Whole genome sequencing assisted outbreak investigation of Salmonella enteritidis, at a hospital in South Africa, September 2022

Abstract

Health authorities were notified of a suspected outbreak of foodborne disease in a hospital in South Africa, where staff and patients reported acute onset of abdominal cramps, diarrhoea, fever and rigours after eating a chicken pasta meal. The aim of this report is to discuss the use of whole genome sequencing (WGS) analysis of bacterial isolates to support an epidemiological investigation. An epidemiological investigation led by the Infection Control Manager of the hospital and supported by an outbreak response team was conducted. Standard microbiological procedures were used to process stool samples and culture/identify diarrhoeal pathogens. Bacterial cultures were investigated using WGS performed using Illumina NextSeq technology, and WGS data were analysed using multiple bioinformatics tools, including those available at the Center for Genomic Epidemiology and EnteroBase. Core genome multilocus sequence typing (cgMLST) was used to investigate the phylogeny of isolates. Forty-nine cases were identified, with stool samples collected from 21 cases, and nontyphoidal Salmonella isolated from 19 out of 21 (90%) of the samples. All isolates were identified as Salmonella enterica serovar Enteritidis and differed from each other by ≤2 allele differences on cgMLST, indicating that isolates are highly genetically related. Delays in testing of food retention samples rendered the negative test results of limited value. A case-control study was conducted; eating chicken pasta was strongly associated with developing gastroenteritis (Odds Ration (OR) = 15.4, Chi-Square test with Yates correction p value = 0.02). The epidemiological evidence suggests that the chicken pasta was the likely vehicle of transmission in this outbreak, although the source of S. enterica serovar Enteritidis remains unknown.

Keywords: Salmonella Enteritidis; case–control; outbreak; whole genome sequencing.

Fig. 1.. Epidemic curve of all cases (suspected and confirmed) from three sub-populations at the hospital during September 2022.

Fig. 2.. Minimum spanning tree drawn using cgMLST data from Salmonella Enteritidis isolates (n = 68) sourced from the Free State province of South Africa for the year 2022. The circular nodes represent isolate(s). The larger the circular node, the more isolates which are reflected. The number of segments within a circular node is indicative of the number of isolates. Isolates positioned within the same circular node show no allelic differences when comparing one isolate against another. The number values between adjacent nodes indicate the number of allele differences between connecting nodes (isolates). The outbreak cluster is indicated.