* Researchers at the New York State Psychiatric Institute reviewed findings from 15 controlled studies on the use of anticonvulsants, lithium, fenfluramine and antidepressants in bulimia patients to reach their conclusions; patients in the studies used vomiting to compensate for food intake, were of normal weight, were predominately female, and were young adults.

* Found that only antidepressant therapy – particularly with fluoxetine – was consistently shown to reduce the frequency of binge eating among normal-weight patients with bulimia nervosa.

* Authors suggest that bulimia patients should demonstrate six months of real improvement before any effort is made to discontinue any antidepressant drug therapy.

In 1999, Leon et al. reported a prospective, naturalistic 15-year follow-up study from the National Institute of Mental Health Collaborative Depression Study. Analyses of this study focused on data from the time fluoxetine was available in the United States. Since case reports, retrospective studies, randomized clinical trials and surveys of prescribers all have limitations as to applicability to the general population of patients taking fluoxetine (Prozac), this NIMH study, with its close and frequent clinical assessments, offered the unique opportunity to evaluate this issue in a naturalistic treatment setting.

Method

The study focused on 643 subjects who were followed after approval of fluoxetine (Prozac) by the U.S. Food and Drug Administration at the end of 1987. Women comprised 64% of the sample, and 77% of subjects were inpatients at intake. Diagnosis at intake was major depressive disorder only (72%), mania or hypomania (21%), and schizoaffective disorder (7%).

A total of 185 subjects received fluoxetine (Prozac), 226 subjects received antidepressants other than fluoxetine (with imipramine [Tofranil] equivalence of at least 100 mg daily) and 232 subjects received no treatment. Over 89% of fluoxetine subjects also received at least one other antidepressant at some point during the follow-up period.

Suicidal risk for other selective serotonin reuptake inhibitors could not be assessed since they were not commonly prescribed or available during the study period. Therefore, the majority of subjects taking other antidepressants were using tricyclic antidepressants (TCAs).

Severity of affective psychopathology was rated on a weekly basis, and information on circumstances of suicide attempts and deaths was recorded. Each suicidal act was rated with regard to intent and lethality. Each of the three treatment groups had similar proportions of suicide attempts in the year prior to entry into the study.

Results

In the 185 subjects who received fluoxetine (Prozac), one suicide occurred and six other subjects had 11 suicide attempts. Of the six subjects with a suicide attempt, two patients each were taking 20 mg, 40 mg and 60 mg daily. Five of the six were taking at least one other psychotropic medication at the time of the suicide attempt-three were taking an additional antidepressant, and four an antipsychotic drug. The proportion who exhibited suicidal behavior was significantly reduced from 38.9% during the study follow-up prior to treatment with fluoxetine to just 3.8% during fluoxetine treatment.

There was one suicide and 42 suicide attempts by 23 subjects among the 226 subjects treated with antidepressants other than fluoxetine (Prozac). In the group of 232 subjects not treated with antidepressants, there were three suicides and 66 suicide attempts by 39 subjects.

Comparison of suicide intent and lethality found the highest ratings among those subjects not taking an antidepressant and the lowest ratings for those subjects taking fluoxetine (Prozac), but these differences did not reach statistical significance. In all subjects, the severity of psychopathology was strongly associated with an elevated risk of suicidal behavior.

Relative to no antidepressant treatment, there were nonsignificant reductions in the risk of suicide attempts associated with the use of fluoxetine and with the use of other antidepressant drugs. When factors of gender, age, diagnosis at intake, psychopathology and number of prior suicide attempts were controlled, fluoxetine was associated with a 56% decrease in risk of suicidal behavior. Antidepressants other than fluoxetine were associated with a 40% decrease in risk of suicidal behavior.

Study Conclusions

Study limitations were those inherent in a naturalistic study. Treatment was not randomly assigned, and antidepressant drugs other than fluoxetine (Prozac) were classified in aggregate for the analyses. In addition, long-term follow-up studies typically have high dropout rates, although two-thirds of fluoxetine (Prozac) patients in this study were followed for at least four years. Finally, even though the study numbers were large, it may have been inadequate for the study of an uncommon event such as suicide, and less serious suicide attempts might have been missed.

The results indicate that although fluoxetine (Prozac) was prescribed to more severely ill subjects, their risk of suicidal behavior was not elevated. Only seven of 185 (3.8%) subjects taking fluoxetine had suicidal attempts, compared to 23 of 226 (10.2%) subjects taking other antidepressants and 39 of 232 (16.8%) subjects receiving no antidepressant drug therapy.

The speculation that fluoxetine may increase the risk of suicide was not supported in this naturalistic study. Rather, a nonsignificant reduction in risk of suicidal behavior among patients treated with fluoxetine (Prozac) was found.

Case Reports

In 1990, Teicher et al. reported six cases of emergence, re-emergence or intensification of suicidal ideation in patients with major depression after two to seven weeks of treatment with fluoxetine (Prozac). While all patients were begun on 20 mg daily, four received doses greater than 40 mg per day, one stayed at 20 mg and another was titrated up to 40 mg then lowered back to 20 mg due to adverse effects.

The increased suicidal ideation persisted from three days to three months after fluoxetine was discontinued. Four of the six patients had comorbid psychiatric diagnoses, three were taking other psychotropic medication and four had histories of suicidal ideation or gestures. Four of the patients complained of inner restlessness.

In 1990, additional cases of fluoxetine-associated suicidal ideation were reported by Dasgupta and Hoover, although in one case, similar symptoms developed during subsequent treatment with imipramine, making specific association with fluoxetine (Prozac) unlikely.

Masand et al. in 1991 reported emergence of suicidal ideation in two cases-one was related to nonresponse to fluoxetine (Prozac), the other had both major depression and bulimia nervosa. In all case reports, no clear relationship was demonstrated between the emergence of suicidal thoughts or behavior and fluoxetine (Prozac) doses.

Fluoxetine is not unique in being the only antidepressant with case reports of paradoxical worsening of depressive symptoms, including suicidal ideation. Prior to the report by Teicher and his colleagues, four cases of worsened depressive symptoms and emergence of suicidal ideation after treatment with desipramine (Norpramin) 75 mg to 100 mg were reported (Damluji and Ferguson, 1988).

Within the first two weeks of treatment, all four patients had a noticeable worsening of depressive symptoms, particularly dysphoria, development of suicidal ideation not present prior to treatment and a worsening of neurovegetative symptoms. These symptoms were unrelated to any adverse effects.

In three patients, discontinuation of desipramine resulted in prompt relief of the worsened symptoms, but switching to either nortriptyline (Pamelor, Aventyl), amoxapine (Asendin) and trazodone (Desyrel) resulted in return of suicidal ideation. Interestingly, in one case, trials of both desipramine and amoxapine resulted in emergence of suicidal ideation, but a third trial of fluoxetine led to complete remission of the patient’s symptoms.

Other Studies

While case reports suggest an association between fluoxetine (Prozac) and emergence of suicidal ideation, surveys of prescribers, retrospective postmarketing studies and evaluation of premarketing clinical study data in thousands of fluoxetine patients failed to demonstrate a relationship between fluoxetine (Prozac) and suicide.

In 1991, Fava and Rosenbaum surveyed 27 psychiatrists who treated 1,017 depressed outpatients with an antidepressant drug. Of the 231 patients treated with fluoxetine alone, eight patients (3.5%) became suicidal after treatment was initiated. This compares to 6.5% of patients treated with both a TCA and fluoxetine (Prozac), 1.3% of patients treated with a TCA or lithium alone, and 3% of patients given other antidepressant drugs. No patients reported intense suicidal ideation of the degree described in Teicher et al.

At Eli Lilly and Co., a meta-analysis of clinical trial data by Beasley et al. in 1991 found that 1.2% of fluoxetine-treated patients experienced emergence of suicidal ideation during treatment, as compared to 2.6% on placebo and 3.6% on TCAs. Suicidal ideation improved in 72% of fluoxetine-treated patients compared to 55% of placebo patients.

Some of these case reports speculated that the association of fluoxetine (Prozac) and suicidality may be related to a drug-induced akathisia or restlessness, since four of Teicher’s six cases reported inner restlessness. All serotonin reuptake inhibitor drugs, including fluoxetine, have been reported to cause akathisia.

An analysis of 17 controlled clinical trials was conducted at Lilly Research Laboratories by Tollefson et al. in 1994 to test the hypothesis that patients who develop activation or akathisia from antidepressants might experience an associated emergent suicidal ideation. In the more than 3,000 patients analyzed, no temporal association was found between emergent suicidal ideation and any adverse effect cluster. Suicidality was associated infrequently with treatment-emergent activation, and at comparable rates, among fluoxetine (Prozac), TCAs and placebo-treated patients.

Discussion

While there are case reports of a temporal relationship between fluoxetine (Prozac) treatment and emergent suicidal ideation, this large naturalistic study adds to the substantial literature of meta-analyses of clinical trials with thousands of patients and retrospective reviews that fail to demonstrate such a relationship. The emergence or intensification of suicidal ideation is possible with any antidepressant drug.

One possible explanation of this observation is the lack of clinical response to the antidepressant and continued worsening of the depressive symptoms. Another explanation is that sleep disturbance and lack of energy are usually the first symptoms responsive to antidepressant drug therapy, occurring prior to the improvement of anhedonia and dysphoric mood. The suicidal depressed patient may thus have improved energy yet continued dysphoria and suicidal ideation. This may lead to an attempt at suicide after drug therapy begins.

All of these possible explanations of emergent or intensified suicidal ideation suggest the need for all depressed patients to be carefully monitored and evaluated in the first month or two of treatment.

A study funded by the National Institute of Mental Health (NIMH) concludes that Prozac is more effective in treating depression in teenagers than so-called talk therapy, but the two combined show the best treatment results. The NIMH has been conducting research to help clarify the potential value and risks of antidepressants and to explore how drug therapy compares with psychotherapy in adolescent depression. The multisite controlled clinical trial — the Treatment for Adolescents with Depression Study (TADS) — was launched in the late 1990s to directly compare the efficacy of fluoxetine, cognitive behavior therapy, and a combination of the two. The findings come amid controversy over the use of antidepressants in children and the possible link between antidepressants and the risk of suicide in children and teenagers. In this study of 439 adolescents suffering from moderate to severe depression, 71% responded well to the combined treatment. The researchers also found a dramatic decrease in the amount of suicidal thinking, although five suicide attempts were made by those given fluoxetine alone.

Intermittent dosages of fluoxetine (Prozac/Lilly) given during the luteal phase of the menstrual cycle may help women suffering from PMDD (premenstrual dysphoric disorder), studies suggest. Premenstrual dysphoric disorder, a severe variant of premenstrual syndrome, affects 3%-5% of women during their reproductive years, causing depression, anxiety, tension, lability, and irritability. A review of 12 clinical trials led researchers to conclude that, at 20 mg/day, fluoxetine reduces mood symptoms, physical symptoms, and social impairment in women with premenstrual dysphoric disorder and is well-tolerated.