The value of restriction spectrum imaging in predicting lymph node metastases in rectal cancer: a comparative study with diffusion-weighted imaging and diffusion kurtosis imaging

Huijia Yin; Wenling Liu; Qin Xue; Chen Song; Jipeng Ren; Ziqiang Li; Dongdong Wang; Kaiyu Wang; Dongming Han; Ruifang Yan

doi:10.1186/s13244-024-01852-z

The value of restriction spectrum imaging in predicting lymph node metastases in rectal cancer: a comparative study with diffusion-weighted imaging and diffusion kurtosis imaging

Insights Imaging. 2024 Dec 19;15(1):302. doi: 10.1186/s13244-024-01852-z.

Authors

Huijia Yin¹, Wenling Liu¹, Qin Xue¹, Chen Song², Jipeng Ren¹, Ziqiang Li¹, Dongdong Wang³, Kaiyu Wang⁴, Dongming Han⁵, Ruifang Yan⁶

Affiliations

¹ Department of MR, The First Affiliated Hospital, Xinxiang Medical University, Weihui, China.
² Hematology Laboratory, The First Affiliated Hospital of Xinxiang Medical University, Xinxiang, China.
³ Department of Radiology, People's Hospital of Zhengzhou, Zhengzhou, 450000, PR China.
⁴ MR Research China, GE Healthcare, Beijing, China.
⁵ Department of MR, The First Affiliated Hospital, Xinxiang Medical University, Weihui, China. 625492590@qq.com.
⁶ Department of MR, The First Affiliated Hospital, Xinxiang Medical University, Weihui, China. yrf718@163.com.

Abstract

Background: To investigate the efficacy of three-compartment restriction spectrum imaging (RSI), diffusion kurtosis imaging (DKI), and diffusion-weighted imaging (DWI) in the assessment of lymph node metastases (LNM) in rectal cancer.

Methods: A total of 77 patients with rectal cancer who underwent pelvic MRI were enrolled. RSI-derived parameters (f₁, f₂, and f₃), DKI-derived parameters (D_app and K_app), and the DWI-derived parameter (ADC) were calculated and compared using a Mann-Whitney U test or independent samples t-test. Logistic regression (LR) analysis was used to identify independent predictors of LNM status. Area under the receiver operating characteristic curve (AUC) and Delong analysis were performed to assess the diagnostic performance of each parameter.

Results: The LNM-positive group exhibited significantly higher f₁ and K_app levels and significantly lower f₃, D_app, and ADC levels compared to the LNM-negative group (p < 0.05). There was no difference in f₂ levels between the two groups (p = 0.783). LR analysis showed that D_app and K_app were independent predictors of a positive LNM status. AUC and Delong analysis showed that DKI (D_app + K_app) exhibited significantly higher diagnostic efficacy (AUC = 0.908; sensitivity = 87.10%; specificity = 86.96%) than RSI (f₁ + f₃) and DWI (ADC), with AUCs were 0.842 and 0.771 (Z = 2.113, 3.453; p = 0.035, < 0.001, respectively). The AUC performance between RSI and DWI was also statistically significant (Z = 1.972, p = 0.049).

Conclusion: The RSI model is superior to conventional DWI but inferior to DKI in differentiation between LNM-positive and LNM-negative rectal cancers. Further study is needed before it could serve as a promising biomarker for guiding effective treatment strategies.

Critical relevance statement: The three-compartment restriction spectrum imaging was able to differentiate between LNM-positive and LNM-negative rectal cancers with high accuracy, which has the potential to serve as a promising biomarker that could guide treatment strategies.

Key points: Three-compartment restriction spectrum imaging could differentiate lymph node metastases in rectal cancer. Diffusion kurtosis imaging and diffusion-weighted were associated with lymph node metastases in rectal cancer. The combination of different parameters has the potential to serve as a promising biomarker.

Keywords: Diffusion kurtosis imaging; Diffusion-weighted imaging; Lymph node metastases; Rectal cancer; Restrictive spectrum imaging.

Abstract

Grants and funding