Risk factors for delirium after on-pump cardiac surgery: a systematic review

Table 2 Best-evidence synthesis of variables associated with the occurrence of delirium reported more than once in multivariable or more than four times in univariable analysis

Variables	Multivariable analysis /randomized				Univariable analysis	Level of evidence
Variables	HQ and positive association	positive association	negative association	no association	HQ and no association	Level of evidence
Predisposing variables
Patient characteristics
Age	[1, 45–50]	[27, 31, 37, 41, 51, 52]		[24, 36]	[53]	Strong
Gender				[41, 47, 49]	[1, 24, 36, 45, 48, 53, 54]	No association
Education				[36, 48]	[24]	No association
Nicotine use	[50]	[37]		[53]	[45, 49]	Inconclusive
Chronic pathology
Cardiac disease/NYHA class				[1, 46]	[24, 36, 45, 49]	No association
Hypertension	[1]	[37]		[46, 49]	[24, 36]	Inconclusive
Peripheral vascular disease/atherosclerosis	[49]	[29]		[1, 46]	[36]	Inconclusive
Cerebrovascular disease	[47, 48]			[46]	[24, 36, 49]	Strong
Diabetes mellitus		[55]		[1, 46]	[36, 45, 49]	Inconclusive
Psychiatric impairment	[1, 24, 36]	[35, 51, 52]		[53]		Strong
Risk scores	[26]	[31]		[33, 41, 45, 48, 53]	[24, 36]	Inconclusive
Cognitive functioning	[24, 36, 48, 56]	[32, 34, 51, 57]			[53]	Strong
Preoperative diagnostics
Peripheral oxygen saturation			[55]			Inconclusive
Cerebral oxygen saturation			[48]			Inconclusive
Lower hemoglobin		[35, 51, 57]		[1, 24, 48]		Inconclusive
Renal dysfunction	[1]	[39]		[24, 36]		Inconclusive

APACHE acute physiology and chronic health evaluation, ACC aortic cross-clamping, ASA American Society of Anaesthesiologists, CPB cardiopulmonary bypass, HQ high quality, + positive, − negative association, NYHA New York Heart Association

ISSN: 1364-8535