis authors noted that specificity of imaging varied more than sensitivity. Additional considerations relevant to interpretation of meta-analysis findings include the range in cancer prevalence and whether two or three criteria used to determine whether or not masses were considered malignant.
Based on the findings gathered during the analysis, the researchers commented that MRI is not yet specific enough to replace the use of other techniques to characterize breast masses, such as pathologic assessment.
January 23, 2008 Related topics: Standards, Imaging, Diagnostic, Radiology