Background: The Polymerase Chain Reaction (PCR) technique has some application limits therefore suggesting the evaluation of the sensibility (SE), specificity (SP), predictive positive value (PPV) and predictive negative value (NPV) of a PCR test detecting a target sequence of OspA gene of Borrelia burgdorferi sensu lato. This technique is currently used in the diagnosis of Lyme borreliosis in our laboratory.
Methods: Ninety-three patients, either examined in our outpatient clinic or coming from other clinical centres, were divided in two groups on the basis of a correlation between clinical picture and serologic tests (ELISA and Western Blot): 42 patients were true positives (tp) and 51 true negatives (tn). The samples (77 sera, 16 cerebrospinal fluids) have been tested with the above PCR technique. The results obtained were used to calculate the SE, SP, PPV and NPV. Subsequently, 13 urine samples of patients with negative serum PCR were examined with the same technique.
Results: A high SP and a low SE, high PPV and a considerably lower NPV have been reported. 50% of the PCR positive results, obtained with serum and cerebrospinal fluid samples corresponded to patients who were true positives at clinical examination but negatives at serologic tests. 62.5% of urine samples positive results belonged to tp patients who had negative serologic and serum PCR RESULTS.
Conclusions: The obtained results suggested a good reliability of positive results obtained with the PCR technique used in this study and allowed the false negatives of serologic tests to be detected, more specifically when urine samples were used.