Estimated incidence of previously undetected atrial fibrillation on a 14-day continuous electrocardiographic monitor and associated risk of stroke: comment—Authors’ reply
Bibliographic record
Abstract
This is a response to the Letter to the Editor, ‘Estimated incidence of previously undetected atrial fibrillation on a 14-day continuous electrocardiographic monitor and associated risk of stroke: comment’ by Adithya Sreeniva and Mahmood Ahmad https://doi.org/10.1093/europace/euac206, about the article, ‘Estimated incidence of previously undetected atrial fibrillation on a 14-day continuous electrocardiographic monitor and associated risk of stroke’ by William F. McIntyre et al.https://doi.org/10.1093/europace/euab324. We thank Sreenivas and Ahmad for their interest in our work. We used pacemaker data from participants in asymptomatic atrial fibrillation and stroke evaluation in pacemaker patients and the atrial fibrillation reduction atrial pacing trial (ASSERT) to simulate 14-day Holter monitors used for atrial fibrillation (AF) screening in patients aged ≤65 with hypertension.1 The proportion of patients who would have a total duration of AF ≥6 min was estimated at 3.1%. This finding was associated with a tripling of the hazard for stroke. The readers request exploration of the relationship between CHA2DS2-VASc and stroke/systemic emboli in ASSERT patients without any AF. A previous analysis of ASSERT reported that among patients with no subclinical AF during follow-up, 19 strokes or systemic emboli occurred in 1811 patients, corresponding to an event rate of 0.54%/year.2 With so few clinical events, we would not have the power to appropriately investigate this relationship. Moreover, we would have no ability to test whether such events in high CHA2DS2-VASc patients without AF would be sensitive to oral anticoagulation (OAC). Larger, observational data sets with contemporary monitoring are better suited to answer the question of baseline risk and appropriately designed randomized trials would be required to assess the role of OAC in this population. Two large randomized trials have already failed to show that OAC was superior to aspirin for the prevention of recurrent stroke in patients with a prior history of embolic stroke of undetermined source (ESUS). Among these, rivaroxaban versus aspirin in secondary prevention of stroke and prevention of systemic embolism in patients with recent embolic stroke of undetermined source (NAVIGATE-ESUS) showed that OAC increased bleeding.3 The readers also asked about the relationship between the burden of premature atrial contractions (PACs) and the risk of stroke/systemic embolism. The pacemakers used in ASSERT did not collect these data. Interestingly, a sub-study of NAVIGATE-ESUS showed that high PAC counts did not predict response to OAC in patients with ESUS but without AF4 The readers wonder about the relationship between AF episodes that lasted <6 min and stroke. ASSERT began in 2000, when device-based AF detection algorithms were less sophisticated compared with today’s technology. In ASSERT, physicians reviewed all device-detected AF lasting ≥6 min, and 50% of shorter episodes.5 Of the more than 10 000 adjudicated episodes lasting <6 min, only 50% were actually AF; these episodes were totally impractical for clinical or research use. As a result, ASSERT focused on device-detected AF episodes that lasted ≥6 min, where the positive predictive value was 83%, although physician review was still necessary. Although the readers are concerned about the risk associated with short AF episodes, it was uncommon for individuals with device-detected AF to have only short episodes. The average and median AF burdens over 14 days of simulated monitoring were 55.3 ± 104.7 h and 6.1 (interquartile range 1.1–38.3) hours, respectively. Each data set has its strengths and weaknesses. Ten years after the original publication, the strengths of ASSERT remain the completeness of monitoring and the very low rate of OAC use. Unfortunately, it cannot tell us about AF events that were shorter than 6 min and the relatively small number of events makes subgroup analyses challenging. We believe the questions raised by Sreenivas and Ahmad are interesting and important but are best left to other studies.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.008 | 0.082 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.037 | 0.030 |
| Insufficient payload (model declined to judge) | 0.006 | 0.007 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".