Evaluation of Accuracy Limits of Countable Colony-forming Units on Agar Plates

Abstract
Objectives: Accurate colony counts are an essential component of many microbiology research projects and clinical laboratory processes. The suggested range of accuracy of CFUs extends from 30 to 300 (Standard Methods for the Examination of Water and Wastewater). This recommendation dates to 1907, and fails to adequately address the numerous sources of inter- and intravariability. Without more detailed analysis it is difficult to estimate the sample size and number of replicates necessary to ensure accurate results. The purpose of this study was to determine the validity of accuracy limits for quantifying CFUs on agar plates.
Methods: Escherichia coli (ATCC 25922) and Staphylococcus epidermidis (ATCC 12228) were used to prepare series of four organism densities ranging from approximately 40-500 CFU, on three different days. On each day, each of the 4 densities for both organisms was plated on SBA and viable organisms were counted following incubation. An average of the margins of error obtained over the 3 days of testing was used to determine the reproducibility of agar plate counts, and to estimate the optimum number of replicate plates (sample size) required for each organism at each concentration.
Results: Margins of error for both organisms were greatest with suspensions yielding approximately 40 CFU, and lowest for suspensions yielding 300 and 500 CFU. Nine replicate plates were required for a suspension of S. epidermidis yielding 40 CFU to achieve the same margin of error as obtained with 3 replicate plates at concentrations yielding 100-300 CFU. Seven replicates plates were required for a suspension of E. coli yielding 40 and 100 CFU to achieve similar margins of error to those obtained with 4 replicate plates at concentrations yielding 300 CFU, and 3 replicate plates at concentrations yielding 500 CFU.
Discussion: We found that the greater the concentration (300 and 500 CFU), the fewer replicate plates necessary to reliably estimate organism concentrations. The lower the organism density (40 CFU), the more plates necessary to reliably estimate CFUs. Contrary to the recommendations described in Standard Methods for the Examination of Water and Wastewater, CFU of 500 were reliably reproducible. For greatest accuracy, experiments should be conducted so as to assure that colony counts are in the range of 300-500

Readers found these studies helpful