Depressed elderly patients commonly present to general practitioners and if recognized they are treated with antidepressants. In this article, Dr Alan Wade discusses his recently published six month study1 which compared mirtazapine and paroxetine in general practice with conclusions emphasizing the importance of faster relief of symptoms and improvement in sleep.

It has been estimated that as many as 15 per cent of all consultations in general practice are associated with depression, the figure for elderly patients is likely to be much higher2. And yet we treat relatively few depressed patients at all and even fewer adequately. Why should this be? Elderly patients (and their relatives) often accept the symptoms of depression as part of the natural ageing process Ð poor sleep, physical and mental slowing, forgetfulness and loss of interest in friends and family. They tend not to complain3. It is inevitable, therefore, that the GPÕs attention will be drawn to the physical complaints with depression easily overlooked. This has led to the conclusion that formal screening of elderly patients should be undertaken and indeed many practices have incorporated this in their annual over 75 screening.

The evidence demonstrating that there is an improved outcome for these patients is, however, currently lacking and outcomes will probably not improve until more rigorous treatment regimes are instituted4,5. The elderly need treatment at an adequate dose of antidepressant and although it is reasonable to start with lower doses, even for the newer antidepressants, it is important to achieve a dose which does not just improve the patientÕs symptoms but which genuinely achieves remission with no residual symptoms6.

The elderly improve more slowly than the young and this may take several months. Having achieved remission, maintenance treatment is important. The elderly are also more likely to relapse and the usual dictum of maintenance treatment for six months after an acute episode can usefully be extended to 12 months in the elderly7.

Treatment considerations
The current drugs of choice are specific serotonin reuptake inhibitors (SSRIs) most of which have been shown to be effective in elderly patients. These are a major improvement on the tricyclic antidepressant (TCA) but nevertheless have significant deficiencies such as slow onset of action, gastro-intestinal disturbance, the potential to cause a transitional increase in anxiety symptoms, sleep disturbance and, at least for some, interactions with other drugs. These problems all contribute to poor compliance and early discontinuation.

In light of this we were determined in our investigation published in International Clinical Psychopharmology1 to assess the value of mirtazapine, a noradrenergic and specific serotonergic antidepressant (NaSSA) compared to an SSRI (paroxetine) in a 24 weeks study of depressed patients in a primary care setting. Enrolled in the study were depressed patients over 18 and with no upper age limit so that the elderly were included. To our knowledge this is the first long term study comparing the efficacy and tolerability of two modern antidepressants in a typical population of primary care patients.

Patient population and doses
Some 197 patients were enrolled from primary care practices in the west of Scotland. Each patient was rigorously assessed both by the GP and a specialist nurse to ensure they met full criteria for major depression (Diagnostic and Statistical Manual of Mental Disorders - DSM IV) of at least of moderate severity. This was defined as a Hamilton Depression Rating Scale (HAM-D) of at least 18. No concomitant psychoactive drug treatment was permitted immediately prior to or during the study. The demographics of the population are typical of those seen in general practice and are outlined in Table 1.

Initial doses were in line with standard recommendations Ð mirtazapine 30mg or paroxetine 20mg. From week 5 the dose could be increased to 45mg and 30mg respectively at the discretion of the physician. For the Intent to Treat (ITT) population, the mean daily doses (SD) were 34.6 (5.70)mg/day of mirtazapine and 23.9 (3.96)mg/day of paroxetine. Similar numbers of patients in both treatment groups required dosage increase during the study (mirtazapine, n=39; paroxetine, n=42).

Adverse events and efficacy
As might be expected in this population, the drop-out rate was significant, particularly during the first eight weeks when 57 per cent of mirtazapine dropouts and 64 per cent of paroxetine dropouts occurred. From Table 2 it can be seen that lack of efficacy was not a prominent reason for discontinuation but that over 20 per cent in both groups discontinued due to adverse events. The most common adverse events are listed in Table 3.

The data demonstrated that both drugs were effective in improving overall depressive symptomatology and producing high percentages of responders and remitters. At all assessments except week 8 the magnitude of reduction in HAM-D was greater for mirtazapine than paroxetine and this reached statistical significance during the crucial early weeks 1, 2, and 4. But sometimes statistical significance means little in clinical practice. The difference in favour of mirtazapine, however, was clinically relevant at weeks 2 and 4 as defined by accepted standards (Figure 1)8. We are all aware that minor improvements in depression symptoms are not enough. In this study we defined ÔrespondersÕ to the drug as demonstrating at least a 50 per cent improvement in HAM-D score and ÔremittersÕ as patients with a HAM-D score of 8 or less. At the end of the study (24 wks) 87.0 per cent of mirtazapine treated completers and 77.8 per cent of paroxetine treated completers had responded and 60.9 per cent and 42.2 per cent respectively were in remission. By these criteria almost two out of every three patients taking mirtazapine at 24 weeks was depression free. During the 24-week study period, only two patients in each treatment group suffered a relapse.

Sleep disturbance
Sleep disturbance is a cardinal symptom of depression and frequently the symptom causing most distress to the patient. Analysis of the sleep related questions of the HAM-D showed a statistically significant difference, in favour of mirtazapine, as early as the first week. Improvements in sleep parameters favouring mirtazapine were also noted on a specific sleep questionnaire where mirtazapine-treated patients reported less awakening, an increase in total number of hours of sleep, and improved quality of sleep. Tolerability If we are to treat more patients with depression adequately it is important that not only are the drugs effective but that they are well tolerated, particularly over the long term. Both drugs were well tolerated with similar numbers of patients discontinuing due to adverse events. Statistically significantly more mirtazapine Ð (n=20/20 per cent) than paroxetine-treated patients (n=7/7 per cent) reported fatigue (P=0.012). In the paroxetine group, statistically significantly more patients reported increased sweating (n=11/11 per cent; mirtazapine, n=2/2 per cent; P=0.010), headache (n=28/29 per cent; mirtazapine n=13/13 per cent, P=0.008) and nausea (n=31/31per cent; mirtazapine n=8/8 per cent; P<0.001). Weight gain was noted in both groups but the pattern was different. In the mirtazapine group, this occurred in the first twelve weeks of treatment (3.0kg) and plateaued or slightly fell thereafter. In contrast the paroxetine group had negligible increase initially but thereafter had a steady increase to six months (2.2kg). A modest improvement in sexual function was noted in both groups over the trial period.

Discussion
It is important that drugs that are used in primary care are also assessed in primary care and trial design should reflect normal clinical practice without excluding important groups such as the elderly. This is the first study to compare two antidepressants in primary care over a six month period. The mean doses employed (34.6mg/day for mirtazapine and 23.9mg/day for paroxetine) fairly reflect both other clinical trials and routine clinical practice.

Compliance is the main challenge facing general practitioners treating patients with depression8. Depressed patients stop taking medication for many reasons but early in the treatment phase, lack of obvious improvement in symptoms can be very disheartening. Although not specifically designed to assess speed of onset of antidepressant action, differences in depressive symptoms favouring mirtazapine were observed between weeks 1 and 4. Not only were these differences statistically significant, they were also clinically meaningful by currently accepted standards9. In particular, sleep, one of the cardinal symptoms of depression improved significantly during the first week on 30mg mirtazapine and this improvement was maintained subsequently. It has been suggested that the early improvement in depression and the beneficial sleep effects seen in the mirtazapine group are merely due to the effect of sedation. They are, however, in accordance with patterns seen in formal sleep studies which demonstrate improvement in the typical sleep disturbances associated with depression accompanied by subjective improvements observed using the Leeds Sleep Questionnaire10. This effect, has been shown, therefore, to be not merely due to sedation but to an antidepressant effect on core depression symptomatology. It is likely that these benefits will be reflected in practice with improved early adherence to treatment.

In the longer term (beyond eight weeks) the improvement in both groups was relatively modest Ð 12.7 per cent of initial HAM-D in the mirtazapine group and 9.0 per cent in the paroxetine. Nevertheless a considerable increase in both responders and remitters was recorded indicating substantial additional benefits to patients.

The side effects noted were in line with the known pharmacology of each agent with significantly more fatigue in the mirtazapine group and more headache, sweating and nausea in patients receiving paroxetine. Weight gain was experienced by both groups but the pattern was strikingly different. Mirtazapine patients experienced weight gain during the early period of the study but plateaued after 12 weeks. By contrast the paroxetine group had minimal weight gain by twelve weeks but increased sharply at that point and steadily thereafter. It would be important to advise patients about this aspect of their treatment as part of their routine clinical management.

The results of this UK study echo the conclusions of a recent US trial which found mirtazapine to alleviate depressive symptoms more quickly in elderly patients than paroxetine. During the study, 255 depressed patients (aged 65 or older) randomly received either mirtazapine or paroxetine. The drugs were equally effective after eight weeks - however, depressive HAM-D-17 symptom scores on day 14 were better in the mirtazapine arm. Median times to a 50 per cent reduction in symptom scores were 26 and 40 days in patients taking mirtazapine and paroxetine respectively. Anxiety scores also declined with mirtazapine and patients reported better sleep quality, especially early on in treatment11.

Conclusion
Both mirtazapine and paroxetine were shown to be effective and well-tolerated antidepressant treatments when used in a population of general practice patients, including the elderly. Patients treated with mirtazapine were seen to have greater improvement from weeks 1 to 4 together with improved sleep which was maintained throughout the study. These benefits may be particularly relevant in primary care. In the next ten years almost 25 per cent of the population of the UK will be over 60 of whom many will suffer from depression. Only by improving our management and treatment of depression will they enjoy maximum quality of life.

References
1.Wade A, Gordon M, Crawford M et al. A randomized, double-blind, 24 week study comparing efficacy and tolerability of mirtazapine and paroxetine depressed patients in primary care. International Clinical Psychopharmacology 2003; 18: 133Ð41

2. Banazak DA, Wills C, Collins C. Late-life depression in primary care: where do we go from here? J Am Osteopath Assoc 1998; 98(9):489Ð97

3. OÕConnor DW, Rosewarne R, Bruce A. Depression in primary care. 1: elderly patients' disclosure of depressive symptoms to their doctors. Int Psychogeriatr 2001;13(3):359Ð65

4. Whooley MA, Stone B, Soghikian K. Randomized trial of case-finding for depression in elderly primary care patients. J Gen Intern Med 2000;15(5): 293Ð300

5. Hustey FM, Meldon SW, Palmer RM. The effect of a short depression screen on the care of older emergency department patients. Acad Emerg Med 2003; 10(5):494.

6. Lyness JM, King DA, Cox C et al. The importance of subsyndromal depression in older primary care patients: prevalence and associated functional disability. J Am Geriatr Soc 1999; 47(6):647Ð52

7. Baldwin RC. Poor prognosis of depression in elderly people: causes and actions. Ann Med 2000; 32(4): 252Ð6

8. Keller MB, Hirschfeld RM, Demyttenaere K et al. Optimizing outcomes in depression: focus on antidepressant compliance. Int Clin Psychopharmacol 2002; 17(6): 265Ð71

9. Thase ME. Methodology to measure onset of action. J Clin Psychiatry 2001; 62(Suppl 15):18Ð21

10.Schittecatte M, Dumont F, Machowski R et al. Effects of mirtazapine on sleep polygraphic variables in major depression (2002). Unpublished data. Poster presented at the 23rd Collegium Internationale Neuro-Psychopharmacologium (CINP) congress, June 23-27, Montreal, Canada.

11.Schatzberg AF et al. Mirtazapine versus Paroxetine Study Group (2002) Mirtazapine v Paroxetine in elderly patients. Am J Geriatr Psychiatry 10: 541Ð50

Alan G. Wade is Medical Director at CPS Research, Clydebank

return to main menu