Prescribing Viloxazine

Why Viloxazine Matters

Viloxazine is atomoxetine's younger sibling: a norepinephrine reuptake inhibitor, unscheduled, taken once a day. The FDA approved the extended-release capsule (Qelbree) for ADHD in children and adolescents 6 to 17 in April 2021 and extended it to adults in 2022. The molecule is old: it has been used as an antidepressant outside the US since the 1970s, but it has no US depression approval.

Compared with atomoxetine, CYP2D6 genotype barely matters (poor metabolizers had only 26% higher viloxazine AUC), the capsule can be sprinkled on applesauce or pudding, and the label carries neither the sudden-death language nor the liver-injury warning that atomoxetine's does. In exchange, viloxazine is a strong CYP1A2 inhibitor with real contraindications (duloxetine, ramelteon, tizanidine and theophylline among them), it raises heart rate and diastolic blood pressure, and its suicidality boxed warning covers adults as well as children. It is brand-only, and 30 capsules of the 200 mg strength list at roughly $400 to $500 cash on GoodRx. A 30-month litigation stay bars FDA approval of generics until October 2, 2028, unless the court rules sooner (Supernus 10-Q, 2026).

Its effect size, estimated mostly from pediatric trials, is about 0.5 to 0.6, against 0.8 to 0.9 for stimulants and 0.4 to 0.6 for atomoxetine. The pooled pediatric NNT is 6 to 7, and the single adult trial was positive but modest.

Part 1: Indications

FDA-Approved Uses

• ADHD in adults and in pediatric patients 6 years and older, as monotherapy

Not approved: children under 6, depression in the US, or add-on use with stimulants (guanfacine ER and clonidine ER carry that indication in children; viloxazine does not).

Off-Label / Emerging

• ADHD with depressive or anxiety symptoms (one open-label trial, below)
• Added to a stimulant after a partial response (open-label safety data only)

The Evidence by Indication

Children 6 to 11. Two trials in the label. In Study 1 (6 weeks, n=477), 100 mg beat placebo by 5.8 points on the ADHD-RS-5 (95% CI 8.9 to 2.6) and 200 mg by 6.9 points (CI 10.0 to 3.8). In Study 2 (8 weeks, n=313), 200 mg beat placebo by 6.0 points and 400 mg by 5.8. Baseline scores were around 44 and placebo improved by 11 to 12 points.

Adolescents 12 to 17. Study 3 (6 weeks, n=310) was positive at 200 mg (4.5 points, CI 8.4 to 0.6) and 400 mg (5.1 points, CI 8.9 to 1.3). A fourth adolescent trial, which is not in the label, tested 400 mg and 600 mg: 400 mg separated from placebo and 600 mg did not, per the trial record.

Pooled pediatric data. An industry-sponsored meta-analysis of the four pediatric trials (Nasser et al., Int J Clin Pract 2021; n=1,354) found an NNT of 6 for a 30% improvement on the ADHD-RS and 7 for a 50% improvement, with one discontinuation for adverse effects per 46 patients treated. That puts its efficacy near atomoxetine's (NNT 5 to 7) and below that of stimulants (NNT 2 to 4). The trials excluded children with other significant psychiatric or neurological disorders.

Adults. One flexible-dose 6-week trial (Study 4; Nasser et al., CNS Drugs 2022; n=374, ages 18 to 65). The mean final dose was 504 mg. Viloxazine beat placebo by 3.7 points on the AISRS (CI 6.2 to 1.2), from a baseline near 38, and separated from placebo at week 2. A 30% symptom reduction was more common on drug (60.0% vs 47.6%), but the difference in 50% responders (39.2% vs 32.9%) was not significant. Both the inattention and hyperactivity-impulsivity subscales improved. Discontinuation for adverse events was 9.0% vs 4.9%. Adult efficacy rests on this one modest trial.

ADHD with depression or anxiety (off-label). An open-label, decentralized phase 4 trial (Adler et al., J Clin Psychiatry 2026; n=161 adults, 14 weeks, 200 to 600 mg) reported large improvements in AISRS, MADRS and HAM-A scores. With no placebo arm, treat this as a signal only.

How Fast It Works

One pediatric study separated from placebo by week 1, where atomoxetine generally separates around week 3, but that is one study of four and there are no head-to-head data. In adults, separation came at week 2. In practice, give it four to six weeks at an adequate dose before calling it a failure, the same as atomoxetine.

Who Is a Good Candidate?

• A child who needs a nonstimulant and cannot swallow capsules (it can be sprinkled)
• A patient with a substance use history or diversion risk (Part 9)
• A patient on fluoxetine, paroxetine or bupropion, or a known CYP2D6 poor metabolizer, for whom atomoxetine dosing becomes awkward
• A patient who did not tolerate or did not respond to atomoxetine and still wants a nonstimulant

Who Is a Poor Candidate?

• A patient taking an MAOI, or who stopped one in the last 14 days (contraindicated)
• A patient who must stay on duloxetine, tizanidine, ramelteon, tasimelteon, alosetron or theophylline (contraindicated), or on clozapine (not recommended)
• Bipolar disorder that is not well controlled; noradrenergic drugs can induce mania
• Uncontrolled hypertension or tachycardia, since it raises heart rate and diastolic pressure
• A patient who needs the largest and fastest ADHD response (a stimulant)
• A patient for whom brand pricing is prohibitive and generic atomoxetine would do

Part 2: Mechanism

Viloxazine's ADHD effect is attributed to norepinephrine reuptake inhibition, though the label calls the mechanism unclear. Two binding facts are measured:

• Norepinephrine transporter inhibitor: Ki 0.13 μM, the same principle as atomoxetine. The label groups it with other noradrenergic drugs when it warns about mania.
• 5-HT2C partial agonist: Ki 0.66 μM. The label makes no clinical claim for it, and nothing in the trials shows what it adds.

Pharmacokinetics. Exposure rises in proportion to dose from 100 to 600 mg. The capsule peaks at about 5 hours (range 3 to 9). The mean half-life is about 7 hours, with wide spread (standard deviation 4.7 hours), and steady state arrives in 2 days without accumulation. Metabolism runs mainly through CYP2D6, UGT1A9 and UGT2B15, and 90% of a dose is recovered in urine within 24 hours, so renal function matters more than liver function. Children get more drug per milligram: at 200 to 400 mg, steady-state exposure was 130% to 250% higher at ages 6 to 11 and 60% to 140% higher at 12 to 17 than in adults.

A high-fat meal lowered exposure by under 10% and delayed the peak by about 2 hours, so food is optional. Alcohol did not raise exposure (40% alcohol lowered it).

Viloxazine is a strong CYP1A2 inhibitor and a weak CYP2D6 and CYP3A4 inhibitor (Part 8). Other drugs change its levels little: paroxetine, a strong CYP2D6 inhibitor, raised its AUC by less than 35% (Wang et al., Clin Drug Investig 2024).

Part 3: Before You Start: Workup and Candidacy

Before the first dose, the label asks for two things: heart rate and blood pressure, and a screen for personal or family history of suicide, bipolar disorder and depression.

Ask about baseline sleep and daytime sleepiness, since both somnolence and insomnia are common, and about tics, anxiety and substance use, which steer the choice among nonstimulants.

Part 4: How to Start and Dose

Children 6 to 11

Raise by 100 mg weekly as response and tolerability allow. Both 100 mg and 200 mg worked in Study 1, and 400 mg did not separate further from placebo than 200 mg in Study 2 (5.8 vs 6.0 points). That is my reading of the label table (there was no formal comparison), but a child doing well at 200 mg has little reason to be pushed higher.

Adolescents 12 to 17

Study 3 showed similar benefit at 200 mg and 400 mg, and 600 mg failed in the separate adolescent trial.

Adults

The adult trial titrated to 400 mg in week 2 and allowed 600 mg after that; the mean final dose was 504 mg. Many adults will end up at 400 mg or more, which means paying for two or three capsules a day.

Sprinkling

The whole mixture must be eaten without chewing, within 15 minutes for pudding or 2 hours for applesauce, and never stored. Sprinkling on applesauce changed exposure by 10% or less.

Timing

Pick the time of day by the side effect, since the label allows any: evening if the patient is sleepy during the day, morning if sleep gets worse (common practice, not label text). The only timing data I found is an open-label phase 4 trial in 56 children and adolescents already on a stimulant, who took viloxazine in the morning for weeks 1 to 4 and in the evening for weeks 5 to 8; both schedules appeared safe and effective (Childress et al., Psych Congress 2023).

Switching From Atomoxetine

No cross-taper appears in the label. A common practice approach is to stop atomoxetine and start viloxazine at its usual starting dose, or overlap for a week if symptom return is a worry, watching heart rate and blood pressure, which both drugs raise.

Adding to a Stimulant

Viloxazine is not labeled as a stimulant adjunct. In pharmacokinetic studies, neither methylphenidate nor lisdexamfetamine changed viloxazine levels in a clinically relevant way, and viloxazine did not change methylphenidate levels (Faison et al., Clin Drug Investig 2021). The open-label phase 4 trial above, in patients with an inadequate stimulant response, reported mean ADHD-RS-5 improvements of 13.5 points at week 4 and 18.2 at week 8, with 3.6% stopping for adverse events. That trial allowed doses up to 600 mg, above the 400 mg pediatric maximum. There is no placebo-controlled trial of the combination in my sources. In adults, adding a nonstimulant to a standard stimulant dose is one alternative to pushing the stimulant past its licensed maximum. Check heart rate and blood pressure after each change.

Part 5: Monitoring

There is no blood level and no routine lab. Monitoring is vital signs, mood, sleep and weight, plus a medication check whenever another prescriber adds something.

The label asks family members and caregivers to watch for suicidal thinking and its possible precursors (anxiety, agitation, panic, irritability, insomnia, impulsivity, aggression, mania) and to report them right away. Tell parents this at the first visit.

Part 6: Side Effects and How to Manage Them

Children mostly get sleepy and lose appetite; adults mostly get insomnia, headache, nausea and dry mouth. About 3% of children and 9% of adults stopped for an adverse reaction in the trials. Rates below are drug vs placebo from the label tables.

Somnolence and Fatigue

In children, somnolence (including lethargy and sedation) ran 16% vs 4%, and rose with dose: 12% at 100 mg, 16% at 200 mg and 19% at 400 mg. Fatigue was 6% vs 2%. In adults, somnolence was 6% vs 2% and fatigue 12% vs 3%. Somnolence was the most common reason children stopped the drug.

• Management: evening dosing (common practice), a slower titration, or staying at a lower effective dose. Warn teens who drive.

Insomnia

In adults, insomnia was 23% vs 7%, the most common adult adverse reaction, and a leading reason adults stopped. In children it was 4% vs 1%. The label lists insomnia among symptoms that may precede suicidal thinking, so ask about mood when sleep changes.

• Management: move the dose to the morning (common practice). If you add a sleep aid, note that ramelteon and tasimelteon are contraindicated, and melatonin is a sensitive CYP1A2 substrate to avoid (Part 8).

Appetite and Weight

Decreased appetite was 7% vs 0.4% in children and 10% vs 3% in adults. Over 6 to 8 weeks, children 6 to 11 gained 0.2 kg on viloxazine vs 1 kg on placebo, and adolescents lost 0.2 kg vs a 1.5 kg gain on placebo. Among 338 children evaluated at 12 months in the uncontrolled extension, the mean change in weight-for-age z-score was -0.2.

• Management: weigh at visits, plot growth in children, and give the dose after breakfast or with a meal if appetite at lunch suffers.

GI Effects and Headache

Adults had nausea 12% vs 3%, dry mouth 10% vs 2%, constipation 6% vs 1% and vomiting 4% vs 1%. In children, nausea was 5% vs 3%, vomiting 4% vs 2% and abdominal pain 5% vs 4%. With either NRI, taking the dose with food may ease GI effects. Headache was 11% vs 7% in children and 17% vs 7% in adults.

Heart Rate and Blood Pressure

A heart rate rise of 20 beats per minute or more at any point in the trial occurred in 22% to 31% of children 6 to 11 on viloxazine (vs 9% to 23% on placebo), in 22% to 34% of adolescents (vs 14% to 17%), and in 29% of adults (vs 13%). A diastolic rise of 15 mmHg or more occurred in 25% of adolescents on 400 mg (vs 13%) and 13% of adults (vs 9%). These are peak-at-any-visit counts and say little about average change.

• Management: recheck after each increase. A sustained rise usually means a lower dose or a different drug (practice).

Irritability and Other Effects

Irritability was 3% vs 1% in children (5% at 400 mg) and 4% vs 3% in adults. The label lists it among possible precursors to suicidal thinking, and the Medication Guide lists it among signs of mania. Pyrexia (2% vs 0.2%) in children; dizziness (4% vs 2%) in adults. The adult adverse reaction table lists no sexual or urinary side effects at the 2% threshold.

Part 7: Toxicity, Overdose, and Boxed Warnings

Contraindications

• MAOIs: concurrent use, or use within 14 days after stopping an MAOI, because of the risk of hypertensive crisis.
• Sensitive CYP1A2 substrates, or CYP1A2 substrates with a narrow therapeutic range: see Part 8 for the named drugs.

Other Warnings

• Blood pressure and heart rate increases: check before starting, after each increase and periodically (Part 6).
• Activation of mania or hypomania: noradrenergic drugs may induce a manic or mixed episode in patients with bipolar disorder. Screen before starting. See Part 9 for bipolar patients.
• Somnolence and fatigue: no driving or hazardous machinery until the patient knows how the drug affects them.
• QT and conduction: no clinically relevant QT prolongation at 3 times the maximum dose, and no effect on PR or QRS in healthy volunteers. Nonclinical studies suggest the drug may inhibit cardiac sodium channels; the label draws no clinical instruction from that.
• Seizures (animal data): dose-dependent convulsions occurred in rats, mice and dogs at doses roughly equal to or slightly above the adult maximum. The label has no human seizure warning; I would still be careful in epilepsy (my inference; the label gives no instruction).

Overdose

Part 8: Drug Interactions

Viloxazine's interactions mostly come from what it does to other drugs. It is a strong CYP1A2 inhibitor: in a healthy-volunteer study at 900 mg a day, caffeine AUC rose about 6-fold (label Figure 3; 4.4-fold over the sampling window in Wang et al.) with no change in its peak. It is a weak CYP2D6 inhibitor (dextromethorphan AUC up about 1.9-fold) and a weak CYP3A4 inhibitor (midazolam AUC up about 1.7-fold) (Wang et al., Clin Drug Investig 2024). The label notes that CYP1A2 inhibition raises total exposure of sensitive substrates without raising their peak levels.

Part 9: Special Populations

Pregnancy

Human data are case series, too limited to estimate the risk of birth defects, miscarriage or adverse maternal outcomes. In rats at about the adult maximum dose, viloxazine increased resorptions, delayed fetal development and possibly caused low rates of malformations (including neural tube and brain anomalies), and in rats and mice exposed through pregnancy and lactation it caused maternal deaths and fetal toxicity at doses at or below the adult maximum. The label says to stop viloxazine when pregnancy is recognized unless the benefit outweighs the potential risk to the mother.

• Framework: pausing ADHD medication or switching to a drug with more pregnancy data are common choices (practice, not label). Discuss it before a planned pregnancy.
• Register exposed pregnancies with the National Pregnancy Registry for Psychiatric Medications (1-866-961-2388).

Lactation

A January 2025 label update added a lactation study in 15 women taking 600 mg daily for three days. The relative infant dose was about 1% for viloxazine and 0.07% for its main metabolite, an estimated 0.085 mg/kg/day for the infant, and the label calls transfer into milk low. There are no data on effects in breastfed infants or on milk production. The label asks you to weigh the benefits of breastfeeding against the mother's need for the drug.

Fertility

Viloxazine did not affect male or female fertility in rats at doses about equal to the adult maximum.

Elderly

Too few patients 65 and older were studied to judge their response, and no geriatric pharmacokinetic study was done. Because the kidneys clear the drug, check renal function, and watch heart rate and blood pressure.

Renal Impairment

• eGFR 30 to 89: no adjustment.
• eGFR under 30: start 100 mg, titrate weekly by 50 to 100 mg, maximum 200 mg. Label Figure 1 shows AUC about 1.9 times higher in severe impairment than in healthy subjects.

Hepatic Impairment

No labeled dose adjustment. Label Figure 1 shows AUC about 20% higher in mild impairment, little change in moderate impairment and about 25% higher in severe impairment.

Pediatric / Adolescent

Approved from age 6, based on three positive placebo-controlled trials in the label. Not established under 6. Children get substantially higher exposure per milligram than adults (Part 2). The label asks you to monitor suicidality and weight.

Bipolar Disorder

Screen before starting, as the label requires. Atomoxetine and viloxazine are controversial in bipolar disorder because, like antidepressants, they can induce mania. Stimulants are often preferred here, since whatever mania risk they carry comes with a larger ADHD benefit. If you use viloxazine anyway, get the mood stabilizer in place first and watch for mania (common practice).

Substance Use

Viloxazine is not scheduled, and the label has no abuse and dependence section, which makes it a reasonable choice for a patient with a substance use history.

Part 10: Discontinuation and Taper

No withdrawal syndrome or taper schedule appears in the label. With a mean half-life of about 7 hours, the drug clears within a couple of days.

• Planned stop: a step-down over a week or so is a common practice choice, not label text, mainly so you can see whether ADHD symptoms return.
• Stopping for a side effect: somnolence, insomnia, nausea and the heart-rate rise should fade as the drug clears over a few days (inferred from the half-life).
• Before an MAOI: the label sets a 14-day gap after stopping an MAOI before viloxazine. It gives no interval for the reverse direction.
• Before a CYP1A2 substrate: the label gives no washout before starting duloxetine, tizanidine or a similar drug. Allowing a few days for viloxazine to clear is common practice.

Part 11: Viloxazine vs the Alternatives

It is unclear whether viloxazine has any advantage over atomoxetine, and there is no cheaper generic. Trying the other NRI after one fails is reasonable practice, but no trial shows that a patient who fails one responds to the other.

The Bedside Cheat Sheet