The new antidepressants

J.W.G. Tiller, Associate Professor and Reader, Department of Psychiatry, University of Melbourne, Royal Melbourne Hospital, Melbourne

Summary

The new antidepressants are moclobemide, fluoxetine, paroxetine and sertraline. They are as effective as the established antidepressants, have fewer adverse effects and are safer in overdose. The new drugs are more likely to be taken in an effective dosage for an adequate period of time. They provide therapeutic options for prescribers seeking an effective tolerated drug for an individual patient. The higher tablet cost of the newer drugs may be offset by their greater tolerability, leading to more effective treatment of depression, and their greater safety. At this time, the new drugs are increasingly being regarded as the most appropriate initial therapy for depression, with the older drugs, such as tricyclic antidepressants, being regarded as secondline alternatives.

Key words: moclobemide, fluoxetine, paroxetine, sertraline

(Aust Prescr 1995;18:92-6)

Why new antidepressants?
Antidepressant pharmacotherapy complements nonpharmacological treatments. Most patients are treated with tricyclic antidepressants (TCAs), but some are treated with the tetracyclic, mianserin, and a few with nonselective irreversible monoamine oxidase inhibitors (MAOIs).

These antidepressants have substantial adverse effects which often limit the dosage and duration of treatment. The TCAs have anticholinergic, antihistaminic and alpha adrenergic blocking effects, while MAOIs have a potentially lethal interaction with dietary amines and certain drugs. Patients may take inadequate doses and prematurely discontinue treatment because of these adverse effects. This results in ineffective treatment and enhances the risk of relapse.

What are the new antidepressants?
Four new antidepressants have been recently marketed in Australia. They are the selective reversible monoamine oxidase A inhibitor, moclobemide, and the selective serotonin reuptake inhibitors (SSRIs), fluoxetine, paroxetine and sertraline. These are not really new medicines; moclobemide has been studied in Australia since 1983, while fluoxetine will shortly be out of patent. Over a million patients have taken these drugs, so it is highly unlikely that any important new adverse effects will emerge. One other antidepressant likely to be marketed in coming months is the phenethyl amine, venlafaxine, which inhibits the reuptake of noradrenaline and serotonin.

What do the new antidepressants do?
By selectively inhibiting the metabolism of noradrenaline and serotonin, moclobemide increases their concentrations in the brain. This is believed to enhance amine transmission and improve depression. Moclobemide could be seen as a functional derivative of the old MAOIs, although it is structurally different and does not have the significant food or drug interactions of the MAOIs.

The SSRIs could be seen as functional derivatives of the TCAs, although they have different structures. Unlike the TCAs which have a common chemical nucleus, the SSRIs are chemically different to one another. Their common function is to inhibit reuptake of neurotransmitters, especially
serotonin, into the presynaptic neuron. This is believed to increase serotonin concentrations in the synaptic cleft, which enhances amine transmission and improves depression.

The newer antidepressants have less effect on muscarinic (anticholinergic), histaminic, alpha adrenergic and dopaminergic receptors than the TCAs. Consequently, they have fewer adverse effects.

Efficacy and effectiveness
No antidepressant is consistently more efficacious than any other. The newer antidepressants are no more efficacious than the TCAs, MAOIs or mianserin. When first prescribed, each drug helps about 60-70% of patients. However, the overall effectiveness of the newer drugs is greater because they are better tolerated. Patients are therefore more likely to take an adequate therapeutic dose for a longer time and so reduce the risk of relapse. They are also less likely to have toxic or lethal adverse events if they overdose.

Adverse effects and toxicity
Approximately half of the patients taking the new drugs will have some adverse effects. When adverse effects do occur, these are usually mild and of short duration, diminishing and usually stopping with continued treatment. The common adverse effects of the SSRIs and moclobemide are headache, nausea, insomnia and dizziness. The reported rates for moclobemide are marginally less than for the SSRIs. The SSRIs can also cause diarrhoea and there are occasional reports of dyskinesia or akathisia, although tardive dyskinesia has not been reported. While it is difficult to compare adverse events between studies, clinically it appears that nervousness, agitation and restlessness are more commonly reported with fluoxetine (15%) than with the other new antidepressants (approximately 5%).

With the shorteracting SSRIs, there are occasional reports of increased anxiety, agitation and insomnia at the end of therapy. These transient symptoms may not occur if the dose is slowly reduced.

If adverse effects are persistent and disabling, they may be diminished by a temporary dosage reduction. Often the adverse effects do not recur if the dose is subsequently increased.

The TCAs and SSRIs have been associated with sexual impairments both in isolated case reports and studies, whereas moclobemide has not. They are less likely to be reported in shortterm studies as depression impairs sexual functioning, and continuing or worsening impairment may not be identified. These adverse effects on arousal and performance occur in women and men and can lead to them stopping longterm treatment because of the effect on their quality of life. Sometimes a lower dose will control the depression with less effect on sexual function. Sexual function may be restored within a few days of stopping a drug with a short halflife (e.g. paroxetine).

Suicidal ideation and behaviour are features of depression and may be present at the onset of treatment or emerge during treatment. There are no scientific data to suggest that patients on the newer antidepressants are more likely to have suicidal thoughts or commit suicide than those on other drugs, or on no pharmacotherapy.

Must an antidepressant be sedative?
Agitation, anxiety and insomnia are common in depression. Prescribing a sedative TCA or tetracyclic to provide some immediate relief is tempting. While this can be effective, there are the problems of continuing sedative effects and the risk of potentiating the effects of alcohol and other sedatives. The development of tolerance diminishes sedation, but many patients remain adversely affected and compromised in driving cars and other activities that require unimpaired psychomotor performance. A further problem arises when discontinuing sedative antidepressants. Increased dreaming, insomnia, daytime agitation and anxiety may occur, particularly if the dose is rapidly reduced.

The newer drugs are generally neither sedating nor stimulating; however, an occasional patient may experience these effects. Treatment may need to be temporarily or permanently reduced or stopped. Sedative effects can be managed by taking the drug at night.

The nonsedating antidepressants all reduce anxiety, agitation and insomnia as depression resolves.

Interactions
Moclobemide has minimal effects on the cytochrome P450 2D6 system (CYP2D6), although it is metabolised by that system. The only clinical interaction of consequence is with cimetidine when the oral dose of moclobemide should be halved.

With the SSRIs, the effects on the CYP2D6 system are more marked with paroxetine and fluoxetine than with sertraline. This interaction is usually of little consequence. Table 1 lists some of the drugs whose metabolism is affected by the microsomal cytochrome P450 system.

Table 1 Some drugs, metabolised by the cytochrome P450 2D6 hepatic microsomes, which may have clinically significant interactions with SSRI antidepressants tricyclic antidepressants prednisone metoprolol propranolol theophylline felodipine warfarin

The major drug interaction of consequence with SSRIs is with TCAs. The coadministration of SSRIs and TCAs can give rise to tricyclic toxicity, or a serotonin syndrome; both can be fatal. TCAs and SSRIs should not be coprescribed. If a hypnotic is required in the short term (2-4 weeks), a benzodiazepine or zopiclone is suitable.

There is an interaction between sertraline and tolbutamide which has not been reported with the other SSRIs. Tolbutamide concentrations can increase by about 16% because sertraline may impair the metabolism of tolbutamide. Clinically, the consequences of any such effect should be detected with routine monitoring.

The new antidepressants do not potentiate the psychomotor impairments of alcohol.

Protein binding
The SSRIs are highly bound to plasma proteins. Although caution is advised when coprescribing other highly protein bound drugs, interactions due to protein binding may not be clinically significant.

Safety in overdosage
The new antidepressants are very much safer in overdose than the TCAs. Prescriptions for TCAs usually provide patients with enough medication to kill themselves, whereas prescriptions for the newer antidepressants do not. However, the newer antidepressants may be lethal if taken in overdose with other drugs.

Differences between the new antidepressants
Moclobemide has a short halflife of about 4 hours, but a duration of action between 12 and 18 hours. While twice a day dosage is common, once a day dosage can be as effective. The disadvantage of a single daily dose is a slight increase in complaints of dizziness. For patients on a normal diet, there is no need for specific dietary proscriptions or a low tyramine diet. Drug interactions are relatively few, although caution should be taken with cimetidine, pethidine and large amounts of orally ingested sympathomimetics.

With the SSRIs, the most potent inhibitor of 5HT uptake is paroxetine, followed by fluoxetine and sertraline, whereas, the most selective inhibitor of 5HT uptake is paroxetine, followed by sertraline and fluoxetine. In practice, the potency and selectivity differences do not appear to alter clinical effectiveness.

Pharmacokinetics
In the normal therapeutic range, the kinetics of moclobemide and sertraline are linear. Paroxetine and fluoxetine have nonlinear kinetics which means that their plasma levels may increase more rapidly at higher doses than with the same milligram increment at lower doses. Potentially, there may be more adverse events at higher doses. This accounts for paroxetine's recommended initial dose of 20 mg and subsequent increments of only 10 mg at a time. For paroxetine, as for fluoxetine, the 20 mg dose suffices for most patients. However, in clinical practice, there is little consequence whether the kinetics are linear or not.

Changing treatment (Table 2)
One important difference between the antidepressants is their halflives. One third of patients do not respond to the first antidepressant prescribed and need to be changed to another drug. The second antidepressant should be commenced as soon as possible, ideally without a long wait for washout to avoid drug interactions. Washout periods of two weeks are required after high doses of paroxetine and sertraline, or 5 weeks after fluoxetine. In practice, moclobemide has the shortest halflife and any other antidepressant can be commenced the day after the last dose of moclobemide with minimal risk of an interaction.

There is little likelihood of adverse interactions if another antidepressant is commenced the day after a low or moderate dose of a SSRI is stopped. The one exception is commencing one of the irreversible MAOIs.

Irreversible MAOIs should not be taken until 5 SSRI halflives have elapsed after stopping the earlier agent. The halflives of paroxetine and sertraline are both approximately 24 hours. However, there are interindividual differences, with the halflives in some patients being up to approximately 3 days. Sertraline also has a major metabolite, N-desmethylsertraline, with an activity of about 12-25% of the parent compound and a halflife of 66 hours. Although the halflife of fluoxetine is 48-72 hours, its equally active metabolite, norfluoxetine, has a halflife of approximately 6-9 days. Two weeks must be allowed after paroxetine or sertraline before an irreversible MAOI is used, and 5 weeks allowed after fluoxetine. The new medicines must be commenced with caution because of interindividual differences, even after these washout periods.

There is no interaction between moclobemide and MAOIs, although in changing to moclobemide, drug and dietary restrictions should continue for two weeks after stopping the old MAOI.

There is generally no adverse effect when directly changing to moclobemide from lower doses of SSRIs, although caution should be exercised in changing from high doses of a SSRI because of the long washout period.

After high doses of TCAs, it is advisable to wait up to a week before commencing a SSRI because of the risk of adverse interactions. This also applies when high doses of clomipramine are replaced by moclobemide.

Do these differences between antidepressants really matter?
The differences in half-life do matter when changing medications and in washout after an adverse event. Inpractice, most of the other differences between these drugs are of limited consequence. There are interindividual differences so that one patient will tolerate one medicine well, while another patient will tolerate a second medicine better. Despite the choice of drugs available, some patients will still fail to respond for long periods of time to treatments.

Costs
The efficacy of the new antidepressants is, in general, no different from that of the old antidepressants. The greater effectiveness of the newer drugs results from better tolerability and safety. However, the tablets are 10-20 times more expensive than most of the older drugs.

Patients are more likely to tolerate the newer drugs in effective doses, rather than the subtherapeutic doses sometimes seen with the older antidepressants. An effective dose of a new antidepressant is more likely to be taken for a minimum of 6 months and this reduces the risk of relapse. If there are fewer adverse effects, fewer medical consultations are needed. Patients are also less likely to be admitted to hospital to establish pharmacotherapy, or for 'treatment resistance'.

The issue is the total cost of treating an episode of depression, not the cost of a tablet. The most sensitive variable is the cost of treatment failure. Medication that appears expensive in terms of cost per day may not be so when patient compliance and total costs of treatment are taken into account. 1

A cheaper option is not to prescribe antidepressants at all.

As Henry2 states, 'At present, the worst outcome of depressive illness suicide absolves carers from further costs. While this reduces the Government's cost burden, it could hardly be seen as a desired outcome.'

Which antidepressant?
The arguments in favour of TCAs, that they have stood the test of time or the (generally) cheaper per tablet cost, cannot be sustained. The new antidepressants have generally been better researched and have now been widely used.

Table 2 Changing from one antidepressant to another
changing to changing from	tricyclic antidepressant	mianserin	fluoxetine	paroxetine	sertraline	mociobemide	irreversible MAOI
tricyclic antidepressant	no antidepressant-free interval required	no antidepressant-free interval required	1-week antidepressant-free interval required*	1-week antidepressant-free interval required*	1-week antidepressant-free interval required*	Note a (1-week antidepressant-free interval required for clomipramine)	1-week antidepressant-free interval required
mianserin	no antidepressant-free interval required		no antidepressant-free interval required	no antidepressant-free interval required	no antidepressant-free interval required	Note a	1-week antidepressant-free interval required d
fluoxetine	commence at low dosage b	commence at low dosage		commence at low dosage	commence at low dosage	Note c
paroxetine	commence at low dosage b	commence at low dosage	commence at low dosage		commence at low dosage	Note c
sertraline	commence at low dosage b	commence at low dosage	commence at low dosage	commence at low dosage		Note c
moclobemide	1-day antidepressant-free interval required a	1-day antidepressant-free interval required a	1-day antidepressant-free interval required a	1-day antidepressant-free interval required a	1-day antidepressant-free interval required a		1-day antidepressant-free interval required a
irreversible MAOI	2-week antidepressant-free interval required	2-week antidepressant-free interval required	2-week antidepressant-free interval required	2-week antidepressant-free interval required	2-week antidepressant-free interval required	Note*
* This includes period of tapering of the tricyclic antidepressant dose. a There is insufficient evidence to support a definite recommendation. However, the manufacturers of moclobemide are prepared to endorse no antidepressant-free period if moderate doses of both drugs are involved. b Note that tricyclic antidepressant levels may be elevated for at least several weeks due to persisting SSRI-induced cytochrome P450 inhibition. c There is insufficient evidence to support a definite recommendation. However, the manufacturers of moclobemide are prepared to endorse no antidepressant-free period if moclobemide is commenced at 150 mg daily. After 2 days, moclobemide can be increased to 300 mg daily and subsequently higher if necessary. This recommendation is only for changing from low or moderate dosses of SSRIs. High doses of SSRIs should be gradually reduced before ceasing, to commence moclobemide. d Five weeks is necessary because of the long half-life of fluoxetine and its active metabolite. e There is insufficient evidence to support a definite recommendation. However, the manufacturers of moclobemide are prepared to endorse no antidepressant-free period if moderate doses of both drugs are involved, but irreversible MAOI dietary restrictions should be continued for 2 weeks.
This table has been reprinted with permission from the third edition of the Psychotropic Drug Guidelines

Fig 1 Changing from one antidepressant to another Antidepressants, community use Note: Data supplied by the Drug Utilization Sub-Committee of the Pharmaceutical Benefits Advisory Committee.

In terms of efficacy, there is nothing to choose between the newer antidepressants, or between newer and older drugs. What matters for the individual patient is whether the medicine works for them and is tolerated.

When selecting an antidepressant, it is important to consider how easy it will be to change to another drug if the treatment does not work. There may be other considerations. For example, patients with obsessivecompulsive disorder are helped by SSRIs, but, to date, there is no published evidence supporting the use of moclobemide in this disorder.

Adopting the principle of primum non nocere , it seems preferable to prescribe a drug that is less likely to cause death in overdosage than others which, generally, are no more effective, have more adverse effects and can be lethal in overdosage.

Summary
The new antidepressants are useful advances in treatment. Their safety and effects on the quality of life may justify their higher cost and their use is likely to continue to increase (Fig. 1).

R E F E R E N C E S
1. Jonsson B, Bebbington PE. What price depression? The cost of depression and the costeffectiveness of pharmacological treatment. Br J Psychiatry 1994;164:665-73.

2. Henry JA. Debits and credits in the management of depression. Br J Psychiatry 1993;163(Suppl):S33-S39.