Of a total of 902 consecutive patients admitted to our department's coronary care unit between July 1, 2009 and June 30, 2011 with a diagnosis of ACS, we prospectively studied 682 patients. The time between the onset of ACS symptoms and the time of blood sampling to assess biomarkers was <24 hours. The remaining 220 patients were excluded for two main reasons: more than 24 hours between symptom onset and the time of blood sampling; or if at least one of the four laboratory predictors had not been assessed on admission (clotted blood sample or failure to request the laboratory study). The excluded patients were slightly older than the study population, but did not differ significantly in other baseline characteristics (Table 1).

The study was approved by the ethics committee and all patients provided written informed consent.

The patients included in the study presented a history of chest pain at rest or symptoms suggestive of ACS, the most recent episode being up to 24 hours before admission. Symptoms could be associated with persistent or transient ST-segment elevation, ST-segment depression, or T-wave inversion on the ECG, or elevated biomarkers of cardiac injury. The biomarkers used were cardiac troponin I (cTnI) and the MB fraction of creatinine kinase (CK-MB) and the 99th percentile of a reference population was employed as the diagnostic threshold. Patients with unstable angina and normal admission ECG were also included (n=24). Left ventricular (LV) ejection fraction was assessed by transthoracic echocardiography 1–2 days after admission. In patients with ST-segment elevation ACS (STE-ACS) echocardiography was performed after cardiac catheterization as long as a mechanical complication was not suspected. In patients with non-ST-segment elevation ACS (NSTE-ACS) echocardiography was performed before cardiac catheterization and usually within 24 hours of admission.

The four laboratory predictors were considered to be elevated when CysC >0.95 mg/l (normal 0.53–0.95); CRP >5.0 mg/l (<5); NT-proBNP >125 pg/ml (<125) and RDW >14% (11.6–14.0), which were the upper limit values of a normal population. Glomerular filtration rate (GFR) was estimated by the Cockcroft-Gault (CG) formula and renal dysfunction was defined as GFR <60 ml/min/1.73 m2.4

The study's primary outcome was cumulative all-cause mortality at six months. Follow-up and confirmation of adverse events were by telephone contact and by consultation of the hospital's databases and records. Six-month follow-up was achieved in 633 patients (92.8% of the study population). The majority of patients lost for follow-up had moved away or were from other areas of reference and so contact was not possible; their baseline characteristics did not differ significantly from those of the study population (Table 2).

Patients were divided into five groups according to the number of elevated laboratory predictors on admission: G0 (n=24), in whom none of the four biomarkers was elevated at admission; G1 (n=158), in whom one biomarker was elevated at admission; G2 (n=281), in whom two biomarkers were elevated at admission; G3 (n=156), in whom three biomarkers were elevated at admission; and G4 (n=63), in whom all biomarkers were elevated at admission. The groups were compared in terms of baseline characteristics. Categorical variables were expressed as percentages and continuous variables as means ± standard deviation or medians with ranges. Between-group differences were compared using the Student's t test or the Mann-Whitney U test for continuous variables and the chi-square test or Fisher's exact test for categorical variables, as appropriate. Multivariate linear regression analysis was performed to determine which of the following variables were associated with a higher number of laboratory predictors elevated on admission: age and gender; history of diabetes, hypertension, smoking, myocardial infarction (MI), renal failure, stroke/transient ischemic attack, atrial fibrillation, or coronary artery bypass grafting (CABG); previous medication with aspirin, thienopyridines, beta-blockers, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers and statins; Killip class, heart rate, GFR, hemoglobin and type of ACS on admission.

To assess the association between a higher number of biomarkers elevated on admission and 6-month mortality, multivariate logistic regression analysis was performed, the models being constructed with the number of predictors elevated on admission as a categorical variable and including four other covariables associated with 6-month mortality, given the small number of events in the study. The threshold for retaining a variable in the final model was set at p<0.1. The variables included were age, Killip class on admission, LV systolic dysfunction, coronary revascularization and the number of elevated biomarkers on admission. To assess whether the study score (number of laboratory predictors elevated on admission) was a predictor of mortality independent of the GRACE score,5 the GRACE score was included in a second multivariate logistic regression model which included the multimarker score, LV systolic dysfunction, coronary revascularization and GRACE score. Age and Killip class were not included in this second model because they are already included in the GRACE score.

Kaplan-Meier survival curves were constructed to compare event-free survival at six months according to the number of elevated laboratory predictors on admission, using the log-rank test. A value of p<0.05 was considered statistically significant. All statistical tests were two-sided. The statistical software used was Statistical Package for the Social Sciences (SPSS) version 17.0 (SPSS Inc., Chicago, Ill).

The hierarchy of the biomarkers in terms of predictive power for 6-month mortality was assessed by comparing the areas under the receiver-operating characteristic curve (AUCs). This calculation and comparison was performed with MedCalc version 12.2.1.0 (MedCalc Software, Mariakerke, Belgium).

Given the known value of the GRACE score in the risk stratification of patients with ACS, we conducted a stratified analysis of 6-month mortality according to the number of elevated biomarkers in patients with low, intermediate and high risk according to their GRACE score (in both NSTE-ACS and STE-ACS).

Of the 682 patients enrolled in the study, 356 (52.2%) had ST-segment elevation MI, 302 (44.3%) had non-ST-segment elevation MI, and 24 (3.5%) had a final diagnosis of unstable angina. On admission, 24 (3.5%) patients were in group G0, 158 (23.2%) in G1, 281 (41.2%) in G2, 156 (22.9%) in G3 and 63 (9.2%) in G4. The demographic characteristics of the study population according to the number of elevated biomarkers on admission are shown in Table 3. Patients with a higher number of elevated biomarkers on admission were older (p for trend <0.001), more likely to be female (p<0.001), and had a higher prevalence of diabetes (p=0.001) and hypertension (p<0.001). These patients more often had a history of MI (p<0.001), CABG (p=0.021) and stroke (p<0.001), and on admission more often presented with NSTE-ACS (p<0.001), Killip class >I (p<0.001) and lower mean hemoglobin (p<0.001).

Table 4 shows patients’ in-hospital treatment and angiographic and echocardiographic characteristics according to the number of laboratory predictors elevated on admission. Three-vessel and left main disease (p<0.001) and LV dysfunction (p<0.001) were more frequent in patients with a higher number of laboratory predictors elevated on admission, who were also less often revascularized (p<0.001). Furthermore, percutaneous coronary intervention (PCI) (p<0.0001) and thienopyridines (p<0.0001) and beta-blockers were used less often in these patients, whereas diuretics (p=0.003), aldosterone antagonists (p<0.0001) and intravenous inotropics (p<0.0001) were used more frequently.

Sixteen (2.4%) patients died during initial hospitalization and 23 (3.4%) died within six months of hospital discharge. The primary outcome (all-cause mortality) was recorded in 39 (5.7%) patients during 6-month follow-up. Patients with a higher number of laboratory predictors elevated on admission had increased in-hospital (p<0.0001) and 6-month mortality (p<0.0001). Table 4 lists the in-hospital events and Figure 1 shows Kaplan-Meier curves for the primary outcome during follow-up according to the number of laboratory predictors elevated on admission.

Figure 1.

Kaplan-Meier curves for 6-month mortality according to the number of biomarkers elevated on admission.

(0.08MB).

In Table 5, the results of multivariate logistic regression analysis are presented for 6-month mortality according to the number of laboratory predictors elevated on admission. After adjustment for the previously mentioned possible confounding factors, patients with four laboratory predictors elevated on admission had a higher incidence of 6-month mortality compared to patients with none or one, with an odds ratio (OR) of 10.66 (95% confidence interval [CI], 2.11–53.93; p=0.004). When the GRACE risk score was included in the second model (Table 6), the number of biomarkers elevated on admission was still an independent predictor of 6-month mortality; patients with four laboratory predictors elevated on admission had almost 14 times greater risk of 6-month mortality than patients with none or one.

The AUCs for 6-month mortality of CysC, NT-proBNP, CRP and RDW (Figure 2) of all the patients included in the study were, respectively, 0.84 (p<0.0001), 0.83 (p<0.0001), 0.75 (p<0.0001) and 0.64 (p=0.0036). There was no statistically significant difference between the predictive power of CysC and NT-proBNP (differences in AUC); however, CysC and NT-proBNP had stronger predictive power than CRP and RDW, and CRP had stronger predictive power than RDW (Figure 2).

AUCs of the four biomarkers for 6-month mortality. AUC: area under the receiver-operating characteristic curve for 6-month mortality; CRP: C-reactive protein; CysC: cystatin C; NT-proBNP: N-terminal pro-brain natriuretic peptide; RDW: red blood cell distribution width.'> AUCs of the four biomarkers for 6-month mortality. AUC: area under the receiver-operating characteristic curve for 6-month mortality; CRP: C-reactive protein; CysC: cystatin C; NT-proBNP: N-terminal pro-brain natriuretic peptide; RDW: red blood cell distribution width.' title='Receiver-operating curves and AUCs of the four biomarkers for 6-month mortality. AUC: area under the receiver-operating characteristic curve for 6-month mortality; CRP: C-reactive protein; CysC: cystatin C; NT-proBNP: N-terminal pro-brain natriuretic peptide; RDW: red blood cell distribution width.'/>

Figure 2.

Receiver-operating curves and AUCs of the four biomarkers for 6-month mortality. AUC: area under the receiver-operating characteristic curve for 6-month mortality; CRP: C-reactive protein; CysC: cystatin C; NT-proBNP: N-terminal pro-brain natriuretic peptide; RDW: red blood cell distribution width.

(0.28MB).

Among patients with a high GRACE risk score, our score was able to further stratify risk, in both patients with STE-ACS (Figure 3) and patients with NSTE-ACS (Figure 4). Patients with low and intermediate risk had a very low mortality in our population. In patients with NSTE-ACS and intermediate GRACE risk, those with four elevated biomarkers on admission had high mortality; in patients admitted for STE-ACS, all patients with four biomarkers elevated on admission were in the high risk group defined by the GRACE score (Figure 3), while in patients admitted for NSTE-ACS and with low risk according to the GRACE score, there were no patients with four elevated biomarkers.

Figure 3.

Six-month mortality in ST-elevation acute coronary syndromes according to GRACE risk score and the number of biomarkers elevated on admission. *No patients with four elevated biomarkers in these risk categories.

(0.1MB).

Figure 4.

Six-month mortality in non-ST-elevation acute coronary syndromes according to GRACE risk score and number of biomarkers elevated on admission. *No patients with four elevated biomarkers in this risk category.

(0.1MB).

The authors declare that no experiments were performed on humans or animals for this study.

The authors declare that they have followed the protocols of their work center on the publication of patient data and that all the patients included in the study received sufficient information and gave their written informed consent to participate in the study.

The authors declare that no patient data appear in this article.