Table 1.
Demographics and baseline characteristics in the derivation and external validation cohorts
| Variable | Derivation N = 14 980 | Validation N = 9369 |
|---|---|---|
| Age (years) | 70.0 (63.0–76.0) | 72.0 (67.0–77.0) |
| Gender: female | 35.6% (5331) | 36.4% (3408) |
| Current smoker | 8.1% (1219) [14] | 7.7% (721) [0] |
| Alcohol | 2.5% (378) | 14.1% (1322) |
| Antiplatelet/NSAID | 38.6% (5778) | 43.4% (4063) |
| Diabetes | 24.7% (3697) | 22.2% (2079) |
| Heart failure | 31.0% (4651) [0] | 28.9% (2711) [1] |
| Hypertension | 87.5% (13 114) | 78.9% (7391) |
| Permanent or persistent AF | 84.9% (12 711) [3] | 67.6% (6333) [4] |
| Prior stroke/TIA | 18.7% (2809) | 19.5% (1823) |
| Prior myocardial infarction | 12.9% (1926) [1] | 17.0% (1589) [0] |
| Peripheral arterial disease | 4.9% (730) [1] | 3.7% (345) [1] |
| Vascular disease | 24.8% (3723) | 19.4% (1819) |
| Renal function (CKD-EPI, mL/min) | 56.0 (45.3–68.0) [9] | 64.8 (53.7–77.0) [91] |
| cTnI-hs (ng/L) | 5.4 (3.3–10.1) [158] | 6.9 (4.3–13.0) [454] |
| cTnT-hs (ng/L) | 11.0 (7.5–16.7) [82] | 12.2 (7.7–19.5) [650] |
| GDF-15 (ng/L) | 1383.0 (977.0–2052.0) [181] | 1514.0 (1109.5–2194.0) [718] |
| Hemoglobin (g/dL) | 14.2 (13.2–15.3) [70] | 14.3 (13.2–15.3) [159] |
| NT-proBNP (ng/L) | 713.5 (363.0–1250.0) [87] | 814.5 (391.0–1438.2) [73] |
Variables are represented as median (Q1–Q3) or percentage (frequency). Numbers in brackets represents the number of missing values. A total of 8548 subjects in the validation cohort had complete data on all variables included in the final model and were used in the validation of the new risk model.
NSAID, non-steroidal anti-inflammatory drugs; AF, atrial fibrillation; TIA, transient ischaemic attack; eGFR, estimated glomerular filtration rate; cTn-hs, cardiac troponin with high sensitivity assay; NT-proBNP, N-terminal fragment B-type natriuretic peptide; GDF-15, growth differentiation factor 15; IL-6, interleukin-6.