QT prolonging drugs list: the classes, the risk factors, and what to check before you prescribe

The interval that matters clinically is the QTc, the QT corrected for heart rate. The thresholds most widely used describe a QTc above roughly 450 milliseconds in men and 460 in women as prolonged, with a QTc above 500 milliseconds, or an increase of more than about 60 milliseconds from a patient's own baseline, treated as the point where torsades risk becomes a genuine concern rather than a theoretical one. Confirm the thresholds your institution uses, since correction formulas differ and the numbers move with them.

Two practical caveats matter more than the exact cutoff. First, an absolute QTc says less than the change from that patient's own baseline, which is why a pre-treatment ECG is worth having before starting a high risk drug. Second, the correction formula affects the answer, particularly at fast heart rates, where Bazett's formula overestimates. A patient labeled as having a prolonged QTc at a heart rate of 110 may look different under a different correction.

Most patients on a QT prolonging drug never have an arrhythmia. The events cluster in patients who carry several risk factors at once, and the list is consistent across the literature: hypokalemia, hypomagnesemia and hypocalcemia, bradycardia, female sex, older age, structural heart disease or reduced ejection fraction, hepatic or renal impairment that raises drug concentrations, congenital long QT syndrome, and the use of two or more QT prolonging drugs together.

The electrolyte piece deserves particular attention because it is both the most common and the most correctable. A patient started on ondansetron after vomiting for two days, or on a diuretic that has quietly dropped the potassium, is in a different risk category from the same patient with normal electrolytes. That is also why QT risk is not a static property of a drug list. It changes with the patient's state on the day you prescribe.

The two classes that come up most often are the macrolides and the fluoroquinolones. Within the macrolides, erythromycin and clarithromycin carry the clearest signal, and clarithromycin adds potent CYP3A4 inhibition that raises concentrations of other QT prolonging drugs the patient may already be taking, which is often the bigger problem. Azithromycin has a smaller direct effect but has been associated with cardiovascular risk in observational work.

Among the fluoroquinolones there is a usable gradient: moxifloxacin has the largest QT effect, levofloxacin sits in the middle, and ciprofloxacin the least. When a fluoroquinolone is genuinely indicated in a patient who already has QT risk, that gradient is the practical lever. Azole antifungals belong in the same conversation, largely because of the interaction effect rather than their direct one.

CredibleMeds, run by the Arizona Center for Education and Research on Therapeutics, maintains the reference lists that most institutional protocols point to. It sorts drugs into known risk of torsades de pointes, possible risk, conditional risk (meaning the risk appears under certain conditions such as overdose, hypokalemia or an interaction), and drugs to be avoided by patients with congenital long QT syndrome. Those categories are the right source of truth for any specific agent, and they are updated as evidence changes.

The category distinction is worth carrying into practice. A conditional risk drug in a patient with normal potassium and no other QT prolonging medication is a very different proposition from the same drug in a vomiting patient on ondansetron and haloperidol. This is exactly the kind of additive, context dependent judgment that a static list cannot make for you.

The failure mode with QT prolongation is almost never that the clinician did not know a drug prolongs the QT. It is that nobody looked at the whole list at once. The antiemetic was added in the emergency department, the antipsychotic on the ward, the antibiotic by the covering clinician, and each decision was individually reasonable.

Prescriber.io takes the full regimen and surfaces the additive QT prolonging combination as a flag with a cited source, in the same card as the drug-drug interaction check, the contraindication and allergy check, and the renal and hepatic dose considerations. It is decision-support for licensed US clinicians and it does not prescribe. You review the flag, weigh it against the patient's electrolytes, ECG and clinical situation, verify against official sources, and sign.