Table 7.

 Summary of ‘bleeding history’ as a risk factor for bleeding in AF patients receiving OACs

Study	n	Type of OACs	Definition	Main findings	OR/HR (95% CI)	P value
Pisters et al., 2010	5333	VKA	Prior major bleeding (ICH, hospitalization, haemoglobin decrease >2 g/L, and/or blood transfusion)	The rate of major haemorrhage was 1.3% in patients without prior major bleeding vs. 14.8% in those with prior major bleeding.	OR 7.51 (3.00–18.78)	<0.001
Fang et al., 2011	9186	VKA	Prior GI haemorrhage	Prevalence of prior GI bleeding in patients with or without major bleeding: 12.1% vs. 6.8%	HR 2.1 (1.5–2.9)	<0.001
Hylek et al., 2014	18 122	Apixaban/VKA	Bleeding history	Rate of ISTH major haemorrhage was 16.5% in patients without bleeding history vs. 25.2% in those with prior bleeding history (apixaban) and 16.4% vs. 22.5% (warfarin).	HR 1.38 (1.17–1.63)	0.002
O'Brien et al., 2015	7411	VKA/dabigatran	Bleeding history	Bleeding history had good ability to identify those who bled vs. not.	HR 1.73 (1.34–2.23)	NR
Šinigoj et al., 2020a	2260	Dabigatran Rivaroxaban Apixaban	Bleeding history	History of bleeding was a significant predictor of major bleeding.	HR 3.32 (1.87–5.90)	<0.001

AF, atrial fibrillation; GI, gastrointestinal; HR, hazard ratio; ICH , intra-cranial haemorrhage; NR, not reported; OACs, oral anticoagulants; OR, odds ratio; VKA, vitamin K antagonists.

^aŠinigoj et al. is restricted to individuals aged 85 and older.