Fa. Manian, A PROSPECTIVE-STUDY OF DAILY MEASUREMENT OF C-REACTIVE PROTEIN IN SERUM OF ADULTS WITH NEUTROPENIA, Clinical infectious diseases, 21(1), 1995, pp. 114-121
A total of 1,622 daily measurements of the level of C-reactive protein
(CRP) in the serum of 40 hospitalized patients with neutropenia were
made during 55 study periods from October 1990 through February 1993 (
mean, 29.5 measurements per period). Clinical events were categorized
into four groups: group I (bloodstream infection), group II (significa
nt bacterial or fungal infection without bloodstream infection), group
III (fever without an obvious source), and group IV (drug-related fev
er). There was a strong association between baseline elevation of the
CRP level (greater than or equal to 100 mg/L) and tumor-associated fev
er (P = .0005); the resolution of such fever coincided with a decrease
in the CRP level following chemotherapy. Levels of CRP increased by g
reater than or equal to 40 mg/L in 10 (31%) of 32 cases during the 48-
72 hours preceding the clinical diagnosis of a subsequently demonstrab
le infection-often pneumonia without bloodstream infection. CRP values
on day 2 (1 day after the diagnosis of a clinical event) were signifi
cantly higher for events in groups I and II than for those in group II
I (P < .01) but not those in group IV. With regard to significant infe
ctions, a day-2 CRP value of greater than or equal to 40 mg/L was 100%
sensitive and an increase in CRP level of greater than or equal to 50
mg/L from day 1 to day 2 had a positive predictive value of 95%. We c
onclude that serial measurements of serum CRP levels may be helpful in
determining the extent of evaluation required for a newly febrile neu
tropenic patient and possibly in accelerating the detection of an othe
rwise unsuspected infection.