Stroke in patients with heart failure and reduced ejection fraction without atrial fibrillation: external validation of a risk model
Bibliographic record
Abstract
Abstract Background Heart failure (HF) ranks only second to atrial fibrillation (AF) as a cause of cardio-embolic stroke. Although anticoagulation reduces this risk in HF patients not in AF, the risk/benefit profile in relatively unselected populations is not favourable. Identification of patients at high risk of stroke may allow targeted and safer use of prophylactic anticoagulant therapy. Previously, we proposed a simple risk model for stroke in patients with HF and reduced ejection fraction (HFrEF). However, this model was derived from the two older trials (published in 2007/2008) and was not externally validated. Purpose We aimed to evaluate the current incidence of stroke in patients with HFrEF not in AF receiving modern pharmacological therapy and to validate our stroke prediction model. Methods We examined patient-level data from the PARADIGM-HF, ATMOSPHERE, and DAPA-HF trials. The risk score was calculated following: 7.39×(insulin-treated diabetes) + 6.53×(previous stroke) + 2.80×[ln(NT-proBNP (pg/ml)) × 0.1182]). According to the tertile of risk score, we divided the patients into three groups. Patients with AF were defined as those with either AF on an ECG or a history of AF. Results Of the total of 20,159 patients (who experienced 590 strokes) enrolled in the three trials, 12,751 patients did not have AF at baseline. Of those, 1,143 patients (9%) had insulin-treated diabetes, 873 patients (6.8%) had a history of the previous stroke, and the median value of NT-proBNP was 1,243 pg/ml. During a median follow-up of 2.0 years, 346 (2.7%) experienced a stroke (11.7 per 1000 patient-years). Figure 1 shows cumulative incidence function plots for stroke according to the tertile of risk score in 12,331 patients whose risk score can be calculated. The number of strokes in tertile 1, 2 and 3 were 80, 102 and 149, respectively. The 3-year cumulative incidence function rates of stroke were 2.0 (95% CI: 1.5–2.5) % in tertile 1, 2.6 (95% CI: 2.1–3.2) % in tertile 2, and 4.3 (95% CI: 3.6–5.2) % in tertile 3, respectively. In patients with tertile 3, the stroke rate was 18.1 per 1000 patient-years (compared to 20.1 per 1000 patient-years in patients with AF not receiving anticoagulation). In the Cox model, risk for stroke increased according to the elevation in the risk score (tertile 2: HR 1.47 (95% CI 1.09–1.97), tertile 3: HR 2.53 (95% CI 1.92–3.33), with tertile 1 as reference). Figure 2 shows calibration plots by comparing observed and predicted probabilities of stroke at 1 to 3 years. Discrimination evaluated using the overall c-index 0.84 (95% CI: 0.75–0.91) was good. Conclusions These findings validate a previously described predictive model and confirm that it is possible to identify a subset of HFrEF patients without AF who have a risk of stroke that approximates to that in patients with AF. In these patients, the risk/benefit balance might justify the use of prophylactic anticoagulation, but this hypothesis needs to be tested prospectively. Funding Acknowledgement Type of funding sources: Foundation.
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 | 0.045 | 0.071 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
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".