Aims-To evaluate a new enzyme immunoassay (EIA) method for detection o
f Clostridium difficile toxin by comparing it to cytotoxicity assay. T
o investigate the nature of false negative and false positive EIA resu
lts by evaluating clinical and therapeutic parameters. Methods-737 con
secutive diarrhoeal specimens collected fi om patients clinically susp
ected of having C difficile colitis were tested for the presence of C
difficile toxin by EIA for toxin A and by cytotoxicity assay. Clinical
data were evaluated in all cases positive by either method. Results-W
ith the cytotoxicity assay as a gold standard, the specificity of EIA
for toxin detection was 99.3% and the sensitivity was 62.2%. No false
negative ELA specimens were obtained from patients already being treat
ed for C difficile colitis. Among patients with cytotoxicity positive
specimens, those with EIA positive samples had no clinical features di
stinguishing them from patients with EIA negative samples. Conclusions
-Although specific, the new EIA method directed against toxin A lacks
sensitivity compared to cytotoxicity, False negative EIA tests are not
associated with concurrent treatment for C difficile colitis nor with
any specific clinical features examined in our study.