Efficacy of milnacipran on cognitive dysfunction with post-stroke depression: Preliminary open-label study


  • SHINJI SATO md ,

    Corresponding author
    1. Department of Psychiatry, Tsukuba Memorial Hospital,
    2. Department of Neuropsychiatry, Institute of Clinical Medicine, Tsukuba University, Tsukuba, Departments of
      Shinji Sato, MD, Department of Psychiatry, Tsukuba Memorial Hospital Kaname 1187-299, Tsukuba, Ibaragi, Japan. Email: ssato@bg8.so-net.ne.jp
    Search for more papers by this author

    1. Department of Neuropsychiatry, Institute of Clinical Medicine, Tsukuba University, Tsukuba, Departments of
    2. Psychiatry and
    Search for more papers by this author

    1. Department of Neuropsychiatry, Institute of Clinical Medicine, Tsukuba University, Tsukuba, Departments of
    Search for more papers by this author
  • HITOSHI OHTA md , phd,

    1. Rehabilitation, Ibaraki Prefectural University of Health Sciences, Ibaraki, Japan
    Search for more papers by this author
  • TAKASHI ASADA md , phd

    1. Department of Neuropsychiatry, Institute of Clinical Medicine, Tsukuba University, Tsukuba, Departments of
    Search for more papers by this author

Shinji Sato, MD, Department of Psychiatry, Tsukuba Memorial Hospital Kaname 1187-299, Tsukuba, Ibaragi, Japan. Email: ssato@bg8.so-net.ne.jp


Abstract  The aim of this study was to assess the therapeutic efficacy of the serotonin norepinephrine reuptake inhibitor (SNRI), milnacipran, on both cognitive impairment and depression in post-stroke depression (PSD) patients. A total of 18 PSD patients, approximately 3 months after stroke, were divided into two groups, milnacipran and control. A total of 10 patients were assigned to the milnaciprane group and eight were assigned to control group. Their cognitive impairment and mood symptoms were measured using the Mini-Mental State Examination (MMSE) and Hamilton Depression Rating Scale (HAM-D) both at the time of admission and at discharge, an interval of approximately 3 months. This study examined the changes in both MMSE and HAM-D scores during the study period. A significant time-by-group interaction for results of the MMSE was observed, although there was no significant difference between the two groups on the HAM-D. Amelioration of cognitive impairment was greater in the milnacipran group than the control group. For PSD patients, milnacipran is effective in improving cognitive dysfunction.


In Japan, cerebral vascular disease is the third highest cause of mortality in adults.1 Cerebral infarction or hemorrhage (stroke) not resulting in death nearly always leads to physical and/or cognitive impairment. Mood disturbance is the most common psychiatric disorder after stroke. In earlier studies, the prevalence rate of post-stroke depression (PSD) varied widely, from 25% to 70%, depending on patient selection and diagnostic criteria.2 PSD correlates with the severity of neurological deficits and impairment of daily activities.3–5 Several investigators have also reported a significant association between PSD and cognitive dysfunction.3,6–9

There are two contrary interpretations for the association between depression and cognitive deficits. One is that depression leads to reversible dementia (dementia of depression) or pseudodementia.10,11 Kimura et al.10 examined 47 PSD patients, dividing them into two groups according to response to therapy. The responder group consisted of those with more than a 50% reduction in Hamilton Self-Rating Depression Scale (HAM-D)12 score and the non-responder group consisted of those that failed to reduce their HAM-D score. The two groups were followed for 3–6 months and analysis conducted using a repeated-measures anova of Mini-Mental State Examination (MMSE)13 score. This analysis demonstrated a significant time-by-group interaction. Those findings were interpreted as preliminary evidence that cognitive deficits may improve, with a recovery of the severity of depression. Cognitive impairment leading to depression is termed depressive dementia.5,14–16 Robinson et al.15 compared antidepressants (noritriptyline and fluoxetine) to placebo in the treatment of depression and cognitive impairment after stroke. A repeated-measure anova analysis comparing HAM-D scores for all patients in antidepressant or placebo groups demonstrated a significant time-by-group interaction. However, no significant correlations with treatment or time were observed for MMSE score. These findings, therefore, support the ‘depression of dementia’ hypothesis. Therefore, characterization of the relationship between recovery of cognitive dysfunction and treatment of depression remains incomplete.

In this study, change in both cognitive function and mood of PSD patients after treatment with milnacipran, a selective serotonin and norepinephrine reuptake inhibitor (SNRI), was investigated in an open-label controlled study. Milnacipran (Tredomin; Pierre Fabre, French) has recently been registered for use in Japan.

The efficacy of milnacipran is similar to amitriptyline17 and imipramine,18 however, it has less anticholinergic adverse effects than both of these tricyclic antidepressants.17,18 Results of meta-analysis indicate that milnacipran is more efficacious than the Serotonin Selective Reuptake Inhibitors (SSRI; fluoxetine and fluvoxamine), in terms of efficacy and that the two are equally well tolerated for the treatment of major depression.19 Furthermore, milnacipran does not inhibit any cytochrome P450 isoenzyme subtypes and does not give rise to an active metabolite.20 These findings suggest that there is a lower risk of drug interactions using milnacipran compared to conventional antidepressants or SSRI. As many PSD patients take other medications, such as antihypertensives, the characteristics of milnacipran suggest that it is likely to be advantageous for such patients.21

The authors hypothesize that the SNRI, milnacipran, will improve both mood symptoms and cognitive impairment in PSD patients compared to PSD patients not treated with this compound. To the authors’ knowledge, this is the first case-control study using milnacipran for the treatment of PSD patients.


Subjects included 18 depressed patients out of a total of 123 patients (93 men and 30 women; mean ± SD age 58.8 ± 10.0 years; range, 34–83 years) that had recently developed strokes. They had been admitted to the Ibaraki Prefectural University of Health Sciences Hospital (IPUHS hospital, Ibaraki, Japan ) from May 2002 to December 2003. In this study, the exclusion criteria were: (i) presence of severe aphasia, especially fluent aphasia; (ii) presence of severe cognitive dysfunction (MMSE < 14); and (iii) presence of another chronic disorder (i.e. severe Parkinson’s disease, cardiac disease, cancer etc.) The study protocol was approved by the ethics committee of Ibaraki Prefectural University of Health Science. All patients gave written informed consent before inclusion into the study.

The IPUHS hospital was established as a rehabilitation center in Ibaraki Prefecture. All stroke patients with physical and/or cognitive dysfunction in this prefecture were eligible for admission after acute or subacute phases of stroke. In the acute phase, generally within 3 months after the onset of stroke, patients receive treatment in the neurological unit of general hospitals. After this treatment phase, they would apply to an IPUHS hospital for rehabilitation.

On admission, each patient was examined for neurological symptoms by a physician and neurologist. In addition, clinical psychologists, physiotherapists, speech therapists, occupational therapists and nurses conducted assessments for every patient. The diagnosis of stroke was based on history, clinical examination and computed tomography scans or magnetic resonance imaging. A stroke was defined as a sudden, focal neurological deficit or cognitive dysfunction persisting for more than 24 h.22 The semistructured Mini-International Neuropsychiatric Interview (MINI)23 was employed for diagnosis of PSD. All patients were required to undergo an interview for diagnosis of mood disorder based on the 4th edition of the Diagnostic and Statistical Manual24 criteria for ‘mood disorder due to stroke with major depressive disorder’ (major depression) or ‘minor depression’. Severity of depression was measured using the 21 items in HAM-D. Severity of cognitive dysfunction was measured using the MMSE. A Functional Independence Measure (FIM) was used for the assessment of level of activity in daily life.25 For FIM score, higher scores indicate greater independence. Interviews were routinely performed on the admission day, by the same psychiatrist. These evaluations were used as the baseline assessment. At the end of the rehabilitation program, that is, at the time of discharge, HAM-D, FIM and MMSE were assessed again.

The authors administered 30 mg/day milnacipran to 10 PSD patients (milnacipran group), when each patient was diagnosed with depression. The dosage was regulated weekly between 30 and 60 mg/day, according to clinical observations. The remaining eight PSD patients (control group) were administered no antidepressants. For these eight control patients, their physicians refused antidepressant therapy due to physical condition of the patient or patients themselves refused to take any antidepressants. Therefore, cases and controls for this study were not randomly assigned.

All patients had rehabilitation treatment. The rehabilitation programs for each patient were designed by a ‘rehabilitation treatment team’, including physicians, physiotherapists, occupational therapists and speech therapists. Individual rehabilitation treatment was performed 5 days a week.

Background variables were compared between the two groups using the Mann–Whitney U-test and the χ2 test. HAM-D and MMSE scores were analyzed by means of two-way repeated measures anova.


The mean age of all patients was 58.2 ± 9.5 years, with a median age of 58.5 and a range of 41–75. In total, 13 patients (72.2%) were male.

There were no significant differences in the milnacipran group (major depression = 3; minor depression = 7) or the control group (major depression = 3; minor depression = 5) in age, gender, handedness, years of education, interval of admission or discharge after onset of stroke, personal history of psychiatric disease, family history of psychiatric disease and FIM total scores (Table 1).

Table 1.  Demographic characteristics for serotonin norepinephrine reuptake inhibitor and control groups
(n = 10)
(n = 8)
  •  The χ2 test.

  •  The Mann–Whitney U-test.

  • FIM, Functional Independence Measure.

Diagnosis (n)
 Major depression 3 3NS
 Minor depression 7 5 
 Age (mean ± SD, years)56.0 ± 9.960.9 ± 8.7NS
 Male (n) 7 6NS
 Right-handed (n)10 7NS
 Education (mean ± SD, years)12.9 ± 2.110.3 ± 2.6NS
 Admission since stoke (mean ± SD, months) 2.5 ± 0.8 2.2 ± 0.6NS
 Discharge since stroke (mean ± SD, months) 6.0 ± 1.5 5.4 ± 0.8NS
 Family psychiatric history (n) 0 0NS
 Personal psychiatric history (n) 2 0NS
 FIM total score at admission (mean ± SD)82.1 ± 27.883.0 ± 16.2NS

Also, no significant differences were found between the groups in frequency of the following neurological symptoms: motor deficit, ataxia, agnosia, apraxia, aphasia, attention disturbance and sensory disturbance (Table 2).

Table 2.  Neurological symptoms for milnacipran and control groups
(n = 10)
(n = 8)
  • *

     The χ2 test.

  • Values are number.

Motor deficit108NS
Ataxia 21NS
Agnosia 10NS
Apraxia 30NS
Aphasia 30NS
Attention disturbance 40NS
Sensory disturbance 87NS

Stroke type, based on computed tomography or magnetic resonance imaging scan findings, for the two groups is shown in Table 3. No significant differences were observed in the rate of infarction or hemorrhage, for right or left hemisphere.

Table 3.  Stroke types for milnacipran and control groups
(n = 10)
(n = 8)
  1. Values are number.

  2. NS, not significant.

Stroke type
Lesion location
 Right hemisphere44NS
 Left hemisphere34NS

An anova analysis on HAM-D scores indicated no significant effect in either the milnacipran or control group (Fig. 1). A comparison between the two groups of cognitive impairment showed a significant time-by-group interaction (F = 5.35, P = 0.034) for the MMSE score (Fig. 2). There was greater improvement for the milnacipran group than the control group.

Figure 1.

Changes of Hamilton Depression Rating Scale scores in patients with post-stroke depression during the study. There were no significant changes among milnacipran (n = 10) and control (n = 8) groups. Error bar represent standard error of the means (SE).

Figure 2.

Changes of Mini-Mental State Examination scores in patients with post-stroke depression during the study. There was significant time-by-group interaction among milnacipran (n = 10) and control (n = 8) groups (F = 5.35, P = 0.034), and there was significant time effect among two groups (F = 8.69, P = 0.009). Error bar represent standard error of the means (SE).

None of the subjects dropped out or showed any adverse effect during medication.


This study is the first to investigate the efficiency of milnacipran treatment for cognitive dysfunction of PSD in a case-control study. In this study, milnacipran was significantly effective for improvement of cognitive impairment measured using the MMSE and was well tolerated, although improvement of depressive symptoms assessed by the HAM-D scores was not observed.

The results of this study suggest cognitive dysfunction in PSD patients is reversible and treatable with administration of milnacipran. This finding also supports the suggestion that the effects of milnacipran on mood and cognitive recovery may be independent. In PSD patients, depression does not appear to lead to dementia and cognitive dysfunction does not necessarily lead to depression.

Kimura et al.21 reported that milnacipran (30–75 mg/day) may be effective for depressive symptoms evaluated with HAM-D for PSD patients. However, this study failed to detect any improvement in mood for PSD patients using milnacipran. A possible explanation is that the milnacipran group consisted of more patients with minor PSD (7 of 10 PSD subjects). There were less minor PSD subjects (4 of 12 PSD patients) in the investigation by Kimura et al.21 compared to this study, therefore, this may account for the contrast in findings. Puech et al.18 have argued that milnacipran is justified for treating severe or resistant depression. Treatment response to milnacipran may be more marked in more severe depressive states, that is, major depression rather than minor depression. Another possible reason for this difference is differences in time since stroke. The mean time since onset of stroke in the study by Kimura et al.21 was 38 weeks. In this study, the mean time since stroke was about 2 months. If administration of milnacipran had been conducted in an earlier phase after stroke, mood recovery of PSD patients may have been observed. Therefore, there could be a temporal therapeutic window for antidepressant therapy for depressive states.

Interestingly, milnacipran showed an effect on improvement of cognitive dysfunction in PSD patients, however, the mechanism is unknown. In prior studies, milnacipran was effective at ameliorating impaired cognitive function in healthy volunteers aged over 65 years26 and depressed patients following traumatic brain injury.27 The pharmacological action of milnacipran, normalizing both serotonin and norepinephrine systems,28 may relate to the cognitive impairment in PSD patients, however, further studies are required to clarify this.

This study had some limitations. First, the patient sample number was relatively small. This was partly due to the inherent difficulty in recruiting depressed patients. Therefore, results from this study may not be representative of all stroke patients. Second, this study was not double-blind placebo-controlled due to the ethical considerations in rehabilitation. The control group in this investigation was selected according to the judgment of either patients or physicians. Therefore, this selection method may have biased the results of this study. Third, the period of investigation was relatively short, ranging from 3 to 6 months after stroke (range of 1–8 months). Astrom et al.3 reported that different factors, such as brain lesions and activity in daily life, affected the development of depression at different evaluation times after stroke in a long-term follow-up study. Therefore, subjects in this study may include etiologically different PSD patients from those of other studies. Finally, the MMSE has a dependence on verbal ability and different degrees of sensitivity of its various items,29 therefore, it also has limitations that may have affected the findings in this study. A more detailed range of neuropsychological testing would more likely have documented whether cognitive impairment occurred in all areas or only in limited areas of functioning.


Findings in this study suggest that the SNRI, milnacipran, may be effective in ameliorating cognitive dysfunction. However, severity of depression was not significantly improved by this compound. Further studies are required to determine the relationship between depression and cognitive impairment in PSD patients treated with milnacipran.


This work was supported in part by a Grant-in-Aid for the Encouragement of Young Scientists from Ibaraki Prefectural University of Health Science.