BACKGROUND Our understanding of the complexity of the oral microbiota continues to improve as new technologies, such as the analysis of 16S rRNA bacterial genes, are utilized. Despite the difficult of cataloguing all microorganisms and determining their pathogenic potential, some species, mostly members of the resident oral microbiota, have been identified as likely periodontal pathogens. However, for microbial diagnosis to be of value, it needs to affect disease diagnosis and/or treatment planning as well as result in superior treatment outcomes and/or provide an economic benefit to the patient. RATIONALE The purposes of this systematic review were to determine if microbial identification influences periodontal patient management and whether treatment outcomes are better compared to patients whose treatment plans are developed without this information. FOCUSED QUESTION In patients with periodontal diseases, does microbial identification influence patient management compared to treatment prescribed without this information? SEARCH PROTOCOL The MEDLINE database was searched for clinical studies in English from 1991 through 2002 by 2 investigators. Hand searches were performed on the Journal of Clinical Periodontology, Journal of Periodontology, Journal of Periodontal Research, Oral Microbiology and Immunology, and Periodontology 2000. In addition, directors of diagnostic laboratories were contacted about unpublished data. SELECTION CRITERIA INCLUSION CRITERIA Articles in which bacterial identification influenced patient treatment were preferred as were those reporting longitudinal data demonstrating a direct relationship between the presence or absence of certain bacteria and subsequent alterations in clinical variables. Because of the limited number of studies, all articles, including case reports, were considered. EXCLUSION CRITERIA Review articles without original data were excluded, although references were examined for possible inclusion. Articles reporting data showing associations between certain microorganisms and disease or health that did not affect treatment were excluded. Clinical trials testing antibacterial agents for their ability to enhance mechanical debridement were not included since bacterial identification had little effect on drug selection or experimental group assignment. Articles dealing with implants rather than natural teeth were omitted. DATA ANALYSIS AND COLLECTION: The heterogeneity of the published data precludes any meaningful pooling of data or meta-analysis. The pertinent literature, including relevant variables of plaque, gingivitis, and bleeding on probing scores; probing depth; clinical attachment level; number of lost teeth; and microbial changes; and patient-centered outcomes including decrease in morbidity, reduced need for surgery, and duration and cost of treatment are summarized. MAIN RESULTS 1. There was a lack of articles with a high evidence rating; most pertinent articles were either case reports or case series without controls. 2. Because reports were heterogeneous regarding study design, patient selection, and data collection, meta-analysis was not feasible and results are summarized in tabular format. 3. This report is based on a total of 24 studies, representing a total patient population of approximately 835. 4. Thirteen studies reported on microbiological identification as an aid in treatment planning. 5. Eleven studies reported a differential clinical response depending on the detection or lack of detection of specific organisms. REVIEWERS' CONCLUSIONS 1. The published material suggests that microbiological monitoring may be useful in management of selected patients who do not respond to standard therapy. 2. Some practitioners consider microbial identification a valuable adjunct to managing patients with certain forms of periodontitis, although there is a lack of strong evidence to this effect. 3. Additional research is needed to address this issue.