These results support the efficacy and tolerability of duloxetine in the treatment of elderly patients with MDD with or without common comorbid medical conditions. Overall, duloxetine-treated patients demonstrated significantly greater improvement compared with placebo-treated patients on outcome measures that included the composite cognitive score, depression severity measures and several of the SF-36 and VAS measures. Few significant treatment-by-comorbidity subgroup interactions occurred for these efficacy variables, or for AEs reported as the reason for discontinuation or common TEAEs, suggesting that the efficacy and tolerability of duloxetine in the treatment of depression in elderly patients were not largely affected by comorbidity status. Specifically, the results regarding the impact of comorbid medical illness on treatment outcomes in MDD were consistent regardless of whether the outcomes were measured with the GDS, the HAMD17 or CGI-S.
This study reaffirms the high prevalence of medical comorbidity in patients with late-life depression because nearly 75% of patients participating in the study met criteria for the presence of medical comorbidity. This proportion may be an underestimate, as patients with unstable acute medical illness were excluded from the study. Moreover, we measured only three common medical comorbidities; a higher prevalence of comorbidities likely would have been measured if more medical illnesses were recorded. Most of the findings in patient baseline characteristics were as expected; for instance, mean age was higher in patients with vascular comorbidity and VAS overall pain severity was higher in patients with arthritis. The fact that patients with vascular comorbidity had more previous episodes and longer duration of current episode of MDD may be attributed to certain vascular-related factors contributing to depression, including brain damage from infarcts or microvascular effects (6
). Such factors are in line with the notion that physical illness and depression tend to have mutually worsening effects.
Although cognitive deficits in patients with MDD, especially the elderly patients, have been demonstrated in many studies (27
), antidepressant drugs do not routinely improve cognition in such patients (32
). Follow-up studies of patients treated for MDD have shown that some patients demonstrate poor performance on cognitive tests even after treatment (34
). In the present study, duloxetine-treated patients demonstrated similar improvement on the composite cognitive score with or without a comorbid illness. However, even though the treatment-by-subgroup interaction was not significant, patients with any of the three comorbid illnesses and treated with duloxetine showed a significantly greater improvement on the composite cognitive score compared with the placebo group. In the subgroup without a comorbid illness, duloxetine did not show significantly greater improvement than did placebo. Although the explanation for this lesser improvement in the group without a comorbid illness is unknown, it could be an artifact as a result of the relatively small number of patients in that subgroup. Nevertheless, the augmenting effect of duloxetine on 5-HT and NE activity may make it particularly beneficial in treating the cognitive deficits associated with depression as imbalance or deficiency in either 5-HT or NE neurotransmission has been found to contribute to cognitive deficits (16
). Treatment of depression with tricyclics, such as imipramine, has been found to improve cognitive function despite the detrimental anticholinergic effects (35
The comorbidity of pain and depression in elderly patients is of special significance when considering the high prevalence of pain in this age group (36
). A direct relationship between pain severity and depression has been identified in elderly patients with chronic pain (37
). In the present analyses, duloxetine demonstrated statistically significant improvements in two of the six pain measures: the VAS for back pain and time in pain while awake. These findings suggest that duloxetine may be effective in relieving pain caused by chronic conditions, including arthritis and vascular disease. Duloxetine has been shown to be effective in the management of diabetic peripheral neuropathic pain (38
) and in improving painful physical symptoms independent of improvement of depression severity in younger MDD populations (40
). However, patients enrolled in this study were not selected specifically for pain and the pain reported was generally not severe. It is possible that more or less robust efficacy for duloxetine might be observed in patients with depression who have a higher baseline pain severity.
The present results are consistent with findings in acute treatment studies of antidepressants, including SSRIs, in medically ill patients with depression (41
). A review of 18 studies in mostly non-elderly patients with depression and having at least one physical disorder found that antidepressant treatment was significantly better than placebo in improving depressive symptoms (42
). In an 8-week study, older patients treated with sertraline had similar responses on the HAMD17
total score, CGI-S and CGI-I whether they had a comorbid medical illness or not (41
). In a study by Small et al. (45
), geriatric patients with major depression and with none, 1, 2, 3, 4 or ≥ 5 chronic physical illnesses were treated with fluoxetine 20 mg/day or placebo for 6 weeks. The analyses showed that the number of chronic illnesses did not have an influence on treatment outcomes. One outcome of interest from that study was the finding that patients with a greater number of illnesses had a greater response to fluoxetine, whereas the opposite was found with the placebo group, although the total number of patients was somewhat small in the subgroup with no chronic illnesses (N
= 73). A similar finding was found with elderly patients with vascular illnesses treated with sertraline (44
). In a pooled analysis of two studies, patients were placed into three groups: with hypertension, with vascular illness and no hypertension, and no hypertension or cardiovascular illness. The patients in the first two groups showed a consistently greater percentage of responders on the HAMD and CGI-I compared with the third group.
In the present study, the absolute response and remission rates were relatively low, which may be attributed to the short 8-week study duration, as well as the fixed dosing schedule. Nevertheless, the relative advantages of duloxetine in response and remission rates were convincing, as evidenced by the fact that they were twice the placebo rates and the differences were statistically significant. In general, depression trials in the elderly patients are more difficult to show positive efficacy results for active treatments over placebo than do trials in the general population (46
). In a recent placebo-controlled study of fluoxetine and venlafaxine in patients ≥ 65 years with MDD, there were no significant differences among the three treatment groups in the change of HAMD21
, MADRS or CGI scores, and the difference in the proportion of remitters at the last on-therapy visit was not statistically significant (47
). Similarly, in a group of community-dwelling elderly patients ≥ 75 years with depression, citalopram was not more effective than placebo for the treatment of depression (48
). However, in two larger studies in patients ≥ 60 years with depression, sertraline or fluoxetine was more effective than placebo (41
). The remission rates in these studies were generally low, and the differences between treatment groups were small (20–35% for active treatments, 18–33% for placebo).
Duloxetine was well tolerated in patients with medical comorbidity as well as in patients with no medical comorbidity. More importantly, common TEAEs and discontinuation caused by AEs were comparable to those seen in younger, more general populations receiving duloxetine (50
). This is also in line with what has been found with the SSRI sertraline (41
). A study of patients with late-life depression and having vascular disease, diabetes mellitus or arthritis actually had a lower percentage of patients experiencing the most common TEAEs than did patients with no comorbid illnesses (41
). In addition, the percentage of patients discontinuing for any reason, or specifically for AEs, were also lower in the comorbid illness group. In the pooled analysis by Krishnan et al. (44
), there were also no differences between the groups with comorbid illness compared with the group with no comorbid illness in rates of TEAEs and in discontinuation caused by TEAEs.
A number of limitations should be kept in mind when evaluating the results of this study. First, the study excluded patients with acute and unstable medical conditions that are common among elderly populations. The 8-week study duration of the trial may be relatively short for a study in elderly patients. Roose and Sackeim (51
) have suggested that a minimum of 12-week duration may be necessary to identify response or remission. Patients had limited-dose flexibility during the study, which may not be typical of clinical practice for an elderly population. Comorbid medical conditions could also be based on medical histories and patient report rather than a diagnosis by a physician during the physical examination. Finally, the number of patients was relatively small in the diabetes group.
In conclusion, results from this placebo-controlled trial support the hypothesis that duloxetine is effective and well tolerated in treating elderly patients with MDD. The duloxetine vs. placebo treatment effect on cognition, depression and quality-of-life in elderly patients with MDD was not largely affected by the presence or absence of one or more of the three medical comorbidities (vascular disease, diabetes and arthritis) that frequently occur in the elderly population.