Ratio of Pulse Pressure to Left Ventricular End-Diastolic Pressure Predicts In-Hospital and One-Year Mortality in Patients With ST-Elevation Myocardial Infarction

Tuesday, May 21, 2019
Belmont Ballroom 2-3 (The Cosmopolitan of Las Vegas)
Cynthia Zhou, B.S. , UNC Hospitals, Chapel Hill, NC
Robert A Rayson, M.D. , UNC Healthcare Systems, Chapel Hiill, NC
Allie Goins, M.D. , UNC Hospitals, Chapel Hill, NC
Michael Yeung, M.D. , Barnes Jewish Hospital at Washington University School of Medicine, Chapel Hill, NC
Joseph Rossi, M.D., FSCAI , University of North Carolina Hospitals, Chapel Hill, NC
Rick Stouffer, M.D. , UNC McAllister Heart Institute , Chapel Hill, NC

Background
Invasive hemodynamic measurements such as left ventricular end-diastolic pressure (LVEDP) and systolic blood pressure (SBP) obtained at the time of primary percutaneous coronary intervention are predictive of mortality in ST-elevation myocardial infarction (STEMI). We hypothesized the combination of pulse pressure (PP) and LVEDP would provide increased ability to predict early and late mortality.

Methods
This was a cross-sectional, single-center study of 536 adult patients with STEMI who had LVEDP measured using 6F fluid filled catheters. Size of infarction was estimated by LV ejection fraction determined by echocardiography and by peak troponin I level. PP was calculated as the difference between SBP and diastolic blood pressure (DBP).

Results
The median [25%, 75%] age was 61 [53, 71] years, heart rate was 80 [68, 93] bpm, LVEDP was 18 [13, 24] mm Hg, SBP was 126 [109, 146] mm Hg, DBP was 75 [64, 87] mm Hg, PP was 52 [39, 64] mm Hg, SBP/LVEDP ratio was 7 [5.0, 10.5] and PP/LVEDP ratio was 2.89 [1.96, 4.50]. PP/LVEDP had a weaker correlation with ejection fraction (r = 0.210, p<0.0001) than did LVEDP (r = -0.380, p<0.0001) or heart rate (r = -0.304, p<0.0001) although the correlation was stronger than PP alone (r = 0.191, p<0.0001) or SBP/LVEDP ratio (r = 0.171, p<0.0001). SBP (r = 0.0813, p = 0.054) and DBP (r = -0.0767, p = 0.071) had no correlation with ejection fraction. PP/LVEDP had a weak, inverse correlation with peak troponin (r = -0.18, p<0.0001); by comparison, correlation of other hemodynamic parameters with peak troponin were PP (r = -0.16, p = 0.0005), SBP/LVEDP ratio (r = -0.16, p<0.0005), LVEDP (r = 0.29, p<0.0001) and SBP (r = -0.17, p<0.0003). In-hospital mortality was 2.2%, 3.7%, 6.7% and 12.7% and one-year mortality was 5.2%, 6.0%, 8.2% and 16.4%, respectively in quartiles of highest to lowest PP/LVEDP (P for trend <0.0001). By comparison, in-hospital mortality was 4.4%, 6.0%, 6.7% and 8.2% and one-year mortality was 7.5%, 8.2%, 8.2% and 11.9%, respectively in quartiles of lowest to highest LVEDP and 1.5%, 1.5%, 4.4% and 17.9% and 4.4%, 5.2%, 9.7% and 17.9%, respectively in quartiles of highest to lowest SBP.

Conclusions
PP/LVEDP was predictive of in-hospital and one-year mortality in STEMI.