Comparison Of Analytical Profile Index, Microscan Walkaway 40 Plus And Bruker Maldi Biotyper Systems For Identification Of Enteric Pathogens In Kenya

ABSTRACT

Developing countries experience diarrheal disease at a higher rate than

industrialized nations. A key contributor to increased morbidity and mortality

from diarrheal disease is the lack of efficient diagnosis methods. In Kenya

diagnosis of enteric pathogens is mainly performed using manual methods which

involves colony morphology, gram stain and biochemical tests. Although these

methods are generally reliable they are labour intensive, have a long turn-around

time and they occasionally fail to differentiate closely related species. There is

therefore need to introduce alternative methods. Fast and reproducible methods

for identification of microbial pathogens have been developed include the

Matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) and

MicroScan WalkAway 40 Plus (MicroScan). This study involves a retrospective

analysis to compare the performance and agreement of MALDI-TOF and

MicroScan against Analytical Profile Index (API) method for diagnosis of

enteric pathogens in a Kenyan environment. The isolates were randomly picked

from archive stocks previously collected in a surveillance study. The samples

were characterized using the three methods. American type concentrate culture

(ATCC) were used as controls. Polymerase chain reaction was used to resolve

discrepant results. Data on continuous variables and costing were coded, entered

in spreadsheet and analysed using SPSS 20. Out of the 196 isolates recovered

100 were Enterobacteriaceae and 96 were yeast. For yeast Candida albicans was

the most frequent (22%) and Magnusiomyces capitulum the least (6%), and all

the three methods had the same identification accuracy. For the bacterial isolates

25 genera were identified and the most frequent were Escherichia coli (27%),

Raoutella ornitholytica (7%), Klebsiella pneumonia (7%), Shigella (6%),

Salmonella (4%) and Citrobacter freundii (4%). MicroScan and MALDI TOF

were able to significantly (p < 0.05) identify more bacterial isolates than the API

method. However, three was no significant difference (p > 0.05) in the number of

yeast isolates identified by the three methods. Identity of 42 isolates with

discrepant results was confirmed using the gold standard PCR method and API,

MicroScan and MALDI TOF were able to correctly identify 19.1%, 71.4% and

76.2% of the isolates, respectively. Identification turnaround time for API was

42-144 hours, MicroScan it was 4-24 hours and for MALDI it was 6-8 minutes.

The cost per sample with API, MicroScan and MALDI TOF was Ksh 1800, 3200

and 380, respectively; therefore making MALDI-TOF the most inexpensive. In

conclusion, MicroScan and MALDI TOF were better than API in accuracy and

performance. Both methods can therefore be introduced in the Kenyan

environment for identification of microbial pathogen.