Improving resolution of public health surveillance for human Salmonella enterica serovar Typhimurium infection: 3 years of prospective multiple-locus variable-number tandem-repeat analysis (MLVA)

Abstract

Background: Prospective typing of Salmonella enterica serovar Typhimurium (STM) by multiple-locus variable-number tandem-repeat analysis (MLVA) can assist in identifying clusters of STM cases that might otherwise have gone unrecognised, as well as sources of sporadic and outbreak cases. This paper describes the dynamics of human STM infection in a prospective study of STM MLVA typing for public health surveillance.

Methods: During a three-year period between August 2007 and September 2010 all confirmed STM isolates were fingerprinted using MLVA as part of the New South Wales (NSW) state public health surveillance program.

Results: A total of 4,920 STM isolates were typed and a subset of 4,377 human isolates was included in the analysis. The STM spectrum was dominated by a small number of phage types, including DT170 (44.6% of all isolates), DT135 (13.9%), DT9 (10.8%), DT44 (4.5%) and DT126 (4.5%). There was a difference in the discriminatory power of MLVA types within endemic phage types: Simpson's index of diversity ranged from 0.109 and 0.113 for DTs 9 and 135 to 0.172 and 0.269 for DTs 170 and 44, respectively. 66 distinct STM clusters were observed ranging in size from 5 to 180 cases and in duration from 4 weeks to 25 weeks. 43 clusters had novel MLVA types and 23 represented recurrences of previously recorded MLVA types. The diversity of the STM population remained relatively constant over time. The gradual increase in the number of STM cases during the study was not related to significant changes in the number of clusters or their size. 667 different MLVA types or patterns were observed.

Conclusions: Prospective MLVA typing of STM allows the detection of community outbreaks and demonstrates the sustained level of STM diversity that accompanies the increasing incidence of human STM infections. The monitoring of novel and persistent MLVA types offers a new benchmark for STM surveillance.A part of this study was presented at the MEEGID × (Molecular Epidemiology and Evolutionary Genetics of Infectious Diseases) Conference, 3-5 November 2010, Amsterdam, The Netherlands.

Figure 1

Trends of STM phage types and MLVA patterns over the 3-year study period. A: Monthly counts of STM phage types. Cumulative counts for phage types DT170, DT135 and DT9 are colour-coded. All other phage types are included in 'other' types. B: Monthly occurrence of total MLVA types and novel MLVA types observed. C: Monthly counts of STM isolates included in MLVA clusters. D: Monthly counts of MLVA clusters and trends in MLVA cluster sizes.

Figure 2

Temporal changes in STM population diversity represented by monthly figures of McIntosh's dominance index of diversity and the ratio of novel MLVA types to all MLVA types observed within a given month of the study.

Figure 3

Box plots of the mean MLVA cluster sizes and time between an initial cluster and its re-occurrence. Mean value is shown as horizontal line across the bar. Ends of a bar designate confidence intervals and dotted lines indicate the spread of values in the subgroup.

Figure 4

Sensitivity of cluster definition.