Abstract
AIM: To evaluate the ability of magnetic resonance imaging (MRI) of the chest to detect malignant pulmonary nodules compared to compute tomography (CT).
MATERIALS AND METHODS: We searched the following databases with the final date of search on April 7th, 2021: PubMed, Google Scholar. We selected studies according to the inclusion and exclusion criteria that assessed the detection of malignant lung nodules by MRI and CT and included information about sensitivity and specificity. Method of the analysis and data grouping was chosen with regard to statistical heterogeneity of the studies included in the analysis. We used the 2 test and I2 statistic to evaluate the heterogeneity.
RESULTS: We selected 168 articles for the systematic review from the PubMed and Google Scholar databases. We included 21 studies on 1,188 patients in the meta-analysis and revealed statistically significant heterogeneity (р0,00001 for 2 test; I2=99%) for sensitivity and specificity. Hence, we used a random-effect model for further analysis. As a result, values of sensitivity for detection of pulmonary nodules with MRI of 70.4%100%, specificity ― from 60.6% to 100%.
CONCLUSIONS: Thus, MRI has sufficient sensitivity and specificity for detecting malignant pulmonary nodules primarily discovered with CT.