Seroquel is the antipsychotic that gets prescribed for everything other than what it’s actually approved for, and the off-label sleep use is the biggest example. There are real reasons to be on Seroquel, including bipolar disorder, schizophrenia, and adjunct depression treatment, but a lot of the patients I inherit on Seroquel are on 25 or 50mg of it because somebody started them on it for sleep five years ago and never came back to the question of whether that was the right call. That’s the version that bothers me, and that’s most of what this post is about.

If we’re being honest, low-dose Seroquel for sleep is one of the more consequential prescribing habits in psychiatry, and most patients on it have no idea they’re paying a metabolic price that builds slowly over years. Any prescriber who pretends low-dose Seroquel for sleep is benign is, honestly, a damn liar, or hasn’t pulled labs on enough long-term Seroquel patients to notice the pattern.

What it does

Quetiapine is an atypical antipsychotic (the newer class of antipsychotics, the ones that came out starting in the 1990s and replaced most of the older haloperidol-class drugs). It hits a lot of receptors: dopamine D2 (the antipsychotic effect), serotonin 5-HT2A, histamine H1 (the sedating effect), alpha-1 adrenergic (more sedation, plus some drop-in-blood-pressure-when-standing effects), and a few others. The receptor profile is broad enough that the drug does different things at different doses, which is part of what makes it useful and part of what makes it tricky.

At low doses (25 to 100mg), the histamine blockade dominates and the drug is essentially a heavy sedative. At moderate doses (200 to 400mg), the antidepressant and mood-stabilizing effects come in. At higher doses (600 to 800mg), the antipsychotic effect is the main game. Same molecule, at different doses, is functionally three different drugs, which is the part that doesn’t get explained well to patients.

FDA approvals and off-label sleep

Seroquel is FDA-approved for schizophrenia, bipolar I (acute mania and depression), bipolar maintenance, and as an adjunct for major depressive disorder when an antidepressant isn’t enough on its own. Those are the on-label uses, and for those indications it’s a solid drug with real data behind it.

The biggest off-label use, by a wide margin, is sleep. Low-dose Seroquel (25 to 100mg) at bedtime is one of the most-prescribed insomnia treatments in psychiatry, despite the fact that it has never been FDA-approved for sleep, the data for sleep specifically is thin, and the medication carries a real metabolic side effect profile.

The reason it gets prescribed for sleep is that it works. It’s heavily sedating, it knocks people out within an hour, and the next morning they wake up having slept. That’s a hard combination to argue with for a patient who’s been white-knuckling insomnia for months and is desperate to sleep again. The problem is that there are better options for most patients (trazodone, mirtazapine at low doses, hydroxyzine, melatonin, and the actual first-line treatment which is CBT-I, cognitive behavioral therapy for insomnia, the structured sleep-rebuilding protocol that almost nobody gets referred to even though it has the strongest data of any sleep treatment). Seroquel for sleep tends to be a path of least resistance rather than a considered choice.

The metabolic price tag

This is the part patients on long-term Seroquel often don’t understand the size of. Seroquel is one of the worse offenders in the atypical antipsychotic class for weight gain, diabetes risk, and the cholesterol-and-triglyceride mess that gets called dyslipidemia. The average weight gain on Seroquel in the first year is something like 7 to 12 pounds, with a meaningful subset of patients gaining substantially more. Fasting glucose creeps up. Triglycerides creep up. HDL (the protective cholesterol) drops. The picture builds slowly, year over year, and most patients don’t connect their lab trend to the medication they’ve been on since the divorce.

This metabolic syndrome picture is the reason any patient on long-term Seroquel needs annual fasting glucose, A1c (the three-month average blood sugar number), and lipid panels, plus weight checks at every visit. Most prescribers know this and most do it. Some don’t. The patient who comes in on low-dose Seroquel for sleep that they’ve been on for five years and never had a metabolic panel is a patient with a problem they don’t know about, and the responsibility for the missed lab work is on the prescriber who didn’t order it, not on the patient who didn’t ask. Not to be Chicken Little about it, but a Seroquel patient five years in without labs is a routine kind of bad medicine that nobody flags because nothing dramatic has happened yet.

At higher doses, the metabolic picture is worse. At very low doses, the picture is real but smaller, which is what lets the for-sleep prescribing slide longer than it probably should.

When it’s the right call

Bipolar I with acute mania, where Seroquel works fast and reliably. Bipolar depression, where the FDA approval and the data support it. Schizophrenia, where it’s a reasonable atypical option, though not always the first one anymore. Treatment-resistant depression where augmentation with an atypical antipsychotic is on the table and Seroquel has the most data for that indication.

For sleep specifically, it’s rarely the right call. Trazodone 50 to 100mg has less metabolic baggage and similar effect on falling asleep. Mirtazapine at low doses (7.5 to 15mg) works for sleep with the added benefit of helping mood and appetite for patients who’ve stopped eating, which sounds like a problem until you meet the patient who hasn’t slept or eaten in three weeks. CBT-I is the actual first-line treatment per every guideline and almost nobody gets referred to it because there aren’t enough trained providers and insurance doesn’t always pay for it. Seroquel for sleep is usually somebody finding the path of least resistance, and the patient bearing the metabolic cost a few years later without anyone having told them that’s the deal.

The long-term Seroquel-for-sleep patient

Say you’ve got the kind of guy who’s been on Seroquel 100mg at bedtime for six years for sleep, originally prescribed after a divorce and never re-examined, one of the more common patients to inherit when somebody changes prescribers. Comes in for a routine visit, labs get pulled because nobody has pulled labs in three years. A1c is 6.4 (which is the prediabetes range). Fasting glucose 118. Weight up 25 to 30 pounds from when he started. Lipids worse than baseline.

Some of that is age and life. Most of it is the medication. The conversation about whether to taper off the Seroquel is its own appointment, and the plan that works is to taper over eight weeks while switching to trazodone for the sleep piece, then sending the patient to primary care for a diabetes prevention conversation and a possible metformin trial. Six months later the weight is down 12 to 14 pounds, the A1c is down toward normal, the sleep is about the same as it had been on the Seroquel. The Seroquel had worked for what he was using it for. The opportunity cost showed up in the blood work, slowly, and it took an actual lab review to surface what had been happening.

What’s nice to hear

For the patient where Seroquel is actually the right drug, the bipolar I patient on 400mg with good monitoring and steady mood, the schizophrenia patient where the drug is keeping psychosis at bay, the treatment-resistant depression patient where the augmentation finally moved the needle, the drug is doing important work and the metabolic monitoring is the deal. The trade is often worth it because the alternative is uncontrolled illness, and uncontrolled bipolar or schizophrenia does more damage than the metabolic side effects of a well-monitored antipsychotic.

The drug itself isn’t the problem. The unexamined long-term off-label use is. For the right indication, with the right monitoring, Seroquel is a real tool and worth being on. The version where the patient knows what they’re trading and the labs are getting pulled is the version where the drug works the way it’s supposed to.

The patient autonomy piece

If you’re on Seroquel for sleep and you want to stay on it because nothing else has worked and you’ve heard the honest take about the metabolic risk, the answer is yes, with the lab schedule attached. Provider, not parent. Appointment isn’t mine. Disapproving yes for the cases where I’d have picked trazodone or mirtazapine first, but yes regardless. I hardly ever say no. The honest take is the whole point, the choice is yours, and the part that’s not optional is the annual fasting glucose, the A1c, and the lipid panel, which is the deal for being on the drug long-term.

If you’re on Seroquel for bipolar or schizophrenia or augmentation and it’s working, the answer is also yes. Same lab schedule. The trade is different from the for-sleep trade because the alternative isn’t a different sleep aid, the alternative is uncontrolled illness, and the math comes out differently.

The same molecule, at different doses, is functionally three different drugs. Patients deserve to know which version they’re on and why.

Bottom line

Seroquel is a useful drug for the indications it’s actually approved for. For sleep, it’s a path of least resistance with a real metabolic price tag that builds slowly over years. If you’re on low-dose Seroquel for sleep and you’ve been on it for a while and nobody’s pulled your labs, that’s the conversation to have with your prescriber at the next visit. If you’re on it for bipolar or schizophrenia or treatment-resistant depression and it’s working, the metabolic monitoring is still the deal, but the trade is often worth it. The drug isn’t the problem. The unexamined long-term off-label use without the monitoring is.

The two-minute version, if your eyes glazed over halfway through, is this. Seroquel for sleep without labs is the thing to fix. Seroquel for a real psychiatric indication with regular labs and a prescriber paying attention is the thing to keep, if it’s working. The middle case, somebody on it for sleep who genuinely can’t sleep without it and has tried alternatives, is the one where the conversation gets honest about whether the metabolic risk is worth the sleep, and that’s the patient’s call to make, not the prescriber’s.

Sources

U.S. Food and Drug Administration. Seroquel (quetiapine fumarate) Prescribing Information. NDA 020639. FDA; 2022. FDA label.
Suttajit S, Srisurapanont M, Maneeton N, Maneeton B. Quetiapine for acute bipolar depression: a systematic review and meta-analysis. Drug Des Devel Ther. 2014;8:827–838. PMID 25028535.
Terao I, Yokoi A, Fukushima H. Comparative efficacy of quetiapine by dose and formulation for psychosis in schizophrenia: A systematic review and dose-response model-based network meta-analysis. J Psychopharmacol. 2023;37(10):953–959. PMID 37740667.