Deep Dive Into COAPT Echo Results Raises More Questions

Patrice Wendling

March 21, 2019

NEW ORLEANS — An echocardiographic substudy of the COAPT trial failed to identify any baseline echo characteristics to differentiate patients with heart failure (HF) and secondary mitral regurgitation (MR) that would or would not benefit from the MitraClip (Abbott).

The MitraClip provided "substantial death and heart failure hospitalization benefits in all the cohort-adjusted subgroups, regardless of the degree of left ventricular (LV) dysfunction, LV dimensions, pulmonary hypertension, severity of TR [tricuspid regurgitation], or individual mitral regurgitation parameters," Federico Asch, MD, Medstar Health Research Institute, Washington, DC, said here at the ACC 2019 Scientific Session (ACC.19). "All subgroups were responders."

Speaking with | Medscape Cardiology, Asch said there is a possibility that the extremes in the cohort actually did behave differently.

"Someone, let's say, with an ejection fraction of 45% and someone with an EF of 22%, those patients probably did behave differently," he said. "We did not perform that analysis here. So we may have to break the population into tertiles or quartiles or find thresholds to define who can do better or not."

"But as of now, the data that we have do not support that we should be cherry-picking patients here," he said. "If patients have the criteria to get into the COAPT trial, they should have a clip."

Last year, COAPT investigators reported that the MitraClip reduced the primary end point of HF hospitalizations at 2 years by 47% over guideline-directed medical therapy alone (GDMT), and reduced mortality by 38%.

The results were met with wild enthusiasm but also skepticism in light of the French MITRA-FR trial, which used the same device but showed no added benefit over medical therapy for the primary composite end point of all-cause death and unplanned hospitalization at 1 year.

Despite being unable to find echocardiographic nonresponders and using a relatively similar composite end point, 45.4% of patients treated with the MitraClip and 67.4% of patients treated with GDMT alone died or had a first HF hospitalization at 2 years, Asch reported.

He noted that expert panels have disagreed on how to define secondary MR severity, resulting in conflicting European and American guidelines. Although MITRA-FR used the European guidelines, COAPT adapted the American guidelines and created a three-tier screening protocol.

At baseline, 85.7% of the 665 patients met the tier 1 criteria of an effective regurgitant orifice area (EROA) of at least 0.3 cm2 or pulmonary vein systolic flow reversal.

Another 10.5% of patients met the tier 2 criteria of an EROA of 0.2 cm2 to less than 0.3 cm2 with one of three additional criteria: regurgitant volume (RV) of at least 45 mL/beat; regurgitant fraction (RF) of at least 40%; or vena contracta (VC) width of at least 0.5 cm.

Finally, 3.8% of patients met the tier 3 criteria of either an EROA not measured or less than 0.2 cm2 with at least two additional criteria: RV of at least 45 mL/beat; RF of at least 40%; VC width of at least 0.5 cm; proximal isovelocity surface area (PISA) radius greater than 0.9 cm but continuous wave Doppler of MR jet not done; large (≥6.0 cm) holosystolic jet wrapping around the left atrium; or peak E velocity of at least 150 cm/s.

Asch stressed the need to follow the COAPT enrollment criteria if the results are to be replicated in clinical practice. This takes on added importance after the US Food and Drug Administration's decision just days ago to extend the MitraClip indication to include secondary or functional MR based on the COAPT results.

For the late-breaking analysis, Asch and colleagues used median values for 16 different baseline echocardiographic parameters to examine their impact on mortality and first HF hospitalization at 2 years. "Remarkably, all of the subgroups showed a benefit for the MitraClip," he said.

Single Predictor

The team also conducted a Cox regression analysis to identify predictors of the composite primary end point after treatment and found the usual suspects for the medical therapy group:

  • tricuspid regurgitation severity (hazard ratio [HR], 1.60; 95% CI, 1.07 - 2.39)

  • LVEF (HR, 0.98; 95% CI, 0.96 - 1.00)

  • RV systolic pressure (HR, 1.01; 95% CI, 1.00 - 1.02)

  • EROA by PISA (HR, 3.15; 95% CI, 1.08 - 9.21)

RV systolic pressure (RVSP) was the only echocardiographic predictor of poor outcomes after MitraClip (HR, 1.02; 95% CI, 1.01 - 1.04).

"I think this is probably the strongest point of this whole presentation," Asch told | Medscape Cardiology. "Once we took care of the mitral regurgitation by using a MitraClip, then the severity of MR, the severity of cardiomyopathy, the severity of tricuspid regurgitation were not predictors anymore."

Responding to a question during the panel discussion, Asch said they had not yet analyzed whether tenting of MV leaflets and apical displacement — a predictor of failure after MitraClip surgery for functional MR — had an impact on outcome.

Dual Disease

During the discussion, panelist Catherine Otto, MD, University of Washington, Seattle, remarked: "I think it is unfortunate that in both arms the death rates remain high," but questioned RVSP as a predictor of poor outcomes because enrollment criteria excluded severe pulmonary hypertension.

"I don't know what number was reached with that, but even so, pulmonary hypertension still turned out to be a predictor of adverse outcomes, so is something else going on?" she said. "Is the pulmonary pressure a reflector of mitral regurgitant severity or do these patients have severe lung disease, pulmonary vascular disease, some other reason? And should a lower pulmonary pressure be used to exclude patients from this procedure based upon that result?"

Asch responded that COAPT excluded patients with a pulmonary pressure greater than 60 mm Hg (mean, 44 mm Hg) and thus did not have patients with severe pulmonary hypertension (PH).

"But I think it's important to understand that the high pulmonary pressures in this population are not necessarily because of the mitral regurgitation by itself, but also because of cardiomyopathy," he continued. "These patients have a dual disease and we are taking care with the MitraClip of the mitral regurgitation only. Is there any other threshold we should be using for this? I really don't know."

Speaking with | Medscape Cardiology Siddharth Singh, MD, Cedars-Sinai Medical Center, New York City, said: "As someone who does pulmonary hypertension, I think this just reflects how important the right heart is. The fact is that even in spite of MitraClip implantation, the mortality rate was still high."

"The question is how do we come up with newer therapies, medical therapies for the right heart," Singh said. "There's so much development being done on device therapies for tricuspid regurgitation, but we still only have three broad groups of drugs for pulmonary hypertension: PDE5 inhibitors, endothelin receptor antagonists, and prostacyclins. But most of those therapies, I cannot use for group 2 PH patients" and "most of these patients have group 2 PH."

Panelist Michael Borger, MD, PhD, Leipzig Heart Center, Germany, followed up on this point, asking: "Should we be tending, in patients with pulmonary hypertension, toward perhaps transcatheter mitral valve replacement in that difficult subset of patients?"

Asch replied: "I think it is an interesting hypothesis to work on. So far that is not a therapy treatment as of today but in the future, it may be. So it's going to be a very interesting aspect to explore in the future."

Limitations of the analysis are that echocardiography was blinded to clinical condition but not to intervention, PISA and hemodynamics after MitraClip have "major limitations," and multiple imputation methods were used to account for missing follow-up data, Asch said.

Experts recently hypothesized that characterization of MR as proportionate or disproportionate to LV end diastolic volume may explain the disparate results between the two pivotal trials. They observe that MITRA-FR enrolled patients who had MR that was proportionate to the degree of LV dilatation, whereas COAPT patients had an EROA about 30% higher but LV volumes that were about 30% smaller, indicative of disproportionate MR.

Commenting on the echo analysis for | Medscape Cardiology, Milton Packer, MD, distinguished scholar in cardiovascular science, Baylor University Medical Center, Baylor Scott & White Health, Dallas, and a coauthor of the theory, said via email that "there are no data here that help identify candidates," but a patient-level meta-analysis of the two trials is planned.

"This meta-analysis is what we are all waiting for," Packer said. "If the patient-level data can be obtained and merged, then the 'proportionate vs disproportionate' framework can actually be formally tested. If our framework is correct, we will then be able to provide very specific guidance about patient selection."

Asch agreed that future analyses are needed to understand why the two trials had such different results.

"Maybe they actually enrolled different patients," he said when interviewed. "I think pooling the data and understanding the results of one and the other will help us better understand which are the right patients for this technology. But the results of both studies are extremely important and extremely strong. I think we need to cope with that and get a good understanding out of the two of them."

COAPT is funded by Abbott. Asch reports that his institution has contracts with Abbott, Boston Scientific, Edwards, Medtronic, Neovasc, Livanova, GDS, and Mitralign. Otto reports no relevant conflicts of interest. Singh reports serving as the echo investigator for the TRILUMINATE trial.

American College of Cardiology (ACC) 2019 Scientific Session: Abstract 404-11. Presented March 17, 2019.

Follow Patrice Wendling on Twitter: @pwendl. For more from | Medscape Cardiology, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.