Skip to main content

TNF Inhibition and Heart Failure

Mar 02, 2016 12:00 am

Clinicians often face the issue of using a TNF inhibitor (TNFi) in patients with heart failure or cardiac disease. Here are the facts and data necessary to make a cogent decision to use or avoid TNFi.

  • The product label for marketed TNF inhibitors state, under warnings, that congestive heart failure, “worsening or new onset”, may occur. It also says that in some studies, a higher rate of serious CHF-related adverse reactions was observed with TNFi use; and to “exercise caution when using TNFi in patients who have heart failure and monitor them carefully. (These warnings are based on the performance of etanercept and infliximab in heart-failure only patients, wherein they were shown to increase CHF, cardiac events or hospitalizations, especially when used at high doses.)    
  • The recent 2015 ACR Guidelines for the Treatment of RA recommends that established rheumatoid arthritis with moderate or high disease activity and NYHA class III or IV CHF should receive combination DMARD therapy, a non-TNF biologic, or tofacitinib rather than a TNF inhibitor (TNFi). (This is a “conditional Recommendation” with Moderate to Very Low Level of Evidence. The authors noted there are no reports of exacerbation of heart failure using non-TNF biologics. A TNFi should only be used if there are no other reasonable options, and then, perhaps, only in compensated heart failure)

Chronic systemic inflammation yields a proportional risk for myocardial infarction, coronary artery disease, heart failure and cardiovascular (CV) death.  The role of vascular inflammation is at the forefront of cardiovascular research as there are currently large trials in cardiology assessing whether chronic methotrexate (CIRT trial) or IL-1 inhibitor (CANTOS trial) use may avert further CV events in at risk individuals. 

Why You Shouldn’t Use a TNF Inhibitor in RA
TNF mediates anorexia, cardiac cachexia and endotoxic shock, and also induces left ventricular hypertrophy (LVH), LV dysfunction/remodeling/dilation and has negative inotropic effects on the myocardium.  Many studies have shown increased circulating levels of TNF in heart failure, especially in those with New York Heart Association (NYHA) class III and IV disease, which correlates with the morbidity and mortality of CHF. Finally, transgenic mice that over express TNFα  tend to die prematurely from dilated cardiomyopathy.   

Three trials have examined the use of etanercept and infliximab in patients with NYHA class III and IV CHF (without any history of RA). Paradoxically these trials failed to show any benefit, but instead both showed worse CV outcomes, especially for those treated with higher doses. The unexpected harm of TNFi therapy in CHF patients may relate to the compensatory effects of TNFα on the failing myocardium and the negative effect of TNF inhibition and LV mass diminution in high-risk CHF patients. 

Can TNFi therapy be safely used in patients with heart failure? Cardiology clinical trials strongly suggest TNF inhibition would be unsafe in CHF patients, especially those with uncompensated or poorly controlled CHF or those with NYHA Class III or IV disease.   But the same trials have not been done in RA; and observational studies with large numbers of RA patients reveal far more optimistic results in those with well compensated and treated CHF.

Why You May Be Able to Use a TNF Inhibitor in RA
When TNF inhibitors have been given to RA patients, several studies have shown no higher rates of new onset CHF or sudden cardiac events and other studies have failed to link higher CV event rates with receiving TNFi (see Table). 

In the drug development trials for TNFi, the risk of new-onset CHF was 0.06%, 0.2% and 0.26% for etanercept, infliximab and adalimumab respectively (7). In 2003 Kwon reported 47 cases of new onset or worsening CHF reported to the FDA; 38 with new-onset CHF and 9 with CHF worsening (8). Half had no identifiable risk factors and 10 patients were younger than 50 years of age when they developed new-onset heart failure. Despite these sporadic reports, CHF while initiating TNFi therapy appears to be a rare event.  

Nevertheless there is a growing body of literature supporting to the cardiac safety or benefits of TNF targeted therapies in patients with RA. Recognizing that patients with CHF and cardiac disease are usually excluded from clinical trials, the early trials of etanercept, infliximab and adalimumab showed more de novo CHF events in those on placebo than the TNFi comparator (6). Numerous studies and registries of RA patients have demonstrated a cardioprotective effect of TNF inhibition on myocardial infarction and new cardiac events. Wolfe and Michaud studied 13171 RA patients in the National Databank, noting more cases of CHF in RA (compared with osteoarthritis); however this was not higher in patients receiving TNFi (11).   A retrospective cohort study of 303 Veterans Administration RA patients, failed to show a higher rate of CHF admissions or mortality amongst those treated with TNF blockers (12).  

  The Frequency and Risk of CHF in RA patients receiving TNF inhibitors#




# Patients

Freq. of CHF in RA

(CHF Freq. on Rx)

TNFi-Related risk of CHF* (95%CI)





RA 3.9%






RA 2.2% w/ CV hx

RA 0.4% without CV hx

HR 1.7 (0.7–4.1)]





RA 0.2%

RA: RR = 4.3 (NS)

CD: RR = 1.2 (NS)






HR 0.9 (0.6-1.1)

# CD: Crohns;   $ in absence of preexisting CVD;    *compared with non-TNFi treated group

Listing examined 4248 RA patients in the RABBIT registry over 3 years and found that CHF was 5 times more likely in those with CV disease compared to those without (2.2% vs. 0.4%). When TNFi was given to patients with or without prior CHF, CHF rates were not higher in those receiving TNFi. Factors most predictive of new onset CHF included age, prior CHF, DAS activity, male sex and higher BMI. Effective reduction of inflammatory activity in RA with TNFi therapy was more likely to be beneficial than harmful with regard to the risk of heart failure, especially if there is no concomitant therapy with glucocorticoids or cyclooxygenase-2 inhibitors.

Curtis studied the use of TNFi in 4018 RA and Crohn’s patients under age 50 yrs. and identified 9 new cases of CHF (0.2%). The relative risk of incident CHF among TNFi-treated RA and CD patients elevated at 4.3 and 1.2, but this was not statistically significant for those exposed to TNFi. Similar claims data analysis by Solomon compared RA patients on MTX who either started a TNFi or nonbiologic DMARD or found the risk of TNFi-related CHF was lower and not increased (HR=0.85, 95% CI 0.63 to 1.14). 

The mistake in the recent RA treatment guidelines is to avoid TNFi based on the concerns raised in cardiology patients (not RA patients) and the package insert warnings; and to recommend other therapies (DMARDs, non-TNFi biologics) that have NOT been studied RA patients with CHF, nearly as much as TNFi has been. Based on a systematic review, Heslinga et al has suggested that there is no contraindication to using TNFi in RA or inflammatory arthritis patients with mild-to-moderate CHF (NYHA class I or II), but that it should be avoided in patients with symptomatic moderate-to-severe CHF (NYHA class III-IV)   Curr Med Chem. 2015;22(16):1892-902.

The author has no conflicts of interest to disclose related to this subject

Add new comment

If you are a health practitioner, you may to comment.

Due to the nature of these comment forums, only health practitioners are allowed to comment at this time.