Effects of Statin Therapy According to Plasma High-Sensitivity C-Reactive Protein Concentration in the Controlled Rosuvastatin Multinational Trial in Heart Failure (CORONA)

Abstract

We examined whether the antiinflammatory action of statins may be of benefit in heart failure, a state characterized by inflammation in which low cholesterol is associated with worse outcomes. We compared 10 mg rosuvastatin daily with placebo in patients with ischemic systolic heart failure according to baseline high sensitivity-C reactive protein (hs-CRP) <2.0 mg/L (placebo, n=779; rosuvastatin, n=777) or > or = 2.0 mg/L (placebo, n=1694; rosuvastatin, n=1711). The primary outcome was cardiovascular death, myocardial infarction, or stroke. Baseline low-density lipoprotein was the same, and rosuvastatin reduced low-density lipoprotein by 47% in both hs-CRP groups. Median hs-CRP was 1.10 mg/L in the lower and 5.60 mg/L in the higher hs-CRP group, with higher hs-CRP associated with worse outcomes. The change in hs-CRP with rosuvastatin from baseline to 3 months was -6% in the low hs-CRP group (27% with placebo) and -33.3% in the high hs-CRP group (-11.1% with placebo). In the high hs-CRP group, 548 placebo-treated (14.0 per 100 patient-years of follow-up) and 498 rosuvastatin-treated (12.2 per 100 patient-years of follow-up) patients had a primary end point (hazard ratio of placebo to rosuvastatin, 0.87; 95% confidence interval, 0.77 to 0.98; P=0.024). In the low hs-CRP group, 175 placebo-treated (8.9 per 100 patient-years of follow-up) and 188 rosuvastatin-treated (9.8 per 100 patient-years of follow-up) patients experienced this outcome (hazard ratio, 1.09; 95% confidence interval, 0.89 to 1.34; P>0.2; P for interaction=0.062). The numbers of deaths were as follows: 581 placebo-treated (14.1 per 100 patient-years of follow-up) and 532 rosuvastatin-treated (12.6 per 100 patient-years) patients in the high hs-CRP group (hazard ratio, 0.89; 95% confidence interval, 0.79 to 1.00; P=0.050) and 170 placebo-treated (8.3 per 100 patient-years) and 192 rosuvastatin-treated (9.7 per 100 patient-years) patients in the low hs-CRP group (hazard ratio, 1.17; 95% confidence interval, 0.95 to 1.43; P=0.14; P for interaction=0.026). In this retrospective hypothesis-generating study, we found a significant interaction between hs-CRP and the effect of rosuvastatin for most end points whereby rosuvastatin treatment was associated with better outcomes in patients with hs-CRP > or = 2.0 mg/L. URL: http://www.clinicaltrials.gov. Unique identifier: NCT00206310.