Development, Stability Testing, and Performance of Xpert MTB/RIF Proficiency Testing Material in National Tuberculosis Reference Laboratory, Nairobi City County, Kenya
Loading...
Date
2024-01
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Kenyatta University
Abstract
Systematic and ongoing proficiency testing (PT) program is a fundamental element of
external quality assurance in diagnosis of TB using Xpert MTB/RIF assay. In many
resource-limited setting, particularly in Kenya, PT is poorly covered and largely
inconsistent following inadequate supply of dried tube specimens (DTS). This has
serious negative implications on patients‘ management and TB control programs. The
aim of this study was to develop, assess stability and performance of local DTS panels
for Xpert MTB/RIF assay PT in Kenya. This study was conducted at the National
Tuberculosis Reference Laboratory (NTRL) and DTS piloting was done in Xpert
MTB/RIF sites in Nairobi City County. An experimental study design was adopted to
develop and assess stability DTS panels, and a cross sectional study design was
adopted to pilot panels. The piloting sites were purposively selected to include all the
twenty (20) TB diagnostic laboratories that were using Xpert machine in Nairobi City
County, from August 2019 to December 2019.Bacterial strains stored in a -80oC
freezer at the NTRL were retrieved and processed following standard bacteriological
procedures. The strains were sub-cultured in liquid culture media (MGIT), growth
detected by the BACTEC® MGIT 960® system, and their purity confirmed. Bacterial
cultures were then heat-inactivated and pre-tested using Xpert MTB/RIF assay before
preparing DTS. To assess the effect of drying on the stability of DTS panels, DTS
panels were left uncapped inside a biosafety cabinet for 7, 10, and 14 days, and
immediately subjected to Xpert MTB/RIF assay. In assessing the stability of the DTS
panels at various temperatures as anticipated in TB diagnostic laboratories that use
Xpert machine across the Kenya, 36 DTS panels from 7, 10, and 14 days drying
periods were held at 18oC, 20oC, 24oC, 33oC, and 40oC for 12 weeks, and subjected to
Xpert MTB/RIF weekly. For piloting, a set of 5DTS panels, 5 disposable sterile
dispensing pipettes, processing instructions, and PT results evaluation form were
delivered to the targeted piloting sites in a sealable transparent bag for testing by the
Xpert MTB/RIF personnel who printed the results, entered them into the reporting
form, and a scanned image send to the principal investigator for analysis. Data were
analyzed using STATA v17 and variables were summed up in medians, interquartile
range, means, and standard deviation (SD). Each of the five pilot DTS panels was
assigned an accuracy score of 20 points, and individual laboratory scores as follows;
incorrect determination of either MTB detection (0 points), unsuccessful result (error,
invalid, or no results) (5 points), RIF-indeterminate result (10 points), and correct
determination of both MTB detection and RIF resistance (20 points). Total scores for
each laboratory were computed, with 100% being considered Excellent, ≥80%
Satisfactory, and <80% Unsatisfactory. The DTS panels‘ probe A Ct values were
within the expected mean range (16-23) and SD limit (≤3), with100% concordance
between those dried for 7, 10 or 14 days and the pretest results. Except for DTS
panels held at -80oC, the probe A mean Ct values of panels dried for 7 and 14 days
increased with increasing temperature (from -20oC to 40oC) during the 12 weeks‘
study period (p= <0.001), with the greatest mean Ct value increment was at
40ᵒC.Ninety-five per cent (18/20) of piloted laboratories reported the expected, with
10% (2/20) of sites giving discordant results, false MTB detection (5%, 1/20) and
error (code 5007) (5%, 1/20). 90% (18/20) of the piloted sites had satisfactory, 80%
had excellent, and 10% (2/20) had unsatisfactory results. This study recommends
adoption of 7-day DTS drying period, -80oC and 40oC storage up to 2 weeks, and up
scaling DTS production for profic
Description
A Research Thesis Submitted in Fulfilment of the Requirements for the Award of the Degree of Master of Science in Infectious Diseases (Clinical Bacteriology) in the School of Health Sciences of Kenyatta University, January 2024.
Supervisors
1. Abednego Musyoki
2. Nelson Menza