Risk Factors for Acute Delirium in Critically Ill Adult Patients: A Systematic Review

Table 2

Studies included in the paper.

Author(s)	Population	Intervention/control	Outcome measures	Results
Afonso et al., 2010 [10]	112 adult postoperative cardiac surgical patients	Patients w/out delirium	RASS and CAM-ICU scores	Increased age (OR = 2.5, C.I. = 1.6–3.9, and 𝑃<0.0001, per 10 years) and increased duration of surgery (OR = 1.3, CI = 1.1–1.5, and 𝑃=0.0002) were the most significant risk factors for postoperative delirium.
Agarwal et al., 2010 [11]	82 adult ventilated burn patients	Patients w/out delirium	CAM-ICU scores	Benzodiazepines were found to be independent risk factors for the development of delirium (OR = 6.8, CI = 3.1–15.0, and 𝑃<0.001). Opiates (𝑃<0.001) and methadone (𝑃=0.02) appeared to have protective effects, being associated with a lower risk of delirium.
Andrejaitiene and Sirvinskas, 2011 [12]	90 patients with postoperative delirium after cardiac surgery on cardiopulmonary bypass	Nil	RASS and CAM-ICU scores	Administering a dose of fentanyl above 1.4 mg increased the possibility of developing severe delirium (OR = 29.4, CI = 4.1–210.3, and 𝑃<0.001). Longer aortic clamping time was also noted as an independent predictor of severe delirium (OR = 8.0, CI = 1.7–37.2, and 𝑃<0.001).; Postoperative delirium prolonged the length of stay in the ICU by 8.4 days.
Angles et al., 2008 [13]	59 patients admitted to the trauma intensive care unit	Patients w/out delirium	CAM-ICU scores	A GCS of 12 or less (12±1.0 versus 15±0.1, 𝑃<0.01), increased blood transfusions (2.8±0.7 versus 0.5±0.3, 𝑃<0.01), and higher multiple organ failure scores (1.2±0.2 versus 0.1±0.1, 𝑃<0.01) were significantly associated with delirium. Subjects with delirium had longer hospital and ICU stays and were more likely to require postdischarge institutionalization.
Bakker et al., 2012 [14]	201 patients who went for cardiac surgery aged 70 years and older	Patients w/out delirium	CAM-ICU scores and MMSE	63 patients developed delirium after cardiac surgery. Lower MMSE scores (OR = 2.32, CI = 1.20–4.46), higher creatinine levels (OR = 1.02, CI = 1.00–1.03), and longer extracorporeal circulation time (OR = 1.01, CI = 1.01–1.02) were independent predictors of delirium.
Balas et al., 2007 [15]	117 SICU patients	Patients w/out delirium	CAM-ICU scores	Older patients admitted to the SICU were at high risk for developing delirium during hospitalization.
Detroyer et al., 2008 [16]	104 patients admitted for elective cardiac surgery	Patients w/out delirium	CAM-ICU scores and DI	Prolonged intubation time (OR = 1.10, CI = 1.05–1.15) and a low intraoperative lowest body temperature (OR = 0.86, CI = 0.74–0.99) were the most significant predictors of delirium.
Eden et al., 1998 [17]	20 elderly patients in a critical care setting	Patients w/out delirium	DSM III and CAM scores	Comorbidity, presence of infection, a blood urea nitrogen/creatinine ratio of 18 or more, and age were the most significant variables, with a sensitivity of 100% and a specificity of 90%.
Guillamondegui et al., 2011 [18]	97 patients with multiple injuries, requiring ICU management	Patients w/out delirium	CAM-ICU scores	55 of 97 ICU patients were CAM-ICU positive for delirium. Number of ventilator days (OR = 1.16, CI = 1.05–1.29) and ED pulse rate (OR = 1.02, CI = 1.00–1.04) were significantly associated with delirium.
Hudetz et al., 2011 [19]	44 patients undergoing elective cardiac surgery aged 55 years or more	Nonsurgical controls and patients undergoing coronary artery bypass graft (CABG) alone	ICDSC scores	A univariate analysis showed that postoperative delirium occurred more frequently in patients undergoing valve surgery with or without CABG as opposed to CABG alone (𝑃=0.01).
Lin et al., 2008 [20]	143 mechanically ventilated patients	Patients w/out delirium	CAM-ICU scores	Hypoalbuminemia (OR = 5.94, CI = 1.23–28.77) and presence of sepsis on admission (OR = 3.65, CI = 1.03–12.9) are significant factors in the development of early onset delirium.
Norkiene et al., 2007 [21]	1367 adult patients undergoing CABG	Patients w/out delirium	DSM IV criteria	Eight factors were independent predictors of delirium, which were age more than 65 years (OR = 3.82, CI = 1.44–10.12), peripheral vascular disease (OR = 2.80, CI = 1.11–7.04), a EuroSCORE (European System for Cardiac Operative Risk Evaluation) more or equal to 5 (OR = 2.46, CI = 1.16–2.51), preoperative intra-arterial blood pressure support (OR = 8.51, CI = 1.81–40.03), blood product usage (OR = 4.59, CI = 2.10–10.06), and postoperative low cardiac output syndrome (OR = 8.04, CI = 1.1–60.6).
Ouimet et al., 2007 [22]	820 ICU patients	Patients w/out delirium	ICDSC and RASS scores	A history of hypertension (OR = 1.88, CI = 1.3–2.6), alcohol use (OR = 2.03, CI = 1.2–3.2), higher APACHE II score (OR = 1.25, CI = 1.03–1.07), and administration of sedative and analgesic drugs were associated with delirium (OR = 3.2, CI = 1.5–6.8).
Pandharipande et al., 2006 [23]	198 mechanically ventilated patients	Patients w/out delirium	RASS and CAM-ICU scores	Lorazepam was an independent risk factor (OR = 1.2, CI = 1.1–1.4) for daily transition to delirium. Midazolam (𝑃=0.09), fentanyl (𝑃=0.09), morphine (𝑃=0.24), and propofol (𝑃=0.18) were not significant, although they were “associated with trends towards significance.”
Pandharipande et al., 2008 [24]	100 surgical and trauma ICU patients requiring mechanical ventilation for >24 hours	Patients w/out delirium	RASS and CAM-ICU scores	Midazolam (OR = 2.75, CI = 1.43–5.26, 𝑃=0.002) was a strong risk factor for transition to delirium. Opiate exposure was inconclusive in that opiates such as fentanyl were a risk factor for delirium in the SICU (𝑃=0.007), but not in the TICU (𝑃=0.936). Opiates such as morphine were linked to a lower risk to delirium (𝑃=0.024).
Peterson et al., 2006 [25]	156 medical intensive care unit (MICU) patients	Younger MICU patients aged lower than 65	RASS and CAM-ICU scores	Patients 65 years and above experienced hypoactive delirium more frequently than younger patients (41.0% versus 21.6%, 𝑃<0.001), and older age was strongly associated with hypoactive delirium (OR = 3.0, CI = 1.7–5.3). Mixed type (hyper-, hypoactive) delirium was the most common (54.9%) amongst other subtypes.
Ranhoff et al., 2006 [26]	401 sub intensive care unit patients 60 years and above	Patients w/out delirium	CAM and MMSE scores	Delirium was found in 29.2% of the patients, of which 13.7% developed delirium in the ICU. Heavy alcohol use (OR = 6.1, CI = 1.8–19.6), polypharmacy (7 or more drugs) (OR = 1.9, CI = 1.1–3.2), and the use of bladder catheter are predictors of delirium (OR = 2.7, CI = 1.4–4.9).
Robinson et al., 2009 [27]	144 patients older than 50 years admitted to postoperative intensive care unit	Patients w/out delirium	RASS and CAM-ICU scores	Several preoperative variables were significantly associated: older age (𝑃<0.001), hypoalbuminemia (𝑃<0.001), impaired functional status (𝑃<0.001), preexisting dementia (𝑃<0.001), and preexisting comorbidities (𝑃<0.001).
Schoen et al., 2011 [28]	231 patients scheduled for elective/urgent cardiac surgery	Patients w/out delirium	RASS, CAM-ICU and MMSE scores	Older age (OR = 4.30, CI = 1.54–12.04, and 𝑃=0.005), lower MMSE scores (OR = 6.50, CI = 1.75–24.13, and 𝑃=0.018), neuropsychiatric disease (OR = 6.22, CI = 2.02–19.16, and 𝑃=0.001), and lower preoperative cerebral oxygen saturation scores (OR = 3.27, CI = 1.14–9.37, and 𝑃=0.027) were independent predictors for postoperative delirium.
Shi et al., 2010 [29]	164 surgical intensive care unit patients after noncardiac surgery	Patients w/out delirium	Nursing Delirium Screening Scale	Predictive factors of delirium were increasing age (OR = 2.646, CI = 1.431–4.890), history of previous stroke (OR = 4.499, CI = 1.228–16.481), high APACHE II score on SICU admission (OR = 1.391, CI = 1.201–1.621), and high serum cortisol level (OR = 3.381, CI = 1.690–6.765) on the first postoperative day.
Taipale et al., 2012 [30]	122 participants requiring nonemergency surgery for coronary artery or valvular heart disease	Patients w/out delirium	RASS and CAM-ICU scores	The prevalence of delirium ranged from 37.7% to 44.3%. For every additional milligram of midazolam administered, patients were 7-8 % more likely to develop delirium (CI: 1.00–1.14, 𝑃=0.06).
Uguz et al., 2010 [1]	212 coronary intensive care unit patients	Patients w/out delirium	DSM IV criteria and DRS	Independent predictors of delirium were advanced age (𝐵=0.10, 𝑃=0.02), higher level of serum potassium at admission (𝐵=1.37, 𝑃=0.04), and experience of cardiac arrest (𝐵=4.85, 𝑃<0.001) during MI.

CAM-ICU: Confusion Assessment Method for the Intensive Care Unit.
DI: delirium index.
DSM: Diagnostic Statistical Manual.
DRS: Delirium Rating Scale.
ICDSC: Intensive Care Delirium Screening Checklist.
MMSE: Mini-Mental Status Examination.
RASS: The Richmond Agitation Sedation Scale.