Carbapenem-resistant Enterobacteriaceae in patients with bacteraemia at tertiary hospitals in South Africa, 2015 to 2018.

Enhanced surveillance for CREs was established at national sentinel sites in South Africa. We aimed to apply an epidemiological and microbiological approach to characterise CREs and to assess trends in antimicrobial resistance from patients admitted to tertiary academic hospitals. A retrospective analysis was conducted on patients of all ages with CRE bacteraemia admitted at any one of 12 tertiary academic hospitals in four provinces (Gauteng, KwaZulu-Natal, Western Cape and Free State) in South Africa. The study period was from July 2015 to December 2018. A case of CRE bacteraemia was defined as a patient admitted to one of the selected tertiary hospitals where any of the Enterobacteriaceae was isolated from a blood culture, and was resistant to the carbapenems (ertapenem, meropenem, imipenem and/or doripenem) or had a positive result for the Modified Hodge Test (MHT) according to the Clinical and Laboratory Standards Institute (CLSI) guidelines. A positive blood culture result obtained after 21 days of the last blood culture result was regarded as a new case. To distinguish hospital-acquired (HA) from the community-acquired (CA) bacteraemia, the following definitions were applied: the HA CRE bacteraemia was defined as a patient with CRE isolated from blood culture ≥ 72 h of hospital admission or with any prior healthcare contact, within 1 year prior to the current episode or referral from a healthcare facility where the patient was admitted before the current hospital. A case of the CA CRE bacteraemia was defined as a patient with CRE isolated from blood culture < 72 h of hospital admission and with no prior healthcare contact. The majority of carbapenem-resistant Enterobacteriaceae (CRE) (70%) were hospital-acquired (HA) with Klebsiella pneumoniae being the predominant species (78%). In-hospital mortality rate was 38%. The commonest carbapenemase genes were bla- (52%) and bla- (34%). The high mortality rate related to bacteraemia with CRE and the fact that most were hospital-acquired infections highlights the need to control the spread of these drug-resistant bacteria. Replacement with OXA-48 is the striking finding from this surveillance analysis. Infection control and antibiotic stewardship play important roles in decreasing the spread of resistance.