Amsterdam, the Netherlands - C-reactive protein (CRP) measurements as well as changes in concentrations of the biomarker after a week of antimicrobial therapy in patients with native-valve endocarditis are strong predictors of bad outcomes, including death or major infection-related complications, concludes a prospective observational study in the February 11, 2008 issue of the Archives of Internal Medicine [1].

According to the authors, led by Dr Dominique W M Verhagen (University of Amsterdam, the Netherlands), serial CRP measurements could potentially guide treatment decisions, perhaps by identifying patients headed for complications early or allowing the intensity of care to be eased in those found to be low risk.

"For example, treatment could be sought in an outpatient or home setting, by oral instead of intravenous antibiotics, or with shorter duration of treatment," they write. "On the other hand, identifying patients with poor prognosis could lead to intensified antimicrobial treatment schedules or earlier timing of cardiac surgery."

The researchers prospectively followed serial CRP measurements in 123 adults with left-sided native-valve infective endocarditis for 12 weeks after the completion of antimicrobial therapy. About 40% of the group had at least one serious comorbidity, such as renal failure, diabetes with complications, autoimmune disorders, or cancer. Their baseline CRP averaged 89 mg/L and ranged from 20 mg/L to 324 mg/L.

After a week of treatment, the group found, the odds ratio for the primary outcome of death or serious infectious complications was significantly higher among patients in the highest CRP-level tertile compared with those in the lowest tertile. The significance increase dissipated by the second treatment week; in fact, at no other measured interval were CRP levels by tertile predictive of outcomes.

The risk of a poor outcome was also inversely and significantly related to percentage change in CRP level after one week of treatment (p=0.009).

Verhagen et al propose that serial CRP measurements need be obtained only out to day 7 of treatment. But, they caution, they were unable to determine a cut-point value for CRP that could be used to distinguish high-risk patients; "consequently, CRP level cannot be used as a sole prognostic factor, but it should be applied in combination with other clinical variables that have yet to be identified."

Dr Blase A Carabello (Baylor College of Medicine, Houston, TX), not a coauthor of the study, said it doesn't add much to what's already apparent and that he doubts that CRP is likely to be prognostically useful in infective endocarditis. Elevated CRP is simply a sign that the patient is sick and has ongoing inflammation, "so it makes intuitive sense that if CRP levels stay high, you're not getting better," he observed for heartwire. "But there are other ways doctors can determine whether patients are sick."

For example, he said, "For a century we've known that one of the first barometers of outcome is appetite. When a patient with endocarditis has a return of appetite, it's a good sign they're getting better. Some very simple things could have been forced into their multivariate analysis, for which CRP may or may not be any better."

CRP is more likely to be prognostically useful in clinical conditions that are not readily apparent and not obviously associated with active inflammation, Carabello said. Coronary disease is a good example—"that's where it has a role to play."