Two recent reviews have noted that certain personality-disordered patients may partially benefit from treatment with psychotropic agents.1,2 In one of these reviews, Gitlin described three scenarios in which to consider the potential usefulness of psychotropics, including treating the Axis II disorder itself; treating associated Axis I disorders, such as psychosis or depression; and, treating symptom clusters within and across various personality disorders. Inherent in this last model is the concept of Axis I and II disorders constituting psychopathological dimensions to form a continuum of symptoms.1
For those considering drug therapy in these patients, one of the authors (F.J.A.) has summarized several critical issues:
* Emphasize that drugs are not a panacea.
* Address unrealistic expectations for therapy.
* Always employ concurrent nonpharmacological therapies in addition to medication.
* Employ serial drug substitutions, which are preferable to concurrent multiple drug use (i.e., avoid polypharmacy).
* If possible, avoid drugs that may cause tardive dyskinesia or paradoxical disinhibition, or lower the seizure threshold.
* Continually update informed consent.3,4
While the majority of drug treatment trials have involved borderline personality disorder, no medication has clearly emerged as efficacious for this or any other Axis II disorder. Presently, we would suggest that drug therapy be symptom oriented, regardless of the specific personality disorder. With this approach, a variety of presentations might benefit from the same agents used to treat a full diagnostic syndrome, including antipsychotics for transient psychosis or paranoia; antidepressants (e.g., SSRIs, MAOIs) for concurrent panic attacks, phobias, compulsive symptoms, as well as depressive syndromes; and mood stabilizers for cyclothymic presentations.
Because reasonable data are available for some but not all personality disorders, our review will specifically cover Cluster A - schizotypal; Cluster B - borderline, antisocial, and histrionic; and Cluster C - obsessive-compulsive and avoidant.
SCHIZOTYPAL PERSONALITY AND RELATED DISORDERS
This Axis II disorder demonstrates a pattern of discomfort in close relationships, cognitive or perceptual distortions, and behavioral eccentricities. A single-blind study involving 17 schizotypal patients treated with modest doses of Haloperidol (i.e., 2 to 12 mg/day) showed this agent produced some benefit; however, many patients were quite sensitive to the neuroleptic's side effects.5 Goldberg's study also found that thiothixene benefited schizotypal disorder alone or when concurrent with borderline personality disorder.6 Similarly, Jensen and Andersen conducted an open study in which five schizotypal personality-disordered patients responded to amoxapine (an antidepressant with purported neuroleptic properties), while five with borderline personality disorder showed no response.7 Finally, Schulz found that low-dose neuroleptics had a modest positive effect in patients with both schizotypal and obsessive-compulsive personality disorders.8
BORDERLINE PERSONALITY AND RELATED DISORDERS
Borderline personality disorder (BPD) is characterized by a pervasive pattern of unstable affect, stormy interpersonal relationships, and behavioral dyscontrol; and about 70% of BPD patients abuse alcohol or drugs. Approximately 1% to 2% of the general population manifest this syndrome. Unfortunately, self-mutilation, suicide attempts, and completed suicides are common. Indeed, it is estimated that 3% to 10% of these patients will take their lives.
BPD can also be a comorbid condition with major mood disorders; however, the relationship is weak in magnitude and somewhat inconsistent (i.e., different studies estimate that from 25% to 75% of these patients have a major depressive episode and 5% to 20% have a bipolar presentation) .9
The symptoms of borderline personality disorder can be divided into three groups: impulsive, consisting of both aggressive and self-damaging acts, eating disorders, and sensation seeking, which includes substance abuse; affective, which includes mood lability and depressive symptoms; and psychotic. In this context, New, Tristman, and Siever have commented that no single drug treats all three groups of symptoms seen in this disorder.10
Psychopharmacotherapy in BPD
Impulsive Group. One of the problems that plagues the investigation of borderline personality disorder is that many patients fit multiple diagnostic criteria. For example, there is a significant overlap in criteria among hysteroid dysphoria, bipolar type II, and borderline personality disorders. It is also extraordinarily difficult to do well-controlled clinical trials with highly disturbed, manipulative patients, such as those with acute mania or severe borderline features. Thus, while recognizing the heroic nature of doing such work, we must also keep in mind that some of the borderline studies include very small sample sizes.
In the multiple crossover study of Gardner and Cowdry that compared various agents to placebo over a 3-week drug trial, there were only six pairs for alprazolam versus placebo, eight pairs for carbamazepine versus placebo, five pairs for trifluoperazine versus placebo, and eight pairs for tranylcypromine versus placebo.11 Four of 16 patients on tranylcypromine had a moderate or marked improvement, while only 1 of 13 placebo patients showed similar improvement. Of note, those receiving alprazolam had significant increases in dyscontrol in comparison to those on placebo. In another study, Hedberg, Houck, and Glueck conducted a double-blind, crossover design in a sample of schizophrenics that included 28 characterized as "pseudo-neurotic," 14 of whom responded to tranylcypromine.12 Thus, these two small-sample studies provide some evidence that tranylcypromine may be beneficial for borderline personality and related disorders.13
Soloff and coworkers carried out a particularly important comparison of phenelzine versus haloperidol in BPD with depression.14 Although this study had reasonable sample sizes (i.e., 38 on phenelzine, 36 on haloperidol, and 34 on placebo), it failed to find a clear advantage for phenelzine over placebo. It did, however, find that phenelzine produced a significantly better improvement in hostility than placebo, and a greater antidepressant response than haloperidol. The authors then maintained their patients on these same regimens for an additional 4 months, with little evidence that phenelzine had a continued greater beneficial effect than placebo.
There have been a number of open trials with selective serotonin response inhibitors (SSRIs) (e.g., fluoxetine and sertraline) in BPD that indicate they may be most helpful for impulsive aggression and irritability.15-20 Salzman has also recently reported a small, placebo-controlled, double-blind, randomassignment study of fluoxetine in patients with mild to moderately severe borderline personality disturbance.21 This was a 13-week trial with 13 patients on fluoxetine and 9 on placebo. There was a striking reduction in measures of anxiety in those on fluoxetine (i.e., seven times as likely to show decrease in anger versus patients on placebo). Furthermore, this change was independent of the level of depression. Norden investigated 12 patients with BPD without major depression and found that fluoxetine produced substantial improvement, which was maintained through a follow-up period of 6 months.19 Seventy-five percent of the patients were assessed as much or very much improved for a wide variety of symptoms, including rejection sensitivity, anger, mood lability, irritability, and impulsivity, including substance abuse and overeating. Of note, six patients who interrupted medication at some point during this lengthy trial experienced an increase in symptoms, which remitted with the resumption of fluoxetine. Thus, this finding supports the Salzman study.21
Links et al reported a small, blind crossover study in BPD patients treated with lithium, desipramine, and placebo.22 While 8 of 13 patients responded to desipramine, 6 of 12 also responded to placebo. Of those patients with high scores for anger and suicide symptoms, 4 of 11 responded to desipramine, in comparison to 5 of 6 with placebo. Finally, one small placebo-controlled study found lithium effective in BPD (i.e., decreased both anger and suicidal symptoms), and another trial also found carbamazepine beneficial.23
Affective Group. Conventional tricyclic antidepressants have had a variable and sometimes lackluster effect on the depressive symptoms seen in borderline personality disorder. Indeed, some studies indicate that these drugs may worsen symptoms, particularly by increasing irritability and agitation.24 By contrast, Cole et al did a retrospective evaluation of drug response in patients with borderline personality and major depression, noting that five of six who received tricyclics responded, whereas BPD patients without depression treated with a tricyclic antidepressant did not respond.25
Although the term pseudo-neurotic schizophrenia has been dropped from our nomenclature, Klein and colleagues made an interesting observation regarding drug treatment of this condition. Specifically, in an investigation comparing the effects of chlorpromazine and Imipramine in a wide variety of patients, they observed that imipramine was beneficial for this diagnosis.26 In this controlled study, imipramine produced a positive outcome in 60% of patients, in contrast to placebo, which produced only a 25% positive outcome.
This group also investigated the effects of antidepressants in patients with emotionally unstable personalities, a condition characterized by excitability and ineffectiveness when confronted with minor stress. This syndrome is primarily found in female adolescents whose moods consist of short periods of intense unhappiness, social withdrawal, depression and irritability, episodes of impulsivity, rejection of religious rules, and pleasure seeking. They found that imipramine produced improvement in 67% and chlorpromazine produced improvement in 81% of these cases, with both agents demonstrating a statistically significant greater improvement in comparison to placebo.27,28
Rifkin also investigated 21 patients with emotionally unstable character disorder whose core psychopathology consisted of depression and/or hypomanic mood swings lasting hours to days.29 Although they had a history of chronic maladaptive behavior patterns (e.g., truancy, poor work history), their symptoms seemed more consistent with bipolar disorder or cyclothymia. While there are certain similarities between emotionally unstable character disorder and BPD, the patients described in this double-blind study may have had a variant of bipolar disorder. In this light, it is noteworthy that lithium produced a beneficial effect on mood swings. Further, Gardner and Cowdry reported that 10 of 11 similar patients also had a modest improvement in mood when given carbamazepine.
While we have reviewed evidence that both MAO inhibitors and SSRIs may benefit the impulsivity cluster of BPD, as expected, some of these same studies also demonstrated that these antidepressants were effective in controlling mood. Because patients with BPD are prone to suicide attempts, it is important to remember that tricyclic antidepressants and MAOIs are lethal in overdose, while the SSRIs are very safe. Further, these latter agents may have antidepressant as well as anti-impulsivity/aggression properties.
Psychotic Group. There are several placebo-controlled studies indicating that antipsychotics are efficacious in BPD. For example, Goldberg et al, in a double-blind, placebo-controlled study of 50 borderline or schizotypal personality-disordered patients,6 found that thiothixene (mean dose=8.7 mg/day) was superior to placebo, with improvement most evident for illusions, ideas of reference, and psychoticism. Of note, there were seven patients with borderline personality disorder only, six with schizotypal personality disorder only, and 11 with both, with meaningful improvement occurring only in the latter two patient groups. In Cowdry and Gardner's multiple crossover design,13 trifluoperazine was compared with tranylcypromine, carbamazepine, alprazolam, and placebo in 16 female BPD patients. While many in the neuroleptic group did not tolerate the drug or clinically deteriorated, the five who completed the trial experienced modest benefit.
Several open studies have also found antipsychotics useful in BPD.6,30,31 For example, Leone found that both loxapine and chlorpromazine were effective and equal in most respects.32 Frankenburg and Zanarini conducted an open study in 15 patients who met the criteria for BPD but also had a substantial degree of psychosis.33 Indeed, their highest rating on the Brief Psychiatric Rating Scale was on the suspiciousness, hallucinations, and unusual thought content items. Treatment with clozapine produced substantial improvement, as was evident from a systematic follow up of these patients, and was effective despite a history of nonresponse to conventional neuroleptics. Because these trials were in hospitalized patients, some with more severe schizophrenia, it may have been the comorbid psychosis that responded to clozapine rather than the borderfine disorder per se. By contrast, alprazolam was ineffective, and indeed, 7 of 12 patients on this agent either attempted suicide or were assaultive.
Soloff et al compared haloperidol, amitriptyline, and placebo in 61 patients and found the neuroleptic clearly superior to amitriptyline or placebo, particularly for psychotic symptoms.32 The study is particularly noteworthy in that the sample size is reasonably large, with approximately 28 in each group. While haloperidol benefited psychotic symptoms (e.g., paranoid ideation, psychoticism), general severity, and depression, it also improved depressive symptoms, as did amitriptyline.
Serban and Siegal investigated a population of borderline-schizotypal personalities and found thiothixene was more effective than haloperidol.35 This population, however, was less psychotic than those in the earlier studies.
Next, Cornelius et al investigated haloperidol, phenelzine, and placebo in a 4-month continuation study.36 Those on haloperidol deteriorated faster, with two thirds dropping out in the first 8 weeks in contrast to only one fourth on placebo. Furthermore, those receiving haloperidol showed more depression, particularly in comparison to those on phenelzine. Overall, there were no significant changes on a global assessment scale.
In an open trial of antipsychotics for BPD outpatients, Teicher and coworkers noted that three patients developed sustained melancholic depression, necessitating removal from the trial and treatment with antidepressants.37 In this context, Gardner and Cowdry also reported that three patients on carbamazepine developed melancholia, which required drug discontinuance and treatment for the depression.38 Hence, clinicians should anticipate that severe depression can develop in some BPD patients and be prepared to treat with antidepressants.
In Cole's previously mentioned retrospective chart review of BPD, all nine patients falling into the schizophrenic group were treated with antipsychotics and improved. In addition, three of four patients with major depression who received antipsychotics alone also responded. By contrast, four patients with core borderline personality disorder without psychosis or depression did not improve on antipsychotics.
Borderline personality disorder can be lifethreatening with a stormy course, at times necessitating inpatient hospitalization; or it can be a milder disorder managed on an outpatient basis. Thus, because one sample of patients may be more severely disturbed or has more psychotic features than another, discrepancies in outcome among studies may be due to these differences. In addition, because several studies combine schizotypal and borderline personality disorders, it is important to separate these patients when analyzing the results.
ANTISOCIAL PERSONALITY DISORDER
This disorder is characterized by a blatant disregard for or violation of others' rights. While the relationship between Type II alcoholism and antisocial personality is unclear, there is excellent evidence that impulsive violence in Type II alcoholics may be related to low brain serotonin levels, as manifested by low CSF 5-HIAA. In this context, while it has been asserted that the SSRIs may precipitate suicidal ideation and violence, there is also excellent evidence that they typically do not induce and can even reduce such impulsivity.39 This finding leads to the hope that elucidating the biology of impulsivity may lead to effective drug treatments. Unfortunately, we know of no clinical trials presently addressing this question.
Sheard investigated men in a maximum security prison who repeatedly committed aggressive acts but were not psychotic or brain damaged. In this blinded study of lithium, he found that it reduced impulsivity and aggression.40 Without knowledge of an individual's psychopathology, however, it is difficult to know for whom lithium may be most useful, but his observation warrants further study. By contrast, Schiff41 reported a case in which lithium exacerbated interictal aggression in patients with complex partial seizures.
HISTRIONIC PERSONALITY DISORDER
This disorder is characterized41 by excessive emotions and attention-seeking behavior. Its drug treatment has not been studied but may be similar to the syndrome Leibowitz and Klein have termed "hysteroid dysphoria" or atypical depression.42,43 In this context, there is evidence that MAOIs, such as phenelzine, may be even more effective in hysteroid dysphoria than tricyclics.44
OBSESSIVE-COMPULSIVE PERSONALITY DISORDER
This disorder includes individuals who are preoccupied with orderliness, perfectionism, and control. Baer et al found that only 6% of those with full obsessive-compulsive disorder (OCD) also had comorbid obsessive-compulsive personality, suggesting that these two conditions are distinct.45 When OCD and psychosis are present in obsessive-compulsive personality, however, antipychotics may be helpful.
AVOIDANT PERSONALITY DISORDER
These patients manifest social inhibition, a sense of inadequacy, and hypersensitivity to negative evaluation. Because most patients with avoidant personality disorder also have social phobias, there is a substantial degree of overlap. Indeed, some investigators estimate that as many as 85% of these patients also meet diagnostic criteria for social phobia.46,47 Therefore, social phobias should always be considered when the diagnosis of avoidant personality is made.
Patients with social phobias feel anxious and are subject to such peripheral symptoms as anxiety-like heart palpitations, blushing, and sweating, especially when they feel people are looking at or evaluating them.
The prevalence of social phobias is about 2.8%, and they can be subdivided into generalized and discrete types. Discrete social phobias consist of anxiety regarding specific activities such as speaking or performing in public. General social phobia is the fear of interacting with other people and the avoidance of groups, an observation that is confirmed by others. This is different from patients who are pathologically shy or fearful of certain social events, like a cocktail party, where they are concerned about embarrassing themselves or believe that people will be unduly critical. Here, the fear does not lead to physical symptoms such as tremor or rapid heartbeat, but is more like embarrassment and fear of being watched or judged.
Because patients formally diagnosed with avoidant personality respond to the same drugs (e.g., MAOIs and SSRIs) that social phobies do, we believe avoidant personalities often include patients with social phobia and should be treated with these drugs.48-49 Given the present evidence and risk-benefit ratio, SSRIs would be our first-line therapy, with MAOIs reserved as a second choice.
Selective Serotonin Reuptake Inhibitors
There is a substantial body of literature showing that the SSRIs are effective for social phobias. For example, Lepóla and coworkers reported an open trial in which three phobic patients responded to Citalopram.50 Other open studies have shown social phobies to have an excellent response to fluoxetine or sertraline.51 van Vliet et al performed a double-blind, random-assignment study of fluvoxamine in social phobies.52 A substantial degree of improvement was noted in 46% of patients on fluvoxamine, whereas only 7% on placebo improved in this trial of 30 patients. This study is noteworthy in that it is the first double-blind, random-assignment study of an SSRI for this disorder. Additionally, patients were excluded from the study if they were experiencing depression, thus eliminating the possibility that the beneficial effect was secondary to an improvement in mood.
In general, social phobias respond to treatment after a 2- to 3-month period. Indeed, in the van Vliet study, significant decreases in anxiety were not reached until week 12. Thus, clinicians must be prepared for a longer therapeutic trial than when these drugs are used as antidepressants.
Monoamine Oxidase Inhibitors
The clinical observation that the MAOIs are effective in treating social phobia was followed by controlled trials showing these agents to be superior to placebo. For example, Leibowitz et al conducted a double-blind, controlled clinical trial using the beta blocker atenolol, phenelzine, or placebo in 74 social phobic patients.53 Over a 16-week period, phenelzine produced a greater improvement than either atenolol or placebo. These results are consistent with a second trial using the MAOI tranylcypromine.54 A reversible and selective MAO-A inhibitor (RIMA), brofaromine, has also been shown effective for social anxiety and social avoidance over a 2month period.55
The use of propranolol or other beta blockers is an entirely different concept, because these agents block the physiologic manifestations of anxiety and are used primarily to alter the peripheral, not the central, nervous system. Symptoms such as palpitations, rapid heartbeat, tremor, tingling, cold sweats, chest constriction, and twitching are in part caused by the secretion of epinephrine during stress. This raises the question of whether the basis of anxiety is the perception of those internal and external epinephrine-induced physiological events (i.e., a hyperawareness of normal adrenergic functioning in a specific social situation, such as performing before an authence).
Some studies divided patients into psychic anxiety and somatic anxiety groups. Although diazepam was superior to placebo for psychic anxiety, both diazepam and propranolol were superior to placebo for somatic anxiety.56 Further evidence that the peripheral betareceptor blocking property is responsible for this improvement is seen with dextropropranolol, an agent that lacks beta-blocking properties and was ineffective in treating anxiety. Also relevant is the fact that practolol, a selective peripheral beta blocker, demonstrated antianxiety properties at a dose of 200 mg/day. One must be aware of the side effects and contraindications for beta blockers, particularly asthma and cardiac conditions, for which a slowing of the heart rate could be detrimental.
Open trials have also shown buspirone and alprazolam to be effective in social phobia.57,58
Until recently in the United States, a great variety of patients, including some who would now be classified as BPD, were diagnosed as schizophrenic or "borderline" schizophrenic. This is understandable because many patients have difficult psychological problems, often requiring aggressive inpatient intervention. In addition, proper diagnosis and optimal psychopharmacotherapy of all personality disorders is hampered due to the lack of a clear separation of comorbid conditions. Thus, when a patient has two disorders (e.g., schizophrenia or depression plus borderline personality disorder), treatment becomes more complicated, and simply managing the Axis I condition should not necessarily be construed as adequate treatment of the comorbid personality disorder. Because there is evidence that SSRIs have a specific effect on obsessive-compulsive symptoms not mediated by their antidepressant effect, and these agents, as well as MAOIs, benefit social phobia not associated with comorbid depression, it may be possible to develop similar drug strategies to disentangle other personality disorders from Axis I comorbid conditions.
It is also possible that there may not be a strict isomorphism between personality disorders and other diagnostic entities. Thus, a "personality disorder" may be a psychological condition that only exists in patients who have a major psychiatric disorder, such as schizophrenia, atypical depression, or social phobia. One example is mania, which can present with interpersonal behaviors such as attacking people's weaknesses, sensitivity to division, and manipulative behavior with staff. This syndrome disappears completely, however, when the patient is treated with a mood stabilizer.
In summary, major psychiatric disorders comorbid with various personality disorders should be recognized and properly treated. While this approach will not alter an underlying and separate personality disorder, it often brings about significant symptomatic improvement. Implicit in this approach is a drug treatment strategy that targets symptom patterns and not specific personality disorders.
1. Gitlin MJ. Pharmacotherapy of personality disorders: conceptual framework and clinical strategies. J Clin Psychopharmacol. 1993; 13:343-353.
2. Soloff P. Pharmacotherapy of borderline disorders. Compr Psychiatry. 1981; 22:535-543.
3. Ayd F. Psychopharmacologic treatment of personality disorders. Int Drug Therapy Newsletter. 1990; 25(4): 1-2.
4. Janicak PG, Davis JM, Preskorn SH, Ayd FJ. Principles and Practice of Psychopharmacotherapy. Baltimore, MD: Williams & Wilkins; 1993.
5. Hymowitz P, Frances A, Jacobsberg L, et al. Neuroleptic treatment of schizotypal personality disorder. Compr Psychiatry. 1986; 27:267-271.
6. Goldberg SC, Schulz SC, Schulz PM, et al. Borderline and schizotypal personality disorders treated with low-dose thiothixene vs placebo. Arch Gen Psychiatry. 1986; 43:680-686.
7. Jensen HV, Anderson J. An open, noncomparative study of amoxapine in borderline disorders. Acta Psychiatr Scand. 1989; 79:89-93.
8. Schulz SC. The use of low dose neuroleptics in the treatment of "schizo-obsessive" patients. Am J Psychiatry. 1986; 143:1318-1319.
9. Widiger TA, Frances AJ. Epidemiology, diagnosis, and comorbidity of borderline personality disorder. In: Tasman A, Hales RE, Frances AJ, eds. Review of Psychiatry. Vol 8. Washington, DC: American Psychiatric Press, Ine; 1989:8.
10. New AS, Trestman RL, Siever LJ. The pharmacotherapy of borderline personality disorder. CNS Drugs. 1994; 2:347-354.
11. Gardner DL, Cowdry RW. Alprazolam induced dyscontrol in borderline personality disorder. Am J Psychiatry. 1985; 142:98-100.
12. Hedberg DL, Houck JH, Glueck BC Jr. Tranylcyprominetrifluoperazine combination in the treatment of schizophrenia. Am J Psychiatry. 1971; 127:1141-1146.
13. Cowdry RW, Gardner DL. Pharmacotherapy of borderline personality disorder. Alprazolam, carbamazepine, trifluoperazine, and tranylcypromine. Arch Gen Psychiatry. 1988;45:111-119.
14. Soloff PH, Cornelius J, George A, et al. Efficacy of phenelzine and haloperidol in borderline personality disorder. Arch Gen Psychiatry. 1993; 50:377-385.
15. Coccaro EF, Astili JL, Herbert JA, et al. Fluoxetine treatment of impulsive aggression in DSM-III-R personality disorder patients. J Clin Psychopharmacol. 1990; 10:373-375.
16. Cornelius JR, Soloff PH, Perei JM, et al. Fluoxetine trial in borderline personality disorder. Psychopharmacol Bull. 1990;26:151-154.
17. Cornelius JR, Soloff PH, Perei JM, et al. A preliminary trial of fluoxetine in refractory borderline patients. J Clin Psychopharmacol. 1991; 11:116-120.
18. Markowitz PJ, Calabrese JR, Schulz SC, et al. Fluoxetine treatment of borderline and schizotypal personality disorder. Am J Psychiatry. 1991;148:1064-1067.
19. Norden MJ. Fluoxetine in borderline personality disorder. Prog Neuropsychopharmacol Biol Psychiatry. 1989; 13:885-893.
20. Kavoussi RJ, Liv J, Coccaro EF. An open trial of sertraline in personality disorder patients with impulsive aggression. J Clin Psychiatry. 1994; 55:137-141.
21. Salzman C. Effect of fluoxetine on anger in borderline personality disorder abstract. Neuropsychopharmacology. 1994; 10(suppl 1):826.
22. Links PS, Steiner M, Boiageo BA, et al. Lithium therapy for borderline patients: preliminary findings. Journal of Personality Disorders. 1990; 4:173-181.
23. Gardner DL, Cowdry RW. Positive effects of carbamazepine on behavioral dyscontrol in borderline personality disorder. Am J Psychiatry. 1986; 143:519-522.
24. Soloff PH, George A, Nathan RS, et al. Paradoxical effects of amitriptyline on borderline patients. Am J Psychiatry. 1986; 143:1603-1605.
25. Cole JO, Salomon M, Gunderson J. Drug therapy in borderline patients. J Clin Psychiatry. 1984; 25:249-254.
26. Klein DF. Importance of psychiatric diagnosis in the prediction of clinical drug effects. Arch Gen Psychiatry. 1967; 16:118-126.
27 Klein DF. Chlorpromazine-procyclidine combination, Imipramine and placebo in depressive disorders. Canadian Psychiatric Association Journal. 1966; lKsuppl 1): 146-149.
28. Fink M, Klein DF, Kramer JC. Clinical efficacy of chlorpromazine-procyclidine combination, Imipramine and placebo in depressive disorders. Psychopharmacologia. 1965;7(1):27-36.
29. Rifkin A, Quitkin F, Carrillo C, et al. Lithium carbonate in emotionally unstable character disorder. Arch Gen Psychiatry. 1972; 27:519-523.
30. Teicher MH, Glod CA, Aaronson ST, et al. Open assessment of the safety and efficacy of thioridazine in the treatment of patients with borderline personality disorder. Psychopharmacol Bull. 1989; 25:535-549.
31. Brinkley JR, Beitman BD, Friedel RO. Low-dose neuroleptic regimens in the treatment of borderline patients. Arch Gen Psychiatry. 1979; 36:319-326.
32. Leone NF. Response of borderline patients to loxapine and chlorpromazine. J Clin Psychiatry. 1982; 43:148-150.
33. Frankenberg FR, Zanarini MC. Clozapine treatment of borderline patients: a preliminary study. Compr Psychiatry. 1993; 34:402-405.
34. Soloff PH, George A, Nathan S, Schulz PM, et al. Amitriptyline versus haloperidol in borderlines: final outcomes and predictors of response. J Clin Psychopharmacol. 1989; 9:238-246.
35. Serban G, Siegel S. Responses of borderline and schizotypal patients to small doses of thiothixene and haloperidol. Am J Psychiatry. 1984; 141:1455-1458.
36. Cornelius JR, Soloff PH, George A, et al. Haloperidol versus phenelzine in continuation therapy of borderline disorder. Psychopharmacol Bull. 1993; 29:333-337.
37. Teicher MH, Glod CA, Aaronson ST, et al. Open assessment of the safety and efficacy of thioridazine in the treatment of patients with borderline personality disorder. Psychopharmacol Bull. 1989; 25:535-549.
38. Gardner DL, Cowdry RW. Development of melancholia during carbamazepine treatment in borderline personality disorder. J Clin Psychopharmacol. 1986; 6:236-239.
39. Mann J, Kapur S. The emergence of suicidal ideation and behavior during antidepressant pharmacotherapy. Arch Gen Psychiatry. 1991;48:1027-1033.
40. Sheard MH, Marini JL, Bridges CL, Wagner E. The effect of lithium on impulsive aggressive behavior in man. Am J Psychiatry. 1976;133:1409-1413.
41. Schiff HB, Sabin TD, Geller A, et al. Lithium in aggressive behavior. Am J Psychiatry. 1982; 139:1346-1348.
42. Leibowitz MR, Klein DF. Interrelationship of hysteroid dysphoria and borderline personality disorder. Psychiatr Clin North Am. 1981; 4:67-87.
43. Leibowitz MR, Quitkin FM, Stewart JW, et al. Antidepressant specificity in atypical depression. Arch Gen Psychiatry. 1988; 45:129-137.
44. Parsons B, Quitkin FM, McGrath PJ, et al. Phenelzine, Imipramine, and placebo in borderline patients meeting criteria for atypical depression. Psychopharmacol Bull. 1989; 25:524-534.
45. Baer L, Jenike MA, Ricciardi JN, et al. Standardized assessment of personality disorders in obsessive-compulsive disorder. Arch Gen Psychiatry. 1990; 47:826-830.
46. Holt CS, Heimberg RG, Hope DA. Avoidant personality disorder and the generalized subtype of social phobia. J Abnorm Psychol. 1992; 101:318-325.
47. Schneier FR, Spitzer RL, Gibbon M, Fyer AJ, Liebowitz MR. The relationship of social phobia subtypes and avoidant personality disorder. Compr Psychiatry. 1991; 32:1-5.
48. Leibowitz MR. Pharmacotherapy of social phobia. J Clin Psychiatry. 1993; 54(12, suppl):31-35.
49. Tancer ME. Neurobiology of social phobia. J Clin Psychiatry. 1993; 54(12, suppl):26-30.
50. Lepola U, Koponen H, Leinonen E. Citalopram in the treatment of social phobia: a report of three cases. Pharmacopsychiatry. 1994; 27:186-188.
51. Van Ameringen M, Mancini C, Streiner D. Sertraline in social phobia. J Affect Disord. 1994; 31:141-145.
52. van Vliet IM, den Boer JA, Westenberg HGM. Psychopharmacological treatment of social phobia; a double blind placebo controlled study with fluvoxamine. Psychopharmacology. 1994; 115:128-134.
53. Leibowitz MR, Schneier F, Campeas R, et al. Phenelzine versus atenelol in social phobia. A placebo controlled comparison. Arch Gen Psychiatry. 1992;49:290-300.
54. Versiani M, Mundin FD, Nardi AE, et al. Tranylcypromine in social phobia. J Clin Psychopharmacol. 1988; 8:279-282.
55. Bakish D. The use of the reversible monoamine oxidase-A inhibitor brofaromine in social phobia complicated by panic disorder with or without agoraphobia. J Clin Psychopharmacol. 1994; 14:74-75. Letter.
56 Lader M. Beta-adrenoceptor antagonists in neuropsychiatry: an update. J Clin Psychiatry. 1988; 49:213-223.
57. Schneier FR, Saoud JB, Campeas R, et al. Buspirone in social phobia. J Clin Psychopharmacol. 1993; 13:251-256.
58. Reich J, Noyes R, Yates W. Alprazolam treatment of avoidant personality traits in social phobic patients. J Clin Psychiatry. 1989; 50:91-95.