Tuesday, March 22, 2022

Medication for Symptoms of Borderline Personality Disorder

A recent review of the literature on the use of medications in cases of people with borderline personality (BPD) disorder (“Pharmacological Treatments for Borderline Personality Disorder: A Systematic Review and Meta-Analysis.”  CNS Drugs. 35(10):1053-1067, 2021 10) concluded that “Despite the common use of pharmacotherapies for patients with BPD, the available evidence does not support the efficacy of pharmacotherapies alone to reduce the severity of BPD.” Additionally, “Second-generation antipsychotics, anticonvulsants, and antidepressants were not able to consistently reduce the severity of BPD.”

Well, duh. This is hardly surprising in light of the fact that personality disorders are primarily disorders of relationships and their subsequent effect on the mental state of participants in problematic interactions. Last I checked, medications do not fix relationships.

So are medications not indicated at all for people with this relatively common disorder? Well that’s nonsense as well, because sufferers often have other co-existing anxiety and mood syndromes (comorbid disorders) for which meds are most useful. The most common one in this population is panic disorder. One study showed that 40% of these people experienced panic attacks, but I think it’s much higher than that, at least in the patients who came to a psychiatrist like me for psychotherapy. It’s also true that rage attacks—another symptom of the disorder—are physiologically identical to panic attacks. You know, fight or flight.

I found out relatively early on that self injurious behavior like self-cutting or burning or bulimia often occurred when a patient found themselves in a hopeless bind in their families in which they felt it was imperative to do something to “fix” the situation but they felt helpless to do so. I discovered this the hard way. 

A patient would call me, often late at night, asking me what to do about something when they knew very well that I did not yet know enough about their situation to make any suggestions which would actually be helpful. If I dared to offer most anything, they immediately would know that I was full of crap. Talk about a sense of helplessness. I later figured out the best response in this situation was to say, “You don’t have to do anything right now. From what you’ve told me, this crisis will soon pass and be replaced by another crisis in short order.” Patients found this comment had a calming effect.

So what medications can reduce the chances of self injurious behavior by lowering the frequency of panic attacks? Oddly, when I first started private practice way back in 1979, a psychoanalyst (of all people) told me the secret: a combination of an antidepressant drug called an MAO inhibitor (this was before there were any Prozac-like drugs, which also fill the bill) with a long acting benzodiazepine like Clonazepam. Prescribing these worked far more quickly for reducing or even stopping self injurious behavior episodes than months of dialectical behavior therapy, and was quite effective.

Naturally, I was criticized for prescribing this combination. With MAOI’s, the patient would have to avoid certain foods and drugs which interact with these medications and cause an attack of severe high blood pressure. (Luckily with the Prozac-like SSRI’s, this is no longer an issue). “You mean you trusted these people to keep to the diet?!? I was asked. My answer, “Yes I do if they tell me they will stick to the diet.” Yes, and if they told me that, lo and behold, they did! I had only one patient take a proscribed medication, ending up in the ER, and I took him off the MAOI immediately.

“And benzo’s can be abused!” was the next attack. Yes, so can pretty much anything. Once again, if the patient agreed to take the meds as prescribed, and I prescribed an adequate dose (patients who were given sub-therapeutic doses tended to raise the dose on their own), seemed not to abuse them. I received further confirmation of this belief when states started to produce a data base of prescriptions for drugs of abuse, and I saw that my patients were only rarely getting them from another doc (in which case I immediately tapered them off the drug). Luckily, with the exception of Xanax and in methadone treatment centers, there is no large street market offering my patients benzo’s.

So are there studies that prove this combination is effective in the way I say? Well I’ve been on the lookout for such studies for decades, and there literally aren’t any! The closest that come are those that study SSRI’s by themselves in this population without the necessary augmentation. They show some very small effects on self-injury, but nothing substantial. Oddly, I asked the guy who did most of these studies if he ever considered doing the add-on one, and he looked at me as if he didn’t understand what I was talking about. He later gave a talk on BPD and chemicals (neurotransmitters) that help brain cells communicate, and he discussed several of them. Except one —GABA—which is the most important one in anxiety and the target of benzo drugs.

Verrrrry interrrresssssting.


  1. Great article as usual, David. I'm a big fan of SSRIs and judicious use of Clonazepam in highly anxious BPD patients, and in keeping with a GABA agonistic treatment strategy, I also prescribe Gabapentin when benzos may not be desirable (eg, previous dependency on benzos or other CNS depressants), or even adjunctively with Clonazepam which facilitates dosing of the latter within conservative limits.

    As you know, inadequately treated anxiety is a significant risk factor for suicide, particularly in these patients. Unfortunately, I frequently encounter colleagues who have an unreasonable anathema of all benzos and those who prescribe them, and thereby deprive their patients of the opportunity to lessen their turmoil.

    I'm curious about your thoughts regarding a number of states' decision to control Gabapentin? Seems to be a growing trend based on the misuse of it in combination with opiates by addicts, which has resulted in some fatalities. But I would argue that it would be far more dangerous to mix opiates and alcohol, which is readily available at any corner market in the country. My concern is that if this happens nationally, we will be removing an invaluable treatment for anxiety and pain from our non-scheduled armamentarium, not unlike what happened with tramadol. Perhaps an idea for another blog?

    1. Thank you for your the compliment and the question. Yes, I was surprised about the control measures for Gabapentin - I never saw anyone get in trouble with it back when I was practicing. The biggest issue was that the drug company that made it was pushing it for everything imaginable short of ingrown toenails.

      I also found doses up to a max of 4 mg of Clono much more effective - maybe higher than most docs who use them go, but in line with the initial dosage recommendations when valium first came out.

      The argument that benzo's when mixed with opiates can be more dangerous than an opiate alone is true but a bit bogus. I always argue that you can overdose on opiates all by themselves.

      Benzo's have been demonized by the drug companies because they are cheap, effective, and don't lead to much of a "high." Every time you see a reference to them in the literature, it's followed by something like, "but of course they can be addictive." When you see discussions of antipsychotics, you don't see that followed with "but of course they can cause diabetes and tardive dyskinesia."