High-sensitivity C-reactive protein in acute heart failure: insights from the ASCEND-HF trial.

BACKGROUND

Inflammation is associated with progression of chronic heart failure (HF). Few data exist on high-sensitivity C-reactive protein (hsCRP) levels and their importance in acute HF.

METHODS AND RESULTS

In this biomarker substudy of the ASCEND-HF trial, we measured hsCRP levels at admission (n = 794), 48-72 hours (n = 677), and 30 days (n = 581) and evaluated their association with outcomes. Levels of hsCRP were considerably elevated at admission (median 12.6 mg/L, interquartile range [IQR] 5.23-30.5) and 48-72 hours (median 11.0 mg/L, IQR 4.87-29.9) and declined only at 30 days (median 4.7 mg/L, IQR 1.83-13.1). Admission hsCRP was not associated with dyspnea improvement at 6 hours (74.1%) and 24 hours (86.2%), in-hospital death or worsening HF (n = 37; 4.7%), 30-day mortality or HF readmission (death: n = 25 [3.2%]; combined death and HF readmission: n = 95 [12.0%]), or 180-day mortality (n = 96; 12.1%). Hospital stay (median 5 days, IQR 3-7) was longer among patients with higher admission hsCRP levels (0.57 days per log2-hsCRP in adjusted models; 95% confidence interval [CI] 0.33-0.81; P < .001). Levels of hsCRP at 48-72 hours did not predict 30-day mortality or HF readmission and were only marginally associated with 180-day mortality. However, higher hsCRP at 30 days among survivors was associated with higher 180-day mortality in models including admission hsCRP (adjusted hazard ratio [HR] per log2-hsCRP: 1.23; 95% CI 1.04-1.45; P = .016). Patients with an hsCRP increase at day 30, defined as >10% increase over baseline value, had higher 180-day mortality risk compared with those with unchanged or decreased 30-day hsCRP (HR 2.29, 95% CI 1.16-4.52; P = .017).

CONCLUSIONS

Levels of hsCRP are elevated among patients with acute HF. Increasing levels at 30 days after discharge are associated with higher 180-day mortality.