by James Russell, MD; Joel Raskin, MD; Curtis Wiltse, PhD; Daniel Walker, PhD; and Olga Brawman-Mintzer, MD

Drs. Russell, Wiltse, and Walker are with Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana; Dr. Raskin is with Lilly Research Laboratories, Eli Lilly Canada, Toronto, Ontario, Canada; and Dr. Brawman-Mintzer is with the Medical University of South Carolina, Charleston, South Carolina.

Financial support: This work was sponsored by Eli Lilly and Company and Boehringer Ingelheim GmbH.

Clinical trial registry number for this study: NCT00062673 at www.clinicaltrials.gov. First patient enrolled in this study (assigned to therapy): 18 March 2003. Last patient completed: 22 July 2004.

Disclosures: Drs. Russell, Raskin, Wiltse, and Walker are employees of Eli Lilly and Company; Dr. Brawman-Mintzer is on the speakers bureau of AstraZeneca, Forest Pharmaceuticals, Pfizer Inc., Eli Lilly, and Janssen Pharmaceutica, has received grant/research support from Pfizer, Forest Pharmaceuticals, AstraZeneca, Janssen Pharmaceutica, Biovail, UCB Pharma, Eli Lilly, Novartiz, and Sanofi-Aventis, and is a consultant to Pfizer, UCB Pharma, Janssen Pharmaceutica, AstraZeneca, Solvay.

Key Words: duloxetine, anxiety, major depressive disorder, elderly, placebo

ABSTRACT

Objective: To compare the efficacy and tolerability of duloxetine 60mg/day versus placebo in treating elderly patients with major depressive disorder (MDD) and concurrent anxiety symptoms.

Methods: Patients (?65) were randomized to eight weeks of treatment with duloxetine 60mg/day (n=207) or placebo (n=104). Anxiety measures were analyzed for all patients, by age (<75 and ?75), and in patients having concurrent high anxiety (HAMD17, item 10; Psychic Anxiety baseline score of 2, 3, or 4). Psychic Anxiety, Somatic Anxiety item 11, and the Anxiety/Somatization subscale were analyzed for all patients and subgroups by mean change from baseline to endpoint and repeated measures. Tolerability was assessed via treatment-emergent adverse events (TEAEs), and adverse events were reported as the reason for discontinuation. The analyses presented are primarily post hoc in nature.

Results: Duloxetine produced significantly greater reductions than placebo in Psychic Anxiety (least-squares mean change: -0.62 vs. -0.18, p<0.001) and the Anxiety/Somatization subscale (-1.88 vs. -0.99, p=0.002). Repeated measures analyses showed separation between the treatment groups beginning at Week 1 for Psychic Anxiety and Week 4 for the Anxiety/Somatization subscale. Significant improvement occurred in the <75 and ?75 age groups for Psychic Anxiety, but only the <75 group for the Anxiety/Somatization subscale. Duloxetine-treated patients with high anxiety showed significant improvement compared with placebo-treated patients on Psychic Anxiety, Anxiety/Somatization subscale, the 17-item Hamilton Depression Rating Scale (HAMD17) total score, and several other measures. Duloxetine and placebo had similar TEAE rates and discontinuation rates due to adverse events.

Conclusion: Duloxetine (60mg/day) was efficacious and tolerable in elderly patients with MDD and concurrent anxiety symptoms.

Introduction

It has been shown that a close association exists between major depressive disorder (MDD) and various anxiety disorders.[1–4] The percentage of patients with MDD and a comorbid anxiety disorder was shown to be 58 percent in the National Comorbidity Survey[4] and 57 percent of patients in a recent US study.[2] In an earlier literature review, Gorman[5] found that approximately 85 percent of patients with depression experience significant anxiety symptoms.

Patients with MDD having comorbid anxiety disorders are more difficult to treat and have more severe symptoms than do patients with MDD alone.[6–8] In older patients with MDD, it has been shown that having a comorbid anxiety disorder often is associated with more severe somatic symptoms,[6] a greater decline in memory,[9] increased suicidal ideation,[6] and increased suicide rates.[10]

Several compounds from the selective serotonin reuptake inhibitors (SSRI) class have been shown to improve both depressive and anxiety symptoms.[11–16] Venlafaxine, a serotonin (5-HT) and norepinephrine (NE) reuptake inhibitor (SNRI), has been shown to be a safe and effective treatment for anxiety associated with depression,[17] as well as for various anxiety disorders.[18] Duloxetine, a relatively balanced SNRI, is an effective and safe treatment in patients with MDD,[19–24] including the elderly,[25] and in patients with generalized anxiety disorder (GAD).[26,27] Also, there are accruing data on the efficacy of the SNRIs in the treatment of comorbid anxiety and depressive symptoms.[28,29] For example, duloxetine 60 to 120mg/day was shown to be effective in rapidly relieving anxiety symptoms associated with MDD in middle-aged patients (mean age ~41 years).[30]

Although there are increasing data on the treatment of adults with depressive and anxiety symptoms, little is known about the treatment of elderly who have both depressive and anxiety symptoms.[31–34] In this paper, we compare the efficacy and tolerability of duloxetine 60mg/day with placebo in elderly patients with MDD and concurrent anxiety symptoms.

Methods

Study design. The findings presented here were from largely post-hoc analyses of a multicenter, parallel, double-blind, placebo-controlled study designed to evaluate the efficacy and safety of duloxetine in patients 65 years of age or older with MDD.[25] Following a screening period (5–9 days) and a 1-week, double-blind, placebo lead-in period, eligible patients (N=311) were randomly assigned to eight weeks of treatment with either duloxetine 60mg/day (n=207) or placebo (n=104) (Figure 1). At the end of eight weeks of treatment, patients entered a one-week, double-blind discontinuation phase, at which time the dosage of duloxetine was tapered.

The objective of this work was to compare the efficacy and tolerability of duloxetine 60mg/day versus placebo in treating elderly patients with major depressive disorder (MDD) and concurrent anxiety symptoms.

Using baseline item 10 (Psychic Anxiety) of the 17-item Hamilton Depression Rating Scale (HAMD17),[35] the patients were placed into one of two subgroups—one subgroup being patients with a baseline item 10 score of 0 or 1 (no or mild anxiety symptoms) or a second subgroup with a baseline score of 2, 3, or 4 (moderate to severe anxiety symptoms). These subgroups will be referred to as low anxiety patients and high anxiety patients, respectively. We also evaluated the efficacy of duloxetine in patients <75 and ?75 years old.

Patients. Patients were male or female outpatients 65 years of age or older who met criteria for MDD as defined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV).[36] The diagnosis of MDD was confirmed by the Mini International Neuropsychiatric Interview.[37] The patients were required to have a HAMD17 total score ?18 at screening (Visit 1) and placebo lead-in (Visit 2), as assessed by a trained clinican, a Mini-Mental State Examination (MMSE)[38] score of ?20 with or without mild dementia, and at least one previous episode of MDD. The study protocol was approved by the ethics committee for each site in accordance with the Declaration of Helsinki. Each patient provided written informed consent prior to any study procedures or administration of study drug.

Patients were excluded for the following reasons: Current primary Axis I diagnosis other than MDD or mild dementia (including dysthymia or psychotic depression); any anxiety disorder as a primary diagnosis within the past year; previous diagnosis of psychotic disorder; organic mental disorder, moderate to severe dementia, or mental retardation diagnosis; serious or unstable medical illness, psychological condition, or clinically significant laboratory abnormality that, in the opinion of the investigator, would compromise participation in this study or be likely to lead to hospitalization during the course of the study; or ALT, AST, or GGT >1.5 times upper limit of normal, based on Lilly reference ranges.[39]

Efficacy measures. Anxiety symptoms were assessed using the Psychic Anxiety item and the Somatic Anxiety item (Item 11) from the HAMD[17], and also the Anxiety/Somatization subscale, which consists of the sum of items 10, 11, 12 (gastrointestinal somatic symptoms), 13 (general somatic symptoms), 15 (hypochondriasis), and 17 (insight). The subscale score ranges from 0 to 18. Additional measures used in the analyses were the Geriatric Depression Scale (GDS),[40] Visual Analog Scales (VAS) for pain,[41] and a composite cognitive score based on four tests including the following: 1) Verbal Learning and Recall Test (VLRT), adapted from the Rey Auditory Verbal Learning Test,[42,43], 2) Symbol Digit Substitution Test (SDST),[44] 3) 2 Digit Cancellation Test (2DCT),[45,46] and 4) Letter-Number Sequencing Test (LNST).[44]

The 36-item Short-Form Health Survey (SF-36)[47] was used to assess health outcomes. The SF-36 consists of 36 items that calculate eight health domains (subscales): bodily pain, general health, mental health, physical functioning, role-physical, role-emotional, social function, and vitality.

Two summary scores also were derived from the 36 items to represent a Mental Component Summary and a Physical Component Summary.

Tolerability measures. Tolerability measures included collection of spontaneously reported treatment-emergent adverse events (TEAEs) and adverse events reported as the reason for discontinuation.

Statistical analyses. All analyses were conducted using the intention-to-treat (ITT) sample, comprising all randomly assigned patients even if the patient did not take the assigned treatment, did not receive the correct treatment, or otherwise did not follow the protocol. Some analyses were defined a priori in the protocol and included the Anxiety/Somatization subscale in the overall study group and by age subgroups, and treatment-emergent adverse events for the overall study group; all other analyses were post hoc. All randomly assigned patients were included in the tolerability analyses, and all patients with a baseline and at least one post-baseline measurement were included in the efficacy analyses. The term significant for treatment comparisons indicates statistical significance (two-sided p?0.05). Because these were secondary analyses, no adjustments were made for multiple comparisons. Throughout this manuscript, mean refers to the raw mean unless the least-squares (LS) mean is specified.

Unless specified otherwise, baseline refers to the last nonmissing pre-randomization observation, and endpoint refers to the last nonmissing observation during the acute treatment phase. Total scores (e.g., composite cognitive score, GDS and HAMD17 total scores, and Anxiety/Somatization subscale) were considered to be missing if any of the item scores were missing.

Changes in continuous efficacy variables from baseline to endpoint, overall and within subgroups, were analyzed using a fixed-effects analysis of covariance (ANCOVA) model that included terms for treatment, investigator, and baseline score. Continuous demographic and baseline variables were evaluated using a fixed-effects analysis of variance (ANOVA) model that included terms for treatment and investigator. These analyses are referred to as mean change (MC) analyses. Some efficacy variables measured repeatedly over time were analyzed using a mixed-effects model, likelihood-based, repeated measures (MMRM) approach. The model included the fixed categorical effects of treatment, investigator, visit, and treatment-by-visit interaction, as well as the continuous fixed covariates of baseline score and baseline score-by-visit interaction. Based on results from previous studies, the unstructured (co)variance structure was used to model the within-patient errors. Categorical measures, overall and within subgroups, were analyzed using Fisher’s exact test.

The consistency of the effect of duloxetine compared with placebo in patient subgroups of age (<75 versus ?75 years) and anxiety (low vs. high) was investigated for continuous variables by adding the relevant subgroup and treatment-by-subgroup interaction terms to the ANCOVA model.

Path analysis[48] was performed to test the direct treatment effect on changes in cognition and anxiety. For example, in this analysis, pre-specified pathways (figure 2) described the causal relationships for the testing: the treatment has effect on change in Psychic Anxiety (direct effect), and the treatment effect on cognition (COG) or depression (GDS) has an impact on change in Psychic Anxiety (indirect effect). The significance of the direct treatment effect was tested by a Student’s t-test in the regression model where change in Psychic Anxiety was a dependent variable, and treatment, baselines (Psychic Anxiety, COG, and GDS), change in COG, and change in GDS were regressors. In addition, with two other regression models for change on COG and GDS, respectively, the indirect treatment effect via change in COG (or GDS) was a result of multiplying the treatment coefficient by the coefficients of the corresponding change in the first described regression model. The percentage of direct and indirect effects on the total treatment effect was computed and presented. The situation with one indirect effect was analyzed in an analogous manner.

TEAEs were defined as events that first occurred or worsened post-randomization during the acute therapy phase as compared with the maximum pre-randomization severity. Events were reported using Preferred Terms for Version 7.0 of the Medical Dictionary for Regulatory Activities (MedDRA).

The association between baseline values of Psychic Anxiety and the individual VAS items was evaluated using Spearman’s Rho correlation coefficient. As specified in the protocol, 200 duloxetine- and 100 placebo-treated patients were to provide 80-percent power to detect an effect size (difference between mean changes in the composite cognitive score divided by the common standard deviation) of 0.35, using a two-sided test at a five-percent significance level and assuming 95 percent of patients have data for the analysis.

Results

Patient characteristics and disposition. Patient demographics and baseline disease characteristics for the low anxiety (n=112; placebo, n=35; duloxetine, n=77) and high anxiety (n=199; placebo, n=69; duloxetine, n=130) subgroups are shown in table 1. Study participants were mainly Caucasian and female, with a mean age of about 73 years. Correlations between the baseline Psychic Anxiety score and baseline VAS item scores were primarily weak. The range of the correlation values (rho) was small (-0.046 to +0.138), with only Severity of Headache reaching statistical significance (rho=0.138, p=0.015).

Overall, 78.3 percent of duloxetine- and 76.9 percent of placebo-treated patients completed the eight weeks of acute therapy. The numbers of patients in each age subgroup were: <75 (n=212; placebo, n=69; duloxetine, n=143), and ?75 (n= 99; placebo, n=35: duloxetine, n=64).

Efficacy. Overall and by age subgroup.Baseline to endpoint improvement on Psychic Anxiety was significantly greater in the duloxetine group compared with placebo group in all patients (least-squares mean change: -0.62 vs. -0.18, p<0.001) and in both the <75-year-old age group
(-0.63 vs. -0.17, p=0.002) and ?75-year-old age group (-0.61 vs. 0.0, p=0.007). The treatment-by-age interaction was not significant (p=0.818), suggesting that the effect of duloxetine versus placebo is consistent across the age subgroups. There were no significant differences on Somatic Anxiety between treatment groups overall (-0.31 vs. -0.22, p=0.342) or by age (<75-year-old age group [-0.28 vs. -0.21, p=0.561] and ?75-year-old age group [-0.43 vs. -0.43, p=0.998]); the treatment-by-age interaction was not significant (p=0.824). Mean change in the Anxiety/Somatization subscale was significantly in favor of duloxetine compared with placebo overall (-1.88 vs. -0.99. p=0.002) and in patients <75 (-1.90 vs. -0.80, p=0.003) but not in patients ?75 (-1.92 vs. -1.25, p=0.312). The treatment-by-age interaction was not significant (p=0.458).

Duloxetine-treated patients demonstrated significantly greater improvement on Psychic Anxiety at each visit compared with placebo-treated patients using the repeated measures analysis (figure 3). There was no difference between treatments on Somatic Anxiety at any visit (figure 4). The duloxetine group showed significant improvement on the Anxiety/Somatization subscale at Weeks 4 and 8 compared with the placebo group (figure 5).

Effects of low and high anxiety on HAMD and GDS. Patients treated with duloxetine showed significantly greater improvement on the HAMD[17] total score in the high anxiety subgroup (least-squares mean change: -7.56 vs. -3.99, p<0.001) but not in the low anxiety subgroup (-4.39 vs. -4.13, p=0.850). The treatment-by-anxiety subgroup interaction was significant (p=0.092). Similarly, patients treated with duloxetine showed significantly greater improvement on the GDS total score in the high anxiety subgroup (-4.70 vs. -2.14, p=0.006) but not in the low anxiety subgroup (-3.14 vs. -0.97, p=0.162). However, there was not a significant treatment-by-anxiety subgroup interaction (p=0.933).

High anxiety subgroup. In the high anxiety subgroup, patients treated with duloxetine showed significantly greater improvement for Psychic Anxiety (least-squares mean change: -1.08 vs. -0.45, p<0.001), Anxiety/Somatization subscale (-2.42 vs. -1.28, p=0.002), and the composite cognitive score (1.63 vs. 0.20, p=0.024). For VAS scores, duloxetine showed significant improvement compared with placebo on only severity of back pain (-3.69 vs. 3.04, p=0.05). For the SF-36, duloxetine was significantly better at improving both the Mental and Physical Component Summaries, as well as six of the eight subscale scores compared with placebo (table 2).

Duloxetine-treated patients demonstrated significantly greater improvement on Psychic Anxiety at Weeks 4 and 8 compared with placebo-treated patients using the repeated measures analysis (figure 3).

There was no difference between treatments on Somatic Anxiety at any visit (figure 4). The duloxetine group showed significant improvement on the Anxiety/Somatization subscale at only Week 8 compared with the placebo group (figure 5).

Path analyses found that improvement in Psychic Anxiety and the Anxiety/Somatization subscale was primarily due to the statistically significant direct effect of duloxetine treatment rather than an indirect effect through the improvement of the composite cognitive score (table 3).

Conversely, the improvement of the composite cognitive score was also primarily a direct effect of duloxetine rather than an indirect effect through Psychic Anxiety or the Anxiety/Somatization subscale. Therefore, duloxetine has a direct effect on the improvement of both cognition and anxiety symptoms. The indirect effect of the GDS accounted for more than one-third of the improvement on Psychic Anxiety and the Anxiety/Somatization subscale; the direct effect was significant for only Psychic Anxiety. This indirect effect was likely due to the composition of the GDS; that is, a number of the items are anxiety related. For change in Psychic Anxiety and also the Anxiety/Somatization subscale, the indirect effects through COG and GDS were similar whether they were used individually or jointly, although as would be expected, the direct effects are quite different.

Tolerability. Overall, treatment-emergent adverse events of anxiety and related symptoms showed similar rates of occurrence in patients taking either duloxetine or placebo, including anxiety (duloxetine 2.9% vs. placebo 1.0%, p=0.431), insomnia (duloxetine 3.9% vs. placebo 4.8%, P=0.766), nervousness (duloxetine 1.9% vs. placebo 0.0%, p=0.305), and agitation (duloxetine 0.5% vs. placebo 0.0%, p=1.000).

High anxiety subgroup. Common TEAEs were similar in the high anxiety subgroup between duloxetine-treated patients and placebo-treated patients (table 4). The exception was a significantly higher incidence of dry mouth with duloxetine (duloxetine 15.4%, placebo 0.0%; p<0.001).

Discontinuation rates due to any adverse event were similar between duloxetine and placebo in the high anxiety subgroup (duloxetine 9.2% vs. placebo 10.1%, p=0.806). For individual events, there was a significant difference between treatments for only nausea, in which the rate was higher for placebo than for duloxetine (duloxetine 0.0% vs. placebo 4.3%, p=0.040). No individual adverse events leading to discontinuation occurred at a rate greater than or equal to five percent.

Discussion

In these analyses, duloxetine demonstrated efficacy in reducing certain symptoms related to anxiety in elderly patients with MDD. For all randomized patients and also for the high anxiety subgroup, duloxetine-treated patients showed significant improvement compared with placebo-treated patients in both Psychic Anxiety and the Anxiety/Somatization subscale, although not in Somatic Anxiety. The high anxiety subgroup taking duloxetine also showed significantly greater improvement than their placebo counterparts on most of the secondary measures, but not on the VAS, except for severity of back pain. Duloxetine was efficacious for Psychic Anxiety in both the <75 and ?75 subgroups, whereas the Anxiety/Somatization subscale was significantly improved with duloxetine in only the younger age group. However, the degree of improvement with duloxetine was nearly the same in the ?75 subgroup and the <75 subgroup. The lack of statistical significance was likely due to a greater placebo response and fewer patients in the ?75 subgroup compared with the <75 subgroup.

The effect of duloxetine treatment on Psychic Anxiety in all patients began at Week 1. However, in the high anxiety subgroup, significant improvement did not begin until Week 4. The placebo response was quite dramatic in the high anxiety subgroup for the first two weeks, which likely contributed to this finding. Interestingly, the high anxiety placebo group started worsening at Week 8. In contrast, the overall placebo group was still showing improvement from Week 4 to Week 8. In a review by Dunner et al,30 middle-aged patients with MDD taking duloxetine 60mg/day (2 studies) showed similar mean improvements compared with placebo on the Psychic Anxiety and the Anxiety/Somatization subscale as in the present study. The effect of duloxetine on Psychic Anxiety began at Week 1 in one study and at Week 2 in the other study.

Duloxetine did not separate from placebo on the Anxiety/Somatization subscale until four weeks and eight weeks after treatment for the overall group and high anxiety group, respectively. In two studies reviewed by Dunner and colleagues,[30] duloxetine 60mg/day did not separate from placebo until Weeks 2 and 7. The Anxiety/Somatization subscale consists of six items, three of which are related to somatic symptoms. In the present study, duloxetine-treated patients did not separate from placebo-treated patients on Somatic Anxiety. This lack of separation on Somatic Anxiety explains the slower onset of significant improvement with duloxetine on the Anxiety/Somatization subscale.

Some potentially interesting differences were noted regarding baseline illness characteristics. For example, the high anxiety subgroup averaged almost two more previous episodes of MDD than did the low anxiety subgroup. This could suggest that having more depressive episodes might lead to increased levels of anxiety. Alternatively, patients that have higher levels of anxiety also may become depressed more often. A recent study found that a prior history of GAD, especially in men, led to increased depressive episodes.[49] The high anxiety subgroup also had a mean HAMD17 total score at randomization that was 4.5 points higher than that of the low anxiety subgroup. Taking into account HAMD items 10 and 11, the difference between high and low anxiety subgroups is more than two points. Therefore, patients with higher levels of anxiety or those who have a comorbid anxiety disorder also may experience more severe MDD symptomatology, which has been suggested in previous studies.[6–8] In support of this observation, the GDS total score is more than five points higher in the high anxiety subgroup compared with the low anxiety subgroup.

An additional noteworthy finding of baseline disease characteristics is that the high anxiety subgroup had a mean VAS overall pain score more than five points higher than the low anxiety subgroup. The high anxiety subgroup also had baseline means at least five points higher compared with the low anxiety subgroup for time in pain while awake and severity of headache. The high anxiety subgroup also had lower (worse) baseline scores for all domains and the Mental and Physical Component Summaries of the SF-36 (data not shown). These findings are supported by Lenze and colleagues[6] who found that older patients with MDD and a comorbid anxiety disorder have more severe somatic symptoms.

Duloxetine has been shown to significantly improve painful somatic symptoms in patients with MDD[50,51] and in managing patients with diabetic peripheral neuropathic pain.[52–54] In the high anxiety subgroup, duloxetine-treated patients showed numerically greater improvement on four of six VAS measures compared with placebo-treated patients, but significant improvement occurred on only severity of back pain. Although the argument that duloxetine might be less beneficial in treating physical pain in elderly patients with MDD has been broached, another explanation is that the baseline VAS scores were primarily mild in severity, which did not allow for much improvement. A post-hoc analysis of patients with postoperative pain by Collins and colleagues[55] found that a VAS score of at least 30 is required to have a level of pain considered to be of moderate intensity. Indeed, most of the baseline VAS scores in the present study were in the 20s or lower.

Path analyses showed that improvement of Psychic Anxiety was primarily a direct treatment effect of duloxetine. Indirect treatment effects were minimal via the COG, although the GDS did account for about one-third of the improvement in Psychic Anxiety. For the Anxiety/Somatization subscale, the indirect treatment effect through the GDS contributed 60 percent of the improvement. Thus, improvement in the GDS after duloxetine treatment accounted for more improvement than the direct effect of duloxetine on the Anxiety/Somatization subscale. These results suggest the benefit of a drug that can improve not only depression, but also anxiety symptoms, which are often comorbid in patients with MDD.

Comorbid anxiety disorders are prevalent in geriatric patients with MDD. A study in the Netherlands found that 47.5 percent of elderly patients with MDD had a comorbid anxiety disorder.[56] A study conducted in the US found that 23 percent of elderly patients with MDD had a concurrent anxiety disorder (excluding generalized anxiety disorder).[6] In addition, nearly 28 percent of the patients had a current diagnosis of generalized anxiety disorder. Overall, one-half of these elderly patients with MDD had a concurrent anxiety disorder. A study of nursing home residents found that 65 percent of those patients with a diagnosis of major depression displayed anxiety symptoms.[57] Thus, a significant number of elderly patients with MDD will need to be treated not only for depressive symptoms, but anxiety symptoms as well.

Unfortunately, there have been few studies that have evaluated the efficacy of pharmacotherapy for MDD with comorbid anxiety, or even anxiety disorders alone, in elderly populations. A recent small prospective trial found that significantly more elderly patients with anxiety disorders responded to citalopram compared with placebo.[58] An open-label trial found that the SSRI fluvoxamine was beneficial in treating anxiety disorders in older patients, as well as those patients having comorbid depressive disorders.[59] Several SSRIs have shown to be equally effective in elderly patients with either anxious depression or non-anxious depression.[60,61] A post-hoc analysis of four double-blind, placebo-controlled trials found that duloxetine was effective in reducing anxiety symptoms in middle-aged patients with MDD.[30] The present study demonstrates that duloxetine is significantly more effective than placebo in reducing anxiety symptoms in patients 65 years of age and older, including those over 75.

Even though nausea was experienced by three times as many patients in the duloxetine group as in the placebo group, the discontinuation rate due to nausea was actually lower with duloxetine. Of the four placebo patients who discontinued due to nausea, two had treatment-emergent nausea and two had nausea prior to randomization that did not worsen. The nausea for the one duloxetine patient who discontinued due to nausea also began prior to randomization and did not worsen. Moreover, it has been shown that of the duloxetine-treated patients who experienced nausea, most rated it as mild to moderate, and also showed that the nausea rapidly resolves with continued treatment.[62]

A limitation of this analysis was that the study excluded patients with primary DSM-IV anxiety disorders in the year before enrollment. The study also excluded patients with acute and unstable medical conditions that are common among elderly patients. The study lacked a specific scale intended to measure anxiety, such as the Hamilton Rating Scale for Anxiety (HAM-A).[62] Long-term effects of duloxetine on anxiety were not assessed due to the short-term treatment (8 weeks) for MDD. Patients had limited dose flexibility during the study, which may not be typical of clinical practice for the elderly. Finally, there were a relatively small number of patients available for statistical analyses by subgroup, especially in the ?75 age group.

These analyses demonstrated that duloxetine was efficacious in reducing the symptoms of anxiety in elderly patients with MDD. Duloxetine was well tolerated, as noted by a low incidence of anxiety-related adverse events, rates of TEAEs similar to placebo, and low rates of discontinuation due to adverse events. Prospective trials designed to determine the tolerability and efficacy of duloxetine in depressed older patients having a comorbid anxiety disorder or having severe anxiety symptoms are needed to confirm the present results.

Acknowledgments

The authors thank Andrew Chen, MS, and Barry M. Brolley, MS, for their assistance with statistical analyses and Elizabeth Agostinelli, MA, for editorial review.

References

1. Small GW, Birkett M, Meyers BS, et al. Impact of physical illness on quality of life and antidepressant response in geriatric major depression. Fluoxetine Collaborative Study Group. J Am Geriatr Soc 1996;44:1220–25.
2. Zimmerman M, Chelminski I, McDermut W. Major depressive disorder and axis I diagnostic comorbidity. J Clin Psychiatry 2002;63:187–93.
3. Rush AJ, Zimmerman M, Wisniewski SR, et al. Comorbid psychiatric disorders in depressed outpatients: demographic and clinical features. J Affect Disord 2005;87:43–55.
4. Kessler RC, Nelson CB, McGonagle KA, et al. Comorbidity of DSM-III-R major depressive disorder in the general population: results from the US National Comorbidity Survey. Br J Psychiatry 1996;168(Suppl 30):17–30.
5. Gorman JM. Comorbid depression and anxiety spectrum disorders. Depress Anxiety 1996; 4:160–8.
6. Lenze EJ, Mulsant BH, Shear MK, et al. Comorbid anxiety disorders in depressed elderly patients. Am J Psychiatry 2000;157:722–8.
7. Lenze EJ, Mulsant BH, Shear MK, et al. Comorbidity of depression and anxiety disorders in later life. Depress Anxiety 2001;14:86–93.
8. Schoevers RA, Deeg DJH, van Tillburg W, Beekman ATF. Depression and generalized anxiety disorder: Co-occurrence and longitudinal patterns in elderly patients. Am J Geriatr Psychiatry 2005;13:31–9.
9. DeLuca AK, Lenze EJ, Mulsant BH, et al. Comorbid anxiety disorder in late life depression: Association with memory decline over four years. International J Geriatr Psychiatry 2005;20:848–54.
10. Allgulander C, Lavori PW. Causes of death among 936 elderly patients with ‘pure’ anxiety neurosis in Stockholm County, Sweden, and in patients with depressive neurosis or both diagnoses. Compr Psychiatry 1993;34:299–302.
11. Bielski RJ, Bose A, Chang CC. A double-blind comparison of escitalopram and paroxetine in the long-term treatment of generalized anxiety disorder. Ann Clin Psychiatry 2005;17:65–9.
12. Dahl AA, Ravindran A, Allgulander C, et al. Sertraline in generalized anxiety disorder: efficacy in treating the psychic and somatic anxiety factors. Acta Psychiatrica Scandinavica 2005;111:429–35.
13. Sheehan DV, Mao CG. Paroxetine treatment of generalized anxiety disorder. Psychopharmacol Bull 2003;37(Suppl 1):64–75.
14. Varia I, Rauscher F. Treatment of generalized anxiety disorder with citalopram. Int Clin Psychopharmacol 2002;17:103–7.
15. Tollefson GD, Holman SL, Sayler ME, Potvin JH. Fluoxetine, placebo, and tricyclic antidepressants in major depression with and without anxious features. J Clin Psychiatry 1994;55:50–9.
16. Russell JM, Koran LM, Rush J, et al. Effect of concurrent anxiety on response to sertraline and imipramine in patients with chronic depression. Depress Anxiety 2001;13:18–27.
17. Rudolph RL, Entsuah R, Chitra R. A meta-analysis of the effects of venlafaxine on anxiety associated with depression. J Clin Psychopharmacol 1998;18:136–44.
18. Katzman M. Venlafaxine in the treatment of anxiety disorders. Expert Rev Neurother 2004;4:371–81.
19. Nemeroff CB, Schatzberg AF, Goldstein DJ, et al. Duloxetine for the treatment of major depressive disorder. Psychopharmacol Bull 2002;36:106–32.
20. Detke MJ, Lu Y, Goldstein DJ, et al. Duloxetine, 60mg once daily, for major depressive disorder: A randomized double-blind placebo-controlled trial. J Clin Psychiatry 2002;63:308–15.
21. Detke MJ, Lu Y, Goldstein DJ, et al. Duloxetine 60mg once daily dosing versus placebo in the acute treatment of major depression. J Psychiatr Res 2002;36:383–90.
22. Detke MJ, Wiltse CG, Mallinckrodt CH, et al. Duloxetine in the acute and long-term treatment of major depressive disorder: A placebo- and paroxetine-controlled trial. Eur Neuropsychopharmacol 2004;14:457–70.
23. Goldstein DJ, Mallinckrodt C, Lu Y, Demitrack MA. Duloxetine in the treatment of major depressive disorder: A double-blind clinical trial. J Clin Psychiatry 2002;63:225–31.
24. Goldstein DJ, Lu Y, Detke MJ, et al. Duloxetine in the treatment of depression: A double-blind placebo-controlled comparison with paroxetine. J Clin Psychopharmacol 2004;24:389–99.
25. Raskin, J, Wiltse CG, Siegal A, et al. Efficacy of duloxetine on cognition, depression, and pain in elderly patients with major depressive disorder: An 8-week, double-blind, placebo-controlled trial. Am J Psychiatry:In press.
26. Rynn M, Russell J, Erickson J, et al. Efficacy and safety of duloxetine in the treatment of generalized anxiety disorder: A flexible-dose, progressive-titration, placebo-controlled trial. Depress Anxiety:In press.
27. Koponen H, Allgulander C, Erickson J, et al. Efficacy of duloxetine for the treatment of generalized anxiety disorder:Implications for primary care physicians. Prim Care Companion J Clin Psychiatry:In press.
28. Akkaya C, Sivrioglu EY, Akgoz S, et al. Comparison of efficacy and tolerability of reboxetine and venlafaxine XR in major depression and major depression with anxiety features: an open label study. Hum Psychopharmacol 2006;21:337–45.
29. Gorman JM, Papp LA. Efficacy of venlafaxine in mixed depression-anxiety states. Depress Anxiety 2000;12[S1]:77–80.
30. Dunner DL, Goldstein DJ, Mallinckrodt C, et al. Duloxetine in treatment of anxiety symptoms associated with depression. Depress Anxiety 2003;18:53–61.
31. Sonawalla SB, Spillmann MK, Kolsky AR, et al. Efficacy of fluvoxamine in the treatment of major depression with comorbid anxiety disorders. J Clin Psychiatry 1999;60:580–3.
32. Perugi G, Frare F, Toni C, et al. Open-label evaluation of venlafaxine sustained release in outpatients with generalized anxiety disorder with comorbid depression or dysthymia: Effectiveness, tolerability, and predictors of response. Neuropsychobiol 2002;46:145–9.
33. Goodnick PJ, Puig A, DeVane CL, Freund BV. Mirtazapine in major depression with comorbid generalized anxiety disorder. J Clin Psychiatry 2006;60:446–8.
34. Silverstone PH, Salinas E. Efficacy of venlafaxine extended release in patients with major depressive disorder and comorbid generalized anxiety disorder. J Clin Psychiatry 2001;62:523–9.
35. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960;23:56–62.
36. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. Washington, DC: American Psychiatric Press Inc., 2000.
37. Sheehan DV, Lecrubier Y, Sheehan KH, et al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59(Suppl 20):22–33.
38. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state.” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–98.
39. Thompson WL, Brunelle RL, Enas GG, Simpson PJ. Routine laboratory tests in clinical trials: interpretation of results. J Clin Research Drug Dev 1987;1:95–119.
40. Yesavage JA, Brink TL, Rose TL, et al. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 1982;17:37–49.
41. DeLoach LJ, Higgins MS, Caplan AB, Stiff JL. The visual analog scale in the immediate postoperative period: intrasubject variability and correlation with a numeric scale. Anesth Analg 1998;86:102–6.
42. Lezak M. Neuropsychological Assessment. New York: Oxford, 1995: 438–45.
43. REY A. L’examen psychologique dans les cas d’encephalopathie traumatique [French]. Archives de Psychologie 1941;28:286–340.
44. Wechsler D. Wechsler Adult Intelligence Scale, Third Edition. San Antonio, TX: The Psychological Corporation, 1997.
45. Neisser U. Visual search. Sci Am 1964; 210:94–102.
46. Mohs RC, Knopman D, Petersen RC, et al. Development of cognitive instruments for use in clinical trials of antidementia drugs: Additions to the Alzheimer’s Disease Assessment Scale that broaden its scope. The Alzheimer’s Disease Cooperative Study. Alzheimer Dis Assoc Disord 1997;11(Suppl 2):S13–S21.
47. Ware JE, Snow KK, Kosinski M, Gandek B. SF-36 Health Survey Manual and Interpretation Guide. Boston, MA: The Health Institute, New England Medical Center, 1993.
48. Retherford RD, Choe MK. Statistical Methods for Causal Analysis. Hoboken, NJ: John Wiley & Sons, Inc., 1993.
49. Hettema JM, Kuhn JW, Prescott CA, Kendler KS. The impact of generalized anxiety disorder and stressful life events on risk for major depressive episodes. Psychol Med 2006;36:789–95.
50. Brannan SK, Mallinckrodt CH, Brown EB, et al. Duloxetine 60mg once-daily in the treatment of painful physical symptoms in patients with major depressive disorder. J Psychiatr Res 2005;39:43–53.
51. Goldstein DJ, Lu Y, Detke MJ, et al. Effects of duloxetine on painful physical symptoms associated with depression. Psychosomatics 2004;45:17–28.
52. Raskin J, Pritchett YL, Wang F, et al. A double-blind, randomized multicenter trial comparing duloxetine with placebo in the management of diabetic peripheral neuropathic pain. Pain Med 2005;6:346–56.
53. Raskin J, Smith TR, Wong K, et al. Duloxetine versus routine care in the long-term management of diabetic peripheral neuropathic pain. J Palliat Med 2006;9:29–40.
54. Goldstein DJ, Lu Y, Detke MJ, et al. Duloxetine vs. placebo in patients with painful diabetic neuropathy. Pain 2005;116:109–18.
55. Collins SL, Moore RA, McQuay HJ. The visual analogue pain intensity scale: What is moderate pain in millimetres? Pain 1997;72:95–7.
56. Beekman ATF, de Beurs E, van Balkom AJLM, et al. Anxiety and depression in later life: Co-occurrence and communality of risk factors. Am J Psychiatry 2000;157:89–95.
57. Parmelee PA, Katz IR, Lawton MP. Anxiety and its association with depression among institutionalized elderly. Am J Geriatr Psychiatry 1993;1:46–58.
58. Lenze EJ, Mulsant BH, Shear MK, et al. Efficacy and tolerability of citalopram in the treatment of late-life anxiety disorders: Results from an 8-week randomized, placebo-controlled trial. Am J Psychiatry 2005;162:146–50.
59. Wylie ME, Miller MD, Shear MK, et al. Fluvoxamine pharmacotherapy of anxiety disorders in later life: preliminary open-trial data. J Geriatr Psychiatry Neurol 2000;13:43–8.
60. Cassano P, Soares CN, Cohen LS, et al. Sex- and age-related differences in major depressive disorder with comorbid anxiety treated with fluoxetine. Arch Womens Ment Health 2004;7:167–71.
61. Lenze EJ, Mulsant BH, Dew MA, et al. Good treatment outcomes in late-life depression with comorbid anxiety. J Affect Disord 2003;77:247–54.
62. Greist J, McNamara RK, Mallinckrodt C, et al. Incidence and duration of antidepressant-induced nausea: Duloxetine compared with paroxetine and fluoxetine. Clin Ther 2004;26:1446–55.
63. Hamilton M. The assessment of anxiety states by rating. Br J Med Psychol 1959;32:50–5.