Paxlovid interaction checker: screen Paxlovid drug interactions and contraindications before you prescribe

US labeling splits the prohibited drugs into two groups with opposite mechanisms. The first is CYP3A substrates whose raised concentrations can cause serious harm: lovastatin and simvastatin, ranolazine, alfuzosin and silodosin, amiodarone, dronedarone, flecainide, propafenone and quinidine, triazolam and oral midazolam, lurasidone and pimozide, the ergot derivatives, eletriptan and ubrogepant, eplerenone, ivabradine and finerenone, tolvaptan, flibanserin, lomitapide, naloxegol, voclosporin, suzetrigine, and sildenafil when it is being used for pulmonary arterial hypertension. Colchicine joins them in renal or hepatic impairment.

The second group is strong CYP3A inducers, which cause the opposite problem: rifampin, rifapentine, carbamazepine, phenytoin, phenobarbital, primidone, apalutamide, enzalutamide, lumacaftor/ivacaftor and St John's wort all accelerate clearance of the antiviral itself, risking loss of virologic response and resistance. There is a detail here that is easy to miss and worth knowing: labeling states Paxlovid cannot simply be started the day after one of these is stopped, because a recently discontinued inducer keeps inducing for some time. Stopping the rifampin this morning does not make Paxlovid safe this afternoon.

It depends entirely on which statin, and this is the single most useful distinction on the page because the answer is different for three groups. Lovastatin and simvastatin are contraindicated. If Paxlovid is judged medically necessary anyway, labeling describes discontinuing them at least 12 hours before starting, through the 5 days of treatment, and for 5 days after finishing. Atorvastatin is different: temporary discontinuation during the course is described, and labeling explicitly states it does not need to be withheld before or after.

Pravastatin, rosuvastatin and pitavastatin are cleared largely outside CYP3A, which is why they generate the fewest questions during a course. That pattern is the same one that governs statin interactions with any strong CYP3A inhibitor, not just ritonavir, so it is worth recognizing as a class rule rather than a Paxlovid rule. A five day statin hold in a patient on secondary prevention carries very little risk, which is usually what makes this an easy interaction to manage rather than a reason to withhold antiviral treatment.

There is no single anticoagulant answer, which is exactly why this one gets handled badly. Apixaban is a substrate of both CYP3A and P-glycoprotein, and ritonavir inhibits both, so exposure rises. Labeling does not give a blanket instruction: it points to the apixaban prescribing information, where the recommendation depends on the apixaban dose the patient is already taking. Rivaroxaban is more restrictive, with labeling advising that concomitant use be avoided. Dabigatran depends on the indication and renal function, with either a dose reduction or avoidance.

Warfarin behaves differently again, because ritonavir can push the INR in either direction, and labeling calls for close INR monitoring if the combination is necessary. The general point that carries across all four is that this is not a class decision. It is a per-agent decision, and it is one of the more common places where a quick lookup on the wrong drug produces confident but wrong advice. The same per-agent logic applies to apixaban and the other direct oral anticoagulants with any strong dual inhibitor.

Labeling answers this directly, and the answer is not the contraindicated list. The drugs most commonly reported in serious, life-threatening or fatal adverse reactions with Paxlovid are calcineurin inhibitors, specifically tacrolimus and cyclosporine, followed by calcium channel blockers. Neither group is contraindicated. Both are common. That combination, a manageable interaction in a widely prescribed drug, is where the real risk concentrates.

The reason is structural rather than pharmacological. A contraindicated drug produces a hard stop that the prescriber cannot miss, so those interactions get caught. A transplant patient on tacrolimus, or an older patient on amlodipine, generates a softer flag that reads as informational, and the course gets prescribed with no dose change and no monitoring plan. Transplant recipients are also exactly the patients for whom antiviral treatment is most worth having, so the answer is coordination with the transplant team rather than avoidance.

Not on the last day, which is the assumption behind a fair number of restart errors. Ritonavir inhibits CYP3A irreversibly, by covalent modification of the enzyme, so activity returns only as the body makes new enzyme rather than as the drug clears. The commonly described figure is that most of the inhibition resolves over about two to three days after the final dose, and more slowly in older adults. That is the reasoning behind the labeled instruction to keep lovastatin and simvastatin held for five days after the course rather than restarting them the next morning.

The practical consequence is that the medication list needs a second look at the end of the course, not just at the start. Whatever was held has to be restarted deliberately, at the right time, by someone who knows it was held. In a patient discharged from urgent care with a five day course and three medications paused, that follow-up step is the one most likely to be dropped.

A list is useful for pattern recognition and much less useful when a patient with COVID-19 is in front of you on nine medications and the treatment window is closing. Prescriber.io takes the actual regimen, checks it against the interaction set, and returns each flag with the mechanism in plain language and the source cited, alongside contraindications, allergy blockers, and renal and hepatic dose considerations, in one card rather than four separate lookups.

It is decision-support for licensed US clinicians, not autonomous prescribing. Every flag is a prompt for you to consider. You verify against the current labeling and your own institutional references, apply clinical judgment, and sign. For the deeper clinical background on the mechanism and the specific pairs, the companion article on what to check before prescribing Paxlovid goes further into the pharmacology.