Comparison of molecular and microscopic techniques for detection of Treponema pallidum in genital ulcers.

We compared the ability of direct immunofluorescent staining (DFA) and the PCR to detect Treponema pallidum in specimens from patients with genital ulcer disease. Touch preparations from 156 patients with genital lesions were fixed in acetone and stained with a fluorescein-labeled monoclonal antibody specific for the 37-kDa antigen of T. pallidum. After microscopic examination, the smear was removed from the slide with a swab. DNA was extracted with phenol-chloroform and precipitated with isopropanol. Ten microliters of the extracted DNA was amplified by PCR using primers for the gene encoding the 47-kDa protein of T. pallidum and hybridized to an internal probe. Twenty-two of 156 specimens were positive for T. pallidum by DFA and PCR, while 127 were negative by both methods, yielding a concordance of 95.5% (kappa = 0.84). Four specimens were positive by PCR and negative by DFA, while three specimens were negative by PCR and positive by DFA. The DFA-negative, PCR-positive specimens may have resulted from the presence of large numbers of leukocytes on the slides, obscuring visualization of treponemes. The DFA-positive, PCR-negative results were not due to inhibition of the PCR since purified T. pallidum DNA was amplified when added to aliquots of these specimens. Negative results in these specimens were most likely due to inefficient recovery of their DNA. These data suggest that DFA and PCR are equivalent methods for detection of T. pallidum on touch preparations of genital lesions. Further refinements of the PCR assay are necessary for it to significantly improve the detection of T. pallidum in genital lesions.