Feature Article
|
Managing Mood Disorders in Older Patients: A Focus on Depression and Anxiety Managing Mood Disorders in Older Patients:
A Focus on Depression and Anxiety
J. Craig Nelson, MD, Helen Lavretsky, MD, and William J. Burke, MD (Program Chair)
A symposium entitled, “Managing Mood Disorders in Older Patients: A Focus on Depression and Anxiety,” was presented on March 4, 2003, at the 16th Annual Meeting of the American Association for Geriatric Psychiatry in Honolulu, HI. The presentations focused on the efficacy and safety benefits of current treatments for depression and anxiety, the role of gender differences in treating late-life depression, and the emerging new treatment strategies for geriatric depression and anxiety.
CME Certified
Educational Objectives
• Identify evidence-based efficacy and safety benefits of selective serotonin reuptake inhibitors (SSRIs) in the treatment of geriatric depression and anxiety.
• Describe the role gender differences play in treating depression and anxiety in older adults.
• Evaluate the advantage of improved tolerability in single isomer SSRI treatment for elderly patients with depression.
• Recognize the potential value of improved efficacy, faster onset, and increased tolerability emerging in new therapies to treat late-life depression and anxiety.
Depression and Anxiety in the Old-Old
J. Craig Nelson, MD, Professor of Psychiatry, Leon J. Epstein, MD, Chair in Geriatric Psychiatry, and Director of Geriatric Psychiatry, University of California at San Francisco, presented a review of available data for antidepressant treatment in older patients. He also examined the significance of older age, late-onset compared to recurrent depression, and nonspecific interventions as factors implicated in the degree of selective serotonin reuptake inhibitor (SSRI) treatment response observed in the elderly.
Evidence-Based Review of Antidepressant Efficacy in the Elderly
The number of older patients age 85 and over is increasing rapidly. People live much longer lives today, and yet few studies have been conducted that examine the pharmacologic treatment of depression and anxiety in the elderly, described as patients age 75 and older, or age 80 and older.1
Nortriptyline
Katz and colleagues2 conducted the first placebo-controlled study of elderly patients over 80 years of age with major depression in a nursing home setting. Of the 79 patients who were determined to need treatment, 35 (44%) entered the study with a mean age of 84. They received 65 mg/day of nortriptyline, a dose similar to that used to treat younger patients, and achieved a mean plasma level of 76 ng/mL. A dropout rate of 34% (12 patients) was observed. Using Clinical Global Impressions (CGI) or Hamilton Depression (HAM-D) rating scales, nortriptyline was significantly more effective than placebo. This is the only study to date that shows an advantage for drug treatment in this group of patients.
Desipramine
Dr. Nelson et al3 conducted a fixed desipramine dose (2.5 mg/kg) plasma level study on 34 nursing home residents with a mean age of 79 (range 75-89). The primary outcome measurement was the CGI scale, which showed 42% of patients to be much improved with an adverse event rate of 21%. Overall, older patients had less of a response than younger participants during a similar duration of treatment. The relationship of response and plasma levels appeared to be similar across different age groups.3-5
Fluoxetine
In 1996, Finkel6 analyzed the subset of 70 patients aged 75 and older who participated in a fluoxetine-placebo study of 671 patients aged 60 and older. All patients were treated with a 20-mg dose of fluoxetine for a 6-week period. Although response to fluoxetine was greater than that to placebo in the entire sample, in patients 75 years and older placebo was just as effective as fluoxetine.6,7
Paroxetine
Burrows and associates8 conducted a placebo-controlled study of paroxetine in 24 nursing home residents aged 80 and older (mean age = 88; range 80-94) with non-major depression, but “with symptoms that warranted treatment.” The 8-week, double-blind, random-assignment trial involved paroxetine up to 30 mg/day (mean dose = 23 mg). Patient- and nurse-rated assessment instruments found a mean Mini-Mental State Examination (MMSE) score of 24 (range 16-30) and a mean HAM-D score of 14 (range 4-23). Twenty of 24 patients completed tûe trial. Using the CGI response as the primary outcome measure, the paroxetine response rate was slightly higher (56% vs 36%), but in this small sample the difference was not significant. Anticholinergic assays did not find evidence that paroxetine was having an anticholinergic effect. The mean change in the MMSE (2.0 vs -0.1) on paroxetine and placebo, respectively, was near significant (P = .07).
Citalopram
Roose and colleagues9 conducted the first large multicenter, randomized, placebo-controlled, double-blind outpatient study in patients aged 75 and older with nonpsychotic unipolar major depressive disorders (MDD). In this study of 170 patients with MDD over age 74 years, response to citalopram was compared with response to placebo over an 8-week period. Mean patient age was 79.8 years in the citalopram group, and 79.3 years for patients in the placebo-control group. Citalopram dosage began with 20 mg/day for 4 weeks followed by an option to systematically increase it to 40 mg/day for the remainder of the trial unless a patient was much improved. Magnetic resonance imaging (MRI) and neuropsychological testing were performed by the study coordinator. Participants were assessed with a 24-item HAM-D scale (score > 20) with mean baseline HAM-D scores of 24.4 in the citalopram-treated group, and 24.3 in the placebo-control group. Patients presenting symptoms of early dementia (MMSE scores > 18) or stable medical illnesses were included as typical characteristics for this type of patient group. Other axis 1 disorders, unstable medical illness, or patients who had already failed to respond to one SSRI were excluded. Completion rate for the study was 87% (placebo) and 76% (citalopram). Using the HAM-D and Hamilton Anxiety (HAM-A) scales as the primary outcome measures, no observed differences in the drug-vs-placebo groups or the responders-vs-nonresponders were observed. Both treatments were effective, with response rates of about 40% in these older patients.
Severity and age of onset appeared related to response. In patients with severe depression (HAM-D score of > 24), citalopram remained effective with 43% of patients responding, but placebo response dropped to 18%. In this smaller group of 45 patients, this difference was not statistically significant. In less severe patients, the nonspecific effects in the placebo group appeared to be just as effective as citalopram. Late-onset depression was defined as the first episode occurring after age 59 years. In late-onset depression, response rates with citalopram and placebo were similar. Patients with early-onset depression by definition were also those with recurrent depression. In this group, citalopram was numerically superior to placebo (40% vs 24%), but again, in this smaller group of 41 patients, the difference was not significant (Figure 1).
Effects of nonspecific interventions—such as the amount of time the study coordinator spent with patients, which was appreciable—appeared to have meaningful effects in patients with less severe or late-onset depression. These supportive interventions may have been quite helpful for older patients who were demoralized about declining health or other age-related stressors. There was considerable site-to-site variation in drug-vs-placebo response, possibly due to the small numbers of patients at each site or other unaccounted interventions that patients received.
Augmenting SSRI Treatment in Primary Care
One example of an “unaccounted-for” intervention was telephone calls to check on the patients’ status. At one of the sites, this was required by the Internal Review Board. Hunkeler and associates10 have reported on the value of this intervention. They looked at 302 patients with MDD who received SSRI treatment with fluoxetine or paroxetine for 6 months. In addition to SSRI treatment, patients were randomly assigned to receive brief telephone calls from the nurse or peer support, or the SSRI alone. The study demonstrated that peer support added no benefit, but nurse telehealth care did. Keeping in touch with the patient, providing reassurance and education, and answering questions made a difference and enhanced response to SSRI treatment.
Other forms of intervention may also prove useful in older patients with depression. Alexopoulos et al11 are researching a type of behavioral treatment called Problem Solving Therapy (PST) for patients with MDD and executive dysfunction. Investigators recently reported on an initial sample of 25 patients over age 65 years with depression (HAM-D > 18) who had evidence of executive dysfunction. Results of this 12-week random-assignment study indicated that PST was more effective than supportive therapy (70% vs 20%).
Benefits of SSRI Treatment of Geriatric Depression
Supported by a review of evidence on the efficacy and safety of SSRIs in older patients, SSRIs may merit primary consideration in the pharmacologic approach to treating geriatric depression. SSRIs may be particularly useful in patients with recurrent depression or those with more severe depression for whom nonspecific supportive interventions may be less effective. However, these nonspecific supportive treatments may prove to be quite important in elderly patients with late-onset or less severe depression. In general, these recent findings challenge our assumptions about late-onset depression and indicate the need to further explore a variety of treatments.
References
1. NIH Consensus Conference: Diagnosis and treatment of depression in late life. JAMA 1992;268(8):1018-1024.
2. Katz IR, Simpson GM, Curlik SM, et al. Pharmacologic treatment of major depression for elderly patients in residential care settings. J Clin Psychiatry 1990;51(suppl):41-47.
3. Nelson JC, Mazure CM, Jatlow PI. Desipramine treatment of major depression in patients over 75 years of age. J Clin Psychopharmacol 1995;15(2):99-105.
4. Nelson JC, Jatlow PI, Quinlan DM, Bowers MB Jr. Desipramine plasma concentration and antidepressant response. Arch Gen Psychiatry 1982;39(12):1419-1422.
5. Nelson JC, Jatlow PI, Mazure C. Desipramine plasma levels and response in elderly melancholic patients. J Clin Psychopharmacol 1985;5(4):217-220.
6. Finkel SI. Efficacy and tolerability of antidepressant therapy in the old-old. J Clin Psychiatry 1996;57(suppl 5):23-28.
7. Tollefson GD, Holman SL. Analysis of the Hamilton Depression Rating Scale factors from a double-blind, placebo-controlled trial of fluoxetine in geriatric major depression. Int Clin Psychopharmacol 1993;8(4):253-259.
8. Burrows AB, Salzman C, Satlin A, et al. A randomized, placebo-controlled trial of paroxetine in nursing home residents with non-major depression. Depress Anxiety 2002;15(3):102-110.
9. Roose S, et al. Treatment of depression in patients over 75. Presented at: 15th Annual Meeting of the American Association for Geriatric Psychiatry; May 24-27, 2002; Orlando, FL.
10. Hunkeler EM, Meresman JF, Hargreaves WA, et al. Efficacy of nurse telehealth care and peer support in augmenting treatment of depression in primary care. Arch Fam Med 2000;9(8):700-708.
11. Alexopoulos GS, Raue P, Arean P. Problem-solving therapy versus supportive therapy in geriatric major depression with executive dysfunction. Am J Geriatr Psychiatry 2003;11(1):46-52.
Sex Differences in Late-Life Depression
Helen Lavretsky, MD, Assistant Professor of Psychiatry, Division of Geriatric Psychiatry, UCLA-Neuropsychiatric Institute, Los Angeles, discussed the existing evidence of sex differences in depression by describing epidemiology, phenomenology, neurobiology, and antidepressant response in geriatric depression. Dr. Lavretsky also reviewed the relationship of depression to sex by age or age of onset, and evaluated the sex differences in risk factors and brain structure in geriatric depression.
Epidemiology of Sex Differences in Major Depression
The most consistent finding in psychiatric research is that of gender differences in the prevalence of depression: a 2:1 ratio in women versus men, and a lifetime prevalence of 4% in women versus 1.7% in men.1 A fact that is frequently overlooked is that this prevalence rate only exists from puberty through menopause. Boys have higher depression rates during childhood, and the ratio becomes nearly 1:1 by age 80. Multiple explanations are used to explain these differences, including disparities in seeking help or reporting symptoms. Men tend to underreport episodes of depression, and women tend to report occurrences more frequently. Prior history of depression would place women at greater risk for recurrent or future episodes.
Another explanation of gender differences in the prevalence of depression may be social risk factors. Adverse childhood experiences such as sexual abuse, which accounts for 35% of the variance in sex differences, show women to be at greater risk. Limitations, lack of choice, and competing societal roles affect women more often. More than 50% of women compared to 25% of men live alone as older adults. This can mean that women face greater financial strain and loss of independence due to declining health. In addition, women are even more likely to be placed in a nursing home.
Stress and coping styles are also offered as an explanation of sex differences. Childhood trauma and life stresses affect men and women differently because men may cope with stress differently. Women are three times more likely than men to develop depression in response to any stressful life event, according to research observations.
Biological theories have been proposed to explain gender differences in depression. Inheriting major depression is about 30% more likely in women because there is a 60% overlap in the genes responsible for depression that are found in men and women. Since no increased depression rates are seen in postmenopausal women and hormone replacement therapy (HRT) has a minimal to modest effect on treatment response, hormonal influences are thought to have less effect on depression rate differences than environmental factors. Hormones of the adrenal and thyroid axes, long considered the missing link, have a contrasting or limited role. The differences in the neurotransmitter systems are unclear, as is brain laterality or lack of it in women.
Two paradoxes exist in gender differences of depression rates among men versus women. The first is that rates of completed suicide in men, particularly white men, are greater even though women have higher rates of depression. Access to guns and gun control laws only partially explain this paradox, because women also have access to guns but choose less lethal methods such as drug overdose, and overall rates of suicide are comparable among men in countries with gun control laws (eg, Canada).
A second paradox is that men live an average of 8 years less than women, even though women reportedly suffer more chronic illnesses and disabilities. This may be because older men, who have a higher prevalence of vascular diseases, may be more prone to developing comorbid, late-onset depression, which increases mortality and disability secondary to vascular conditions.
Sex Differences in Antidepressant Response
Kornstein and associates2 reported on a randomized trial of 635 patients, of which 400 were female, showing that women have a preferential response to SSRIs (sertraline) and men to tricyclic antidepressants (TCAs; imipramine). In a meta-analysis conducted by Quitkin and colleagues,3 a number of studies including over 1700 patients ages 18-65 were analyzed. Patients were then divided into two groups, age 50 and over and under 50, and then divided into men versus women. Researchers found no differences in the response rate to TCAs and SSRIs, but a greater rate of response to monoamine oxidase inhibitors (MAOIs) in women overall with a preferential response from women under 50. Frank et al4 reported that men had a greater overall speed of response compared to women, and relapses were seen more often in women.
Postmenopausal women have an increased volume of distribution and decreased absorption and bioavailability. The effect of estrogen tends to increase certain neurotransmissions (5-HT). Drugs are metabolized via hydroxylation in women and by demethylation in CYP2D6 in men. Hydroxymetabolites are more toxic; therefore, women are more prone to developing drug toxicity, a fact supported by greater dropout rates of women versus men in clinical trials due to adverse reactions.
Relationship of Age, Age at Onset, and Sex to Depression in Elderly Patients
Dr. Lavretsky and colleagues5 performed a cross-sectional comparison of 96 outpatients with moderate MDD (HAM-D > 15; MMSE > 24) by dividing patients into an early-onset depression (EOD) group (33 patients; onset before age 50; 64% women) and a late-onset depression (LOD) group (63 patients; onset after age 50; 43% women). Investigators conducted on each participant a neuropsychiatric assessment and evaluation of medical burden, as well as history of medical illness, such as hypertension and smoking. MRI was conducted to calculate ventricular brain ratio, a measure of atrophy, and total white-matter hyperintensities (WMH) lesion size. Comparison of the two groups, LOD versus EOD, by clinical variables differed only in the lower number of episodes in the LOD group.
Sex differences were found in rates of LOD, EOD, degree of severity, hypothyroidism, neurovegetative factors, suicidal ideations, and history of smoking. Comparing the effect of sex versus age of onset on phenomenology, rates of depression in the four groups (LOD men, LOD women, EOD men, EOD women) showed the presence of suicidal ideations, and vegetative and cognitive signs, which resulted in more severe depression based on HAM-D ratings. Men also had more severe vegetative signs and suicidal ideations compared to women. Comparing brain structural changes by sex versus age of onset showed that men had greater ventricle-to-brain ratio and areas of WMHs (Figure 2). The LOD group had greater percentages of WMHs (defined as greater than 10 ccs). Results showed that men were more severely depressed with neurovegetative signs present and a history of smoking. The LOD group had increased WMH, larger ventricle-to-brain ratio (VBR) and a history of hypertension. Age also influenced WMH, atrophy, hypertension, and a lack of social support. Researchers concluded that the relationship of sex to the age of onset may be an influence in the phenomenology and neurobiology of late-life depression. Older men may be at greater risk for LOD associated with brain structural changes and cerebrovascular disease.
Pathophysiology of Geriatric Depression
In another study using a similar approach, Dr. Lavretsky and associates6 recruited 22 patients age 60 and older (9 men, 13 women; mean age, 71.4) with major depression, and 15 control patients (4 men, 11 women; mean age, 72), whose age, education, and sex ratio matched normal controls, in order to compare total brain, frontal gray and white matter volumes, and measures of medical burden and neuropsychiatric symptoms. All patients received a three-dimensional MRI segmentation into gray matter, white matter, and cerebrospinal fluid (CSF) volumes. Regional volumes were calculated automatically and total frontal lobe volume was examined. Depression and quality of life were assessed using a 21-item HAM-D scale (short form [SF]-36) and questionnaire. Medical burden was assessed using the Cumulative Illness Rating Scale (CIRS), geriatric version, and a cerebrovascular risk factor (CVRF) scale to determine stroke factors. Unified Parkinson’s Disease Rating Scale (UPDRS) was used to evaluate psychomotor retardation. The Apathy Evaluation Scale (AES) was used to look at the severity of apathy due to its prevalence in the population.
Observations based on clinical variables compared to the control group showed no differences by age. Patients with depression had lower MMSE scores that were consistent across all samples. CIRS score was higher in these patients (ie, they were much more ill than other patients with only depression). Groups did not differ in stroke risk factors. Compared to controls, apathy, retardation, and lower quality of life were more prevalent in the patients with depression.
Corrected volumes of the intracranial volume (ICV) were used because patients vary widely based on their circumference and brain volumes, and men tend to have larger brain ICVs. Patients with depression had smaller corrected volumes in frontal, left frontal, and right frontal areas, and smaller frontal white matter volumes.
Age accounted for a greater medical burden, vascular and nonvascular factors, and psychomotor retardation. No sex differences were found; however, differences between diagnostic groups remained. Patients with depression were still more apathetic, more medically ill, and had poorer quality of life.
Diagnosis by sex revealed that age accounted greatly for atrophy in frontal lobe volumes (Table I). Men had smaller corrected volumes throughout frontal, left frontal, right frontal, and frontal white matter volumes. Differences between diagnostic groups were observed. Logistic regression showed that frontal total volume and frontal white matter volume predicted sex assignment after controlling for age in the group with depression, but not in the control group.
Another study of brain structure conducted by Gur and colleagues7 reported that age correlates with reduced gray matter volume and increased sulcal CSF. Therefore, atrophy is increased in men because of age, and the association is prominent in frontal, temporal, and basal ganglia structures, but less in subcortical regions. Men experience a greater volume decrement across ages, particularly in dorsolateral prefrontal cortex regions, which may place them at greater risk for late-onset depression.
Geriatric depression is associated with greater severity of medical comorbidity and brain structural changes. Sex differences in neuroanatomy may be important in the pathophysiology of geriatric depression.
Gender-Sensitive Approach
The ultimate goal in treating geriatric depression in men and women should be the optimization of pharmacotherapy and other forms of treatment by maximizing therapeutic benefits and minimizing negative effects. However, before that can be achieved, underlying sex differences and other neurobiological variables of this illness must be understood.
References
1. Blazer D, Williams CD. Epidemiology of dysphoria and depression in an elderly population. Am J Psychiatry 1980;137(4):439-444.
2. Kornstein SG, Schatzberg AF, Thase ME, et al. Gender differences in treatment response to sertraline versus imipramine in chronic depression. Am J Psychiatry 2000;157(9):1445-1452.
3. Quitkin FM, Stewart JW, McGrath PJ, et al. Are there differences between women’s and men’s antidepressant responses? Am J Psychiatry 2002;159(11):1848-1854.
4. Frank E, Thase ME, Spanier CA, et al. Gender-specific response to depression treatment. J Gend Specif Med 1999;2(4):40-44.
5. Lavretsky H, Lesser IM, Wohl M, Mill BL. Relationship of age, age at onset, and sex to depression in older adults. Am J Geriatr Psychiatry 1998;6(3):248-256.
6. Lavretsky H, Kumar A, Pham D, et al. Sex differences in brain structure in geriatric depression [abstract]. Presented at: 2002 Annual Meeting of the Society of Biological Psychiatry; May 2003; Philadelphia, PA.
7. Gur RC, Gunning-Dixon FM, Turetsky BI, et al. Brain region and sex differences in age association with brain volume: a quantitative MRI study of healthy young adults. Am J Geriatr Psychiatry 2002;10(1):72-80.
New Treatments for Depression and Anxiety
William J. Burke, MD, Professor and Vice-Chair for Research, Department of Psychiatry, and Director of Geriatric Psychiatry, University of Nebraska Medical Center, Omaha, discussed the benefits of increased efficacy and safety in new SSRI therapies for t“e treatment of mood and anxiety disorders. Dr. Burke also evaluated new and developing pharmacotherapeutic strategies for depression and anxiety.
Very little age-specific data are available for the newer antidepressants and anxiolytics. Although it is important to treat older adults, they are less likely to participate in clinical trials, and often get excluded. Therefore, current data tend to be of very healthy individuals, making the applicability uncertain.
New SSRI Strategies
Escitalopram
Escitalopram, the most recently approved antidepressant, is the pure S-enantiomer (single isomer) of citalopram, and was developed specifically to see if the tolerability of citalopram could be improved by eliminating the R-enantiomer. Gorman and colleagues1 conducted a pooled analysis of three randomized, double-blind, placebo-controlled, 8-week trials comparing escitalopram and citalopram in the treatment of MDD. One trial used a fixed dose and the other two trials used flexible doses. The primary outcome measure of efficacy was the Montgomery-Asberg Depression Rating Scale (MADRS; Figure 3).1 In this analysis, escitalopram was statistically superior to placebo from week 1 through week 8, and to citalopram at week 1 and week 8. However, at the time of Food and Drug Administration (FDA) approval of escitalopram, no specific depression studies had been done in the elderly. Analysis of covariance (ANCOVA) was performed on the available data on MADRS scores, using age as a covariate to look at the effect of age on response.2 No significant age effect or age-by-treatment interaction was observed. Improvement in MADRS scores produced by escitalopram was not related to patient age.
Escitalopram was well tolerated in patients age 60 and over (Table II).6 No side effect was present in more than 10% of patients, and nausea and abdominal pain were the only side effects that occurred in more than 5% of patients and more frequently than placebo.6 Three percent of patients taking placebo discontinued for adverse events compared to 9% of the escitalopram-treated patients.
There are no specific data available on efficacy of escitalopram for anxiety in older patients, although a number of positive trials have been completed in younger adults with generalized anxiety disorder (GAD), panic disorder, and social anxiety disorder.1,3-5
Duloxetine
Dr. Burke believes the next available drug may be duloxetine, which has shown results in balanced uptake inhibition of both serotonin and norepinephrine (NE) at starting doses. Evidence suggests NE is an important neurotransmitter with concentrations that are reduced in CSF of patients with depression.7-9 Duloxetine has some similarities to venlafaxine, except that venlafaxine affects serotonin at lower doses and only begins to have noradrenergic effects once the dosage passes 150 mg/day. Imipramine, a tricyclic antidepressant, also has a similar balanced effect on serotonin and NE uptake inhibition.
Nelson and colleagues10 performed a pooled analysis on outpatients age 55 and older with Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV)-defined major depression. Data were extracted from two identical randomized, multicenter, placebo-controlled, double-blind trials. Patients received 60 mg/day of duloxetine over a 9-week study period. The primary efficacy measure was the HAM-D. Duloxetine was significantly superior to placebo at week 9 on the HAM-D.
Tran and associates11 conducted a pooled analysis of 753 patients evaluating the efficacy of duloxetine using three studies in which the drug was statistically superior to placebo in patients under age 55 versus patients age 55 and older. Only 61 patients receiving duloxetine and 50 patients receiving placebo, respectively, were age 55 and older. Researchers found no significant differences across subgroups; older patients appeared to do as well as younger patients.
Tran et al11 conducted a second pooled analysis study of duloxetine safety by looking at data from 1755 patients in seven double-blind, placebo-controlled trials. Only 34 patients receiving duloxetine and 31 patients receiving placebo, respectively, were age 65 and older in these trials. No significant differences were observed across subgroups. In patients aged 65 and older who received duloxetine, discontinuation rates due to adverse events were 26%, compared to 12% in the placebo group. For patients taking duloxetine, the mean change in systolic blood pressure (BP) was 2 mm Hg; in diastolic BP, 4 mm Hg; and in pulse rate, -1 bpm. Adverse events occurring |