AFib screening cuts hospitalizations and ED visits

NEW ORLEANS – People diagnosed with atrial fibrillation by screening with a wearable ECG patch had significantly fewer emergency department visits or hospital admissions, compared with similar people diagnosed with atrial fibrillation by usual-care surveillance in an observational study with 5,109 total participants.

People diagnosed with atrial fibrillation (AFib) through screening had a statistically significant 80% relative cut in hospitalizations and a 65% cut in emergency department visits during 12 months of follow-up, compared with controls in the study who had their AFib identified and diagnosed as part of routine practice, Steven R. Steinhubl, MD, said at the annual meeting of the American College of Cardiology.

^{Mitchel L. Zoler/MDedge News}

Dr. Steven R. Steinhubl

The data also showed no difference between the screened and control patients identified with AFib in the average number of cardiologist consultations during a year of follow-up, and a trend that missed statistical significance for 16% fewer primary care physician visits in Afib patients diagnosed by screening rather than by routine surveillance.

These findings provided some insight into the potential clinical impact of AFib screening in at-risk people. Dr. Steinhubl and his associates plan to report on the incidence of strokes and MIs in the two study subgroups after 3 years of follow-up, but he noted that preliminary findings for these two outcomes after 1 year indicated that active screening for AFib also had reduced these rates, compared with waiting for the arrhythmia to become apparent by emergence of symptoms.

The data came from the mSToPS (mHealth Screening to Prevent Strokes) study, which randomized 2,659 U.S. residents enrolled in a large health plan who had risk factors for AFib to either immediate or delayed arrhythmia assessment by ECG patches. Half the participants used a patch for about 14 days immediately and then a second time 3 months later, while the other half waited 4 months and then wore an ECG patch for 2 weeks and again 3 months later. The primary endpoint, first reported at the ACC annual meeting a year before and subsequently published, was the incidence of newly diagnosed AFib during the first 4 months in the actively monitored cohort, compared with a cohort followed by usual care. The results showed that screening identified AFib in 3.9% of people, while no screening and usual-practice follow-up identified a 0.9% incidence of AFib, showing that screening worked better for AFib case identification (JAMA. 2018 Jul 10;320[2]:146-55).

To examine the clinical impact of screening and an increased incidence of diagnosed AFib cases, Dr. Steinhubl and his associates focused on 1,725 of the original 2,659 patients who underwent ECG patch assessment, either immediate or delayed, and continued through 12 months of follow-up, and compared them with 3,384 matched controls who never underwent ECG patch screening but were also followed for 12 months for incident AFib identified during routine care and surveillance. This resulted in a cumulative incidence of newly diagnosed AFib of 6.3% in those who had worn two ECG patches and 2.3% among the matched controls.

During follow-up, use of various interventions was more common among the screened people than the controls. Initiation of anticoagulation treatment started in 4.0% of the entire screened group, compared with 1.9% of the controls, The screened people also had a 0.9% rate of receiving a pacemaker or defibrillator, a 0.8% rate of starting on treatment with an antiarrhythmic drug, and a 0.3% rate of undergoing catheter ablation, compared with none, 0.3%, and one of the controls, respectively, said Dr. Steinhubl, director of digital medicine at the Scripps Research Translational Institute in La Jolla, Calif.

The mSToPS study was funded by Janssen. Dr. Steinhubl has received research funding from DynoSense, EasyG, Janssen, the Qualcomm Foundation, and Striv.

[email protected]

SOURCE: Steinhubl SR et al. J Am Coll Cardiol. 2019 Mar 12;73(9)suppl 1:296.