D

3 yr

2.8 (1.2-6.3)

CI, confidence interval; D, death; NSTEACS, non-ST-elevation ACSs; RA, refractory angina; UA, unstable angina; UR, urgent revascularization.

CI, confidence interval; D, death; NSTEACS, non-ST-elevation ACSs; RA, refractory angina; UA, unstable angina; UR, urgent revascularization.

CRP > 3 mg/L CRP < 3 mg/L Fig. 1. Incremental prognostic value of CRP in addition to troponin T for prediction of major adverse coronary events at 6 mo in patients with ACSs. The highest risk is confined to patients with elevations in both markers, whereas double-negative markers predict more favorable outcome. cTnT, cardiac troponin T; CV, cardiovascular. (Modified from ref. 69.)

CRP > 3 mg/L CRP < 3 mg/L Fig. 1. Incremental prognostic value of CRP in addition to troponin T for prediction of major adverse coronary events at 6 mo in patients with ACSs. The highest risk is confined to patients with elevations in both markers, whereas double-negative markers predict more favorable outcome. cTnT, cardiac troponin T; CV, cardiovascular. (Modified from ref. 69.)

CRP > 10 mg/L CRP <10 mg/L Fig. 2. Risk of cardiac death through 2 yr of follow-up after presentation with non-ST-elevation ACS stratified by CRP and cardiac troponin T (cTnT) at presentation. (Data from ref. 19.)

Clinical Cut Points for CRP in ACS

A variety ofcut points have been applied in clinical studies ofCRP in ACS, leaving some uncertainty with respect to the optimal decision limit. Although available data, especially those deriving from primary prevention studies, show no definite threshold below which CRP levels are not associated with cardiovascular risk, the relationship is graded with a very low risk reported for CRP levels <2.0 mg/L (17). Expert guidelines from the American Heart Association (AHA)/Centers for Disease Control (CDC) provided recommendations regarding the preferred cut points for primary prevention but did not give specific recommendations for risk stratification in ACS; rather, they suggested that a higher cutoff value, such as 10 mg/L, is appropriate for unstable patients (3). Liuzzo et al. (9) applied a cut point of 3 mg/L, based on the 90th percentile of a normal distribution. Dichotomized at this cut point, CRP identified patients at increased risk of in-hospital recurrent ischemic events. Other investigators have specifically evaluated the predictive value of CRP for death after ACS and found superior prognostic performance at a higher decision limit (e.g., 15 mg/L)

(14). This observation has been validated by subsequent studies, including a study by Biasucci et al. (15), who later confirmed that levels of CRP >15 mg/L distinguished those patients at higher risk of death. Among patients with unstable coronary disease, CRP > 2 mg/L identified a population at higher risk, with a particularly high mortality in patients with CRP > 10 mg/L (19). Similarly, Mueller et al. (16) found that a serum CRP >3 mg/L was associated with a higher risk of death in patients with non-ST-elevation ACS, with even higher mortality in those with a CRP >10 mg/L, despite aggressive management with early invasive therapy. With validation ofthis decision limit in two additional large studies in ACS (11, 17), a CRP concentration >10 mg/L appears to be the optimal cut point (when applied during the index hospitalization) for prediction of new AMI and death in secondary prevention. Nevertheless, the clinician should recognize the graded nature ofthe risk relationship to the probability of a recurrent event increasing significantly when CRP levels are above standard reference values for healthy people (3 mg/L) and even more so when CRP levels are higher than 10 mg/L. In addition, some experts propose that although a cut point of 10 mg/L may be optimal for prediction of mortality, it may not be best with respect to assessing the risk of recurrent ischemic events (8). Moreover, use of the standard cut points recommended by the AHA/CDC is likely preferable when CRP is measured during the convalescent phase (e.g., beyond 30 d) after ACS.

Was this article helpful?

0 0

Post a comment