Incomplete Revascularization Linked to Poorer Outcomes in TAVR Patients With CAD

Download this article's Factoid (PDF & PPT for Gold Subscribers)

Patients with reduced left ventricular ejection fraction (LVEF) who undergo transcatheter aortic valve replacement (TAVR) face diminished LVEF recovery and higher mortality if they do not receive prior complete revascularization of occluded arteries, suggests a small series published online August 28, 2014, ahead of print in Catheterization and Cardiovascular Interventions. Such patients should not be denied TAVR if percutaneous coronary intervention (PCI) is not feasible, the authors say, but complete or at least partial revascularization should be performed first whenever possible.

Anita W. Asgar, MD, and colleagues from the Montreal Heart Institute (Montreal, Canada), looked at data from 56 patients with severe aortic stenosis and LVEF ≤ 50% who underwent TAVR at their institution between March 2006 and May 2012. The type of device (Sapien or Sapien XT [Edwards Lifesciences; Irvine, CA] or CoreValve [Medtronic; Minneapolis, MN]), access route, and bioprosthesis size were chosen by a multidisciplinary team.
CAD was defined as the presence of at least 1 stenosis ≥ 70% by visual estimation or a history of revascularization. CAD severity was assessed using the Duke Myocardial Jeopardy Score (DMSJ); a score of 0 indicates no CAD or complete revascularization prior to TAVR, while a score of greater than 0 indicates nonrevascularized CAD or incomplete revascularization.

Overall, 44 patients (78.5%) had CAD, but half of these had a DMJS of 0, while scores of 2-4 and > 4 were observed in 39.3% and 10.7% of CAD patients, respectively. While there was a higher prevalence of segmental wall-motion abnormalities in patients with > 0 DMJS, there were no significant differences between the CAD groups in LVEF, aortic mean gradient, aortic valve area, and degree of aortic or mitral regurgitation. Procedural characteristics were well matched between the groups.

After January 2010, patients also underwent assessment by cardiac magnetic resonance imaging (CMR). Among the 17 with a complete viability study, there were no significant differences between the groups in LVEF, LV end-systolic or end-diastolic volumes, wall motion score, delayed enhancement score, or number of viable segments.

CAD Severity Linked to In-Hospital, 1-Year Mortality

Six patients (10.9%) died during the initial hospitalization. Mortality was higher in those with DMJS > 0 compared with 0 both in the hospital (22.2% vs 0; P = .010) and at 1 year (25.9% vs 3.5%; P = .019). The former also had a higher rate of the composite of death, MI, and readmission for heart failure (37% vs 3.7%; P = .002). However, there were no differences in NYHA functional class between the groups.

At 3 months, patients with a DMJS of 0 demonstrated better LVEF recovery (primary endpoint) than those with a DMJS > 0, with the difference appearing before discharge (P = .007) and persisting through 12 months (P = .020; table 1). The same pattern was seen for the subgroup with only a history of CAD (P = .008).

 Table 1. LVEF Recovery at 3 Months by DMJSa



3 Months

DMJS = 0

39.5 ± 9.3%

51.4 ± 8.6%

DMJS > 0

41.2 ± 7.4%

44.7 ± 10.2%

a P < .001 for the difference in LVEF improvement between DMJS groups.

The severity of CAD correlated with the degree of LVEF improvement—patients with a DMJS of 0 had the highest degree of recovery (P < .001), and patients with a DMJS between 2 and 4 fared better than those with a score of > 4 who had reduced LVEF at 3 months (P = .01).

On multivariable analysis, the only independent predictors of LVEF recovery at 3 months were:

  • Higher baseline LVEF (P = .004)
  • DMJS of 0 (P < .001) 

However, in analysis of the subgroup with CMR evaluation, LVEF did not correlate with the number of viable segments or wall motion score. Extent of delayed enhancement on gadolinium-enhanced CMR was the only variable that showed a trend toward a correlation with LVEF recovery (P = .09). In this model, independent predictors of LVEF recovery at 3 months were DMJS (P = .039) and delayed enhancement transmurality score (P = .022). 

Disease Severity More Important Than Completeness of Revascularization 

In a telephone interview with TCTMD, Philippe Généreux, MD, of Columbia University Medical Center (New York, NY), noted that debate over the impact of complete revascularization in TAVR parallels that in the coronary arena. But, he said, in TAVR the completeness of revascularization is less relevant than the baseline disease because many patients with multivessel CAD are unsuitable for revascularization, either because they are too sick or the lesions are too complex.

“You have to weigh the risks and benefits of the procedure in very elderly patients,” Dr. Généreux commented. “For example, in the case of a chronic total occlusion with good collaterals in a patient who is 80 or 90 years old, most operators would leave it alone and do the TAVR first.”

Clinicians should perform PCI whenever possible, Dr. Généreux said, but if it is not possible, knowing the patient’s CAD burden and its adverse impact on prognosis at least suggests a need for closer follow-up and perhaps intensified medical treatment. 

Conflicting Results

The current findings should be viewed in the context of other recent studies, Dr. Généreux said. One, by Van Mieghem et al, found that complete revascularization is not a prerequisite to a good prognosis following TAVR, while another, presented at the Transcatheter Cardiovascular Therapeutics symposium in October 2013 (Stefanini GG, et al. J Am Coll Cardiol;62;SuppB:B222) showed that a higher Syntax score predicts worse outcomes.

He added that while the DMJS employed in the current study is a valuable metric in that it defines overall disease burden, it fails to account for CAD complexity. In contrast, the Syntax score incorporates factors that affect the likelihood of lesion revascularization, such as bifurcations, calcification, and tortuosity.

Dr. Généreux suggested the conflict in the medical literature may in part reflect differing answers to key questions related to the degree of disease severity that warrants revascularization and whether ‘reasonable’ revascularization is as good as complete revascularization.

Against the backdrop of controversy over the value of complete revascularization, the current study helps underline the need for a future prospective study, he concluded.

Freixa X, Chan J, Bonan R, et al. Impact of coronary artery disease on left ventricular ejection fraction recovery following transcatheter aortic valve implantation. Catheter Cardiovasc Interv. 2014;Epub ahead of print.

Related Stories:

  • Dr. Asgar reports serving as a consultant to Medtronic.
  • Dr. Généreux reports receiving speaker fees from Edwards Lifesciences.

We Recommend