Clinical Psychopharmacology Seminar
Original Author: Paul Perry, Ph.D, BCPP
Latest Revisers: Paul Perry, Ph.D, BCPP, Brian C. Lund,
Pharm.D.
Creation Date: 1996
Last Revision Date: June 2004
Peer Review Status: Internally Peer Reviewed
Among the SSRIs, there are more similarities than differences. Nevertheless, there are differences between SSRIs that could be clinically significant. The objective of this review is to describe the significant differences between the currently available SSRIs and to identify their role in the treatment of psychiatric disorders.
Pharmacology
As the name implies, the primary pharmacologic effect of the SSRIs is to block the presynaptic serotonin transporter receptor (Grimsley et al 1992). Among the SSRIs, there is variability in the potency of their serotonergic effects (Johnson 1991). Clinically, however, these variations have no known effect on efficacy or adverse effects. In contrast to tricyclic antidepressants (TCAs), SSRIs are more potent inhibitors of serotonin reuptake, and they have less of an effect on a1, a2, histaminic, and muscarinic receptors. These disparities may explain adverse effect profile differences between TCAs and SSRIs. The specific relationship between the pharmacologic and clinical effects of SSRIs seen in psychiatric illnesses is unknown, although the serotonergic effect is thought to be fundamental to their clinical action (Grimsley et al 1992). From animal studies, it is known that somatodendritic autoreceptor desensitization occurs over 14 days with a net result of increased serotonin release and transmission. This 14-day delay may, in part, explain the delayed onset of action of SSRIs in the treatment of depression and other psychiatric disorders.
Stahl (2000) proposes three neurotransmitter deficiency syndromes that are associated with depressed mood. These include a serotonin, a norepinephrine, and a dopamine deficiency syndrome. Additionally, the 5 available SSRI differ from each other in that some affect not only the serotonin ((5HT) presynaptic transporter but also to a lesser degree the norepinephrine and/or the dopamine transporters. Based on these observations, the selection of an SSRI can theoretically be based on the symptom presentation of the patient. Table 1 summarizes these observations.
|
Table 1. Neurotransmitter deficiency syndromes and SSRIs that affect the deficiency syndrome specific neurotrasmitter presynaptic reuptake (transporter) receptor. |
||
|
Deficiency Syndrome |
Symptoms |
SSRI |
|
Serotonin |
|
|
|
Norepinephrine |
|
|
|
Dopamine |
|
|
Pharmacokinetics
The SSRIs share a number of pharmacokinetic characteristics (Sommi et al 1987, Kaye et al 1989, Ronfeld et al 1988). They are well absorbed orally. While the presence of food in the stomach slightly alters the absorption of some SSRIs, this effect has no known clinical importance (Grimsley et al 1992). The SSRIs are highly lipophilic and are widely distributed and are highly protein plasma protein bound. All of the SSRIs are extensively metabolized to polar metabolites, which are then excreted in urine and feces (Sommi et al 1987, Kaye et al 1989, Ronfeld et al 1988). There is no evidence that serum drug monitoring of SSRIs is a useful strategy to predict response (Grimsley et al 1992).
Despite these similarities, some of the key differences among SSRIs are due to differences in their pharmacokinetic properties (Table 2). Fluoxetine is unique because of its long half-life and the long half-life of its active metabolite norfluoxetine (Sommi et al 1987). Although sertraline also has an active metabolite, it is 10 times less potent than sertraline and is probably not clinically relevant (Ronfeld et al 1988). Fluvoxamine and paroxetine have no active metabolites (Kaye et al 1989, Perucca et al 1994). Because of the differences in half-lives and activities of metabolites, a much longer washout period is necessary when switching from fluoxetine (long acting SSRI) to a monoamine oxidase inhibitor than it is with paroxetine or sertraline (both short acting SSRIs). These differences can cause considerable therapeutic delays in treatment-refractory patients.
Depression. A meta-analysis of antidepressant efficacy studies calculated SSRI response rates using an intent-to-treat sample (Depression Guideline Panel 1993b). The response rates were 54% and 47% for inpatients and outpatients, respectively and were similar to the TCA, MAOI and heterocyclic antidepressant response rates. The inpatient response rate was 26% better than placebo while the outpatient response rate was 20% better than placebo. There are few published studies of SSRIs in inpatient depression or treatment refractory-depression. Compared to TCAs, there are relatively few studies using SSRIs to prevent recurrence of depression, although the few existing studies suggest that SSRIs will prevent relapse for up to one year. Summaries of the available studies of these less studied populations are listed below.
Although in the past, the tricyclic antidepressants (TCAs) were considered the drugs of choice for the treatment of non-psychotic unipolar depression, the cardiotoxicity associated with tricyclic overdoses in conjunction with the high rate of suicide and suicide attempts associated with depression renders these agents less than desirable in many cases. Anderson (2000) performed a meta-analysis that contrasted the efficacy of the SSRIs to the TCAs. Overall, there was no difference in the efficacy of the two antidepressant classes. However, the TCAs were more effective in depressed inpatients. On the other hand the SSRIs were better tolerated. The discontinuation rates because of adverse effects were less with the SSRIs (RR=0.73, 95% CI (0.67,0.80). Another way to interpret this finding is that the number of patients needed to be treated with a SSRI to avoid a TCA discontinuation because of side effects is 33. Anderson (2001) reviewed 108 meta-analyses of the use of antidepressants in depressive disorders. The review yielded a number of useful generalizations regarding the antidepressant drugs. Findings that can be taken with a high level of confidence include the following: 1) there is little difference in efficacy between most new (post-1980) and older TCA and monoamine oxidase inhibitor (MAOI) antidepressants; 2) the serotonin and noradrenaline re-uptake inhibitors (SNRIs) are superior in efficacy to the SSRIs; 3) fluoxetine has a slower onset of therapeutic action than the other SSRIs; 4) the different antidepressant class adverse effect profiles makes the SSRIs more tolerable than the TCAs; 5) fluvoxamine is the least tolerable SSRIs; and 6) there is no increased the risk of suicidal acts or ideation in fluoxetine compared with TCAs (or placebo) in low-risk patients. Findings that can be taken with a lower level of confidence include the following: 1) there is greater efficacy of TCAs than SSRIs among in-patients; 2) the greater efficacy of amitriptyline than SSRIs; 3) the better tolerability of the MAOI, moclobemide (not available in the US) than TCAs; 4) there is no demonstrable difference in tolerability between SSRIs and TCAs in the elderly; 5) there is no better tolerability of fluvoxamine than TCAs; 6) better tolerability of dothiepin (not available in the US) than SSRIs; 7) better tolerability of sertraline and; 8) greater frequency of agitation on fluoxetine than other SSRIs in a within group comparison. Generalizing from mostly short-term randomized controlled studies to clinical practice requires caution.
Thase et al (2001) surveyed data from eight randomized controlled trials contrasting venlafaxine (n=851), SSRIs (fluoxetine, paroxetine, fluvoxamine, n=748), and placebo (4 studies, n=446) in order to compare remission rates (HAMD ² 7). Remission rates were 45% (382/851) for venlafaxine, 35% for the SSRIs (260/748), and 25% (110/446) for placebo. The odds ratio for remission was was 1.5 times greater for venlafaxine versus the SSRIs. The difference in the remission rates was apparent between venlafaxine and the SSRIs at 2 weeks of treatment while the difference between SSRIs and placebo was apparent after 4 weeks of treatment.
Melancholia. There is now substantial evidence that suggests that SSRIs are less effective than TCAs in the treatment of the melancholic subtype of major depressive disorder (see Table 6) (Danish University Antidepressant Group 1990, Heiligenstein et al 1994, Peselow et al 1992, Roose et al 1994, Tignol et al 1992). In general, the relative efficacy of the SSRIs in melancholia is dependent on the definition of response used in each study. In studies that defined response as a 50% reduction in psychometric rating (usually the HAM-D scale), the SSRIs generally appear equivalent to TCAs and more effective than placebo. This is misleading because 50% reduction may indicate improvement, but the patient may not yet be euthymic. Euthymia is the goal of treatment of melancholia. Among studies using a 50% symptom reduction as the definition of response, the response rates ranged from 25-41% with placebo, 44 to 67% with SSRIs, and 59-70% with TCAs (Heiligenstein et al 1994, Peselow et al 1992, Tignol et al 1992). In the Peselow et al study, the fluoxetine and imipramine response rates were 56% and 61%, respectively, while the placebo response rate was 23%. From these three studies, the SSRIs seem to be more effective than placebo and roughly equivalent to TCAs. However, in the three studies that set an arbitrary threshold psychometric score as the definition of response (e.g. final HAM-D score less than 8 or 10), the efficacy of the SSRIs in the treatment of melancholia is less clear (Danish University Antidepressant Group 1990, Roose et al 1994, Tignol et al 1992). Using this definition of response is a better measure of response because a final HAM-D score less than 8 or 10 is a reasonable estimate of euthymia, the goal of treatment. In these studies, the placebo response rate was 15% (one study), the SSRI response rates ranged from 8-30%, and the TCA response rate was 57-63%. From these studies it appears that TCAs may produce a more robust response than SSRIs in melancholic depression.
In a multicenter trial, 68 inpatients received either venlafaxine 200 mg/day or fluoxetine 20 mg/day for a maximum of 42 days, with 6 days of tapering. A response was defined as a decrease in the MADRS or HAM-D total score of 50% or more from baseline. At week 4, 76 vs 47% of patients responded according to the MADRS and 76 vs 41% responded on the HAM-D for venlafaxine and fluoxetine, respectively (significant). At week 6 the MADRS 70 vs 50% showed a response and 73 vs 50% a response on the HAM-D for venlafaxine vs. fluoxetine, respectively (not significant). A similar pattern was observed for the CGI ratings (Clerc et al 1994). Based on limited data it appears that venlafaxine may also be superior to SSRIs for melancholic depression.
Inpatient Depression. There is one published study that compares imipramine with fluoxetine in the treatment of non-melancholic inpatient depression (Beasley et al 1993). A baseline HAM-D score > 19 was required for inclusion in the study and psychotic patients were excluded. In this double-blind comparison study of 118 patients with DSM-III major depressive disorder, fluoxetine was titrated to a median dose of 80 mg/day in one group and imipramine was titrated to a median dose of 200 mg/day in the other group. The study was 6 weeks long. There was no difference in baseline or final HAM-D scores between the two groups. "Response" was defined as a 50% decrease in HAM-D, and "remission" was defined as a final HAM-D less than 8. Response rates and remission rates did not differ between treatment groups. Among fluoxetine patients, 54.5% were responders and 21.2% achieved remission. Among imipramine patients, 60% were responders and 34.3% achieved remission. Imipramine caused significantly more complaints of dry mouth and sweating. There was no difference between groups for other adverse effects.
Psychotic Depression. There is one open study that suggests that a combination of fluoxetine plus and antipsychotic (perphenazine in this case) may be an effective approach to treating patients with DSM-III-R major depressive disorder with psychotic features (Rothschild et al 1993). In this study, 73% of the 30 patients responded (defined as a 50% decrease in HAM-D and BPRS scores and a final HAM-D < 12). Eight patients were also given lithium augmentation during the study.
In a controlled trial, Zanardi et al (1996) titrated 46 unipolar and bipolar delusional depressed patients to doses of sertraline 150 mg/d and paroxetine 50 mg/d over 8 days. Nine paroxetine but no sertraline patients dropped out because of adverse effects that were characterized by the jitteriness syndrome. None of the patients received antipsychotics. The intent-to-treat analysis defined response as a final HAMD score of less than 8. The response rate in the unipolar patients favored sertraline (13/18, 72%) over paroxetine (3/14, 21%). However, although the trend is obvious the difference in the bipolar depressed patients was not significant, i.e., sertraline (5/6, 83% versus paroxetine (3/8, 38%). All of the bipolar patients were receiving lithium augmentation. These are impressive response rates in this typically difficult to treat population.
Atypical Depression. There are four published studies that examine the efficacy of SSRIs in the treatment of "atypical" depression (i.e. depressed patients with unusual symptoms such as hypersomnia, weight gain, interpersonal rejection sensitivity, leaden paralysis) (Reimherr et al 1984, Stratta et al 1991, Pande et al 1996). Reimherr et al (1984) in a controlled study treated a group of atypical depressives with either imipramine 150-300 mg/d or fluoxetine 40-80 mg/d for 4 weeks. Sixty-five percent (13/20) of the fluoxetine treated patients while only 13% (1/8) of the imipramine patients experienced a moderate or marked improvement in their depression. In the Stratta (1991) study, patients were given either fluoxetine 20 mg/d or imipramine 125 mg/d for 5 weeks. Both groups exhibited similar response rates with HAM-D scores dropping over the course of the five-week study from approximately 15 to 7 in both groups. However, symptoms of anxiety, sleep, and interpersonal problems did better on fluoxetine. Pande et al (1996) compared the effectiveness of fluoxetine 20-60 mg/d to phenelzine 45-90 mg/d in 40 patients. The decrease in the HAMD, CGI and the remission rates (fluoxetine, 80% vs phenelzine, 85%) did not differ between treatments. Phenelzine caused more adverse effects. The primary ADRs observed were orthostatic hypotension, sedation, and asthenia. McGrath et al (2000) contrasted the effectiveness of fluoxetine 20-60 mg/d to imipramine 50-300 mg/d vs placebo in a 10-week trial of 154 patients with atypical depression. Based on the CGI-improvement scores of much or very much improved, both drugs were similarly more effective than placebo, i.e., fluoxetine (51%), imipramine (53%) and placebo (23%). Analysis of the HAMD scores yielded the same results. These findings differ from those of the older studies that suggested that TCAs were less effective than either MAOIs or SSRIs in atypical depression.
Bipolar Depression. MAOIs, TCAs, bupropion, and the SSRIs have been evaluated in the treatment of acutely ill bipolar depressed patients (Zornberg and Pope 1993, Gelenberg and Hopkins 1993, Srisurapanont et al 1995, Angst and Stabl 1992, Dantzler and Salzman 1995). SSRIs and MAOIs appeared to be more effective than the TCAs possibly because of the TCAs being more predisposed toward having patients "switching" to a manic state. Most recently Nemeroff et al (2001) randomized 117 outpatients with bipolar disorder, depressive phase to treatment with either imipramine (mean=167 mg/d), paroxetine (mean=33 mg/d) or placebo for 10 weeks. All patients were being treated with lithium at the time of the recurrence. The treatment groups were stratified according to lithium concentration, i.e., > 0.8 mEq/L versus ² 0.8 mEq/L. Overall, imipramine, paroxetine, and placebo were equally effective. However, they were more effective than placebo in the treatment groups with the subtherapeutic (² 0.8 mEq/L) lithium levels. It could be inferred from this finding the antidepressant effects of lithium were more prominent in patients with high or therapeutic serum levels.
Dysthymia. Thase et al (1996) randomized 410 dysthymic without major depression (DSM-III-R) patients to treatment with either imipramine (mean = 160 mg/d), sertraline (mean = 90 mg/d) or placebo. Seventy percent of the subjects had an Axis II personality disorder diagnosis. Dropouts occurred more often with the imipramine and placebo treatment groups than the sertraline treatment group. A full response, defined as having no dysthymia criteria and a rating of 0 on the depressed mood item of the HAMD, occurred more often in the sertraline (50%) and imipramine (44%) treatment groups than the placebo (28%) group. A HAMD of ² 4 also occurred more often in the sertraline (53%) and imipramine (56%) treatment groups than the placebo (37%) group. Despite there being no difference in effectiveness between the antidepressants, the threefold greater incidence (18% vs 6%) of adverse drug reaction dropouts would favor the use of the sertraline over imipramine. Ravindran et al (2000) replicated these findings in a randomized controlled 12-week trial of sertraline versus placebo in 310 early- and late-onset dysthymic patients without major depression. Dropouts occurred in similar number of patients, sertraline (25%) and placebo (23%). The mean sertraline dose was 128 mg/d although the modal dose was only 50 mg/d. Sertraline was more effective than placebo in producing remissions (HAMD ² 8), 34% versus 22%, and responses (> 50% HAMD decrease), 52% versus 34%. Both treatment groups had similar numbers of adverse effect complaints, 75% for sertraline and 65% for placebo.
Pindolol augmentation. Perez et al (1997) randomized 111 patients with major depression, unipolar subtype (DSM-IV), to treatment with either fluoxetine 20 mg/d alone or fluoxetine plus the beta-blocker, pindolol 7.5 mg/d for 6 weeks. Response rates (50% decrease in HAMD) favored the combination (75% vs 59%), sustained response (<10% change in HAMD positive change) (69% vs 48%), and remission rate (HAMD < 8) (60% vs 45%). Reanalysis of the data found that the combination produced the endpoint of a response within an average 19 days versus 29 days in the fluoxetine only treated patients (Perez et al 2001). Pindolol immediately blocks the presynapatic somatodendritic 5HT1A receptors, ('brake receptors') thereby increasing serotonin neuronal transmission at a earlier than would be normally be observed with an SSRI alone. These neuroreceptors do not require prolonged exposure to excessive amounts of serotonin for several weeks to be down regulated. This results in augmentation and acceleration of the antidepressant effect of the SSRIs. Bordet et al (1998) demonstrated the accelerated antidepressant response. Paroxetine 20 mg/d was administered to 100 patients with major depression (DSM-IV). Half the patients received pindolol 5mg po tid while the other half were randomized to placebo. After 10 days of treatment the response rate (HAMD < 11) favored the combination therapy. However, this difference was not apparent at study days 15, 21, 25, and 28. Contrary findings were generated by Berman et al (1999) who contrasted the effectiveness of fluoxetine 20 mg/d versus fluoxetine 20 mg/d plus pindolol 7.5-10 mg/d over 6 weeks in 86 patients with major depression. At days 14, 28, and 42 were was no significant difference in the partial (HAMD decrease > 50%, final <16) or full remssion (HAMD decrease > 50%, final <11) rates. Fifty percent of the patients had a chronic major depression diagnosis. The authors hypothesized that this demographic may have adversely affected the outcome of the study.
Other Indications. In addition to the affective and anxiety disorders, the SSRIs have been used with some success to treat a variety of other psychiatric and non-psychiatric disorders (Boyer 1992, Adly et al 1992). Promising indications include bulimia, anorexia nervosa, substance abuse disorders, borderline personality disorder, and premenstrual tension (Boyer 1992). A 2-month controlled trial comparing fluoxetine to placebo in the treatment of 50 withdrawing alcoholics demonstrated that although the anxiety and depression scores change equally between the treatment groups, there was greater abstinence rate in the fluoxetine group (62%) than the placebo group (35%) (Janiri et al 1996). The common SSRI side effect of decreased appetite and subsequent weight loss appears to be most pronounced in obese patients and may be a useful effect as an adjunct to diet and exercise in cases of severe obesity (Boyer 1992). During a four-week double-blind trial, fluoxetine 20 mg significantly reduced pain intensity, as did amitriptyline 25 mg, compared to placebo in treating chronic rheumatic pain. At the end of the fourth week, fluoxetine was superior in efficacy to amitriptyline (Rani 1996). Fluoxetine was found to prevent migraine headache in a small double-blind, placebo-controlled study (Adly et al 1992). Paroxetine (mean = 31 mg/d) was found to be more effective than placebo in preventing interferon alfa-induced major depression in 40 patients being treated for malignant melanoma who were eligible for high-dose therapy. Major depression developed in over the 12 weeks of treatment in 11% (2/18) of the paroxetine randomized patients while it occurred more often in 45% (9/20) of the placebo treated patients. The interferon therapy was discontinued because of the depression in one patient on paroxetine but 7 subjects receiving the placebo. Impressively, the final HAMD scores were 15.2 in the placebo group in contrast to 8.4 in the paroxetine group. Continued research of SSRIs in the treatment of these and other disorders may lead to the discovery of additional uses.
Dosage
Depression. The SSRIs are expensive, and it has been common to have non-compliant patients admitted to the hospital due to relapse because they could not afford their medications. However, fluoxetine (Prozac) will be going off patent in February of 2001, although generic fluoxetine will not be available reportedly until October of 2001. Persuading the patient to take his or her medication as prescribed is extremely important in potentially suicidal patients with depression. Because of this, it is exceedingly important that patients receive the lowest effective (and thereby the most cost effective) dose of any drug they are prescribed.
For citalopram, an initial dose of 20 mg once daily, with an increase to 40 mg/day is generally done. Dose increases should usually occur in increments of 20 mg at intervals of no less than one week. Although certain patients may require a dose of 60 mg/day, this has not been demonstrated to be advantageous. Citalopram should be administered once daily, in the morning or evening, with or without food (Citalopram package insert 1998).
For fluoxetine, it is well established that doses higher than 20 mg per day do not produce a better therapeutic response in most patients (Schweizer et al 1990). Lower doses have also been studied. A single, six-week, double-blind multicenter study that compared placebo and fluoxetine at 5, 20, or 40 mg per day in 354 patients failed to find any substantial difference in response between the three active treatment groups (Wernicke et al 1988). Because of fluoxetine and its primary metabolite, norfluoxetine, having extremely long half-lives, it is observed that a dose of 20 mg every other day produces an effect similar to 10 mg per day. This dose should be seriously considered in the elderly, who tend to require lower doses for all medications, and in patients who have trouble paying for their treatment. For even lower doses, fluoxetine is also available as an oral solution and as a 10 mg capsule. Doses higher than 20 mg per day should only be used in patients who have had an adequate therapeutic trial at lower doses.
The starting dose for fluvoxamine is 50 mg/day (Wilde et al 1993). Titration to doses greater than 50 mg/day should be done slowly to prevent nausea. There is little evidence that total daily doses greater than 150 mg provide added efficacy. A lower fluvoxamine dose of 50-150 mg/day (mean 92.7±46.9 mg) was shown to be as effective as imipramine 80-240 mg/day (mean 136.4±71.7 mg) vs. placebo in 150 outpatients in lowering HAM-D scores from baseline (Claghorn et al 1996). In a trial demonstrating equal efficacy in reducing depressive symptoms mean doses at the end of 7 weeks were 101.85±25.22 vs. 34.17±18.84 for fluvoxamine and fluoxetine, respectively (Rapaport et al 1996). Fluvoxamine should be administered in single daily dose at bedtime at doses less than 150 mg/day. It is recommended that fluvoxamine be administered in divided doses if doses greater than 100 mg/day are necessary.
The established minimum effective dose of paroxetine is 20 mg per day (Grimsley et al 1992). Doses higher than 20 mg per day have not been documented to improve percentage of response or response rate (Grimsley et al 1992). Higher doses should be reserved for patients who have failed a three to four week trial at 20 mg per day. A starting dose of 10 mg per day is recommended in the elderly and in patients with significant hepatic or renal disease.
The minimum effective dose of sertraline is not as apparent because the data have not been extensively published. Summaries of the pre-marketing studies presented by the manufacturer for FDA approval are available and they indicate that the efficacy of sertraline at a dose of 50 mg per day is comparable to higher doses (Lapierre et al 1991). However, physician prescribing data suggests an average dose of approximately 100 mg/d.
Speed of Response is also an issue in the treatment of depression. It has been asserted that SSRIs do not work as fast as TCAs. There are two explanations for this. First, when new psychotropic drugs are marketed, they are often reserved for more difficult or treatment-refractory patients. By definition, these patients are less likely to respond. Second, one study suggests that a 6-8 week trial is necessary to evaluate the efficacy of fluoxetine in a given patient (Schweizer et al 1990). However, a more definitive report evaluated all of the available comparisons between fluoxetine and other antidepressants (mostly TCAs in studies of outpatients) (George et al 1991). In this report, no difference in the time to onset of action was found between fluoxetine and standard antidepressants. There is no reason to believe that other SSRIs are any different from fluoxetine in this regard.
In the Iowa medicaid population, data collected in the fall of 1996 represent 36% of the population taking sertraline, vs. 37.7% on fluoxetine and 26.3% on paroxetine. The average doses for each SSRI in this population are 102 mg sertraline, 28 mg fluoxetine and 26 mg paroxetine.
Adverse Effects
SSRIs are reported to have fewer side effects than TCAs. This assertion is misleading because nearly every published study comparing an SSRI with a TCA uses one of the more poorly tolerated TCAs (e.g., amitriptyline, imipramine, doxepin, or clomipramine) as the comparison drug (Grimsley et al 1992, Sommi et al 1987). It is not surprising that these TCAs, which have strong anticholinergic properties, caused more adverse effects than the SSRIs. There are now two studies comparing an SSRI (fluoxetine) with nortriptyline (Fabre et al 1991) or desipramine (Bowden et al 1993), two of the better tolerated TCAs. In the Fabre et al study (1991), although significantly more patients in the nortriptyline group dropped out of the study because of adverse effects, there were few differences in frequency of adverse effects between the two groups (based on analysis of all patients in the study, including dropouts). The nortriptyline group experienced more complaints of dry mouth, while the fluoxetine group had more reports of insomnia. These results suggest that, if fluoxetine is truly better tolerated than nortriptyline, the difference is very small. In the Bowden et al study (1993), the desipramine treatment group experienced more orthostatic hypotension, flushing, constipation, dry mouth, and dizziness. The fluoxetine group had more complaints of diarrhea, and anxiety. Dropout rates did not differ between groups. Finally, Trindade et al (1998) conducted a meta-analysis of the adverse effect rates among 84 randomized controlled trials involving SSRIs and TCAs. There were no differences in dropout rates between the two classes of antidepressants due to either lack of efficacy, worsening of symptoms, or adverse effects. Thus similar compliance could be assumed between the two classes. ADRs observed in ³ 5% of patients included nausea (14%), diarrhea (12%), anxiety (6%), agitation, insomnia (5%) for the SSRIs and constipation (11%), dizziness (9%), hypotension (6%) and dry mouth (5%), and blurred vision (5%) for the TCAs. These findings were replicated by Thompson et al (2000). They compared the compliance with fluoxetine to that with the TCA, dothiepin, in a randomized open 12-week trial of outpatients diagnosed with major depression. The study found no differences in compliance between the two antidepressants despite the varying adverse effect profiles for the drugs.
A meta-analysis of 42 published randomized controlled studies compared the discontinuation rates of the tricyclics and the SSRIs due to side effects. Significantly fewer patients discontinued SSRIs than TCAs, 15% to 19% in the head-to-head studies and 19 to 27% in the placebo-controlled studies, respectively (Montgomery et al 1994). However, a subsequent meta-analysis of the discontinuation data concluded that this is only true of the older TCAs, amitriptyline and imipramine and not nortriptyline, desipramine, doxepin, and clomipramine (Hotopf et al 1997).
Among the SSRIs, there are few differences in adverse effects, and most differences are minor (see Table 4) (SmithKline Beecham 1992, Roerig 1992, Dista Products Company 1992, Solvay Pharmaceuticals 1994). The reported differences may be due to chance or differences in study design. Until there are objective comparisons between SSRIs, their adverse effect profiles should be considered roughly equivalent. In a post-marketing study in which patients who received fluoxetine (1577 patients) or sertraline (1209 patients) were asked to report any new or unusual symptoms while on either of the two SSRIs. The following side effects were reported more frequently for sertraline (mean dose 73.5 mg) vs. fluoxetine (mean dose 24.9 mg); urinary frequency, excited/overactive, daytime sedation, dizzy/faint/light-headed, shakiness/tremors, diarrhea, nausea (Fisher et al 1995).
Sexual Function. The effect of SSRIs on sexual function was originally very poorly studied and characterized. The manufacturers' literature underestimates the nature of this effect because of the voluntary nature of the adverse effect reporting system used in early clinical trials. From the manufacturers' data, estimates of the incidence of sexual dysfunction resulting from SSRIs ranges from 1.9-15.9% (Table 4). Although this appears to be a wide range, it is likely that these apparent differences among SSRIs are probably a result differing methodology among studies over the past decade. The basic pharmacologic similarities among these drugs suggest that the effects on sexual function should be similar for each agent. The incidence of SSRI-induced sexual dysfunction is higher in reports that specifically ask the patients about their sexual function. In these reports, the incidence of sexual dysfunction ranged from 34-75% (Jacobson 1992, Patterson 1993). In the post-marketing surveillance done by Fisher et al (1995), orgasm difficulties and impotence occurred more frequently in the sertraline group compared to the fluoxetine group. In patients receiving mean doses of sertraline 148 mg and nefazodone 456 mg, negative effects on sexual function and satisfaction in both men and women were observed with sertraline but not with nefazodone. The doses used in this study may not have been comparable; a more common dose of sertraline to compare to nefazodone would have been 100 mg (Feiger et al 1996). Please see nefazodone chapter for a more detailed discussion of this trial.
Sexual dysfunction has been variously reported as either ejaculatory incompetence, ejaculatory retardation, decreased libido, anorgasmia, or inability to obtain or maintain an erection. Furthermore, sexual dysfunction was reported in both men and women. In a retrospective chart review done in 110 women placed on an SSRI (fluoxetine, paroxetine or sertraline), 35 (32%) reported sexual adverse effects. These patients had either orgasmic disturbances or loss of or decreased libido, or both, as the initial sexual adverse effect (Shen and Hsu 1995). There was no one SSRI more likely to cause the reported sexual adverse effects than another.
Forty-two outpatients diagnosed with depression with or without comorbid obsessive-compulsive disorder, who received fluoxetine 20-60 mg/day, sertraline 50-200 mg/day or paroxetine 20 mg/day for 8 weeks were assessed by utilizing the Rush Sexual Inventory (RSI) (Zajecka et al 1997). This is a comprehensive, succinct, patient-rated scale designed to provide sexual function and satisfaction information. Males (n=20) reported treatment emergent sexual dysfunction at a rate of 60%, whereas females (n=22) reported sexual dysfunction at a rate of 57%. There was no significant variation in the reporting of changes in sexual function/satisfaction between responders and nonresponders as assessed by the use of the Hamilton Depression Scale. In males who received fluoxetine, there was a significant increase in the frequency of desires to initiate sexual activity, increase frequency of initiating sexual activity, and an increase in overall degree of sexual satisfaction attained. In females that received fluoxetine, significant increases in frequency of pleasurable sexual thoughts and frequency of desire to initiate sexual activity were seen. No statistically significant variation in incidence of sexual function/satisfaction within the female population was seen between the 3 treatment groups. Sertraline and paroxetine were associated with a statistically significant decrease in frequency of pleasurable sexual thoughts and decrease in frequency of initiating sexual activity compared with baseline measures. Overall, this report indicates that there may be differences in the sexual effects of the individual SSRIs. Of the three SSRIs studied, sertraline had the greatest negative impact on sexual function, whereas fluoxetine was associated with improvements in several aspects of sexual functioning. Unfortunately this study was not controlled or randomized and it is unknown if there were underlying differences among treatment groups that lead to the observed differences in sexual functioning. Further randomized and controlled studies are needed to establish differences in sexual effects between SSRIs.
It is important to remember that SSRI-induced changes in sexual function can actually be beneficial in some clinical situations (Lee et al 1996). In an eight-week open-label study involving eleven males with DSM-III-R criteria for premature ejaculation or ejaculation rapidly before or immediately after vaginal intromission during sexual intercourse received 20 mg/day of fluoxetine for 1-2 weeks, titrating to 60 mg/day. The Yonsei Sexual Function Inventory-II (YSFI-II) scale showed that sexual desire, performance anxiety for rapid ejaculation, patient's or partner's satisfaction with ejaculation, and overall sexual functions revealed a significant improvement difference between pre- and post-treatment within the subjects (Lee et al 1996).
Some evidence suggests that SSRI-induced sexual dysfunction is dose related and may be treated by simply lowering the dose of the offending agent. In one report 30/30 patients who opted to change their daily fluoxetine dose from 20 mg daily to 20 mg every other day reported improved sexual function (Patterson 1993). Furthermore, 15/15 patients who discontinued fluoxetine reported reversal of their symptoms. In the product literature for fluvoxamine, the incidence of abnormal ejaculation is 8% among OCD and depressed patients combined (Solvay Pharmaceuticals 1994). However, among the OCD patients alone, the incidence of 17.9%. Since OCD patients tend receive higher SSRI doses, this also data supports the dose-related effect of SSRIs on sexual function. In patients who cannot have their SSRI dosage reduced, another option is to simply wait and reassess sexual function after several months. While there are cases in whom sexual dysfunction was transient, most cases will continue to have some dysfunction (Herman et al 1990, Feder 1991).
If the above measures are ineffective in managing SSRI-induced sexual dysfunction, the next step is to consider an alternative antidepressant. In one study of 31 patients with sexual dysfunction secondary to fluoxetine, improvement was noted in 29 patients at the end of an eight-week trial of bupropion (Walker et al 1993). Twenty-five out of 31 (81%) patients reported their sexual functioning was "much or very much" improved on bupropion. Bupropion administration to normal healthy males does not affect sexual functioning or nocturnal erections (Labbate et al 2000). Additionally, citalopram (20-40 mg/d) substitution for fluoxetine (40 mg) and paroxetine (20 mg/d) was reported effective in two cases of sexual dysfunction (Pallanti and Koran 1999). Controlled studies are necessary to confirm these findings.
As a last resort, polypharmacy may be considered as a treatment for SSRI-induced sexual dysfunction. The addition of amantadine, cyproheptadine, yohimbine, or sildenafil has been anecdotally reported to be effective in some patients with SSRI-induced sexual dysfunction. Each of these treatments are poorly studied, but may be reasonable to try in selected patients.
Cyproheptadine, a serotonin antagonist, has been evaluated in two reports involving five patients (Feder 1991, McCormick et al 1990). In McCormick's report, one patient was given 8 mg of cyproheptadine 1.5 hours before intercourse resulting in successful ejaculation (1990). In McCormick's other case, cyproheptadine 16 mg/day was necessary to achieve normal sexual functioning. However, in three other cases reported by Feder (1991), cyproheptadine treatment of SSRI-induced sexual dysfunction was associated with a return of depressive symptoms, presumably because of interference with the serotonergic effect of the fluoxetine. In seven patients being treated with SSRIs for OCD (five had concurrent MDD) for at least ten weeks and complaining of dysfunction were given cyproheptadine 4 mg 1-2 hours before sexual activity, the dose was increased gradually to a maximum of 12 mg. Three of seven experienced improvement in sexual function, 2 had initial improvement which subsided after the third week and two patients did not show any improvement following cyproheptadine administration (no clinical deterioration in depressive symptoms were observed, although no psychometric rating scales were used) (Aizenberg et al 1995). As a result, cyproheptadine cannot be recommended as a first-line treatment of SSRI-induced sexual dysfunction.
Two anecdotal reports have suggested that amantadine may also be an effective treatment for SSRI-induced sexual dysfunction (Balogh et al 1992, Shrivastava et al 1995). Balogh et al noted improvement in 5/7 cases of fluoxetine-induced sexual dysfunction treated with amantadine 100-200 mg/day (1992). Response took 3-21 days. Two of the five cases that improved with amantadine were women with anorgasmia. One woman did not improve. In Shrivastava's report, 6/6 male patients with paroxetine-induced sexual dysfunction improved on amantadine. The patients in this study took amantadine only during the 48-hour period before intercourse at a dose of 200-400 mg/day.
Two open reports have suggested that yohimbine is an effective treatment for SSRI-induced sexual dysfunction. In the first report, 8/9 patients reported improved sexually function on yohimbine 5.4 mg three times daily (Jacobsen 1992). Five patients reported adverse effects due to the yohimbine including anxiety, insomnia, and nausea. Two patients stopped the yohimbine because of these adverse effects. The author suggested slowly titrating the yohimbine to prevent adverse effects. In a second report, Hollander and McCarley reported improved sexual function in 5/6 patients treated with yohimbine (1992). The patients in this report received 2.7-16.2 mg on a prn basis 2-4 hours before intercourse. They also reported that very small changes in dosage (0.25-0.5 tablets, or 1.4-2.7 mg) was sometimes the difference between efficacy and lack of efficacy or adverse effects and lack of adverse effects. Adverse effects of insomnia, sweating and shakiness were reported.
Sildenafil 50 or 100 mg has been reported either fully or partially effective in 6 of 7 males and 3 of 3 female patients having SSRI-induced sexual dysfunction problems. The female patients complained of a 24-hour period of adverse effects that included lethargy, headache, and flushing (Schaller and Behar 1999, Rosenberg 1999, Nurberg et al 1999, Gupta et al 1999).
Bleeding. Clinical depression has been identified as an independent risk factor for increased mortality in patients following acute coronary events. Platelet factor 4 (PF4) and beta-thromboglobulin (bHG) are two proteins released into the blood when platelets are activated. Their values double in patients with ischemic heart disease, especially after a myocardial infarction. Importantly, similar values are observed in patients with major depression who are free of cardiovascular disease. Platelet activation has been hypothesized but not proven to be a mechanism by which MDD becomes a risk factor for ischemic heart disease and cardiovascular disease and/or morbidity and mortality after MI (Musselman et al 1996). Serotonin potentiates platelet activation. SSRIs decrease serotonin uptake from blood to platelets. Since platelets do not synthesize serotonin, SSRIs are associated with increases in bleeding episodes. Pollock et al (2000) contrasted treatment of patients with nortriptyline (50-120 ng/ml) versus paroxetine 20 mg/d. At baseline the bHG and PF4 levels were elevated. However, after a week of treatment paroxetine decreased the levels significantly whereas nortriptyline has no effect on the platelet stickiness factors. These findings may explain why De Abajo et al (1999) observed a 3-fold increase in the relative risk (2.1-4.4, 95% CI) of bleeding episodes similar to ibuprofen in patients treated with clomipramine or the SSRIs. Similar bleeding risk was not associated with nortriptyline, protriptyline, desipramine, trimipramine, maprotiline, and amoxapine.
Withdrawal. It may be important to taper a patient off an SSRI prior to discontinuation to avoid withdrawal symptoms particularly if the patient is receiving one of the SSRIs with a shorter half-life and no active metabolites i.e., paroxetine and fluvoxamine. In a comparison study, those who received paroxetine, fluvoxamine, sertraline, and fluoxetine experienced a 14, 20, 2.2 and 0% respective frequency of withdrawal symptoms (Coupland et al 1996). Criteria for SSRI withdrawal include two or more of the following symptoms within one to seven days following discontinuation of a month or more of SSRI treatment: dizziness, light-headed, vertigo, paresthesias, anxiety, diarrhea, fatigue, gait instability, headache, insomnia, irritability, nausea, tremor, visual disturbances (Black et al 2000).
Violence. The SSRIs have been unfairly and unjustly accused of inducing homicidal and suicidal behavior. A significant amount of literature has examined this issue and shows that decreasing the serotonin in the brain results in aggressive behavior in both animals and man while increasing serotonin decreases aggressive behavior. The most common animal model for aggression is that of mouse-killing behavior, i.e, muricidality in rats. Such behavior is consistently decreased by the administration of drugs that increase serotonin levels in the brain. There are numerous animal studies that have demonstrated this finding (Berzsenyi P et al 1983, Molina et al 1987). Additionally, Salzman et al (1995) conducted a controlled trial contrasting the anti-aggression effects of fluoxetine to placebo. Twenty-one patients with mild to moderate borderline personality disorder were randomized to 12 weeks of treatment with either fluoxetine 20-60 mg/d or placebo. Measurements of anger were decreased in the fluoxetine treated subjects. This finding was independent of the change in the patients' mood.
Pregnancy. Pastuzak et al (1993) compared pregnancy outcome following first-trimester fluoxetine exposure with pregnancy outcome in two matched control groups. They prospectively collected and followed up 128 pregnant women exposed to a mean daily dose of 25.8 mg (+/- 13 mg) of fluoxetine during the first trimester and compared pregnancy outcome with two matched groups of women exposed during pregnancy to either nonteratogens or tricyclic antidepressants. Rates of major malformations were comparable within the three groups and did not exceed those expected in the general population. However, women treated with fluoxetine had a tendency for increased risk for miscarriage when compared with women exposed to nonteratogens (relative risk, 1.9; 95% confidence interval, 0.92 to 3.92). The rate of miscarriages in the fluoxetine group was comparable with the tricyclic group (13.5% and 12.2% vs 6.8% in the nonteratogens). The data suggest that the use of fluoxetine during embryogenesis is not associated with an increased risk of major malformations. Women exposed to both fluoxetine and tricyclic antidepressants tended to report higher rates of miscarriage.
Chambers et al (1996) in a non-randomized prospective trial of women who contacted by phone the California Teratogen Information Service contrasted the anomaly rates of 228 women exposed to fluoxetine during pregnancy to 254 controls. Exposed patients were further divided into two groups of early exposure (before week 24) and late exposure (week 24 or later exposure). Thirty percent of the exposed patients were taking other psychotropic drugs. Contrasting the fluoxetine group to controls, the rates of major structural anomalies were 5.5% versus 4.0% and the miscarriage rates were 10.5% versus 9.1%, neither of which were different. However 3 or more minor anomalies occurred more often in the fluoxetine group at 15.5% versus 6.5%. Additionally, in the late exposure fluoxetine group there was a higher rate of premature births, admissions to the neonatal ICU, and poor neonatal adaptation.
The interaction between monoamine oxidase inhibitors (MAOIs) and SSRIs is the most important drug interaction limiting SSRI use (see Table 5). This combination may lead to the development of a hyperserotonergic syndrome consisting of excitement, diaphoresis, rigidity, hyperthermia, tachycardia, hypertension, and possibly death. There are three reports of death in patients who received an MAOI subsequent to fluoxetine discontinuation (Ciraulo et al 1990a). The severity of this interaction necessitates a five-week washout when switching a patient from fluoxetine to an MAOI to allow complete elimination of the fluoxetine. A one- to two-week washout is recommended before starting an MAOI for patients receiving fluvoxamine, paroxetine, or sertraline. This difference in washout time between fluoxetine and fluvoxamine, paroxetine, or sertraline when switching from an SSRI to an MAOI is one of the key differences between SSRIs and should be remembered if an MAOI is planned as a possible subsequent treatment in the event of SSRI failure. However, the differences among SSRIs are not important in the case in which a patient is switched from an MAOI to an SSRI. In this case, a 10 to 14 day washout is necessary regardless of which SSRI is used to allow regeneration of monoamine oxidase.
Since all of the SSRIs are extensively metabolized in the liver, it is a possible that other drugs that inhibit or induce hepatic microsomal enzyme systems (e.g. the cytochrome P450 enzyme system) may alter SSRI plasma concentrations. However, the clinical significance of these possible interactions is doubtful since there is no known correlation between plasma concentration and therapeutic response for any of the SSRIs. If an interaction is suspected, the patient's SSRI dosage can be easily adjusted.
The SSRIs have variable effects on the pharmacokinetic or pharmacodynamic parameters of other drugs. All five SSRIs may affect the pharmacodynamic effect of warfarin. However, the changes appear to be clinically significant only for fluoxetine, fluvoxamine, and paroxetine (Grimsley et al 1992, Sommi et al 1987; Solvay Pharmaceuticals 1994). It is likely that the SSRIs either inhibit the metabolism of warfarin or displace warfarin from plasma protein binding sites. Close monitoring of prothrombin time and INR is necessary if these drugs are used together.
All five SSRIs may inhibit the metabolism of other hepatically metabolized drugs (Table 7) (DeVane 1994). Potential interactions exist between the various SSRIs and other frequently prescribed drugs including some antipsychotics, TCAs, some benzodiazepines, and phenytoin. Currently, the manufacturer of fluvoxamine (Solvay Pharmaceuticals) recommends avoiding concomitant administration of fluvoxamine with triazolam or alprazolam. Theophylline is a substrate for CYP1A2, and fluvoxamine has been shown in an in vitro study to have a significant inhibitory effect on CYP1A2 (Brosen et al 1993). It is recommended that doses of theophylline be reduced by one third if coadministered with fluvoxamine (Ereshefsky et al 1996). Fluvoxamine also inhibits metabolism of diazepam through cytochrome P450 2C9 (Solvay Pharmaceuticals 1994). Since both fluoxetine and sertraline may interfere with these enzymes, it possible that similar interactions may occur (DeVane 1994).
The SSRIs may inhibit the metabolism of the TCAs and carbamazepine (Grimsley et al 1992, SmithKline Beecham Pharmaceuticals 1992, Ciraulo 1990b, Solvay Pharmaceuticals 1994). If there are any differences among SSRIs in this regard, they have not yet been established. Although it has been suggested that fluoxetine may be more likely to interact with TCAs, at least one report has found a possible interaction between and sertraline and desipramine (Barros et al 1993). The interaction between TCAs and SSRIs is of particular importance because of the potential for the development of toxic TCA concentrations and subsequent adverse effects (Ciraulo et al 1990a). Fluoxetine inhibits the metabolism of clozapine and is reported to produce a mean 76% increase in clozapine plasma concentrations (Centorrino et al 1994). Fluoxetine inhibits hepatic oxidative enzyme such that haloperidol plasma levels increase by approximately 20% (Goff et al 1991a). Objective evidence suggests that the SSRIs do not potentiate the effects of alcohol (Grimsley et al 1992, Sommi et al 1987, Roerig 1992).
Safety in Overdose
The SSRIs are dramatically safer in acute overdose than the TCAs or MAOIs (Barbey and Roose 1998). Fatal SSRI overdoses have been reported, typically with extremely large doses (approximately 150 times the usual daily dose) or with co-ingestions of alcohol or other drugs. Ingestions of up to 30 times the normal dose are commonly associated with few or no symptoms. Larger doses in the range of 50-75 times the normal dose can result in drowsiness, tremor, nausea, and vomiting. More serious events including seizures, ECG changes and coma may occur as the magnitude of the overdose increases.
For a complete discussion of SSRI overdose, please refer to the antidepressant overdose chapter.
Cost
Given the small therapeutic differences among SSRIs, the choice of a particular agent may be dictated by cost. Amitriptyline and imipramine are very inexpensive and should be used in patients who have history of good tolerance and response to these drugs. However, for nortriptyline (a better tolerated TCA) there is no difference in cost compared to SSRIs.
It may be possible to decrease the patient's financial burden and, therefore, improve compliance, by using fluoxetine 20 mg every other day. This approach may be justified because of fluoxetine's long half-life and the fact that doses less than 20 mg per day have been shown to be effective in the treatment of depression. Likewise, the cost of sertraline can be halved by using 100 mg half tablets to provide a daily dose of 50 mg per day in those patients who respond to this dose. In patients who require a high daily dose of their SSRI (e.g. in OCD), fluvoxamine, paroxetine, and sertraline are comparable in price, and both are substantially less expensive than similar doses of fluoxetine. The cost of drug therapy can adversely affect compliance and should be strongly considered in patients who may have difficulty paying for their medications.
The Iowa Medicaid Drug Utilization Review Commission reported that in fiscal year 2001 there were 197,698 SSRI prescriptions for Celexa (16.7%), Luvox (2.7%), Paxil (29.5%), Prozac (23.9%), and Zoloft (27.3%). The estimated mean dosages for these drugs were Celexa (25 mg/d), Luvox (139 mg/d), Paxil (22 mg/d), Prozac (22 mg/d), and Zoloft (82 mg/d). The mean cost of these prescriptions was Celexa ($63.67), Luvox ($144.39), Paxil ($72.79), Prozac ($108.26), and Zoloft ($70.88).
Conclusion
The SSRIs represent an advance in the area of psychotropic drug development. They are proven treatments for depression, OCD, and panic disorder and have been found to be helpful in a variety of other conditions, as well. Perhaps the most substantial benefit of the SSRIs is substantially less toxicity in overdose than both the MAOIs and the TCAs. The adverse effect profile of the SSRIs differs from the TCAs and is slightly better tolerated. Although the SSRIs have become the most commonly prescribed drugs for depression, there are clinical situations in which TCAs may be more appropriate. In melancholic depression for example, SSRIs and TCAs have a similar likelihood of producing a 50% reduction in symptoms. However, numerous trials have found that TCAs are significantly better at producing remission.
Differences among SSRIs are of limited significance for most patients. One difference is the very long effective half-life of fluoxetine compared to the other agents. Additionally, there are clinically meaningful differences in CYP450 enzyme inhibition. Though studies claim that one SSRI is more cost-effective than another, these studies are biased and there is probably not a consistent and meaningful difference.
|
Table 2. Selected Pharmacokinetic Variables |
||||||
|
Drug |
Protein Binding |
Half-life |
Active Metabolite? |
Time to steady state |
Effect of renal disease |
Effect of liver disease |
|
Citalopram5 |
80% |
35 hours |
no |
1 week |
No change |
May decrease clearance (therefore decrease dose) |
|
Fluoxetine1 |
94% |
2-4 days |
Yes (t1/2=7-15 days) |
1-2 months |
No change |
decrease clearance (therefore decrease dose) |
|
Fluvoxamine4 |
77% |
ª 20 hours |
No |
7 days |
No known change |
May decrease clearance (therefore decrease dose) |
|
Paroxetine2 |
95 % |
ª 20 hours (highly variable) |
No |
7-14 days |
May Ø clearance (therefore Ø dose) |
May decrease clearance (therefore decrease dose) |
|
Sertraline3 |
99% |
ª 26 hours |
Yes, but not significant |
7-14 days |
Unknown |
Unknown |
1Sommi et al 1987; 2Kaye et al 1989; 3Ronfeld et al 1988; 4Perucca et al 1994, 5Citalopram package insert
|
Table 3. Typical SSRI Dosages |
|||
|
Drug |
Major Depressive Disorder |
Obsessive-Compulsive Disorder |
Panic Disorder |
|
Citalopram |
20-40 mg/day11 |
|
|
|
Fluoxetine |
5-20 mg/day1,2 |
80 mg/day3 |
5-20 mg/day titrated slowly4 |
|
Fluvoxamine |
50-150 mg/day9 |
150-300 mg/day10 |
50 mg/day titrated slowly to 150-300 mg/day10 Use divided doses for > 100 mg/day |
|
Paroxetine |
20 mg/day5 |
60 mg/day6 |
Unclear. Start at 10 mg/day |
|
Sertraline |
50 mg/day5,7 |
50-200 mg/day8 |
Unclear. Start at 25 mg/day. |
1Schweizer et al 1990; 2Wernicke et al 1988; 3Jenike et al 1989; 4Schneier et al 1990; 5Grimsley et al 1992; 6Black 1993; 7Adly et al 1992; 8Chouinard 1992; 9Wilde et al 1993; 10Palmer and Benfield 1994; 11Citalopram package insert*Acquisition cost based on Average Wholesale Price listed in Drug Topics Red Book 1998. Price to the patient may vary depending on individual pharmacy pricing policies.
#Cost comparison with reference TCAs: amitriptyline (generic) = $0.14 per 150 mg tablet; nortriptyline (generic) = $2.05 per 75 mg capsule.
|
Table 4. Typical Adverse Effects |
|||||
|
Adverse Effect |
Fluoxetine1 |
Fluvoxamine4 |
Paroxetine2 |
Sertraline3 |
Citalopram5 |
|
Nausea |
21.1 |
40 |
25.7 |
26 |
21 |
|
Headache |
20.3 |
22 |
17.6 |
20 |
|
|
Nervousness |
14.9 |
12 |
5.2 |
3.4 |
|
|
Sedation |
11.6 |
22 |
23.3 |
13 |
18 |
|
Insomnia |
13.8 |
21 |
13.3 |
16 |
15 |
|
Dry Mouth |
9.5 |
14 |
18.1 |
16 |
20 |
|
Diarrhea |
12.3 |
11 |
11.6 |
18 |
8 |
|
Weakness/Fatigue |
4.2 |
14 |
15.0 |
11 |
5 |
|
Anxiety/Agitation |
9.4 |
5 |
5.0 |
5.6 |
4 |
|
Sexual Dysfunction |
1.9 |
8 |
12.9 |
16 |
6 |
|
Anorexia |
8.7 |
6 |
6.4 |
2.8 |
4 |
1Dista Products Company 1992; 2SmithKline Beecham Pharmaceuticals 1992; 3Roerig 1992; 4Solvay Pharmaceuticals 1994, 5Forest Pharmaceuticals
|
Table 5. MAOI/SSRI Drug Interaction/Washout Period |
|||||
|
Time Course of Switch |
Fluoxetine |
Fluvoxamine |
Paroxetine |
Sertraline |
Citalopram |
|
SSRI -> MAOI |
5 weeks |
1-2 weeks |
1-2 weeks |
1-2 weeks |
14 days |
|
MAOI -> SSRI |
10-14 days |
10-14 days |
10-14 days |
10-14 days |
14 days |
|
Table 6. Efficacy of SSRIs in Melancholia. |
||||
|
|
|
|
|
|
|
|
Clomipramine |
|
|
57% (clomipramine) |
|
|
Nortriptyline |
|
|
63% (nortriptyline) |
|
|
Placebo |
|
|
15% (placebo) |
|
|
Placebo |
|
|
25% (placebo) |
|
|
Placebo |
|
|
41% (placebo) |
|
|
Imipramine Placebo |
|
|
61% (imipramine) 23% (placebo) |
|
Table 7. Effect of SSRIs on Cytochrome P450 Enzymes |
|||
|
|
|
|
|
|
|
|
|
Phenytoin Theophylline Caffeine |
|
|
|
Fluvoxamine Sertraline |
Diazepam Tolbutamide |
|
|
|
Fluvoxamine Paroxetine Sertraline Citalopram |
Haloperidol Perphenazine Thioridazine Clozapine Risperidone b-Blockers Type 1C antiarrhythmics |
|
|
|
Fluvoxamine Sertraline Citalopram |
Carbamazepine Alprazolam Triazolam Terfenadine Astemizole |
References
Abajo FJ, Rodriquez, Montero D (1999). Association between selective serotonin reuptake inhibitors and upper GI bleeding:population based case-control study. BMJ 319:1106-9.
Adly C, Straumanis J, Chesson A (1992). Fluoxetine prophylaxis of migraine. Headache 32:101-4.
Aizenberg D, Zemishlany Z, Weizman A (1995). Cyproheptadine treatment of sexual dysfunction induced by serotonin reuptake inhibitors. Clin Neuropharmacol 18:320-4.
Anderson IM (2000). Selective serotonin reuptake inhibitors versus tricyclic antidepressants: a meta-analysis of efficacy and tolerability. J Affective Disord 58:19-36.
Anderson IM (2001). Meta-analytical studies on new antidepressants. Br Med Bull 57;161-78.
Angst J, Stabl M (1992). Effiacy of moclobemide in different paitent groups: a meta-analysis of studies. Psychopharmacoogy (Berl) 106:S109-13.
Anonymous (1993). The top 200 survey. American Druggist 207(Feb):49-53.
Balough S, Hendricks SE, Kang J (1992). Treatment of fluoxetine-induced anorgasmia with amantadine (letter). J Clin Psychiatry 53:212-3.
Barbey JT, Roose SP (1998). SSRI safety in overdose. J Clin Psychiatry 59(suppl 15):42-8.
Barros J, Asnis G (1993). An interaction of sertraline and desipramine (letter). Am J Psychiatry 150:1751.
Beasley CM, Holman SL, Potvin JH (1993). Fluoxetine compared with imipramine in the treatment of inpatient depression: a multicenter trial. Ann Clin Psychiatry 5:199-207.
Berman RM, Anand A, Cappiello A, et al (1999). The use of pindolol with fluoxetine in the treatment of major depression: Final results from a double-blind, placebo-controlled trial. Biol Psychiatry 45:1170-7.
Berzsenyi P, Galateo E, Valzelli L (1983). Fluoxetine activity on muricidal aggression induced in rats by p-chlorophenylalanine. Aggressive Behav 9:333-8.
Black D (1993). University of Iowa. Personal communication.
Black K, Shea C, Dursun S et al (2000). Selective serotonin reuptake inhibitor discontinuation syndrome: proposed diagnostic criteria. J Psychiatry Neurosci 25:255-61
Bordet R, Thomas P, Bupuis B (1998). Effect of pindolol on onset of action of paroxetine in the treatment of major depression: Intermediate analysis of a double-blind, placebo-controlled trial. Am J Psychiatry 155:1346-51.
Bowden CL, Schatzberg AF, Rosenbaum A, et al (1993). Fluoxetine and desipramine in major depressive disorder. J Clin Psychopharmacol 13:305-10.
Boyer WF (1992). Potential indications for the selective serotonin reuptake inhibitors. Int Clin Psychopharmacol 6(suppl 5):5-12.
Brosen K, Skjelbo E, Rasmussen BB, et al (1993). Fluvoxamine is a potent inhibitor of cytochrome P-450 1A2. Biochem Pharmacol 45:1211-4.
Centorrino F, Baldessarini RJ, Kando J, et al (1994). Serum concentrations of clozapine and its major metabolites: effects of cotreatment with fluoxetine or valproate. Am J Psychiatry 151:123-5.
Chambers CD, Johnson KA, Dick LM, et al (1996). Birth outcomes in pregnant women taking fluoxetine. N Engl J Med 335:1010-5.
Ciraulo DA, Shader RI (1990a). Fluoxetine drug-drug interactions: I. antidepressants and antipsychotics. J Clin Psychopharmacol 10:48-50.
Ciraulo DA, Shader RI (1990b). Fluoxetine drug-drug interactions: II. J Clin Psychopharmacol 10:213-7.
Claghorn JL, Earl CQ, Walczak DD, et al (1996). Fluvoxamine maleate in the treatment of depression: a single-center, double-blind, placebo-controlled comparison with imipramine in outpatients. J Clin Psychopharmacol 16:113-20.
Clerc GE, Ruimy P, Verdeau-Pailles J (1994). A double-blind comparison of venlafaxine and fluoxetine in patients hospitalized for major depression and melancholia. Int Clin Psychopharmacol 9:139-43.
Coupland NJ, Bell CJ, Potokar JP (1996). Serotonin reuptake inhibitor withdrawal. J Clin Psychopharmacol 16:356-62.
Danish University Antidepressant Group (1990). Paroxetine: a selective serotonin reuptake inhibitor showing better tolerance, but weaker antidepressant effect than clomipramine in a controlled multicenter study. J Affective Disorders 18:289-99.
Dantzler A, Salzman C (1995). Treatment of bipolar depression. Psychiatr Serv 46:229-30.
Depression Guideline Panel (1993a). Depression in Primary Care: Volume 1. Detection and Diagnosis. Clinical Practice Guideline, Number 5. Rockville, MD. U.S. Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. AHCPR Publication No. 93-0550. April.
Depression Guideline Panel (1993b). Depression in Primary Care: Volume 2. Treatment of Major Depression. Clinical Practice Guideline, Number 5. Rockville, MD. U.S. Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. AHCPR Publication No. 93-0551. April.
DeVane CL (1994). Pharmacokinetics of the newer antidepressants: clinical relevance. Am J Med 97(suppl 6A):13S-23S
Dista Products Company (1992). Prozac package insert. Carolina, Puerto Rico: March.
Ereshefsky L, Riesenman C, Lam YF (1996). Serotonin selective reuptake inhibitor drug interactions and the cytochrome P450 system. J Clin Psychiatry 57:17-25.
Fabre LF, Scharf MB, Itil MM (1991). Comparative efficacy and safety of nortriptyline and fluoxetine in the treatment of major depression: a clinical study. J Clin Psychiatry 52 (6 suppl):62-7.
Feder R (1991). Reversal of antidepressant activity of fluoxetine by cyproheptadine in three patients. J Clin Psychiatry 52:163-4.
Feiger A, Kiev A, Shrivastava RK, et al (1996). Nefazodone versus sertraline in outpatients with major depression: focus on efficacy, tolerability, and effects on sexual function and satisfaction. J Clin Psychiatry 57 (suppl 2):53-62.
Fisher S, Kent TA, Bryant SG (1995). Postmarketing surveillance by patient self-monitoring: preliminary data for sertraline versus fluoxetine. J Clin Psychiatry 56:288-96.
Gelenberg AJ, Hopkins HS (1993). Report on efficacy of treatment for bipolar disorder. Psychopharmacol Bull 29:447-56.
George MS, Lydiard RB (1991). Speed of onset of action of the newer antidepressants--fluoxetine and bupropion. Int Clin Psychopharmacol 6:209-17.
Goff DC, Midha KK, Brotman AW, et al (1991). Elevation of plasma concentrations of haloperidol after the addition of fluoxetine. Am J Psychiatry 148:790-2.
Grimsley SR, Jann MW (1992). Paroxetine, sertraline, and fluvoxamine: new selective serotonin reuptake inhibitors. Clin Pharm 11:930-57.
Gupta S, Droney T, Masand P, et al (1999). SSRI-induced sexual dysfunction treated with sildenafil. Depression Anxiety9:180-2.
Heiligenstein JH, Tollefson GD, Faries DE (1994). Response patterns of depressed outpatients with and without melancholia: a double-blind, placebo-controlled trial of fluoxetine versus placebo. J Affective Disorders 30:163-73.
Herman JB, Brotman AW, Pollack MH, et al (1990). Fluoxetine-induced sexual dysfunction. J Clin Psychiatry 51:25-7.
Hoehn-Saric R, McLeod DR, Hipsley PA (1993). Effect of fluvoxamine on panic disorder. J Clin Psychopharmacol 13:321-6.
Hollander E, McCarley A (1992). Yohimbine treatment of sexual side effects induced by serotonin reuptake blockers. J Clin Psychiatry 53:207-9.
Hotopf M, Hardy R, Lewis G (1997). Discontinuation rates of SSRIs and tricyclic antidepressants: a meta-analysis and investigation of heterogenicity. Br Med J 170:120-7.
Jacobsen FM (1992). Fluoxetine-induced sexual dysfunction and an open trial of yohimbine. J Clin Psychiatry 53:119-22.
Janiri L, Gobbi G, Mannelli, et al (1996). Effects of fluoxetine at antidepressant doses on short-term outcome of detoxified alcoholics. Int Clin Psychopharmacol 11:109-17.
Johnson AM (1991). The comparative pharmacological properties of selective serotonin re-uptake inhibitors in animals. In: Feighner JP, Boyer WF, eds. Perspectives in psychiatry volume 1: Selective serotonin re-uptake inhibitors. Chichester, England: John Wiley & Sons:37-70.
Kaye CM, Haddock RE, Langley PF, et al (1989). A review of the metabolism and pharmacokinetics of paroxetine in man. Acta Psychiatr Scand 80(suppl 350):60-75.
Labbate LA, Brodrick PS, Nelson RP, et al (2000). Effects of bupropion sustained release on sexual functioning and nocturnal erections in healthy men. American Psychiatric Association Annual Meeting, Chicago, 2000.
Lapierre YD (1991). Controlling acute episodes of depression. Int Clin Psychopharmacol 6(suppl 2):23-35.
Lee HS, Song DH, Kim CH, et al (1996). An open clinical trial of fluoxetine in the treatment of premature ejaculation. J Clin Psychopharmacol 16:379-82.
McCormick S, Olin J, Brotman AW (1990). Reversal of fluoxetine-induced anorgasmia by cyproheptadine in two patients. J Clin Psychiatry 51:383-4.
McGrath PJ, Stewart JW, Janal MN, et al (2000). A placebo-controlled study of fluoxetine versus imipramine in the acute treatment of atypical Depression. Am J psychiatry 157:344-50.
Molina VA, Gobaille S, Mandel P (1986). Effects of serotonin-mimetic drugs on mouse-killing behavior. Aggressive Behav 12:201-11.
Montgomery SA, Henry J, McDonald G, et al (1994). Selective serotonin reuptake inhibitors: meta-analysis of discontinuation rates. Int Clin Psychopharmacol 9:47-53.
Musselman DL, Lawson DH, Gumnick JF, et al (2001). Paroxetine for the prevention of depression induced by high dose interferon alfa. N Engl J Med 344:961-6.
Musselman DL, Tomer A, Manatunga AK,et al (1996) Exaggerated platelet reactivity in major depression. Am J Psychiatry 1996;153:1313-17.
Nemeroff CB, Evans DL, Gyulai L et al (2001). Double-blind, placebo-controlled comparison of imipramine and paroxetine in the treatment of bipolar depression. Am J Psychiatry 158:906-12.
Nurberg HG, Lauriello J, Hensley PL et al (1999). Sildenafil for iatrogenic serotonergic antidepressant medication-induced sexual dysfunction in four patients. J Clin Psychiatry 60:33-5.
Pallanti S and Koran L (1999). Citalopram and sexual side effects of selective serotonin reuptake inhibitors. Am J Psychiatry 156:796.
Palmer KJ, Benfeild P (1994). Fluvoxamine: an overview of its pharmacological properties and review of its therapeutic potential in non-depressive disorders. CNS Drugs 1:57-87.
Pande AC, Birkett M, Fechner-Bates S, et al (1996). Fluoxetine versus phenelzine in atypical depression. Biol Psychiatry 40:1017-20.
Pastuzak A, Schick-Boschetto B, Zuber C, et al (1993). Pregnancy outcome following first-trimester exposure to fluoxetine (Prozac). JAMA 269:2246-8.
Patterson WM (1993). Fluoxetine-induced sexual dysfunction (letter). J Clin Psychiatry 54:71.
Perez V, Gilaberte I, Faries, D, et al (1997). Randomized, double-blind, placebo-controlled trial of pindolol in combination with fluoxetine antidepressant treatment. Lancet 349:1594-7.
Perez V, Puigdemont D, Gilaberte I, et al (2001). Augmentation of fluoxetine's antidepressant action by pindolol: Analysis of clinical, pharmacokinetics, and methodologic factors. J Clin Psychopharmacol 21:36-45.
Perucca E, Gatti G, and Spina E (1994). Clinical pharmacokinetics of fluvoxamine. Clin Pharmacokinet 27:175-90.
Peselow ED, Sanfilipo MP, Difiglia C, et al (1992). Melancholic/endogenous depression and response to somatic treatment and placebo. Am J Psychiatry 149:1324-32.
Pollock BG, Laghrissi-Thode F, Wagner WR (2000). Evaluation of platelet activation in depressed patients with ischemic heart disease after paroxetine or nortriptyline treatment. J Clin Psyhcopharmacol 20:137-40.
Rani PU, Naidu MUR, Prasad VBN, et al (1996). An evaluation of antidepressants in rheumatic pain conditions. Anesth Analg 83:371-5.
Rapaport M, Coccaro E, Sheline Y, et al (1996). A comparison of fluvoxamine and fluoxetine in the treatment of major depression. J Clin Psychopharmacol 16:373-8.
Ravindran AV, Guelfi JD, Lane RM, et al (2000). Treatment of dysthymia with sertraline: A double-blind, placebo controlled trial in dysthymic patients with major depression. J Clin Psychiatry 61:821-7.
Reimherr FW, Wood DR, Byerley B, et al (1984). Characteristics of responders to fluoxetine. Psychopharmacol Bull 20:70-2.
Roerig (1992). Zoloft package insert. New York, NY: June.
Ronfeld RA, Shaw GL, Tremaine LM (1988). Distribution and pharmacokinetics of the selective 5-HT uptake blocker sertraline in man, rat and dog. Psychopharmacology 96(suppl):269[Abstract].
Roose SP, Glassman AH, Attia E, et al (1994). Comparative efficacy of selective serotonin reuptake inhibitors and tricyclics in the treatment of melancholia. Am J Psychiatry 151:1735-9.
Rosenberg, KP (1999). Sildenafil citrate for SSRI-induced sexual side effects. Am J Psychiatry 156:157.
Rothschild AJ, Samson JA, Bessette MP, Carter-Campbell JT (1993). Efficacy of the combination of fluoxetine and perphenazine in the treatment of psychotic depression. J Clin Psychiatry 54:338-42.
Salzman C, Wolfson AN, Schatzberg A, et al (1995). Effect of fluoxetine on anger in symptomatic volunteers with borderline personality disorder. J Clin Psychopharmacol 15:23-9.
Schaller JL, Behar D (1999). Sildenafil citrate for SSRI-induced sexual side effects. Am J Psychiatry 156:156.
Schweizer E, Rickels K, Amsterdam JD, et al (1990). What constitutes an adequate antidepressant trial for fluoxetine? J Clin Psychiatry 51:8-11.
Shen WW and Hsu JH (1995). Female sexual side effects associated with selective serotonin reuptake inhibitors: a descriptive clinical study of 33 patients. Int J Psychiatry Med 25:239-48.
Shrivastava RK, Shrivastava S, Overweg N, Schmitt M (1995). Amantadine in the treatment of sexual dysfunction associated with selective serotonin reuptake inhibitors (letter). J Clin Psychopharamcol 15:83-4.
Slaughter RL, Edwards DJ (1995). Recent advances: The cytochrome P450 enzymes. Ann Pharmacother 29:619-24.
SmithKline Beecham Pharmaceuticals (1992). Paxil package insert. Philadelphia, PA: December.
Solvay Pharmaceuticals (1994). Luvox product monograph. Marietta, GA: December.
Sommi RW, Crismon ML, Bowden CL (1987). Fluoxetine: a serotonin-specific, second-generation antidepressant. Pharmacotherapy 7:1-15.
Srisurapanont M, Yatham LN, Zis AP (1995). Treatment of acute bipolar depression: a review of the literature. Can J Psychiatry 40:533-44.
Stahl S (2000). Essential Psychopharmacology. Neuroscientific Basis andPractical Applications 2nd Ed. Cambridge: New York.
Stratta P, Bolino F, Cupillari M, Casacchia M (1991). A double-blind parallel study comparing fluoxetine with imipramine in the treatment of atypical depression. Int Clin Psychopharmacol 6:193-6.
Thase ME, Entsuah AR, Rudolph RL (2001). Remission rates during treatment with venlafaxine or selective serotonin reuptake inhibitors. Br J Psychiatry 178:234-41.
Thase ME, Fava M, Hallbreich U, et al (1996). A placebo-controlled, randomized clinical trial comparing sertraline and imipramine for the treatment of dysthymia. Arch Gen Psychiatry 53:577-84.
Thomson C, Peveler RC, Stephenson D, McKendrick J (2000). Compliance with antidepressant medication in the treatment of major depressive disorder in primary care: a randomized comparison of fluoxetine and a tricyclic antidepressant. Am J Psychiatry 157:338-43.
Tignol J, Stoker MJ, Dunbar GC (1992). Paroxetine in the treatment of melancholia and severe depression. Int Clin Psychopharmacol 7:91-4.
Trindade E, Menon D, Topfer L, et al (1998). Adverse effects associated with selective serotonin reuptake inhibitors and tricyclic antidepressants: a meta-analysis. CMAJ 159:1245-52.
Walker PW, Cole JO, Gardner EA, et al (1993). Improvement in fluoxetine-associated sexual dysfunction in patients switched to bupropion. J Clin Psychiatry 54:459-65.
Wernicke JF, Dunlop SR, Dornseif BE, et al (1988). Low-dose fluoxetine therapy for depression. Psychopharmacol Bull 24:183-8.
Wilde MI, Plosker GL, Benfield P. Fluvoxamine: an updated review of its pharmacology and therapeutic use in depressive illness. Drugs 46:895-924.
Zajecka J, Mitchell S, Fawcett J (1997). Treatment-Emergent changes in sexual function with selective serotonin reuptake inhibitors as measured with the Rush Sexual Inventory. Psychopharmacol Bull 33:755-60.
Zanardi R, Franchini L, Gasperini M, et al (1996). Double-blind controlled trial of sertraline versus paroxetine in the treatment of delusional depression. Am J Psychiatry 153:1631-3.
Zornberg GL, Pope HG (1993). Treatment of depression in bipolar disorder: new directions for research J Clin Psychopharamcol 13:397-408.