Use of C-reactive protein in differentiation between acute bacterial and viral otitis media.

ABSTRACT

OBJECTIVES: The objectives of this investigation were (1) to determine degree of elevation of serum C-reactive protein (CRP) in uncomplicated acute otitis media (AOM); (2) to compare serum CRP levels in bacterial and viral otitis media; and (3) to determine whether a single serum CRP level, obtained early in the course of AOM, could be used to differentiate between viral and bacterial otitis media. DESIGN AND METHODS: Sera were obtained from otherwise healthy infants and children with AOM who were 3 months to 7 years of age between 1989 and 1991. Tympanocentesis, bacterial and viral studies of the middle ear fluids, virologic studies of nasal wash specimens, measurements of serum antibody titers to respiratory viruses, blood counts, and quantitation of serum CRP concentrations were performed. After the initial tympanocentesis, an oral antibiotic was given for the next 10 days. The patients were clinically reevaluated over next 4 weeks. OUTCOME MEASURES: Serum CRP concentrations were compared among subjects with AOM who were divided into four groups based on the results of bacteriologic and virologic studies group I, Bacterial infection (n = 82); group II, bacterial and viral infections (n = 69); group III, viral infection (n = 12); and group IV, no identifiable pathogen (n = 22). RESULTS: There was no statistical difference in serum CRP values among the four groups. The ranges of CRP were less than 0.6 to 22.8, less than 0.6 to 17.8, less than 0.6 to 2.0, and less than 0.6 to 6.8 mg/dL in groups I through IV, respectively. However, when CRP values in bacteria-positive cases were compared with CRP concentrations in bacteria-negative cases (1.58 +/- 3.16 vs 0.64 +/- 1.24 mg/dL), the difference was statistically significant. Furthermore, a significantly higher proportion of bacteria-positive cases had serum CRP concentrations greater than 2 mg/dL, compared with those in bacteria-negative cases. There was no correlation between initial CRP values and clinical findings and/or the clearance of bacteria from the middle ear. After 10 days of antibiotic treatment, CRP values returned to normal (< 0.6 mg/dL) in all cases. CONCLUSION: In AOM, the range of serum CRP varied from less than 0.6 to 22.8 mg/dL. High CRP values (> 2.0 mg/dL) were associated with 22% of cases of bacterial AOM but only with 6% of nonbacterial AOM. High levels of serum CRP were found to be very specific in detecting bacterial AOM, and no cases of viral AOM without a concurrent bacterial infection were found to exhibit high serum levels of CRP.