# Diagnostic Performance of Advanced Diffusion MRI Parameters Versus Conventional Diffusion-Weighted Imaging (DWI) for Detecting Lymph Node Metastasis in Colorectal Cancer: A Systematic Review and Meta-Analysis

## TL;DR

This study compares advanced MRI techniques with traditional imaging to better detect cancer spread to lymph nodes in colorectal cancer patients.

## Contribution

The study shows that advanced diffusion MRI parameters, especially when combined, outperform conventional DWI in detecting lymph node metastasis.

## Key findings

- IVIM-derived D significantly outperformed ADC in detecting lymph node metastasis.
- Multiparametric models combining IVIM and DTI metrics achieved the highest diagnostic accuracy (AUC ≈ 0.95).
- DTI metrics like FA and AD showed high sensitivity and specificity for metastasis detection.

## Abstract

Accurate preoperative identification of lymph node metastasis (LNM) in colorectal cancer is essential for optimal staging and treatment planning. Conventional diffusion-weighted imaging (DWI) is limited by overlap in apparent diffusion coefficient (ADC) values between benign and malignant lymph nodes. Advanced diffusion MRI techniques, including intravoxel incoherent motion (IVIM) parameters (diffusion coefficient (D), pseudodiffusion coefficient (D*), and perfusion fraction (f)) and diffusion tensor imaging (DTI) metrics such as fractional anisotropy (FA), axial diffusivity (AD), mean diffusivity (MD), and radial diffusivity (RD), may improve diagnostic accuracy. This systematic review and meta-analysis evaluated the diagnostic performance of these advanced diffusion parameters compared with conventional DWI. A comprehensive search of PubMed, Google Scholar, ScienceDirect, and the Cochrane Library identified studies assessing diffusion MRI for LNM detection with extractable true-positive, false-positive, false-negative, and true-negative data. Five studies met the inclusion criteria. Risk ratios (RRs) were pooled using RevMan (The Cochrane Collaboration, London, United Kingdom), and bivariate summary receiver operating characteristic (SROC) analysis was performed in R (R Development Core Team, Vienna, Austria). Study quality was assessed using QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies-2). Five studies involving 239 patients and 358 lymph nodes evaluated IVIM, DTI, ADC, and multiparametric models. IVIM-derived D showed significantly superior performance compared with ADC (RR=9.11, 95% CI: 4.86-17.09; I²=0%), while D* also demonstrated strong discriminatory ability (RR=3.12, 95% CI: 2.10-4.65; I²=0%). SROC analysis revealed high diagnostic accuracy for IVIM D (area under the curve (AUC)≈0.93), moderate accuracy for D* (AUC≈0.82), and the highest accuracy for combined parameters (AUC≈0.95). DTI metrics, particularly FA and AD, consistently showed high sensitivity (88-100%) and specificity (85-100%). Overall risk of bias was low. Advanced diffusion MRI parameters outperform conventional DWI for detecting LNM in colorectal cancer, with multiparametric approaches offering the greatest diagnostic value.