The final assignment for this class will be a 10-page critical review of the drug treatment for a psychiatric disorder (broadly defined to include psychological and neurological disorders as well). Th
CLINICAL PSYCHOPHARMACOLOGY UPDATE Rafael A. Rivas-Vazquez, Editor Pharmacologic Treatment of Personality Disorders Rafael A. Rivas-VazquezNeurologic Center of South Florida and Miami Research Associates Mark A. Blais Harvard Medical School and Massachusetts General Hospital The era of evidence-based medicine has fostered heightened scrutiny of the manner in which therapeutic interventions are formulated, taught, and implemented. In psychopharmacology, this has translated into an emerging paradigm of developing re- search activities on the basis of specific neurobiological theories and subsequently utilizing the results of well-controlled trials to guide clinical practice. Such data ultimately inform the develop- ment of health care policy and guidelines for treating specific disorders. The application of this paradigm to the pharmacologic treatment of personality disorders (PDs) has introduced some challenges based primarily on the inherent complexities of these pathological entities. Long since felt to be the purview of psycho- social treatments, PDs are increasingly being investigated to de- termine the impact that psychotropic medications may have on their expression and course. Concern has been raised as to the underdetection and undertreatment of PDs, particularly in light of their resultant functional impairment and disability, their adverse impact on the treatment of comorbid Axis I disorders, and the increased use of health resources associated with their presence (compared with patients without an Axis II disorder; Bender et al., 2001). In this article, we review the recent literature on pharma- cologic treatment of PDs, discussing the existing models that guide pharmacotherapy of these disorders and reviewing specific classes of medications that appear to play a beneficial role. Existing Paradigm for Pharmacological Treatment of Personality Disorders Treatment of PDs has traditionally been viewed with some pessi- mism. The deeply ingrained nature of personality characteristics pro- duced a sentiment that root-cause treatment—typically psychothera- py—would have to be long-term, intensive, and guarded in its prognosis. From this perspective, pharmacotherapy was secondary, short-term, and highly symptom-oriented. Over the past 10 –15 years, pharmacological and biological studies— coupled with clinical expe- rience— have prompted a reconsideration of pharmacotherapy forPDs. Medications are now considered as important as psychotherapy in the treatment of PDs, particularly given findings that 81% of patients with PDs report that they have received psychotropic medi- cations at some time during their treatment (Bender et al., 2001).
Medications are now clearly valued and used in the treatment of borderline PDs, to the point that some psychiatrists consider it uneth- ical to withhold medications from these patients; psychologists as a group are also highly prone to advocate for the use of medications in their treatment of borderline PDs (Gunderson, 2001).
The current pharmacotherapeutic paradigm for PDs has evolved along three separate lines of reasoning (Cloninger & Svrakic, 2000; Kapfhammer & Hippius, 1998; Siever & Davis, 1991):
1.Pharmacotherapy directly influences PDs. The central as- sumption of this approach (also referred to as causal pharmaco- therapy) is that medications alter underlying neurobiological dis- positions that cause maladaptive traits. From this perspective, pharmacological manipulation of the neurobiological systems modifies the disposition and promotes better functioning (includ- ing enhanced learning and experiential events), gradually produc- ing long-term changes in behavior and adaptation.
2.Pharmacotherapy exerts an effect over core or nuclear symptom clusters.This approach targets core or nuclear symptom clusters felt to be highly characteristic of certain PDs. A body of literature sup- ports the presence of four factors or dimensions reflecting fundamen- tal dysfunction in the areas of cognition or perceptual organization, behavioral control, affective stability, and anxiety suppression (Siever & Davis, 1991). This approach appears to be the one most often implemented in current clinical practice.
3.Pharmacotherapy exerts its therapeutic effect by treating comor- bid Axis I disorders.This view is based upon the finding that Axis I disorders are commonly found to be comorbid with PDs. In this model, the therapeutic effect of medications occurs primarily at the comorbid Axis I level, which improves the clinical status of the patient and increases the probability of better outcome produced by other forms of therapy. This view, although valid and practical, is probably most consistent with the traditional approach of treating PDs with medications and has received the least amount of attention in the reconceptualized paradigm (Cloninger & Svrakic, 2000). Identifying the Target:
Neurobiology of Clusters and Dimensions A basic tenet of pharmacology is that of clearly identifying the target symptoms at which intervention is aimed and for which data will be collected and monitored over time to determine response and efficacy. In the case of treating PDs pharmacologically, this may CORRESPONDENCE CONCERNING THIS COLUMN should be addressed to Rafael A. Rivas-Vazquez, Neurologic Center of South Florida, Baptist Medical Arts Building, 8940 North Kendall Drive, 802-E, Miami, Florida 33176.
E-mail: [email protected] M ARK A. B LAIS received his PsyD in clinical psychology from Nova Southeastern University in 1990 and is chief psychologist of inpatient service at Massachusetts General Hospital and associate professor at Har- vard Medical School. Professional Psychology: Research and Practice, 2002, Vol. 33, No. 1, 104 –107 Copyright 2002 by the American Psychological Association, Inc. 0735-7028/02/$5.00 DOI: 10.1037//0735-7028.33.1.104 104 require a bit of a conceptual shift. A PD is defined as a pattern of inner experience and external behavior (manifested by cognitive processing of the environment, affective regulation, interpersonal functioning, and impulse control) that remains inflexible over time, impairs social and occupational functioning, and pervades across a wide range of situations (American Psychiatric Association [APA], 2000). In the current classification system, a categorical approach (which assumes that PDs are qualitatively distinct syndromes) groups the disorders into three clusters:Cluster A,including paranoid, schizoid, and schizotypal PDs, which may manifest in cognitive distortions and an interpersonal style that is odd, eccentric, or detached;Cluster B, consisting of antisocial, borderline, histrionic, and narcissistic PDs, which often involve behavior that appears dramatic, erratic, impul- sive, aggressive, or affectively dysregulated; andCluster C,which includes avoidant, dependent, and obsessive-compulsive PDs that tend to involve fear, anxiety, apprehension, or perceived avoidance of harm (APA, 2000). A review of the literature reveals that the greatest number of biological and pharmacologic studies have focused almost exclusively on borderline, schizotypal, and antisocial PDs.
Research geared at identifying specific neurobiological markers for various PDs has yielded interesting—albeit preliminary—results. A consistent finding with borderline PD is serotonin dysregulation, manifested by reduced levels of serotonin metabolites in cerebrospinal fluid, blunted neuroendocrine responses to serotonergic agonists, and diminished levels of serotonin synthesis and reactivity in prefrontal cortex and corticostriatal pathways, as evidenced by positron emission tomography (PET) imaging (Gurvits, Koenigsberg, & Siever, 2000; Leyton et al., 2001). This finding may correlate with impulsive, aggressive, and self-injurious activity, including suicidal behaviors.
Anatomically, volumetric analysis of magnetic resonance imaging (MRI) scans conducted on a small series of patients with borderline PD revealed some structural variability in the frontal lobes and hip- pocampus (the latter is most prominent in borderline patients report- ing a history of abuse; Driessen et al., 2000).
Studies of patients with schizotypal PD, often described as a clinical entity existing along the schizophrenia spectrum, have consistently yielded a pattern of processing deficits—sensory gat- ing, working memory, verbal learning, and sustained attention— that are similar to those noted with schizophrenic patients, al- though the pattern of dysfunction may be less severe and more focal (Cadenhead, Light, Geyer, & Braff, 2000; Kirrane & Siever, 2000). Results from MRI studies have suggested abnormalities in the temporal lobe, corpus callosum, and thalamic nuclei (Byne et al., 2001). Furthermore, schizotypal patients do not demonstrate the same degree of diminished frontal lobe volume as do patients with schizophrenia, which may correlate with the comparably less severe pattern of cognitive and social deterioration demonstrated by schizotypal patients (Kirrane & Siever, 2000).
There is less neurobiological data appearing in the recent liter- ature regarding antisocial PD, although data continue to implicate subtle frontal dysfunction (as measured by imaging, neuropsycho- logical, and evoked potentials studies) and reduced autonomic arousal and reactivity in the face of distressing stimuli. Recently, Raine, Lencz, Birhle, LaCasse, and Colletti (2000) identified an 11% reduction in prefrontal gray matter volume in patients with antisocial PD (who had no history of acquired brain lesions or insult), and they asserted that this finding may subserve the clin- icopathological features observed with this disorder.
Although the categorical approach is used in the present diag- nostic system, this may not be the most useful paradigm whenapplying pharmacological principles to PDs (Cloninger & Svrakic, 2000). Bolstered in part by biological studies, a dimensional ap- proach organized around specific core symptoms—rather than a specific diagnosis—may have greater utility for initiating and monitoring pharmacotherapy with PDs. As noted above, four dimensions of PD symptom clusters have been characterized that show some degree of correspondence to the current PD categories and assume a neurobiological substrate that can potentially serve as a rationale for treatment selection (Siever & Davis, 1991):
1. Thecognitive–perceptual dimension,generally correlating with Cluster A disorders, includes psychosis, unusual perceptual experiences, dissociation, suspiciousness, and odd or magical thinking. Behaviors from this dimension may be the result of dysfunction within the dopaminergic system.
2. Theimpulsive–aggressive dimensionserves as a prominent factor for Cluster B PDs. Anger, aggression, and behavioral dis- inhibition constitute primary impairments in this dimension and may implicate dysfunction in the serotonergic system.
3. Theaffective instability dimensionis also closely tied with the Cluster B disorders and consists of mood dysregulation, depres- sion, dysphoria, and emotional lability. These behaviors may have broad neurotransmitter underpinnings, possibly related to dysfunc- tion in serotonergic, cholinergic, or noradrenergic systems.
4. Theanxiety–inhibition dimension,predominantly associated with Cluster C disorders, consists of anxiety, fear, and manifesta- tions of behavioral inhibition geared at reducing exposure to anxiety-provoking situations. This dimension may reflect a distur- bance in autonomic function and may be mediated by serotonergic and noradrenergic systems.
Speculation and identification of specific neurotransmitters sub- serving the PD dimensions translate into important pharmacolog- ical targets (Soloff, 2000). It should be noted that while a certain degree of concordance exists between clusters and dimensions, the heterogeneity of PDs may result in expression of behaviors from several different dimensions within a single PD. For example, although patients with borderline PD typically demonstrate behav- iors from the impulsive–aggressive and affective instability dimen- sions, some borderline patients may exhibit behaviors along the cognitive–perceptual dimension (APA, 2001). This reinforces the notion of basing treatment selection on target-symptom groups rather than on a specific diagnosis. Although results from clinical trials reported in the literature typically refer to the application of a medication to a specific PD diagnosis, efficacy data are typically collected on one of the dimensions noted above (in addition to ratings of effect on global functioning). Application of Specific Classes of Medications to PDs Antipsychotics One of the earliest reports of applying medications to PDs involved the antipsychotic trifluoperazine (Vilkin, 1964). 1Since 1Drug names for the medications discussed in this article are as follows:
alprazolam (Xanax), carbamazepine (Tegretol), chlorpromazine (Thor- azine), clonazepam (Klonopin), clozapine (Clozaril), divalproex sodium (Depakote), fluoxetine (Prozac), haloperidol (Haldol), loxapine (Loxitane), naltrexone (ReVia), olanzapine (Zyprexa), perphenazine (Trilafon), risperi- done (Risperdal), sertraline (Zoloft), thiothixene (Navane), trifluoperazine (Stelazine), and venlafaxine (Effexor). 105 CLINICAL PHARMACOLOGY UPDATE then, several conventional antipsychotics, including haloperidol, perphenazine, thiothixene, chlorpromazine, and loxapine, have been investigated with variable and modest therapeutic efficacy in PDs (Coccaro, 1998). Patients in these studies demonstrated either borderline or schizotypal PDs. Efficacy was noted for cognitive– perceptual distortions, as would be expected, but also for symp- toms from other dimensions, including the impulsive–aggressive and affective instability dimensions (Kapfhammer & Hippius, 1998; Soloff, 2000). The advent of the atypical antipsychotics (with their relatively decreased tendency to produce motor side effects and expanded efficacy for improving negative symptoms [e.g., alogia, avolition, apathy, affective flattening] associated with schizophrenia) has prompted a decrease in the use of the conven- tional antipsychotics with PDs.
Several studies have provided preliminary data supporting the efficacy of the atypical antipsychotic clozapine for treating psy- chotic symptoms demonstrated by borderline PD, as well as for treating impulsivity, aggression, and self-mutilation (Benedetti, Sforzini, Colombo, Maffei, & Smeraldi, 1998; Chengappa, Ebel- ing, Kang, Levine, & Parepally, 1999). However, given clozap- ine’s tendency to produce agranulocytosis, clinicians are increas- ingly beginning to use the other atypical antipsychotics (APA, 2001). Case reports and open-label trials have suggested that olanzapine and risperidone may have efficacy in patients with borderline PD who exhibit either self-mutilation or mild depres- sion (Coccaro, 1998; Hough, 2001; Schulz, Camlin, Berry, & Jesberger, 1999). There has been one report of risperidone reduc- ing aggression and impulsivity associated with antisocial PD (Hirose, 2001).
Mood Stabilizers As a group, this class of medications shows particular promise for treating patients with PDs that load high on the impulsive– aggressive and affective instability dimensions. Lithium was the first mood stabilizer to be studied, with results suggesting thera- peutic efficacy for aggression and emotional lability displayed by patients with PDs (Coccaro, 1998). The anticonvulsant carbamaz- epine has also been studied for PDs associated with behavioral impulsivity, but it has demonstrated equivocal results (Soloff, 2000). Recent attention has focused on another mood-stabilizing anticonvulsant—divalproex sodium—with results from several open-label studies indicating efficacy for borderline patients who demonstrate affective instability, irritability, and impulsive– aggressive behavior (Kavoussi & Coccaro, 1998; Townsend, Cam- bre, & Barbee, 2001). These findings have recently been supported by results from a double-blind, placebo-controlled trial in which divalproex sodium was more effective than placebo in reducing aggression and depression in patients with borderline PD (Hol- lander et al., 2001).
Antidepressants Older antidepressants—the monoamine oxidase inhibitors (MAOIs) and tricyclic antidepressants (TCAs)—have demon- strated variable results with PDs. With the advent of newer anti- depressants, coupled with the cardiotoxicity and other undesirable side effects associated with the MAOIs and TCAs, these older agents have fallen into relative disuse. The selective serotonin reuptake inhibitors (SSRIs) have received the most attention, pri-marily with borderline PD. These agents are thought to represent the usual starting point as first-line agents for impulsive– aggressive symptoms and affective dysregulation (APA, 2001; Gunderson, 2001). Fluoxetine has been the most extensively stud- ied, with results from three randomized, placebo-controlled studies of primarily borderline patients demonstrating efficacy on mea- sures of anger, rage, aggression, depression, irritability, and self- mutilation (APA, 2001). Data are also available for sertraline, primarily from case reports and open trials. A similar pattern of response should be expected from the other SSRIs, and there is no reason to suspect that one SSRI will have significantly greater efficacy over another. The application of the novel antidepressant venlafaxine, which has both serotonin and norepinephrine reuptake blockade, should be of interest. One open trial has noted a decrease in self-injurious behavior and overall symptomatology in a group of patients with borderline PD (Markovitz & Wagner, 1995). SSRI treatment of avoidant PD has been discussed in the literature, although there is some difficulty in overlap between this PD and a generalized form of the Axis I disorder social phobia (which has been found to be responsive to SSRIs and other newer antidepressants). Benzodiazepines Despite an early finding of efficacy for diazepam in patients with PDs (Vilkin, 1964), this class of agents has not been consis- tently or systematically studied. Case reports and open-label trials of alprazolam and clonazepam suggest some efficacy in decreasing anxiety and fear-related symptoms associated primarily with bor- derline PDs (Kapfhammer & Hippius, 1998; Soloff, 2000). How- ever, there is at least one report that application of these agents to PDs can lead to behavioral disinhibition and dyscontrol (Coccaro, 1998). This, in conjunction with the high potential for abuse and dependence, has served as a bit of contraindication for using this class with PDs (Cloninger & Svrakic, 2000). Application of these antianxiety agents to PDs associated with the anxiety–fear dimen- sion, most notably avoidant, dependent, and obsessive-compulsive PD, has not been reported.
Opioid Antagonists On the basis of findings that self-injurious behavior may be partially produced by dysregulation of endorphins (endogenous opioid-like neuropeptides), the opioid antagonist naltrexone has been used with some success in various populations (e.g., children with mental retardation and other developmental disabilities).
There is one case report of naltrexone reducing self-mutilation in a patient with borderline PD (McGee, 1997). Results from an open-label trial of 13 patients with borderline PD suggested that naltrexone may reduce the duration and intensity of cognitive– perceptual symptoms of dissociation (Bohus et al., 1999). Summary Mounting data would appear to justify more than a secondary role for medications in the overall treatment strategies for PDs.
The current paradigm posits that pharmacological intervention impacts either underlying neurobiological dysfunction or core symptoms that are fundamental aspects of these PDs (rather than simply treating comorbid Axis I symptoms). Of the recognized 106 CLINICAL PHARMACOLOGY UPDATE disorders in the current classification system, borderline, schizo- typal, and antisocial PDs have been investigated to the greatest degree, with little if any work being done with the other PDs.
Dimensions of symptom clusters represent the most likely phar- macological targets, and these include cognitive–perceptual dis- tortions, aggression–impulsivity, affective instability, and fear– anxiety. Of the currently available agents, preliminary data suggest that the atypical antipsychotics and mood stabilizers may provide the broadest spectrum of efficacy in the most troublesome dimen- sions (notably, cognitive distortions, aggression–impulsivity, and affective instability). However, algorithms developed for guiding the pharmacological treatment of borderline PD indicate that the SSRIs and newer antidepressants occupy a first-line position for affective instability and aggression–impulsivity (APA, 2001). In light of the current data and clinical application, there is general agreement that more well-controlled studies are needed to better elucidate the efficacy, safety, and application of medications to various PDs.
Initiation of pharmacotherapy with this population also requires increased vigilance regarding risk-management issues. Medication abuse, in the form of overdose or noncompliance, takes place at least 50% of the time during the treatment of patients with bor- derline PD (Gunderson, 2001), requiring that clinicians be active in monitoring for adverse and potentially lethal events. Psycho- therapy and psychosocial interventions remain critical for many PDs, and it has been well stated that the art of pharmacotherapy for PDs involves employing the concurrent use of nonpharmacologi- cal therapies, addressing potentially unrealistic expectations, and emphasizing that medications are not a panacea (Janicak, Davis, Preskorn, & Ayd, 1997). References American Psychiatric Association. (2000).Diagnostic and statistical man- ual of mental disorders(4th ed., rev.). Washington, DC: Author.
American Psychiatric Association. (2001). Practice guideline for the treat- ment of patients with borderline personality disorder.American Journal of Psychiatry, 158,1–52.
Bender, D. S., Dolan, R. T., Skodol, A. E., Sanislow, C. A., Dyck, I. R., McGlashan, T. H., et al. (2001). Treatment utilization by patients with personality disorders.American Journal of Psychiatry, 158,295–302.
Benedetti, F., Sforzini, L., Colombo, C., Maffei, C., & Smeraldi, E. (1998).
Low-dose clozapine in acute and continuation treatment of severe bor- derline personality disorder.Journal of Clinical Psychiatry, 59,103– 107.
Bohus, M. J., Landwehrmeyer, G. B., Stiglmayr, C. E., Limberger, M. F., Bohme, R., & Schmahl, C. G. (1999). Naltrexone in the treatment of dissociative symptoms in patients with borderline personality disorder:
An open-label trial.Journal of Clinical Psychiatry, 60,598–603.
Byne, W., Buchsbaum, M. S., Kemether, E., Hazlett, E. A., Shinwari, A., Mitropoulou, V., & Siever, L. J. (2001). Magnetic resonance imaging of the thalamic mediodorsal nucleus and pulvinar in schizophrenia and schizotypal personality disorder.Archives of General Psychiatry, 58, 133–140.
Cadenhead, K. S., Light, G. A., Geyer, M. A., & Braff, D. L. (2000).
Sensory gating deficits assessed by the P50 event-related potential in subjects with schizotypal personality disorder.American Journal of Psychiatry, 157,55–59.
Chengappa, K. N., Ebeling, T., Kang, J. S., Levine, J., & Parepally, H.
(1999). Clozapine reduces severe self-mutilation and aggression in psy- chotic patients with borderline personality disorder.Journal of Clinical Psychiatry, 60,477–484.Cloninger, C. R., & Svrakic, D. M. (2000). Personality disorders. In B. J.
Sadock & V. A. Sadock (Eds.),Comprehensive textbook of psychiatry (7th ed., pp. 1723–1764). Philadelphia: Lippincott/Williams & Wilkins.
Coccaro, E. F. (1998). Clinical outcome of psychopharmacologic treatment of borderline and schizotypal personality disordered subjects.Journal of Clinical Psychiatry, 59(Suppl. 1), 30–35.
Driessen, M., Herrmann, J., Stahl, K., Zwaan, M., Meier, S., Hill, A., et al.
(2000). Magnetic resonance imaging volumes of the hippocampus and the amygdala in women with borderline personality disorder and early traumatization.Archives of General Psychiatry, 57,1115–1122.
Gunderson, J. G. (2001).Borderline personality disorder: A clinical guide.
Washington, DC: American Psychiatric Publishing.
Gurvits, I. G., Koenigsberg, H. W., & Siever, L. J. (2000). Neurotransmit- ter dysfunction in patients with borderline personality disorder.Psychi- atric Clinics of North America, 23,27–40.
Hirose, S. (2001). Effective treatment of aggression and impulsivity in antisocial personality disorder with risperidone.Psychiatry & Clinical Neurosciences, 55,161–162.
Hollander, E. H., Allen, A., Prieto Lopez, R., Bienstock, C. A., Grossman, R., Siever, L. J., et al. (2001). A preliminary double-blind, placebo- controlled trial of divalproex sodium in borderline personality disorder.
Journal of Clinical Psychiatry, 62,199–203.
Hough, D. W. (2001). Low-dose olanzapine for self-mutilation behavior in patients with borderline personality disorder.Journal of Clinical Psy- chiatry, 62,296–297.
Janicak, P. G., Davis, J. M., Preskorn, S. H., & Ayd, F. J. (1997).
Principles and practice of psychopharmacotherapy.Baltimore: Wil- liams & Wilkins.
Kapfhammer, H.-P., & Hippius, H. (1998). Pharmacotherapy in personality disorders.Journal of Personality Disorders, 12,277–288.
Kavoussi, R. J., & Coccaro, E. F. (1998). Divalproex sodium for impulsive aggressive behavior in patients with borderline personality disorder.
Journal of Clinical Psychiatry, 59,676–680.
Kirrane, R. M., & Siever, L. J. (2000). New perspectives on schizotypal personality disorder.Current Psychiatry Reports, 2,62–66.
Leyton, M., Okazawa, H., Diksic, M., Paris, J., Rosa, P., Mzengeza, S., et al. (2001). Brain regional alpha-[11C]methyl-L-tryptophan trapping in impulsive subjects with borderline personality disorder.American Jour- nal of Psychiatry, 158,775–782.
Markovitz, P. J., & Wagner, S. C. (1995). Venlafaxine in the treatment of borderline personality disorder.Psychopharmacology Bulletin, 31,773– 777.
McGee, M. D. (1997). Cessation of self-mutilation in a patient with borderline personality disorder treated with naltrexone [Letter].Journal of Clinical Psychiatry, 58,32–33.
Raine, A., Lencz, T., Birhle, S., LaCasse, L., & Colletti, P. (2000).
Reduced prefrontal gray matter volume and reduced autonomic activity in antisocial personality disorder.Archives of General Psychiatry, 57, 119–127.
Schulz, S. C., Camlin, K. L., Berry, S. A., & Jesberger, J. A. (1999).
Olanzapine safety and efficacy in patients with borderline personality disorder and comorbid dysthymia.Biological Psychiatry, 46,1429– 1435.
Siever, L. J., & Davis, K. L. (1991). A psychobiological perspective on personality disorders.American Journal of Psychiatry, 148,1647–1568.
Soloff, P. H. (2000). Psychopharmacology of borderline personality dis- order.Psychiatric Clinics of North America, 23,169–192.
Townsend, M. H., Cambre, K. M., & Barbee, J. G. (2001). Treatment borderline personality disorder with mood instability with divalproex sodium: A series of ten cases.Journal of Clinical Psychopharmacol- ogy, 21,249–251.
Vilkin, M. I. (1964). Comparative chemotherapeutic trial in treatment of chronic borderline patients.American Journal of Psychiatry, 120,1004. 107 CLINICAL PHARMACOLOGY UPDATE