Unenhanced CT-based predictive model to identify small bowel necrosis in patients with mechanical small bowel obstruction

Table 4 Logistic regression statistics for general characteristics and imaging data

Variables	Univariate analysis			Multivariate analysis
Variables	P	OR	95% CI	P	OR	95% CI
Gender	0.297	1.497	0.701–3.196
Age	0.056	1.021	0.999–1.044	0.850	1.003	0.973–1.033
BMI	0.327	1.055	0.948–1.174
History of abdominal surgery	< 0.001	0.217	0.098–0.478	0.060	0.353	0.119–1.045
Increased attenuation of small bowel wall	<0.001	6.621	2.980–14.710	0.002	5.778	1.895–17.618
Small bowel wall annular thickening	0.001	3.658	1.697–7.882	0.789	1.165	0.280–3.575
Small bowel lumen dilation	0.816	1.204	0.252–5.760
Maximum diameter of Small bowel lumen dilation	0.921	0.980	0.652–1.471
Mesenteric ground glass sign	0.030	9.461	1.247–71.805	0.718	0.640	0.057–7.180
Mesenteric vascular density	0.118	1.026	0.994–1.059
Difference of CT values between mesenteric vessel and aorta	0.003	0.898	0.836–0.964	0.025	0.899	0.819–0.986
Mesenteric vasodilation	0.754	0.878	0.390–1.978
Diffuse mesenteric haziness	<0.001	10.700	4.331–26.433	0.010	4.649	1.450-14.905
Small bowel fecal gas sign	0.140	1.783	0.828–3.842
U-/C-shaped small bowel loop	< 0.001	5.218	2.387–11.405	0.010	4.105	1.398–12.053
Ascites	<0.001	8.933	2.619–30.476	0.131	3.071	0.716–13.173
Peritoneal thickening	0.002	3.921	1.671–9.201	0.999	0.999	0.218–3.140
Pneumatosis intestinalis	0.079	8.848	0.779-100.509	0.284	5.612	0.239–131.770
Whirlpool sign	0.573	1.365	0.464–4.017

ISSN: 1471-2342