High-sensitivity C-reactive protein in patients with suspected myocardial infarction classified to the observe zone
J Lehmacher, B Strauss, B Toprak, P M Haller, A Schock, N Thiessen, F Dehkordi, L Scharlemann, B Abdennadher, B Voelschow, R Twerenbold, N A Soerensen, J T NeumannAbstract
Background
Of all patients presenting with suspected myocardial infarction (MI), around 30% remain in the observe group after initial triage using rapid diagnostic algorithms. These patients are at high risk for mortality and major adverse cardiovascular events (MACE), making early risk stratification crucial. As a strong prognostic biomarker which has been thoroughly evaluated in different patient groups, high-sensitivity C-reactive protein (hs-CRP) may be useful in this setting.
Purpose
We aimed to evaluate the prognostic utility of hs-CRP in patients in the observe group, initially presenting with suspected MI, and comparing it to the established risk prediction model of GRACE Score.
Methods
We prospectively enrolled patients presenting to the emergency department with suspected MI and measurements of hs-cTn and hs-CRP were obtained and GRACE-Score (1.0) was calculated in each patient. Individuals with STEMI and patients with infectious cause of hs-CRP elevation were excluded from further analyses. Patients were categorized into three groups using the 0/1h hs-cTnI (Architect) algorithm: Rule-in, observe, or rule-out. For patients in the observe group, quintiles of hs-CRP were defined. All-cause mortality and MACE at 90 days, 1, 3, and 5 years were assessed. Event rates, Cox-regression, and c-indices were calculated for hs-CRP, GRACE-Score, and their combination, unadjusted as well as adjusted for age, sex and cardiovascular risk factors.
Results
Of 2,843 patients (median age 63 years; 36.1% female), 917 remained in the observe group. Median hs-CRP concentration in the observe group was 2.9mg/l. Patients with hs-CRP concentrations in the highest quintile (>12.57mg/l) had significantly higher 1-year (16% vs. 1.7%), 3-year- (29.4% vs. 4.2%) and 5-year mortality (41.2% vs. 6%) as well as 3-year- (39.5% vs. 24%) and 5-year MACE (47.5% vs. 28.2%) compared to the lowest quintile (≤0.95mg/l) (Figure 1). Hs-CRP resulted in a higher c-index compared to the GRACE-Score for 3-year (C-index 0.716 vs. 0.649) and 5-year mortality (C-index 0.705 vs. 0.649) (Table 1). Combined interpretation of hs-CRP and GRACE Score even further improved model performance (C-index5-year mortality 0.761). No relevant differences between hs-CRP quintiles and GRACE Score could be identified regarding MACE. Elevated hs-CRP concentrations in the highest quintile independently predicted 1-year- (adjusted HR1-year 6.72 [2.03, 22.23], p < 0001), 3-year- (adjusted HR3-years 5.37 [2.42, 11.93], p < 0.001) and 5-year mortality (adjusted HR5-years 6.44 [95%CI 3.07–13.50], p < 0.001) as well as 5-year MACE (adjusted HR 1.57 [95% CI 1.06, 2.31], p = 0.023) compared to the lowest quintile.
Conclusion
We identified hs-CRP as a strong and independent predictor of mortality and MACE in patients with suspected MI, classified into the observe group, further improving risk-prediction beyond established models.Figure 1:Kaplan-Meier curvesTable 1:Regression analyses and C-indices