Multi-parametric (mp) MRI for the diagnosis of abdominal wall desmoid tumors

Desmoid tumors are benign myofibroblastic neoplasms, originating from the muscle aponeurosis and classified as deep fibromatoses. The aim of this study was to evaluate the utility of multi-parametric (mp)-MRI for the diagnosis of abdominal wall desmoid tumor (awdt).

This Institutional review board approved retrospective study compared 10 patients (mean age±SD; 38.2±13years; 9 females and 1 male) with awdt to 14 subjects (mean age±SD; 45.6±14.7years; 9 females and 5 males) with non-desmoid abdominal wall tumors (ndawt). All included subjects underwent mp-MRI, which included conventional, diffusion weighted and dynamic contrast-enhanced (DCE) MRI. Two blinded experienced fellowship trained radiologists (MK and SR) evaluated each lesion characteristics qualitatively and quantitatively which included margin, homogeneity, T2W/T1W signal intensity (SI), T2 dark strands, and fascial tail together with measurements of apparent diffusion coefficient (ADC) and semi-quantitative DCE analysis. Inter-observer agreement was assessed using Cohen's kappa and data were compared between groups using independent sample t-tests and Chi-square tests.

No significant differences in age or gender appeared between groups. On qualitative analysis, T2 dark strands were identified in 90% by both radiologist (K=0.82) of awdt, while fascial tail was identified in 70% by radiologist 1 and 80% by second radiologist (k=0.91) of awdt; however no other lesions showed these findings. Other subjective imaging findings did not significantly differ between groups with moderate-to-strong agreements (k=0.7-1.0). On quantitative measurements, diffusion imaging awdt lesions showed higher mean ADC value compared to other lesions, although it did not reached at the level of significance. While on DCE MRI, all awdt lesions showed type 1 (progressive) DCE curve, however no significant difference was observed between groups.

T2 dark strands and fascial tail are characteristic features of awdt, whereas other subjective/qualitative findings are not useful. Quantitative findings such as ADC measurements and DCE curve analysis may have additional value to differentiate awdt from ndawt, but will require further analysis.