Lamotrigine for “Borderline Personality Disorder”: should we prescribe it? #BIGSPD19


“Borderline Personality Disorder” is one of the most contentious diagnoses within mental health. Associated with significant levels of stigma, it is frequently held up as an invalid and damaging way to characterise the difficulties faced by someone. While change is coming with the ICD-11, “Borderline Personality Disorder” is a name given to the relationship difficulties and self-destructive methods of coping that cause significant problems for some people. It is a common diagnosis with a lifetime prevalence of around 6% (at some point in their lives, 6% of the population will have met the criteria), with numbers rising as you move into higher levels of mental health service provision.

While the NICE guidelines for BPD say “Drug treatment should not be used specifically for borderline personality disorder” (NICE, 2009), the guidelines are rarely followed. Polypharmacy (the treatment of one condition using more than one medication (or via provision of a parrot)) is common (Moeller et al, 2016). In my practice, I regularly come into contact with people on an antidepressant, mood stabiliser and antipsychotic, often with some benzos and sleeping tablets to wash them down. Sometimes more than one of each. Something gets in the way of NICE guidelines being followed…

This study looked at whether lamotrigine, a mood stabiliser, would be that one drug to confound the NICE guidance and be something that could directly target BPD.

Polypharmacy is common in people with "Borderline Personality Disorder", despite the fact that NICE does not recommend drug treatment "specifically for BPD”.

Polypharmacy is common in people with “Borderline Personality Disorder”, despite the fact that NICE does not recommend drug treatment “specifically for BPD”.


276 people with a primary diagnosis of “Borderline Personality Disorder” were randomised to receive either Lamotrigine or placebo. People were excluded if they met criteria for bipolar disorder or a psychotic disorder. Participants and researchers were blinded as to which tablets people were taking. An automated randomisation system was used and everyone received ‘treatment as usual’ except that they couldn’t be prescribed another mood stabiliser or have an additional lamotrigine prescription.

The main outcome measure was the Zanarini rating scale for borderline personality disorder (ZAN-BPD). I’m not familiar with it, but it is a 36-point scale where the higher your score the worse your difficulties. The rating scale was used at the start of the trial and 12 months later. In addition it was run after 3, 6 and 9 months and other measures were used to rate depression, deliberate self-harm, social functioning, use of alcohol and other drugs, health-related quality of life and the extent to which the medication was taken as prescribed.

The description of the statistical analysis baffles me somewhat, but the authors describe a large enough sample to have 95% confidence in the data, as well as adjusting to account for site, baseline ZAN-BPD score, severity of personality disorder (simple or complex), and the extent of bipolarity. To my ignorant eyes, it looks like you can trust the results.


Everyone made a slight improvement after 12 weeks, but after a year there was no difference between those taking lamotrigine and those on placebo for any of the measures used. Adherence was fairly poor for both groups, but potentially better than usual due to the increased time staff spent with people. Where the results were adjusted for adherence…it still didn’t make any difference.


The authors conclude that:

Treating people with borderline personality disorder with lamotrigine is not a clinically effective or cost-effective use of resources.

This evidence suggests that treating people with "borderline personality disorder" with lamotrigine is not a clinically effective or cost-effective use of resources.

This evidence suggests that treating people with “borderline personality disorder” with lamotrigine is not a clinically effective or cost-effective use of resources.

Strengths and limitations

This is the first large scale study to look at the effect of lamotrigine for people diagnosed with BPD.

One strength (which is potentially a weakness) is that everyone completed a formal assessment to ascertain whether BPD was a fitting diagnosis for them. In my experience this doesn’t happen in services where ‘gut feeling’ is often used more than formal tools.

This study had a large population, meaning that the results carry more sway, and the participants were followed up over a year. Previous studies in this area were smaller samples over a shorter period.

A strength (for me) is that this study had few exclusion criteria. It could be argued that those involved were not ‘pure BPD’ (whatever that might mean), but they had the complex presentations typically found in the patrons of your local community mental healthcare team.

While there was an acknowledgement that the extra staff contact may have artificially increased adherence, it would be interesting to know if there was a positive effect of this contact regardless of whether lamotrigine, sugar pills or nothing was imbibed.

Implications for practice

The NICE guidelines are explicit that medication should not be prescribed for BPD. Despite this it is regularly dished out and this study is explicit that Lamotrigine in particular isn’t helpful. It’s interesting to think about what leads to increasingly lengthy prescriptions for this client group, in the face of guidance not to do it.

Perhaps one answer lies in an article commenting on the paper in hand by the late John Gunderson. He says “Does this mean that psychiatrists should cease prescribing medications for their patients with borderline personality disorder? No. Clinical experience has shown that medications can help build an alliance, can relieve states of distress, and can sometimes be helpful for comorbid disorders.” (Gunderson 2018)

It’s the ‘build an alliance’ part of this that worries me. Crawford et al were clear that Lamotrigine doesn’t work, it costs a fair chunk of money and it has nasty side effects. These are all good reasons not to prescribe but in the clinic, when medication is expected and the only tool on offer, how tempting is it to prescribe something on the off chance of success? As Gunderson points out, it builds a relationship. Now many people would be deeply uncomfortable giving out something against NICE guidelines that can be toxic, but perhaps there is something powerful about Drs only having certain tools in the box and patients expecting those tools to be used that leads to actions that we wouldn’t take under less pressure. Medication should definitely be given out if there’s a clear problem that it is likely to ameliorate. It definitely shouldn’t be prescribed to improve your relationship with the patient.

My favourite article ever is The Ailment by Tom Maine (Maine, 1957). Here he warns of the ‘Heroic surgical attack’ where the patient is subject to “a frenzy of treatment, each carrying more danger than the last” where the patient is made unconscious, near death, mutilated or poisoned. He warns “He who frustrates his therapist by failing to improve is always in danger of meeting primitive human behaviour disguised as treatment”. Not knowing how to help can leave us feeling incompetent and powerless, perhaps even more so for consultants where everyone will look to them to have all the answers. Before sliding down the slope that leads to ECT, better to thoughtfully work with people and risk ‘not knowing’ than to placate in ways that can harm.

It’s worth turning to Gunderson again for another lesson from this paper. While he was one of the people who promoted the label of BPD, he was also someone who was a strong advocate for helping people without resorting to specialist, exclusive therapies. His ‘good psychiatric management’ (Gunderson, 2014) emphasised the value of a structured, goal-focused relationship, but also the value of talking and listening to people. It would help us all to remember that help for the people diagnosed with BPD is unlikely to mean the contents of a blister-pack. It is more likely to be found in someone who can be consistent, honest and available.

While this paper is clear that Lamotrigine shouldn’t be prescribed, I’m aware one author is doing another study on the effectiveness of clozapine for BPD. Perhaps we will need a separate study for every possible medication before the advice not to prescribe for BPD can be heeded. Should that ever happen, I’m not sure it would slow down the prescribing.

The condensed version of the above is stop prescribing lamotrigine and think twice before prescribing anything ‘for BPD’.

"Not knowing how to help can leave us feeling incompetent and powerless".

“Not knowing how to help can leave us feeling incompetent and powerless”.

Conflicts of interest

I’ve spent some very pleasant time in the company of two of the authors, one of whom I’ve had a few discussions with around diagnosis, inclusion and ways to communicate ideas around this very topic.   I have a good social media relationship with another.

Thanks to Dr Javier Rodriguez-Mendieta for his valuable contribution.


Primary paper

Crawford MJ, Sanatinia R, Barrett B, et al (2018) The clinical effectiveness and cost-effectiveness of lamotrigine in borderline personality disorder: a randomized placebo-controlled trial (PDF). Am J Psychiatry; 175:756–764

Other references

Karen E. Moeller, Amad Din, Macey Wolfe, and Grant Holmes (2016) Psychotropic medication use in hospitalized patients with borderline personality disorder. Mental Health Clinician: March 2016, Vol. 6, No. 2, pp. 68-74.

Gunderson JG, Choi-Kain LW. (2018) Medication Management for Patients With Borderline Personality Disorder. Am J Psychiatry. Aug 1;175(8):709-711. doi: 10.1176/appi.ajp.2018.18050576.

Gunderson, J. G., & Links, P. S. (2014). Handbook of good psychiatric management for borderline personality disorder. Washington, DC: American Psychiatric Publishing.

Main T (1957) The Ailment. British Journal of Medical Psychology 30 (3), 129-145

NICE (2009). Borderline personality disorder: recognition and management. NICE clinical guideline 78. [online] Available at: [Accessed 1 Apr. 2019]

Photo credits

Share on Facebook Tweet this on Twitter Share on LinkedIn Share on Google+