Comparison of commercially available Escherichia coli enumeration tests: implications for attaining water quality standards.

Many states are replacing microbiological water quality standards based on "fecal" or thermotolerant coliforms (ThCs) with new standards that employ Escherichia coli as the indicator organism. Implicit in these new standards are assumptions about the equivalence of E. coli enumeration tests and the E. coli levels that will provide protection equivalent to former ThC standards. To investigate these assumptions, E. coli levels in split samples (tests conducted on portions of the same grab sample) collected from small urban streams were determined using enzyme-specific media (Colilert, m-ColiBlue24, and nutrient agar with MUG) and compared to levels determined with conventional culture media (m-FC and m-TEC). Although levels observed with all tests were highly correlated, significantly fewer E. coli were enumerated with m-TEC than with enzyme-specific media (paired Student's t-test, alpha>99%). In addition, E. coli were found to comprise a larger fraction (84-104%, depending on the test) of the total presumptive ThC concentration than that suggested by the United States Environmental Protection Agency (63%). Both of these observations (1) the improvement in E. coli yields observed with enzyme-specific media, and (2) the greater proportion of ThC organisms that are E. coli, indicate that more water quality violations will occur when enzyme-specific media are used for testing than if conventional culture media are used.