Even if you’re a by-the-book psychopharmacologist, you’re likely to inherit patients on unusual combinations of meds. In a recent issue, we looked at how to handle patients on benzos and stimulants (see the October/November 2022 issue of The Carlat Psychiatry Report). We’ll continue that theme here with another unusual pairing: the antipsychotic-stimulant combo.

Combining a dopamine blocker with a dopamine agonist sounds questionable on its face, but the pairing has a long history, dating back to the 1950s when the chlorpromazine-dextroamphetamine combo pill ThoraDex was marketed for “anxious neurotics” and “alcoholics.” That combo quietly left the shelves after the FDA required proof of efficacy, but the pairing is on the rise, particularly among the young. Around one in five children who are prescribed a stimulant are also prescribed an antipsychotic (Kamble P et al, Psychiatr Serv 2015;66(4):404–410).

The concern here is not of a dangerous drug interaction. Rather, stimulants might worsen problems that the antipsychotic is treating, especially psychosis and mania, by increasing dopamine at the D2 receptor. Antipsychotics, meanwhile, might blunt the cognitive benefits of stimulants by blocking the cortical D1 receptor. 

Stimulants in psychosis

A single dose of a stimulant causes psychosis 30% of the time when given to someone in remission from a psychotic disorder and worsens psychosis 50%–70% of the time when given during an active episode, according to an analysis of 54 studies (Curran C et al, Br J Psychiatry 2004;185:196–204). This risk is higher with amphetamine than methylphenidate. In a cohort study of 221,000 adolescents and young adults with ADHD, amphetamines were associated with double the risk of psychosis compared to methylphenidate formulations (Moran LV et al, N Engl J Med 2019;380(12):1128–1138). 

Given those risks, we’d need to see significant benefits with stimulants to justify their use in psychotic disorders. Unfortunately, stimulants did not mitigate cognition problems, fatigue, or weight gain in a meta-analysis of 22 trials in psychotic disorders (Solmi M et al, CNS Spectr 2019;24(5):479–495). Likewise, the modafinils failed on those outcomes, and these novel stimulants bring an additional risk by lowering blood levels of many antipsychotics through CYP3A4 induction (Wittkampf LC et al, Ther Adv Psychopharmacol 2012;2(3):115–125).

Stimulants in mania

In bipolar disorder, stimulants raise the risk of mania, even when paired with an antipsychotic. As with psychosis, the risk of mania is worse with amphetamines, which are even used as an animal model for mania. For methylphenidate, well-designed studies are lacking, and epidemiologic studies find both benefits and harms. 

Two studies have attempted to shed light on the situation by comparing rates of mania before and after starting methylphenidate in large populations of patients with bipolar disorder. These before-and-after or “mirror image” designs are prone to false conclusions, and here the results were mixed. One study found higher rates after starting the drug (but only in those not taking mood stabilizers), while the other found no difference or lower rates (regardless of mood stabilizer status) (Jefsen OH et al, J Clin Psychopharmacol 2023;43(1):28–34).

More reassurance comes from a long-term follow-up study in 289,840 children with ADHD and their age-matched controls. Compared to those with ADHD who never took methylphenidate, those who took it long term had a 30% lower risk of developing bipolar disorder (Wang LJ et al, J Psychiatr Res 2016;72:6–14). 

Some have even speculated that methylphenidate might improve mania by stabilizing wakefulness and attention. That idea did not pan out in a randomized controlled trial of actively manic patients. The brief, three-day trial found no benefit but—reassuringly—also found no worsening of mania in the patients who took methylphenidate (Hegerl U et al, Eur Neuropsychopharmacol 2018;28(1):185–194).

Stimulants in complex patients

Stimulants are often used as a last resort in patients with complex comorbidities, many of whom are also taking an antipsychotic. Here we lack good data but see a similar pattern favoring methylphenidate over amphetamines. For example, methylphenidate improved ADHD symptoms and decision-making in a controlled trial of borderline personality disorder with ADHD, while amphetamine has a record of causing aggression and paranoia in the borderline population (Gvirts HZ et al, Int Clin Psychopharmacol 2018;33(4):233–237).

In a series of large observational studies from Denmark, methylphenidate but not amphetamine use was associated with a decrease in suicidality and self-harm in mood and personality disorders. These studies were limited by the “mirror image” design discussed above, which lacks randomization and cannot control for all variables (Rohde C et al, Aust N Z J Psychiatry 2021;55(4):422–424).

The psychiatric risks of both methylphenidate and amphetamine are magnified in the higher doses, as a study in a non-psychiatric population of 112 patients with narcolepsy demonstrates. Compared to patients on normal doses of stimulants, high doses were associated with a three-fold to 12-fold increase in psychosis, addiction, suicide attempts, and psychiatric hospitalizations (Auger RR et al, Sleep 2005;28(6):667–672).

Antipsychotics do have a role in stimulant use disorders, where they reliably reduce psychotic symptoms brought on by excessive use. They also reduce the rewarding effects of stimulants, rendering subjects in laboratory settings unable to tell if they were given a stimulant or a placebo. 

Antipsychotics in ADHD

Turning to ADHD, it is the stimulant that is therapeutic and the antipsychotic that is controversial. This combination has the most support in children with ADHD who continue to have problems with aggression after successful stimulant treatment. Risperidone (0.5–3.5 mg/day) reduced aggression in two large randomized trials of that population, although valproate brought similar improvements with fewer side effects (Blader JC et al, J Am Acad Child Adolesc Psychiatry 2021;60(2):236–251).

Those studies did not look at whether the additional antipsychotic worsened ADHD symptoms, but three controlled studies suggest it can. In both healthy adults and children with ADHD, risperidone and haloperidol blocked the beneficial effects of stimulants on objective tests of working memory and attention (Markowitz JS et al, Clin Pharmacokinet 2001;40(10):753–772). 

Stimulants exert their cognitive benefits through the D1 receptor, raising hopes that antipsychotics with low D1 blockade might be able to augment stimulants in ADHD (eg, aripiprazole, brexpiprazole, and cariprazine). This idea yielded mixed results in a few small trials, and it was recently tested in a large controlled trial of brexpiprazole as augmentation to stimulants in ADHD. The antipsychotic neither worsened nor improved ADHD symptoms in this manufacturer-supported trial (Reimherr FW et al, J Clin Psychopharmacol 2022;42(5):445–453).

Practical tips

We’ve found little reason to start an antipsychotic-stimulant combination, but what should we do when a patient presents on it? The decision to taper depends on the risk. Stimulants are riskiest in psychotic disorders, somewhat less risky in bipolar I, and safer when there is a valid diagnosis of ADHD. Antipsychotics are not particularly risky in ADHD, but they may worsen cognition and their side effect profile rarely outweighs their benefits in this disorder.

Unless the combination is posing a dire threat, it’s best to proceed gingerly and avoid the impulse to make sudden wholesale changes. Both medications are associated with withdrawal syndromes, and little is known about long-term accommodation to the pair. Check with past providers for a possible rationale and taper slowly, if necessary, over at least two weeks and often longer. Make sure you have a good alliance with your patient before disrupting any attachment they’ve developed with the medicine.

CARLAT VERDICT

As antipsychotic indications have spread, more patients are taking them in combination with their opposite: a stimulant. We look at the risks, search for scenarios where it might make sense, and come up with an action plan for this paradoxical pairing.