A 58-year-old male who had been diagnosed as hepatic cirrhosis four ye
ars previously was admitted to our hospital because his serum C-reacti
ve protein (CRP) level had gradually risen, reaching 139 mg/dI. No inf
lammation findings were observed subjectively or objectively. Close ex
amination revealed his CRP reaction to be false positive. His serum CR
P showed positive only in a latex agglutination method using goat anti
-CRP IgG. This false-positive reaction was thought to be owing to the
abnormally glycosylated IgM, which has an affinity for the goat serum
IgG.