Provigil and Rivaroxaban Interaction: What Patients and Clinicians Need to Know

Modafinil (brand name Provigil), a wakefulness-promoting agent FDA-approved for narcolepsy, obstructive sleep apnea with residual sleepiness, and shift-work sleep disorder, is a documented inducer of the liver enzyme CYP3A4. Rivaroxaban (brand name Xarelto), a direct oral factor Xa inhibitor used to prevent stroke in non-valvular atrial fibrillation and to treat or prevent venous thromboembolism, is cleared in part through that same enzyme and through the P-glycoprotein (P-gp) transporter. That shared pathway is the reason the combination is flagged as a potential interaction. No dedicated clinical trial measuring the actual change in rivaroxaban blood levels when modafinil is added has been identified for this review, so the concern rests on mechanism and analogy rather than a direct measurement in patients taking both drugs.

The core answer

Modafinil's induction of CYP3A4 is established in its FDA prescribing information. Rivaroxaban's dependence on CYP3A4 and P-gp for part of its clearance, and the label instruction to avoid combining rivaroxaban with combined P-gp and CYP3A4 inducers, is established in the Xarelto label. What has not been directly studied and published is the size of the effect when modafinil specifically is the inducer paired with rivaroxaban specifically. Because of that gap, this should be treated as a plausible, mechanism-based moderate interaction that warrants clinician awareness, not as a precisely quantified risk with a validated percentage attached to it.

What is established

• Modafinil's own label states that it can induce CYP3A4/5 activity and specifically warns about the need for closer monitoring with cyclosporine, a sensitive CYP3A4 substrate, when modafinil is started (per the modafinil prescribing information).
• Rivaroxaban's label states that a meaningful fraction of its clearance depends on CYP3A4 and P-gp, and it advises against combining rivaroxaban with drugs that are combined P-gp and CYP3A4 inducers, giving rifampin, carbamazepine, phenytoin, and St. John's Wort as named examples, because such combinations can lower rivaroxaban exposure and may raise thromboembolic risk (per the rivaroxaban prescribing information).
• Rivaroxaban is not cleared by CYP3A4 alone. A substantial portion is excreted unchanged by the kidney, and other metabolic routes besides CYP3A4 contribute, which means the drug is not as fully dependent on CYP3A4 as some other substrates are.
• Modafinil is not named in the rivaroxaban label's list of inducers to avoid. That omission does not establish safety; it likely reflects that the modafinil-rivaroxaban pair has not been formally studied, since rivaroxaban was approved after modafinil.

What is pharmacologically plausible but unproven

• Because modafinil induces CYP3A4 and rivaroxaban is partly cleared through that enzyme, it is biologically plausible that adding modafinil could lower rivaroxaban blood levels to some degree.
• Some in vitro and pharmacology literature describes modafinil as activating the pregnane X receptor (PXR), the same nuclear receptor pathway that drives CYP3A4 and P-gp induction together. If that translates into clinically meaningful P-gp induction in patients, the combined effect on rivaroxaban exposure could be larger than CYP3A4 induction alone would predict, because P-gp limits how much rivaroxaban is absorbed in the gut before it ever reaches the liver.
• Induction from CYP3A4 inducers generally takes one to two weeks to reach its full effect and takes a similar amount of time to wear off after the inducing drug is stopped. This general pharmacology principle is well established for inducers as a class; the exact timeline specific to modafinil and rivaroxaban together has not been separately confirmed.
• By analogy with other CYP3A4 inducers of varying strength, a moderate inducer might be expected to produce a smaller reduction in rivaroxaban exposure than a strong inducer such as rifampin. This is a reasonable inference from general pharmacokinetic principles, not a measured value for this specific pair.

What is not established

• There is no published head-to-head pharmacokinetic study quantifying how much modafinil reduces rivaroxaban AUC or trough concentration in humans. Any specific percentage attached to this pair should be treated as an estimate requiring verification against the primary literature, not a confirmed figure.
• There is no published outcome data (stroke, systemic embolism, VTE recurrence) specifically comparing patients on rivaroxaban who did versus did not add modafinil.
• The degree of P-gp induction by modafinil in humans, as opposed to in vitro receptor activation, has not been clearly established at therapeutic modafinil doses.
• Whether occasional or as-needed modafinil dosing (for example, only on shift-work nights) produces a clinically meaningful reduction in rivaroxaban levels is unknown; it is plausible that intermittent dosing produces a smaller and less predictable effect than daily continuous dosing, but this has not been separately tested.

Evidence-status interaction assessment

Use this table to locate a specific claim about this interaction before acting on it or repeating it to a patient or colleague.

Why this matters clinically even without a direct study

Rivaroxaban's efficacy in atrial fibrillation and venous thromboembolism depends on maintaining factor Xa inhibition through the dosing interval. If an enzyme inducer meaningfully lowers trough drug levels, the period of lowest protection (just before the next dose) is affected disproportionately, because trough concentrations sit on the steep, low end of the concentration-time curve. This is a general pharmacokinetic principle for once-daily anticoagulants with induction-sensitive clearance, not a finding specific to this pair. It explains why prescribers treat any CYP3A4-inducer combination with rivaroxaban cautiously even before a specific number is available.

Monitoring if the combination cannot be avoided

Standard coagulation tests, PT and aPTT, do not reliably reflect rivaroxaban's anticoagulant activity. Anti-Xa assays calibrated specifically for rivaroxaban can provide a quantitative measure of drug exposure, but calibration, reference ranges, and availability vary by laboratory and are not universal at point-of-care labs. A patient and prescriber who decide the combination is necessary should:

• Confirm the indication and current rivaroxaban dose before modafinil is started.
• Ask the local lab or anticoagulation clinic whether a rivaroxaban-calibrated anti-Xa assay is available, and at what timepoints it is meaningful (peak versus trough).
• Set a follow-up window around two weeks after starting modafinil, when enzyme induction is expected to be near its peak, and again after any dose change.
• Review symptoms consistent with under-anticoagulation at each visit: new leg swelling or pain, sudden shortness of breath or chest pain, or new neurological symptoms such as weakness, facial droop, or speech difficulty. Any of these warrants urgent evaluation, not a routine follow-up appointment.
• Avoid adding other CYP3A4 inducers, especially St. John's Wort, on top of this combination.

Alternatives worth discussing with a prescriber

Alternatives to modafinil: Armodafinil (Nuvigil) is closely related to modafinil and should not be assumed to be free of the same CYP3A4 induction concern. Solriamfetol (Sunosi) and pitolisant (Wakix) have different mechanisms of action and are not established CYP3A4 inducers at labeled doses, though the interaction literature for these newer agents is more limited; current labeling should be checked before substituting.

Alternatives to rivaroxaban: Apixaban (Eliquis) is also partly cleared through CYP3A4 and P-gp, so it does not remove the concern, though its twice-daily dosing may behave differently than rivaroxaban's once-daily dosing under partial induction; this is a pharmacologic consideration, not a confirmed clinical advantage. Warfarin is cleared through CYP2C9 and CYP1A2, not CYP3A4, so modafinil's CYP3A4 induction would not be expected to reduce warfarin's anticoagulant effect directly. Warfarin's practical advantage is that INR monitoring detects a change in anticoagulant effect from any cause within days, which is not true of rivaroxaban unless specialized anti-Xa testing is used.

Any change in anticoagulant should be made by the prescriber managing the underlying indication (atrial fibrillation, VTE, or valve disease), not self-directed by the patient.

Special populations

Hepatic impairment: CYP3A4 is a hepatic enzyme, and rivaroxaban is already contraindicated in moderate to severe hepatic impairment (Child-Pugh B or C) associated with coagulopathy, independent of modafinil. This population should not be combined with either drug without specialist input.

Renal impairment: Because roughly a third of a rivaroxaban dose is excreted unchanged by the kidney, patients with reduced renal function already have less reserve if metabolic clearance is also altered. This makes careful monitoring more, not less, important if modafinil is added.

Older adults: Baseline CYP3A4 activity tends to be lower with age, and age-related decline in renal function adds another source of variability. There is no published data quantifying whether older adults experience a larger relative change from modafinil specifically; this is a general pharmacologic consideration to raise with a prescriber, not a confirmed finding.

When to seek urgent care

Anyone taking rivaroxaban, with or without modafinil, who develops sudden chest pain or shortness of breath, one-sided weakness, facial drooping, sudden difficulty speaking, or new leg swelling with pain should seek emergency evaluation immediately. These can be signs of pulmonary embolism, stroke, or deep vein thrombosis, and time to treatment affects outcomes. Do not wait for a scheduled follow-up appointment to report these symptoms.

Patients should not stop rivaroxaban on their own because of interaction concerns. Abrupt discontinuation carries its own stroke or clot risk in patients being treated for atrial fibrillation or VTE. Any change should go through the prescriber.

A note on this page's limits

This page cannot substitute for a pharmacist or physician review of an individual patient's full medication list, renal and hepatic function, indication for anticoagulation, and bleeding or clotting risk. It also cannot provide a validated numeric estimate of how much modafinil lowers rivaroxaban levels, because that specific study does not appear to have been published. Where a precise percentage would be reassuring, the honest answer is that it does not currently exist for this drug pair, and that gap itself is a reason for caution rather than a reason to dismiss the concern.