Borderline personality disorder: go easy on SSRIs.

It's been said that psychiatrists are the stepchildren of medicine, and personality disorders are the stepchildren of psychiatry.

Not so long ago, patients with personality disorders were considered beyond professional help. Their symptoms were considered "character flaws," and they were relegated to an Axis II wasteland rarely considered worthy of research dollars or serious study.

Dr. S. Charles Schulz, head of psychiatry at the University of Minnesota, Minneapolis, recalls being told early in his career not to bother studying such patients, because they had no insight into their condition and were completely unreliable.

"They won't even come for a second appointment," he was advised.

Today, a small but growing body of evidence (much of it from Dr. Schulz's lab) suggests that structured psychotherapy and drugs from several classes might have therapeutic benefit for significant symptoms of some personality disorders, especially borderline personality disorder.

Dr. Schulz quickly learned that many patients with personality disorders indeed have insight into the difficulties in their lives, and may participate enthusiastically in therapy and clinical trials. He views his pursuit of a deeper understanding into personality disorders "one of the most rewarding things that I've done."

Increasing attention to personality disorders has come in part as a result of recognition of comorbid links between Axis I and Axis II disorders, with the addition of a personality disorder making depression, bipolar disorder, or schizophrenia harder to treat.

Between 50% and 80% of patients in acute care settings do have a personality disorder, and the psychiatric community has begun to sit up and take notice, said Dr. Deanna Mercer of the University of Ottawa and Ottawa Hospital.

"Personality disorders may be the hypertension and high cholesterol of cardiac disease. If we keep ignoring them, we won't be as effective at treating Axis I disorders," she noted.

Many clinicians historically just followed the symptoms in an attempt to treat patients with personality disorders, perhaps leaning toward atypical antipsychotics for Cluster A disorders (paranoid personality disorder, schizoid personality disorder, and schizotypal personality disorder); anticonvulsants and mood stabilizers for Cluster B (antisocial personality disorder, borderline personality disorder, histrionic personality disorder, and narcissistic personality-disorder); and maybe antidepressants and/or anxiolytic agents for Cluster C disorders (avoidant personality disorder, dependent personality disorder, and obsessive-compulsive personality disorder).

Certainly, psychiatry has a long way to go to find equivalents to ACE inhibitors and statins that will control patients' underlying issues and thereby improve care of acute psychiatric syndromes. But a burgeoning library of clinical trials is beginning to light the way to a more systematic, scientific approach to prescribing.

Borderline personality disorder is by far the most studied in psychopharmacology, although limited studies are also being conducted in other personality disorders. Mount Sinai Medical Center in New York, for example, is conducting extensive studies of approaches to schizotypal personality disorder, with particular attention to the possible usefulness of low-dose atypical antipsychotics.

In 2008, Dr. P. Francis Abraham coauthored a review of 28 randomized, double-blind, controlled trials of drugs used in the treatment of borderline personality disorder, concluding that a wide variety of agents showed evidence of improvement "although often circumscribed and variable" (J. Affect. Disord. 2008;111:21-30).

In general, anticonvulsants and atypical antipsychotics appeared to show greater benefit in this population than did antidepressants, reported Dr. Abraham, a psychiatrist with The Nord Center, a community-based comprehensive behavioral mental health center in Lorain, Ohio.

Dr. Abraham's clinical experience parallels what the studies showed, he said in an interview. "It is my perception that SSRIs are used all too frequently as an indiscriminate, default option due to their ease in prescribing and perceived safety," he said. "However, in my opinion, their efficacy in borderline personality disorder based on the scientific data is fairly unimpressive."

In general, as with bipolar disorder and schizophrenia, "positive" and maniclike symptoms of borderline personality disorder seem more responsive to medication than depression. For mood-related symptoms, he tends to turn to mood stabilizers and atypical antipsychotics.

"As long as one is prepared to recognize and manage potentially complicating factors which can be more likely in personality disordered patients, such as medication transference issues, diversion, and medication lethality in overdose, I think prescribing medicine can be quite useful in patients with personality disorders in general]; and, in borderline personality, is supported by the medical literature," he said.

Dr. Mercer and associates from the University of Ottawa conducted a second meta-analysis, this one specifically looking at symptoms of depression and anger in borderline personality disorder (J. Pers. Disord. 2009;23:156-74).

The findings were somewhat confounding, but again suggestive of benefit.

This study found a large pooled effect size for mood stabilizers in targeting anger symptoms, with the exception of divalproex. On the other hand, divalproex and carbamazepine had a moderate effect on depression.

Antidepressants had only a small effect on depression, but a moderate impact on anger symptoms.

Antipsychotics also had a moderate effect on anger. None of the antipsychotics showed benefit for depression.

The authors pointed to several serious limitations in attempting to draw conclusions from studies of medications in borderline personality disorder, including highly variable treatment duration, dropout rates, and patients receiving concomitant psychotherapy, the latter ranging from 0% to 100% in the trials they reviewed.

The pioneering nature of the study of psychopharmacology for personality disorders, compounded by limited resources, leaves solid, generalizable evidence in the field "very, very thin," said Dr. Schulz, leaving clinicians in a difficult spot when they attempt to select the very best agent (and psychotherapeutic approach) to help a specific patient.

His reading of the literature and clinical experience suggest that SSRIs do play a role, but more for symptoms of impulsivity than for depression.

Divalproex and lamotrigine may help impulsivity, aggressive behavior, and moodiness in patients with borderline personality, while risperidone and quetiapine may improve anxiety and paranoic symptoms.

Medications that may be overused, without evidence of benefit, include benzodiazepines, which may be disinhibiting; and tricyclic antidepressants, which may be overly stimulating as well as posing a risk of lethality in overdose, he said.

Despite a host of studies including some by Dr. Schulz--that allude to possible benefits of olanzapine in this population, he recently published a large negative study comparing the novel antipsychotic to placebo (Brit. J. Psychiatry 2008;193:485-92).

In that study, 155 patients assigned to receive olanzapine experienced more weight gain and higher rates of abnormal prolactin levels, but did not experience significantly more benefit in terms of borderline personality disorder symptoms than did 159 patients assigned to placebo.

One important take-home message from the trial is that patients in both groups did very well, said Dr. Schulz in an interview.

"We have noticed since the mid-'80s that patients appreciate the structure of the clinical trial set-up," he said.

Some express relief at being given a diagnosis that explains life-long symptoms.

Many seem to benefit from completing weekly rating scales and symptom checklists, so that they can track their own progress.

They also have the security of knowing there is a crisis coordinator on call and a consistent place to turn for help.

"I try to incorporate some of these elements in my clinical practice," he said.

The positive response of patients to structure also may go a long way to explain the robust response of patients with borderline personality disorder to highly structured forms of psychotherapy, including University of Washington, Seattle, psychologist Marsha Line-han's dialectical behavior therapy, and the STEPPS Group Treatment Program developed at the University of Iowa, Iowa City.

Arranging for a multidisciplinary approach that incorporates structured cognitive-behavioral therapy and medication can be a challenge for community psychiatrists, but the effort makes a real difference in patients, experts consulted for this column agreed.

"I interpret the evidence to show that there is a real and significant, but likely limited, benefit to prescribing drugs for personality disorders," said Dr. Abraham. "Therefore, employing any other helpful modality, especially psychotherapies with proven efficacy, would seem entirely appropriate."

The employment of psychotherapy along with a thoughtful, limited range of medications for symptom stabilization, makes sense to Dr. Mercer as well.

"Some patients come to me on five or six medications, maybe two from the same class, which tells me someone was desperate to help this individual and got caught on a medication merry-go-round," she said.

A more reasonable approach is to target one to two important symptoms and treat the patient with the aim of improvement, not "cure," that might allow full participation in a structured, psychotherapeutic program designed to enhance insight, awareness, and productive behavior, she said.

Dr. Abraham and Dr. Mercer reported no relevant financial disclosures. Dr. Schulz reported that he has conducted research on atypicals and personality disorders for the past 10 years, including studies supported by several companies, including Eli Lilly Pharmaceuticals and Abbott Pharmaceuticals.

By Betsy Bates. Share your thoughts and suggestions at [e-mail].