Prescribing two antipsychotics to a single patient is a common practice in clinical settings. One study estimated that up to 50% of psychiatric inpatients are receiving antipsychotic polypharmacy (Faries D et al, BMC Psychiatry;5:26). Increasingly, we are being scolded for this practice. A variety of guidelines discourage polypharmacy, and recent studies have documented that we are often not following such guidelines.
Clearly, psychiatrists aren’t combining drugs in order to be obstinate, to cost the system more money, or to harm patients. We’re doing it because we want to make our patients feel better, and in many cases a single drug is not doing the trick. Studies indicate that the patients who are more likely to be prescribed multiple antipsychotics are younger, single, unemployed, and male; they also present with more severe psychotic symptoms. Patients with frequent hospital admissions, involuntary admissions, and those prescribed long-acting injectables (LAI) are also more likely to eventually be on two antipsychotics at the same time (Fleischhacker WW and Uchida H, Int J Neuropsychopharmacol 2014;17(7):1083–1093).
If possible, stick with a single antipsychotic. But if the situation seems to call for polypharmacy, here are some factors to guide your decision-making. The published evidence base for any combination regimen is either slim or non-existent, so consider these to be commonsense guidelines drawn from clinical experience (when there are actual studies endorsing these practices, we cite them below).
Consider the differing receptor-binding profiles of the many antipsychotics available for use (see table below). First-generation antipsychotics, such as Haldol, bind tightly to dopamine receptors and occupy the sites longer. Some second-generation antipsychotics, especially quetiapine and clozapine, are known to have a quick “on-off” effect, meaning they dissociate from the dopamine receptor quickly (McIntosh DM et al, The Canadian Journal of Diagnosis 2011; (4):33–44). How does receptor affinity translate to clinical practice? Generally, longer and tighter binding to D2 will trigger more extrapyramidal side effects (EPS), whereas quick on-off binding translates to minimal or no risk of EPS development. Also take note of other receptor binding activity, such as histamine, alpha, and muscarinic, all of which are described in the table. Using this information, for example, you would predict that two highly histaminergic or muscarinic blocking antipsychotics, like clozapine and olanzapine, will likely lead to a very tired, cognitively impaired, and constipated patient.
Some common polypharmacy scenarios
Our clinical experience indicates that there are a few common scenarios leading to multiple antipsychotics. We’ll describe each in turn, and comment on which ones are more or less reasonable.
1) Attempting to gain more rapid control of psychotic symptoms
Current treatment: quetiapine or other low-potency agent. Potential addition: haloperidol 1–2 mg every 6 hours PRN agitation, po or IM, depending on situation and setting. This scenario is often appropriate. Haloperidol can be used to control acute agitation on top of an atypical antipsychotic, and it frequently works. One could also use other higher-potency antipsychotics, such as aripiprazole, risperidone, or paliperidone. Overall, we suggest limiting such combinations to short term use (1–4 weeks). Too many antipsychotics with complicated dosing schedules will only lead to difficulty with treatment adherence.
2) Refractory cases
Current treatment: clozapine. Potential addition: risperidone 1 mg daily due to poor symptom control: low-potency oral agent.
Clozapine is the clear favorite when it comes to the treatment of refractory schizophrenia. But what do we do when clozapine isn’t enough? Adding risperidone to clozapine has become a favorite of clinicians, based on two placebo-controlled trials (Freudenreich O et al, J Clin Psychiatry 2009;70(12):1674–1680 and Josiassen RC et al, Am J Psychiatry 2005;162(1):130–136), though one larger trial showed no beneficial effect of adding risperidone (Honer et al, N Engl J Med 2006;354(5):472–482). We suggest trying this option only after a solid 12-week trial of clozapine monotherapy at therapeutic doses of 300–900 mg daily.
3) Management of acute symptom exacerbations during treatment with an LAI antipsychotic
Current treatment: paliperidone LAI IM q 4 weeks, or other LAI. Potential addition: low-potency oral agent. Your patient has a worsening of psychosis, despite taking the maximum recommended dose of an LAI. There is no evidence-based guidance on to how to approach this problem. You will probably be reluctant to discontinue the shot if your patient has responded well to it in the past. You could consider shortening the interval of the injection first (eg, move from every 4 weeks to every 3 weeks). When you do add on a second antipsychotic, try one with a differing receptor-binding profile (see table). Be mindful of worsening EPS and hyperprolactinemia with these combinations. Ensure that you reevaluate the regimen routinely, and attempt to remove the additional oral antipsychotic periodically.