Clinical Editor's Corner: Kern

When Should You Perform Coronary Angiography in a Patient With New Onset CHF?

Morton J. Kern, MD, with contributions from Samuel M. Butman, MD, Heart & Vascular Center of Northern Arizona, Cottonwood, Arizona; Lloyd W. Klein MD, Rush University, Chicago, Illinois; Gregory J. Dehmer, MD, Texas A&M College of Medicine, Temple, Texas; Ajay Kirtane, MD, Columbia University/New York-Presbyterian Hospital, New York City, New York; Neal Kleiman, MD, Houston Methodist, Weill Cornell Medical College, Houston, Texas; Bonnie H. Weiner, MD, University of Massachusetts, Worchester, Massachusetts; Barry Borlaug, MD, Mayo Clinic, Rochester, Minnesota; Paul Teirstein, MD, Scripps Clinic, La Jolla, California; Kirk Garrett, MD, Christiana Hospital System, Wilmington, Delaware.

Morton J. Kern, MD, with contributions from Samuel M. Butman, MD, Heart & Vascular Center of Northern Arizona, Cottonwood, Arizona; Lloyd W. Klein MD, Rush University, Chicago, Illinois; Gregory J. Dehmer, MD, Texas A&M College of Medicine, Temple, Texas; Ajay Kirtane, MD, Columbia University/New York-Presbyterian Hospital, New York City, New York; Neal Kleiman, MD, Houston Methodist, Weill Cornell Medical College, Houston, Texas; Bonnie H. Weiner, MD, University of Massachusetts, Worchester, Massachusetts; Barry Borlaug, MD, Mayo Clinic, Rochester, Minnesota; Paul Teirstein, MD, Scripps Clinic, La Jolla, California; Kirk Garrett, MD, Christiana Hospital System, Wilmington, Delaware.

Questions regarding when we should proceed or wait on cardiac catheterizations for certain groups of patients are often difficult to answer, even with the help of guidelines or appropriate use criteria. At many interventional cath conferences, these questions come up frequently. For example, when should we stage the cath procedure for the multivessel ST-elevation myocardial infarction (STEMI) patient? Should every out-of-hospital cardiac arrest survivor be taken directly to the cath lab and if not, when? 
 
In this editor’s page, our colleague Dr. Sam Butman asks our group of cath lab experts, “What is your approach to a patient admitted with new-onset congestive heart failure (CHF) without recent myocardial infarction or angina? His left ventricular ejection fraction (LVEF) is <30% by echocardiography. Specifically, should we perform the catheterization within 24 or 48 hours for all new CHF patients or if not, when? Do you mandate medical therapy first, then cath all CHF patients when optimally medicated? Do you require viability studies after medical therapy and if viability is positive, cath with an eye toward revascularization?”
 
Mort Kern, Long Beach, California: Great question and while it seems straightforward, there are no easy answers. Since I’m first to answer, I’ll share our approach before my colleagues chime in. 
All new-onset CHF patients are going to get a catheterization for anatomy and perhaps a right heart for assessment of pulmonary hypertension. Stress testing in such patients is problematic, since for extensive 3-vessel coronary disease, balanced ischemia may present as false negative. The use of nuclear testing for viability is also questionable at best (although some will argue with this). The value of early cath is that once coronary artery disease (CAD) is excluded, the patient can then be evaluated for other types of cardiomyopathies. As for timing of the catheterization, we usually perform it before discharge, after a thorough echo study, and after trying to optimize medical therapy for CAD (if present), risks, and CHF. 
 
A bigger question is the benefit of revascularization for 3-vessel disease absent evidence of ischemia. That is a theoretical concern, as any symptoms in a low EF CHF patient with CAD will be labeled “anginal equivalent” and undergoing revascularization (either CABG or stenting) to eliminate any contribution of ischemia to the patient’s recovery. 
 
Lloyd W. Klein, Chicago, Illinois: Almost everyone gets a cath. The exceptions are based on candidacy for procedures, i.e., important comorbidities or other, obvious, untreated causes (e.g., alcohol or drug abuse). For those patients, I may do only a stress test. But the real issue is when to cath early. I cath sooner when there is a high degree of suspicion for ischemia demonstrated by chest pain, electrocardiogram (ECG) changes, or abnormal echo findings. I’d cath the patient later but before hospital discharge, even without obvious ischemia. The severity of CHF or the need to know about pulmonary hypertension would also play into the timing of the procedure. 
 
Neal Kleiman, Houston, Texas: I think that in most institutions, such a patient would get a coronary angiogram in the first few days of presentation. The problem arises when there is mild coronary artery disease that is not severe enough to explain the clinical picture, someone brings up the term “ischemic cardiomyopathy”, and this diagnosis is cut and pasted into the patient’s EMR for time immemorial. [No avoiding a cath after that – MK]
 
Bonnie Weiner, Worchester, Massachusetts: Virtually all of these folks get cathed at some point. The question is: when? I see many of my colleagues cath too soon when heart failure is not optimally treated, especially when there is no evidence of active ischemia precipitating it. Not only are the angiograms better, but the risk is lower if the hemodynamics are optimized first. Moreover, even if the patient winds up being a candidate for revascularization, they are in better shape to undergo additional procedures.
 
Sam Butman, Cottonwood, Arizona: That may have been my real question, Bonnie. In a patient without angina or in a peri-MI situation, revascularization is not going to improve function for days to weeks, and therefore optimization and assurance of stability should trump the early (24-48 hour) need for cath.
 
Ajay Kirtane, Columbia University, New York: We published on this using administrative claims data (Figure 1)1, and found that while we may think that cath (or even stress testing) happens frequently, less than one-third of patients with new HF actually get worked up for ischemic heart disease! We found this remarkable (almost unbelievable), but it is consistent with a prior study by Farmer et al2. It may have to do with the fact that many of these patients never get to see a cardiologist, let alone one who recognizes ischemia as a potentially addressable cause of HF. Even if one accepts the possibility that there may be underreporting in this analysis, even doubling the rate of an ischemic workup would yield rates that are still likely too low. 
 
One other point to consider is that whether or not revascularization is indicated, considered, or performed, there are additional data that suggest that simply making the diagnosis of concomitant CAD (with HF) is associated with a greater utilization of guideline-directed medical therapies for HF.
 
As noted above, cath before discharge makes good sense. I have found anecdotally though, that if these folks are deferred, sometimes they never end up getting the workup they need. If the patient is clinically stable (not needing escalation of therapies such as left ventricular (LV) support, not infracting, or presenting with acute coronary syndromes [ACS]), diuresis first can be very reasonable. This was the logic of doing the right heart cath before getting arterial access in “classical” fellowship training. Sometimes this can be taken too far, though, and doing a diagnostic coronary angiogram unless the patient is in florid shock or acute kidney injury (AKI) is rarely harmful.
 
Lloyd W. Klein, Chicago, Illinois: Really? There is a definite risk of invasive procedures in those who are volume overloaded, hypoxic, oliguric, or who are otherwise suboptimal candidates. Unless there is apparent ischemia, or the LV dysfunction is clearly related to an acute event, there is no rush to do anything. Otherwise, the conversation typically transitions to topics like “viability” and “ischemic burden”, which are sure signs of ambiguity.
 
Barry Borlaug, Rochester, Minnesota: Unless they are not candidates for revascularization, I would vigorously exclude CAD as a cause of heart failure with reduced ejection fraction (HFREF) and do this definitively and anatomically (i.e., angiography rather than stress testing). Coronary disease is the cause in 67% of all patients with systolic heart failure in the U.S., so it must be excluded, because it may be reversible and you can potentially cure them of heart failure. Somewhere between 7-10% of apparent “primary dilated cardiomyopathy” turns out to be CAD, when you look for it. No one wants to miss that. 
 
In terms of timing, it would be best to cath when the clinical hemodynamics are optimized (no more JVD or S3, etc.) if possible. Then you can determine what the patient looks like in terms of pulmonary artery (PA) pressures, etc., when he is euvolemic. A right heart cath is incredibly useful for that purpose. The patients are too often sent home still too wet, in which case they are more likely to bounce back [be readmitted – MK]. So you can save your hospital a dreaded readmission by checking volume status in the definitive way at the time of angiography. 
 
Paul Teirstein, La Jolla, California: This indication for cath used to be inappropriate. Has the appropriate use criteria (AUC) changed now so that this is at least an appropriate indication? 
 
Greg Dehmer, Temple, Texas: The original question from Sam was: what is your approach to new-onset CHF? No worries, Sam, you won’t go to AUC jail for performing angiography in your patient. From the 2012 AUC for Diagnostic Cath, there are 2 scenarios that would fit this patient: 
 
#24: The patient has new CHF (implying symptoms) and a EF<40% = Appropriate (Table 1).
 
Or #93: Suspected cardiomyopathy with heart failure = Appropriate (Table 2).
 
The AUC Task Force is moving away from single modality (in this example, Dx Cath) in favor of multimodality documents such as the 2013 Multimodality AUC for the Detection and Risk Assessment of SIHD (J Am Coll Cardiol 2014;63:380-406). Performing coronary angiography in a patient with newly diagnosed systolic heart failure was also scored as “Appropriate” in this document (Table 3).
 
A new multimodality imaging document is currently under development. 
 
Kirk Garrett, Wilmington, Delaware: On the appropriateness question Paul raises, remember we’re talking about diagnostic cath here. If a newly diagnosed heart failure patient has had symptoms for >6 weeks and no active ischemic findings, we have to consider them as having stable ischemic heart disease for AUC purposes (if different, they can be classified as unstable). The 2013 AUC paper on multi-modality assessment of stable ischemic heart disease (SIHD) finds invasive coronary angiography is APPROPRIATE for symptomatic patients with a high pre-test probability of CAD. Asymptomatic patients are RARELY APPROPRIATE for cath as initial diagnostic step (Wolk M et al, J Am Coll Cardiol. 2014;63:380-406). I think this means that truly asymptomatic patients with high functional capacity should get tested for evidence of ischemia before cath.
 
Is it appropriate to revascularize the patient if you find bad CAD? The new AUC paper on revascularization (Patel M, J Am Coll Cardiol. 2017;69:2212-2241) says that patients with “severe resting LV dysfunction (LVEF <35%) not reliably explained by noncoronary causes” count as high risk; if LVEF is 35-49%, then they’re intermediate risk. This very helpful, since PCI does not have a RARELY APPROPRIATE finding for any patient with high-risk non-invasive findings, even if they’re on no meds. Meanwhile, revascularization guidelines put CABG for LV dysfunction as either IIa or IIb (depending on degree of LV dysfunction, whether a LIMA can be used, etc.), while PCI isn’t ranked because of insufficient data (Finn S, J Am Coll Cardiol. 2012;60:e44-164). 
 
So, diagnostic cath for newly admitted patients with symptoms is appropriate, and if they have high-grade lesions suitable for percutaneous coronary intervention, you can go ahead, even if they’re not on any anti-ischemia medications. For many with 3-vessel disease (especially diabetics), CABG may be a better option.
 
Paul Teirstein, La Jolla, California: Kirk and Greg, please note this was not in the original AUC and has been added, which is appropriate. It’s one of the major reasons why the percentage of “inappropriate” fell so dramatically over the past few years, i.e., it’s not because the doctors got better at doing appropriate PCI — it’s because the definitions got tightened. 
 
The Bottom Line
 
For the patient with new-onset CHF, we should consider these approaches: 1) medically stabilize, and 2) cath before discharge. The decision to revascularize will depend on disease extent, viability (perhaps), and comorbidities. According to the AUC, it is appropriate to cath such patients. I thank my colleagues for helping us to understand how to manage this important patient group.
 
 
References
  1. Doshi D, Ben-Yehuda O, Bonafede M, et al. Underutilization of coronary artery disease testing among patients hospitalized with new-onset heart failure. J Am Coll Cardiol. 2016; 68: 450-458.
  2. Farmer SA, Lenzo J, Magid DJ, et al. Hospital-level variation in use of cardiovascular testing for adults with incident heart failure: findings from the cardiovascular research network heart failure study. JACC Cardiovasc Imaging. 2014 Jul; 7(7): 690-700.
Disclosure: Dr. Kern is a consultant for Abiomed, Merit Medical, Abbott Vascular, Philips Volcano, ACIST Medical, Opsens Inc., and Heartflow Inc.