Key to antimicrobial stewardship success: Surveillance by diagnostic microbiology laboratories.

The important role of laboratories in enhancing antimicrobial stewardship activities through improved diagnostics and provision of surveillance data is globally recognised.[1] Consider the aim of an antimicrobial stewardship programme: ‘optimize clinical outcomes while minimizing the unintended consequences of antimicrobial use’. [2] The clinical microbiology laboratory plays a critical role in achieving these aims through the provision of culture and susceptibility data that are both patient­specific (optimisation of clinical outcomes) and informative for surveillance activities that guide empirical antimicrobial selection (minimising unintended consequences of antimicrobial use). For this reason, the World Health Organization (WHO) included the strengthening of surveillance and laboratory capacity in its 2011 World Health Day six­point plan to combat antimicrobial resistance (AMR).[3] South African laboratories in the public and private sectors have the means to provide surveillance data and, through a collaborative approach, the capacity to create antimicrobial resistance maps. In line with the WHO’s recommendations and under the auspices of the South African Society of Clinical Microbiology, efforts are currently underway to improve national AMR surveillance data for typical healthcare­ associated pathogens. The generation and provision of these data is, however, only half of the challenge. The analysis and interpretation thereof is equally important, as highlighted in this month’s SAMJ by McKay and Bamford[4] in their study comparing community­ with healthcare­acquired bloodstream infections at Groote Schuur Hospital (GSH), Cape Town, a tertiary public sector hospital. The study is a retrospective description of bloodstream isolates submitted to the GSH laboratory over a 1­year period. Using predefined criteria, the isolates were classified as community or healthcare acquired and a comparative analysis of resistance patterns was undertaken. The issue of differing resistance patterns in the community v. hospital is a well­established and much­publicised phenomenon.[5] The article serves to reinforce these established differences, highlighting the profound resistance associated with hospital pathogens: (i) a 47.6% difference in the methicillin­ resistant Staphylococcus aureus (MRSA) rate; and (ii) a ~35% difference in extended­spectrum β­lactamase (ESBL) production by Enterobacteriaceae. The authors have provided substantive local surveillance data to support their recommendations for empirical antimicrobial prescribing based on an assessment of whether the infection is community or healthcare acquired. They have therefore highlighted a crucial element that, for various reasons relating to surveillance capacity, has unfortunately not been available in similar published local AMR surveillance data.[6­8] This element is the clinical and epidemiological context that is required for interpretation of the data. Unfortunately without this context the data become blurred and the issue of resistance is magnified disproportionately. Aggregated data (which is what we have seen to date) give an indication of whether overall resistance is on the increase or decrease, but do not provide sufficient information to guide practice at a local level. Nearly three­quarters of all bloodstream infections (BSIs) from GSH, as McKay and Bamford show, were healthcare acquired; when