External validation of the BEST-J score and a new risk prediction model for ESD delayed bleeding in patients with early gastric cancer

Table 5 Univariate and multivariate logistic regression analysis of delayed bleeding in patients with EGC

	Univariate logistic regression				Multivariate logistic regression
	OR value	95% CI	P value		OR value		95% CI	P value
Smoking
Yes	2.221	1.006–4.906		0.048	2.564	1.103–5.963			0.029*
No	1				1
Tumor size
> 20 mm	2.545	1.162–5.571		0.019	2.630	1.148–6.029			0.022*
≤ 20 mm	1				1
Intraoperative use of hemoclips
Yes	0.303	0.103–0.893		0.030	0.282	0.093–0.854			0.025*
No	1				1
Horizontal margin
Negative	3.598	0.968–13.378		0.056	NA	NA			NA
Positive	1
Multiple tumors
Yes	3.283	1.153–9.348		0.026	3.621	1.161–11.291			0.027*
No	1				1
Absolute indication
Yes	0.498	0.227–1.091		0.082	NA	NA			NA
No	1
Prolongation of APTT
Yes	6.419	1.897–21.722		0.003	4.923	1.282–18.907			0.020*
No	1				1

APTT, activated partial thromboplastin time; CI, confidence interval; EGC, early gastric cancer; NA means that this factor was not included in the multivariate logistic regression analysis; OR, odds ratio
*P < 0.05

ISSN: 1471-230X