DELIRIUM is a common medical condition, reported to

Transcription

1 Journal of Gerontology: MEDICAL SCIENCES 2005, Vol. 60A, No. 12, Copyright 2005 by The Gerontological Society of America Are Preoperative Depressive Symptoms Associated With Postoperative Delirium in Geriatric Surgical Patients? Jacqueline M. Leung, 1 Laura P. Sands, 2 E. Ann Mullen, 1 Yun Wang, 3 and Linnea Vaurio 1 1 Department of Anesthesia and Perioperative Care, University of California, San Francisco. 2 School of Nursing and 3 Department of Statistics, Purdue University, West Lafayette, Indiana. Background. Delirium is common in older surgical patients and predicts negative health outcomes. Whether depressive symptoms are prevalent and predict the development of postoperative delirium in elderly surgical patients has not been investigated. Our study aimed to examine the prevalence and prognostic importance of preoperative depressive symptoms in older surgical patients. Methods. Patients at least 65 years of age who were scheduled for major noncardiac surgery were recruited. Preoperatively, patients were screened for symptoms of depression using the 15-point Geriatric Depression Scale (GDS). Postoperative delirium was measured using the Confusion Assessment Method. The associations between covariates and preoperative depression, and postoperative delirium were determined by multivariate logistic regression. Results. In the 219 patients studied, the mean age was years, 12% of patients reported six or more depressive symptoms, and 32% reported three to five depressive symptoms. By multivariate logistic regression, patient characteristics associated with six or more symptoms of depression included,12 years of education, moderate to severe limitation in functional status, and drinking more than two alcoholic beverages per day. Postoperatively, 46% of patients developed delirium. Patients with a greater number of preoperative depressive symptoms were more likely to develop postoperative delirium ( p ¼.048) and experience a longer duration of postoperative delirium ( p ¼.027). Even after adjusting for covariates associated with depression and postoperative delirium including age, educational level, functional status, and preoperative alcohol use patients with more than six preoperative depressive symptoms were still significantly more likely to have a longer duration of postoperative delirium than did those patients with fewer than two depressive symptoms (odds ratio ¼ 2.69, confidence interval ¼ ). Conclusion. Preoperative screening for the presence of depressive symptoms can be performed easily in elderly patients, and yields useful prognostic information relating to postoperative delirium. DELIRIUM is a common medical condition, reported to occur in 14% 50% of hospitalized medical patients, with an associated mortality rate ranging from 10% to 65% (1,2). Elderly surgical patients, in particular, are at risk of developing postoperative delirium, which occurs in 10% 70% of patients (3 8). The conventional approaches in surgery and anesthesia research have focused on using disease-specific concepts to risk prediction and outcome measurements. For example, previous studies of postoperative delirium have identified some personal factors (such as age and preoperative medical conditions) and surgical factors (including surgery duration, blood loss, and postoperative hematocrit level) that increase the likelihood of postoperative delirium (9 11). However, delirium is likely a multifactorial syndrome (12), whereby elderly individuals who are vulnerable preoperatively, when subjected to the stress of major surgery, develop postoperative delirium. This concept is analogous to recent developments in geriatrics which have demonstrated the importance of including not just the disease status of the individual, but other geriatric risk factors such as cognitive status, the presence of depression, and functional dependence as baseline risk factors that increase geriatric patients vulnerability for negative health outcomes (13 17). In the perioperative period, evaluation of patients psychological status has often been overshadowed by the focus on multiple comorbid medical conditions that are common in elderly individuals awaiting surgery. Preoperative assessment of patients psychological status may be informative in predicting subsequent poor surgical outcomes, because psychological stress may be a harbinger for the development of delirium (18). Accordingly, our study sought to determine whether the presence of multiple depressive symptoms preoperatively is common in older surgical patients and predictive of the development of postoperative delirium. METHODS Patient Recruitment Institutional review board approval and written informed consent from each patient were obtained. The study was conducted from 2001 through 2003 at the University of California, San Francisco Medical Center, and the patients included in the current report are the subset of a larger ongoing study evaluating the pathophysiology of postoperative delirium in elderly patients. Included in the study were consecutive English-speaking patients 65 years of age who were scheduled for major noncardiac surgery requiring anesthesia and who were expected to remain in the hospital postoperatively for more than 48 hours. Excluded from the study were patients who were incapable of providing written informed consent. Three patient interviews were conducted in person by the same interviewer. The preoperative interview, occurring 1563

2 1564 LEUNG ET AL. Table 1. Characteristics of Patients (N ¼ 219) Demographic and Clinical Data Patients (%) Age, y (mean 6 SD) Sex Female 51.6 Male 48.4 Race White 86.2 Non-White 13.8 Educational level High school graduate or incomplete 27.1 Incomplete college or above 72.9 Living arrangements Home of spouse or close relatives 67.1 Home alone 30.6 Other 2.3 Daily alcohol intake 0 2 drinks 90.6 >2 drinks 9.4 Functional status No limitation 33.8 Some limitation of activities 66.2 Independent in performing five ADLs 82.2 Independent in performing seven IADLs 64.2 Pain (VAS at rest) Pain (VAS with movement) Preoperative self-reported sleep disturbance 49.4 TICS scores (mean 6 SD) (range 10 39) Notes: SD ¼ standard deviation; TICS ¼ telephone interview of cognitive status; ADL ¼ Activities of Daily Living; IADL ¼ Instrumental Activities of Daily Living; VAS ¼ Visual Analog Scale (0 10), where 0 ¼ no pain, and 10 ¼ the most severe pain.,48 hours prior to surgery in the preoperative clinic, included the assessment of depressive symptoms, medical history focusing on neurological status, and assessment of patients cognitive status, pain, and functional status. The two postoperative interviews which focused on the assessment of cognitive status were conducted approximately 24 and 48 hours after surgery in the patients hospital rooms. Assessment of Depressive Symptoms and Cognitive Status The number of symptoms of depression was determined by administration of the 15-item Geriatric Depression Scale (GDS) (19), which is adapted from the 30-item GDS and has high validity compared to the Hamilton Rating Scale for Depression and the Zung Self-Rating Depression Scale (20). The GDS was selected because it is an easy selfadministered survey that can be applied in the preoperative period without substantial time involvement. The score on the GDS is the total number of symptoms of depression reported by the patient. We categorized the number of symptoms such that 6 points represented a high number, 3 5 points represented an intermediate number, and 0 2 points represented few to no symptoms of depression. Preoperative cognitive function was measured using the Telephone Interview for Cognitive Status (TICS) test (adapted from the Mini-Mental Status Examination), which can be administered in person or over the phone (21). Trained interviewers administered the TICS in person during the preoperative interview. Potential Factors Associated With Preoperative Depressive Symptoms Potential covariates of preoperative depressive symptoms were measured through direct patient interview during the preoperative interview and a review of medical records. The covariates measured included age, sex, race, educational level, alcohol intake, and living arrangement. Other selfreported symptoms included pain (at rest and with movement), which was measured using a verbal version of the visual analog scale (22), in which 0 ¼ no pain and 10 ¼ maximal pain experienced. Functional status was determined using the Activities of Daily Living (ADL) scale (23), the Instrumental Activities of Daily Living (IADL) scale (24), and self-perceived overall limitation in daily activities (25), which was measured as none, intermittent, mild, moderate, or severe. The number and severity of preoperative coexistent conditions were measured using the Charlson comorbidity index (26). Other perioperative data obtained from chart review included type of surgery and the American Society of Anesthesiologists (ASA) classification, a clinically based risk index, which incorporated the number and severity of preoperative comorbid conditions (27). Surgical risk was estimated using the guidelines from the American College of Cardiology and the American Heart Association update for the perioperative cardiovascular evaluation for noncardiac surgery which takes into consideration the type and duration of surgery and intraoperative blood loss (28). Outcomes Measurement During the three interviews, the trained interviewers determined the presence and duration of delirium using the Confusion Assessment Method (29). All assessments of postoperative delirium were validated by a second investigator (L.P.S.). Onset of delirium was determined by the patient meeting Confusion Assessment Method criteria for delirium on either the first or second postoperative day assessments. Duration of delirium refers to whether delirium was present 0, 1, or 2 days after surgery. The potential covariates of postoperative delirium were selected on the basis on findings from previous studies in surgical patients (9), and included variables listed in Tables 1 and 2. Statistical Analysis Exploratory data analyses were carried out using univariate statistics to examine distributions of key variables. To determine the association between presence of preoperative depressive symptoms and potential predictors, we conducted

3 DEPRESSION AND POSTOPERATIVE DELIRIUM 1565 analyses of variance, Wilcoxon rank-sum tests, chi-square tests, and Fisher s Exact test using SAS 8.2 (Cary, NC). To determine which variables were independently associated with a high number of preoperative depressive symptoms (six or more symptoms), variables that had significant bivariate association with preoperative depressive symptoms (p value.2) were entered in a forward multivariable logistic regression model. Odds ratios (ORs), 95% confidence intervals (CIs), and p values were reported. A p value less than.05 was considered statistically significant. Comparison of the onset of delirium and duration of postoperative delirium between patients with and without preoperative depression was performed using the chi-square test for trend. For each outcome, multivariate models were run in two stages. First, all variables listed in Tables 1 and 2 that had a significant bivariate association with depression and the outcome were entered in a forward multivariable logistic regression model. The model was then trimmed to include only those covariates that significantly and independently predicted the outcome. RESULTS Two hundred nineteen patients were included in the study. The mean age of the patients was years (range years) (Table 1). Forty-eight percent were male. Preoperative self-reports indicated that 44% of the patients complained of moderate to severe limitations in regular activities. However, analysis of ADL and IADL revealed that 82% could independently perform ADLs and 64% could independently perform IADLs. Sixteen percent of patients had pain scores.4 at rest, and 43% had pain scores.4 with movement. The majority of patients underwent orthopedic (47%), urologic (22%), or gynecological (13%) surgery, and 57.8% of patients had 2 coexistent chronic medical conditions. Fifty percent of patients were classified as ASA class 3, and 51% of patients had Charlson comorbidity indices of 2 (Table 2). By preoperative screening, 12% of patients had a GDS score of 6, 32% had scores of 3 5, and 56% had scores of 0 2. In contrast, only three patients (1.37%) had depression noted by medical record review. Nine percent of patients were taking antidepressant medications preoperatively (Table 2). Bivariate analyses of variables associated with preoperative depressive symptoms as determined by GDS scores are shown in Table 3. The variables included in the multivariate logistic regression analysis included age, sex, educational level, self-report of overall functional status, preoperative pain at rest and with movement, dependency in performing 1 ADL or IADL, TICS score, ASA classification, alcohol intake, and type of surgery. By multivariate logistic regression, a large number of preoperative depressive symptoms (GDS score 6) is associated with lower educational level (incomplete high school vs higher, OR ¼ 2.89, 95% CI ¼ , p ¼.026), moderate to severe versus no or mild limitation in functional status (OR ¼ 7.83, 95% CI ¼ , p ¼.0005), and having more than two alcoholic drinks per day (OR ¼ 10.1, 95% CI ¼ , p ¼.0012). Table 2. Surgical and Medical Data Patient (%) Types of surgery* Orthopedic 46.8 Urologic/gynecologic 34.3 General 9.1 Vascular/neurologic 9.7 Other 1.4 Surgical risk Low 3.2 Intermediate 80.4 High 16.4 Coexisting medical conditions Neurologic 9.2 Cardiovascular 20.2 Renal/endocrine 17.8 Cancer 48.4 Number of coexistent medical conditions Charlson comorbidity index Mean ASA classification 1 and Preoperative medications Antidepressants 9.1 Benzodiazepines 11.4 Narcotics 24.2 Notes: ASA classification ¼ American Society of Anesthesiologists physical status classification (1 ¼ healthy patient; 2 ¼ patient with mild systemic disease; 3 ¼ patient with moderate systemic disease; 4 ¼ patient with severe systemic disease that is incapacitating; 5 ¼ patients who are moribund). *Patients may have more than one type of surgery concurrently. Postoperatively, 46% of patients were determined to have developed delirium by day 1 and/or day 2 after surgery. Bivariate analysis (Table 4) showed that there was an ordered association between the number of depressive symptoms and postoperative delirium: The greater the number of depressive symptoms, the greater the incidence ( p ¼.048) and the longer the duration of postoperative delirium (p ¼.027). Two-stage multivariable analyses revealed that a large number of depressive symptoms was significantly associated with the duration of delirium (OR ¼ 2.69, 95% CI ¼ ) even after statistically adjusting for age (OR ¼ 2.32, 95% CI ¼ ), educational level (OR ¼ 1.29, 95% CI ¼ ), functional status (OR ¼ 0.89, 95% CI ¼ ), and preoperative alcohol intake (OR ¼ 0.70, 95% CI ¼ ). Other potential covariates of postoperative delirium which were included in the first stage of the multivariable analyses but were not significant included sex, educational level, IADL score, ASA classification, diagnoses, and TICS score.

4 1566 LEUNG ET AL. Table 3. Bivariate Analysis of Factors Associated With Preoperative Depression GDS Score Variable 3 5 6þ p Value Age % 9%.70 36% 14% Sex.002 Female 40% 23% Male 24% 8% Highest level of education.001 y High school or less 39% 23% Incomplete college or above 30% 9% Functional status.003 y No limitations of activities 26% 5% Some limitations of activities 35% 16% Preoperative VAS pain score at rest.03 y 0 32% 10% % 13% 5 35% 24% Preoperative VAS pain score with movement.01 y 0 27% 9% % 6% 5 39% 18% Independent in five ADLs,.0001 No 44% 31% Yes 29% 8% Independent in seven IADLs,.0001 No 38% 23% Yes 29% 6% TIC score (mean 6 SD).003 (for GDS score 0 2: ) ASA classification.04 y 1 or 2 30% 8% 3 or 4 34% 16% Alcohol intake.03* 0 2 drinks/day 32% 10%.2 drinks/day 32% 32% Orthopedic surgery.06 No 26% 11% Yes 39% 14% Urologic surgery.02 No 34% 15% Yes 23% 4% Vascular surgery.08 No 33% 11% Yes 15% 31% Note: The percentages of patients with GDS scores 3 5 vs 6 were compared according to the presence or absence of the variables. ASA classification ¼ American Society of Anesthesiologists physical status classification. Covariates which were analyzed but not significant included race, living arrangement, history of depression, history of stroke, history of myocardial infarction, history of congestive heart failure, history of renal disease, history of cancer, number of preoperative comorbid conditions, preoperative use of antidepressants, benzodiazepines, narcotics, general/orthopedic/gynecological/ neurological surgery, and preoperative self-reported sleep disturbance. *By Fisher s Exact test. y By Mantel Haenszel chi-square test. Only three patients had chart documentation of a history of depression. GDS ¼ Geriatric Depression Scale; VAS ¼ Visual Analog Scale; ADL ¼ Activities of Daily Living; IADL ¼ Instrumental Activities of Daily Living; SD ¼ standard deviation. Table 4. Relationships of Preoperative Depression and Postoperative Delirium Variable GDS, 2 GDS 3 5 GDS 6 p Value Delirium by CAM 38.7% 56.5% 56.0%.048 Delirium duration day 79.8% 62.7% 52% 2 days 20.2% 37.3% 48% Notes: CAM ¼ Confusion Assessment Method; GDS ¼ Geriatric Depression Scale. DISCUSSION Our study demonstrates that preoperative depressive symptoms are common in a consecutive cohort of elderly patients awaiting elective noncardiac surgery. Importantly, patients with a large number of depressive symptoms are more likely to develop postoperative delirium and have a longer duration of postoperative delirium. Even after adjusting for covariates of postoperative delirium, a large number of preoperative depressive symptoms remain a significant and independent predictor of longer duration of postoperative delirium. The findings also reveal a pattern of increasing vulnerability to a greater number of preoperative depressive symptoms; this pattern was not assessed by previous work (30), which suggests that the associations between preoperative depressive symptoms and postoperative outcomes cannot be attributed to a subgroup of patients with severe depression. Our results also suggest that the association between preoperative depression and postoperative delirium is due to preoperative symptoms of depression rather than to a history of depression. Depressive symptoms can fluctuate with circumstances such as stress and anxiety. It is likely that patients preoperative mood states are quite different from their baseline mood state. Therefore, this dynamic mood state further underscores the importance of screening for the number of preoperative depressive symptoms. Patient Characteristics Associated With Preoperative Depression Increased medical burden has been demonstrated to increase depressive symptoms (31). Patients in our study have multiple coexistent medical conditions and many have functional limitations which may increase their likelihood of having depressive symptoms. In a meta-analysis of risk factors for depression in elderly community-dwelling persons, level of functioning and a prior history of depression were determined to be important prognostic indicators of a diagnosis of depression (32). Other risk factors for having a greater number of preoperative depressive symptoms were alcohol use and educational level; both of these factors have been validated by population studies to be risk factors for depression (33). In various clinical studies (34 36), alcohol use has been shown to be associated with depressive disorders, and it may cause or worsen depression or, alternatively, depressed subjects may be more likely to drink alcohol (37). Documenting the amount of alcohol use preoperatively may guide the perioperative care of those who abuse alcohol to minimize the development of alcohol withdrawal

5 DEPRESSION AND POSTOPERATIVE DELIRIUM 1567 symptoms, but also, the amount of alcohol use identifies a group at risk of having depressive symptoms preoperatively. Our study also demonstrates that preoperative pain is prevalent in elderly patients. It has been suggested that depression amplifies the perception of pain and persistent pain worsens depression (38). Our results suggest that the association between pain and depression are moderated by other factors. Clinical Implications Although routine screening of the general older population for depression is not warranted, our study suggests that screening in a subgroup of elderly individuals awaiting major surgery is indicated. In particular, patients with a) a complaint of moderate to severe limitations in functional status, b) moderate alcohol use, or c) lower educational level are particularly at risk for depression. Training perioperative physicians in their recognition of depressive symptoms in elderly surgical patients is indicated, because measurement and recording of depression in medical records provides little information about the presence of multiple preoperative depressive symptoms. Potential Limitations This study was not designed to determine the causal pathway between preoperative symptoms of depression and postoperative delirium. It remains to be proven whether surgery induces psychological stress resulting as depressive symptoms, which may lead to changes in neurochemistry, ultimately increasing patients vulnerability for developing postoperative delirium (18). Because our study is part of a larger study investigating the effect of perioperative management on postoperative delirium, we focused on measuring the prevalence of delirium in the early postoperative period. As a result, we may have missed the occurrence of delirium which may occur in the late postoperative period. In this study, we excluded non-english-speaking patients, who accounted for approximately 7% of the total available study patients. Direct generalization of our results to other non-english-speaking minority patients may not be appropriate. In contrast to those in prior studies of cognitive outcomes after surgery (39,40), our patients were highly educated. However, the health status of our study population is similar to that reported in another population study (41). Summary Our results show that preoperative screening for multiple depressive symptoms can be performed easily in elderly patients awaiting surgery, and it yields useful prognostic information relating to postoperative delirium. Studies on postoperative delirium should include the measurement of preoperative depression and account for its possible impact. Future studies should determine whether the timing of occurrence and duration of preoperative symptoms of depression are predictive of the subsequent development of postoperative delirium, and whether pharmacologic treatment is effective in reducing postoperative delirium in patients with significant preoperative depressive symptoms. ACKNOWLEDGMENTS This project was supported in part by institutional funds and by grant #1K24 AG (to JML) from the National Institute on Aging, National Institutes of Health. Address correspondence to Jacqueline M. Leung, MD, MPH, University of California, San Francisco, Department of Anesthesia and Perioperative Care, 521 Parnassus, San Francisco, CA REFERENCES 1. Lipowski Z. Delirium in the elderly patient. N Engl J Med. 1989; 320: Inouye S. The dilemma of delirium: clinical and research controversies regarding diagnosis and evaluation of delirium in hospitalized elderly medical patients. Am J Med. 1994;97: Williams-Russo P, Urquhart B, Sharrock N, Charlson M. Postoperative delirium: predictors and prognosis in elderly orthopedic patients. J Am Geriatr Soc. 1992;40: Williams-Russo P, Sharrock N, Mattis S, Szatrowski T, Charlson M. Cognitive effects after epidural vs general anesthesia in older adults. JAMA. 1995;274: Williams-Russo P, Sharrock N, Mattis S, Liguori GA, Mancuso C, Peterson MG, et al. Randomized trial of hypotensive epidural anesthesia in older adults. Anesthesiology. 1999;91: Williams M, Holloway J, Winn M, Wolamin MO, Lawler ML, Westwick CR, et al. Nursing activities and acute confusional states in elderly hip-fractured patients. Nursing Res. 1979;26: Dyer C, Ashton C, Teasdale T. Postoperative delirium. A review of 80 primary data-collection studies. Arch Intern Med. 1995;155: 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: Marcantonio E, Goldman L, Mangione C, Ludwig LR, Muraca B, Haslauer CM, et al. A clinical prediction rule for delirium after elective noncardiac surgery. JAMA. 1994;271: Marcantonio E, Goldman L, Orav E, Cook E, Lee T. The association of intraoperative factors with the development of postoperative delirium. Am J Med. 1998;105: Zakriya KJ, Christmas C, Wenz JF Sr, Franckowiak S, Anderson R, Sieber FE. Preoperative factors associated with postoperative change in confusion assessment method score in hip fracture patients. Anesth Analg. 2002;94: Inouye S, Charpentier P. Precipitating factors for delirium in hospitalized elderly persons: predictive model and interrelationship with baseline vulnerability. JAMA. 1996;275: Covinsky KE, Fortinsky RH, Palmer RM, Kresevic DM, Landefeld CS. Relation between symptoms of depression and health status outcomes in acutely ill hospitalized older persons. Ann Intern Med. 1997;126: Roach MJ, Connors AF, Dawson NV, et al. Depressed mood and survival in seriously ill hospitalized adults. The SUPPORT Investigators. Arch Intern Med. 1998;158: Covinsky KE, Kahana E, Chin MH, Palmer RM, Fortinsky RH, Landefeld CS. Depressive symptoms and 3-year mortality in older hospitalized medical patients. Ann Intern Med. 1999;130: Sands LP, Yaffe K, Covinsky K, Chren MM, Counsell S, Palmer R, et al. Cognitive screening predicts magnitude of functional recovery from admission to 3 months after discharge in hospitalized elders. J Gerontol A Biol Sci Med Sci. 2003;58: Ponzetto M, Zanocchi M, Maero B, et al. Post-hospitalization mortality in the elderly. Arch Gerontol Geriatr. 2003;36: Lipowski ZJ. Update on delirium. Psychiatr Clin North Am. 1992;15: Brink T, Yesavage J, Lum O, Heersema P, Adey M, Rose T. Screening tests for geriatric depression. Clin Gerontol. 1982;1: Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, Ade M, et al. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res. 1982;17: Brandt J, Spencer M, Folstein M. The telephone interview for cognitive status. Neuropsychiatry Neuropsychol Behav Neurol. 1988;1:

6 1568 LEUNG ET AL. 22. Jensen MP, Karoly P, Braver S. The measurement of clinical pain intensity: a comparison of six methods. Pain. 1986;27: Katz W, Ford A, Moskokwitz R, Jackson B, Jaffe M. Studies of illness in the aged: the index of ADL: a standardized measure of biological and psychosocial function. JAMA. 1963;185: Lawton M, Brody E. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9: CASS Coordinating Center. Manual of Operations II: Data Collection and Storage: Collaborative Studies in Coronary Artery Surgery. Seattle (WA): University of Washington; Charlson M, Pompei P, Ales K, MacKenzie C. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis. 1987;40: Longnecker D, Murphy F. Dripps/Eckenhoff/Vandam Introduction to Anesthesia, 9th ed. Philadelphia (PA): Saunders; ACC/AHA guideline update for the perioperative cardiovascular evaluation for noncardiac surgery-executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). Anesth Analg. 2002;94: Inouye S, van Dyke C, Alessi C, Balkin S, Siegal A, Horwitz R. Clarifying confusion: the confusion assessment method. Ann Intern Med. 1990;113: Galanakis P, Bickel H, Gradinger R, Von Gumppenberg S, Forstl H. Acute confusional state in the elderly following hip surgery: incidence, risk factors and complications. Int J Geriatr Psychiatry. 2001;16: Meeks S, Murrell S, Mehl R. Longitudinal relationship between depressive symptoms and health in normal older and middle-aged adults. Psychol Aging. 2000;14: Cole M, Dendukuri N. Risk factors for depression among elderly community subjects: a systematic review and meta-analysis. Am J Psychiatry. 2003;160: Barkow K, Maier W, Ustun TB, Gansicke M, Wittchen HU, Heun R. Risk factors for depression at 12-month follow-up in adult primary health care patients with major depression: an international prospective study. J Affect Disord. 2003;76: Fink A, Hays RD, Moore AA, Beck JC. Alcohol-related problems in older persons. Determinants, consequences, and screening. Arch Intern Med. 1996;156: Feldman T, Chua K, Childers R. R wave of the surface and intracoronary electrogram during acute coronary artery occlusion. Am J Cardiol. 1986;58: Liappas J, Paparrigopoulos T, Tzavellas E, Christodoulou G. Impact of alcohol detoxification on anxiety and depressive symptoms. Drug Alcohol Depend. 2002;68: Patten SB, Charney DA. Alcohol consumption and major depression in the Canadian population. Can J Psychiatry. 1998;43: Alexopoulos G, Borson S, Cuthbert B, Devannand DP, Mulsant BH, Olin JT, et al. Assessment of late life depression. Biol Psychiatry. 2002;52: Moller J, Cluitmans P, Rasmussen L, et al. Long-term postoperative cognitive dysfunction in the elderly: ISPOCD1 study. Lancet. 1998; 351: Rasmussen LS, Johnson T, Kuipers HM, Kristensen D, Siersma VD, Vila P, et al. Does anaesthesia cause postoperative cognitive dysfunction? A randomised study of regional versus general anaesthesia in 438 elderly patients. Acta Anaesthesiol Scand. 2003;47: Schoenfeld DE, Malmrose LC, Blazer DG, Gold DT, Seeman TE. Selfrated health and mortality in the high-functioning elderly a closer look at healthy individuals: MacArthur field study of successful aging. J Gerontol. 1994;49:M109 M115. Received September 9, 2004 Accepted January 27, 2005 Decision Editor: John E. Morley, MB, BCh