Added value of advanced over conventional magnetic resonance imaging in grading gliomas and other primary brain tumors

Table 5 Variables with independent predictive value obtained from combining different magnetic resonance techniques (multivariate logistic regression) in gliomas

Technique	N	Predictor variables	OR	95% CI	P	Coefficient (β)	Standard error	AUC	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)
MRI	118	Enhancement	58.82	9.35-333.33	<0.0001	4.06	0.93	0.940	97.8	76.9	93.7	90.9
MRI	118	Necrosis	13.89	2.91-66.67	0.001	2,63	0.80	0.940	97.8	76.9	93.7	90.9
MRI	67	Enhancement	25	3.60-166.67	0.001	2.46	0.99	0.940	96.2	64.3	91.1	81.8
PWI		Enhancement	25	3.60-166.67	0.001	2.46	0.99
DWI		Necrosis	8.26	1.53-43.48	0.014	2.11	0.86
MRS		Necrosis	8.26	1.53-43.48	0.014	2.11	0.86
MRI	110	Enhancement	50	8.13-333.33	<0.0001	3.94	0.94	0.940	97.6	76	93.3	90.5
PWI
DWI
DWI		Necrosis	13.89	2.91- 66.67	0.001	2.64	0.80

AUC, area under the receiver operating characteristic curve; CI, confidence interval; DWI, diffusion-weighted imaging; MRI, magnetic resonance imaging; MRS, MR spectroscopy; N, number of cases from which each classifier was constructed; PWI, perfusion-weighted imaging; rADC, relative apparent diffusion coefficient; OR, odds ratio; PPV, positive predictive value; NPV, negative predictive value.

ISSN: 1470-7330