Non-delirious Patients Research Articles

Readmissions and mortality in delirious versus non-delirious octogenarian patients after aortic valve therapy: a prospective cohort study

ObjectivesTo determine whether postoperative delirium predicts first-time readmissions and mortality in octogenarian patients within 180 days after aortic valve therapy with surgical aortic valve replacement (SAVR) or transcatheter aortic valve...

The impact of delirium on healthcare utilization and survival after transcatheter aortic valve replacement.

We assessed whether post-operative delirium is associated with healthcare utilization and overall survival after trans-catheter aortic valve replacement. Delirium, a common syndrome among hospitalized older adults, is associated with increased morbidity and mortality. We reviewed 294 transcatheter aortic valve replacement cases between June 2008 and February 2015 at a tertiary care academic medical center. Post-operative delirium was identified by confusion assessment method screening and clinician diagnosis. Delirium was identified in 61 patients (21%). Non-femoral access for trans-catheter aortic valve replacement was more common in delirious patients than in non-delirious patients (41% vs. 27%, P = 0.04). Delirious patients had diminished overall survival after trans-catheter aortic valve replacement compared to non-delirious patients (1-year survival 59% vs. 84%, log-rank P = 0.002). After adjusting for age, Society of Thoracic Surgeons predicted 30-day mortality, and access type; delirium remained independently associated with diminished overall survival (hazard ratio 2.01, 95% confidence interval 1.21-3.33, P = 0.007). The delirium group had longer mean hospital stay (13.3 ± 9.5 days vs. 6.7 ± 3.8 days, P < 0.001) and a higher rate of discharge to a rehabilitation facility (61% vs. 27%, P < 0.001), but there was no difference in 30-day hospital re-admission rates or 30-day mortality based on delirium status. Delirium occurs in one out of five patients after trans-catheter aortic valve replacement and is associated with diminished survival and increased healthcare utilization. Further studies are needed to clarify whether strategies aimed at reducing delirium after trans-catheter aortic valve replacement may improve outcomes in this high-risk subset. © 2016 Wiley Periodicals, Inc.

Absence of association between whole blood viscosity and delirium after cardiac surgery: a case-controlled study.

BackgroundDelirium after cardiothoracic surgery is common and associated with impaired outcomes. Although several mechanisms have been proposed (including changes in cerebral perfusion), the pathophysiology of postoperative delirium remains unclear. Blood viscosity is related to cerebral perfusion and thereby might contribute to the development of delirium after cardiothoracic surgery. The aim of this study was to investigate whether whole blood viscosity differs between cardiothoracic surgery patients with and without delirium.MethodsIn this observational study postoperative whole blood viscosity of patients that developed delirium (cases) were compared with non-delirious cardiothoracic surgery patients (controls). Cases were matched with the controls, yielding a 1:4 case–control study. Serial hematocrit, fibrinogen, and whole blood viscosity were determined pre-operatively and at each postoperative day. Delirium was assessed using the validated Confusion Assessment Method for the Intensive Care Unit or Delirium Screening Observation scale.ResultsIn total 80 cardiothoracic surgery patients were screened of whom 12 delirious and 48 matched non-delirious patients were included. No significant difference was found between both groups in fibrinogen (p = 0.36), hematocrit (p = 0.23) and the area under curve of the whole blood viscosity between shear rates 0.02 and 50 s-1 (p = 0.80) or between shear rates 0.02 and 5 s-1 (p = 0.78).ConclusionIn this case control study in cardiothoracic surgery patients changes in whole blood viscosity were not associated with the development of delirium.

Abstract 16245: The Impact of Delirium After Trans-catheter Aortic Valve Replacement

Introduction: Delirium is associated with increased healthcare utilization, morbidity, and mortality. We assessed the impact of delirium on outcomes after trans-catheter aortic valve replacement (TAVR). Methods: We reviewed 294 consecutive cases of TAVR at a tertiary care academic medical center from June, 2008 to February, 2015. Delirium was identified by prospective confusion assessment method screening and retrospective chart review. The primary outcome was overall survival. Secondary outcomes included length of hospital stay, discharge destination, hospital re-admission within 30 days, and 30 day mortality. Results: Delirium was identified in 61 patients after TAVR (21%). Non-femoral access for TAVR was more common in delirious than non-delirious patients (41% vs 28%, P=0.04). There was no difference in the post-operative stroke rate between delirious and non-delirious patients (8% vs 6%, P=0.45). Post-TAVR respiratory failure was more common in delirious patients (21% vs 3%, P&lt;0.001). Of 13 delirious patients who developed respiratory failure, delirium preceded respiratory failure in 8 (62%). Median follow-up was 8.2 months. Delirious patients had decreased overall survival compared to non-delirious patients (figure, log-rank P&lt;0.01). After adjusting for Society of Thoracic Surgeons risk score and TAVR access type, delirium remained an independent predictor of overall survival (hazard ratio 1.97, 95% confidence interval 1.18 to 3.23, P&lt;0.01). Delirium was associated with longer mean length of hospital stay (13.3 +/- 9.5 days vs 6.7 +/- 3.8 days, P&lt;0.001) and higher rates of discharge to a rehabilitation facility (61% vs 27%, P&lt;0.001). There was no difference in 30 day hospital re-admission rates or 30 day mortality between groups. Conclusions: Delirium, a common syndrome after TAVR, is associated with decreased survival and increased healthcare utilization. Early detection of delirium after TAVR may help identify a high risk subset of patients.

Incidence of and Risk Factors for Delirium After Cardiac Surgery at a Quaternary Care Center: A Retrospective Cohort Study.

Delirium after cardiac surgery is associated with persistent cognitive deficits and increased mortality. The authors' objective was to determine the incidence of and risk factors for delirium in a mixed cohort of patients undergoing on-pump and off-pump cardiac surgery and transcatheter aortic valve implantations (TAVI) in a Canadian quaternary care center. This study followed a pilot from the same center on patients treated in 2007. A retrospective cohort study. A quaternary care center in Vancouver, B.C., Canada. Patients undergoing cardiopulmonary bypass grafts (CABG), conventional valve replacements, combined CABG-valve replacements, transfemoral TAVI, or transapical TAVI in 2008. Data from 679 charts on demographics, medical history, medications, laboratory results, surgical procedure, and anesthesia were abstracted and analyzed using univariate and multivariate analyses. Nurses screened for delirium using the Confusion Assessment Method, and the final diagnoses were made clinically by physicians. Risk factors were identified using logistic regression and bootstrapping. Delirium occurred in 28% of patients. Delirium was most common in transapical TAVI (47%), and least common in transfemoral TAVI (17%). Delirious patients were older and had greater preoperative cardiac and neurologic burdens than nondelirious patients. Age≥64 years, history of delirium, history of stroke/transient ischemic attack, cognitive impairment, depression, and preoperative use of beta-blocker(s) were associated independently with delirium. The incidence of delirium varied greatly with the type of procedure. The authors' logistic regression model showed that age and certain pre-existing neurologic conditions could predict delirium after cardiac surgery.

Postoperative delirium in elderly after elective and acute colorectal surgery: A prospective cohort study

ObjectiveTo assess the frailty and the incidence of delirium in elderly patients undergoing elective and acute colorectal surgery in correlation with morbidity and mortality. MethodsPatients aged 65 years and older having elective and acute colorectal surgery, between April 2013 and December 2013 were included in a prospective database. Patients diagnosed with a colorectal carcinoma or diverticulitis who were operated on were included. Factors that characterize frailty of patients were noted. The incidence rates of delirium after elective and acute surgery were recorded. Delirium was diagnosed using the Delirium Observation Screening Scale (DOSS). Preoperative evaluation, surgical outcome including morbidity, hospital stay and mortality were analyzed. ResultsPatients ≥65 years were included, 83 (75%) received elective and 28 (25%) acute surgery. The overall incidence of delirium was 21%, 18% for elective and 29% for patients having urgent surgery (p = 0.24). Patients with delirium were older than the non-delirious patients (median 82 years vs. 74 years; p < 0.001). Delirious patients showed higher incidence of adverse events. Hospital stay, mortality and discharge to a nursing home were significant higher in the delirious compared to the non-delirious group (p = 0.01; 0.01; 0.02 respectively). ConclusionHigh incidence of delirium was found in both acute and elective colorectal surgery. Delirium was associated with adverse outcomes.

Intraoperative burst suppression is associated with postoperative delirium following cardiac surgery: a prospective, observational study.

BackgroundPostoperative delirium (POD) occurs frequently after cardiac surgery and is associated with increased morbidity and mortality. We analysed whether perioperative bilateral BIS monitoring may detect abnormalities before the onset of POD in cardiac surgery patients.MethodsIn a prospective observational study, 81 patients undergoing cardiac surgery were included. Bilateral Bispectral Index (BIS)-monitoring was applied during the pre-, intra- and postoperative period, and BIS, EEG Asymmetry (ASYM), and Burst Suppression Ratio (BSR) were recorded. POD was diagnosed according to the Confusion Assessment Method for the Intensive Care Unit, and patients were divided into a delirium and non-delirium group.ResultsPOD was detected in 26 patients (32%). A trend towards a lower ASYM was observed in the delirium group as compared to the non-delirium group on the preoperative day (ASYM = 48.2 ± 3.6% versus 50.0 ± 4.7%, mean ± sd, p = 0.087) as well as before induction of anaesthesia, with oral midazolam anxiolysis (median ASYM = 49.5%, IQR [47.4;51.5] versus 50.6%, IQR [49.1;54.2], p = 0.081). Delirious patients remained significantly (p = 0.018) longer in a burst suppression state intraoperatively (107 minutes, IQR [47;170] versus 44 minutes, IQR [11;120]) than non-delirious patients. Receiver operating analysis revealed burst suppression duration (area under the curve = 0.73, p = 0.001) and BSR (AUC = 0.68, p = 0.009) as predictors of POD.ConclusionsIntraoperative assessment of BSR may identify patients at risk of POD and should be investigated in further studies. So far it remains unknown whether there is a causal relationship or rather an association between intraoperative burst suppression and the development of POD.Trial registrationclinicaltrials.gov NCT01048775

Prevalence, risk factors, and outcomes of delirium in mechanically ventilated adults.

Delirium is common during critical illness and associated with adverse outcomes. We compared characteristics and outcomes of delirious and nondelirious patients enrolled in a multicenter trial comparing protocolized sedation with protocolized sedation plus daily sedation interruption. Randomized trial. Sixteen North American medical and surgical ICUs. Four hundred thirty critically ill, mechanically ventilated adults. All patients had hourly titration of opioid and benzodiazepine infusions using a validated sedation scale. For patients in the interruption group, infusions were resumed, if indicated, at half of previous doses. Delirium screening occurred daily; positive screening was defined as an Intensive Care Delirium Screening Checklist score of 4 or more at any time. Delirium was diagnosed in 226 of 420 assessed patients (53.8%). Coma was identified in 32.7% of delirious compared with 22.7% of nondelirious patients (p = 0.03). The median time to onset of delirium was 3.5 days (interquartile range, 2-7), and the median duration of delirium was 2 days (interquartile range, 1-4). Delirious patients were more likely to be male (61.1% vs 46.6%; p = 0.005), have a surgical/trauma diagnosis (21.2% vs 11.0%; p = 0.030), and history of tobacco (31.5% vs 16.2%; p = 0.002) or alcohol use (34.6% vs 20.9%; p = 0.009). Patients with positive delirium screening had longer duration of ventilation (13 vs 7 d; p < 0.001), ICU stay (12 vs 8 d; p < 0.0001), and hospital stay (24 vs 15 d; p < 0.0001). Delirious patients were more likely to be physically restrained (86.3% vs 76.7%; p = 0.014) and undergo tracheostomy (34.6% vs 15.5%; p < 0.0001). Antecedent factors independently associated with delirium onset were restraint use (hazard ratio, 1.87; 95% CI, 1.33-2.63; p = 0.0003), antipsychotic administration (hazard ratio, 1.67; 95% CI, 1.005-2.767; p = 0.047), and midazolam dose (hazard ratio, 0.998; 95% CI, 0.997-1.0; p = 0.049). There was no difference in delirium prevalence or duration between the interruption and control groups. In mechanically ventilated adults, delirium was common and associated with longer duration of ventilation and hospitalization. Physical restraint was most strongly associated with delirium.

B-type natriuretic peptide and estimated glomerular filtration rate at ICU admission as a predictor of delirium

Delirium in the ICU is a predictor of mortality and cognitive impairment at hospital discharge. Although several pathways for delirium have been described, it is very difficult to predict the occurrence of delirium. In this study, we examined plasma biomarkers in delirious and nondelirious patients at admission and whether the biomarkers can predict onset of delirium.

MMSE items that predict incident delirium and hypoactive subtype in older medical inpatients

Because hypoactive delirium is especially under-recognized, we analyzed which Mini-Mental State Examination (MMSE) items predicted incident delirium and its hypoactive motor presentation. Over a 1-year period, older medical inpatients (n=291) were consecutively screened on admission with the Confusion Assessment Method-Spanish (CAM-S) to exclude prevalent delirium. Nondelirious patients were evaluated the same day with the MMSE, followed by daily ratings with the CAM-S. Those who became CAM-S positive were rated using the Delirium Rating Scale—Revised-98 to assess severity and motor subtype. Disorientation to time (OR 4.4, 95% CI 1.7–11.1) and place (OR 3.8, 95% CI 1.7–8.2) at admission were risk factors for delirium at follow-up and together correctly classified 88.3% of subjects as to delirium status. Disorientation to time and place, and visuoconstructional impairment were each associated with either hypoactive or mixed subtype (p<0.05 χ2 test). Simple bedside evaluation of cognitive function in nondelirious patients revealed deficits that detected patients at risk for developing incident delirium at follow-up (especially hypoactive or mixed). We recommend patients with orientation deficits be monitored closely for emergence of delirium. A separate evaluation for possible dementia or other causes of cognitive impairment at admission should be considered too.

Cerebrospinal fluid markers of neuroinflammation in delirium: A role for interleukin-1β in delirium after hip fracture

ObjectiveExaggerated central nervous system (CNS) inflammatory responses to peripheral stressors may be implicated in delirium. This study hypothesised that the IL-1β family is involved in delirium, predicting increased levels of interleukin-1β (IL-1β) and decreased IL-1 receptor antagonist (IL-1ra) in the cerebrospinal fluid (CSF) of elderly patients with acute hip fracture. We also hypothesised that Glial Fibrillary Acidic Protein (GFAP) and interferon-γ (IFN-γ) would be increased, and insulin-like growth factor 1 (IGF-1) would be decreased. MethodsParticipants with acute hip fracture aged >60 (N=43) were assessed for delirium before and 3–4 days after surgery. CSF samples were taken at induction of spinal anaesthesia. Enzyme-linked immunosorbent assays (ELISA) were used for protein concentrations. ResultsPrevalent delirium was diagnosed in eight patients and incident delirium in 17 patients. CSF IL-1β was higher in patients with incident delirium compared to never delirium (incident delirium 1.74pg/ml (1.02–1.74) vs. prevalent 0.84pg/ml (0.49–1.57) vs. never 0.66pg/ml (0–1.02), Kruskal–Wallis p=0.03). CSF:serum IL-1β ratios were higher in delirious than non-delirious patients. CSF IL-1ra was higher in prevalent delirium compared to incident delirium (prevalent delirium 70.75pg/ml (65.63–73.01) vs. incident 31.06pg/ml (28.12–35.15) vs. never 33.98pg/ml (28.71–43.28), Kruskal–Wallis p=0.04). GFAP was not increased in delirium. IFN-γ and IGF-1 were below the detection limit in CSF. ConclusionThis study provides novel evidence of CNS inflammation involving the IL-1β family in delirium and suggests a rise in CSF IL-1β early in delirium pathogenesis. Future larger CSF studies should examine the role of CNS inflammation in delirium and its sequelae.

Preventing Delirium After Cardiothoracic Surgery

Preventing Delirium After Cardiothoracic Surgery

Delirium Detection Based on Monitoring of Blinks and Eye Movements

To investigate whether delirious patients differ from nondelirious patients with regard to blinks and eye movements to explore opportunities for delirium detection. Using a single-center, observational study in a tertiary hospital in the Netherlands, we studied 28 delirious elderly and 28 age- and gender-matched (group level) nondelirious elderly, postoperative cardiac surgery patients. Patients were evaluated for delirium by a geriatrician, psychiatrist, or neurologist using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria. Blinks were automatically extracted from electro-oculograms and eye movements from electroencephalography recordings using independent component analysis. The number and duration of eye movements and blinks were compared between patients with and without delirium, based on the classification of the delirium experts described above. During eyes-open registrations, delirious patients showed, compared with nondelirious patients, a significant decrease in the number of blinks per minute (median: 12 [interquartile range {IQR}: 5-18] versus 18 [IQR: 8-25], respectively; p= 0.02) and number of vertical eye movements per minute (median: 1 [IQR: 0-13] versus 15 [IQR: 2-54], respectively; p= 0.01) as well as an increase in the average duration of blinks (median: 0.5 [IQR: 0.36-0.95] seconds versus 0.34 [IQR: 0.23-0.53] seconds, respectively; p<0.01). During eyes-closed registrations, the average duration of horizontal eye movements was significantly increased in delirious patients compared with patients without delirium (median: 0.41 [IQR: 0.15-0.75] seconds versus 0.08 [IQR: 0.06-0.22] seconds, respectively; p<0.01). Spontaneous eye movements and particularly blinks appear to be affected in delirious patients, which holds promise for delirium detection.

Prediction of delirium in critically ill patients with elevated C-reactive protein

Background and PurposeDelirium is thought to be associated with systemic inflammatory response. However, its association with the most widely used inflammatory biomarker C-reactive protein (CRP) has not been well established. We aimed to examine whether CRP on intensive care unit (ICU) entry was associated with subsequent development of delirium. Design and SettingThis prospective observational study was conducted in a mixed 24-bed ICU in a tertiary teaching hospital. MethodsAll patients admitted to the ICU from February 2011 to June 2012 were screened for eligibility. Demographic data and clinical characteristics of included patients were recorded. Patients were screened for the presence of delirium by using the tool Confusion Assessment Method for the ICU (CAM-ICU). C-reactive protein was obtained on ICU entry and 24 hours thereafter. Eligible patients were followed up for 28 days or until death. Univariate and multivariate analyses were performed to evaluate independent risk factors for delirium. Clinical outcome included the length of stay (LOS) in the ICU, 28-day mortality, and duration of mechanical ventilation. Two-tailed P < .05 was considered statistically significant. ResultsA total of 223 patients were included during study period. In univariate analysis, patients with delirium showed significantly higher CRP values than those without (120.5 vs 57.5 mg/L; P = .0001). By adjusting for confounding variables (including age, sex, Acute Physiology and Chronic Health Evaluation II, intubation, living alone, physical restraint, alcohol drinking, smoking, type of medical condition, and hospital LOS before ICU admission) in logistic regression model, CRP remained an independent predictor of delirium (odds ratio, 1.07; 95% confidence interval, 1.01-1.15). As compared with nondelirious patients, those with delirium showed longer LOS in ICU (13 vs 5 days; P < .001) and duration of mechanical ventilation (6 vs 1 days; P < .001). An increase in CRP greater than 8.1 mg/L within 24 hours was associated with 4-fold increase in the risk of delirium (odds ratio: 4.47, 95% confidence interval, 1.28-15.60). ConclusionC-reactive protein measured on ICU entry and its changes within 24 hours are risk indicators of delirium. Further studies exploring the treatment of delirium according to CRP levels are warranted.

Lack of association of S100β and neuron-specific enolase with mortality in critically ill patients

To evaluate the relationship between brain damage biomarkers and mortality in the intensive care unit (ICU). The sample comprised 70 patients admitted to an ICU. Blood samples were collected from all patients on ICU admission, and levels of S100β and neuron-specific enolase (NSE) were determined by ELISA. Acute Physiologic and Chronic Health Evaluation (APACHE II) score was associated with mortality, but NSE and S100β were not associated with this outcome. In contrast, S100β levels were significantly higher in delirious and non-delirious patients who required mechanical ventilation during ICU stay. Levels of brain biomarkers at the time of ICU admission did not predict mortality in critically ill patients.

Intensive care delirium – effect on memories and health‐related quality of life – a follow‐up study

To investigate the effects of delirium in the intensive care unit on health-related quality of life, healthcare dependency and memory after discharge and to explore the association between health-related quality of life and memories, patient diaries and intensive care unit follow-up. Up to 83% of intensive care unit patients experience delirium. In addition to increased risk of mortality, morbidity and cognitive impairment, the experience itself is unpleasant. A number of studies have focused on memories associated with delirium, but the association between delirium, memories and health-related quality needs further investigation. We used an observational multicentre design with telephone interviews. Adult intensive care unit patients (n = 360) were consecutively recruited and interviewed using the intensive care unit-Memory Tool one week after intensive care unit. Interviews were repeated after two and six months and supplemented with Short Form-36 and the Barthel Index. Delirium was detected in 60% of the patients in our study, and delirious patients had significantly fewer factual memories and more memories of delusion than nondelirious patients up to six months postintensive care unit discharge. Delirium, memories and intensive care unit diaries with follow-up did not affect health-related quality of life and healthcare dependency. Memories of delusions might have an impact on patients assessed as nondelirious. More than half of the patients in intensive care unit experience delirium, which is associated with fewer factual memories and more memories of delusions. Short Form-36 might not be sensitive to delirium-related outcomes. Future research should include the development of better assessment tools to determine the long-term consequences of intensive care unit delirium. We recommend regular assessment to prevent, detect and treat delirium. We also recommend an intensive care unit follow-up programme providing an opportunity for postintensive care unit patients, particularly previously delirious patients, to discuss their memories and experiences with intensive care unit professionals.

Actigraphy for Measurement of Sleep and Sleep-Wake Rhythms in Relation to Surgery

Patients undergoing surgery have severe sleep and sleep-wake rhythm disturbances resulting in increased morbidity. Actigraphy is a tool that can be used to quantify these disturbances. The aim of this manuscript was to present the literature where actigraphy has been used to measure sleep and sleep-wake rhythms in relation to surgery. A systematic review was performed in 3 databases (Medline, Embase, and Psycinfo), including all literature until July 2012. Thirty-two studies were included in the review. Actigraphy could demonstrate that total sleep time and sleep efficiency was reduced after surgery and number of awakenings was increased in patients undergoing major surgery. Disturbances were less severe in patients undergoing minor surgery. Actigraphy could be used to differentiate between delirious and non-delirious patients after major surgery. Actigraphy measurements could determine a differential effect of surgery based on the patient's age. The effect of pharmacological interventions (chronobiotics and hypnotics) in surgical patients could also be demonstrated by actigraphy. Actigraphy can be used to measure sleep and sleep-wake rhythms in patients undergoing surgery.

Reliability of delirium rating scale (DRS) and delirium rating scale-revised-98 (DRS-R98) using variance-based multivariate modelling

Delirium’s characteristic fluctuation in symptom severity complicates the assessment of test–retest reliability of scales using classical analyses, but application of modelling to longitudinal data offers a new approach. We evaluated test–retest reliability of the delirium rating scale (DRS) and delirium rating scale-revised-98 (DRS-R98), two widely used instruments with high validity and inter-rater reliability. Two existing longitudinal datasets for each scale included DSM-IV criteria for delirium diagnosis and repeated measurements using the DRS or DRS-R98. To estimate the reliability coefficients RT and RΛ for each scale we used a macros provided by Dr. Laenen at http://www.ibiostat.be/software/measurement.asp. For each dataset a linear mixed-effects model was fitted to estimate the variance–covariance parameters. A total of 531 cases with between 4 and 9 measurement points across studies including both delirious and non-delirious patients. Comorbid dementia in the datasets varied from 27% to 55%. Overall RT for the DRS were 0.71 and 0.50 and for DRS-R98 0.75 and 0.84. RΛ values for DRS were 0.99 and 0.98 and for DRS-R98 were 0.92 and 0.96. Individual RT measures for DRS-R98 and DRS across visits within studies showed more range than overall values. Our models found high overall reliability for both scales. Multiple factors impact a scale's reliability values including sample size, repeated measurements, patient population, etc in addition to rater variability.

Long-term outcome of delirium in critically ill patients

In ICU patients, little research has been performed on the relationship between delirium and long-term outcome, including health-related quality of life (HRQoL), cognitive functioning and mortality. In addition, results seem to be inconsistent. Furthermore, in studies that reported increased mortality in delirious patients, no proper adjustments were made for severity of illness during ICU admission. This study was conducted to investigate the association between ICU delirium and long-term HRQoL, cognitive functioning and mortality. The hypothesis was that delirious patients have worse long-term outcome in comparison with nondelirious patients.

Fluctuations in sedation levels may contribute to delirium in ICU patients

Delirium in patients admitted to the intensive care unit (ICU) is a serious complication potentially increasing morbidity and mortality. The aim of this study was to investigate the impact of fluctuating sedation levels on the incidence of delirium in ICU. A prospective cohort study of adult patients at three multidisciplinary ICUs. The Richmond Agitation and Sedation Scale (RASS) and the Confusion Assessment Method for the ICU were used at least twice a day. Delirium was detected at least once in 65% of the patients (n = 640). Delirious patients were significantly older, more critically ill, more often intubated, had longer ICU stays, and had higher ICU mortality than non-delirious patients. The median duration of delirium was 3 days (interquartile range: 1;10), and RASS was less than or equal to 0 (alert and calm) 91% of the time. The odds ratio (OR) for development of delirium if RASS changed more than two levels was 5.19 when adjusted for gender, age, severity of illness, and ICU site and setting. Continuous infusion of midazolam was associated with a decrease in delirium incidence (OR: 0.38; P = 0.002). Fluctuations in sedation levels may contribute to development of delirium in ICU patients. The risk of developing delirium might be reduced by maintaining a stable sedation level or by non-sedation.