Investigation of the Molecular Epidemiology of Clostridium difficile in Scotland by MLVA

Background: Clostridium difficile PCR ribotype 027 (NAP1) was first identified in Scotland in late 2007 and is now the third most common type observed in Scotland after types 106 and 001, accounting for between 9.1% and 15.5% of human isolates submitted to Scottish Salmonella, Shigella & Clostridium difficile Reference Laboratory (SSSCDRL) for typing each quarter (see Figure).  While PCR ribotyping continues to be a valuable typing method for surveillance purposes, more discriminatory methods must be applied for epidemiological investigation of the spread of ribotype 027.

Objective: To evaluate the application of Multi Locus Variable Number Tandem Repeat Analysis (MLVA) in the investigation of outbreaks of C.difficile PCR ribotype 027 in Scotland.

Methods: Sixty-seven isolates of C.difficile 027, submitted from hospital laboratories around Scotland between November 2007 and end September 2009 were examined.  Isolates were from suspected outbreaks, sporadic hospital cases and community acquired cases.  All were subjected to MLVA analysis using the method described by van den Berg et al. (2007).  Seven different tandem repeat loci were examined.  Results were analysed using eBurst v3.0.

Results: Fifty six different MLVA types were identified among the isolates examined.  Six separate clusters, based on the existence of single locus variants (SLVs) were identified containing between 3 and 20 isolates (mean = 8.5).  Clusters correlated with known outbreaks and/or individual hospitals.  The temporal spread within clusters ranged up to 18 months.  Isolates from one suspected hospital outbreak were found in two different clusters.  Further epidemiological investigation indicated that two separate incidents were occurring simultaneously.  Fourteen isolates had no SLVs and existed as singletons.

Conclusions: MLVA showed a high degree of discrimination, able to discriminate 56 different types within the collection of 027 isolates examined.  The results from cluster analysis of different types were found to be congruent with the available epidemiological information.  MLVA was able to identify six particular clusters associated with individual hospitals or outbreaks over a period of time up to 18 months.  This would indicate the spread of individual clones within distinct geographical areas and highlight the potential value of MLVA typing for epidemiological analysis.  MLVA can be used as a highly discriminatory tool, supplementary to PCR ribotyping, for investigation of the molecular epidemiology of C.difficile PCR ribotype 027.