Table 3

Adjusted association of the continuous BNP and MACNE

Adjusted HR (95% CI)P values
Overall1.11 (1.01 to 1.22)0.0244
No CHF§1.06 (0.93 to 1.20)0.3673
CHF§1.10 (0.96 to 1.26)0.1577
No CKD1.12 (0.98 to 1.27)0.0983
CKD1.07 (0.93 to 1.23)0.3459
Age <75 years1.15 (1.01 to 1.32)0.0367
Age ≥75 years1.07 (0.94 to 1.23)0.3104
Men∗∗1.15 (1.02 to 1.30)0.0276
Women∗∗1.05 (0.90 to 1.22)0.5649
CHA2DS2-VASc score 0–2¶1.03 (0.78 to 1.35)0.8349
CHA2DS2-VASc score 3–9¶1.13 (1.03 to 1.25)0.0135
LAD type (normal/mild enlargement)††1.16 (1.02 to 1.32)0.0248
LAD type (moderate/severe enlargement)††1.05 (0.90 to 1.21)0.5492
  • *Log-transformed BNP was included in the model.

  • †HRs and 95% CIs of every doubling in BNP value.

  • ‡Covariates for adjustment include age, sex, body mass index, systolic blood pressure, diastolic blood pressure, heart rate, intraventricular conduction, cognitive impairment/dementia, frailty, rhythm control, diabetes, anaemia, hypertension, hyperlipidaemia, estimated glomerular filtration rate, left atrial enlargement, cancer, haematocrit, smoking, chronic obstructive pulmonary disease, New York Heart Association (NYHA) class, atrioventricular (AV) node/His bundle ablation, prior cardioversion, prior antiarrhythmic medication use, significant valvular disease, left ventricular ejection fraction type, peripheral vascular disease, atrial fibrillation (AF) type, history of stroke or transient ischaemic attack, aldosterone antagonist, beta-blocker, ACE inhibitor, angiotensin receptor blocker and loop diuretic.

  • §For tests within the CHF subgroups, which was defined as any of New York Heart Association class 1, class 2, class 3 or class 4, New York Heart Association and AV node/His bundle ablation were not included in the model, and moderate+severe dysfunction was combined due to small group sizes.

  • ¶For testing within the CHA2DS2-VASc subgroups (0–2, ≥3), peripheral vascular disease, history of stroke/transient ischaemic attack, AV node/His bundle ablation, and cognitive impairment/dementia were not included in the model, and some of the levels for the following variables were combined (NYHA: I+II, intraventricular conduction: LBBB+RBBB, AF type: persistent+permanent AF).

  • **For testing within the subgroup of men or women, sex and AV node/His bundle ablation were not included in the model.

  • ††For testing within the subgroup of LAD severe/moderate versus other patients, AV node/His bundle ablation was not included in the model.

  • ‡‡Haematocrit was found to violate the linearity assumption. Linear splines of haematocrit (linear spline when haematocrit ≤36.4%, linear spline when 36.4% < haematocrit ≤41.7%, linear spline when haematocrit >41.7%) were used.

  • BNP, B-type natriuretic peptide; CHF, chronic heart failure; CKD, chronic kidney disease; LAD, left atrial diameter; MACNE, major adverse cardiovascular or neurological event.