Risperidone/Aripiprazole for Irritability in ASD
The one FDA-approved indication in autism spectrum disorder — real efficacy for aggression and irritability, weighed against real metabolic risk (weight gain, dyslipidemia) in a young, otherwise metabolically healthy child.
The bruise on the forearm of K.B.'s mother is four days old and still clearly finger-shaped, and she shows it to the team not for sympathy but as evidence, the way she might show an X-ray. K.B. is a 7-year-old boy, diagnosed with autism spectrum disorder at 3, nonverbal, and has always had a sensory profile that makes certain textures — wet grass, the tags in shirt collars, the specific hum of the refrigerator when the kitchen is otherwise quiet — genuinely intolerable in a way that used to resolve with removal of the trigger. In the past three months that hasn't been enough anymore. What began as distress escalated into hitting, first himself, then his mother when she tries to intervene, and twice in the last two weeks into a level of aggression that required both parents to physically de-escalate a single episode.
Applied behavior analysis, which K.B. has received consistently for four years, has been intensified over the past two months without measurable change in the aggression's frequency or severity. The behavioral team's own data show the escalations are not clearly tied to identifiable sensory triggers anymore, which is itself a change from his baseline pattern — the ABA consultant's notes describe several episodes beginning with no identifiable antecedent at all, a shift from the clear trigger-response chains his team had reliably mapped and interrupted for years. That unpredictability is part of what has made this feel, to both parents, like a different problem than the one they've spent four years learning to manage, not simply a worsening of the same one.
He is a healthy-weight, healthy child by every other metric — no diabetes risk factors, no family history of early cardiometabolic disease — which is precisely what makes the medication conversation feel, to his mother, like trading one worry for a different one she doesn't yet have any experience managing.
Starting the one FDA-approved medication class against a real metabolic cost
Risperidone and aripiprazole are the only two agents with FDA approval specifically for irritability associated with autistic disorder, and the trial evidence behind that approval is real: the RUPP Autism Network's randomized trial of risperidone found substantial reduction in aggression and self-injury against placebo, and Marcus and colleagues' fixed-dose aripiprazole trial found similar separation. Given two weeks of physically dangerous episodes and a four-year-established behavioral program that has stopped working, the safety argument for starting one of these now is not a close call.
I'm not disputing the efficacy data or the acute safety concern. I want the metabolic cost stated with equal specificity rather than treated as a generic caveat: both agents carry real weight-gain and dyslipidemia risk in children, and that risk compounds over years of exposure in a way a single risk-benefit conversation today doesn't fully capture. Aripiprazole's trial data show a somewhat more favorable metabolic profile than risperidone's on average, which is a real, relevant distinction if we're starting an antipsychotic in a 7-year-old with an otherwise clean cardiometabolic baseline.
That distinction is worth building into the choice of agent, but it's an argument for which antipsychotic, not whether one is warranted given what's actually happening at home right now.
Both points hold together cleanly: start aripiprazole, given its comparatively better metabolic profile in the trial data, at the lowest effective dose, continue ABA rather than treating medication as a replacement for it, and build metabolic monitoring — weight, fasting lipids, glucose — into the same follow-up schedule from day one rather than adding it reactively after a problem appears.
Agreed: low-dose aripiprazole started, baseline metabolic panel drawn today, ABA continued unchanged, four-week follow-up with repeat weight check and a three-month lipid recheck already scheduled rather than left to be arranged later.
The endocrinologist's caution about long-term cumulative metabolic exposure was documented explicitly in the plan as an ongoing consideration for future visits, not a concern this single decision resolves.