Susceptibility testing practices for Streptococcus pneumoniae: results of a proficiency testing survey of clinical laboratories.

In June 2003, a test sample was sent to 355 laboratories enrolled in a proficiency testing program to assess their ability to detect low-level penicillin resistance in a strain of Streptococcus pneumoniae. One hundred fifty participants reported results for antimicrobial susceptibility testing. Of the 62 respondents using disk diffusion, 34 (55%) failed to report a result that was acceptable for detecting penicillin resistance and 30 (48%) reported a result for one or more drugs not approved for testing S. pneumoniae. Moreover, 12 (14%) of the 88 respondents using minimum inhibitory concentration methods reported results for at least one unapproved drug. These findings support the conclusions of other studies that antimicrobial susceptibility testing practices are suboptimal in many laboratories. Resolution of this problem will require continued educational initiatives, studies to discover the reasons why laboratories fail to follow published standards, and enforcement of the use of standards by regulatory and credentialing agencies.