A proposed inclusion of magnetic resonance imaging features to the VI RADS to enhance its accuracy in predicting muscle invasion

Egyptian Journal of Radiology and Nuclear Medicine

Table 1 Categorical imaging variables showing statistically significant differences between NMIBC and MIBC

Categorical variables with statistically significant differences	State of muscle invasion				χ²	p
	NMI (n = 41)		MI (n = 39)
	No	%	No	%
Multiple lesions	31	75.6	14	35.9	12.809	< 0.001*
Shape
Papillary	37	90.2	19	48.7	16.413	< 0.001*
Sessile	4	9.8	20	51.3	16.413	< 0.001*
Tumor–wall contact length (cm)
≤ 3	25	61.0	5	12.8	19.776	< 0.001*
> 3	16	39.0	34	87.2	19.776	< 0.001*
Distal ureteric backpressure	11	26.8	25	64.1	11.2196	0.001*
VUJ involvement	23	68.3	31	79.5	4.894	0.025*
Fibrovascular stalk
Absent	17	41	30	76.9	10.370	0.001*
Present	24	58.5	9	23.1	10.370	0.001*
Central (inchworm sign)	12	50	0	0	15.985	< 0.001*
Eccentric/distorted	12	50	9	100	15.985	< 0.001*
VI RADS SC score
≤ 3	34	82.9	4	10.3	42.327	< 0.001*
> 3	7	17.1	35	89.7	42.327	< 0.001*
VI RADS DW score
≤ 3	34	82.9	4	10.3	42.327	< 0.001*
> 3	7	17.1	35	89.7	42.327	< 0.001*
VI RADS CE score
≤ 3	33	80.5	4	10.3	39.658	< 0.001*
> 3	8	19.5	35	89.7	39.658	< 0.001*
VI RADS final score
≤ 3	33	80.5	4	10.3	39.658	< 0.001*
> 3	8	19.5	35	89.7	39.658	< 0.001*

NMI non-muscle invasive, MI muscle invasive, ADC apparent diffusion coefficient, VUJ vesicoureteric junction, χ² Chi-square test, *p value statistically significant at < 0.05