Secondary Logo

Share this article on:

A Systematic Review and Meta-Analysis Examining the Impact of Sleep Disturbance on Postoperative Delirium

Fadayomi, Ayòtúndé B., MBBS, MPH1; Ibala, Reine, BS2; Bilotta, Federico, MD, PhD3; Westover, Michael B., MD, PhD4; Akeju, Oluwaseun, MD, MMSc2

doi: 10.1097/CCM.0000000000003400
Online Review Articles

Objectives: Basic science and clinical studies suggest that sleep disturbance may be a modifiable risk factor for postoperative delirium. We aimed to assess the association between preoperative sleep disturbance and postoperative delirium.

Data Sources: We searched PubMed, Embase, CINAHL, Web of Science, and Cochrane from inception until May 31, 2017.

Study Selection: We performed a systematic search of the literature for all studies that reported on sleep disruption and postoperative delirium excluding cross-sectional studies, case reports, and studies not reported in English language.

Data Extraction: Two authors independently performed study selection and data extraction. We calculated pooled effects estimates with a random-effects model constructed in Stata and evaluated the risk of bias by formal testing (Stata Corp V.14, College Station, TX),

Data Synthesis: We included 12 studies, from 1,238 citations that met our inclusion criteria. The pooled odds ratio for the association between sleep disturbance and postoperative delirium was 5.24 (95% CI, 3.61–7.60; p < 0.001 and I 2 = 0.0%; p = 0.76). The pooled risk ratio for the association between sleep disturbance and postoperative delirium in prospective studies (n = 6) was 2.90 (95% CI, 2.28–3.69; p < 0.001 and I 2 = 0.0%; p = 0.89). The odds ratio associated with obstructive sleep apnea and unspecified types of sleep disorder were 4.75 (95% CI, 2.65–8.54; p < 0.001 and I 2 = 0.0%; p = 0.85) and 5.60 (95% CI, 3.46–9.07; p < 0.001 and I 2 = 0.0%; p = 0.41), respectively. We performed Begg’s and Egger’s tests for publication bias and confirmed a null result for publication bias (p = 0.371 and 0.103, respectively).

Conclusions: Preexisting sleep disturbances are likely associated with postoperative delirium. Whether system-level initiatives targeting patients with preoperative sleep disturbance may help reduce the prevalence, morbidity, and healthcare costs associated with postoperative delirium remains to be determined.

1Division of Epidemiology, Department of Obstetrics and Gynecology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA.

2Department of Anesthesia, Critical Care, and Pain Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA.

3Department of Anesthesia and Critical Care Medicine, Sapienza University, Rome, Italy.

4Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, MA.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (

Supported, in part, by grants National Institutes of Health National Institute of Aging R01 R01AG053582.

Drs. Fadayomi, Westover, and Akeju received support for article research from the National Institutes of Health (NIH). Dr. Akeju received funding from the NIH. The remaining authors have disclosed that they do not have any potential conflicts of interest.

For information regarding this article,

Postoperative delirium (POD), commonly encountered after surgery (1), is an acute brain dysfunction characterized by disturbances in attention, awareness, and cognition not explained by a preexisting neurocognitive disorder (1 , 2). POD remains a leading cause of morbidity in hospitalized patients (3). Recent studies suggest that clinical care protocols directed at elderly postsurgical patients result in a reduced prevalence of POD (4). Thus, principled strategies to preemptively identify and target patients at risk for POD may result in significantly improved perioperative outcomes.

Sleep is a naturally occurring state of decreased arousal that is crucial for normal immune and cognitive function (5). Sleep disturbance is associated with increased levels of proinflammatory cytokines (6–8). Because diagnoses with a high prevalence of delirium are also characterized by increased levels of proinflammatory cytokines (systemic inflammation) (9–23), systemic inflammation may be involved in the pathophysiology of delirium, and sleep disturbance may constitute a modifiable risk factor for POD (24). Therefore, our objective was to determine the association between preexisting sleep disturbance and POD.

Back to Top | Article Outline


This systematic review and meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The study protocol was registered with the international prospective register of systematic reviews (PROSPERO) (CRD42017070607).

Back to Top | Article Outline

Search Strategy

A library sciences specialist designed and implemented a comprehensive literature search of five databases: PubMed, Embase, CINAHL, Cochrane, and Web of Science from inception until May 31, 2017 using the following key word terms: (“Confusion” (25) OR delirium[tiab] OR delirious[tiab] OR confusion*[tiab] OR metabolic encephalopathy[tiab] OR toxic encephalopathy[tiab] OR acute brain dysfunction[tiab] OR acute organic psychosyndrome*[tiab] OR acute psychoorganic[tiab] OR acute psycho organic[tiab] OR clouded[tiab] OR clouding[tiab] OR exogenous psycho*[tiab] OR toxic psycho*[tiab]) AND (“Sleep” (25) OR “Sleep Wake Disorders” (25) OR sleep*[tiab] OR restless leg*[tiab] OR dyssomn* OR parasomn*[tiab] OR narcolep* OR somnolen*[tiab] OR hypersomn*[tiab] OR insomnolen*[tiab] OR hyposomn*[tiab] OR myoclonus syndrome[tiab] OR hypnogenic paroxysmal[tiab] OR somnamb*[tiab]) AND (“Anesthesia” (25) OR “Anesthetics” (25) OR “Anesthetics”[Pharmacological Action] OR “Surgical Procedures, Operative” (25) OR “Postoperative Care” (25) OR “Postoperative Complications”[Mesh:noexp] OR “surgery”[subheading] OR surgical[tiab] OR surgery[tiab] OR postsurg*[tiab] OR operative[tiab] OR postoperative[tiab] OR anesthes*[tiab] OR anaesthes*[tiab] OR anesthet*[tiab] OR anaesthet*[tiab]) (Supplemental Table 1, Supplemental Digital Content 1,

Back to Top | Article Outline

Study Selection

Two reviewers (A.B.F., R.I.) assessed the retrieved studies independently, including titles, abstracts, and citations to determine whether each citation met inclusion criteria. The full texts of the citations classified as include or unclear were reviewed independently with reference to the predetermined inclusion and exclusion criteria. Consensus was achieved through discussions with a third reviewer (O.A.) in cases of disagreement. All citations associated with selected studies were screened for articles that met our inclusion criteria but were not captured by our literature search.

Eligibility was determined if studies were in adult human population greater than 18 years old, quantitative with calculated effect measures, randomized controlled trials or prospective or retrospective cohort or case-control studies, electronically accessible, done within 30 days POD period. We excluded cross-sectional studies, case reports, and studies not reported in English language.

Back to Top | Article Outline

Data Extraction

Two reviewers (A.B.F., R.I.) independently extracted data from eligible studies using predesigned forms. Disagreements between the two reviewers were resolved after discussions with a third reviewer (O.A.). The following data were collected: 1) name of first author; 2) year of publication; 3) country; 4) study design; 5) study population; 6) mean/median age and age range; 7) type of sleep disorder assessed; 8) period of sleep disorder assessment; 9) sleep quality assessment tool; 10) delirium assessment tool; 11) type of surgery; 12) numerical data on the number of participants in each arm, names of comparators, number of events in each arm, and reported odds ratios (ORs) and risk ratios (RRs); and 13) variables used for data adjustment or matching. Where multiple effect estimates were reported, we extracted both the crude and adjusted effect estimates reflecting the greatest degree of control for potential confounders.

Back to Top | Article Outline

Quality of Individual Studies and Publication Bias

The quality of included studies was independently assessed by two authors (A.B.F., R.I.) using the Newcastle-Ottawa form (26). The Newcastle-Ottawa form is a standardized and published tool for assessing nonrandomized studies. High-quality studies were defined as scores of 7 or higher, moderate quality as scores of 5–6, and low quality as less than 5 (26). Publication bias was assessed by visual inspection of a funnel plot and formal testing with the Egger’s test and Begg’s test (27–29).

Back to Top | Article Outline

Data Analysis

Analyses were performed using Stata software (Stata Corp V.14, College Station, TX). All tests were two sided, and we considered a p value of less than 0.05 statistically significant. The crude OR and 95% CI of each study were estimated using number of events on each arm where available or SE when the CI was not reported. Heterogeneity among studies was quantified using visual inspection of forest plot, the Cochran’s Q statistic (p < 0.05), and I 2 statistic. We used the conservative random effects model of DerSimonian and Laird (30) to pool all ORs. Heterogeneity was explored in subgroup analysis of 1) study size, 2) study design, 3) mean population age, 4) sleep disorder type, 5) timing of sleep disorder, 6) delirium assessment tool, and 7) type of surgery.

Back to Top | Article Outline

Sensitivity Analysis

The crude RR of prospective cohort studies were computed, and the random effects model was used to calculate the overall pooled estimate.

Back to Top | Article Outline


Study Selection

We identified 1, 238 citations (328 studies in the MEDLINE database [PubMed], 939 in Embase, 86 in CINAHL, 266 in Web of Science, and 76 in the Cochrane Library Database) and included 12 unique studies enrolling 1, 878 patients. The PRISMA flow diagram of the studies’ selection is presented in Figure 1.

Figure 1

Figure 1

Back to Top | Article Outline

Study Characteristics

All studies were reported between 2001 and 2015. They included seven prospective cohort studies (31–37) and five retrospective studies (38–42). Nine studies had a sample size greater than or equal to 60 (32 , 33 , 35–37 , 39–42). The largest study was a retrospective cohort study involving 432 patients (42), whereas the smallest was a prospective cohort study with 40 patients (31).

Four studies specified obstructive sleep apnea (OSA) as the type of sleep disorder evaluated (32 , 34 , 35 , 39). However, the other studies did not explicitly specify the type of sleep disorder that was evaluated (31 , 33 , 37 , 38 , 40–43). Eight studies assessed preoperative sleep disorder (31 , 32 , 34 , 35 , 38–40 , 43), three studies assessed sleep disorder immediately post surgery but before the onset of POD (37 , 41 , 42), and one study assessed sleep disorder occurring after the onset of POD (33). Five studies evaluated orthopedic surgeries (32 , 36 , 38 , 39 , 41), four studies evaluated cardiac surgeries (31 , 33 , 35 , 37), and others evaluated thoracic and noncardiac surgeries (34 , 40 , 42). Eight studies used the Confusion Assessment Method (CAM), a standardized evidence-based tool (20 , 21), to diagnose delirium (31 , 32 , 34–37 , 40 , 41), two studies used the Diagnosis and Statistical Manual of Mental Disorders (DSM)–IV (33 , 42), and two studies used patient chart records and information obtained from caregivers to diagnose POD (38 , 39). Detailed characteristics of these studies are described in Supplemental Table 2 (Supplemental Digital Content 2,

Back to Top | Article Outline

Pooled Analysis

The crude ORs were analyzed in this meta-analysis. Yildizeli et al (42) and Jeong et al (40) did not report sufficient information to calculate crude ORs. Thus, data from this study were excluded in the analysis. A total of 1,199 patients and 244 cases of POD were analyzed. The pooled OR for the association between sleep disturbance and POD of 4.89 (95% CI, 3.45–6.94; p < 0.001 and I 2 = 0.0%; p = 0.73) (Fig. 2A) was statistically significant. We next restricted our analysis to studies that evaluated sleep disorder before onset of POD. A total of 1,096 patients and 225 cases of POD were analyzed. The pooled OR for the association between pre-POD sleep disturbance and POD of 5.24 (95% CI, 3.61–7.60; p < 0.001 and I 2 = 0.0%; p = 0.76) (Fig. 2B) was statistically significant.

Figure 2

Figure 2

Back to Top | Article Outline

Sub Group Analysis

Sample Size. The pooled OR for the association between sleep disturbance and POD of 3.37 (95% CI, 0.74–15.27; p = 0.12 and I 2 = 39.8%; p = 0.19) (Supplemental Fig. 1, Supplemental Digital Content 3,; legend, Supplemental Digital Content 9, for the studies with less than 60 patients was not statistically significant (31 , 34 , 38). However, the pooled OR for the association between sleep disturbance and POD of 5.52 (95% CI, 3.72–8.17; p < 0.001 and I 2 = 0.0%; p = 0.96) (Supplemental Fig. 1, Supplemental Digital Content 3,; legend, Supplemental Digital Content 9, for the studies with greater than 60 patients (32 , 35–37 , 39 , 41) was statistically significant.

Study Design. The pooled OR for the association between sleep disturbance and POD of 5.77 (95% CI, 3.82–8.73; p < 0.001 and I 2 = 0.0%; p = 0.89) (Fig. 3A) for the prospective studies (31 , 32 , 34–37) was statistically significant. Similarly, the pooled OR for the association between sleep disturbance and POD of 3.42 (95% CI, 1.39–8.44; p = 0.008 and I 2 = 9.1%; p = 0.33) (Fig. 3A) for retrospective studies (38 , 39 , 41) was statistically significant.

Figure 3

Figure 3

Age (≥ 65 or ≤ 65). The pooled OR for the association between sleep disturbance and POD of 4.83 (95% CI, 3.00–7.78; p < 0.001 and I 2 = 0.0%; p = 0.97) (Fig. 3B) for studies with median or mean patient age greater than 65 years old (32 , 34–36 , 39 , 41) was statistically significant. Similarly, the pooled OR for the association between sleep disturbance and POD of 4.91 (95% CI, 1.35–17.86; p = 0.02 and I 2 = 47.9%; p = 0.15) (Fig. 3B) for studies with median or mean patient age less than 65 years old (31 , 37 , 38) was statistically significant.

Obstructive Sleep Apnea Versus Unspecified Sleep Disorder. The pooled OR for the association between sleep disturbance and POD of 4.75 (95% CI, 2.65–8.54; p < 0.001 and I 2 = 0.0%; p = 0.85) (Fig. 4A) for obstructive sleep apnea (32 , 34 , 35 , 39) was statistically significant. Similarly, the pooled OR for the association between sleep disturbance and POD of 5.60 (95% CI, 3.46–9.07; p < 0.001 and I 2 = 0.0%; p = 0.41) (Fig. 4A) for studies with unspecified types of sleep disorder (31 , 36–38 , 41) was statistically significant.

Figure 4

Figure 4

Timing of Sleep Disorder. The pooled OR for the association between sleep disturbance and POD of 4.59 (95% CI, 2.82–7.48; p < 0.001 and I 2 = 0%; p = 0.64) (Fig. 4B) for patients with preoperative sleep disorder (31 , 32 , 34–36 , 38 , 39) was statistically significant. Similarly, the pooled OR for the association between sleep disturbance and POD of 6.29 (95% CI, 3.55–11.17; p < 0.001 and I 2 = 0%; p = 0.82) (Fig. 4B) for patients with postsurgical, but pre-POD, sleep disorder (37 , 41) was statistically significant.

Delirium Assessment Tool. The pooled OR for studies which used CAM screening tools (31 , 34–37 , 41) for delirium assessment was 5.97 (95% CI, 3.93–9.08; p < 0.001 and I 2 = 0%; p = 0.92), the study which used DSM-IV (32) had an OR of 4.33 (95% CI, 1.39–13.47; p = 0.01), and two studies which made use of chart review or care giver assessment (38 , 39) had a pooled OR of 2.21 (95% CI, 0.55–8.98; p = 0.27 and I 2 = 24.6%; p = 0.25) (Supplemental Fig. 2, Supplemental Digital Content 4,; legend, Supplemental Digital Content 9,

Type of Surgery. The pooled OR for orthopedic/ noncardiac surgeries (32 , 34 , 36 , 38 , 39 , 41) was 3.92 (95% CI, 2.27–6.75; p < 0.001 and I 2 = 0%; p = 0.79), whereas studies with cardiothoracic surgeries (31 , 35 , 37) had a pooled OR of 6.77 (95% CI, 4.06–11.26; p < 0.001 and I 2 = 0%; p = 0.78) (Supplemental Fig. 3, Supplemental Digital Content 5,; legend, Supplemental Digital Content 9,

Sensitivity Analysis. We conducted a sensitivity analysis using RRs obtained from the prospective studies (31 , 32 , 34–36 , 44). The pooled RR for the association between sleep disturbance and POD of 2.90 (95% CI, 2.28–3.69; p < 0.001 and I 2 = 0.0%; p = 0.89) was significant (Supplemental Fig. 4, Supplemental Digital Content 6,; legend, Supplemental Digital Content 9,

Back to Top | Article Outline

Quality of Individual Studies and Publication Bias

Using the Newcastle-Ottawa form, all but the study by Koster et al (33) were defined as high quality (Supplemental Table 3, a and b, Supplemental Digital Content 7, The symmetrical nature of our funnel plot argues against observable publication bias (Supplemental Fig. 5, Supplemental Digital Content 8,; legend, Supplemental Digital Content 9, We further assessed for publication bias by formal testing with both the Begg’s and Egger’s tests and confirmed a null result for publication bias (p = 0.371 and 0.103, respectively).

Back to Top | Article Outline


Despite the heterogeneous design of available studies, we observed an association between sleep disturbance and POD. Overall, patients with sleep disturbance were approximately five times more likely to develop POD compared with those without known history of sleep disturbance. Further subanalyses suggested that these findings were conserved for OSA and unspecified types of sleep disturbances.

Consistent with our findings, Evans et al (45), in a pilot prospective study that used electroencephalogram recording to objectively study predictors of POD, found that diminished sleep time and increased sleep latency on postoperative day 1 is associated with increased prevalence and severity of POD. However, whether sleep disturbance is causal to POD to suggest that improved sleep hygiene would result in reduced prevalence of POD, or whether sleep disruption is an epiphenomenon that is associated with POD vulnerable brains remains an open question.

Sedative drugs alter the level of arousal to achieve a behavioral state that closely approximates sleep. These drugs, most of which modulate the gamma amino butyric acid A receptor (GABAA), are instead associated with an increased prevalence of POD and do not produce the neurophysiologic oscillations of sleep (46). GABAA receptor modulation is not a physiologically relevant mechanism for sleep promotion (46 , 47). Thus, it is not surprising that these medications are associated with neurophysiologic dynamics such as electroencephalogram frontal beta oscillations, frontal alpha oscillations, burst suppression, and isoelectricity (46). These oscillatory dynamics, which are fundamentally distinct from the oscillatory dynamics encountered during normal sleep, may reflect disruption in cognitive processing circuits (46).

Zhang et al (48) performed a network meta-analysis of sedation strategies in mechanically ventilated patients and reported that dexmedetomidine is associated with a lower prevalence of delirium. However, the generalizability of this finding to nonmechanically ventilated postsurgical patients was unclear until Su et al (49) demonstrated that the administration of a prophylactic low-dose infusion of dexmedetomidine, an a2a adrenergic receptor agonist, resulted in significantly reduced prevalence of POD in postsurgical noncardiac patients. Investigators from this research group also demonstrated that this prophylactic low-dose infusion paradigm is associated with increased duration of stage N2 sleep, prolonged total sleep time, and increased sleep efficiency in nonmechanically ventilated postsurgical patients (50). This finding is consistent with results from neurophysiologic (51–53) and clinical polysomnography studies (54 , 55) confirming that dexmedetomidine sedation closely approximates stage N2 sleep. Thus, pharmacologic approximation of sleep may result in decreased prevalence of POD.

To our knowledge, this is the first meta-analysis to assess the association between sleep disorder and POD. Key strengths of our meta-analysis include consistency of results and lack of publication bias. We note that our meta-analysis was limited by the lack of randomized clinical trials, unspecified types of sleep disorder, and variability in metrics that were used to diagnose sleep disturbance and POD. Despite the limitation of variability in delirium assessment tools, CAM screening tools have been validated against the DSM-IV and found to be highly sensitive and specific (56 , 57). We also stratified by sleep disorder to reduce any confounding by unspecified types of sleep disorder. Our present findings underscore the need for large randomized controlled studies of POD prevention that couple interventions such as IV dexmedetomidine (48), physiologically relevant concentrations of oral melatonin (58), and cognitive behavioral sleep therapy (59 , 60) with objective sleep polysomnography metrics. These studies will enable causal inferences and make clear the extent to which the prevalence of POD may be modified by sleep in the perioperative period.

Back to Top | Article Outline


We conclude that preoperative sleep disturbances are likely associated with POD. Whether system-level initiatives targeting patients with preoperative sleep disturbance may help modify this association remains to be determined.

Back to Top | Article Outline


We thank Paul Bain, MS, PhD, of Countway Library of Medicine, Harvard Medical School, for his assistance with literature search on the different databases.

Back to Top | Article Outline


1. Gleason LJ, Schmitt EM, Kosar CM, et al. Effect of delirium and other major complications on outcomes after elective surgery in older adults. JAMA Surg 2015; 150:1134–1140
2. American Psychiatric Association, DSM-5 Task Force: Diagnostic and Statistical Manual of Mental Disorders: DSM-5. 2013, pp Fifth Edition. Washington, DC, American Psychiatric Association, 947
3. Inouye SK, Charpentier PA. Precipitating factors for delirium in hospitalized elderly persons. Predictive model and interrelationship with baseline vulnerability. JAMA 1996; 275:852–857
4. Chen CC, Li HC, Liang JT, et al. Effect of a modified hospital elder life program on delirium and length of hospital stay in patients undergoing abdominal surgery: A cluster randomized clinical trial. JAMA Surg 2017; 152:827–834
5. Brown RE, Basheer R, McKenna JT, et al. Control of sleep and wakefulness. Physiol Rev 2012; 92:1087–1187
6. Shearer WT, Reuben JM, Mullington JM, et al. Soluble TNF-alpha receptor 1 and IL-6 plasma levels in humans subjected to the sleep deprivation model of spaceflight. J Allergy Clin Immunol 2001; 107:165–170
7. Haack M, Sanchez E, Mullington JM. Elevated inflammatory markers in response to prolonged sleep restriction are associated with increased pain experience in healthy volunteers. Sleep 2007; 30:1145–1152
8. Vgontzas AN, Zoumakis E, Bixler EO, et al. Adverse effects of modest sleep restriction on sleepiness, performance, and inflammatory cytokines. J Clin Endocrinol Metab 2004; 89:2119–2126
9. Siami S, Annane D, Sharshar T. The encephalopathy in sepsis. Crit Care Clin 2008; 24:67–82, viii
10. O’Keeffe ST, Ní Chonchubhair A. Postoperative delirium in the elderly. Br J Anaesth 1994; 73:673–687
11. Lenz A, Franklin GA, Cheadle WG. Systemic inflammation after trauma. Injury 2007; 38:1336–1345
12. Shimaoka M, Park EJ. Advances in understanding sepsis. Eur J Anaesthesiol Suppl 2008; 42:146–153
13. Weichhart T, Haidinger M, Hörl WH, et al. Current concepts of molecular defence mechanisms operative during urinary tract infection. Eur J Clin Invest 2008; 38(Suppl 2):29–38
14. Bjornsson GL, Thorsteinsson L, Gudmundsson KO, et al. Inflammatory cytokines in relation to adrenal response following total hip replacement. Scand J Immunol 2007; 65:99–105
15. Kragsbjerg P, Holmberg H, Vikerfors T. Serum concentrations of interleukin-6, tumour necrosis factor-alpha, and C-reactive protein in patients undergoing major operations. Eur J Surg 1995; 161:17–22
16. Keck M, Herndon DH, Kamolz LP, et al. Pathophysiology of burns. Wien Med Wochenschr 2009; 159:327–336
17. Marx N, Neumann FJ, Ott I, et al. Induction of cytokine expression in leukocytes in acute myocardial infarction. J Am Coll Cardiol 1997; 30:165–170
18. Asimakopoulos G. Mechanisms of the systemic inflammatory response. Perfusion 1999; 14:269–277
19. Sun Y, Tawara I, Toubai T, et al. Pathophysiology of acute graft-versus-host disease: recent advances. Transl Res 2007; 150:197–214
20. Stinghen AE, Gonçalves SM, Martines EG, et al. Increased plasma and endothelial cell expression of chemokines and adhesion molecules in chronic kidney disease. Nephron Clin Pract 2009; 111:c117–c126
21. Shawcross DL, Shabbir SS, Taylor NJ, et al. Ammonia and the neutrophil in the pathogenesis of hepatic encephalopathy in cirrhosis. Hepatology 2010; 51:1062–1069
22. van Munster BC, Korevaar JC, Zwinderman AH, et al. Time-course of cytokines during delirium in elderly patients with hip fractures. J Am Geriatr Soc 2008; 56:1704–1709
23. Culley DJ, Snayd M, Baxter MG, et al. Systemic inflammation impairs attention and cognitive flexibility but not associative learning in aged rats: Possible implications for delirium. Front Aging Neurosci 2014; 6:107
24. Maldonado JR. Neuropathogenesis of delirium: Review of current etiologic theories and common pathways. Am J Geriatr Psychiatry 2013; 21:1190–1222
25. Taylor K, Kim WT, Maharramova M, et al. Intraoperative management and early postoperative outcomes of pediatric renal transplants. Paediatr Anaesth 2016; 26:987–991
26. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010; 25:603–605
27. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994; 50:1088–1101
28. Song F, Gilbody S. Bias in meta-analysis detected by a simple, graphical test. Increase in studies of publication bias coincided with increasing use of meta-analysis. BMJ 1998; 316:471
29. Sterne JA, Gavaghan D, Egger M. Publication and related bias in meta-analysis: Power of statistical tests and prevalence in the literature. J Clin Epidemiol 2000; 53:1119–1129
30. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7:177–188
31. Cheraghi MA, Hazaryan M, Bahramnezhad F, et al. Study of the relationship between sleep quality and prevalence of delirium in patients undergoing cardiac surgery. Int J Med Res Health Sci 2016; 5:38–43
32. Flink BJ, Rivelli SK, Cox EA, et al. Obstructive sleep apnea and incidence of postoperative delirium after elective knee replacement in the nondemented elderly. Anesthesiology 2012; 116:788–796
33. Koster S, Hensens AG, van der Palen J. The long-term cognitive and functional outcomes of postoperative delirium after cardiac surgery. Ann Thorac Surg 2009; 87:1469–1474
34. Leung JM, Sands LP, Newman S, et al. Preoperative sleep disruption and postoperative delirium. J Clin Sleep Med 2015; 11:907–913
35. Roggenbach J, Klamann M, von Haken R, et al. Sleep-disordered breathing is a risk factor for delirium after cardiac surgery: A prospective cohort study. Crit Care 2014; 18:477
36. Todd OM, Gelrich L, MacLullich AM, et al. Sleep disruption at home as an independent risk factor for postoperative delirium. J Am Geriatr Soc 2017; 65:949–957
37. Zhang WY, Wu WL, Gu JJ, et al. Risk factors for postoperative delirium in patients after coronary artery bypass grafting: A prospective cohort study. J Crit Care 2015; 30:606–612
38. Bosmak FS, Gibim PT, Guimarães S, et al. Incidence of delirium in postoperative patients treated with total knee and hip arthroplasty. Rev Assoc Med Bras (1992) 2017; 63:248–251
39. Gupta RM, Parvizi J, Hanssen AD, et al. Postoperative complications in patients with obstructive sleep apnea syndrome undergoing hip or knee replacement: A case-control study. Mayo Clin Proc 2001; 76:897–905
40. Jeong DM, Kim JA, Ahn HJ, et al. Decreased incidence of postoperative delirium in robot-assisted thoracoscopic esophagectomy compared with open transthoracic esophagectomy. Surg Laparosc Endosc Percutan Tech 2016; 26:516–522
41. Wang J, Li Z, Yu Y, et al. Risk factors contributing to postoperative delirium in geriatric patients postorthopedic surgery. Asia Pac Psychiatry 2015; 7:375–382
42. Yildizeli B, Ozyurtkan MO, Batirel HF, et al. Factors associated with postoperative delirium after thoracic surgery. Ann Thorac Surg 2005; 79:1004–1009
43. Todd O. Poor sleep-An early sign of incipient delirium, or a risk factor in itself? a clinical observational study. Sleep Med 2013; 14:e285
44. Zhang H, Lu Y, Liu M, et al. Strategies for prevention of postoperative delirium: A systematic review and meta-analysis of randomized trials. Crit Care 2013; 17:R47
45. Evans JL, Nadler JW, Preud’homme XA, et al. Pilot prospective study of post-surgery sleep and EEG predictors of post-operative delirium. Clin Neurophysiol 2017; 128:1421–1425
46. Akeju O, Brown EN. Neural oscillations demonstrate that general anesthesia and sedative states are neurophysiologically distinct from sleep. Curr Opin Neurobiol 2017; 44:178–185
47. Zhdanova IV. Melatonin as a hypnotic: Pro. Sleep Med Rev 2005; 9:51–65
48. Zhang Z, Chen K, Ni H, et al. Sedation of mechanically ventilated adults in intensive care unit: A network meta-analysis. Sci Rep 2017; 7:44979
49. Su X, Meng ZT, Wu XH, et al. Dexmedetomidine for prevention of delirium in elderly patients after non-cardiac surgery: A randomised, double-blind, placebo-controlled trial. Lancet 2016; 388:1893–1902
50. Wu XH, Cui F, Zhang C, et al. Low-dose dexmedetomidine improves sleep quality pattern in elderly patients after noncardiac surgery in the intensive care unit: A pilot randomized controlled trial. Anesthesiology 2016; 125:979–991
51. Akeju O, Kim SE, Vazquez R, et al. Spatiotemporal dynamics of dexmedetomidine-induced electroencephalogram oscillations. PLoS One 2016; 11:e0163431
52. Akeju O, Pavone KJ, Westover MB, et al. A comparison of propofol- and dexmedetomidine-induced electroencephalogram dynamics using spectral and coherence analysis. Anesthesiology 2014; 121:978–989
53. Huupponen E, Maksimow A, Lapinlampi P, et al. Electroencephalogram spindle activity during dexmedetomidine sedation and physiological sleep. Acta Anaesthesiol Scand 2008; 52:289–294
54. Alexopoulou C, Kondili E, Diamantaki E, et al. Effects of dexmedetomidine on sleep quality in critically ill patients: A pilot study. Anesthesiology 2014; 121:801–807
55. Oto J, Yamamoto K, Koike S, et al. Sleep quality of mechanically ventilated patients sedated with dexmedetomidine. Intensive Care Med 2012; 38:1982–1989
56. Wei LA, Fearing MA, Sternberg EJ, et al. The Confusion Assessment Method: A systematic review of current usage. J Am Geriatr Soc 2008; 56:823–830
57. Ely EW, Margolin R, Francis J, et al. Evaluation of delirium in critically ill patients: Validation of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU). Crit Care Med 2001; 29:1370–1379
58. Al-Aama T, Brymer C, Gutmanis I, et al. Melatonin decreases delirium in elderly patients: A randomized, placebo-controlled trial. Int J Geriatr Psychiatry 2011; 26:687–694
59. Kamdar BB, King LM, Collop NA, et al. The effect of a quality improvement intervention on perceived sleep quality and cognition in a medical ICU. Crit Care Med 2013; 41:800–809
60. Park SY, Kim HS, Choe YH, et al. Early reality-orienting assuring and sleep assurance for delirium in intensive care unit(ICU): A quality improvement(QI) project. Intensive Care Med 2014; 40:S250

anesthesia; delirium; meta-analysis; postoperative; sleep disturbance; systematic review

Supplemental Digital Content

Back to Top | Article Outline
Copyright © by 2018 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.