Copyright 1995 by James D. Hegarty. All rights reserved.

This report is distributed through the Drugs and Devices Information Line at the Harvard School of Public Health. If you have comments, please mail them to the Line at ddil@episun1.harvard.edu. We will compile readers' views periodically and append them to the article.

The serotonin reuptake inhibitor antidepressant fluoxetine (Prozac®) was introduced by Eli Lilly and Company to the US market in January of 1988. Within two years it had become the most prescribed antidepressant in history. The popularity of fluoxetine was attributed to its low side effect profile and lack of toxicity in overdose. In fact, on March 26, 1990 Fluoxetine appeared on the cover of Newsweek magazine with the banner headline "Prozac: A Breakthrough Drug for Depression".(1) The article went on to expound the safety and efficacy of fluoxetine therapy in the treatment of major depression. Ironically, only one month earlier, a brief report appeared in the American Journal of Psychiatry by Martin Teicher et. al. at McLean Hospital describing six case reports of the emergence of violent suicidal ideation during fluoxetine therapy.(2) The resulting controversy spilled over from academic journals into the tabloids, talk shows and popular press. It spawned scores of lawsuits targeting Eli Lilly as the responsible party for suicides and violent acts that occurred during fluoxetine therapy. Of note, the only lawsuit to come to trial was dismissed by a Louisville, Kentucky jury on December 12, 1994 after less than a day of deliberation.(3) Over the past five years fluoxetine has been the subject of, at times, intense debate with the accusants ranging from skeptical psychiatrists to The Church of Scientology.(1)

This paper will attempt to avoid the more sensationalistic aspects of this story and, instead, focus on the empirical evidence available concerning a possible association between fluoxetine use and either violence or suicide. The pertinent literature was identified by performing a Medline® search in January of 1995 for the keywords fluoxetine with either violence or suicide appearing as MESH headings and/or as title/abstract text in articles published between 1988 and 1995. The search yielded 59 references published between 1990-95 and 3 published between 1988-90. Lets begin with a re-examination of the Teicher et. al. study of February 1990.

The paper by Teicher et. al. paper described six depressed patients "free of recent serious suicidal ideation who developed intense, violent suicidal preoccupation after 2-7 weeks of fluoxetine treatment". The authors estimated that the risk of developing violent suicidal preoccupation on fluoxetine to be 3.5% with a 95% CI (1.3%-7.5%). These findings were alarming, given the high estimated frequency of occurrence and the severity of this adverse effect.

The idea of suicidal action occurring early in the treatment of depression was not new. Detre and Jarecki (1971) described the "rollback phenomenon" in which suicidal ideation may reemerge during pharmacological treatment of severe depression. This idea was paralleled by the observation that severely depressed patients may not have the energy or initiative to act on suicidal thoughts until partial recovery occurs with the initiation of antidepressant medication.(4) In fact, the incidence of suicide attempts in depressed patients is approximately 15%, and SI can be expected to occur in as many as 90% of depressed patients.(5)

What was new was the idea that classes of antidepressants could differentially influence the occurrence of suicide. That fluoxetine might be associated with such an effect was particularly disturbing since its marketing focused on its unique status as an antidepressant agent with significantly fewer side effects (and toxicity) than earlier drugs including tricyclic (TCA) and monoamine oxidase inhibitor antidepressants (MAOI).

As with other spontaneous reports, an immediate effect of the February, 1990 Teicher article was to open up a stream of similar reports. Reports of fluoxetine-associated suicidality dominated the Letters to the Editor section of the American Journal of Psychiatry during the Fall of 1990. In October, Papp and Gorman confirmed the general observation that early fluoxetine therapy can be associated with a transient worsening of symptoms and postulated "a sudden increase in serotonergic transmission" as a biological mechanism.(6) Dasgupta and Hoover each reported a single case similar to those in the Teicher series.(7,8) King published a report of six adolescents with obsessive compulsive disorder who experienced new, or more intense, self-injurious ideation during fluoxetine treatment.(9) Yet, despite this flow of similar adverse case reports, there was much skepticism of Teicher's findings. For example, in the same AJP issue as the Dasgupta and Hoover letters, Miller reported his own anecdotal history of fluoxetine use in 100 patients with no evidence of violent suicidal ideation.(10) Miller questioned the high doses of fluoxetine reported by Teicher (80 mg/qd) and wondered if the observed increase in suicidal preoccupation was simply to do iatrogenic induction of akathisia (severe restlessness). Berkley similarly reported treatment of "over 300 patients with fluoxetine with startling success" and speculated that the Teicher findings were due to the fact all six patients had received MAOI therapy in the past, perhaps engendering a state of "serotonergic hypersensitivity".(11) Berkley concluded that "It would be catastrophic if Dr. Teicher and associates' article gave rise to hysteria that prevented patients from having the benefit of this unique and effective drug."

The initial criticisms of the Teicher report included the representativeness of his small (six patient) case series at a national referral center, the high dosages of fluoxetine employed, the use of concomitant benzodiazepines and the extensive prior psychotropic use including MAOI antidepressants. For Eli Lilly, 1990 began with Prozac on the cover of Newsweek but ended with the safety of the drug under intense media scrutiny. The fluoxetine and suicide debate, previously contained to the back pages of the AJP, moving onto the talk shows and tabloids. The Church of Scientology, long a vocal opponent of mainstream psychiatric practice, targeted fluoxetine for its attacks and amplified any reports connecting fluoxetine to violence or suicide. Meanwhile, several empirical studies neared completion in an attempt to more systematically address this question.

The first major study to examine the possible association of fluoxetine and suicidal preoccupation or violence was published by Fava and Rosenbaum of the Massachusetts General Hospital in March, 1991. The design involved a survey of 27 psychiatrists treating 1017 depressed outpatients at MGH.(12) The authors emphasized that, because they were aware of the Teicher report before its publication, they were able to complete their survey before the study participants had learned of Teicher's findings and, thereby, be biased by its findings. The Fava and Rosenbaum survey found 3.5% (8/231) of the MGH depressed outpatients treated with fluoxetine, 6.5% (4/62) of those treated with fluoxetine and tricyclics, 1.3% (5/385) of those treated with tricyclics alone or with lithium, and 3.0% (3/101) of those treated with other antidepressants became suicidal only after treatment with these antidepressants was initiated. The difference in the incidence of suicidal ideation (SI) was not significantly different between patients treated with fluoxetine and those receiving other antidepressant treatments. Furthermore, none of the 27 practitioners surveyed reported the emergence of the intense suicidal ideation described in the Teicher case series.

One month after the appearance of the Fava and Rosenbaum survey, Dr. Hoover wrote again to the editor of the AJP, cited the MGH study, and essentially retracted her original case report.(8) In her letter, Dr. Hoover detailed how her patient, who had originally developed intense SI on fluoxetine went on to experience an identical episode on imipramine (TCA) therapy. Dr. Hoover's rather sweeping conclusion? "Given these facts, it is unlikely that fluoxetine has any relationship to the development of suicidal ideation."(13) In September of 1991, Eli Lilly itself weighed in with a meta-analysis performed by Beasley et. al. of the 17 double blind clinical trials which were conducted prior to the release of fluoxetine in 1988.(14) This trials were pooled into a dataset including 1765 fluoxetine, 731 TCA and 569 placebo subjects. The pooled incidences for the emergence of substantial SI was 1.2% for fluoxetine, 3.6% for TCAs and 2.6% for placebo subjects. Similarly, the pooled incidences for worsening of SI was 15.3% for fluoxetine, 16.3% for TCAs and 17.9% for placebo. If only suicidal acts were considered, the pooled incidences were; 0.3% for fluoxetine, 0.4% for TCAs and 0.2% for placebo. The data were limited by the fact that these clinical trials lasted only from five to six weeks, but the emergence of new SI during early treatment, should have been captured in this rather narrow window. The Beasley et. al. meta-analysis did not contradict the null findings of the MGH study.

The first thorough review of the available literature on all antidepressants and the emergence of suicidal ideation was completed by Mann and Kapur in November, 1991.(15) This paper focused on all double-blind placebo-controlled randomized clinical trials (DBRCT) of antidepressants which reported findings on the emergence of suicidality and aggression. The authors reported that 2/2 amitriptyline, 1/1 alprazolam, 0/6 nonfluoxetine serotonin reuptake inhibitor DBRCTs found an increased risk for increased aggression or emergent SI. All three fluoxetine DBRCTs found a decreased incidence of SI and aggression. The authors pointed out that all reports of increased SI and aggression have appeared only in very small case series ranging in size from two 7 to six patients.(2)

While evidence seemed to be accumulating against an association between fluoxetine and SI/aggression, the popular press focused attention on the more titillating possibility that there was a relationship. On February 28, 1991 a "Donahue" TV talk show presented a group of individuals who reported compulsively self-destructive and violent behaviors after taking Prozac.(1) The "Prozac Defense" was introduced unsuccessfully in at least two murder trials and the New York Times employed the provocative headline "Suicidal Behavior Tied Again to Drug: Does Antidepressant Prompt Violence?".(16) Dozens of lawsuits were filed against Eli Lilly and Company, though only one has made it to trial.(17)

Of the 13 related publications appearing during 1992, only one continued to support an association between fluoxetine and increased SI or aggression. In this paper, Hamilton and Opler offered a neuroanatomical model to explain the alleged association. They proposed a syndrome, the "Extrapyramidal-Induced Dysphoric Reaction," in which fluoxetine mediates increased serotonergic activity by inhibiting the nigrostriatal dopamine tract and ultimately induces extrapyramidal side effects.(18) They interpreted the proposed increase in SI and aggression with fluoxetine as "one extreme manifestation" of extrapyramidal side effects. However, the rest of the psychiatric literature published in 1992 was dismissive of any such causal association. Power and Cohen noted that "the view that an SSRI like fluoxetine could exacerbate suicidal thinking and behavior is at variance with current biochemical formulations of affective disorders, which suggest that suicidal patients are particularly likely to have evidence of decreased brain 5-HT (serotonin) functioning.(19) Ashleigh and Fesler reported "our patient population of more than 200 patients failed to demonstrate even a single case of intense suicidal preoccupation, despite the incidence of 1.3%-7.5% predicted by Dr. Teicher and associates".(20) Additional negative reports included Dominguez and Goldstein's(21) case report of suicidal and homicidal ideation emerging during a placebo period and Wheadon et. al. reporting that there was a lower 2.0% (fluoxetine) vs. 3.8% (placebo) risk of substantial suicidal ideation in a placebo controlled DBRCT of fluoxetine in bulimic subjects.(22)

By 1993 the aggregate evidence weighted against a causal relationship. There were no published reports of fluoxetine associated SI or aggression in 1992-94. Ioannou raised the question of "Media coverage versus fluoxetine as the cause of suicidal ideation" suggesting that the spontaneous case reports might be the product of a dysphoric placebo effect created by expectations as a result of the pervasive negative publicity surrounding fluoxetine use in 1991.(23) Morton et. al. reported no increase in suicidal ideation during the 661 patient-months of follow-up in 252 outpatients who were subject to prospective surveillance from the initiation of fluoxetine therapy.(24)

The available data suggest that there may be an increased risk of suicide in depressed patients early in treatment. However, the risk does not appear to vary by type of antidepressant therapy. 1994 saw US sales of Prozac exceed $1 billion a year with an estimated 5 million Americans currently receiving the medication.(25) The debate about whether Prozac might precipitate homicidal aggression has shifted to a consideration of fluoxetine's "transformative powers," its ability not only to treat illness but to help some patients feel better than they have ever been.(26) The exuberant case report's in Listening to Prozac may well prove as difficult to generalize as Teicher's.

In conclusion, the available evidence supports only an equivalent or decreased risk for emergent SI or aggression during fluoxetine therapy as compared to other antidepressant medications. While academia and industry appear to have reached a consensus regarding the safety of fluoxetine, the final word may be left to the judiciary. The first case to come before a jury was decided only at the end of 1994 when a Louisville, Ky. jury rejected a lawsuit blaming Lilly and fluoxetine for a 1989 shooting rampage that left nine people dead.(3) The jury sat through six weeks of testimony but required only a few hours to acquit. A Lilly spokesman welcomed the verdict stating "The science was with us, there's just no evidence [that fluoxetine causes] suicidal and aggressive behavior." Lilly was able to present evidence that the shooter had a long history of homicidal and suicidal ideation before ever taking fluoxetine. One wonders how another jury may respond without such a well documented premorbid history. As of September, 1994 at least 160 fluoxetine-related civil cases remained pending against Eli Lilly and Company.(3)

--------------------------------------------------------------------------------

References

Breggin P. Toxic Psychiatry. New York: St. Martin's Press, 1991:464.
Teicher MH, Glod C, Cole JO. Emergence of intense suicidal preoccupation during fluoxetine treatment . Am J Psychiatry 1990; 147:207-10.
Prozac, Eli Lilly not responsible for shooting rampage, jury replies. Psychiatric News, January, 1995:1.
Detre TP, HG J. Modern Psychiatric Treatment. Philadelphia: JB Lippincott, 1971:59.
Montgomery SA, Montgomery DB, Green M, Bullock T, D. B. Pharmacotherapy in the prevention of suicidal behavior. J Clin Psychopharmacol 1992; 12:27-31.
Papp LA, Gorman JM. Suicidal preoccupation during fluoxetine treatment . Am J Psychiatry 1990; 147:1380-1.
Dasgupta K. Additional cases of suicidal ideation associated with fluoxetine. Am J Psychiatry 1990; 147:1570-1.
Hoover C. Additional cases of suicidal ideation associated with fluoxetine. Am J Psychiatry 1990; 147:1570-1.
King RA, Riddle MA, Chappell PB, et al. Emergence of self-destructive phenomena in children and adolescents during fluoxetine treatment . J Am Acad Child Adolesc Psychiatry 1991; 30:179-86.
Miller R. Discussion of fluoxetine and suicidal tendencies . Am J Psychiatry 1990; 147:1571. Berkley R. Discussion of fluoxetine and suicidal tendencies . Am J Psychiatry 1990; 147:1571-2.
Fava M, Rosenbaum JF. Suicidality and fluoxetine: is there a relationship? . J Clin Psychiatry 1991; 52:108-11.
Hoover CE. Suicidal ideation not associated with fluoxetine. Am J Psychiatry 1991; 148:543. Beasley CM, Jr., Dornseif BE, Bosomworth JC, et al. Fluoxetine and suicide: a meta-analysis of controlled trials of treatment for depression. Bmj 1991; 303:685-92.
Mann JJ, Kapur S. The emergence of suicidal ideation and behavior during antidepressant pharmacotherapy. Arch Gen Psychiatry 1991; 48:1027-33.
Angier N. Suicidal Behavior Tied Again to Drug: Does Antidepressant Prompt Violence? New York Times. New York, February 7,1991.
More Cases Have Been Filed Against Manuacturer of Antidepressant. Psychiatric Times, 1990.
Hamilton MS, Opler LA. Akathisia, suicidality, and fluoxetine . J Clin Psychiatry 1992; 53:401-6.
Power AC, Cowen PJ. Fluoxetine and suicidal behaviour. Some clinical and theoretical aspects of a controversy . Br J Psychiatry 1992 Dec;161:735-41 1992.
Ashleigh EA, Fesler FA. Fluoxetine and suicidal preoccupation . Am J Psychiatry 1992; 149:1750.
Dominguez RA, Goldstein BJ. Suicidal and homicidal ideations emerging during a placebo period . J Clin Psychiatry 1992; 53:171.
Wheadon DE, Rampey AH, Jr., Thompson VL, Potvin JH, Masica DN, Beasley CM, Jr. Lack of association between fluoxetine and suicidality in bulimia nervosa. J Clin Psychiatry 1992; 53:235-41.
Ioannou C. Media coverage versus fluoxetine as the cause of suicidal ideation. Am J Psychiatry 1992; 149:572.
Morton WA, Sonne SC, Lydiard RB. Fluoxetine-associated side effects and suicidality. J Clin Psychopharmacol 1993; 13:292-5.
Editorial. Artificial paradise encapsulated. Lancet 1994; 343:865-6.
Kramer P. Listening to Prozac. London: Fourth Estate, 1994:409.

http://www.hsph.harvard.edu/Organizations/DDIL/prozac.html

SMHAI Home | About Suicide | About Mental Health | Suicide Prevention | Suicide Survivors
Suicide Attempters | Self-Injury - Cutters | Crisis | Donate | SMHAI Library | Online Support & Resources
Speakers & Presentations | Memorials, Remebrances & Celebrations Of Life | Healing Music
Suggested Reading - Survivors | Suggested Reading - Attempters & Self-Injurers | Mental Health Pros.
Upcoming Events | Dr. Roerich's Welcome | Ann Gay's Welcome | Legal & About SMHAI
Privacy Policy | Copyright Notice | Awards Honoring SMHAI | SMHAI Awards Program | Contact

© SMHAI 2004 - 2006 All Rights Reserved.
No copying or redistribution without expressed written permission of SMHAI.
Logo Design by Allen R. Jacobson.
Site launched July 01, 2004.