TCPR: A lot of meds have warnings about their use in “heart disease.” Does that term include cardiac risk factors like hypertension or high cholesterol?

Dr. Funk: These warnings specifically refer to known heart abnormalities (eg, arrhythmias or other conduction problems), structural problems like valve disease, congenital defects, damage from myocardial infarction (MI), and congestive heart failure. Now, people with cardiovascular risk factors may have undiagnosed underlying heart disease. We are also worried about people who have vascular disease, which could include coronary artery disease (CAD) in people who’ve never had a heart attack or peripheral vascular disease in people who’ve never had a stroke. So, a better term than “heart disease” is “cardiovascular disease,” since we are concerned about the entire vascular system.

TCPR: What do we worry about with psychiatric medications and cardiovascular disease (CAD)?

Dr. Funk: Let’s start with CAD, which means the patient has had angina on a stress test, ischemia on the electrocardiogram (ECG), or atherosclerosis detected on imaging studies. I avoid stimulants and tricyclics in CAD because they can raise the risk of stroke and MI, and in fact tricyclics are absolutely contraindicated in the six months following an MI. Medications that block cardiac sodium channels, including tricyclics and many anticonvulsants, may delay cardiac conduction or unmask conduction abnormalities, resulting in life-threatening arrhythmias. With clozapine, we worry about the risk of myocarditis and cardiomyopathy. Another problem is prolongation of the QT interval. Many medications prolong the QT, and when combined with other risk factors, that can result in torsades de pointes (TdP), a serious arrhythmia.

TCPR: Pharmacists often warn us when our patient is on multiple medications that prolong the QT interval. When do you put the brakes on?

Dr. Funk: One where I would put the brakes on is methadone. The risk of TdP is high with this one, so I would avoid it with other medications that prolong the QT. Methadone’s QT effects are dose-dependent, so the risk is even higher with CYP2D6 inhibitors like fluoxetine and paroxetine, which raise methadone levels (Chou R et al, J Pain 2014;15(4):321–337). Other high-risk medications include the antiarrhythmics sotalol and dofetilide. Aside from these medications, which necessitate extreme caution, it is essential to perform a risk-benefit assessment for each patient, weighing the risk of TdP versus the psychiatric risk from not giving a particular psychotropic. The QT is only one component of that assessment.

TCPR: What are other risk factors?

Dr. Funk: There are modifiable and nonmodifiable risk factors (Editor’s note: See the table “Risk Factors for Torsades de Pointes”). The nonmodifiable ones include older age, female sex, and personal history of arrhythmia or of structural or functional cardiac disease. Look for a family history of sudden cardiac death (SCD), as this can point to familial long-QT syndrome or genetic cardiomyopathies that predispose to SCD. Modifiable risk factors include other medications and the patient’s overall health status. Bradycardia is a big risk for TdP, as well as electrolyte imbalance, specifically hypokalemia, hypomagnesemia, and hypocalcemia. This includes people with eating disorders and those who require hemodialysis. People in the ICU are at higher risk, as well as those with poor kidney or liver functioning, in part because those problems can raise levels of higher-risk medications.

Table. Risk Factors for Torsades de Points
Click to view a full PDF

TCPR: Are there meds that raise the QT interval without raising the risk of TdP?

Dr. Funk: The link between prolonged QT and TdP is not a straight line, but it’s difficult to be precise here because TdP is a difficult outcome to study. Most of the data used to categorize the risk of TdP in registries like CredibleMeds are based on weak levels of evidence like case reports and retrospective cohort studies.

TCPR: Which antidepressants are safest in cardiovascular disease?

Dr. Funk: Sertraline. We have a lot of safety data to back this one up (Parissis J et al, Expert Opin Pharmacother 2007;8(10):1529–1537). I’m also comfortable with bupropion. It causes a mild tachycardia, which may actually protect against TdP. Sometimes, though, bupropion can raise blood pressure. SNRIs and mirtazapine are generally safe. Venlafaxine can raise blood pressure and has some mixed data regarding QT prolongation, but in general the risk is negligible. Trazodone prolongs the QT in overdose, but not in normal doses. The other SSRIs are safe as well, except citalopram, which has a more complicated safety profile.

TCPR: Citalopram has a specific warning about TdP, correct?

Dr. Funk: Yes. In 2011, the FDA warned that citalopram “should not be used” at doses above 40 mg/day, and the cutoff is even lower—20 mg/day—if the patient 1) is over 60, 2) is a poor metabolizer at CYP2C19, or 3) has hepatic impairment. Interestingly, this guidance was based on the FDA’s assertion that doses of 60 mg/day were no more efficacious than 40 mg/day. In 2012, the FDA downgraded the language in both statements from “should not be used” to “is not recommended.” They also recommended discontinuing citalopram in anyone with a QTc greater than 500 ms. These recommendations are controversial, though. Citalopram does prolong the QT interval more than other SSRIs, but we lack clinical evidence that it raises the risk of TdP. In fact, two very large studies compared high doses of citalopram and escitalopram to equivalent doses of other SSRIs and found that neither was associated with an increased risk of ventricular arrhythmia, sudden cardiac death, or mortality (Ray WA et al, J Clin Psychiatry 2017;78(2):190–195; Zivin K et al, Am J Psychiatry 2013;170(6):642–650).

TCPR: What is your bottom-line recommendation? Should we feel safe prescribing citalopram in patients without cardiovascular illness?

Dr. Funk: Yes, and I would also caution against reflexively changing citalopram. There are many data showing adverse psychiatric outcomes when citalopram was reflexively lowered in response to these warnings (Rector TS et al, Am J Psychiatry 2016;173(9):896–902).

TCPR: What about tricyclics and MAOIs?

Dr. Funk: With tricyclics, the concern is conduction delays. They should be avoided in patients with bundle-branch block or a widened QRS. I would also avoid MAOIs in cardiac disease. These patients tend to be on a lot of medications, and many are on special diets, so the risk of food and drug interactions goes up, and the complications of hypertensive crisis are much more serious.

TCPR: Stimulants have cardiac warnings. Is that related to TdP?

Dr. Funk: No, it isn’t. Stimulants raise blood pressure and heart rate, can cause coronary vasospasm, and are linked to very rare cases of sudden cardiac death. An ECG is not required before starting a stimulant, because the pretest probability is so low, but if you do have access to an ECG, you should check it. What you’d be looking for is ventricular conduction delay, arrhythmias, heart block, and recent MI. One problem with stimulants is that we don’t have many data in older adults. What I would recommend in older adults is to monitor heart rate and blood pressure at least every six months, and get an ECG once a year, depending on the patient’s other cardiac risk factors.

TCPR: Are there mood stabilizers you worry about in cardiovascular disease?

Dr. Funk: Yes. Many of the anticonvulsants block sodium channels like tricyclics do, so we worry about ventricular conduction delay and heart block. Carbamazepine, oxcarbazepine, topiramate, and lamotrigine are all on that list, and lamotrigine has a new black box warning about this. I would avoid these medications in people with ventricular conduction delays, meaning a QRS complex that is 110 ms or greater (normal is 80–100 ms). Depakote, gabapentin, and pregabalin are fairly safe here.

TCPR: When would you worry about lithium in cardiovascular disease?

Dr. Funk: Lithium is associated with sinus node dysfunction, atrioventricular block, Brugada syndrome, and QT prolongation at high doses. In general, if a patient is in treatment for one of these conditions and follows with a cardiologist, I would consult with the cardiologist before starting lithium. It is not necessary to order an ECG before starting lithium in every patient, but it is appropriate for patients who have known QT prolongation, are taking QT-prolonging medications, or have other cardiac risk factors.

TCPR: We often see QT prolongation with antipsychotics. Which ones have the highest and lowest risk of TdP?

Dr. Funk: In the lower risk category are aripiprazole and lurasidone. The highest risk is with thioridazine, chlorpromazine, and possibly ziprasidone. Olanzapine, quetiapine, risperidone, and haloperidol are middle of the road. One that is interesting is IV haloperidol, which has a reputation for causing more TdP than the oral form. This misconception was largely debunked in a recent paper, which found that much of the increased risk was due to the fact that patients in the ICU are not only more likely to get the IV form, but also typically have numerous risk factors for TdP by nature of being severely ill (Beach SR et al, Gen Hosp Psychiatry 2020;67:42–50). Basically, I would avoid the high-risk antipsychotics in patients with multiple additional risk factors for TdP and be careful about the middle-of-the-road ones.