J Clin Periodontol. 2005 Jul;32(7):778-83. doi: 10.1111/j.1600-051X.2005.00740.x.
Comparison of culture and real-time PCR for detection and quantification of five putative periodontopathogenic bacteria in subgingival plaque samples.
Journal of clinical periodontology
P-M Jervøe-Storm, M Koltzscher, W Falk, A Dörfler, S Jepsen
- Department of Periodontology, Operative and Preventive Dentistry, University of Bonn, 53111 Bonn, Germany. [email protected]
OBJECTIVES: Bacterial cultivation is a well-established method for analyzing plaque samples. Real-time polymerase chain reaction (PCR) is a novel rapid method for the identification and quantification of periodontopathogenic bacteria that has been recently introduced. In this study, we compared real-time PCR with conventional anaerobic cultivation.
METHOD: A total of 78 subgingival plaque samples were harvested from pockets > or =5 mm in 22 patients with advanced chronic periodontitis and immediately transferred into transport medium. Aliquots were evaluated with species-specific probes by real-time PCR (meridol Perio Diagnostics, GABA) and anaerobic bacteria culture on selective media for the detection of Actinobacillus actinomycetemcomitans, Fusobacterium nucleatum, Porphyromonas gingivalis, Prevotella intermedia and Tannerella forsythensis. The analysis was performed by two separate, blinded examiners.
RESULTS: When real-time PCR was compared with the culture method (golden standard) for the detection of putative periodontopathogenic bacteria, the sensitivity and specificity for A. actinomycetemcomitans were 67% and 100%, respectively (kappa: 0.79); for F. nucleatum 73% and 53%, respectively (kappa: 0.21); for P. gingivalis 94% and 84%, respectively (kappa: 0.77); for P. intermedia 33% and 94%, respectively (kappa: 0.26) and for T. forsythensis 92% and 56%, respectively (kappa: 0.51). Spearman's correlation coefficients for the quantitative results of both methods were 0.82 for A. actinomycetemcomitans, 0.33 for F. nucleatum, 0.83 for P. gingivalis, 0.38 for P. intermedia and 0.67 for T. forsythensis.
CONCLUSION: Overall, the agreement between both test methods was excellent for A. actinomycetemcomitans and P. gingivalis, fair for T. forsythensis and poor for F. nucleatum and P. intermedia. The discrepancies in the results may be explained by the inability of cultivation methods to distinguish between close related taxa, and the problems of keeping periopathogenic bacteria viable, which is required for standard cultivation.