Inflammatory markers, amino-terminal pro-brain natriuretic peptide, and mortality risk in dyspneic patients

Shafiq Rehman, Donald M. Lloyd-Jones, Abelardo Martinez-Rumayor, James L. Januzzi

Research output: Contribution to journalArticlepeer-review

8 Scopus citations


Dyspnea is a common emergency department (ED) complaint, and it may be associated with significant mortality risk. We studied 599 dyspneic subjects enrolled in an ED. At 1 year, the role of inflammatory markers (including C-reactive protein [CRP]) and amino-terminal pro-brain natriuretic peptide (NT-proBNP) as independent predictors of mortality was assessed. By 1 year, 91 subjects (15.2%) had died. Among patients who died, the median CRP concentration at admission was significantly higher than in survivors: 47.2 mg/L (449.5 nmol/L; interquartile range [IQR], 10.2-101.9 mg/L [97.1-970.5 nmol/L]) vs 7.25 mg/L (69.5 nmol/L; IQR, 2.2-29.6 mg/L [21.0-281.9 nmol/L]; P < .001). For 1-year mortality, CRP had an area under the receiver operating characteristic curve of 0.76 (95% confidence interval [CI], 0.69-0.80; P < .001). In multivariable analysis, a CRP concentration greater than 14 mg/L was a strong predictor of mortality at 1 year (hazard ratio, 2.47; 95% CI, 1.51-4.02; P < .001). In multivariable models, CRP and NT-proBNP demonstrated independent and additive prognostic value. Among dyspneic patients, CRP levels are significantly associated with mortality at 1 year and show additive value to natriuretic peptide testing for prognosis.

Original languageEnglish (US)
Pages (from-to)305-311
Number of pages7
JournalAmerican journal of clinical pathology
Issue number2
StatePublished - Aug 2008


  • Amino-terminal pro-brain natriuretic peptide
  • C-reactive protein
  • Dyspnea
  • Inflammatory markers
  • Mortality

ASJC Scopus subject areas

  • Pathology and Forensic Medicine


Dive into the research topics of 'Inflammatory markers, amino-terminal pro-brain natriuretic peptide, and mortality risk in dyspneic patients'. Together they form a unique fingerprint.

Cite this