Applying Diffusion-Weighted Imaging to Detect Metastatic and Non-Metastatic Lymph Nodes in Abdomen: A Meta-Analysis and Systematic Review

The objective of the study was to evaluate the clinical significance of diffusion-weighted imaging (DWI) in detecting abdominal metastatic and non-metastatic lymph nodes. We searched for articles about assessing the diagnostic performance of DWI in detecting metastatic and non-metastatic lymph nodes in abdomen in PUBMED until October 2014. A meta-analysis was used to assess the clinical significance of DWI through pooling sensitivity, specificity, calculating likelihood ratios and constructing summary receiver operating characteristics. A total of 1029 patients and 4620 metastatic and non-metastatic lymph nodes were enrolled in the 18 studies. The threshold effect did not exist (Spearman correlation coefficient = 0.084, P = 0.741). The pooled sensitivity and specificity of DWI were 0.82 (95 % CI 0.74-0.89) and 0.90 (95 % CI 0.82-0.94), respectively. The positive likelihood ratio (PLR), negative likelihood ratio (NLR), and area under the curve (AUC) were 6.20 (95 % CI 4.30-8.95), 0.21 (95 % CI 0.11-0.39), and 0.92 (95 % CI 0.90-0.94), respectively. When we defined the pretest probability as 0.25, 0.46, and 0.75, the corresponding post-probability positive (PPP) and post-probability negative (PPN) were 0.73, 0.87, 0.96, and 0.6, 0.14, 0.37. No publication bias was found (bias = -0.69, P = 0.499). Diffusion-weighted imaging may be an expedient tool for detecting abdominal metastatic lymph nodes and non-metastatic lymph nodes. Larger number and high-quality prospective studies still need to be conducted.