First Validated Risk Tool for Frequent PVCs Shows Potential

Patrice Wendling

March 13, 2020

An independently validated risk score may help predict adverse events in patients with frequent premature ventricular contractions (PVCs), new research suggests.

Over the past decade, there has been increasing recognition that frequent PVCs are a cause of cardiomyopathy and, if suppressed with antiarrhythmics or catheter ablation, heart function often improves or returns to normal.

But along with this recognition, and wearable devices like the Apple Watch, came a flood of young patients referred with PVCs and normal hearts, explained senior author Edward Gerstenfeld, MD, chief of cardiac electrophysiology and arrhythmia at the University of California, San Francisco (UCSF).

"That was sort of what led to this, is seeing in clinics all these healthy people with PVCs that are referred to their cardiologists and are worried because they did a Holter for whatever reason, felt their pulse being irregular or felt a skipped beat, and have 20% of their beats as PVCs," he said.

Most of these patients will do fine for years and some will even have spontaneous resolution. But yearly echocardiograms and monitors eat up healthcare resources and may not allay their worries, Gerstenfeld noted.

"A lot of people, after a couple of years, say: 'Doc, can't you just ablate these and be done with the monitoring. I don't want to come back every year'," he said. "So the question is, are all PVCs the same, and can we somehow figure out — even if it's a small group that's going potentially to be at risk of progressing — which of those people are at higher risk and which are not."

To that end, the investigators, co-led by Aleksandr Voskoboinik, MBBS, PhD, and Alexios Hadjis, MD, both from UCSF, analyzed 206 consecutive patients (62% male; mean age, 65 years) from 2012 to 2017 with a PVC burden more than 5% who underwent 14-day single-lead electrocardiographic patch monitoring (Ziopatch) and transthoracic echocardiography within 3 months.

Custom software was designed to analyze the raw ECG data, in particular the coupling interval, which isn't typically available from a patch or Holter monitor, Gerstenfeld noted.

Most patients (81%) had preserved left ventricular ejection fraction (>45%), and 55% had more than one PVC morphology, typically left ventricular in origin (62%) with an inferiorly directed axis (65%).

The average PVC burden was 11.6% but varied considerably from day-to-day (minimum 7.3% vs maximum 17.9%), the authors report February 25 in Heart Rhythm.

In univariate analysis, minimum PVC burden on any day of the 14-day patch monitor was the strongest predictor of adverse LV remodeling (LVEF <45% or LV end-diastolic volume index >75 mL/m2).

Multivariate predictors of adverse remodeling were:

  • nonsustained ventricular tachycardia (VT) (odds ratio [OR], 5.3)

  • longer PVC coupling interval of at least 500 ms (OR, 4.7)

  • PVC burden (OR, 4.4 for >20%; OR, 3.5 for 10% to 20%)

  • Superiorly directed PVC axis (OR, 2.7)

These independent risk factors were used to create the ABC-VT score, with 1 point assigned for superior axis, 2 points for PVC burden of 10% to 20%, 3 points for PVC burden of at least 20%, and 4 points for coupling interval of at least 500 ms or presence of nonsustained VT.

Scores from 0 to 4 were classified as low risk; from 5 to 8, intermediate risk; and from 9 to 12, high risk.

Validation Cohorts

The study involved two validation cohorts without structural heart disease at baseline: follow-up involved 134 healthy patients in the derivation cohort and 559 patients from the Korean PVC registry with at least 5% PVC burden on 24-hour Holter (mean PVC burden, 19.1%).

In the first cohort, 57% of patients were low risk, based on the ABC-VT score, 33% were intermediate risk, and 10% were high risk (mean ABC-VT score, 4.2).

Over 3.2 years of follow-up, 13 patients experienced an adverse event, defined as absolute LVEF decline by 10%, heart failure hospitalization, or cardiovascular mortality. All had an intermediate or high ABC-VT score (range, 5 - 11).

In the slightly younger Korean cohort (55% women; mean age, 56 years), 19 patients had adverse events over 4-year follow-up. ABC-VT scores ranged from 2 to 11 in those with adverse events. Five patients were low risk (1.5% of all patients in this group), 10 were intermediate (4.9% of all patients from this group), and four were high risk (13.8% of patients from this group).

The authors note that use of a 24-hour Holter rather than a 2-week patch in the Korean cohort resulted in a lower pick-up rate for nonsustained VT and, thus, lower overall ABC-VT risk scores.

Still, a higher ABC-VT risk score predicted future adverse events in both the follow-up (hazard ratio (HR), 1.43; 95% CI, 1.19 - 1.73) and Korean (HR, 1.22; 95% CI, 1.05 - 1.42) cohorts.

As for how the ABC-VT risk score itself should be used in clinical practice, Gerstenfeld said it provides reassurance for low-risk patients (score, 0 - 4).

"If I saw someone in the office and their PVC burden's 20% but they have none of the other risk factors, that means their score is 2," he said. "I would say, you're fine but maybe check in with your cardiologist and get it monitored in 5 years, but I'm not so worried about that person; whereas someone who's got a score of 10, I might say, maybe we should get an echo in 6 months and keep a closer eye on them."

It would be a stretch, however, to use the ABC-VT risk score to intervene on people without prospective data, Gerstenfeld added.

"The general sort of challenge with PVCs is the chicken-and-egg phenomenon, and whether this study identifies patients who are at risk of developing PVC-induced cardiomyopathy, or are we just identifying patients with some early subclinical structural heart disease," first author Voskoboinik said in an interview. "This study was a retrospective analysis and, as such, it's very hard to tease that out. Nevertheless, I think a low score, even in that context, is reassuring. The implications of a high score are still unknown."

In terms of the risk factors used to create the score, the study held some surprises.

"We've previously associated shorter coupling intervals with a higher risk of sudden death, so it was very surprising that, in fact, the longer coupling interval appears to be associated with more events, more remodeling over time. So that was certainly significant," Voskoboinik said.

Longer coupling interval also appeared to be a stronger predictor than high PVC burden above 20%, which has always been considered a strong risk factor and often used to initiate pharmacological PVC suppression.

Variability in the PVC coupling interval has been reported to increase the risk of cardiomyopathy and sudden death, but fell out of the multivariate model despite an OR of 15.2 in the univariate analysis.

The coupling interval is not measured in a traditional Holter but could be easily added to the report, Gerstenfeld said. In the meantime, "a simple thing a cardiologist can do is look at an EKG and just take your calipers and measure the coupling interval."

Luigi Di Biase, MD, PhD, head of electrophysiology and director of arrhythmia services at Montefiore Medical Center, New York City, commended the authors for creating a simple, easy to use score to help predict something complex and often grey — when PVC may be malignant for a patient and whether these PVCs should be eliminated or not.

"The idea and the goal of the authors are very good because going from complex to simple is great, but there are some limitations in this score," he told | Medscape Cardiology. "The main weakness or limitation of the ABC-VT score is that the authors have built this score on the assumption that the decline in EF is correlated to increased mortality, but they actually didn't have any data on mortality. And, surprisingly, also they have not utilized in their score two previously reported parameters that have shown increased mortality in the presence of PVC."

First of all, the ABC-VT risk score does not take into consideration the PVC coupling interval dispersion, which when greater than 60 or 90 ms, is associated to increase mortality and profound LV decline, respectively.

Second, although the score included a longer PVC coupling interval of at least 500 ms, the authors failed to include a shorter PVC coupling interval of 400 ms or less, which may result in R-on-T phenomenon and cause polymorphic ventricular tachycardia or even ventricular fibrillation, said Di Biase, who is also a professor of medicine at Albert Einstein College of Medicine and author of a forthcoming editorial on the study.

The authors report no relevant disclosures. Di Biase is a consultant for Stereotaxis, Biosense Webster, Boston Scientific, and Abbott (formerly St.  Jude Medical) and has received speaker honoraria/ travel from Medtronic, Atricure, Pfizer, Bristol Myers, and Biotronik.

Heart Rhythm. Published online February 25, 2020. Abstract

Follow Patrice Wendling on Twitter: @pwendl. For more from | Medscape Cardiology, join 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.
Post as: