Both patients had a control serum with a negative LFA test, 5 and 7days later, respectively. These two serum samples were considered as true false positives. Per-sample analysis gave sensitivity, specificity, positive and negative predictive
values calculated on the 205 patients studied, were all at 100% (2 positive/2 samples) for CSF, and at 100% (10 positive/10 samples), 98.9% (2 positive/262 samples) IC95 [97.3e100%], 75% and 100%, respectively for serum
samples.