Comparison of various detection methods for periodontopathic bacteria: can culture be considered the primary reference standard?

The development of diagnostic tests for a periodontal infection raises the issue as to what the appropriate reference standard, or "gold standard," should be for the evaluation of a new test. The present research was initiated to compare the ability of several detection methods, i.e., a serial dilution anaerobic culture and/or microscopic procedure, a DNA probe procedure, and immunological reagents using both an enzyme-linked immunosorbent assay and an indirect immunofluorescence assay to detect Treponema denticola, Porphyromonas gingivalis, Bacteroides forsythus, and Actinobacillus actinomycetemcomitans in subgingival plaque samples taken from 204 periodontally diseased tooth sites. The prevalence of the four monitored species varied as a function of both the species and the detection method. Spirochetes were present in 99% of the plaques, whereas A. actinomycetemcomitans was detected at the lowest frequency. The culture method yielded the lowest prevalence values for the three cultivable species. This raised the question as to which results, those obtained by culture or those obtained by the DNA probes and the immunological reagents, were the most reliable. This issue was addressed by looking at the prevalence profile of the monitored organisms, as determined by all the detection methods. If the species was detected by three or four of the detection methods, then it was considered present, whereas if it was absent by three or four of the detection methods, then it was considered absent. This approach showed the DNA probes and immunological reagents to be significantly superior (P less than 0.05) to the culture approach for the detection of P. gingivalis, A. actinomycetemcomitans, and B. forsythus and to be comparable to the microscopic approach in the detection of T. denticola.