Managing Mood Disorders in Older Patients: A Focus on Depression and Anxiety

At an industry-supported symposium held at the American Association for Geriatric Psychiatry’s 2003 Annual Meeting, a panel of experts examined the use of selective serotonin reuptake inhibitors and new medications for the treatment of depression and anxiety in geriatric patients.

The panel also reviewed existing data regarding gender differences in geriatric patients with depression.

This program was supported by an unrestricted educational grant from Forest Laboratories, Inc.

Depression and Anxiety in the “Old-Old”

J. Craig Nelson, MD, Professor of Psychiatry, Leon J. Epstein Chair in Geriatric Psychiatry, and Director of Geriatric Psychiatry at the University of California at San Francisco, California, reviewed the data regarding the use of antidepressants in patients aged > 75 years (the “old-old”) and examined the effects of older age on response.

Use of Antidepressants in Patients Aged >75 Years
Antidepressants have only a modest effect on depression in patients aged > 75 years, reported Dr. Nelson. In fact, only one study to date, involving nortriptyline, has shown a difference between an antidepressant and placebo in patients aged > 80 years (Katz, et al. J Clin Psychiatry. 1990;51(suppl):41). Other studies involving desipramine (Nelson, et al. J Clin Psychopharmacol. 1995; 15:99), fluoxetine (Finkel, et al. J Clin Psychiatry. 1996;57(suppl):23), and paroxetine (Burrows, et al. Depression and Anxiety. 2002;15:102) have found low response rates, with moderate (~ 40%) placebo response rates and therefore minimal drug-placebo differences in patients aged >75 years.

Likewise, citalopram was as effective as placebo in improving depression and anxiety in nonpsychotic patients aged >75 years with unipolar major-depressive disorder (Roose S. Presentation. 15th Annual Meeting Am Assoc Geriatric Psychiatry, 2002). However, analyses of these responses by severity of depression and age at onset of depression had
interesting implications, Dr. Nelson revealed.

Response to Citalopram and Placebo by Severity of Depression
When categorized according to the severity of the patient’s depression, response rates to citalopram treatment were similar in severely depressed (Hamilton depression rating score > 24) and less depressed patients. Furthermore, in the less severely depressed patients, no differences were noted between the citalopram and placebo groups. However, among the 45 severely depressed patients, the 18% response and 13% remission rates in the placebo group were markedly lower than the 43% response and 30% remission rates of the citalopram group (Figure 1). These results suggest that in severely depressed patients, the placebo response (i.e., response to nonspecific effects) is lower and the drug-placebo difference is more marked.

Response to Citalopram and Placebo by Age at Onset of Depression
A similar trend was observed in an analysis of the response and remission rates according to age at onset of depression. Among the 41 patients who had early-onset depression, those in the placebo group had lower rates of response (24%) and remission (19%) than did the citalopram-treated patients (40% and 30%, respectively). However, among patients with late-onset (> 60 years) depression, placebo response and remission rates were not reduced. Thus, no drug-placebo differences were observed.

Unlike with placebo, the response and remission rates with citalopram were similar in patients with early-onset and late-onset depression. One possible explanation for this might be that the effects of nonspecific treatment (e.g., regular contact with site coordinator) were reduced in patients with early-onset depression, as they were in those with severe depression. In contrast, patients with late-onset depression and those with less severe depression responded fairly well to both citalopram and to nonspecific effects. These nonspecific effects also might have contributed to the considerable inter-site variability observed in this trial, Dr. Nelson suggested.

The Effects of Nonspecific Treatment
Some sites in this trial required that patients be closely monitored by telephone because of concerns about possible patient suicides. Dr. Nelson pointed out that these extra telephone contacts (i.e., nonspecific treatment) might have provided an added benefit to the patients, thereby augmenting their response to the medication. Indeed, a separate study, in which all patients received a selective serotonin-reuptake inhibitor, has shown that patients who received regular telephone calls from nurses had significantly better response rates than did patients who had usual treatment without extra support (P = 0.003 at 6 months) (Hunkeler, et al. Arch Fam Med. 2000;9:700).

Dr. Nelson also mentioned that problem-solving therapy, a form of behavioral treatment, has been shown to be more effective than nonspecific approaches in older patients with depression and executive dysfunction (Alexopoulos, et al. Am J Geriatric Psychiatry. 2003;11:46). Thus, problem-solving therapies may play a role in the future direction of care, Dr. Nelson hypothesized.

Sex Differences in Late-Life Depression

The commonly quoted lifetime prevalence of major depression in women, from puberty through menopause, is 4%, whereas the rate for men is 1.7% in a similar age range. Helen Lavretsky, MD, Assistant Professor of Psychiatry, Division of Geriatric Psychiatry at the ULCA-Neuropsychiatric Institute in Los Angeles, California, presented several explanations for these different rates of depression and described how gender differences may affect responses to antidepressants.

Explanations for Sex Differences in Rates of Depression
Women tend to report prior history of depression more frequently than men. In addition, older women, who tend to live longer than men, are more likely to experience chronic illnesses and be placed in nursing homes, thereby increasing their risk of depression.

Although women do not have a higher rate of adverse life events, they are believed to be at greater risk for sexual abuse and are three times more likely than men to develop depression in response to any stressful life event. However, contrary to popular belief regarding attachment patterns, recent studies indicate that men may be more impaired and depressed than women when unmarried, or after a divorce or the loss of a loved one (van Grootheest, et al. Soc Psychiatry Psychiatr Epidemiol. 1999;34(7):391).

Multiple biologic theories exist regarding the sex differences in rates of depression. Women reportedly exhibit an approximately 30% greater heritability of depression than do men (Kendler, et al. Psychol Med. 2001;31(4):605), although the genes involved in men and women are not identical. In addition, hormonal differences historically were considered to affect rates of depression, but gonadal factors probably have less influence than environmental conditions. Gender differences in adrenal and thyroids axes, neurotransmitter systems, and brain laterality also have been proposed to influence the different rates of depression. However, the effects of these factors remain uncertain. In contrast, gender differences have been shown to account for different responses to antidepressants.

Sex Differences in Antidepressant Response
Several studies have demonstrated sex differences in response to antidepressants. Women have preferential responses to monoamine oxidase inhibitors (Quitkin, et al. Am J Psychiatry. 2002; 159:1848) and selective serotonin reuptake inhibitors (sertraline), whereas men have better responses to imipramine (Kornstein, et al. Am J Psychiatry. 2002;157:1445).

Gender differences also exist in pharmacokinetics and pharmacodynamics. Women tend to metabolize drugs through hydroxylation, whereas drug metabolism in men involves demethylation. The metabolites of hydroxylation are more toxic, which may account for the greater incidence of adverse effects (e.g., gastrointestinal toxicity) and higher dropout rates among women in clinical trials.

Clinical Studies Examining the Impact of Sex on the Pathophysiology of Depression
In a cross-sectional comparison of 96 outpatients with moderate major-depressive disorder, patients with late-onset depression experienced significantly (P = 0.001) fewer episodes of depression than those with early-onset depression (P = 0.001) (Lavretsky, et al. Am J Geriatr Psychiatry. 1998;6:248). Men had a greater severity of depression,
vegetative signs, and suicidal ideation compared with women (Figure 1). In addition, men had a greater ventricle-to-brain ratio and greater areas of white matter hyperintensities, as did the late-onset group. Thus, the interaction between sex and age at onset of depression may influence the phenomenology and neurobiology in late-life depression, with older men being at greater risk for late-onset depression associated with brain structural changes and cerebrovascular disease.

A separate study compared elderly men and women with major depression and age-matched control subjects in terms of total brain, frontal gray, and white matter volumes (Lavretsky, et al. Presentation. Annual Meeting Soc Biol Psychiatry, 2002). The depressed patients had lower Mini-Mental State Examination scores and a greater medical burden, with more apathy and psychomotor retardation. They also had a lower quality of life compared with the control subjects. Moreover, a logistic regression found that the frontal total volume and frontal white matter volume predicted gender assignment, after controlling for age in the depressed group, but not in the control subjects.

Understanding these sex differences and other neurobiological variables in the pathophysiology of geriatric depression is important for optimizing treatment approaches in geriatric depression, concluded Dr. Lavretsky.

New Treatments for Depression and Anxiety in Geriatric Patients

When cholinesterase inhibitors, such as donepezil, galantamine, and rivastigmine, were first introduced for the treatment of Alzheimer’s disease (AD), the most important variables considered were cognition and improvement in the patient’s orientation and memory. However, Daniel D. Christensen, MD, Clinical Professor of Psychiatry and Neurology and Adjunct Professor of Pharmacology at the University of Utah Neuropsychiatric Institute in Salt Lake City stated that increasing attention is now being directed at preserving the patient’s activities of daily living and maintaining appropriate behavior.

Dr. Christensen hypothesized that if treatment can be initiated at the pre-Alzheimer’s stage, i.e., when the patient is exhibiting only mild cognitive impairment (MCI), the progression to AD may be delayed and the patient can be self-sufficient for a longer period (Figure 1). This reduces the burden on caregivers and delays placement of the patient in a skilled-nursing facility. To illustrate the impact of early initiation of drug therapy, Dr. Christensen showed 2-year data for a
patient diagnosed with AD.

Use of Donepezil in the Treatment of AD: A Case Report
Esther, an AD patient, was considered by her family to be unable to live independently and had a Mini-Mental State Examination (MMSE) score of 23 at baseline. She received donepezil 5 mg/d for the first month and then 10 mg/d. At 3 months, her MMSE score was 27, but subsequently declined to 26 at 6 months and to 24 at 12 months. Thus, over a year, her condition was remarkably stable, especially considering that most AD patients experience a decline within 6 months and often may lose 2-3 points a year. In addition, Esther’s scores in the cognitive subscale of the Alzheimer’s Disease Assessment Scale (ADAS-cog) were relatively stable (approximately 46). She still lived independently at the end of the first year of drug treatment.

At 2 years, Esther’s MMSE and ADAS-cog scores were 22 and 48, respectively—markedly better than the projected scores for patients who have not received drug treatment. Moreover, Esther was still living independently, whereas patients at this stage of the disease would probably have been placed in a nursing home.

Explaining the significance of these scores, Dr. Christensen noted that a 7-point decrease in ADAS-cog scores is defined as a significant improvement and can be considered to represent an approximate 1-year delay in the patient’s decline. Even a 0-point change in the ADAS-cog score is important, Dr. Christensen observed, because it represents a stabilization of the patient’s condition. Thus, in Esther’s case, early initiation of drug treatment prolonged her period of self-sufficiency.

Other Treatment Options for Dementia
Dr. Christensen reviewed some of the other treatment options for dementia. Currently under development are several drugs that reduce levels of the polypeptide Aß42 in the brain. Aß42 is formed through abnormal cleavage of the amyloid precursor protein by gamma secretase, and accumulation of Aß42 can constitute the molecular basis of AD. In a murine model, the nonsteroidal anti-inflammatory drug flurbiprofen altered the cleavage site of gamma secretase and thereby prevented production of Aß42 (Unpublished data. Todd Golde. Mayo Clinic, Jacksonville, Florida, 2003). Dr. Christensen believes that drugs that reduce Aß42 levels in the brain may produce better outcomes in AD patients and might even prevent the condition.