Search tips
Search criteria 


Logo of mayoclinprocLink to Publisher's site
Mayo Clin Proc. 2010 January; 85(1): 12–14.
PMCID: PMC2800293

Delirium: A Cognitive Cost of the Comfort of Procedural Sedation in Elderly Patients?

Gregory Crosby, MD and Deborah J. Culley, MD
Department of Anaesthesia, Harvard Medical School, Brigham and Women's Hospital
Boston, MA

Cognitive morbidity is a common complication of hospitalization in elderly patients.1,2 This morbidity takes several forms, with delirium, a confusional state defined clinically by acute and fluctuating changes in consciousness and attention, being the most frequent. The prevalence of delirium is about 20% in elderly hospitalized medical patients, can be as high as 60% after some types of surgery, and approaches 80% among patients admitted to a critical care unit.1,3,4 Delirium is also a costly complication in terms of length of stay and dollars, with a recent study5 estimating that the economic burden of delirium to the US health care system is more than $100 billion annually—enough for a small down payment on the health care legislation currently working its way through Congress. It is also costly from the patient's point of view because it is associated with a 2- to 5-fold higher incidence of other complications, a 3-fold greater likelihood of placement in a long-term care facility, and up to a 2-fold higher 1-year mortality.1,6,7 Against this background, the study by Sieber et al8 in this issue of Mayo Clinic Proceedings is important because it raises the tantalizing possibility that excessive sedation is a risk factor for delirium. If so, the implications are enormous because procedural sedation is widespread, and controlling depth of sedation might be a simple, inexpensive way to reduce the incidence of this cognitive morbidity.

Sieber et al8 examined the incidence of postoperative delirium in elderly patients undergoing hip fracture repair, a procedure that required about 90 minutes. All patients received spinal anesthesia plus, by random assignment, either light or deep sedation (n=57 per group). Sedation was provided by intravenous infusion of propofol, a drug used widely for procedural sedation in the operating room as well as in nonoperative settings, such as endoscopy suites, cardiac catheterization laboratories, and emergency departments. In the Seiber et al study, commercially available processed electroencephalography (EEG) (Bispectral Index [BIS]; Aspect Medical Systems, Norwood, MA) was used to determine the level of sedation. Patients were assessed for delirium preoperatively and on each of the first 3 postoperative days using a validated and widely accepted tool, the Confusion Assessment Method. In study design terms, clinical relevance is a strong point of the Sieber et al report. Propofol is commonly used for procedural sedation, and patients who sustain a hip fracture are typically elderly and have a prevalence of delirium that is among the highest of all surgical patients, facts that the data of Sieber et al confirm (eg, age, approximately 81 years; prevalence of delirium, 20%-40%). The results are remarkable. Patients managed to a light level of sedation had a 50% lower incidence of postoperative delirium than those deeply sedated with the same drug (19% vs 40%; P=.02).8 On a number-needed-to-treat basis, this means that 1 case of delirium would be prevented for every 3.5 to 4.7 patients managed by light instead of deep propofol sedation.8

See also page 18

This is an exciting and potentially actionable observation but, before applying a consciousness monitor on all elderly patients needing sedation and reducing the rate of propofol infusion, a few important caveats must be recognized.

First, despite randomization, the groups were different at baseline. More patients in the light sedation group lived independently before surgery (74% vs 56%; P=.08) and had higher Mini-Mental State Examination scores, suggesting they were more functional and cognitively intact.8 Because preoperative cognitive and functional impairments have been shown to be strongly associated with postoperative delirium,1,9 these baseline imbalances in the 2 groups could partially explain the protective effect of light sedation. Likewise, with a prevalence of postoperative delirium of 20% to 40% instead of the more typical 50% observed in previous studies in similar patients having the same surgery,3 the study may have been biased toward finding aggravation of the problem with deep sedation.

Second, propofol was the main drug used for sedation. From a practical perspective, this is a good choice because propofol, a γ-aminobutyric acid receptor modulator, is widely used for this purpose. However, the results could be propofol-specific. Sedative drugs differ in their mechanisms of action and propensity to produce delirium in elderly persons. Benzodiazepines, which are also γ-aminobutyric acid receptor modulators, are widely used for procedural sedation but probably should be used with caution. This is highlighted by 2 recent randomized, prospective trials that compared long-term sedation of patients in a critical care unit with a benzodiazepine (midazolam, lorazepam) or the α2 adrenergic receptor agonist dexmedetomidine.10,11 Even when the drugs were administered at dosages that achieved the same clinical level of sedation, patients who received a benzodiazepine had a substantially higher incidence and longer duration of delirium than those who received the α2 agonist.10,11 Although sedating a patient for days is undoubtedly different (as far as the brain is concerned) than doing so for minutes to hours, the take-home messages are that (1) clinically used sedative agents differ meaningfully in their pharmacology and end effects and (2) the agent one chooses seems likely to be as important a determinant of cognitive outcome as the depth of sedation achieved.

Third is the issue of what is meant by light and deep sedation and how to measure it. Light and deep are terms of art and have no scientific meaning. During conscious sedation, the depth of sedation is determined on the basis of responsiveness to verbal or nonverbal stimuli. For example, on the Observer's Assessment of Alertness/Sedation Scale (5 = awake; 4 = sedated but oriented and arousable to verbal commands; 3 = responds only to loud and repeated verbal commands; 2 = unresponsive to all but prodding and shaking; 1= reactive only to pain, and 0 = unresponsive), a score of 3 to 4 is considered light sedation; 1 to 2, deep sedation; and 0, general anesthesia.12 A variety of “consciousness monitors” have been developed to enhance and refine assessment of consciousness, and these have been used during general anesthesia as well as sedation.13 Sieber et al8 used one of these, the BIS, and defined light sedation as a BIS value of 80 (arbitrary units) and deep sedation as a BIS of 50. As a surrogate, objective measure for how propofol affects the brain, this is reasonable but not faultless. All consciousness-monitoring devices are controversial because putting a number on consciousness is an inexact science, and the algorithms that calculate the number that appears on the display screen are proprietary.13 On the basis of validation data obtained in adult volunteers and surgical patients, a BIS value of near 100 is expected in an awake, conscious person, a value of 60 to 90 represents sedation, and a value less than 60 corresponds to general anesthesia.13 One problem, however, is that the relationship between an absolute BIS value and the state of consciousness varies considerably across a population. During propofol sedation of a given person, unresponsiveness to all but painful stimulation (ie, deep sedation) may require a BIS as deep as 40 or as light as 70; conversely, at clinically different depths of sedation the BIS value may be similar.12,14,15 Sieber et al used the absolute BIS value rather than a within-patient change as the measure of depth of sedation and, moreover, did not include a simultaneous clinical assessment of sedation. This is a problematic omission because a good correlation between BIS values is reported by some studies16,17 but not by others, especially when it comes to discriminating between different levels of sedation.12,15 In fact, it is precisely in the range of interest—the zone from light to deep sedation—that the BIS algorithm appears to have greatest difficulty.12,15 Therefore, simultaneous assessment of level of consciousness by the BIS and a clinical sedation measure, such as the Observer's Assessment of Alertness/Sedation Scale, would have enhanced confidence that the propofol-treated groups were in fact differentially sedated. It would also have made the study more applicable to the many settings in which sedatives are administered to clinical rather than EEG targets.

Fourth, administering sedation in elderly patients to a BIS target that was defined in younger persons may be misleading because persons older than 65 years lose consciousness at a higher BIS value than young adults; the median BIS to loss of consciousness in patients older than 65 years is about 70 (range 58-91) vs 58 (range 40-70) for those aged 20 to 40 years.18 This implies that the octogenarians in the study by Sieber et al who were maintained at a BIS of 80 were deeply sedated and those at 50 were almost certainly under full general anesthesia. As it turns out, this may make the study more relevant, not less, because deep procedural sedation is more common than generally appreciated. For example, when the BIS is used to guide sedation for colonoscopy, empirical data reveal that values of 59 to 73 are typical, with a range from the 20s to 90s.17,19,20 Likewise, sedation by anesthesia professionals often results in general anesthesia by clinical sedation scores or BIS criteria.12 Importantly, the EEG-guided sedation targets are seldom adjusted for age. The point is that deep sedation, at least as determined by processed EEG, is probably common during procedural sedation in elderly patients, regardless of who gives the drugs. The data of Sieber et al8 suggest this is undesirable and that greater attention to sedation practice in elderly patients is called for because excessive sedation, even for a relatively short time, may have important cognitive implications.

Nevertheless, it would be wrong to conclude from this study that deep sedation alone causes delirium. In the Sieber et al study, the deeply sedated group received roughly 5 times more propofol than the lightly sedated group (10.5 vs 2.5 mg/kg); however, when examined on a patient-by-patient basis, there was no independent association between the dose of propofol and the incidence of delirium. The authors conclude from this that depth of sedation offends, not propofol per se. Another way to explain this seemingly paradoxical result is that the patient is the problem. That is, at any given dose of propofol, the patient at risk of delirium by virtue of some preexisting condition (eg, unrecognized mild cognitive impairment) would develop a lower BIS score than a patient without a delirium-prone brain. Unfortunately, data on the dose of propofol required to achieve the targeted levels of sedation in the subgroups of delirious and nondelirious patients are not provided, but this hypothesis predicts that the delirium-prone patients would require less propofol to reach a given sedative depth/BIS score. Put another way, sedation may be a precipitating factor for delirium in an elderly patient already predisposed to it by virtue of poor preexisting cognitive reserve. Therefore, one reasonable conclusion is that it is prudent to assess patient's cognitive and functional status before surgical and nonsurgical procedures requiring sedation to help identify those in whom an exaggerated sedative response may occur and for whom depth of sedation ought to be limited.

So where does this leave us? Ever since the first public demonstration of ether anesthesia for surgery in 1846, patients have come to expect that procedures will be performed painlessly. They expect to have no recollection of the event, even when they undergo mildly uncomfortable procedures, so the demand for sedation is not going away. The proliferation of mildly invasive procedures performed outside of operating rooms combined with the advent of sedative and anesthetic drugs with shorter half-lives and fewer cardiopulmonary adverse effects have made it both necessary and possible to satisfy patient expectations in this regard. However, at least in elderly persons, a cognitive cost may be paid for that comfort. Traditionally, deliberation about the wisdom and safety of providing procedural sedation has focused almost entirely on potential cardiopulmonary risk. The study by Sieber et al,8 in conjunction with controversial recent data proposing an association between deep sedation or anesthesia and 1-year mortality,21,22 suggests that in elderly patients it is time to move beyond the focus on cardiopulmonary risk alone and add cognitive risk to the equation. After all, the brain is the main target of sedative agents, and increasingly the brain that is being targeted is old and frail. Under those conditions, even sedation may have undesirable cognitive consequences.


This work was supported by National Institutes of Health grants RO1GM088817, R01AG030618, R03AG028189, R21AG027549, and KO8077057.


1. Inouye SK. Delirium in older persons [published correction appears in N Engl J Med. 2006;354(15):1655] N Engl J Med. 2006;354(11):1157-1165 [PubMed]
2. Rudolph JL, Marcantonio ER, Culley DJ, et al. Delirium is associated with early postoperative cognitive dysfunction. Anaesthesia 2008;63(9):941-947 [PMC free article] [PubMed]
3. Marcantonio ER, Flacker JM, Wright RJ, Resnick NM. Reducing delirium after hip fracture: a randomized trial. J Am Geriatr Soc. 2001;49(5):516-522 [PubMed]
4. Girard TD, Pandharipande PP, Ely EW. Delirium in the intensive care unit. Crit Care 2008;12(suppl 3):S3 [PMC free article] [PubMed]
5. Leslie DL, Marcantonio ER, Zhang Y, Leo-Summers L, Inouye SK. One-year health care costs associated with delirium in the elderly population. Arch Intern Med. 2008;168(15):27-32 [PubMed]
6. Marcantonio ER, Flacker JM, Michaels M, Resnick NM. Delirium is independently associated with poor functional recovery after hip fracture. J Am Geriatr Soc. 2000;48(6):618-624 [PubMed]
7. Kiely DK, Marcantonio ER, Inouye SK, et al. Persistent delirium predicts greater mortality. J Am Geriatr Soc. 2009;57(1):55-61 [PMC free article] [PubMed]
8. Sieber FE, Zakriya KJ, Gottschalk A, et al. Sedation depth during spinal anesthesia and the development of postoperative delirium in elderly patients undergoing hip fracture repair. Mayo Clin Proc. 2010;85(1):18-26 [PMC free article] [PubMed]
9. Robinson TN, Raeburn CD, Tran ZV, Angles EM, Brenner LA, Moss M. Postoperative delirium in the elderly: risk factors and outcomes. Ann Surg. 2009;249(1):173-178 [PubMed]
10. Pandharipande PP, Pun BT, Herr DL, et al. Effect of sedation with dexmedetomidine vs lorazepam on acute brain dysfunction in mechanically ventilated patients: the MENDS randomized controlled trial. JAMA 2007;298(22):2644-2653 [PubMed]
11. Riker RR, Shehabi Y, Bokesch PM, et al. SEDCOM (Safety and Efficacy of Dexmedetomidine Compared With Midazolam) Study Group Dexmedetomidine vs midazolam for sedation of critically ill patients: a randomized trial. JAMA 2009;301(5):489-499 [PubMed]
12. Chisholm CJ, Zurica J, Mironov D, Sciacca RR, Ornstein E, Heyer EJ. Comparison of electrophysiologic monitors with clinical assessment of level of sedation. Mayo Clin Proc. 2006;81(1):46-52 [PMC free article] [PubMed]
13. American Society of Anesthesiologists Task force on Intraoperative Awareness Practice advisory for intraoperative awareness and brain function monitoring: a report by the American Society of Anesthesiologists task force on intraoperative awareness. Anesthesiology 2006;104(4):847-864 [PubMed]
14. Drummond JC. Monitoring depth of anesthesia: with emphasis on the application of the bispectral index and the middle latency auditory evoked response to the prevention of recall. Anesthesiology 2000;93(3):876-882 [PubMed]
15. Nishiyama T. Auditory evoked potentials index versus bispectral index during propofol sedation in spinal anesthesia. J Anesth 2009;23(1):26-30 [PubMed]
16. Kearse LA, Jr, Rosow C, Zaslavsky A, Connors P, Dershwitz M, Denman W. Bispectral analysis of the electroencephalogram predicts conscious processing of information during propofol sedation and hypnosis. Anesthesiology 1998;88(1):25-34 [PubMed]
17. von Delius S, Thies P, Rieder T, et al. Auditory evoked potentials compared with bispectral index for monitoring of midazolam and propofol sedation during colonoscopy. Am J Gastroenterol. 2009;104(2):318-325 [PubMed]
18. Lysakowski C, Elia N, Czarnetzki C, et al. Bispectral and spectral entropy indices at propofol-induced loss of consciousness in young and elderly patients. Br J Anaesth. 2009;103(3):387-393 [PubMed]
19. Chen SC, Rex DK. An initial investigation of bispectral monitoring as an adjunct to nurse-administered propofol sedation for colonoscopy. Am J Gastroenterol. 2004;99(6):1081-1086 [PubMed]
20. Drake LM, Chen SC, Rex DK. Efficacy of bispectral monitoring as an adjunct to nurse-administered propofol sedation for colonoscopy: a randomized controlled trial. Am J Gastroenterol. 2006;101(9):2003-2007 [PubMed]
21. Monk TG, Saini V, Weldon BC, Sigl JC. Anesthetic management and one-year mortality after noncardiac surgery. Anesth Analg. 2005;100(1):4-10 [PubMed]
22. Lindholm ML, Träff S, Granath F, et al. Mortality within 2 years after surgery in relation to low intraoperative bispectral index values and preexisting malignant disease. Anesth Analg. 2009;108(2):508-512 [PubMed]

Articles from Mayo Clinic Proceedings are provided here courtesy of The Mayo Foundation for Medical Education and Research