Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Email this article to a Colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Articles & Searches Download to citation manager Citing Articles via Google Scholar Articles by Seaman, J. S. Articles by Brown, T. M. Search for Related Content PubMed Citation Articles by Seaman, J. S. Articles by Brown, T. M. Delirium Syndromes Secondary to General Medical Disorders

Impaired Oxidative Metabolism Precipitates Delirium: A Study of 101 ICU Patients

Received July 13, 2004; revised February 14, 2005; accepted March 7, 2005. From the University of Oklahoma Health Sciences Center (JSS); the Department of Psychiatry, University of Texas Health Sciences Center, San Antonio (JS); Wilford Hall Medical Center, Lackland Air Force Base, TX (DC); and the Department of Psychiatry, Audrey Murphy Veterans Hospital, San Antonio, TX (TMB). Address correspondence and reprint requests to Dr. Seaman, OUHSC Dept of Psychiatry, P.O. Box 26901, OK City, OK 73190. e-mail: jeff-seaman{at}ouhsc.edu

ABSTRACT

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 
Data from 101 consecutively admitted intensive care unit (ICU) patients were examined to determine whether oxidative metabolic stress existed within the 48 hours before delirium onset. The occurrence of pneumonia and sepsis at any time during hospitalization was also recorded. Delirium was defined retrospectively with the Confusion Assessment Method (CAM). Older patients were found to develop delirium more frequently than younger patients. There were no differences in illness severity (APACHE II) between those who developed delirium and those who did not. Three measures of oxygenation (hemoglobin, hematocrit, pulse oximetry) were worse in the patients who later developed delirium. Two measures of oxidative stress (sepsis, pneumonia) occurred more frequently among those diagnosed with delirium. Hence, patients with indicators of oxidative dysfunction developed delirium more frequently, and this was not linked to illness severity.

INTRODUCTION

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 
Delirium is often viewed as a nonspecific syndrome, a synonym for any acute confusional state without a clear cause. Indeed, delirium is at times regarded as merely a symptom of other medical illnesses. This syndromal paradigm has provided little impetus to search for a distinct pathophysiology of delirium or for treatments targeting a distinct etiology.

There are growing arguments that delirium is more than a symptom of other medical disorders.^1–4 This concept is not entirely new—in fact, Engel and Romano argued in favor of such a classification 50 years ago.⁵ If delirium is not merely a syndrome, but also a disease entity, there should be a consistent symptom pattern, an identifiable pathology linked to the symptoms, and treatments specifically targeted to the pathology.

Previous efforts^6–8 have established a consistent symptom phenomenology for delirium. There has been little investigation, however, regarding delirium pathogenesis. Many studies have focused on reporting risk factors, but without a link to understanding core etiology. Fortunately, there are descriptions of abnormal neurotransmitter patterns that occur within delirium.^2,9 Several authors^3–5 have also asked whether dysfunction in the brain’s oxygen supply, demand, and utilization may underlie these neurotransmitter changes. This study questioned whether patients with evidence of impaired oxidative metabolism develop delirium more often than those without such impairment, regardless of overall illness severity.

METHOD

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 
We performed a retrospective review of 101 consecutive admissions to the intensive-care unit (ICU) at Wilford Hall Medical Center, a Level-1 Trauma and Tertiary Care Hospital in San Antonio, Texas. Data were extracted from inpatient medical records and the hospital’s computerized laboratory results system by the 2nd and 3rd coauthors (JS, DC). The information was recorded by the investigators in such a manner that subjects could not be identified directly or through identifiers linked to the subjects. Standard hospital laboratory tests for complete blood count, blood gases, temperature, and bedside "finger-probe" oxygen saturations were used. The accuracy and precision of these instruments were in keeping with CLIP standards.

The dependent variable was whether delirium developed or not at any time during the ICU stay. The charts were examined for patient descriptions that would lend clear evidence to diagnosing delirium or not. A standardized diagnosis of delirium was made via the Confusion Assessment Method (CAM).¹¹ The CAM is a nine-item diagnostic instrument used to detect four key features of delirium: acute onset and fluctuating course of mental status changes, inattention, disorganized thinking, and altered level of consciousness. To diagnose delirium with the CAM, the patient must have documentation of the first two features and then either one or both of the second two features. The entire chart was reviewed, including physician, allied health, and nursing notes. We looked for specific documentation of cognitive and behavioral descriptions that satisfied the CAM diagnostic criteria. The CAM is a valid and reliable indicator of delirium, with a sensitivity, specificity, positive predictive accuracy, and negative predictive accuracy all greater than 90%. The inter-observer reliability of the CAM is also high (=0.81–1.0). Although this instrument was designed for prospective use, it has been successfully applied retrospectively.¹²

We investigated whether there were relative differences between the two groups (delirium versus non-delirium) across the following independent variables that might suggest increased oxidative stress. These measures sought to represent a sampling of oxygen availability, delivery, and requirement.

1. Hemoglobin (average of the two lowest values for the 48 hours before documented delirium onset, or average of two lowest values in the first 48 hours of admission, for non-delirium patients. In rare instances, if there was only one sample drawn during that period, that value was used as the sole data-point)

2. Hematocrit (same as Hemoglobin format above)

3. O₂ saturation (same as above)

4. PaO₂ (same as above)

5. Temperature (average of the two highest readings for the 48 hours before documented delirium onset, or average of the two highest readings in the first 48 hours after admission, for non-delirium patients)

6. Chart diagnosis of pneumonia or not

7. Chart diagnosis of sepsis or not

8. Number of recorded episodes of bradycardia (up to 48 hours before delirium onset of any heart rate (HR)<60, or, for the non-delirium group, first 48 hours in hospital. No more than one episode per hour could be recorded)

We also applied The APACHE II Severity-of-Disease Classification System.¹³ It is readily applicable for retrospective use. For estimating death rates, for example, at a 47% sensitivity level, the specificity is 95%, positive predictive value is 70%, and negative predictive value is 88%.¹³

Included were all medical patients (N=101) consecutively admitted to the Wilford Hall Medical Center’s Medical ICU from February 5, 2002 to May 17, 2002. The ICU patients were chosen over patients admitted directly to regular medical or surgical wards because of the high incidence of delirium (30%–50%) in the ICU population,¹⁷ as well as the extensive documentation and laboratory data available for these high-acuity patients. If a patient was transferred to a general-medical ward after their ICU stay, they remained in the study.

A total of 25 patient charts were excluded from the study. Twelve of these had evidence of alcohol or illicit drug intoxication or withdrawal. These patients can display some symptoms that are shared with delirium, but may require manipulation of the GABA/glutamate systems for treatment. In fact, these patients are believed to have a different etiopathogenesis, which is consistent with their unique EEG pattern.¹⁴ Six patients with hepatic encephalopathy were also excluded for similar reasons, even though this population has been demonstrated to have some commonalities with non-liver-failure-related delirium.^8,15,16 Also excluded were seven patients intubated at the time of admission, and those below the age of 18. There is little published research on delirium in young patients, and, thus, it was unclear what differences there might be in delirium manifestations and etiology between these two populations.

The statistical analysis was accomplished with SPSS V.11 for Windows. For the parametric data, we used Student’s t-tests, with Levine’s test for equal variances. For the nonparametric data, Mann-Whitney U tests were applied. The ordinal data results (sepsis and pneumonia incidence) were examined with Pearson chi-square, with Fisher’s exact test. All were analyzed as one-tailed, with an alpha value of 0.05.

RESULTS

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 
Overall, 30 of the 101 patients reviewed were found to have documented evidence of delirium during their inpatient stay in the ICU. As with many other delirium studies,^18–21 the average age of patients who developed delirium was older than those who did not (68.8 versus 60.8; z = –2.517; p=0.012). The two groups did not, however, differ in illness severity scores as measured by the APACHE II (Table 1).

View this table: [in this window] [in a new window]

TABLE 1. Age and APACHE II Score

The recorded hemoglobin (10.35 g/dl versus 11.61 g/dl; F=7.6; df=75.3; p<0.004) and hematocrit values (31.71% versus 35.38%; F=6.8; df=74.8; p<0.006) were both lower in the patients who later developed delirium (see Figure 1 and Figure 2). Averages of the two lowest pulse-oximetry readings 48 hours before delirium onset or in the first 48 hours after admission were also found to be significantly lower in the delirium group (88.25%) versus the non-delirium group (93.27%; F=12; df=31.9; p<0.006; see Figure 3). Arterial oxygen did not differ between the two groups, although only 23/30 of the delirium and 24/101 of the non-delirium patients had an arterial blood gas determination made.

View larger version (18K): [in this window] [in a new window]

FIGURE 1.  Premorbid Hemoglobin Note: p<0.004; Student’s t-test with equal variances not assumed (Levene’s test).

View larger version (19K): [in this window] [in a new window]

FIGURE 2.  Premorbid Hematocrit Note: p<0.006; Student’s t-test with equal variances not assumed.

View larger version (20K): [in this window] [in a new window]

FIGURE 3.  Premorbid Pulse-Oximetry Note: p<0.006; Student’s t-test with equal variances not assumed.

No significant group differences were identified in temperature (delirium group: mean 99.7°F, and non-delirium group: mean 99.1°F) or bradycardic episodes (both groups had medians of zero) in the defined study period. Sepsis, however, occurred more frequently among patients in the delirium (27%) than the non-delirium group (6%; ²=8.9; df=1; p<0.006; see Table 2). Similarly, pneumonia occurred more often in the delirium (40%) than in the non-delirium group (13%; ²=9.6; df=1; p<0.003).

View this table: [in this window] [in a new window]

TABLE 2. Sepsis and Pneumonia

DISCUSSION

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 
To the authors’ knowledge, this is the first study to objectively correlate the development of delirium with findings suggestive of premorbid oxidative metabolism impairment. Deficits and stressors consistent with impaired oxidative metabolism occurred more frequently or to more severe degrees among patients who later developed delirium.

In hindsight, there were several limitations to this study design. The raters were not blind to the CAM scores and thus may have introduced a bias into the data collection. On the other hand, the data-points were either strictly objective or tightly-defined, and this limited the opportunity for bias effects. Also, it was likely that a number of hypoactive delirium patients were overlooked and not identified as such. It is well established that hyperactive delirium patients are more likely to be noticed and documented than hypoactive patients.²² Hence, it is suspected the sample was skewed toward the hyperactive subset that has previously been found to be more resilient and to have a better prognosis.^22,23 Theoretically, had there been more hypoactive, and thus more severely ill, delirium patients included in the data set, the results may have shown a difference in APACHE II scores as well as larger differences in the studied variables. The solution to recapturing the hypoactive patients would be a prospective design, which would not rely solely on chart documentation of delirium symptoms.

Temperature and bradycardic episodes were chosen as potential markers of oxidative stress. Increases in temperature are known to increase oxygen demand.²⁴ We further assumed that pathologic bradycardia would be a harbinger of inadequate oxygen delivery in an older patient population with limited compensatory reserves. Neither of these two variables, however, demonstrated an effect on delirium diagnosis. Bradycardia events were rare, and temperature differences were very small. A larger sample size, with greater statistical power, may have identified a temperature effect. The bradycardia variable might be substituted for a more clinically relevant and yet commonly encountered measure of cardiac output for future study.

There was a large group disparity in the number of patients who did not have arterial blood-gas data; 77% of the delirium patients and only 34% of the non-delirium patients had PaO₂ levels drawn. This represented a marked selection bias for this variable and made the results regarding the PaO₂ suspect. Hence, we would argue this null finding did not disprove our hypothesis. Namely, it is suggested that as oxygen deficits and oxidative metabolic impairments increase in severity, delirium risk increases, and nonreversible brain insults might even occur.^8,25 This possibility is supported by the findings on pulse-oximetry, in which the patients who later developed delirium had modest, but clinically significant, deficits of oxygen saturation. But again, a more ubiquitous sampling of arterial blood gases is required to clarify if and at what thresholds PaO₂ deficits can precipitate delirium in a vulnerable population.

It was striking to find sepsis and pneumonia as being associated with greater rates of delirium in the absence of a difference in illness severity. This supports the hypothesis of oxidative metabolic stress as having a role in the pathogenesis of delirium. Pneumonia is known to lower the oxygen supply:demand ratio, and sepsis has been documented²⁶ to result in decreased brain-oxygen delivery and utilization. Both sepsis and pneumonia have been associated with impaired mitochondrial functioning,^27,28 which strains oxidative metabolism further, thus increasing the risk of brain dysfunction. Future work is needed to clarify other metabolic processes associated with infection that affect delirium-genesis, independent of oxidative stress.

Understanding the role of oxygen systems in the pathophysiology of delirium may permit earlier recognition of important metabolic deficits, improved prevention and treatment of those deficits, and an improvement in the otherwise "grave prognosis" of patients who develop delirium. Optimizing brain oxygen supply, transport, and utilization with delirium-prevention in mind could lead to a re-setting of treatment-decision thresholds currently in place. For instance, the decision to transfuse packed red blood cells might now include a discussion of the need to prevent delirium. Protection of the brain against delirium would be a new indication for vigorous interventions intended to optimize neuronal oxidative metabolism.

Some authors²⁹ note that impairments of oxidative metabolism, if they involve the mitochondria, have reached a level of dysfunction at which simple improvements in oxygenation are inadequate to restore mitochondrial functioning. The brain is the largest consumer of ATP at rest, and deficits of oxidative metabolism may well be manifested early as delirium. Such deficits not only impair the functioning of the neuronal plasma membrane, but of interneuronal signaling, as well.^30,31 Protection of the brain against the consequences of impaired oxidative metabolism, such as the altered release of neurotransmitters,^2,3,8 impaired functioning of the neuronal plasma membrane Na+/K+-ATPase, and the antagonism of neuronal glycolytic energy production and mitochondrial functioning^32,33 may have a role in more precise delirium-prevention and treatment options. With further clarification of the pathophysiology of delirium, the ability to identify and resist the oxidative dysfunction associated with delirium may provide a pathway to organ protection and resiliency.

ACKNOWLEDGMENTS

 
This study was performed without grant support. The authors do not have any financial or commercial interests related to this work.

The authors thank George Grossberg, M.D., for his insightful editorial suggestions.

REFERENCES

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 

1. Trzepacz PT: The neuropathogenesis of delirium: a need to focus our research. Psychosomatics 1994; 35:374–391[Abstract/Free Full Text]
2. Trzepacz PT: Is there a final common neural pathway in delirium? focus on acetylcholine and dopamine. Sem Clin Neuropsychiatry 2000; 5:132–148[Medline]
3. Brown TM: Basic mechanisms in the pathogenesis of delirium, in The Psychiatric Care of the Medical Patient, 2nd Ed. Edited by Stoudemire A, Fogel BS, Greenberg DB. New York, Oxford University Press, 2000, chap. 35
4. Francis J, Kapoor WN: Prognosis after hospital discharge of older medical patients with delirium. J Am Geriatr Soc 1992; 40:601–606[Medline]
5. Inouye SK, Rushing JT, Foreman MD, et al: Does delirium contribute to poor hospital outcomes? a three-site epidemiologic study. J Gen Intern Med 1998; 13:234–242[CrossRef][Medline]
6. Marcantonio ER, Flacker JM, Michaels M, et al: Delirium is independently associated with poor functional recovery after hip fracture. J Am Geriatr Soc 2000; 48:618–624[Medline]
7. O’Keeffe ST, Lavan J: The prognostic significance of delirium in older hospital patients. J Am Geriatr Soc 1997; 45:174–178[Medline]
8. Bourgeois JA, Seaman JS, Servis ME: Delirium, dementia, and amnestic disorders, in American Psychiatric Press Textbook of Clinical Psychiatry, 4th Ed. Edited by Hales R, Yudofsky S. Arlington, VA, American Psychiatric Publishing, 2003, pp 259-308
9. Fann JR, Burington B, Roth-Roemer S, et al: Delirium and affective and cognitive outcomes in patients undergoing stem-cell transplantation (abstract). Ann Behav Med 2001; 23(suppl):S064
10. Engel GL, Romano J: Delirium: a syndrome of cerebral insufficiency. J Chron Dis 1959; 9:260–277[CrossRef][Medline]
11. Inouye SK, van Dyck CH, Alessi CA, et al: Clarifying the confusion: the Confusion Assessment Method. Ann Intern Med 1990; 113:941–948[Medline]
12. Bates DW, Pruess K, Souney P, et al: Serious falls in hospitalized patients: correlates and resource utilization. Am J Med 1995; 99:137–143[CrossRef][Medline]
13. Knaus WA, Draper EA, Wagner DP, et al: APACHE II: a severity-of-disease classification system. Crit Care Med 1985; 13:818–829[Medline]
14. Kenard MA, Bueding E, Wortis WB: Some biochemical and electroencephalographic changes in delirium tremens. Q J Stud Alcohol 1945; 6:4–14
15. Meagher DJ, Trzepacz PT: Delirium phenomenology illuminates pathophysiology, management, and course. J Geriatr Psychiatry Neurol 1998; 11:150–156[Medline]
16. Trzepacz PT, Maue FR, Coffman G, et al: Neuropsychiatric assessment of liver-transplantation candidates: delirium and other psychiatric disorders. Int J Psychiatry Med 1986; 7:101–111
17. McNicoll L, Pisani MA, Zhang Y, et al: Delirium in the intensive care unit: occurrence and clinical course in older patients. J Am Geriatr Soc 2003; 51:591–598[CrossRef][Medline]
18. Foy A, O’Connell D, Henry D, et al: Benzodiazepine use as a cause of cognitive impairment in elderly hospital inpatients. J Gerontol 1995; 50(suppl):M99-M106
19. Henon H, Lebert F, Durieu I, et al: Confusional state in stroke: relation to preexisting dementia, patient characteristics, and outcome. Stroke 1999; 30:773–779[Abstract/Free Full Text]
20. Marcantonio ER, Goldman L, Mangione CM, et al: A clinical predictive rule for delirium after elective noncardiac surgery. JAMA 1994; 271:134–139[Abstract]
21. Schor JD, Levkoff SE, Lipsitz LA, et al: Risk factors for delirium in hospitalized elderly. JAMA 1992; 267:827–831[Abstract]
22. Liptzin B, Levkoff SE: An empirical study of delirium subtypes. Br J Psychiatry 1992; 161:843–845[Abstract/Free Full Text]
23. Reyes RL, Bhattacharyya AK, Heller D: Traumatic head injury: restlessness and agitation as prognosticators of physical and psychological improvement in patients. Arch Phys Med Rehabil 1981; 62:20–23[Medline]
24. Weissman C, Kemper M, Damask M, et al: Effect of routine intensive care interactions on metabolic rate. Chest 1984; 86:815–818[Abstract/Free Full Text]
25. Rolett EL, Azzawi A, Liu KJ, et al: Critical oxygen tension in rat brain: a combined (31)P-NMR and EPR oximetry study. Am J Physiol Regul Integr Comp Physiol 2000; 279:R9-R16
26. Maekawa T, Fujii Y, Sadamitsu D, et al: Cerebral circulation and metabolism in patients with septic encephalopathy. Am J Emerg Med 1991; 9:139–143[CrossRef][Medline]
27. Chuang YC, Tsai JL, Chang AY, et al: Dysfunction of the mitochondrial respiratory chain in the rostral ventrolateral medulla during experimental endotoxemia in the rat. J Biomed Sci 2002; 9:542–548[Medline]
28. Crouser ED, Julian MW, Blaho DV, et al: Endotoxin-induced mitochondrial damage correlates with impaired respiratory activity. Crit Care Med 2002; 30:276–284[CrossRef][Medline]
29. Fink MP: Cytopathic hypoxia: is oxygen use impaired in sepsis as a result of an acquired intrinsic derangement in cellular respiration? Crit Care Clin 2002; 18:165–175[CrossRef][Medline]
30. Gibson GE, Blass JP: Impaired synthesis of acetylcholine in brain accompanying mild hypoxia and hypoglycemia. J Neurochem 1976; 27:37–42[CrossRef][Medline]
31. Khakh BS, Gittermann D, Cockayne DA, et al: ATP modulation of excitatory synapses onto interneurons. J Neurosci 2003; 23:7426–7437[Abstract/Free Full Text]
32. Muriel P, Castaneda G, Ortega M, et al: Insights in the mechanism of erythrocyte N+/K+-ATPase inhibition by nitric oxide and peroxynitrite anion. J Appl Toxicol 2003; 23:275–278[CrossRef][Medline]
33. Brorson JR, Schumaker PT, Zhang H: Nitric oxide inhibits neuronal energy production. The Committees on Neurobiology and Cell Physiology. J Neurosci 1999; 91:147–158

Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Email this article to a Colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Articles & Searches Download to citation manager Citing Articles via Google Scholar Articles by Seaman, J. S. Articles by Brown, T. M. Search for Related Content PubMed Citation Articles by Seaman, J. S. Articles by Brown, T. M. Delirium Syndromes Secondary to General Medical Disorders

Get information about faster international access.

Privacy Policy

Copyright © 2006 Academy of Psychosomatic Medicine. All rights reserved.

Home | Search | Current Issue | Past Issues | Subscribe | All APPI Journals | Help | Contact Us