Relationship of Serial High-Sensitivity C-Reactive Protein Changes to Long-term Clinical Outcomes in Stabilised Patients After Myocardial Infarction.

Abstract

Little is known about the association between serial high-sensitivity C-reactive protein (hsCRP) measurements and long-term outcomes in post-myocardial infarction (MI) patients. We aimed to investigate the usefulness of serial hsCRP measurements for risk stratification in stabilised post-MI patients after percutaneous coronary intervention (PCI). A total of 1018 patients who had hsCRP values at both baseline and 1 year after MI were included. High inflammatory status was defined as hsCRP > 2 mg/L. Patients were classified into 4 groups: persistently low, falling (first high then low hsCRP), rising (first low then high hsCRP), and persistently high hsCRP. The primary outcome was major adverse cardiac and cerebrovascular events (MACCE: a composite of all-cause of death, MI, and cerebrovascular accident) within 4 years after the second hsCRP measurement. At 1 year after MI, the numbers of patients in the persistently low, falling, rising, and persistently high hsCRP groups were 394 (38.7%), 358 (35.2%), 69 (6.8%), and 197 (19.4%), respectively. The incidence of MACCE was progressively elevated from the persistently low to the falling, rising, and persistently high hsCRP groups (4.8%, 8.1%, 10.1%, and 13.2%, respectively; P= 0.004). Persistently high hsCRP was an independent predictor of MACCE (adjusted hazard ratio 2.55; 95% confidence interval 1.35-4.81; P= 0.004) and provided incremental prognostic value beyond that of the baseline clinical risk model (net reclassification improvement= 0.397; integrated discrimination improvement= 0.025; all P < 0.001). Among stabilised post-MI patients who underwent PCI, persistently high hsCRP was frequently seen 1 year after MI and was strongly associated with long-term adverse clinical outcomes. Serial measurements of hsCRP during clinical follow-up after MI may help to identify patients at higher risk for mortality and morbidity.