The One Medication Every Psychiatrist Needs to Know About

Getting to know metformin is one way child and adolescent psychiatrists can be more prepared to address metabolic side effects of atypical antipsychotic medications.  

Metformin (Trade names: Glucophage, Glucophage XR, Glumetza, Fortamet, Riomet)

Metformin is first-line treatment for overweight patients with Type II diabetes and is FDA approved for diabetic children age 10 and older. Metformin improves sensitivity to insulin and limits the production of glucose in the liver. Typical pediatric dosing of metformin starts at 500 mg per day for the first week with increases to a maximum dose of 1000 mg twice a day. Modest weight loss can occur in overweight youth at risk for diabetes (Park MH et al Diabetes Care 2009;32:1743-1745). Typical side effects of metformin include nausea, vomiting, gas, bloating, diarrhea, and loss of appetite.  

Practice parameters for the use of atypical antipsychotic medications in children and adolescents were published by the American Academy of Child & Adolescent Psychiatry (AACAP) in 2011. Weight management interventions are encouraged if medication-induced weight gain results in a BMI >90th percentile for age. Current practice guidelines do not address routine monitoring of insulin levels, but refer to checking hemoglobin A1c (HbA1c) as indicated. I personally follow fasting insulin levels in overweight patients as I worry that if I wait for an elevated HbA1C, it may be too late to intervene. Most of my patients on metformin stop gaining weight or even lose weight—up to 30 lbs.

This clinical observation is supported by the literature. Children who gained significant (more than 10%) weight within the first year of treatment with olanzapine, quetiapine, or risperidone were selected for a 16-week, double-blind, placebo-controlled study. BMI stabilized in the metformin treatment group while patients receiving placebo continued to gain 0.31 kg/wk (Klein DJ et al Am J Psychiatry 2006;163:2072-2079). Children and adolescents with autism spectrum disorder (ASD) also demonstrated a significant reduction in weight gain in a more recent randomized trial (Anagnostou E et al JAMA Psychiatry 2016;73(9):928-937). Metformin was reported to be well tolerated in the 60 patients with ASD who received doses of up to 500 mg twice a day (ages 6-9) or 850 mg twice daily (ages 10-17).

Case example

Charlie was a 17-year-old male with a longstanding diagnosis of ADHD when his impulsivity and irritability escalated to the level of a mood disorder. He was frequently suspended for fighting. Although he did not yet meet criteria for bipolar disorder, his family was desperate to try a mood stabilizer as he failed intensive behavioral therapies and several antidepressant trials. He started a risperidone trial following a discussion of risks of off-label prescribing. The conversation focused on lifestyle interventions to prevent weight gain given his body mass index (BMI) of 38.5 kg/m2. The improvement in his behavior was immediate but his BMI peaked at 42.1. At that time, his fasting metabolic labs revealed a normal glucose (92 mg/dl) but an elevated insulin level of 43.3 mU/L (normal range for reference lab 3-25 mU/L). Attempts to taper risperidone failed due to breakthrough irritability and his family resisted a major medication change in the middle of his senior year. The use of metformin 500 mg twice daily was supported by his pediatrician. No side effects were noted at his follow-up appointment and the dose was increased to 850 mg twice daily. Three months later his BMI had decreased to 40.7 and his repeat insulin level was 39.2. These numbers were far from ideal, but his family appreciated efforts to mitigate risks as he graduated from high school and transitioned to college.  

Weighing off-label prescribing

The stakes for patients like Charlie are high as effective treatment carries potential lifelong medical implications. Despite stronger labeling, antipsychotic drugs are still prescribed off-label for children with eating disorders, tic disorders, OCD, PTSD, and disruptive aggression. There is a lack of long-term data on the benefits of metformin for medication related weight gain. A fair number of patients can’t tolerate metformin due to GI side effects and parents often have difficulty administering two doses a day with food on school days. Lactic acidosis is rare but the muscle pain could be confused with dystonia or even neuroleptic malignant syndrome (NMS).

Metformin is not a “magic pill” for weight loss but appears to be clinically useful in children and adolescents who require long term treatment with atypical antipsychotics. Families that worry about weight gain will likely be glad that their prescriber “met” metformin.

One Psychiatrist’s Take on Managing Tardive Dyskinesia

I routinely order ECGs, check weights, and draw labs to assess metabolic status for my patients taking second generation antipsychotics (SGAs). Before 2017, I didn’t rigorously screen for tardive dyskinesia (TD). Effectively this meant I didn’t care about it. In my private practice, I hadn’t noticed anyone with obvious TD symptoms, and no one had complained of involuntary movements, so I assumed that TD was a “solved problem”.

However, in March 2016, I met a 27-year-old man who had schizophrenia and obvious evidence of TD, with dyskinesia in facial muscles, extremities, and even his trunk. The best option I could offer at the time was to remove one of the two SGAs he was on. His father (and legal guardian) adamantly refused any change in medications, because that was the only regimen that had ever kept his son out of the hospital. In January 2017, when excitement began to build about two new medications poised to become the first FDA-approved treatments for TD, my patient’s father tragically passed away from sudden cardiac death. Throwing myself into TD research was one way to overcome the sense of powerlessness I felt from my inability to help him and his mother with their grief.

My situation was analogous to the decades of powerlessness clinicians had felt in trying to treat TD with various off-label medications and supplements, none of which had ever merited more than a Level B recommendation by the American Academy of Neurology. The two investigational medications represented a great opportunity for the field, and I wanted to bring that hope to my patient and his family in their difficult time. I offered that if the first of the two VMAT2 inhibitors was approved, we would try it. I was surprised how deeply invested they were in this offer. To them it was more than simply a treatment for just another condition he had. It had become the fulfillment of a promise I wasn’t able to make to his father at that first appointment— that I could improve his TD without risking his hard-won psychiatric stability.

It had become that very beacon of hope I’d tried to ignite for him. When he was finally able to take the medication, it reduced his movements to what would be almost unnoticeable to the average person. Just a month after starting treatment, the young man who couldn’t walk a straight line down the hallway when I met him, became able to remain nearly as still as I did in session.

Over the next few months, I worked to make myself into a “DIY expert” on TD by reading Neurology journals and movement disorder textbooks. Armed with this knowledge, I found that I did have a number of other patients who had mild to moderate TD, to which I had simply been blind. I had inadvertently set my bar for diagnosing TD at the level of the severe, state hospital-level cases. The dyskinetic movements I began to notice might have been easy to miss in a short medication management appointment, but they were often the cause of significant functional impairments that I had totally overlooked before I made the effort to ask. For instance, a 67-year-old woman stopped singing in her church choir because of embarrassment over her facial movements, and a 59-year-old had to swallow her pride and ask coworkers to put on her earrings and necklaces because her hands had too many dyskinetic movements to make the precise movements.

Everyone in clinical practice today trained in an era in which there was effectively nothing we could do to address Tardive Dyskinesia, so we didn’t have much incentive to look for it. Our skills for diagnosing it withered, and many newer clinicians never learned them in the first place. I actually did have the opportunity to train under a renowned TD expert, so I really had no excuse, but I let myself become complicit in “la belle indifference” that seems to have overtaken the field since the advent of the SGAs. That era is over. Today, safe and effective VMAT2 inhibitors are available, and we need to collectively wake up and find every patient with TD in our practices. Now I’ve made assessing for TD a part of my standard mental status examination for every patient on any antipsychotic, during every visit.