Borderline Personality Disorder

John G. Gunderson, M.D.

N Engl J Med 2011; 364:2037-2042May 26, 2011

 Comments open through June 1, 2011

Article

References

Comments (1)

This Journal feature begins with a case vignette highlighting a common clinical problem. Evidence supporting various strategies is then presented, followed by a review of formal guidelines, when they exist. The article ends with the author's clinical recommendations.

A 26-year-old woman is brought to the emergency room by an anxious-looking man who explains that she became angry and suicidal, stating that her “life had no value” and that she would “like to end it all” after he criticized her. Her history includes five previous emergency room visits (twice involving self-inflicted cuts that required sutures) and two psychiatric hospitalizations after overdoses. Adolescent adjustment reaction and major depressive disorder have been diagnosed in the past, and she has been treated with sertraline, alprazolam, and aripiprazole. How should she be evaluated and treated?

THE CLINICAL PROBLEM

Borderline personality disorder (BPD) is present in about 6% of primary care patients1 and persons in community-based samples and in 15 to 20% of patients in psychiatric hospitals and outpatient clinics.2 In clinical settings, about 75% of persons with the disorder are female, although this percentage is lower in community-based samples.3,4 Patients with BPD usually enter treatment facilities after suicide attempts or after episodes of deliberate self-injury.5 Such episodes result in an average hospital stay of 6.3 days per year and nearly one emergency room visit every 2 years, rates that are 6 to 12 times those among patients with a major depressive disorder.6,7

Table 1TABLE 1Criteria for the Diagnosis of Borderline Personality Disorder. summarizes the criteria for the diagnosis of BPD, according to theDiagnostic and Statistical Manual of Mental Disorders, fourth edition.Recurrent suicidal threats or acts, when combined with fears of abandonment, are by themselves strongly indicative of the diagnosis.8Though these signature criteria make BPD easy to recognize, the diagnosis is often underused.9,10, A major reason for this is the perception that the recurrent crises, emotional volatility, and self-injurious behavior that characterize patients with BPD are willful and manipulative episodes rather than signs of an illness. 11,12

BPD is significantly heritable, with 42 to 68% of the variance associated with genetic factors13,14 — rates in the same range as those reported for hypertension.15 All the major components of the disorder (e.g., interpersonal hypersensitivity, affect
ive dysregulation, and impulsivity) have likewise been shown to track in families.14 Studies involving the use of magnetic resonance imaging or positron-emission tomography in patients with BPD have shown a hyperresponsive amygdala and impaired inhibition from the prefrontal cortex during tasks involving exposure to facial expressions, reactions to emotionally charged words, and interpersonal cooperation.16-18 There is evidence that neurohormones, such as oxytocin and opioids, mediate the exaggerated fears of rejection and abandonment that are characteristic of BPD.19 Environmental influences also appear to be important in the pathogenesis of the disorder; insecure attachment, childhood neglect or trauma, and family marital or psychiatric problems are recognized risk markers.2

Whereas BPD has long been considered a chronic and largely untreatable disorder, more recent data indicate a high remission rate (about 45% by 2 years and 85% by 10 years), with remission defined as no more than two diagnostic criteria being met for at least 12 months, and a low relapse rate (about 15%).20 In other respects, however, the prognosis remains discouraging. The suicide rate is about 8 to 10%, which is particularly high for young women, in whom the suicide rate is typically low. Moreover, even after remission, most patients with BPD have severe functional impairment, with only about 25% of patients employed full time and about 40% receiving disability payments after 10 years. 20 In addition, BPD negatively affects the course and treatment of coexisting medical conditions21,22 and other psychiatric disorders.23,24

Costs of the disorder include those related to heavy utilization of expensive health care resources and the persistent lack of productivity of patients.25 There are also considerable emotional and other costs, including those related to a variety of behaviors that are more common among patients with BPD than among those without the disorder, including reckless driving,26 domestic violence,27imprisonment,28 and pathological gambling.29

STRATEGIES AND EVIDENCE

Diagnosis

The most distinctive characteristics of patients with BPD are their hypersensitivity to rejection and their fearful preoccupation with expected abandonment (Table 1).30 Patients with the disorder feel that their lives are not worth living unless they feel connected to someone they believe really “cares,” yet their perception of “caring” generally involves unrealistic levels of availability and validation. Within such relationships, an initial idealization can dramatically shift to devaluation when rejection is perceived. In addition to this external “splitting,” patients with BPD typically have internal splitting (i.e., vacillation between considering oneself a good person who has been mistreated, in which case anger predominates, and a bad person whose life has no value, in which case self-destructive or even suicidal behavior may occur). This splitting is also evident in black-and-white or all-or-nothing dichotomous thinking.

BPD is usually diagnosed in young adulthood, but signs of it often become evident in adolescence.31 Early markers include body-image problems, severe shame, the search for exclusive relationships, extreme sensitivity to rejection, and behavioral problems, including deliberate self-harm.32 Although these phenomena may occur in adolescents without BPD, their presence is predictive of long-term social disability and an increase in the risk of adult BPD by a factor of nine.31

The diagnosis of BPD is most easily established by asking patients whether they believe that the criteria for the disorder characterize them. Clinical experience suggests that by participating in the diagnosis, patients with BPD may be more likely to accept it. Although clinicians who think of a BPD diagnosis as pejorative (chronic and untreatable) may be reluctant to disclose it,8,12 patients and their families often find it helpful to be informed of the diagnosis. Such disclosure frames treatment plans, establishes reasonable expectations, and may improve compliance.33 Clinical experience suggests that patients and their families are often relieved to learn that there are other people with similar symptoms and that there are effective treatments.

Many cases of BPD are initially misdiagnosed as depression or bipolar disorder and are treated with antidepressants or mood stabilizers.9,10 This occurs despite the fact that distinguishing BPD from such mood disorders is usually not difficult. Unlike major depressive disorder, depressive episodes in patients with BPD are marked by emptiness, shame, and a long-standing negative self-image. Unlike patients with bipolar disorders, patients with BPD are extremely sensitive to rejection and do not have periods of mania or elation.

Treatment

Psychotherapy

The primary treatment for BPD is psychotherapy.2,34 Randomized trials involving patients with BPD support the efficacy of several forms of psychotherapy, characteristically involving 2 to 3 hours per week of outpatient care for 1 or more years by psychiatrists or psychologists who have received specific training plus ongoing supervision during the trials.2,35,36 Table 2TABLE 2Four Evidence-Based Treatments for Borderline Personality Disorder.describes four types of psychotherapy that have been shown to be effective for this condition, all of which have been outlined in manuals so that therapists' adherence to the particular approach can be reliably assessed. The best studied of these methods is dialectical behavior therapy,37 but all the methods have proved to be superior to usual care (a mixed array of therapies), with significant reductions of 80 to 90% in the need for other treatments (hospitalizations, emergency room visits, and use of medication) and about 50% in episodes of self-harm or suicidality. Follow-up studies have shown that the clinical and cost benefits are maintained during a period of 2 to 5 years.38,39 However, extensive training is required to deliver these therapies, and they are not widely available.

Nonetheless, there are several principles common to these effective types of psychotherapy that can be applied in practice (Table 3TABLE 3Basic Principles of Effective Treatment for Patients with Borderline Personality Disorder.). These principles are central to a simpler-to-learn and less-intensive therapy called general psychiatric management, which in a large, multicenter trial resulted in reductions in the need for other treatments and in episodes of self-harm or suicidality that were similar to those with dialectical behavior therapy.40

The principles that are derived from effective therapies are relevant for all clinicians caring for patients with BPD. Initial visits should include a discussion of the diagnosis, expression of concern that acknowledges the patient's distress, and establishment of goals for change that are short-term and feasible. Examples of such goals are taking steps to feel better (i.e., leaving a high-stress situation or taking a medication), calling for help before losing control, improving sleep or exercise schedules, attending a self-help group (e.g., Alcoholics Anonymous), and reopening communications with an alienated friend or family member. Clinical experience indicates that such basic initial interventions can be surprisingly helpful.41

Perhaps the most difficult problems for providers to manage are deliberate self-harm, impulsive behaviors with potentially self-destructive consequences (e.g., driving under the influence of drugs or alcohol or engaging in unprotected sex), and recurrent suicidal threats. These behaviors often raise concern about suicidal intentions or plans, but they usually signify self-punishment or a way to manage feelings, and they frequently occur in association with intoxication.42 The distinction can usually be made by inquiring about the patient's intention. If suicidal intent is confirmed, the patient's safety is a priority. This may require hospitalization and usually requires contact with the patient's family members, often despite the patient's protests. In cases in which suicidal intent is denied, self-harming behaviors or threats may be effectively managed by concerned attention (including attention from others involved with the patient as much as from the clinician) and by establishing a plan for management of crises. In clinical trials of various types of psychotherapy, emergency department visits have been used as needed in such situations, but alternatives in practice include telephone or e-mail contact, use of an Alcoholics Anonymous sponsor, and referral for substance-abuse treatment as indicated. When a patient engages in repeated or escalating self-destructive behavior, a psychiatrist who has experience with BPD should be involved in the patient's care.

Another challenge for clinicians is determining how to respond to patients' frequent anger toward and alienation from their families.43 Family members often have the same feelings of anger and helplessness that clinicians experience in caring for patients with BPD.44 Failure on the part of clinicians to recognize this can aggravate the alienation of family members, yet their emotional and financial support is often needed. Clinical experience suggests that involving family members can increase their understanding of and support for patients and facilitates communication between patients and their families, as well as decreasing the emotional and financial burdens.2

Pharmacotherapy

Selective serotonin-reuptake inhibitors and other antidepressants are frequently prescribed to patients with BPD, but in randomized trials such drugs have had little if any benefit over placebo.45Data from randomized trials support the benefits of atypical antipsychotic agents (e.g., olanzapine) and mood stabilizers (e.g., lamotrigine), particularly for reducing impulsivity and aggression, in patients with BPD. However, these effects are typically modest, and side effects are common (e.g., obesity and associated hypertension and diabetes with atypical antipsychotic agents or sedation and possibly toxic effects to kidneys and during pregnancy with mood stabilizers).46,47, Thus, treatment with medications should be initiated with the understanding that they are adjuncts to psychotherapy. 33 In practice, prescribing medications may help to facilitate a positive alliance by concretely showing the physician's wish to help the patient feel better, but unrealistic expectations regarding the benefits of medications can undermine the patient's work on self-improvement.

Common concerns in prescribing medications to patients with BPD include risks of overdosing and noncompliance, but experience suggests that medications can be used without undue risk as long as patients are regularly seeing and communicating with their clinicians.2,36,48 Another common concern in practice is polypharmacy, which may occur when patients want to continue or add medications despite a lack of demonstrable benefit. In one study, 80% of patients with BPD were taking three or more medications.49 Medications with an unclear benefit should be discontinued before a new medication is initiated.

AREAS OF UNCERTAINTY

Genes that confer susceptibility to BPD have not yet been identified. Studies are needed to identify childhood and adolescent risk factors for adult BPD. Data on how these factors interact with genetic factors will allow for the identification of children at risk and the development of early-intervention strategies. More research is also needed to identify predictors of poor outcomes, such as suicide or chronicity. Our understanding of the neurobiology of BPD is incomplete; greater knowledge in this area would facilitate the development of effective pharmacotherapies. Effective treatments are also needed for associated social dysfunction. Improved dissemination of information about the types of psychotherapy that have been shown to be effective for BPD is warranted, since their availability remains limited.

GUIDELINES

The American Psychiatric Association's guidelines for the treatment of BPD note the need to actively engage patients in the process of changing themselves, the primary role of psychotherapy, and the need for a primary clinician to coordinate care.34 These recommendations antedate current knowledge about genetic and neurobiologic factors in BPD, the limitations of antidepressants, and the effectiveness of evidence-based psychotherapies.

CONCLUSIONS AND RECOMMENDATIONS

The woman in the vignette has a history of impulsive and self-destructive behavior that is suggestive of BPD. She should be questioned about other features of the disorder, including anger and fear of rejection or abandonment, in order to make a definite diagnosis. Once the diagnosis is made, clinicians should educate the patient about genetic and environmental contributors and the likelihood of a favorable response to psychotherapy. If the patient has relied on medications for treatment, she should have them reevaluated, and a referral should be made for psychotherapy with a clinician who is experienced in BPD. A thoughtful evaluation of the patient's self-harming behaviors can avert unnecessary hospitalization. By recognizing the patient's likely sensitivity to rejection, doctors can also help her develop plans to have support available, should her problems recur.

Disclosure forms provided by the author are available with the full text of this article at NEJM.org.

No potential conflict of interest relevant to this article was reported.

An audio version of this article is available at NEJM.org.

SOURCE INFORMATION

From the Psychosocial and Personality Research Program, McLean Hospital, Belmont, MA.

Address reprint requests to Dr. Gunderson at the Psychosocial and Personality Research Program, McLean Hospital, 115 Mill St., Belmont, MA 02478, or at jgunderson@mclean.harvard.edu.