 |
Back to 4.2 |
Introduction In the last ten years, consumption of serotonin re-uptake inhibitors (SSRI) has dramatically increased all over the world. Italian data show that in 1996 SSRI accounted for more than 30% of the total antidepressants sold. Despite this wide spread use and at least ten years on the drug market, the safety of SSRI has not been completely assessed. Controlled clinical trials conducted before the registration highlight only the common, and easily detectable adverse effects, but in some circumstances the association between the drug and the adverse effects has been unpredictable. This review focuses on available evidence concerning some specific complications – extrapyramidal reactions, sexual dysfunction, emergency symptoms after withdrawal of SSRI – all sharing the same link with SSRI use.
Extrapyramidal reactions (EPR) Many EPR have been reported with SSRI, especially fluoxetine, though the majority are case reports. At the moment, prevalence or overall incidence estimates on large population samples do not exist, so only a qualitative analysis is possible. Three systematic reviews concerning EPR during SSRI therapy have been published.-4 One evaluated fluoxetine adverse effects in a sample of 5555 patients followed in New Zealand. Fifteen cases reported EPR such as tremor, dystonic reactions, akathisia, tardive dyskinesia, severe trismus, muscle twitch in the lower extremities. None of the patients were older than 65 years. Anyway drug withdrawal and subsequent restarting were associated with an initial recovery followed by recurrence of symptoms.
Based on the number of case reports, akathisia seems more frequent than other EPR. In some reports the patient had previously taken a neuroleptic drug, so the directed association was difficult to assess. In other settings, monotherapy with paroxetine (20 mg/day), fluoxetine (20 mg/day or more), citalopram (20 mg/day) and fluvoxamine (300 mg/day) was detected. In the majority, akathisia occurred a few days after the first prescription of the antidepressant drug, associated with muscular rigidity, insomnia, anxiety and, rarely, suicidal ideation. There are, of course, problems in establishing the causal relations between the drug and the adverse effects reported, because depressive patients typically present many of these symptoms anyway.
Associations between EPR and SSRI have suggested a possible increased risk of tardive dyskinesia, which is rare but sometimes irreversible and potentially fatal. Patients showed abnormal involuntary movements after long term SSRI treatment or after drug interruption. A 42 year-old woman experienced tardive dyskinesia after four years of fluoxetine monotherapy. Drug withdrawal as well as add-on therapy of a neuroleptic drug worsened the clinical feature. Gradual withdrawal of fluoxetine, in the long run, reduced the incidence of dyskinetic movements, although they were still present after one year.
We do not know what the main risk factors are for EPR, but patients potentially at major risk of onset can be identified (table 1).
Concerning the management of EPR in patients treated with SSRI, evidence based data are very scanty. It is common sense to try reducing the dosage up to total withdrawal. When needed, different antidepressant con be tried, for example nortriptyline. Anticholinergic drugs and benzodiazepines are effective in reducing particular EPR due to neuroleptic therapy and can be used for the same purpose, but different outcomes have been reported depending on the type of EPR.
Effects on sexual dysfunction The effects on sexual dysfunction during antidepressant treatment are underestimated. Sexual dysfunction is often practically a symptom of the disease being treated: loss of libido and decreased sexual activity are important characteristics of the depressive syndrome.5 Further confounding factors are concomitant organic disease, which can aggravate drug-induced disturbances. This adverse effect is often unrecognised: many patients with depression are hesitant to communicate sexual problems and clinicians may not ask about them.6 One study detected the importance of a correct relation, using structured questionnaires, between the patient and doctor about whether antidepressant therapy induced sexual adverse effects or not.7 This may lead to the recommendation that the first appropriate intervention is to detect these adverse effects and follow them up in patients treated with SSRI.
Virtually all antidepressant from all classes have been associated with sexual dysfunction such as: decreased libido up to impotence, delayed orgasm and anorgasmia. It is still not known whether there is any difference between SSRI and TCA in the frequency of the onset of adverse effects. Clinical trials give an incomplete picture. Male sexual dysfunction is reported more often, though women, have much more difficulty dealing with it.
5 Some observational studies have followed cohorts of patients and the outcomes vary widely. Longitudinal studies reported 60% of prevalence while retrospective studies, underestimating the problem, indicate 10-20%. It seems that the frequency is higher in patients with major depression rather than other pathologies (panic attack, obsessive-compulsory disorder, etc). One possible explanation is that the higher dosages used to treat major depression could themselves be a cause of the higher incidence of adverse reactions.The management of antidepressant-induced sexual dysfunction has not been well studied. Two strategies are suggested: to act directly on current therapy or to try specific measures. The former is related to the SSRI treatment and in some patients the disorder may spontaneously regress. However, in all cases it is recommended to lower the dosages gradually, trying not to go below the minimum therapeutic doses. Another approach is to switch to a antidepressant of a different class, though there is no evidence which drug should be preferred.
8Finally, some drugs might be used like ‘antidotes’. However, the literature in this area consists mainly of case reports or open case series. There is a surprisingly wide range of alternatives. The most common antidote suggested is cyproheptadine, even yohimbine is reported to be effective in treating decreased libido, erectile dysfunction and anorgasmia caused by the SSRI in both men and women. There are also a few case reports of patients successfully treated with amantadine, dextroamphetamine, methylphenidate, pemoline, bupropion, nefazodone and buspirone, but solid data is lacking to support the use for this indication.
Withdrawal syndrome Charles Medawar first reported the withdrawal syndrome in SSRI users: in a detailed and provocative review he suggested that dependence and tolerance symptoms due to SSRI were similar to those reported with benzodiazepines.9 English adverse effects monitoring system10 has been used to evaluate the rates of occurrence of the events. Paroxetine withdrawn accounted for 5.1% of all the cases of emergency symptoms reported. For all the other SSRI the percentages were: sertraline 0.9%, fluvoxamine 0.4%, fluoxetine 0.06%. When the case report frequency per 1000 prescriptions ratio was taken into account, the estimates became: paroxetine 0.3%, sertraline and fluvoxamine 0.03%, fluoxetine 0.002%.
Similar results were achieved using the WHO database, which collects spontaneous case reporting from 47 countries.
11 The rate of spontaneous reporting, given the total number of cases and the pieces sold, translated into DDD per year, was higher for paroxetine than for the other SSRI. Table 2 summarises the seven studies on incidence published to date. A longitudinal and prospective experimental design may lead to higher estimate than one based on spontaneous reporting.Paroxetine, the best studied SSRI, may induce a withdrawal syndrome in 20 to 50% of patients. Similar percentages have been calculated for fluvoxamine, while for the other antidepressants data are scanty. To start with it seemed that fluoxetine caused fewer withdrawal symptoms than the others. However, Blomgren et al.
12 conducted a controlled clinical trial randomizing patients well treated with SSRI either to continue or to stop the treatment. A withdrawal syndrome appeared more frequently among patients who discontinued sertraline and paroxetine than fluoxetine. Zajecka et al. 13 report a similar picture: 395 patients successfully treated with fluoxetine were randomized either to continue or to stop. The authors did not report any particular syndrome on suspension of the drug, suggesting the absence of a real withdrawal syndrome. Data concerning fluoxetine must anyway be considered with caution. Fluoxetine pharmacokinetic properties are very much different from other drugs: its half-life is about six days, but its metabolite lasts 4-16 days. A delayed withdrawal syndrome may therefore occur, after some days or weeks. This might explain the encouraging results of drug suspension studies, which followed patients up only for a few weeks. However, there is evidence of a fluoxetine withdrawal syndrome 25 days after discontinuation and still present 56 days later.The spontaneous reporting data published in the literature are summarized in Table 3. In the majority of cases the reaction regressed within a couple of weeks. In some instances reintroduction of the antidepressant was necessary to resolve the symptoms it had itself produced. A new gradual withdrawal always led to definitive discontinuation of the drug.
The relation between withdrawal emergency symptoms due to SSRI and antidepressant dosage has not been established. However, more gradual discontinuation is recommended in patients treated with high doses. A preventive attitude seems appropriate. Antidepressant users should be warned that abrupt interruption of therapy can cause several disturbances, but that these can frequently be avoided by gradual dosage escalation.
14 The British National Formulary (1998) recommends a gradual dose increase within about one month, and a very carefully evaluation particularly of patients who report a history of drug withdrawal syndrome.
Discussion Extrapyramidal disturbaces, sexual dysfunction and withdrawal syndrome may be major complications during SSRI treatment, and their frequency, severity, and possible management strategies remain unclear. The major difficulty in recognizing the relation between EPR and SSRI probably arises because of the rarity of the symptoms. Moreover, very often these symptoms arise in patients who had taken other psychotropic drugs in the past, increasing the risk of EPR. Then too, comorbidity with neurological problems may complicate the clinical evidence so that risk attribuition become more complicated. On the contrary, difficulties in detecting sexual dysfunctions and withdrawal symptoms are hard to understand, given their high frequency in SSRI users.
Some explanations are related to the attribution of symptoms: both sexual dysfunction and withdrawal syndrome, which is generally aspecific, can easily be induced by the disease itself. Thus it is suggested that case reports may have been burgeoned only after withdrawal syndrome from paroxetine therapy was reported.
15On the other hand, confusion due to the adverse drug effects and worsening of the disease under treatment not only may be dangerous for individual patients, but may also influence all the information obtained from groups of patients. Long-term studies which randomized responders to SSRI to continue or stop therapy are an example. As Medawar suggested, the better conditions of those who continue therapy might be due not only to the drug preventing relapses of depression, but also to the fact that a percentage of those who discontinue therapy may experience a withdrawal syndrome erroneously classified as a relapse of the depressive illness.
16The example of the safety profile of SSRI, even after ten years on the market, suggests some considerations about pharmacovigilance systems. Spontaneous reporting models are considered effective for rapid identification of new and rare adverse effects. Single case reports are often criticized as not proving the association between a new adverse effect and the drug. There are some clear problems: such reports are extremely selective, they may report false-positive results, or under-report the real occurrence, and do not allow valid estimates of the frequency or the incidence of the adverse effect.
17, 18 Information collected in morbidity and mortality registries have the same limitations. For example they calculated fetal risk of malformation after in utero exposure to fluoxetine.19 The main limitation of this approach is that it cannot be used without a firm hypothesis to test, in a setting where routine drug prescription needs to be assessed so as to identify or quantify any problem - adverse effect or other – that comes to light.In order to go straight to this target, a real change is needed in attitudes and behaviour.
20 A research approach must be transferred to psychiatric community services, which are not the real setting where research is commonly done, but are the natural place to meet real patients who take drugs and stop taking them, experience rare adverse effects and withdrawal symptoms.21 In this context, longitudinal observational studies are urgently needed for all patients taking SSRI.
Table 1 -- probable risk factors for Extrapyramidal reactions (EPR) during SSRI therapy:
1. Concomitant or previous neuroleptic therapy 2. Use of drugs which reinforce serotonin transmission (lithium or other antidepressants) 3. High doses of SSRI 4. Withdrawal of MAO-Inhibitors in the previous four weeks 5. Metoclopramide 6. Cerebral damage (e.g. stroke) 7. Neurological disease (e.g. Parkinson’s disease) 8. Family history of drug-induced EPR 9. Emergency symptoms such as agitation, anxiety, insomnia, restlessness
| Table
2 Incidence of withdrawal symptoms associated with SSRI
therapy: observational studies |
||
| Drug | Study | Incidence of symptoms after drug withdrawal |
| Paroxetine | Keuthen 1994 Barr 1994 Bhaumik 1996 Coupland 1996 Oehrberg 1995 |
5/13 19% 3/6 50% 5/12 42% 10/50 20% 19/55 34% |
| Mallya 1993 Black 1993 Coupland 1996 |
4/17 24% 12/14 86% 10/50 20% |
|
| Fluoxetine | Bhaumik 1996 Coupland 1996 |
0
(?) 0/20 0% |
| Sertraline | Coupland 1996 | 1/45 2.2% |
| Table 3 Outcome of withdrawal symptoms associated with SSRI therapy: spontaneous reports | ||
| Drug | Number of case-reports | Outcome |
| Paroxetine | 20 | In 14 cases symptoms regressed within 1-2 weeks. In five cases paroxetine was reintroduced; in three a gradual discontinuation was possible without residual symptoms. In one case the drug was withdrawn and restarted twice; escalating the dosage permitted at last to stop the drug. |
| Sertraline | 6 | In four cases symptoms resolved spontaneously after a few days. In two cases the reintroduction of the drug was necessary to allow a gradual withdrawn |
| Venlafaxine | 6 | In three cases reintroduction of the drug was necessary, in three cases a gradual discontinuation achieved resolution of the symptoms |
| Fluoxetine | 5 | In two cases symptoms resolved spontaneously within 2-3 weeks. In three cases fluoxetine was reintroduced; in one was then gradually withdrawn |
| Fluvoxamine | 2 | In one case the drug was reintroduced |
References
1. Barbui C, Campomori A, D’Avanzo B, Negri E, Garattini S. Antidepressant drug use in Italy since the introduction of SSRIs: national trends, regional differences and impact on suicide rates. Social Psychiatry & Psychiatric Epidemiology 1999; 34: 152-156. 2. Caley CF. Extrapyramidal reactions and the selective serotonin-reuptake inhibitors. Annals Pharmacotherapy 1997; 31: 1481-9.3.
Lane RM. SSRI-Induced extrapyramidal side-effects and akathisia: implications for treatment. Journal Psychopharmacology 1998; 12: 192-214.4.
Gerber PE, Lynd LD. Selective serotonin-reuptake inhibitor-induced movement disorders. Annals Pharmacotherapy 1998; 32: 692-698 5. Rosen RC, Lane RM, Menza M. Effects of SSRIs on sexual function: a critical review. Journal Clinical Psychopharmacology 1999; 19: 67-85.6. Woodrum ST & Brown CS: Management of SSRI-induced sexual dysfunction. Annals of Pharmacotherapy 1998; 32: 1209-15.
7. Montejo-Gonzales AL, Llorca G, Izquierdo JA, Ledesma A, Bousono M, Calcedo A, Carrasco JL, Ciudad J, Daniel E, De La Gandara J, Derecho J, Franco M, Gomez MJ, Macias JA, Martin T, Perez V, Sanchez JM, Viciens E. SSRI-induced sexual dysfunction: fluoxetine, paroxetine, sertraline, and fluvoxamine in a prospective, multicentre, and descriptive clinical study of 344 patients. Journal Sexual Marital Therapy 1997; 23: 145-52.
8. Walker PW, Cole JO, Gardner EA, Hughes AR, Johnston JA, Batey SR, Lineberry CG. Improvement in fluoxetine-associated sexual dysfunction in patients switched to bupropion. Journal Clinical Psychiatry 1993; 54: 459-65.
9. Medawar C. The antidepressant web. Marketing depression and making medicines work. International Journal Risk & Safey Medicine 1997; 10: 75-126.
10. Price JS, Waller PC, Wood SM, Mackay AVP. A comparison of the post-marketing safety of four selective serotonin re-uptake inhibitors, including the investigation of symptoms occurring on withdrawal. British Journal Clinical Pharmacology 1996; 42: 757-763.
11. Stahl MMS, Lindquist M, Petterson M, Edwards IR, Sanderson JH, Taylor NFA, Fletcher AP, Schou JS. Withdrawal reactions with selective serotonin re-uptake inhibitors as reported to the WHO system. European Journal Clinical Pharmacology 1997; 53: 163-169.
12. Blomgren SL, Krebs W, Wilson M, Ascroft R, Fave M, Rosembaum J. SSRI dose interruption study: interim data. American Psychiatric Association, 150th Annual Meeting: NR 188.
13. Zajecka J, Fawcett J, Amsterdam J, Quitkin F, Reimherr F, Rosenbaum J, Michelson D, Beasley C. Safety of abrupt discontinuation of fluoxetine: a randomised, placebo controlled study. Journal Clinical Psychopharmacology 1998; 18: 193-197.
14. Haddad P, Lejoyeux M, Young A. Antidepressant discontinuation reactions are preventable and simple to treat. British Medical Journal 1998; 316: 1105-1106.
15. Committee of Safety of Medicines. Dystonias and withdrawal symptoms with paroxetine. Current Problems Pharmacovigilance 1993; 19: 1.
16. Barbui C. La gestione dell’insonnia nella medicina di base. Informazioni Sui Farmaci 1998; 22: 91-97.
17. Rawlins MD. Spontaneous reporting of adverse drug reactions I: The data, II: Uses. British Journal Clinical Pharmacology 1988; 26:1-11.
18. Ottervanger JP, Valkenburg HA, Grobbe DE, Stricker BHC. Differences in perceived and presented adverse drug reactions in general practice. Journal Clinical Epidemiology 1998; 51: 795-799.
19. Pastuszak A, Schick-Boschetto B, Zuber C. Pregnancy outcome following first-trimester exposure to fluoxetine. Journal American Medical Association 1993; 269: 2246-8.
20. Tognoni G. Servizi: tra valutazione ed epidemiologia dei diritti. Rivista Sperimentale di Freniatria 1998; CXXII: 145-154.
21. Campomori A, Turconi R, Addis A, Barbui C, Terzian E, Castoldi T, Miglio D, Bonati M. Uso di psicofarmaci nelle donne in età fertile che accedono ai servizi psichiatrici territoriali. Rivista Sperimentale di Freniatria 1998; CXXII: 252-260.
|
Back to 4.2 |