Background: Experimental evidence suggests that autonomic markers such as heart-rate variability and baroreflex sensitivity (BRS) may contribute to postinfarction risk stratification. There are clinical data to support this concept for heart-rate variability. The main objective of the ATRAMI study was to provide prospective data on the additional and independent prognostic value for cardiac mortality of heart-rate variability and BRS in patients after myocardial infarction in whom left-ventricular ejection fraction (LVEF) and ventricular arrhythmias were known.
Methods: This multicentre international prospective study enrolled 1284 patients with a recent (<28 days) myocardial infarction. 24 h Holter recording was done to quantify heart-rate variability (measured as standard deviation of normal to normal RR intervals [SDNN]) and ventricular arrhythmias. BRS was calculated from measurement of the rate-pressure response to intravenous phenylephrine.
Findings: During 21 (SD 8) months of follow-up, the primary endpoint, cardiac mortality, included 44 cardiac deaths and five non-fatal cardiac arrests. Low values of either heart-rate variability (SDNN <70 ms) or BRS (<3.0 ms per mm Hg) carried a significant multivariate risk of cardiac mortality (3.2 [95% CI 1.42-7.36] and 2.8 [1.24-6.16], respectively). The association of low SDNN and BRS further increased risk; the 2-year mortality was 17% when both were below the cut-offs and 2% (p<0.0001) when both were well preserved (SDNN >105 ms, BRS >6.1 ms per mm Hg). The association of low SDNN or BRS with LVEF below 35% carried a relative risk of 6.7 (3.1-14.6) or 8.7 (4.3-17.6), respectively, compared with patients with LVEF above 35% and less compromised SDNN (> or = 70 ms) and BRS (> or = 3 ms per mm Hg).
Interpretation: ATRAMI provides clinical evidence that after myocardial infarction the analysis of vagal reflexes has significant prognostic value independently of LVEF and of ventricular arrhythmias and that it significantly adds to the prognostic value of heart-rate variability.