Diagnostic Performance of Diffusion MRI for differentiating Benign and Malignant Nonfatty Musculoskeletal Soft Tissue Tumors: A Systematic Review and Meta-analysis

Qian Wang; Xinguang Xiao; Yanchang Liang; Hao Wen; Xiaopeng Wen; Meilan Gu; Cuiping Ren; Kunbin Li; Liangwen Yu; Liming Lu

doi:10.7150/jca.62131

Diagnostic Performance of Diffusion MRI for differentiating Benign and Malignant Nonfatty Musculoskeletal Soft Tissue Tumors: A Systematic Review and Meta-analysis

J Cancer. 2021 Oct 28;12(24):7399-7412. doi: 10.7150/jca.62131. eCollection 2021.

Authors

Qian Wang¹, Xinguang Xiao¹, Yanchang Liang², Hao Wen³, Xiaopeng Wen⁴, Meilan Gu¹, Cuiping Ren⁵, Kunbin Li³, Liangwen Yu², Liming Lu⁶

Affiliations

¹ Department of Medical Imaging, Zhengzhou Central Hospital Affiliated to Zhengzhou University, 195 Tongbai Road, 455007, Zhengzhou, China.
² Guangzhou University of Chinese Medicine, 510006, Guangzhou, China.
³ Department of Neurology, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, 107 Yanjiang West Road, Guangzhou, China.
⁴ Department of neurological rehabilitation, Zhengzhou Central Hospital Affiliated to Zhengzhou University, 450000, Zhengzhou, China.
⁵ Department of Medical Imaging, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China.
⁶ Clinical Research and Data Center, South China Research Center for Acupuncture and Moxibustion, Medical College of Acu-Moxi and Rehabilitation, Guangzhou University of Chinese Medicine, Guangzhou, China.

Abstract

Objective: To evaluate the diagnostic performance of standard diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), and diffusion kurtosis imaging (DKI), for differentiating benign and malignant soft tissue tumors (STTs). Materials and methods: A thorough search was carried out to identify suitable studies published up to September 2020. The quality of the studies involved was evaluated using Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2). The pooled sensitivity (SEN), specificity (SPE), and summary receiver operating characteristic (SROC) curve were calculated using bivariate mixed effects models. A subgroup analysis was also performed to explore the heterogeneity. Results: Eighteen studies investigating 1319 patients with musculoskeletal STTs (malignant, n=623; benign, n=696) were enrolled. Thirteen standard DWI studies using the apparent diffusion coefficient (ADC) showed that the pooled SEN and SPE of ADC were 0.80 (95% CI: 0.77-0.82) and 0.63 (95% CI: 0.60-0.67), respectively. The area under the curve (AUC) calculated from the SROC curve was 0.806. The subgroup analysis indicated that the percentage of myxoid malignant tumors, magnet strength, study design, and ROI placement were significant factors affecting heterogeneity. Four IVIM studies showed that the AUCs calculated from the SROC curves of the parameters ADC and D were 0.859 and 0.874, respectively. The AUCs for the IVIM parameters pseudo diffusion coefficient (D*) and perfusion fraction (f) calculated from the SROC curve were 0.736 and 0.573, respectively. Two DKI studies showed that the AUCs of the DKI parameter mean kurtosis (MK) were 0.97 and 0.89, respectively. Conclusion: The DWI-derived ADC value and the IVIM DWI-derived D value might be accurate tools for discriminating musculoskeletal STTs, especially for non-myxoid SSTs, using more than two b values, with maximal b value ranging from 600 to 800 s/mm², additionally, a high-field strength (3.0 T) optimizes the diagnostic performance.

Keywords: diffusion weighted imaging; meta-analysis; musculoskeletal; soft tissue tumors.