Current issues of ACP Journal Club are published in Annals of Internal Medicine


High levels of C-reactive protein and positive troponin T test results predicted 14-day mortality in unstable angina or non-Q-wave MI

ACP J Club. 1998 Nov-Dec; 129:73. doi:10.7326/ACPJC-1998-129-3-073

Source Citation

Morrow DA, Rifai N, Antman EM,et al. C-reactive protein is a potent predictor of mortality independently of and in combination with troponin T in acute coronary syndromes: a TIMI 11A substudy. J Am Coll Cardiol. 1998 Jun;31:1460-5.



Can C-reactive protein (CRP) levels or cardiac-specific troponin T test results, alone or together, predict 14-day mortality in patients with unstable angina or non-Q-wave myocardial infarction (MI)?


Cohort study using data from the Thrombolysis in Myocardial Infarction (TIMI) 11A trial.


45 clinical centers.


437 patients with confirmed unstable angina or non-Q-wave MI. Exclusion criteria were thrombolysis in the previous 24 hours, coronary artery bypass surgery within 2 months, other serious illness, contraindications to or need for continuous anticoagulants, history of heparin-induced thrombocytopenia, or missing CRP levels or troponin T test results.

Assessment of risk factors

CRP and troponin T levels were ascertained in a blinded manner using stored frozen serum collected after 6 hours of symptom onset. Patients were categorized as having a CRP level of ≥ 1.55 mg/dL or < 1.55 mg/dL. Early positive rapid troponin T test results could be read bedside within 10 minutes rather than the standard 20 minutes.

Main outcome measure

All-cause mortality at 14 days.

Main results

7 patients died during follow-up. 346 patients (80%) had negative troponin T test results. 43 of the 91 positive tests were classified as early. 329 patients had a CRP level < 1.55 mg/dL. The mean CRP level in patients who died was 7.21 mg/dL and 1.29 mg/dL in patients who did not (P = 0.004). The various combinations in the Table show that the mortality rate was lowest in patients who had low levels of CRP and negative troponin T test results (0.4%) and was highest inpatients who had high levels of CRP and early positive test results (9.1%) (P < 0.001 for trend).


For patients with unstable angina or non-Q-wave myocardial infarction, mortality was highest in patients with high levels of C-reactive protein and early positive cardiac-specific troponin T test results.

Sources of funding: Behring Diagnostic; Boehringer Mannheim Corporation; Rhône-Poulenc Rorer.

For correspondence: Dr. D.A. Morrow, Cardiovascular Division, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115, USA.

Table. Mortality rates at 14 days for patients with unstable angina or non-Q-wave myocardial infarction for levels of C-reactive protein (CRP) and cardiac-specific troponin T test results

CRP level Combiner Tropinin T result Number of patients Mortality rate
< 1.55 mg/dL and Negative 227 0.4%
≥ 1.55 mg/dL or Positive 160 3.8%
≥ 1.55 mg/dL or Early positive 129 4.7%
≥ 1.55 mg/dL and Positive 39 5.1%
≥ 1.55 mg/dL and Early positive 22 9.1%


Providing accurate prognostic information is an important goal for clinicians and researchers involved in the care of patients who are critically ill. General outcome prediction scores like APACHE (Acute Physiology and Chronic Health Evaluation) and SAPS (Simplified Acute Physiology Score) have been developed and validated to do this (1). However, the predictive accuracy of these methods is usually in the range of 75% to 85% and they have not been validated for use in such specific populations as transplant, trauma, and burn patients. These issues have contributed to the development of disease-specific outcome models that attempt to improve the predictive accuracy of thegeneral scoring systems (1).

Serologic markers of disease activity that have been studied as predictors of patient outcomes include measurement of levels of circulating hormones, peptides, and products of specific disease states (e.g., tumor markers, endotoxin, and cytokines) (2). The main goal of these prediction methods is to provide an objective quantitative estimate of a specific clinical outcome, such as mortality, length of stay, or costs. To date, these prediction instruments have found their most productive application in clinical trials and quality improvement efforts to ensure the comparability of patient groups (1).

Morrow and colleagues have identified a strong relation among troponin test results, CRP levels, and 14-day mortality and unstable angina or non-Q-wave MI. Clinicians and researchers caring for these patients may be able to use these markers to improve medical management and the conduct of clinical trials. For example, it may be more reasonable to subject patients with a relatively high risk for mortality (early positive troponin T test results and CRP level ≥ 1.55 mg/dL) to a more aggressive early intervention than patients at low risk for mortality (negative test results and CRP level < 1.55 mg/dL). If validated, these markers could also be used to stratify patients who enter future clinical trials.

Marin H. Kollef, MD
Washington University School of MedicineSt. Louis, Missouri, USA


1. Kollef MH, Schuster DP. Predicting intensive care unit outcome with scoring systems. Underlying concepts and principles. Crit Care Clin. 1994;10:1-18.

2. Pepin CJ. Prognostic markers in thrombolytic therapy: looking beyond mortality. Am J Cardiol. 1996;78:24-7.