Study objective: To assess the range of plasma C-reactive protein (CRP
) in patients presenting with community-acquired pneumonia and to comp
are the serial changes of this acute-phase protein with clinical outco
me. Design: Prospective hospital-based study, including separate retro
spective case series. Patients: Twenty-eight consecutive patients (mea
n age, 60 years) admitted to our hospital with community-acquired pneu
monia were studied. Serial daily plasma samples were taken and assayed
for CRP, tumor necrosis factor-alpha (TNF-alpha), and interleukin 6 (
IL-6), Clinical parameters, laboratory data, and response to treatment
were recorded. Four other patients considered to be antibiotic failur
es (three empyemas, one death) were studied separately. Results: Two p
atients died. Of those who survived, mean (+/-SD) CRP values for days
1, 2, 3, 4, and 5 were as follows: 136+/-43, 96+/-44, 53+/-36, 54+/-43
, and 44+/-31 mg/L. CRP levels on day 1 in patients who had received a
ntibiotics prior to hospital admission were significantly lower than t
hose who had not, 107+/-42 and 152+/-44 mg/L (p<0.05). CRP levels did
not correlate with other laboratory parameters or with recognized pred
ictors of mortality, A CRP value that continued to rise despite antibi
otic treatment was associated with infective complications or death, O
nly 52% of patients had detectable TNF-alpha and 24% detectable IL-6 a
t some point during their hospital stay. Conclusions: CRP is a sensiti
ve marker of pneumonia, A persistently high or rising CRP level sugges
ts antibiotic treatment failure or the development of an infective com
plication, These results suggest that CRP, rather than TNF-alpha or IL
-6, may have a role as a clinical marker in pneumonia.