A COMPARISON OF TWO COMMERCIAL STRIPS WITH PREDEFINED ANTIBIOTIC CONCENTRATION GRADIENTS FOR SUSCEPTIBILITY TESTING OF PERIODONTAL BACTERIAL PATHOGENS

Abstract

Objectives: Systemic antibiotics are generally recognized as providing a beneficial impact in treatment of both aggressive and chronic periodontitis. Since strains of periodontal pathogens among periodontitis patients may vary in their antibiotic drug resistance, the American Academy of Periodontology recommends antimicrobial susceptibility testing of suspected periodontal pathogens prior to administration of systemic periodontal antibiotic therapy, to reduce the risk of a treatment failure due to pathogen antibiotic resistance. E-test and MIC Test Strip assays are two in vitro antimicrobial susceptibility testing systems employing plastic- and paper-based, respectively, carriers loaded with predefined antibiotic gradients covering 15 two-fold dilutions. To date, no performance evaluations have been carried out comparing the Etest and MIC Test Strip assays in their ability to assess the in vitro antimicrobial susceptibility of periodontal bacterial pathogens. As a result, the purpose of this study was to compare the in vitro performance of E-test and MIC Test Strip assays in assessing minimal inhibitory concentration (MIC) values of four antibiotics frequently utilized in systemic periodontal antibiotic therapy against 11 fresh clinical subgingival isolates of the putative periodontal pathogen, Prevotella intermedia/ nigrescens, and to compare the distribution of P. intermedia/ nigrescens strains identified with interpretative criteria as "susceptible" and "resistant" to each of the four antibiotics using MIC values determined by the two antimicrobial susceptibility testing methods. Methods: Standardized cell suspensions, equivalent to a 2.0 McFarland turbidity standard, were prepared with 11 fresh clinical isolates of P. intermedia/nigrescens, each recovered from the subgingival microbiota of United States chronic periodontitis subjects, and plated onto to the surfaces of culture plates containing enriched Brucella blood agar. After drying, pairs of antibiotic-impregnated, quantitative, gradient diffusion strips from two manufacturers (E-test, bioMérieux, Durham, NC, USA, and MIC Test Strip, Liofilchem s.r.l., Roseto degli Abruzzi, Italy) for amoxicillin, clindamycin, metronidazole, and doxycycline were each placed apart from each other onto the inoculated enriched Brucella blood agar surfaces, so that an antibiotic test strip from each manufacturer was employed per plate against each P. intermedia/ nigrescens clinical isolate for antibiotic susceptibility testing. After 48-72 hours anaerobic jar incubation, individual MIC values for each antibiotic test strip against P. intermedia/nigrescens were read in μg/ml at the point where the edge of the bacterial inhibition ellipse intersected with the antibiotic test strip. MIC50, MIC90, and MIC range were calculated and compared for each of the test antibiotics, with essential agreement (EA) values determined per test antibiotic for the level of outcome agreement between two antimicrobial susceptibility testing methods. In addition, the identification of antibiotic "susceptible" and "resistant" strains among the P. intermedia/nigrescens clinical isolates was determined for each test antibiotic using MIC interpretative criteria from the MIC interpretative standards developed by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) for gram-negative anaerobic bacteria for amoxicillin, clindamycin, and metronidazole findings, and from the French Society of Microbiology breakpoint values for anaerobic disk diffusion testing for doxycycline data. Results: For amoxicillin, higher MIC50 and MIC90 values against the P. intermedia/ nigrescens strains were found with the MIC Test Strip assay than with E-test strips, resulting in a relatively low EA value of 45.5% between the two susceptibility testing methods. A higher percentage of amoxicillin "resistant" P. intermedia/nigrescens strains (72.7%) were identified by MIC Test Strips as compared to E-test strips (54.5%), although both methods found the same proportion of amoxicillin "susceptible" strains (27.3%). For clindamycin, both susceptibility testing methods provided identical MIC values (EA value = 100%), and exactly the same distributions of "susceptible" and "resistant" strains of P. intermedia/nigrescens. For metronidazole, only very poor agreement (EA value = 9.1%) was found between the two susceptibility testing methods, with MIC Test Strips exhibiting markedly higher MIC50 and MIC90 values against P. intermedia/nigrescens as compared to E-test strips. However, the distribution of "susceptible" and "resistant" P. intermedia/ nigrescens were identical between the two susceptibility testing methods. For doxycycline, relatively good agreement (EA value = 72.7%) was found in MIC concentrations between the two susceptibility testing methods, although generally lower MIC values were associated with MIC Test Strips. In addition, identical distributions of "susceptible" and "resistant" P. intermedia/nigrescens were provided by both susceptibility testing methods. Conclusions: Relative to MIC values measured against periodontal strains of P. intermedia/nigrescens, MIC Test Strips gave higher MIC values with amoxicillin and metronidazole, equal MIC values with clindamycin, and lower MIC values with doxycycline, as compared to MIC values measured with the E-test assay. Relative to the identification of antibiotic "susceptible" periodontal P. intermedia/ nigrescens strains, both susceptibility testing methods provided identical findings, suggesting that both methods appear to be interchangeable for clinical decision making in regard to identification of antibiotic-sensitive strains of periodontal P. intermedia/nigrescens. However, for epidemiologic surveillance of drug susceptibility trends, where exact MIC values are important to track over time, the relatively higher proportion of non-exact MIC differences between the two susceptibility testing methods argues against using them interchangeably. Instead, one or the other method should be used consistently for such studies. Further comparative studies of the E-test and MIC Test Strip assays are indicated using other periodontopathic bacterial species besides P. intermedia/ nigrescens, and to assess the reproducibility of MIC values provided by both in vitro susceptibility testing methods over time.