Tollefson and co-workers (1998) reported that fluoxetine (Prozac) was significantly (p < 0.05) more effective than placebo in treating both anxious and non-anxious depression (N = 3183). Fluoxetine was also significantly more effective than placebo in reducing the Hamilton Depression and Anxiety (HAMD) scales anxiety/somatization factor score. The efficacy of fluoxetine (Prozac) and tricyclic antidepressants (TCAs) was comparable on all measures and in all groups. Furthermore, fluoxetine-treated patients were less likely to drop out of treatment due to anxiety than those given a TCA. This was true whether or not the patient was considered “anxious” at baseline.

A similar analysis examined the effect of fluoxetine (Prozac) on psychomotor agitation. These data came from double-blind trials involving 4737 patients with major depression randomly assigned to fluoxetine (Prozac), a comparison antidepressant or placebo. Item 9 of the Hamilton Depression Rating Scale was used to assess psychomotor agitation. Agitation occurred at a similar incidence across treatment groups. The rate of increased agitation (after starting treatment) was comparable between fluoxetine (Prozac), placebo and tricyclic antidepressants. Improvement in agitation occurred significantly more often among fluoxetine-treated than placebo-treated patients.

The previous two studies primarily concerned patients treated with average doses of fluoxetine (Prozac) (20 mg/day). Higher doses are often used for non-responsive or partially responsive patients. There is also evidence that high-dose fluoxetine (Prozac) is also associated with anxiety reduction. In the earliest clinical trials, fluoxetine (Prozac) was typically prescribed at a dose of 60-80 mg/day. Beasley and colleagues reviewed data from 706 outpatients with DSM-III major depression treated with either high-dose fluoxetine (Prozac) (median 80 mg/day) or imipramine (Tofranil) (median 200 mg/day). Imipramine and fluoxetine (Prozac) were comparable in overall antidepressant effect as well as in reduction of sleep disturbance and scores on the anxiety/somatization factor. Reduction in sleep disturbance and anxiety was again independent of baseline psychomotor agitation or retardation. Baseline psychomotor state predicted activation-sedation side effects only for imipramine (Tofranil), which caused more sedation in patients with baseline psychomotor retardation. There was a non-significant trend (p = 0.92) for more drug discontinuations due to activation for fluoxetine (Prozac) than imipramine (Tofranil). Such a small difference, however, is unlikely to be clinically significant given the very large sample size.

Prospective data support this pattern of findings. Three hundred and thirty-six primary care patients with major depression were randomly assigned to fluoxetine (Prozac) or imipramine (Tofranil) in one study. Improvement was essentially identical in the two groups. Neither baseline anxiety nor insomnia scores predicted group differences. Patients assigned to fluoxetine (Prozac) were again significantly less likely to change or discontinue medication.

These findings contrast with those reported by Fava and colleagues (1997). These investigators studied 294 outpatients with major depressive disorder who were treated with fluoxetine (Prozac) 20 mg/day for 8 weeks. They found that non-anxious patients improved slightly but significantly more during treatment than anxious depressives on all outcome measures. However, this study was not strictly comparable to the others reviewed here, since anxious depression was defined as meeting full criteria for a comorbid anxiety disorder. As a general rule the presence of one or more comorbid disorders is associated with poorer outcome across a variety of psychiatric conditions.

The large epidemiological studies that have been carried out in recent years have drawn attention to the high rates of comorbidity of psychiatric disorders in the general community. Major depression is known to have a high comorbidity with separate anxiety disorders and a problem can arise in attributing certain anxiety symptoms as part of either the depression or the anxiety disorder. The overlap of major depression and an anxiety disorder where both conditions satisfy the full diagnostic criteria is perceived as comorbidity. In this case separate diagnoses may be considered. The overlap of major depression and anxiety disorders, where neither disorder satisfies the full criteria, has come to be known as mixed anxiety depression (MAD). This chapter attempts to address the degree of overlap which is so confusing for the clinician and to assess the data relating the role of selective serotonin reuptake inhibitors (SSRIs) in treatment.

Anxious depression or comorbid anxiety and depression

Mixed anxiety depression is a common but poorly defined condition with multiple possible aetiologies. Both anxiety and depression may occur as a symptom of or reaction to a primary psychiatric or medical disorder. The concept of major depression includes a variety of subgroups, including the more severe (psychotic features, melancholia) and chronic subtypes. Anxiety disorders are classified according to whether and how the anxiety is limited to particular situations (phobias, compulsions), thoughts (obsessions) or times (panic attacks). Generalized anxiety disorder (GAD) may be thought of as chronic anxiety which is not limited to any of these dimensions. Some basic researchers believe that anxiety and depression exist on a continuum and that depression may represent under-activity of serotonergic pathways while anxiety results from over-activity in serotonergic neurones.

There are two general meanings for mixed anxiety depression. One is the depressed patient who has signs or symptoms of an anxiety disorder which does not meet the threshold for a separate diagnosis, such as the patient with panic attacks which do not occur often enough or with the requisite number of symptoms to be diagnosed as panic disorder. The other is the patient with generalized anxiety disorder who intermittently fulfils criteria for major depression. Mixed anxiety depression also appears as an experimental diagnosis in DSM-IV (Table: Research criteria for mixed anxiety depressive disorder (DSM-IV)). Its anxiety symptom criteria are very similar, and in some cases identical, to those for GAD. However, DSM-IV requires that the disorder does not meet, and never has met, criteria for generalized anxiety disorder, major depression, dysthymic disorder and/or panic disorder. This is a problem, especially because the symptom threshold for dysthymic disorder is so low (requiring only two symptoms) that chronically depressed and anxious patients will almost never meet criteria for mixed anxiety depression. One might summarize this by saying that mixed anxiety depression (MAD), as it is currently regarded, is an admixture of a subsyndromal depressive disorder with a subsyndromal anxiety disorder. In a sense it is a testament to the importance of MAD that it has received as much research attention as it has despite this lack of satisfactory diagnostic criteria.

Table: Research criteria for mixed anxiety depressive disorder (DSM-IV)

Anxiety symptoms in depression

The scales employed for rating the severity of depression reflect the nature of these anxiety symptoms as integral to the depression. The scores on these anxiety items are seen to reduce as the depression improves and the items have good face validity in depression. For example, the Hamilton Scale for Depression (Hamilton, 1960) includes items to measure agitation, somatic anxiety and psychic anxiety. Three items that measure sleep disturbance and an item to assess the phenomenon of depersonalization feelings also form part of the scale. Moreover, it can be argued that the items for assessing obsessional symptoms and hypochondriasis also register anxiety symptoms. Most studies of the efficacy of various treatments for depression have shown that the symptoms measured by these anxiety items in the scales improve at the same rate as other symptoms of depression.

Some of the items in the Hamilton Rating Scale are more sensitive to treatment change than others and Hamilton himself recognized that the depersonalization and obsessional items, which are less sensitive, should be used more appropriately for diagnostic purposes only.

SSRI in the Treatment of Anxiety Symptoms Within Depression

Some treatments are acknowledged to be more effective than others in treating particular anxiety symptoms that occur with depression. The items that measure disturbed sleep improve more rapidly in response to sedative antidepressants with marked histaminergic receptor properties such as the sedative tricyclic antidepressants or mianserin. This has been shown in a number of comparisons with non-sedative antidepressants such as the selective serotonin reuptake inhibitors. However, the advantage seen with the sedative antidepressants, which is more evident at the start of treatment, tends to diminish as the sleep improves as part of the general improvement of the depression in response to SSRIs. By the end of the acute treatment period, the advantage of the sedative antidepressants on sleep is no longer evident. On the other hand, the negative effects on psychomotor function of the histaminergic activity become apparent with daytime drowsiness and the need to desist from driving cars or operating heavy machinery.

The early effect of anxiolytic drugs on disturbed sleep and certain other anxiety, symptoms in depression has been reported but an antidepressant effect cannot be attributed to these drugs merely on these grounds. The studies of benzodiazepines in the treatment of depression show their effects on improving the sleep items and anxiety but also show that they are less effective, or not effective at all, in improving other features of depression, and it is for this reason that, independently of the associated long-term problems of tolerance and dependence, benzodiazepines are not licensed or recommended for the treatment of major depression. Some studies have reported an increase in paradoxical aggression with benzodiazepines, possibly mediated by a mechanism involving disinhibition, and this characteristic makes these drugs unsuitable for depression, where there is an elevated risk of suicide attempts.

The selective advantages of selective serotonin reuptake inhibitors (SSRIs) in treating the anxiety symptoms of depression compared with sedative tricyclic antidepressants came as a surprise. Zimelidime, an early non-sedative SSRI that was withdrawn from the market, was found to be more effective than amitriptyline in treating the anxiety symptoms of depression in a 6-week treatment study. This finding raised the possibility that serotonin reuptake might have a special beneficial effect on the symptoms of anxiety within depression. Subsequent analysis for paroxetine (Paxil) and fluvoxamine (Luvox) have confirmed this hypothesis, with a differential advantage in treating anxiety symptoms in depression reported for the SSRIs compared with reference tricyclic antidepressants. These observations led to the hypothesis that SSRIs might have particular advantages in treating separate anxiety disorders, such as panic disorder and social phobia.

The first generation of antidepressant medication included tricyclics (TCAs) and monoamine oxidase inhibitors. A number of investigators have shown these drugs to be helpful in the treatment of anxious depression. Several trials have also shown TCAs to be equivalent or even superior to benzodiazepines in the treatment of this syndrome. SSRIs improved over tricyclic antidepressants and monoamine oxidase inhibitors (MAOIs) in two main areas: tolerability and safety. However, anxiety and agitation have been reported as occasional side effects with all selective serotonin reuptake inhibitors, which might lead to reluctance to use these agents in anxious depression. A review of the available data concerning the use of SSRIs in anxious depression, especially relating to whether baseline anxiety is a poor prognostic sign, is therefore timely.

The first line of evidence is indirect. It stems from the observation that patients with anxiety disorders often have concomitant depressive symptoms. There is considerable direct evidence of the efficacy of the SSRIs in anxiety disorders. All four SSRIs approved for marketing in the USA have shown sufficient efficacy to garner an official indication for at least one anxiety disorder, obsessive-compulsive disorder (OCD). Paroxetine (Paxil) and sertraline (Zoloft) are also indicated for the treatment of panic disorder (PD). While patients with bona-fide major depression are usually excluded from trials of an antidepressant in an anxiety disorder, it is likely that a substantial number of the subjects in OCD and PD trials had clinically significant depressive symptoms.

Filteau and colleagues analysed data from 10 studies of SSRIs (sertraline (Zoloft); zimelidine; fluvoxamine (Luvox); fluoxetine (Prozac)); selective noradrenergic uptake inhibitors — desipramine, maprotiline, oxaprotiline; mixed uptake inhibitors — amitriptyline, imipramine (Tofranil); and partial 5-HT2 antagonists — ritanserin, trazodone (Desyrel), nefazodone. The data showed no differences between the efficacy of these classes in agitated or retarded depression. In a subsequent analysis the same investigators found that SSRI responders tended to be initially more anxious and agitated than non-responders.

Prevention of the reuptake of monoamine transmitters such as serotonin, which potentiates their action in the brain, appears to be associated with antidepressant activity. SSRIs such as fluoxetine preferentially inhibit the reuptake of serotonin compared with noradrenaline, and have limited direct action at other neurotrans-mitter sites, including muscarinic receptors. They therefore cause fewer antimuscarinic or sedative adverse effects than the tricyclic antidepressants and are less cardiotoxic. Citalopram is the most selective of the SSRIs currently available, whereas paroxetine is the most potent.

SSRIs provide an alternative to the tricyclics for the treatment of depression. As with the tricyclics, it may be several weeks before an antidepressant effect is seen. Once depression has then resolved, maintenance therapy should be continued for at least 4 to 6 months (12 months in the elderly) to avoid relapse on stopping therapy. Patients with a history of recurrent depression should continue to receive maintenance treatment for at least 5 years and possibly indefinitely.

Some SSRIs are also used as part of the management of generalised anxiety disorder, obsessive-compulsive disorder, panic disorders with or without agoraphobia, social phobia, and post-traumatic stress disorder, and as part of the management of bulimia nervosa. Fluoxetine is also used in the treatment of premenstrual dys-phoric disorder.

Fluoxetine, a phenylpropylamine derivative, is given orally as the hydrochloride doses are expressed in terms of the base. Fluoxetine hydrochloride 22.4 mg is equivalent to about 20 mg of fluoxetine.

In the treatment of depression the usual initial dose of fluoxetine is 20 mg once daily US product information recommends giving this dose in the morning. If no clinical response is seen after several weeks, the daily dose may be gradually increased, up to a maximum of 80 mg daily (60 mg in the elderly). Doses above 20 mg daily may be given in 2 divided doses, for example in the morning and at noon, or as a once daily dose. A once-weekly, modified-release preparation equivalent to 90 mg of fluoxetine is available in the USA for use in patients whose depressive symptoms have stabilised, and who require long-term treatment it is recommended that weekly dosing is started 7 days after the last daily dose of fluoxetine.

Fluoxetine is also used for the treatment of depression in children aged 8 years and over. Initial doses of 10 mg should be increased to 20 mg daily after 1 week (except in low-weight children when such increases should not be made for several weeks, and then only if the clinical response is insufficient). Because of concerns about the use of SSRIs in children (see Effects on Mental State, above) its use is licensed in the European Union only as an adjunct to psychological therapy in children and adolescents with moderate to severe depression who have not responded to psychological therapy alone.

Fluoxetine is used in recommended doses of 60 mg once daily in the management of bulimia nervosa. In the management of obsessive-compulsive disorder the initial dose of fluoxetine is 20 mg once daily increased after several weeks if there is no response to up to 60 mg daily. Up to 80 mg daily has been used, sometimes divided into 2 doses. Fluoxetine is also licensed in the USA for use in children aged 7 years and over for obsessive-compulsive disorder. The starting dose is 10 mg daily in low-weight children this is increased after several weeks to 20 to 30 mg daily, if required. Adolescents and heavier children may be increased to 20 mg daily after 2 weeks further increases to 60 mg daily may be made after several weeks, as necessary.

Fluoxetine may be used in the treatment of panic disorder in initial doses of 10 mg once daily. After a week the dose should be increased to 20 mg daily further increases to 60 mg daily may be considered after several weeks if no improvement is seen.

A dose of 20 mg daily is used in the treatment of premenstrual dysphoric disorder. Intermittent dosing is also permitted: for each new cycle, fluoxetine should be started 14 days before the onset of menstruation and continued until the first full day of menstruation. Treatment may be continued for 6 months benefit should then be reassessed before continuing further.

A lower or less frequent dosage is recommended in elderly patients. For dosage in hepatic or renal impairment see below.

It should be noted that because fluoxetine and nor-fluoxetine have prolonged half-lives several weeks of therapy are required before steady-state concentrations are attained similarly after dosage adjustments a time lag will occur before steady-state concentrations are again achieved. Although SSRIs should generally be withdrawn gradually to reduce the risk of withdrawal symptoms, the long half-life may reduce the need for dose tapering with fluoxetine.

Administration in hepatic or renal impairment. Fluoxetine is subject to hepatic metabolism, and, therefore, lower doses, such as alternate-day dosing, have been recommended in patients with significant hepatic impairment.

It is also excreted by the kidneys and licensed information for some UK products recommends a similar dose reduction in patients with mild to moderate renal impairment and that fluoxetine should be avoided in those with severe impairment. However, other UK and US product information states that plasma concentrations of fluoxetine or its metabolite norfluoxetine in patients with severe impairment requiring dialysis did not differ from those in controls with normal renal function when given fluoxetine 20 mg daily for 2 months.

Anorexia nervosa. Counselling and psychotherapy form the major part of treatment of anorexia nervosa and there is little or no role for specific drug therapy. Antidepressants may be indicated when there is co-existing depression or obsessive-compulsive disorder but malnourished anorexic patients may be more susceptible to adverse effects and less responsive than other patients with depression. Fluoxetine has been tried with some success particularly when used to prevent relapse once weight gain has been achieved.

Anxiety disorders. SSRIs have been given in a variety of anxiety disorders but their role in these disorders is most well established in the treatment of obsessive-compulsive disorder. Efficacy in obsessive-compulsive disorder appears to have been best demonstrated for fluvoxamine and fluoxetine but other SSRIs are also effective and patients unresponsive to one SSRI may respond to another. SSRIs are also of use in the treatment of generalised anxiety disorder, panic disorder, and post-traumatic stress disorder. SSRIs are considered to be the first choice for the treatment of social anxiety disorder (see under Phobic Disorders). Fluoxetine is one of the SSRIs that has been tried in the treatment of trichotillomania.

Bipolar disorder. Treatment of the depressive phase of bipolar disorder with antidepressants needs caution since these drugs may precipitate mania or hypomania. SSRIs such as fluoxetine have nonetheless been used in bipolar disorder with some success. In some countries, fluoxetine is also available as a fixed-dose combination with the atypical antipsychotic olanzapine for use in the depressive phase of bipolar disorder.

Bulimia nervosa. A combination of counselling, support, psychotherapy, and antidepressants is the usual treatment for bulimia nervosa. Fluoxetine and the tricyclic desipramine have been suggested as the antidepressants of choice because they have been used extensively and are considered to be well tolerated. Other S SRIs that have been tried include sertraline, fluvoxamine, and paroxetine. Antidepressants in general do not appear to alter the patient’s disturbed self-image, although disturbed attitudes might improve during short-term therapy with fluoxetine. References.

Depression. As discussed, there is very little difference in efficacy between the different groups of antidepressant drugs, and choice is often made on the basis of adverse effect profile. SSRIs such as fluoxetine are widely used as an alternative to the older tricyclics as they have fewer adverse effects and are safer in overdosage.

Combination therapy with differing classes of antidepressants, including the SSRIs, has been used in the treatment of drug-resistant depression. However, such therapy may result in enhanced adverse reactions or interactions and is considered unsuitable or controversial by some workers. For further details see Antidepressants, under Interactions of Phenelzine. References to the use of SSRIs in general and to the use of fluoxetine are given below.

Disturbed behaviour. SSRIs appear to have been of some benefit in controlling symptoms such as impulsiveness and aggression when tried in various disorders for the management of disturbed behaviour. There have been several case reports of fluoxetine being used with some success in the control of fantasies associated with various paraphilias.

Headache. The results of several studies have suggested that SSRIs may be of benefit in the treatment of chronic tension-type headache however, results in the prophylaxis of migraine have been conflicting. References.

Hot flushes. HRT with oestrogens is usually the mainstay of acute treatment for symptoms such as hot flushes associated with the menopause however, HRT has potentially tumour-stimulating effects and may be unsuitable in some patients, particularly those with a history of breast cancer (see Malignant Neoplasms, under Precautions of HRT). Preliminary studies have shown that some SSRIs (fluoxetine, paroxetine, and sertraline) have a modest effect on alleviating hot flushes and may be an alternative to HRT in peri- and postmenopausal women and in women with a history of breast cancer. However, there is some concern that the SSRI paroxetine may interact with tamoxifen treatment in patients with breast cancer (see Antidepressants under Interactions in Tamoxifen). In addition, paroxetine has been tried in the treatment of hot flushes in men receiving anti-androgen therapy for prostate cancer.

The serotonergic antidepressant venlafaxine has also been tried with some success in men and women with hot flushes in whom other treatments were unsuitable.

Hyperactivity. When drug therapy is indicated for attention deficit hyperactivity disorder (ADHD) initial treatment is usually with a central stimulant. SSRIs such as fluoxetine have produced beneficial effects as an adjunct to central stimulants in small numbers of patients with comorbid disorders such as depression or obsessive-compulsive disorder although there is insufficient evidence to assess their efficacy in ADHD alone.

Hypochondriasis. SSRIs may be of benefit in patients with hypochondriasis. Fluoxetine in an initial dose of 20 mg daily gradually increased up to 80 mg daily produced some beneficial results in 10 of 14 patients with hypochondriasis who completed 12 weeks of treatment. Fluvoxamine and paroxetine have also been tried.

Hypotension. SSRIs have been used in patients with neurally mediated hypotension refractory to standard treatment, although evidence of benefit is mainly anecdotal. However, a small study found that paroxetine reduced the incidence of both tilt-induced and spontaneous syncope. See also Orthostatic Hypotension, below.

Obesity. Fluoxetine has been tried with some success as part of the management of obesity. Fluoxetine’s mechanism of action in obesity is unknown. Serotonin is believed to be involved in the regulation of satiety but fluoxetine has also been shown to increase resting energy expenditure and raise basal body temperature. A common dose for fluoxetine in the management of obesity has been 60 mg daily it appears that fluoxetine has a dose-related effect on weight loss. Reviews agree that fluoxetine can aid weight reduction in the short term but after 16 to 20 weeks some patients have started to regain weight and its long-term efficacy remains to be established. Troublesome adverse effects can occur. Some patients treated with fluoxetine for depression have experienced an increase of appetite and some have gained weight. There has been a report of a patient who lost weight during treatment with fluoxetine for depression despite an increased appetite and food intake.

Orthostatic hypotension. Although orthostatic hypotension has been reported in some patients taking SSRIs, there has been a report that fluoxetine 20 mg daily for 6 to 8 weeks produced beneficial effects in 4 of 5 patients with chronic symptomatic orthostatic hypotension refractory to other treatment. Modest benefits have also been seen in patients with orthostatic hypotension associated with parkinsonism.

Pain. SSRIs have been tried in the treatment of painful disorders including fibromyalgia and diabetic neuropathy. See also Headache, above.

Parkinsonism. It has been suggested that fluoxetine might be of use in the management of selected patients with Parkinson’s disease who have levodopa-induced dyskinesias unresponsive to other measures. However, although fluoxetine has been reported to have produced beneficial results in such patients there has also been a report of increased disability in patients with Parkinson’s disease given fluoxetine. Extrapyramidal effects have also been reported in other patients taking fluoxetine (see under Adverse Effects, above). Fluoxetine has been tried in parkinsonism-related orthostatic hypotension (above).

Pathological crying or laughing. Inappropriate or uncontrolled crying or laughing can occur in patients with lesions in certain areas of the brain. Attempts at treatment have mostly been with antidepressant drugs, including SSRIs. Beneficial effects have been claimed for fluoxetine in a number of uncontrolled studies and reports.

Peripheral vascular disease. Anecdotal reports’ and a small pilot study of fluoxetine (in a daily dose of 20 to 60 mg) suggest it may produce favourable therapeutic responses in patients with Raynaud’s syndrome (see Vasospastic Arterial Disorders).

Premenstrual syndrome. Fluoxetine is used to control both the psychological and somatic symptoms of women with premenstrual dysphoric syndrome, a severe form of premenstrual syndrome. Other SSRIs also appear to be useful, although evidence is so far more limited.

Sexual dysfunction. Impotence or ejaculatory problems have been reported as adverse effects of SSRIs (see Effects on Sexual Function in Adverse Effects, above). Such properties of the SSRIs have been studied as a potential form of treatment for men with premature ejaculation. The relative effects of the SSRIs on delaying ejaculation have also been studied Paroxetine was found to cause the strongest delay, followed by fluoxetine and then sertraline fluvoxamine caused a slight delay although the effect was not significantly different from that seen with placebo. (Citalopram was unavailable at the time of the study later studies of its effect have been conflicting.)

Preparations

BP 2008: Fluoxetine Capsules Fluoxetine Oral Solution

The United States Pharmacopeia 31, 2008: Fluoxetine Capsules Fluoxetine Delayed-Re I ease Capsules Fluoxetine Oral Solution Fluoxetine Tablets.

Proprietary Preparations

Argentina: Alental Animex-On Anzolden Captaton Eburnate Equilibrane Faboxetina Felixina Fibrotina Fluopiram Foxetin Lapsus Mitilase Nervosal Neupax Neuro Laz Prozac Saurat

Australia: Auscap Erocap † Fluohexal Fluoxebell Lovan Prozac Zactin

Austria: Felicium Fluctine Fluoxenorm Fluoxibene Fluoxistad Fluxil FluxoMed Mutan NuFluo Positivum

Belgium: Docfluoxetine Fluox Fluoxemed Fluoxone Fontex Prosimed Prozac

Brazil: Daforin Deprax Depress Eufor Fluox Fluxene Nortec Prozac Frozen Psiquial Verotina

Canada: FXT Prozac

Chile: Actan Alentol Anisimol Clinium Dominium Pragmaten Prozac Sostac Tremafarm †

Czech Republic: Deprenon † Deprex Floxet Flumirex † Fluocim † Fluogal Fluoxin Fluval Fluxonil † Fluzak Framex † Magrilan Milezin † Portal Prozac

Denmark: Afeksin Flutin Fluxantin Folizol Fondur † Fonigen † Fontex Fonzac

Finland: Fluoxal † Fluxantin † Fontex † Seromex Seronil

France: Prozac

Germany: Fluctin Fluneurin Fluox Fluox-Puren Fluoxa † Fluoxe-Q FluoxeLich Fluoxemerck † Fluoxgamma Fluxet Fysionorm †

Greece: Dagrilan Dinalexin Exostrept Flonital Fluocalm Fluxadir Fokeston FHapilux Ladose Orthon Sartuzin Sofelin Stephadilat-S Stressless Thiramil Zinovat

Hong Kong: Atd CP-Fluoxet Deprexin Fluxetin Fluxil Magrilan † Nopres Plazeron † Provatine † Prozac Qualisac

Hungary: Deprexin Fefluzin † Floxet FluWinox Portal Prozac

India: Depzac Fludac Flufran † Nuzac Platin

Ireland: Affex Biozac Gerozac Norzac Prozac Prozamel Prozatan Prozit

Israel: Affectine Flutine Prizma Prozac

Italy: Azur Clexiclor Cloriflox Deprexen Diesan Flotina Fluoxeren Fluoxin Grinflux † Ibixetin † Ipsumor Prozac Serezac † Xeredien Zafluox †

Malaysia: Fluran Fluxetil Prozac Salipax

Mexico: Aponeusak Auroken † Axtin Deprozin Farmaxetina Flocet Florexal Fluctine Fluneurin Fluoxac Fluralex Flutinax Indozul Lebensart Ovisen Prozac Quanilene Regultron Siquial Ulmely Zatin

The Netherlands: Fluoxstad Flustad Prozac

Norway: Fontex

New Zealand: Fluox Plinzene Prozac

Philippines: Adepssir Deprexone Deprizac Prozac

Poland: Andepin Bioxetin Deprexetin Salipax Seronil

Portugal: Digassim Mizac † Nodepef Prozac Psipax Salipax Selectus Tuneluz

Russia: Apo-Fluoxetine Fluval Framex Portal Prodep Proflusak Prozac

South Africa: Deprozan Lorien Nuzak Prohexal Prozac Ranflocs Sanzur †

Singapore: Deprexin Fluxetil Fluxetin Fluxil Foxtin Nagrilan Prozac Zactin

Spain: Adofen Astrin † Augort † Ledmar Luramon Nodepe Prozac Reneuron Zaxetina †

Sweden: Fluxantin † Fontex Seroscand †

Switzerland: Fluctine Fluesco Fluocim Fluoxbasan † Fluoxifar Flusol

Thailand: Actisac † Anzac Dawnex Deproxin Flulox Flurried Fluoxine Flusac Flutine Fluxetil Fluxetin Fluzac Hapilux † Loxetine Nagrilan Oxetine Oxsac Prodep Prozac Unprozy

Turkey: Depreks Fulsac Loksetin Prozac Seronil Zedprex

United Arab Emirates: Flutin

UK: Prozac Prozit

USA: Prozac Sarafem

Venezuela: Anoxen Antipres Fluxet Fluzac Prozac Psiquial

Multi-ingredient

Argentina: Combined Symbyax †

Chile: Symbyax

India: Fludep Plus

Mexico: Symbyax

USA: Symbyax

Because of their epileptogenic effect SSRIs should be used with caution in patients with epilepsy or a history of such disorders (and should be avoided if the epilepsy is poorly controlled). Treatment should be stopped if seizures develop or when there is an increase in seizure frequency. Care is advised in patients receiving ECT as prolonged seizures have occurred rarely. SSRIs should also be used with caution in patients with cardiac disease or a history of bleeding disorders. Although SSRIs are preferred to tricyclics for the treatment of depression in patients with diabetes, they may alter glycaemic control and therefore caution is also warranted in diabetic subjects. SSRIs should be used with caution in patients with angle-closure glaucoma. Fluoxetine should be stopped in patients who develop a rash since systemic effects, possibly related to vasculitis, have occurred in such patients. Fluoxetine undergoes hepatic metabolism and should be used with caution and in reduced doses in patients with impaired hepatic function.

Patients should be closely monitored during early therapy until significant improvement in depression is observed because suicide is an inherent risk in depressed patients. For further details, see under Depression. For a discussion of the concern that SSRIs may

increase suicidal ideation, and concerns about their use for depression in children and adolescents, see Effects on Mental State in Adverse Effects, above. Suicidal thoughts and behaviour may also develop during early treatment with antidepressants for other disorders the same precautions observed when treating patients with depression should therefore be observed when treating patients with other disorders. If SSRIs are given for the depressive component of bipolar disorder, mania may be precipitated. Symptoms may also worsen during the initial treatment of panic disorder with SSRIs. SSRIs may impair performance of skilled tasks and, if affected, patients should not drive or operate machinery.

Some licensed product information recommends reduced or less frequent dosage of SSRIs for elderly patients.

SSRIs should generally be withdrawn gradually to reduce the risk of withdrawal symptoms although this may be unnecessary for fluoxetine because of its long half-life.

Abuse. There have been occasional reports of individuals abusing fluoxetine.

Blood disorders. For a reference recommending cautious use of SSRIs in patients with a history of bleeding disorders, see Effects on the Blood in Adverse Effects, above.

Breast feeding. The American Academy of Pediatrics considers that all antidepressants, including SSRIs (fluoxetine, fluvoxamine, paroxetine, and sertraline) are drugs whose effect on nursing infants is unknown but may be of concern. In addition, most licensed drug information advises that SSRIs should be avoided by the mother during breast feeding.

• Citalopram and its metabolites have been detected in breast milk however, in one study, the plasma concentrations in exposed infants were either very low or undetectable and no adverse effects were reported. In another study, 3 out of 31 breast-fed infants whose mothers were taking citalopram had adverse effects, specifically one case each of colic, decreased feeding, and irritability. However, there was no significant increase in the risk of adverse events in this group of infants when compared with either infants of depressed mothers not taking citalopram, or infants of healthy controls.

• Symptoms of colic were reported in a 6-week-old infant whose mother was taking fluoxetine 20 mg daily. The concentrations of fluoxetine and its active metabolite norfluoxetine were 69 nanograms/mL and 90 nanograms/mL respectively in breast milk, and 340 nanograms/mL and 208 nanograms/mL respectively in the infant’s plasma. The infant’s symptoms resolved when he was formula fed. Postnatal weight gain has been reduced in infants exposed to fluoxetine during breast feeding, although in all cases the reduction was less than 2 standard deviations below the norm. In another report, several seizure-like episodes occurred in a breast-fed infant whose mother was taking fluoxetine in addition to carbamazepine and buspirone however, plasma drug concentrations in the infant were significant only for fluoxetine and norfluoxetine. In a study of 10 women taking fluoxetine while breast feeding 11 infants, breast milk concentrations of fluoxetine ranged from 17.4 to 293 nanograms/mL and of norfluoxetine from 23.4 to 379.1 nanograms/mL. No adverse effects were noted in the infants. Similar levels have occurred in other breast-fed infants without any apparent drug-induced adverse effects. Fluoxetine and norfluoxetine were detected in the milk of 14 nursing women. Blood samples were taken from 9 of the infants in the study, and of these, fluoxetine was detected in the plasma of 5 and norfluoxetine in 7. Although it was felt that many infants would tolerate the mean combined dose of fluoxetine and norfluoxetine transmitted via breast milk in this study, there was considerable interpatient variability in estimated infant dose and caution should be exercised neonates in particular exhibited higher concentrations of norfluoxetine than older infants. Moreover, since both fluoxetine and norfluoxetine have long half-lives, neonates already exposed in utero may have an additional risk of adverse effects during breast feeding.

• The excretion of fluvoxamine into breast milk was studied in a woman who had been receiving fluvoxamine maleate 100 mg twice daily for 2 weeks. The concentration of fluvoxamine base 4.75 hours after a dose was 310 nanograms/mL in maternal plasma and 90 nanograms/mL in breast milk. It was estimated that an infant would ingest only 0.5% of the daily maternal intake. It was considered that these data supported the notion that the use of fluvoxamine by nursing mothers posed little risk to the infant. A subsequent study found no detectable drug levels in the plasma of breast-fed infants exposed to fluvoxamine the authors suggested that fluvoxam-ine was a reasonable choice for nursing mothers requiring treatment for depression.

• Although paroxetine was detected in measurable concentrations in the breast milk of a group of 10 nursing mothers receiving paroxetine no adverse effects were reported in any of their breast-fed infants. Paroxetine could not be detected in the plasma of 7 of the 8 infants from whom samples were obtained and in the other infant, concentrations were not quantifiable. Another study involving 7 women suggested that the dose of paroxetine to suckling infants would be 0.7 to 2.9% of the weight-adjusted maternal dose. A later study also found no detectable drug levels in the plasma of breast-fed infants exposed to paroxetine the authors suggested that paroxetine was a reasonable choice for nursing mothers requiring treatment for depression. In a prospective cohort study, weight gain at 6 and 12 months of age in infants whose mothers took paroxetine during breast feeding was not adversely affected when compared with the infants of mothers who either did not breast feed or breast fed without taking any drugs during lactation in addition, there was no difference in reaching the usual developmental milestones between the infants of the 3 groups.

• Plasma concentrations of sertraline were undetectable in a breast-fed infant despite the presence of concentrations in the mother’s breast milk ranging from 8.8 to 43 nanograms/mL over a 24-hour period. However, the authors pointed out that metabolite levels were not measured and that sertraline may have been present in the infant at a concentration below the level of sensitivity of the assay. Other studies have detected desmethylsertraline in breast milk, which was also detected in the plasma of some of the infants in a number of the studies but not all. The authors of at least one study suggested that sertraline was a reasonable choice for nursing mothers requiring treatment for depression. In addition, some authors recommend expressing and discarding breast milk 8 to 9 hours after a maternal dose, when levels of desmethylsertraline and sertraline are maximal, in order to significantly reduce infant exposure to sertraline.

Children. SSRIs are associated with an increased risk of potentially suicidal behaviour when used for the treatment of depression in children and adolescents under 18 years old for further details, see under Effects on Mental State, above.

Diabetes m el I it us. A patient with type 1 diabetes mellitus experienced a loss of hypoglycaemic awareness following the start of treatment with fluoxetine. Awareness returned on tapered withdrawal of fluoxetine. Changes in blood sugar concentrations may occur in patients with diabetes treated for depression with SSRIs (see also Effects on the Endocrine System, above) however, these may represent an improvement in glycaemic control.

Driving. While affective disorders probably adversely affect driving skill, treatment with antidepressants can also be hazardous, although patients may be safer drivers with medication than without. Impairment of performance is largely related to sedative and antimuscarinic effects. These are more pronounced with older antidepressants such as the tricyclic antidepressants than with the SSRIs, but a comparative study of fluoxetine (an SSRI) and dosulepin (a tricyclic) in healthy subjects showed a similar but apparently small potential for impairing psychomotor and driving performance. A later epidemiological study was unable to confirm any increased risk of road-traffic accidents in those drivers receiving tricyclic antidepressants or SSRIs. In the UK, the Driver and Vehicle Ticensing Authority considers that drugs such as SSRIs may have fewer adverse effects on drivers than antidepressants with pronounced antimuscarinic or antihistaminic adverse effects, such as tricyclic antidepressants. However, all drugs acting on the CNS can impair alertness, concentration, and driving performance, particularly at the start of treatment or when the dose is increased driving must cease if patients are adversely affected. Patients with severe depressive illnesses complicated by significant memory or concentration problems, agitation, behavioural disturbances, or suicidal thoughts should also cease driving pending the outcome of medical enquiry.

Gastrointestinal disorders. For the opinion that the SSRIs may produce a clinically important increase in the risk of upper gastrointestinal bleeding in patients with a high risk of such bleeding, see under Effects on the Gastrointestinal Tract, above.

Glaucoma. For reference to SSRIs precipitating or exacerbating symptoms of glaucoma, see Effects on the Eyes, above.

Mania. Hypomania or mania have been reported with the SSRIs consequently, UK licensed drug information recommends that SSRIs should be withdrawn in any patient entering a manic phase.

Fluvoxamine was associated with manic behaviour in 8 patients who were being treated for major depression 3 also had obsessive-compulsive disorder. Daily doses of fluvoxamine ranged from 75 to 300 mg and duration of therapy to development of manic behaviour from 2 to 6 weeks. The authors were unable to determine whether fluvoxamine had induced mania or unmasked latent bipolar disorder in these patients. However, they recommended that fluvoxamine-treated patients should be monitored for manic behaviour.

Symptoms of manic behaviour also developed in a 7-year-old girl after taking sertraline for about 2 weeks for the treatment of major depression. She recovered within a few weeks of stopping sertraline.

Pregnancy. In an early prospective study comparing 128 pregnant women exposed to a mean daily dose of about 26 mg of fluoxetine during their first trimester with control groups receiving tricyclic antidepressants or non-teratogens, the incidence of neonatal malformations was similar in all groups and did not exceed that in the general population. However, there was a tendency to a higher incidence of miscarriages in the groups receiving fluoxetine or tricyclics. A more recent prospective study comparing 228 pregnant women taking fluoxetine with a control group taking non-teratogens also failed to find a significant increased incidence in major fetal abnormalities in the fluoxetine group it also did not reveal an increased risk of miscarriage. There was an increase in the incidence of minor fetal abnormalities in infants exposed to fluoxetine during the first trimester. Also, infants exposed to fluoxetine during the third trimester experienced more perinatal complications such as prematurity, low full-term birth-weight and length, and poor neonatal adaptation compared with infants exposed only during the first and second trimesters. However, the design of this study was criticised because of several methodological problems such as unmatched controls and a higher maternal age in the fluoxetine group, which may partly explain the excess of poor perinatal outcomes.

The manufacturer evaluated the outcome of 796 pregnancies in which the mother received fluoxetine during the first trimester and considered that it was unlikely that fluoxetine increased the risk of miscarriage or fetal malformation A prospective controlled study on pregnancy outcome in women exposed to fluvoxamine, paroxetine, or sertraline also found that, when used in recommended doses, there appeared to be no increase in the risk of major congenital malformations, miscarriages, or still-births when compared with women exposed to non-teratogens. Nonetheless, the results from a more recent meta-analysis which included some of the above studies have suggested that maternal exposure to antidepressant treatment (specifically SSRIs, tricyclics, nefazodone, trazodone, or venlafaxine) may significantly increase the risk of miscarriage in comparison to women not exposed to antidepressants. However, the authors acknowledged that the underlying depression itself might be a contributing factor to the increased risk.

There is some evidence that paroxetine may be more teratogenic than other antidepressants. The manufacturer GlaxoSmithKline has reported that overall the data from a retrospective US epidemiological study and a study using the Swedish national birth registry have indicated that there was a twofold increase in cardiovascular malformations, particularly ventricular septal defects, in infants born to mothers who had taken paroxetine during pregnancy compared with the general population. However, whereas the US study also showed an overall risk of major congenital malformations (inclusive of the cardiovascular defects), the Swedish study found no such increase.

Maternal use of SSRIs has been associated with neonatal complications. CNS toxicity and an increased heart rate were reported in a neonate whose mother had received 20 mg of fluoxetine daily throughout most of her pregnancy. The neonate’s symptoms resolved 96 hours after delivery. In another neonate whose mother took up to 30 mg daily of fluoxetine throughout the third trimester cardiac arrhythmias were noted. In a matched-control study the rate of complications after delivery in 55 infants exposed to paroxetine during the third trimester was higher than in a control group who had been exposed to paroxetine or non-ter-atogenic agents during the first or second timesters. Complications that occurred in the infants exposed in the third trimester included respiratory distress (9), hypoglycaemia (2), bradycardia (1), jaundice (1), and suckling problems (1). More recently, another matched-control study has suggested that exposure to SSRIs (in this case fluoxetine, paroxetine, and sertraline) after the 20th week of gestation may increase the risk of persistent pulmonary hypertension of the newborn (PPHN). Of 377 infants with a confirmed diagnosis of PPHN, the mothers of 14 (3.7%) had taken an SSRI after the 20th week of gestation, compared with only 6 out of 836 infants (0.7%) in the matched-control group. Although these figures represent about a sixfold increase in the risk of PPHN in infants exposed to SSRIs in utero, the absolute risk remains relatively low (about 6 to 12 per 1000 women).

A number of reports have described symptoms such as jitteriness, irritability, sleep disturbances, and altered muscle tone in neonates who had been exposed to SSRIs in utero, especially during the third trimester in the majority of cases the symptoms are mild and self-limiting. Although withdrawal symptoms have been reported with most SSRIs, they have been more commonly reported in those neonates exposed to paroxetine. More recently, some authors have used the term neonatal behavioural syndrome to refer to such symptoms. It is unclear whether the symptoms represent a withdrawal syndrome or direct serotonin toxicity however, it has been suggested by some that in utero exposure to SSRIs with a short half-life such as paroxetine may lead to a neonatal withdrawal syndrome whereas exposure to an SSRI with a long half-life, particularly fluoxetine, may manifest as neonatal serotonin toxicity. There have been case reports of intraventricular haemorrhage in neonates whose mothers took SSRIs during late pregnancy but there is currently not enough data to determine whether the frequency of such bleeds in infants exposed to SSRIs is higher than normal.

The effects of fluoxetine on fetal neurodevelopment were studied in 55 pregnant women by later assessing global IQ of the children no differences were seen in those exposed to fluoxetine in utero during the first trimester compared with those exposed to tricyclic antidepressants or adverse developmental influences. A subsequent study indicated that exposure to fluoxetine or tricyclic antidepressants throughout gestation did not appear to affect cognition adversely. In another follow-up study, subtle differences in motor development and control, in particular tremulousness and inappropriate fine motor movements, were noted in the infants and children of depressed mothers who had taken S SRIs during pregnancy when compared with the infants of depressed mothers who had not taken any medication. However, in other measures of mental development there were no observed differences between the two groups.

Surgery. In patients undergoing orthopaedic surgery, the risk of perioperative blood loss was significantly increased in those taking serotonergic antidepressants (specifically clomipramine, fluoxetine, fluvoxamine, paroxetine, sertraline, and venlafaxine) when compared with those on non-serotonergic antidepressants. In addition, there was a significant increase in the need for blood transfusion during surgery in those on serotonergic antidepressants compared with those not receiving antidepressant medication.

Withdrawal. Withdrawal reactions have been reported for all SSRIs and the related antidepressants mirtazapine and venlafaxine on dosage reduction or stopping treatment, although the frequency of such reactions may vary. Paroxetine and venlafaxine have been associated with withdrawal reactions more often than other serotonergic antidepressants in the case of paroxetine this may be due, in part, to its short half-life. Fluvoxamine also has a short half-life and has been shown in some studies to have a high risk of withdrawal reactions. The apparent lower risk of withdrawal reactions with fluoxetine may be due to its long half-life. Other factors that increase the risk of withdrawal reactions include abrupt withdrawal, the use of high doses, and prolonged therapy.

In general, withdrawal reactions tend to occur within 3 days of stopping an SSRI or related antidepressant, although a delay of up to 2 weeks may be noted with fluoxetine. Common symptoms include dizziness, numbness and tingling, gastrointestinal disturbances (particularly nausea and vomiting), headache, sweating, anxiety, and sleep disorders. In some cases withdrawal symptoms may be severe and disabling. There has also been a report of 2 patients without a history of major psychiatric disorder who developed severe behavioural symptoms when paroxetine was withdrawn. Withdrawal was abrupt in one patient and more gradual, over a 12-day period, in the other. Symptoms were mainly hypomanic over the first few days, followed by a period of escalated ego-dystonic aggression, behavioural dyscontrol, and suicidal ideation.

Antidepressant dose tapering appears to reduce the frequency and severity of withdrawal reactions. The BNF recommends that any antidepressant, including an SSRI, that has been taken regularly for 8 weeks or more should be stopped gradually over a period of about 4 weeks, or as much as 6 months in patients who have been receiving long-term maintenance therapy. The withdrawal syndrome of the SSRIs is not considered to be a consequence of dependence.’

See also Extrapyramidal Effects under Adverse Effects, above. For debate about whether a withdrawal syndrome exists in neonates whose mothers have received SSRIs see Pregnancy, above.

Interactions

SSRIs interact with other drugs mainly as a result of their inhibitory activity on hepatic cytochrome P450 isoenzymes. Individual SSRIs do not all exhibit the same degree of inhibition nor do they react with the same isoenzymes. The drugs inhibited by specific SSRIs depends on the isoenzyme affected. As SSRIs have occasionally been associated with bleeding disorders and other effects on the blood, caution is advised when they are given with drugs known to affect platelet function.

Although different antidepressants have been used together under expert supervision in refractory cases of depression, severe adverse reactions including the serotonin syndrome may occur. Sequential prescribing of different types of antidepressant may also produce adverse reactions, and an appropriate drug-free interval should elapse between stopping one type of antidepressant and starting another. SSRIs should not generally be given to patients receiving MAOIs or for at least 2 weeks after their use. No treatment-free period is necessary after stopping a reversible inhibitor of monoamine oxidase type A (RIMA) and starting an SSRI. At least one week should elapse between withdrawing an SSRI and starting any drug liable to provoke a serious reaction (e.g. phenelzine) in the case of the SSRI sertraline the drug-free interval is extended to 2 weeks, and for fluoxetine 5 weeks, because of their longer half-lives. (For fluoxetine, the interval may need to be further extended if therapy has been prolonged or if high doses have been given.) Adverse effects such as the serotonin syndrome may also occur when the SSRIs are given with other drugs known to act on the same neurotransmitter, a consequence of synergistic interaction.

Further details concerning some of these interactions, and others, are given below.

Antibacterials. Rapid development of delirium was reported in a patient when clarithromycin was added to his existing regimen of fluoxetine and nitrazepam It was suggested that his delirium was a result of increased plasma-fluoxetine concentrations produced by the inhibition of cytochrome P450 enzymes by clarithromycin. Serotonin syndrome developed in a patient given erythromycin in addition to sertraline this was attributed to inhibition of CYP3A4 by the antibacterial, resulting in accumulation of the SSRI. There have also been reports of serotonin syndrome when linezolid was given with fluoxetine, sertraline, paroxetine, and citalopram in the latter case the patient developed complications including metabolic acidosis and ultimately fatal cardiac arrest. Reviews of serotonin syndrome associated with the use of linezolid suggest that SSRIs are the interacting drug most often implicated. However, it has been suggested that if warranted, linezolid may be given to patients receiving SSRIs provided the patient is carefully monitored for signs and symptoms of serotonin syndrome.

Anticoagulants. SSRIs may increase the anticoagulant activity of some anticoagulants including acenocoumarol and warfarin.

Antidepressants. Combination therapy with differing classes of antidepressants has been used successfully in the treatment of drug-resistant depression. It should be emphasised, however, that such combinations may result in enhanced adverse reactions or interactions, and should be used only under expert supervision. This practice is considered unsuitable or controversial by some authorities. For further details of the interactions between different antidepressants when given together, see Phenelzine. For details of the serotonin syndrome that can arise when two serotonergic drugs with different mechanisms of action are given, see under Adverse Effects of Phenelzine.

Antiepileptics. Antidepressants may antagonise the activity of antiepileptics by lowering the convulsive threshold. There has been a report of the serotonin syndrome developing in a patient 14 days after fluoxetine had been added to carbamazepine therapy.

Phenobarbital has been reported to reduce serum concentrations of paroxetine. Steady-state serum concentrations of paroxetine were found to be lower in patients taking phenytoin than in those taking carbamazepine or valproate.

Low serum concentrations of citalopram have been reported in 2 patients also taking carbamazepine. Serum concentrations increased when carbamazepine was changed to oxcarbazepine. Some SSRIs have been reported to increase plasma concentrations of carbamazepine and phenytoin. For conflicting reports of the effect of fluoxetine on serum-valproate concentrations.

Antihistamines. Cyproheptadine given to male and female patients as treatment for sexual dysfunction induced by fluoxetine or paroxetine has produced re-emergence of previously controlled depressive symptoms or bulimia nervosa in some patients. Citalopram, fluoxetine, and fluvoxamine may increase plasma concentrations ofastemizole or terfenadine by inhibition of their hepatic cytochrome P450 metabolism, increasing the risk of ventricular arrhythmias use together should be avoided.

Antimalarials. For mention of the effect of the SSRI fluvoxamine on the metabolism of proguanil.

Antimigraine drugs. There have been rare reports of serotonin syndrome associated with the use of SSRIs with serotonin (5-HT1) agonists such as sumatriptan. Fluvoxamine may inhibit the metabolism of frovatriptan and zolmitriptan. Fortheeffectswhensome SSRIs are used with dihydroergotamine.

Antimuscarinics. For the effect of SSRIs on benzatropine. For the effect of paroxetine on procyclidine.

Antineoplastics. Paroxetine may inhibit the metabolism of tamoxifen for further details.

Antipsychotics. For reports of adverse effects in patients treated with SSRIs and antipsychotics, see under Chlorpromazine. Interactions between SSRIs and atypical antipsychotics are also mentioned under clozapine, olanzapine, risperidone, sertindole, and zotepine.

Antivirals. Plasma concentrations of fluoxetine and other SSRIs are possibly increased by HIV-protease inhibitors, such as ritonavir, which may inhibit metabolism of the SSRI. Unexpectedly, however, total exposure to paroxetine was approximately halved by a ritonavir-boosted fosamprenavir combination in a study in healthy subjects. Although the free fraction of paroxetine in plasma was increased, suggesting that it had been displaced from protein binding, the maximum concentration of free paroxetine was reduced.

The serotonin syndrome has been described in a few patients given regimens that included fluoxetine and antiretroviral-dose ritonavir. The reaction also occurred in another patient given fluoxetine and efavirenz.

Anxiolytics. Fluoxetine and fluvoxamine increase plasma concentrations of some benzodiazepines (see under Diazepam). There is a report of hyponatraemia and serotonin syndrome developing in a patient who received high doses of citalo-pram and buspirone.

Beta blockers. For the effect of fluoxetine and fluvoxamine on beta blockers.

Ciclosporin. For the effect of fluoxetine and fluvoxamine on ciclosporin.

Cough suppressants. For the effects when using fluoxetine or paroxetine with dextromethorphan.

Dopaminergics. Selegiline is an irreversible selective inhibitor of monoamine oxidase type B. Serious adverse effects have been reported when selegiline and S SRIs have been used together. In some instances, these reactions resemble the potentially fatal serotonin syndromes reported when SSRIs are given with non-selective MAOIs.

SSRIs should not generally be given to patients receiving selegiline, or for at least 2 weeks after it has been stopped. Similarly, at least one week should elapse between withdrawing an SSRI and starting selegiline this interval should be increased to 2 weeks for sertraline, and to 5 weeks for fluoxetine because of their longer half-lives.

Gastrointestinal drugs. Acute dystonia has been noted in a patient given fluvoxamine and metoclopramide. Similar reports have been published for other SSRIs (fluoxetine or sertraline) and metoclopramide. Involuntary twitching, tremor, and stiffness of the jaw and tongue occurred on both occasions after the use of intravenous metoclopramide in a patient also taking sertraline. The authors considered the adverse effects to be features of the serotonin syndrome.

For the effect of fluvoxamine on alosetron, and a recommendation that the combination be avoided. For the effect of fluvoxamine on proton pump inhibitors, including omeprazole.

General anaesthetics. For a report of a generalised tonicclonic seizure in a patient receiving paroxetine and methohexital sodium.

Hypnotics. For reference to visual hallucinations in patients receiving an SSRI concomitantly with zolpidem.

Levothyroxine. For mention of a decreased effect of levothy-roxine in patients given sertraline concomitantly.

Local anaesthetics. For the effect of fluvoxamine on ropivacaine.

Muscle relaxants. For a report of QT prolongation in a patient taking fluoxetine and cyclobenzaprine. For the effect of fluvoxamine on tizanidine.

NSAIDs. For reference to an increased risk of upper gastrointestinal bleeding in patients taking SSRIs and NSAIDs together, see under Effects on the Gastrointestinal Tract, above.

Opioid analgesics. A possible case of serotonin syndrome has been reported with tramadol and sertraline, and another when sertraline was given with high doses of oxycodone There have also been occasional reports of the syndrome in patients given tramadol with citalopram, fluoxetine, or paroxetine. Other reports of serotonin syndrome were associated with use of oxycodone and fluvoxamine, pethidine and fluoxetine, and citalopram with fentanyl or pethidine. For reference to SSRIs enhancing the effects and toxicity of methadone.

Parasympathomimetics. For the effect of fluvoxamine on tacrine. For the effect of some SSRIs on galantamine.

Sibutramine. There is a risk of CNS toxicity due to synergistic serotonergic actions when an SSRI is given with sibutramine.

Smoking. Serum concentrations of fluvoxamine were lower in smokers than non-smokers in a single-dose study. It was proposed that the polycychc hydrocarbons present in cigarette smoke stimulated hepatic metabolism of fluvoxamine by cytochrome P450 isoenzymes.

Stimulants. For the effect of paroxetine on the metabolism of atomoxetine.

Theophylline. For the effect of fluvoxamine on theophylline.

Pharmacokinetics

Fluoxetine is readily absorbed from the gastrointestinal tract with peak plasma concentrations appearing about 6 to 8 hours after oral doses. Systemic bioavailability does not appear to be affected by food. Fluoxetine is extensively metabolised, by demethylation, in the liver to its primary active metabolite norfluoxetine. Excretion is mainly via the urine. Protein binding is reported to be about 95%.

Fluoxetine used clinically is a racemic mixture consisting of R- and S-enantiomers in equal amounts. Both enantiomers are active according to animal studies, but S-fluoxetine is eliminated more slowly. Metabolism is believed to be mediated by cytochrome P450 isoen-zyme CYP2D6 (but see below), and leads to R- and S-enantiomers of norfluoxetine, with the S-enantiomer being considered as active as the parent drug the R-enantiomer is considered to be much less active. This metabolism is subject to genetic polymorphism. While the small proportion of the population known as slow metabolisers do show a different spectrum of parent drug and metabolite, the overall activity does not appear to be altered.

Fluoxetine is widely distributed throughout the body. Fluoxetine has a relatively long elimination half-life of about 1 to 3 days after acute use and 4 to 6 days after long-term use that of its metabolite, norfluoxetine, is even longer, being about 4 to 16 days. These long half-lives have clinical implications. Steady-state plasma concentrations will only be attained after several weeks. Additionally, fluoxetine and its metabolites may persist for a considerable time after treatment, and this has led to precautions concerning the subsequent use of other serotonergic drugs (see Interactions, above).

Fluoxetine and norfluoxetine are distributed into breast milk (see Breast Feeding under Precautions, above).

Metabolism. Although fluoxetine is stated by the manufacturers to be metabolised by the cytochrome P450 isoenzyme CYP2D6, which is supported by studies indicating that its disposition is altered in poor metabolisers of debrisoquine (a substrate for this enzyme), others have suggested that CYP2C19, and perhaps CYP2C9, play an important role.

(BANM, US Adopted Name, rINNM)

International Nonproprietary Names (INNs) in main languages (French, Latin, Spanish):

Note. The following terms have been used as ‘street names’ or slang names for various forms of fluoxetine: Distas; Green and whites; Greens; Limes; Pros; Zacs.

Pharmacopoeias. In Europe and US.

European Pharmacopoeia, 6th ed. (Fluoxetine Hydrochlonde). A white or almost white crystalline powder. Sparingly soluble in water and in dichloromethane freely soluble in methyl alcohol. A 1% solution in water has apH of 4.5 to 6.5.

The United States Pharmacopeia 31, 2008 (Fluoxetine Hydrochlonde). A white to off-white crystalline powder. Sparingly soluble in water and in dichloromethane freely soluble in alcohol and in methyl alcohol practically insoluble in ether. Store in airtight containers.

Adverse Effects

SSRIs such as fluoxetine are less sedating than tricyclic antidepres sants and have fewer antimuscarinic and cardiotoxic effects. Adverse effects reported with SSRIs include dry mouth and gastrointestinal disturbances such as nausea, vomiting, dyspepsia, constipation, and diarrhoea. Anorexia and weight loss may also occur. Neurological adverse effects have included either anxiety, restlessness, nervousness, and insomnia, or drowsiness and fatigue headache, tremor, dizziness, seizures, hallucinations, confusion, agitation, extrapyramidal effects, depersonalisation, mania, panic attacks, sexual dysfunction, and symptoms suggestive of a serotonin syndrome have also occurred. The concern that SSRIs may be associated with increased suicidal ideation is discussed under Effects on Mental State, below.

Excessive sweating, pruritus, skin rashes, alopecia, photo sensitivity, and urticaria have also been reported. Angioedema and anaphylactoid reactions have occurred. In some patients who have developed rashes while taking fluoxetine, systemic hyper sensitivity reactions involving the lungs, kidneys, or liver, and possibly related to vasculitis, have developed it has therefore been advised that fluoxetine therapy should be stopped in any patient who develops a skin rash. Hyponatraemia, possibly due to inappropriate secretion of antidiuretic hormone, has been associated with the use of antidepressants, particularly in the elderly. Hyperprolactinaemia and galactorrhoea have occurred, as have changes in blood sugar, in patients receiving SSRIs.

Arthralgia and myalgia have been reported and there have also been cases of orthostatic hypotension, yawning, urinary retention, and abnormal vision including blurred vision and mydriasis. Abnormal liver function tests have been reported rarely. SSRIs have occasionally been associated with bleeding disorders such as ecchymosis and purpura and other effects on the blood. In overdosage nausea, vomiting, and excitation of the CNS are considered to be prominent features death has been reported.

Incidence of adverse effects. In June 1992 the UK CSM had received 1236 reports of adverse effects with fluvoxamine (from about 280 000 prescriptions) compared with 2422 for fluoxetine (from about 480 000 prescriptions). The overall patterns of adverse effects were similar but dermatological reactions were more likely with fluoxetine and gastrointestinal reactions with fluvoxamine. Reports of attempted suicide increased after adverse publicity about SSRIs in 1990, and the number of reports per million prescriptions were similar for the 2 drugs (25 for fluoxetine and 20 for fluvoxamine) at that time such figures were not considered disconcerting given that features of depression, including attempted suicide, can worsen after the introduction of any antidepressant (see also Effects on Mental State, below). A later review by the CSM of the 5 SSRIs available in the UK (citalopram, fluoxetine, fluvoxamine, paroxetine, and sertra-line) found that the SSRIs were broadly similar with respect to their safety profile. A list of adverse reactions common to all SSRIs was provided.

A review of 1861 adverse reactions to citalopram, fluoxetine, fluvoxamine, paroxetine, or sertraline reported to the Swedish Adverse Drug Reactions Advisory Committee found that the most commonly reported reactions were neurological (22.4% of all reports), psychiatric (19.5%), and gastrointestinal (18.0%). Compared with other SSRIs, gastrointestinal symptoms were more common with fluvoxamine, psychiatric symptoms with sertraline, and dermatological symptoms with fluoxetine. A more recent meta-analysis has compared the adverse effect profile of fluoxetine with other antidepressants including the tricyclics and other SSRIs. The overall risk of any adverse effect with fluoxetine was less than that for the tricyclic antidepressants however, there was no difference in risk when fluoxetine was compared with other SSRIs. When considering individual adverse reactions, fluoxetine was more likely to cause activating effects such as insomnia, agitation, tremor, and anxiety, and gastrointestinal disturbances such as nausea, vomiting, diarrhoea, weight loss, and anorexia than other antidepressants. In contrast, the tricyclics were associated with a greater risk of sedation, an-timuscarinic effects (such as dry mouth, dizziness, and blurred vision), constipation, and weight gain, than fluoxetine.

Effects on the blood. Abnormalities in platelet aggregation were associated with fluoxetine given to a severely underweight patient. Platelet activity returned to normal when fluoxetine was stopped. Fluoxetine was also suspected of being the cause of bruising in a patient whose blood clotting parameters were within normal limits. Purpura and bruising have been reported to be the commonest adverse blood effects associated with fluoxetine, paroxetine, or sertraline although cases of thrombocytopenia have been recorded for all three antidepressants. The suggested mechanism was inhibition of uptake of serotonin into platelets, thereby disrupting platelet aggregation caution was recommended when treating patients with a history of bleeding disorders with SSRIs. However, a subsequent cohort study based on prescription-event monitoring provided only weak evidence of a link between the use of SSRIs and the development of bleeding disorders. A similar study found no evidence of a major increased risk of intracranial haemorrhage with the use of SSRIs, although smaller increases in risk could not be ruled out.

For mention of a possibly increased risk of gastrointestinal bleeding, see Effects on the Gastrointestinal Tract, below.

Effects on the cardiovascular system. SSRIs are not associated with the same degree of cardiotoxicity as the tricyclic antidepressants, although orthostatic hypotension has been reported in some patients. A decrease in heart rate with ECG changes has been noted with fluvoxamine. However, a study on long-term fluvoxamine treatment in 311 patients followed for 1 year revealed no significant effect on ECG findings compared with patients given placebo.

Concern over the use of sertraline in patients with coronary heart disease was raised after a report of a 53-year-old man with a history of coronary heart disease who experienced attacks of sudden precordial chest pain on starting treatment with sertraline. The pain responded to glyceryl trinitrate. The manufacturers pointed out that there had been no ECG changes confirming an ischaemic origin of the disorder in this patient and that in studies sertraline had had no demonstrable clinical effects on intraven-tricular conduction or ECG intervals. Furthermore, no significant changes in cardiovascular indices had been recorded in patients who had taken overdoses of up to 6 g of sertraline. It was suggested that this might have been an effect on the gastrointestinal tract possibly at the oesophageal level.

Effects on the cerebrovascular system. There have been rare reports of cerebral ischaemic events associated with the use of SSRIs. In one case a 57-year-old man receiving long-term treatment for atrial fibrillation and hypercholesterolaemia presented with a facial droop and slurred speech 3 days after starting paroxetine 20 mg twice daily. Symptoms resolved after anticoagulant therapy and stopping paroxetine but recurred when paroxetine was reintroduced at a dose of 10 mg twice daily. Paroxetine was stopped and no further episodes had occurred within 4 months at follow-up.

For a report that there is no major increased risk of intracranial haemorrhage associated with the SSRIs, see Effects on the Blood, above.

Effects on the endocrine system. The syndrome of inappropriate antidiuretic hormone secretion (SIADH) with hyponatraemia has been reported in patients receiving antidepressants. The UK CSM, commenting on reports it had received of hyponatraemia associated with antidepressants (fluoxetine, paroxetine, lofe-pramine, clomipramine, and imipramine), considered that it was likely to occur with any antidepressant, and usually involved elderly patients. However, the results of a later study have suggested that cases are more likely to occur with serotonergic antidepressants such as the SSRIs, clomipramine, and venlafaxine. Case reports of hyponatraemia in 16 patients treated with SSRIs have been summarised. A further review of reports on 15 patients with hyponatraemia with SIADH induced by fluoxetine (12 cases), fluvoxamine (2 cases), and paroxetine (1 case) showed that the risk was greatest during the early treatment phase. This is borne out by single-case reports of hyponatraemia with SIADH in elderly patients receiving either citalopram, paroxetine, or sertraline. A retrospective study of hyponatraemia associated with either fluoxetine or paroxetine use also showed the early onset of the condition and identified low body-weight as being another risk factor for developing hyponatraemia. Not unexpectedly, replacing one SSRI with another has resulted in a recurrence of hyponatraemia however, in one report, the symptoms of hyponatraemia did not recur until about 16 months after switching SSRIs.

SSRI-associated hyperprolactinaemia has been reported. Lactation and raised prolactin levels occurred in a teenager 3 days after fluoxetine was added to her existing therapy which included pimozide. Stopping fluoxetine had no effect on lactation, which only ceased after withdrawing pimozide. In another report, hyperprolactinaemia and galactorrhoea in an elderly woman receiving fluoxetine resolved on stopping the drug. Gynaecomastia, unrelated to prolactin concentrations, was associated with the start of fluoxetine therapy in a 49-year-old man. Symptoms subsided 10 months after withdrawing fluoxetine. Although S SRIs may be favoured for the management of depression in patients with diabetes, there is some evidence that sertraline and fluoxetine can induce hypoglycaemia. Licensed product information for other SSRIs also warns of similar risks with these products.

Effects on the eyes. Symptoms of glaucoma that developed in a patient receiving fluoxetine subsided within 2 days of drug withdrawal. Similar symptoms have been reported with citalopram, fluvoxamine, paroxetine, and sertraline. In some cases, the SSRI may have aggravated pre-existing glaucoma. Intra-ocular pressure after doses of fluoxetine was recorded in 20 patients in a placebo-controlled crossover double-blind study. Significant increases were found in all patients 2 hours after receiving fluoxetine by mouth some patients still had raised intraocular pressure after 8 hours. A review of case reports documenting SSRI-associated changes in intra-ocular pressure concluded that such changes were difficult to predict however, it was recommended that those with risk factors for glaucoma, such as elderly patients with a family history, should be considered for ophthalmic consultations before starting an SSRI and regularly throughout treatment.

Anisocoria (uneven pupillary dilatation) has been reported in a patient taking paroxetine and in another taking sertraline. It was noted that the UK CSM had received 21 reports of mydriasis associated with paroxetine but it appeared that noticeably asymmetrical mydriasis as seen in these 2 patients had not previously been reported.

Effects on the gastrointestinal tract. A case-control study suggested that treatment with SSRIs produced a moderately increased risk of upper gastrointestinal bleeding (adjusted relative risk 3.0). The risk was greatly increased if SSRIs were given with NSAIDs (relative risk 15.6). Treatment with SSRIs did not appear to increase the risk of ulcer perforation. The absolute risk of bleeding was estimated at one case per 8000 prescriptions, a risk similar to that of low-dose ibuprofen. A more recent cohort study found a similar increase in risk. However, others have questioned whether such an association exists. A retrospective cohort study in elderly patients found that there was an increasing risk of upper gastrointestinal bleeding as the extent of inhibition of serotonin reuptake by the antidepressant used increased. The effect was considered to be clinically important for patients with a high risk of such bleeding, namely the very elderly and those with a history of previous upper gastrointestinal bleeding.

Some consider that gastroprotection is unlikely to be justified in those given SSRIs alone and furthermore there are no studies to suggest that gastroprotective drugs reduce the risk of SSRI-associated haemorrhage. However, it has been recommended that such protection should be considered when SSRIs and NSAIDs are used together because of the increased risk.

Effects on the hair. A report on 2 patients who had hair loss associated with the use of fluoxetine noted 4 other published cases and stated that, up to the end of 1991, the US manufacturers had received 498 reports of fluoxetine-associated alopecia.

Effects on the liver. Acute hepatitis occurred in 2 patients after several months of fluoxetine treatment it was noted that 5 other cases of acute hepatitis with fluoxetine had been reported. Abnormal liver function tests were seen in a patient after a suicide attempt with sertraline and cefalexin. The patient was then started on venlafaxine but, again abnormal liver function tests were noted. When these abnormal values had decreased, sertraline was restarted at therapeutic doses, with a subsequent increase in liver function tests. Values returned to normal once all medications were stopped. Autoimmune hepatitis has been reported after therapeutic doses of sertraline rechallenge in this case was also positive.

Hepatotoxicity has also been rarely associated with citalopram and with paroxetine use.

Effects on mental state. As early as 1990 there was concern that the SSRIs increased the risk of suicidal ideation after the publication of a case series of such events with fluoxetine. Meta-analyses performed around that time (criticism of statistical power notwithstanding) did not confirm an increased risk and this was supported by the results of prescription-event monitoring. Nevertheless, reports of suicidal ideation as well as suicide and self-harm have continued to be published for the SSRIs and the issue remains controversial. A subsequent meta-analysis of 702 randomised controlled studies found an increased risk for attempted suicide in those taking SSRIs when compared with placebo but not when compared with tricyclic anti depressants.

Risk analysis of suicidal behaviour with any antidepressant is confounded by the rarity of suicide even in patients with depression, and by the possibility that such behaviour is a manifestation of the underlying depression. Furthermore, the more favourable safety profile of the SSRIs, particularly in overdosage (see Depression), when compared to the older MAOIs and tricyclic anti depressants may have resulted in the SSRIs being prescribed to those patients at greater risk of suicidal behaviour. In 2003, the UK CSM established an Expert Working Group to address increasing public concerns regarding the safety of the SSRIs. The serotonergic antidepressants venlafaxine and mirtazapine were also included and the conclusions given below also apply to these drugs. In its final report issued in December 2004, the Working Group concluded that the risk of suicide may increase in the early stages of treatment for depression in adults and consequently careful and frequent monitoring is important, particularly if a patient has worsening of symptoms or new symptoms after starting therapy. However, the Working Group noted that increases in the prescribing of SSRIs have not been associated with an increase in suicide rates, although they acknowledged that the interpretation of these findings was difficult. They could not rule out that there might be a modest increase in the risk of suicidal ideation and self-harm with the SSRIs when compared with placebo. However, there was insufficient evidence from clinical trials to determine any differences between the SSRIs as a group, or between the SSRIs and other antidepressants regarding the risk of suicidal behaviour evidence from the General Practice Research database has suggested that there is no increased risk of such behaviour with the SSRIs when compared with the tricyclic antidepressants. The Working Group also concluded that the evidence for a relationship between suicidal behaviour and a change in dose was not robust however, patients should be monitored for any new symptoms or worsening of disease around the time of any dose change.

The Expert Working Group of the CSM has also commented on the use of SSRIs in young adults. They concluded that although there was no clear evidence of an increased risk of self-harm or suicidal thoughts in young adults of 18 years or over, such patients have a higher background risk of suicidal behaviour than older adults and consequently those treated with SSRIs should be closely monitored. Furthermore, the results of a meta-analysis undertaken by the FDA found that although the overall risk of suicide was not increased in adult patients receiving antidepressants, there was a non-significant trend toward increased risk with younger age. The FDA considered the trend sufficiently convincing to warn that, like children and adolescents, young adults aged between 18 and 24 years treated with antidepressants of any type may be at increased risk of suicidal thinking and behaviour.

In 2003 the CSM recommended (on the basis of their Expert Working Group’s finding) that paroxetine should not be used to treat depressive illness in children under 18 years old. Data from studies received by CSM in May 2003 failed to show that paroxetine was effective in depressive illness in this age group and indicated that the risk of harmful outcome, including self-harm and potentially suicidal behaviour, was 1.5 to 3.2 times greater in those who received paroxetine when compared with placebo. Following a further review, the CSM extended their recommendation to include the SSRIs citalopram, escitalopram, and sertraline subsequent analysis had also associated these antidepressants with an unfavourable risk to benefit ratio in the treatment of depression in children under 18. The CSM also included fluvoxamine in their recommendation as its risk to benefit ratio was unassessable. Fluoxetine was not included and the CSM acknowledged that clinical trials had shown a favourable risk to benefit ratio for fluoxetine in the treatment of depression in young patients. The European Medicines Agency (EMEA) has also recommended that serotonergic antidepressants, including the SSRIs, should not be used in children and adolescents except within their approved indications, but considered that studies with fluoxetine in children and adolescents have shown a positive effect that outweighs any potential risks, and recommended that it should be licensed for use where needed, as an adjunct to psychological therapies in children from 8 years of age.

The FDA have not issued advice contra-indicating the use of these antidepressants in those under 18 years old in the USA, although they have stressed that all patients, including adolescents and children, should be closely monitored for worsening depression or suicidal behaviour, especially at the beginning of treatment. They also comment that, apart from fluoxetine, the SSRIs are not licensed in the USA for the treatment of depression in young patients.

The use of SSRIs has been associated with the development of akathisia, restlessness, and psychomotor agitation such as an inability to sit or stand still, particularly in the first few weeks of treatment. In some patients these symptoms have precipitated suicidal behaviour. However, the Expert Working Group of the CSM considered that it was not possible to draw any conclusions on the link between the development of these symptoms and the risk of suicidal behaviour, as such data were not included in the majority of cases.

There have been suggested links between the use of fluoxetine and irritability, hostility, and aggression However, one review noted that an unpublished analysis had indicated that patients taking fluoxetine for a variety of disorders were not more likely to be aggressive than those taking placebo. Prescription-event monitoring has also found no evidence to suggest that fluoxetine increases the frequency of aggression. Initiation of antidepressant therapy with paroxetine or sertraline has been associated with either worsening or a new onset of flashback syndrome in 2 patients with a history of lysergide abuse.

There have been isolated reports of memory loss associated with the use of SSRIs.

For further effects on mental function, see also under Withdrawal and under Mania in Precautions, below.

Effects on sexual function. Sexual dysfunction is often noted in patients with depression and may be due to their antidepressant medication or to the disease itself. Complaints include a decrease in or loss of libido, delayed ejaculation, erectile difficulty, or anorgasmia in men loss of libido, delayed orgasm, or anorgasmia have been reported in women. Early identification is important since drug-induced sexual dysfunction is a common cause of non-compliance in addition, it may adversely affect the quality of life of patients and hamper their recovery. It has been considered that sexual dysfunction occurs in up to 1.9% of patients taking fluoxetine, with impotence or ejaculatory problems occurring in less than 1% of patients. However, these figures, which were based on information supplied by the US manufacturer, have been disputed’ and the incidence of sexual dysfunction may be higher than the manufacturer’s data suggest. Earlier studies and anecdotal reports quoted rates of 7.8 to 75% for sexual dysfunction with fluoxetine but it appears that only small numbers of men were studied. A later review also estimated the incidence of SSRI-induced sexual dysfunction at between 10 and 75%.

The incidence of sexual dysfunction may differ between types of antidepressants. A large, observational study found the rate of sexual dysfunction to be higher with the SSRIs (citalopram, fluoxetine, paroxetine, and sertraline) and the SNRI venlafaxine than with bupropion or nefazodone. However, no significant difference in rate was found among the SSRIs as a group. The study also identified other risk factors for developing sexual dysfunction which included increasing age, the use of high doses, and use with other medications. Gender, race, and length of treatment were not associated with an increased risk. Suggested strategies for managing SSRI-induced sexual dysfunction include reducing the dosage of the SSRI or altering the timing of doses, or changing to another antidepressant. In some cases tolerance may develop, particularly if the dysfunction occurred early in treatment. One small study has indicated that spontaneous improvement may occur even up to 6 months after the start of therapy. Evidence of efficacy for drug treatments is mainly anecdotal. Cyproheptadine seems to have been tried most often, but the SSRI may become less effective (see Antihistamines, under Interactions, below) and patients should be monitored for worsening symptoms of depression. The effects of the SSRIs on sexual function have been studied as a potential form of treatment for men with premature ejaculation (see Sexual Dysfunction in Uses, below).

Effects on the skin. Toxic epidermal necrolysis developed in a 16-year-old girl 8 days after beginning fluvoxamine therapy. Other drugs, which included metoclopramide, clorazepate, and clomipramine were discounted as possible causes. Amitriptyline and fluoxetine have been implicated in the development of atypical cutaneous lymphoid hyperplasia in 8 patients, 7 of whom either had an underlying immunosuppressant systemic disease or were also receiving immunomodulatory drugs. The lesions improved or resolved on stopping the antidepressant, although in some patients other factors may have contributed to the improvement.

Bullous pemphigoid induced by fluoxetine has been reported in a 75-year-old woman. Spontaneous resolution followed within 3 weeks of stopping the drug.

Paroxetine treatment has been associated with the development of cutaneous vasculitis, which involved several fingers, in a 20-year-old woman rechallenge was positive. On both occasions the patient recovered when paroxetine was withdrawn.

A severe bullous reaction with full-thickness skin necrosis has been reported in a 48-year-old woman after 6 months of treatment with sertraline. The lesions required extensive skin grafts and recovery was prolonged, with drug-induced scleroderma developing in the affected area.

Epileptogenic effect. Generalised seizures have been reported in 2 patients with no history of seizures after starting fluoxetine therapy. Although convulsions have been noted in patients taking fluvoxamine (see Incidence of Adverse Effects), a small clinical study involving 35 depressed epileptic patients found no change in the number of seizures or in their nature when fluvoxamine was given in doses of up to 200 mg daily.

Extrapyramidal effects. Extrapyramidal effects, such as tics and akathisia, have been reported with fluoxetine. By 1993, the UK CSM had received 39 reports of extrapyramidal reactions with paroxetine including 15 of dystonia of the face and mouth. Although extrapyramidal effects might occur with other SSRIs, orofacial dystonias appeared to be more common with paroxetine. However, evidence from monitoring prescriptions within the UK showed that the overall incidence of extrapyramidal effects was the same for paroxetine as for other SSRIs. Orofacial dystonias (teeth clenching) or dyskinesias (teeth grinding), resulting in severe damage to teeth and gums in many cases, have been reported in 6 patients receiving fluoxetine, fluvoxamine, paroxetine, or sertraline. The authors concluded that these adverse effects were not specific for any particular SSRI. Analysis of spontaneous adverse reaction reports received by the national pharmacovigilance centre in The Netherlands found that there had been 41 reports of extrapyramidal effects associated with the SSRIs over a period of nearly 15 years parkinsonism and dystonia were the most frequently reported effects. Over the same time period, 14 reports had been received in total for other anti-depressants. The authors commented that the results might be biased by the selective reporting of adverse reactions to the SSRIs. Dyskinesia associated with withdrawal of citalopram andrisperi-done has been reported in a patient.

The onset of akathisia may be linked to suicidal ideation for further details, see Effects on Mental State, above.

Hypersensitivity. Hypersensitivity reactions to SSRIs are well documented. Interestingly, despite structural dissimilarities, there have been a few reports of cross-sensitivity between SSRIs. A young man who had previously developed a maculopapular rash while taking paroxetine suffered a similar reaction after starting sertraline both episodes resolved after the SSRI was withdrawn.

Hyponatraemia. See Effects on the Endocrine System, above.

Overdosage. SSRIs are generally regarded as being less toxic in overdosage than tricyclic antidepressants or MAOIs. A review of SSRI overdosage, covering the period 1985 to 1997, noted that there had been remarkably few fatal overdoses when taken alone. Moderate overdoses (up to about 30 times the usual daily dose) were generally associated with minor symptoms at most only at very high doses (more than 75 times the usual daily dose) did more serious effects such as seizures, ECG abnormalities, and decreased consciousness tend to occur. Toxicity was greatly increased, however, when overdoses of SSRIs were taken with alcohol or other drugs. There was no evidence of a difference in the various SSRIs with respect to safety in overdosage. The results of a more recent cohort study have also confirmed the relative safety of the SSRIs in overdosage. However, the study also found that citalopram was potentially more cardiotox-ic than other SSRIs in overdose, causing significant prolongation of the QT interval (see below). In addition, serotonin syndrome was noted as being a common feature of SSRI overdosage although in most cases symptoms were not severe or life-threatening.

• Fatal overdose was reported with citalopram in 6 patients, although the suggested cause of death as cardiac dysfunction rather than seizures was disputed. Nonetheless, cases’ of cardiac abnormalities (including prolongation of the QT interval) associated with citalopram overdose have continued to be reported, and routine ECG monitoring may be necessary in the management of overdosage. Some authors consider that a metabolite, didemethylcitalopram, rather than citalopram itself, may be responsible for the cardiotoxicity seen with citalopram overdose. At therapeutic doses, concentrations of didemethylcitalopram are much lower than those of citalopram and, in general, significant cardiotoxicity is not seen however, in overdosage the amount of didemethylcital opram may be sufficient to induce cardiac conduction abnormalities.

• A report involving 87 cases in which fluoxetine was taken in overdosage without other drugs found that the main symptoms were tachycardia, drowsiness, tremor, and nausea and vomiting. These were considered relatively minor, and were of short duration, and supportive care was considered to be the only intervention necessary.

• Of 41 cases of self-poisoning with fluvoxamine, only one patient died and even here fluvoxamine was not implicated. Prolonged cerebral depression occurred in a patient after fluvoxamine overdose, but this may have been due to an interaction with temazepam which the patient also took in overdose.

• One hour after taking 2 g of sertraline in a suicide attempt a 42-year-old woman was flushed, angry, emotionally labile, and easily distracted but not psychotic. Apart from watery bowel movements recovery was mainly uneventful after treatment with gastric lavage, oral activated charcoal with sorbitol, and intravenous hydration. In another case, symptoms resembling serotonin syndrome developed in a 51-year-old woman after an overdose attempt with sertraline it was believed that the patient may have taken as much as 8 g of sertraline. She recovered with supportive treatment. Abnormal liver function tests have also been noted after a suicide attempt with sertraline (see Effects on the Liver, above).

For a discussion of choice of antidepressant with respect to toxicity in overdosage, see under Depression.

Treatment of Adverse Effects

Treatment of overdosage with an SSRI involves appropriate symptomatic and supportive therapy. Activated charcoal may be given by mouth if the amount ingested was large (see below) and treatment is within an hour of ingestion. Dialysis, haemoperfusion, exchange transfusion, and measures to increase urine production are considered unlikely to be of benefit.

Activated charcoal. The UK Poisons Information Service considers the benefit of gastric decontamination in the management of overdosage with SSRIs to be uncertain. However, it is suggested that oral activated charcoal may be considered if this is given within 1 hour of ingestion and the quantity of SSRI exceeds the following amount:

• citalopram: 5 mg/kg (adult) 5 mg/kg (child)

• escitalopram: 2.5 mg/kg (adult) 2.5 mg/kg (child)

• fluoxetine: 500 mg (adult) 5 mg/kg (child)

• fluvoxamine: 1 g (adult) 100 mg (child)

• paroxetine: 600 mg (adult) 5 mg/kg (child)

• sertraline: 1 g (adult) 10 mg/kg (child)

Answer. I would need to know more about your medication regimen; i.e., do you take the 200 mg of Serzone as a single dose or in two or three doses? At what times of day? If you are now taking the Serzone as a daytime dose, I would try shifting most or all of it to bedtime. Splitting up the total dose into 2 small (25 mg) doses and taking the remaining 150 mg at bedtime might work for some patients, without compromising efficacy. If you are already taking the Serzone in this way and are still feeling heavily sedated all the time, there are two equally reasonable options, in my view:

1. Switch to a less sedating agent (e.g., fluoxetine [Prozac] or sertraline [Zoloft]); or

2. Add a small amount of a stimulating agent to the Serzone, such as methylphenidate (Ritalin), bupropion (Wellbutrin) or caffeine. Caffeine, however, may exacerbate anxiety in some patients.

Some patients may tolerate an alternating schedule of, say 200 mg of Serzone one day, 150 mg the next, etc. Have you tried cutting down the Serzone by just 25 mg/day? It may be that if you can cut it down to the point that you no longer feel so drowsy, a small amount of Wellbutrin (e.g., 37.5 mg per day) could be added to augment the Serzone’s antidepressant effect–Wellbutrin does not have very good antianxiety properties. If the first option is used, I would start very low with the Prozac or Zoloft dose, in order to avoid initial worsening of anxiety; e.g., 5 mg per day of Prozac or 12.5-25 mg of Zoloft. You might need to buy a pill-cutter.

Which path to take would depend, in part, on whether you wanted to accept he risk-benefit ratio of a second (augmenting) agent, vs. the risk-benefit ratio of trying a new and hence uncertain, medication. My general rule is “build on strength.” Try to work with and around the first successful agent, if possible. By the way, are you certain that your TSH is now normal? (In most labs, below 4.5-5.0.) Borderline hypothyroidism can certainly contribute to low energy and delay antidepressant response.

[1] Which of the following agents is contraindicated in a patient with epilepsy?

A. Bupropion

B. Fluoxetine

C. Mirtazapine

D. Venlafaxine

[2] The antidepressant action of imipramine is thought to be caused by which of the following?

A. Blockade of prejunctional α₂-adrenoceptors

B. Blockade of prejunctional neuronal norepinephrine and serotonin uptake transporters in the CNS

C. Increased numbers of β-adrenoceptors

D. Inhibition of monoamine oxidase

[3] Which of the following antidepressant agents inhibits hepatic microsomal enzymes to cause clinically significant drug-drug interactions?

A. Fluoxetine

B. Imipramine

C. Phenelzine

D. Trazodone

Answers

[1] A. Bupropion causes seizures in a small but significant number of patients. This number is reduced with use of the slow-release form.

[2] B. Imipramine and other TCAs block prejunctional neuronal norepinephrine and or serotonin uptake transporters in the CNS. Phenelzine and tranylcypromine inhibit monoamine oxidase. The heterocyclic agent mirtazapine blocks prejunctional α₂-adrenoceptors to enhance serotonin and norepinephrine neurotransmission.

[3] A. The SSRI fluoxetine inhibits cytochrome P450 and therefore can significantly elevate the level of other drugs metabolized by these hepatic enzymes.

Pharmacology pearls

SSRIs are the most commonly prescribed antidepressants because of their favorable side effect profile. Sexual disturbances and GI effects are common, however.

TCAs may lead to toxicity as a result of cardiac arrhythmias.

The antidepressant agents are roughly equivalent in their therapeutic action. However, individual patients may respond to, or tolerate, one better than another.

Small beginning doses of many antidepressant agents are usually preferred because with time tolerance may occur to some of their adverse effects.

Bupropion is contraindicated in patients with seizure disorders.

What is the mechanism of action of fluoxetine?

What are the common side effects of fluoxetine?

Answers to case: Antidepressant agents

Summary: A 30-year-old woman with major depression is prescribed fluoxetine.

Mechanism of action of fluoxetine: Inhibition of the reuptake of serotonin (5-hydroxytryptamine, or 5-HT) at the prejunctional nerve terminal.

Common side effects: Headache, nausea, agitation, insomnia, and sexual disturbances (loss of libido and erectile dysfunction).

Clinical correlation

Selective serotonin reuptake inhibitors (SSRIs) are the most frequently prescribed antidepressant medications. They act by inhibiting the reuptake of serotonin by the prejunctional nerve terminal, allowing more serotonin to interact with postjunctional neurons in the central nervous system (CNS).

This is thought to mediate their therapeutic effect. They have been highly effective in the treatment of major depressive disorders and have an excellent safety profile. Unlike tricyclic antidepressants (TCAs), which have multiple severe and potentially fatal effects in an overdose, SSRIs have relatively few severe toxicities and a very low potential for fatality in an overdose. SSRIs do have several side effects of clinical significance. They often cause headache and gastrointestinal (GI) side effects such as nausea. In some cases, agitation, anxiety, and insomnia can be exacerbated. Many of these SSRI side effects tend to be temporary and can often be improved with dose reduction. Another common side effect of SSRIs is sexual disturbance. Decreased libido and erectile dysfunction occur frequently and do not generally spontaneously resolve while continuing SSRI therapy, often leading to reduced patient compliance.

Approach to pharmacology of antidepressant drugs

Objectives

1. List the classes of antidepressant agents.

2. Contrast the mechanisms of action of the antidepressant agents.

3. Contrast the adverse effects and toxicities of the antidepressant agents.

4. Describe the indications and contraindications to antidepressant drug use.

Definitions

Major depressive disorder: Unexplained, long-term difficulty coping with life events characterized by an inability to experience pleasure, abnormal sleep, decreased libido and appetite, feelings of guilt, and suicidal ideation.

Continuation: Case: Antidepressant agents. Discussion

Reboxetine, produced by Pharmacia & Upjohn, is the first of a new class of antidepressant medications known as selective norepinephrine reuptake inhibitors. It works through a novel mechanism of action, affecting the norepinephrine system of the body. Researchers have long suspected that the neurotransmitter norepinephrine plays a major role in depression. In fact, reduced levels of this brain chemical have been linked to several specific symptoms that characterize this disorder, including diminished energy, interest, and motivation.

More specifically, reboxetine works by selectively inhibiting the reuptake process, whereby norepinephrine is drawn back up into the brain. By impairing reuptake, there is an increased supply of norepinephrine available for use. Transmission is concomitantly enhanced in various brain areas.

The results from two eight-week clinical trials of 549 patients with major depression in Europe, Latin America, and Australia demonstrated greater reductions in depression scores (as measured by the Hamilton Depression Scale) with reboxetine versus placebo. When compared to fluoxetine (Prozac®), reboxetine showed larger reductions in depressive symptoms, despite a similar rate of adverse events. Additionally, reboxetine treatment enabled more patients to achieve a normal score on a scale measuring social functioning by the end of the two studies. Social functioning encompasses active social behavior in both personal and occupational settings, as well as self-perception.

Although generally well tolerated in both studies, the most commonly reported side effects of reboxetine are dry mouth, insomnia, constipation, increased sweating, hypotension, urinary hesitancy/retention, paresthesia, tachycardia, and impotence.

Study author Juan Massana, M.D., concludes that these studies provide valuable information on the role of norepinephrine in depression and a possible new treatment option for patients suffering from this disease.

Researchers from the University of Oregon Health Sciences Center conducted a study of 10,000 people, 5,000 with Type II diabetes and 5,000 without it. The two groups were similar in other ways such as age and gender. The team hoped to answer three questions:

· Are diabetics more likely to be depressed than those who don’t have diabetes?
· What types of medications are depressed diabetic people likely to be taking?
· What are the costs of treating a person with diabetes, depression, or both?

They found that 62 percent of diabetics are more likely to be depressed than people without the disease. Among diabetics, those who are depressed are more likely to be taking insulin than managing their diabetes through diet and oral medications.

Diabetics are less likely to be taking antidepressant drugs. Of those who are being treated, most are on the same medications as any depressed patient—serotonin re-uptake inhibitors, such as Prozac, Zoloft, or Paxil. They also were taking more tricyclic antidepressants, such as Elavil. However, the researchers were uncertain if they were taking the antidepressants at the higher doses needed to treat depression or at the lower doses used to treat peripheral neuropathy, a common complication of diabetes.

As for costs, depressed people use more services and spend more money than average, whether or not they also have diabetes. In fact, it costs about the same amount to treat a diabetic without depression as it does to treat a depressed person without diabetes. Having both diseases increases costs considerably.

Treating the depression was shown to increase an individual’s adherence to his or her treatment plan, improve blood glucose control, as well as improve the individual’s psychological outlook.

From any perspective, untreated depression in Type II diabetes carries a high cost, financially, physically, and mentally.