Date of Award

5-26-2017

Document Type

Dissertation

Degree Name

Master of Medicine (MMed)

First Supervisor/Advisor

Dr. Gunturu Revathi

Second Supervisor/Advisor

Prof. Zul Premji

Third Supervisor/Advisor

Dr. Claudia Daubenberger

Department

Pathology (East Africa)

Abstract

Background: MALDI-TOF MS is an analytical method that has recently become integral in the identification of microorganisms in clinical laboratories. It relies on databases that majorly employ pattern recognition or fingerprinting. Biomarker based databases have also been developed and there is optimism that these may be superior to pattern recognition based databases. This study compared the performance of ribosomal biomarker based MALDI-TOF MS and conventional methods in the identification of selected bacteria and yeast.

Methodology: The study was a cross sectional study carried out in collaboration with the Swiss Tropical and Public Health Institute, Switzerland and Mabritec laboratory, Riehen, Switzerland. Clinically relevant bacteria and yeasts identified from clinical specimens submitted to the Aga Khan University Hospital, Nairobi (AKUH, N) laboratory were included in the study. The processing of the clinical specimens and identification of the isolates using conventional Vitek 2 automated system and serotyping was performed at AKUH, N while the identification of the isolates using MALDI-TOF MS and analysis of spectra using SARAMISTM and PAPMIDTM databases was performed at Mabritec laboratory, Riehen, Switzerland. Comparison of sensitivities was then carried out using Pearson Chi-Square test or Fischer’s exact test and P value of < 0.05 was considered statistically significant. Secondary outcomes analyzed include the proportion of major and minor errors reported in percentages.

Results: Of the 386 isolates MALDI-TOF MS and conventional methods identified 97.6 % and 95.8 % (p = 0.155) to the genus level and 97.5 % and 87.3 % (p < 0.001) to the species level respectively. Biomarker based MALDI-TOF MS was significantly superior to Vitek 2™ in the identification of Gram negative bacteria and Gram positive bacteria to the species level. PAPMIDTM correctly identified isolates of Klebsiella variicola and Acinetobacter genomospecies 13BJ/14TU that were misidentified by SARAMISTM and Vitek 2™. For the Gram positive bacteria, significant difference was observed in the identification of Coagulase negative Staphylococcus (p < 0.001) and Enterococcus (p = 0.008). Significant difference was also observed between serotyping and MALDI-TOF MS (p = 0.005) and this was attributed to the lack of identification of Shigella species by MALDI-TOF MS. No significant difference was observed in the identification of yeast however some species of Candida were unidentified by MALDI-TOF MS.

Conclusion: Biomarker based MALDI-TOF MS had good performance in a clinical laboratory setting with high sensitivities in the identification of clinically relevant microorganisms.

Included in

Pathology Commons

Share

COinS