Survival Benefit Signal After Atrial Fib Screening With Patch ECG Monitor: mSToPS

November 18, 2020

Active screening for atrial fibrillation (AF) not only yielded more diagnoses of new AF compared to standard management in a study, it may have improved clinical outcomes such as stroke, death, and hospitalization for bleeding over a 3-year follow-up.

Still, oral anticoagulants were prescribed just as often for the more than 1700 participants following the screening strategy compared to the almost 3500 control persons who received usual care in the prospective but nonrandomized trial, called mHealth Screening to Prevent Strokes (mSToPS).

The study's participants were selected from an insurance company database and required to have risk factors for AF and no history of AF and be free of symptoms. They either agreed to wear the monitor — a small, skin-adhesive patch electrocardiographic (ECG) monitor (Zio XT) for up to 4 weeks — or they declined and became the control group. Their physicians used the monitor data to help guide management decisions.

There seems to be a mindset in cardiology that AF management means measures aimed at stroke prevention, mused Steven R. Steinhubl, MD, to | Medscape Cardiology. "Now, that's hugely valuable," he said. In the setting of AF, "few things in cardiology are better in terms of preventing events than anticoagulation."

But the impact of AF screening on mortality was unusual and one of the more interesting results of mSToPS, said Steinhubl, of Scripps Translational Science Institute, La Jolla, California. The risk of death from any cause fell by nearly 40% in monitored patients compared with controls, and by nearly 60% solely among those who had received a diagnosis of AF during the study.

The trial's three-year findings, which Steinhubl unveiled November 16 during the American Heart Association (AHA) Scientific Sessions 2020 virtual meeting, are more than just an extension of its 1-year findings presented and published in 2018. The earlier findings included greater use of oral anticoagulants, antiarrhythmic medications, and catheter ablation in the monitored group, but didn't include clinical outcomes such as mortality or stroke.

In the prespecified 3-year analysis, however, the monitored group showed a 21% decline in risk for the composite of stroke, myocardial infarction (MI), systemic embolism, or death (P < .01), which included a 39% drop in mortality alone (P = .047). The risk of hospitalization for bleeding, the primary safety endpoint, fell 53% (P = .01).

Solely among those who received an AF diagnosis during the study, the risk of the composite clinical endpoint fell 47% and risk of stroke went down 50% (P < .01 for both) for participants in the monitoring group.

The Zio XT patches worn in mSToPS, available for more than a decade, can retain up to 2 weeks of continuous single-lead ECG monitoring data. The participants were given two patches to use, each for 2 weeks separated by 3 months for a total maximum of 4 weeks per person. The median total monitoring time was 24.7 days, unusually long for a large AF screening study.

But given the study's limitations, especially any biases that escaped its 1:2 case-control matching, the 3-year results would require independent confirmation "to be confident that aggressive pursuit of diagnosing atrial fibrillation in people at high risk but without symptoms is warranted," concluded Steinhubl when presenting mSToPS.

Management recommendations in the guidelines tend to consider simply whether or not AF is present, observed Rod S. Passman, MD, director of the Center for Arrhythmia Research at Northwestern University Feinberg School of Medicine, Chicago, Illinois.

"But I think that as clinicians, now that we have long-term monitoring, we're realizing that this is not a dichotomous variable. And we are subconsciously using this in our decision-making process, I believe," he told | Medscape Cardiology.

"We're learning with more prolonged monitoring that there's a lot of richness in how much atrial fibrillation one can have. And there may be an interaction between AF duration and underlying stroke risk. But the burden may not be the same for everyone. The amount of atrial fib needed for stroke may change according to underlying vascular risk factors," he noted.

"mSToPS suggests that there are some potential long-term benefits of monitoring, as we might expect if we find people earlier in the course of the disease," said Passman, who isn't connected with the study.

He agreed the study has major limitations and raises some questions, and added the highly selected population and a surprising and unexplained impact on mortality to the list. "I think that until we have a randomized trial showing that a patch technology finds more atrial fibrillation and that treating it reduces hard endpoints, I think it's only hypothesis-generating."

The 2659 monitored intention-to-treat (ITT) mSToPS participants were among more than 100,000 Aetna fully insured commercial and Medicare Advantage members who met the entry criteria, were invited, and consented to use the device as part of the study. Eligibility required participants be at least age 75, (except for those with at least one comorbidity associated with AF risk; the eligibility age there was at least 55 for men and at least 65 for women). Entry also called for no history of AF or other atrial arrhythmias and no other indication for oral anticoagulation.

Invitees who agreed to participate but did not consent to wear the patch monitor were matched to the monitoring group by age, sex, and CHA2DS2-VASc score to yield the 5310-member observational control cohort.

All subjects were managed by their own physicians, who for monitored participants received little information from the trialists on how to use the device's ECG results in management decisions.

Three-year follow-up was available for the "as-treated" cohorts of 1718 monitored and 3371 control subjects, who had an average age of 74 years at entry with a median CHA2DS2-VASc score of 3.

Over a mean of 29 months, 11.4% of the monitored group and 7.7% of the controls received a new diagnosis of AF (P < .001); oral anticoagulation was prescribed in 44% and 45.2%, respectively (P = .84), among those with available pharmacy records. Of all new AF diagnoses in the monitored group, 32% were based on data from the ECG monitor and 68% were made clinically.

The adjusted hazard ratio (HR) for the composite primary endpoint, monitoring group vs the control group, was 0.79 (P < .01) across the cohorts and 0.53 (P <.01) solely among those in either group who received a new diagnosis of AF, Steinhubl reported.

The primary endpoint rate, as events per 100 person-years, was 13.8 among those in the control group who received a new diagnosis of AF; 10.9 for monitored subjects with a new AF diagnosis first made clinically; and 2.6 for subjects in whom a new AF diagnosis was first made based on the ECG monitor (P <.01 compared to the other two groups).

 Table. Primary and Secondary Endpoints in mSToPS
3-Year Analysis


Subjects with a New AF Diagnosis, HR (95% CI) P value

All subjects, HR (95% CI) P value

Stroke, MI, Systemic Embolism, or Death

0.53 (0.37-0.77)
< .01

0.79 (0.66-0.93)
< .01


0.50 (0.29-0.84)
< .01

0.79 (0.62-1.01) .06


0.68 (0.38-1.19) .18

0.82 (0.61-1.10) .19

Systemic Embolism

0.44 (0.18-1.07) .07

0.81 (0.55-1.20) .29


0.41 (0.18-0.98) .046

0.61 (0.37-0.99) .047

*Prespecified primary endpoint for the 3-year analyses, data obtained from claims and membership records.

"I think there's value to identifying atrial fibrillation in an asymptomatic population. It may not mean they need to be anticoagulated. For that, you have to look at the whole picture, their CHA2DS2-VASc score and other things," Steinhubl said when interviewed.

"But it does mean you can look for sleep apnea, you can recommend weight loss, you can recommend alcohol abstinence — all things that have been proven to decrease atrial-fib burden and prevent or minimize the progression of atrial fibrillation."

Steinhubl reports receiving grants from Janssen, Qualcomm Foundation and other funding from DynoSense, EasyG, Livongo, Otsuka, Spry Health, and Striiv. Passman discloses receiving honoraria from Medtronic and Abbott; having a relationship with UpToDate; and receiving research grants from Medtronic and Pfizer.

American Heart Association Scientific Sessions 2020: Abstract LBS.06, 3-year Clinical Outcomes in a Nationwide, Randomized, Pragmatic Clinical Trial of Atrial Fibrillation Screening - Mhealth Screening to Prevent Strokes (mSToPS). Presented November 16, 2020.

Follow Steve Stiles on Twitter: @SteveStiles2. For more from, 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.