乌鸦传媒

Whether screening for atrial fibrillation (Afib) improved detection rates depended on how screening was administered and in whom, research suggested.

The VITAL-AF trial showed that point-of-care screening did not result in more new Afib diagnoses in primary care -- perhaps except in the oldest patients -- whereas mSToPS found that continuous monitoring with a wearable ECG patch did lead to more Afib detected and even better outcomes.

Both trials were presented during a late-breaking session at this year's American Heart Association virtual meeting.

"If you look, you find more Afib than if you don't look," said session discussant Ben Freedman, MBBS, PhD, of Concord Hospital and the University of Sydney in Australia.

He noted that this is true with more intense ECG screening, hence the uptick in Afib diagnoses in mSToPS but not VITAL-AF. In contrast, screening may not work out with single timepoint screening and if the usual care comparator is already doing a good job prescribing anticoagulation for stroke prevention, he said.

VITAL-AF

Routine screening for Afib at primary care visits did not yield more diagnoses -- except in the oldest patients, a large cluster-randomized trial found.

In , primary care practices in the Massachusetts General Hospital (MGH) network were randomized to point-of-care rhythm assessment in patients ages 65 and older or usual care. This resulted in similar rates of new Afib cases at 12 months (1.74% vs 1.60%, P=0.33), according to Steven Lubitz, MD, MPH, of Massachusetts General Hospital.

Nevertheless, the effectiveness of screening appeared to differ by age, such that people older than 85 years saw a larger effect against controls. The number needed to screen to detect a case of Afib was 53 in these oldest individuals, Lubitz said.

Thus, screening everyone over 65 years of age is not an efficient way to detect undiagnosed Afib, whereas screening older individuals may be effective, he concluded.

Patients in the routine screening arm had increased likelihood of Afib diagnosis at a primary care encounter (~25% vs ~18%, P=0.02).

Among people with newly diagnosed Afib, there was no between-group difference in new anticoagulation prescriptions, which was already high (above 70%) in both routine screening and usual care, according to Lubitz.

Usual care is therefore already very good in the MGH network, Freedman commented. "You don't need screening if you're doing a good job already."

Conflicting evidence for routine Afib screening has led to conflicting guidance, Lubitz said, citing with pulse palpitation or ECG rhythm strips and the USPSTF's determination that there is insufficient evidence to recommend screening with ECG.

VITAL-AF is now the largest study so far in this domain, Freedman noted.

Investigators had identified 22 practices that were eligible for the trial. Ultimately, 16 were randomized, with over 15,000 patients ages 65 and older in each arm. Each practice spent 12 months performing routine Afib screening or usual care.

Study participants averaged 74 years of age. Slightly over half were women, and 83% were non-Hispanic white. The mean CHA₂DS₂-VASc score was 3.4.

In the intervention arm, 91% of eligible patients were screened (vs 2% in the control arm).

Afib screening was performed with a handheld, single-lead ECG at vital sign assessment by clinic staff. ECG tracings were integrated in the health system's electronic health records and adjudicated manually by cardiologists.

Lubitz acknowledged that the study may not be generalizable to other institutions and was limited by the lack of direct ascertainment of pulse palpation among controls, the use of 30-second single-lead tracings that may not identify paroxysmal Afib, and the possibility of contamination due to increased attention to Afib in control practices.

mSToPS

Afib screening with a wearable ECG patch improved Afib diagnosis and clinical outcomes over routine care, according to a trial of Aetna members.

The 1,659 individuals who agreed to undergo continuous monitoring (median monitoring time 24.7 days) received an Afib diagnosis in 11.4% of cases compared with 7.7% among matched controls (P<0.01), reported Steven Steinhubl, MD, of Scripps Research Translational Institute in La Jolla, California.

In the entire mSToPS cohort, for the composite endpoint of death, stroke, systemic embolism, or MI, there were 4.5 events versus 5.5 events per 100 person-years in the two groups (adjusted HR 0.79, 95% CI 0.66-0.93), with the difference driven by reduction in death over a median 29 months of follow-up.

Among Afib-diagnosed individuals, continuous monitoring remained associated with fewer events (8.4 vs 13.8 per 100 person-years, adjusted HR 0.53, 95% CI 0.37-0.77), with the difference driven by reductions in stroke.

For safety data, active monitoring was associated with a lower incidence of hospitalization for a primary bleeding diagnosis (0.32 vs 0.71 per 100 person-years, adjusted incident RR 0.47, 95% CI 0.26-0.85) and total hospitalizations (12.9 vs 18.9 per 100 person-years, adjusted incident RR 0.69, 95% CI 0.63-0.76).

"Active screening for Afib, as part of a prospective, pragmatic, direct-to-participant, nationwide study, was associated with a significant improvement in clinical outcomes and safety at 3 years relative to routine care," Steinhubl concluded.

"Independent replication of these findings is required in order to be confident that aggressive pursuit of diagnosing Afib in people at high risk, but without symptoms, is warranted," he cautioned.

Notably, the use of anticoagulants was under 50% in both cohorts.

The enrolled 1,659 Aetna members by contacting them directly via email or physical letters. Another 5,310 people served as matched controls. At 3 years' follow-up, there were 1,718 participants in the monitoring arm and 3,371 matched controls.

Earlier findings indicated that the wearable monitoring increased rates of Afib diagnosis.

Mean participant age in the current analysis was 73.8 years, and nearly 41% were women. The median CHA₂DS₂-VASc score was 3 in both groups.

Despite matching, the active monitoring arm was more likely to have history of stroke (12.7% vs 9.6%, P<0.01) and sleep apnea (26.7% vs 20.8%, P<0.o1). However, controls were more likely to have prior MI (6.8% vs 5.3%, P=0.03) and chronic obstructive pulmonary disease (10.1% vs 8.0%, P=0.01).

Outcomes differed between active monitoring patients who received Afib diagnosis with a patch first compared with a clinical diagnosis first: the primary combined endpoint reached 2.6 versus 10.9 per 100 person-years (P<0.01).

Steinhubl noted the possibility of unmeasured confounding in mSToPS and that clinical follow-up was limited to the duration of health plan enrollment, which was less than 3 years for some people.

Results on prognosis in the trial are important, but only hypothesis-generating, Freedman cautioned.

The main caveat of the study, he said, was that study participants represented just 1.7% of Aetna members who had been invited. There may be important differences between these individuals and those who did not respond, he suggested.

Event rates may be better with a strategy of screening versus no screening, but a randomized controlled trial would be required to convince the USPSTF and payers, according to Freedman.