Different Imaging Modalities for the Diagnosis of Axillary Lymph Node Metastases in Breast Cancer: A Systematic Review and Network Meta‐Analysis of Diagnostic Test Accuracy

Zhifan Li,Ya Gao,Hengxin Gong,Wen Feng,Qinqin Ma,Jinkui Li,Xingru Lu,Xiaohui Wang,Junqiang Lei
DOI: https://doi.org/10.1002/jmri.28399
IF: 4.4
2022-08-31
Journal of Magnetic Resonance Imaging
Abstract:Background Accurate diagnosis of axillary lymph node metastasis (ALNM) of breast cancer patients is important to guide local and systemic treatment. Purpose To evaluate the diagnostic performance of different imaging modalities for ALNM in patients with breast cancer. Study Type Systematic review and network meta‐analysis (NMA). Subjects Sixty‐one original articles with 8011 participants. Field Strength 1.5 T and 3.0 T. Assessment We used the QUADAS‐2 and QUADAS‐C tools to assess the risk of bias in eligible studies. The identified articles assessed ultrasonography (US), MRI, mammography, ultrasound elastography (UE), PET, CT, PET/CT, scintimammography, and PET/MRI. Statistical Analysis We used random‐effects conventional meta‐analyses and Bayesian network meta‐analyses for data analyses. We used sensitivity and specificity, relative sensitivity and specificity, superiority index, and summary receiver operating characteristic curve (SROC) analysis to compare the diagnostic value of different imaging modalities. Results Sixty‐one studies evaluated nine imaging modalities. At patient level, sensitivities of the nine imaging modalities ranged from 0.27 to 0.84 and specificities ranged from 0.84 to 0.95. Patient‐based NMA showed that UE had the highest superiority index (5.95) with the highest relative sensitivity of 1.13 (95% confidence interval [CI]: 0.93–1.29) among all imaging methods when compared to US. At lymph node level, MRI had the highest superiority index (6.91) with highest relative sensitivity of 1.13 (95% CI: 1.01–1.23) and highest relative specificity of 1.11 (95% CI: 0.95–1.23) among all imaging methods when compared to US. SROCs also showed that UE and MRI had the largest area under the curve (AUC) at patient level and lymph node level of 0.92 and 0.94, respectively. Data Conclusion UE and MRI may be superior to other imaging modalities in the diagnosis of ALNM in breast cancer patients at the patient level and the lymph node level, respectively. Further studies are needed to provide high‐quality evidence to validate our findings. Evidence Level 3 Technical Efficacy Stage 2
radiology, nuclear medicine & medical imaging
What problem does this paper attempt to address?