Antimicrobial susceptibility testing (AST) allows for appropriate antibiotic therapy and monitoring of bacterial antibiotic resistance that threatens our ability to treat potentially fatal infections.^1,2 For monitoring to happen, susceptibility testing results must be of good quality. However, in Africa, the quality of microbiological results on antimicrobial resistance (AMR) is a significant challenge.³ Indeed, very unusual results combinations have been found in several African studies. For instance, a study conducted in Egypt revealed that over 80% of Proteus isolates were resistant to imipenem, 80% were sensitive to ceftriaxone, and 50% were sensitive to vancomycin.⁴ Other studies in Ethiopia, Namibia, and Rwanda have shown resistance to penicillin in Streptococcus pyogenes and vancomycin resistance in Staphylococcus aureus isolates from cerebrospinal fluid.^5,6,7 The susceptibility test result quality depends on the equipment (antibiotic discs and culture media), compliance with standard operating procedures, and good laboratory practices.⁸

To improve data quality, the World Health Organization encourages countries to develop a national action plan to tackle AMR.¹ One of the priorities of this national action plan is the implementation of laboratory-based surveillance for AMR. Establishing a laboratory network under the coordination of a reference laboratory is essential to provide factual data to estimate the impact and evolution of AMR in a country. The quality of the data collected from laboratories is one of the significant challenges of this surveillance. The quality of the data collected depends on the availability of qualified human resources, appropriate equipment, and suitable methods. Harmonisation of practices and use of standardised procedures improve the quality of the data collection and correct interpretation. They also allow inter-laboratory data comparison and sharing at several levels (national, regional, and international) through the data entry and processing software provided by World Health Organization.⁹

In 2017, Burkina Faso’s Ministry of Health developed an AMR national action plan, which included the implementation of laboratory-based surveillance of AMR.¹⁰ As part of the Global Health Security Agenda programme, the African Society for Laboratory Medicine, with financial and technical support from the United States Centers for Disease Control and Prevention, supported the Ministry of Health in establishing the laboratory-based surveillance system of AMR in 2018. The overarching goal of the system is to provide comparable and validated data on the prevalence and trends of AMR in a group of bacteria of interest for both the World Health Organization Global Antimicrobial Resistance Surveillance System and the country.¹¹ A national guideline for laboratory-based surveillance of AMR and a standardised national AST procedure was developed and endorsed by the Ministry of Health¹² to support the implementation of this surveillance system. To ensure the accuracy and reliability of susceptibility testing results, the quality of the reagents used, and the competence of the staff performing and interpreting the tests, the guideline required laboratories to carry out weekly quality control of antibiotic discs and the culture media after preparation.^13,14 However, no study data on quality control of sentinel laboratories for AMR have been found since implementing laboratory surveillance in Burkina Faso. To ensure compliance with the national AST procedure and the quality of data produced by sentinel laboratories, the National Antimicrobial Resistance Surveillance Laboratory, with support from the United States Centers for Disease Control and Prevention through the African Society for Laboratory Medicine and Johns Hopkins Program for International Education in Genecology and Obstetrics (Jhpiego), initiated this first quality control study of sentinel AMR laboratories. This study aimed to assess the compliance of sentinel laboratory practices with the national AST procedure and propose an action plan to correct non-conformities.

This study is the first to assess the implementation of quality indicators in the AMR sentinel site laboratories since Burkina Faso implemented the AMR surveillance in 2018. This activity has provided baseline data on the sentinel laboratories’ compliance with the national AST procedures manual, particularly in areas performing quality control checks or meeting quality indicators for culture media and antibiotic discs. The Ministry of Health meets the requirement to promote good practices in sentinel laboratories by conducting this assessment. An impartial review of documents and records and direct observation of practices allowed an objective assessment of compliance with the national procedure. We analysed the non-conformities identified during the evaluation to determine their impact and severity on operations, processes, patient outcomes, and safety considerations. Thus, the assessors proposed urgent actions depending on the criticality of each detected non-conformity. Each laboratory received reports with proposed improvement plans and recommendations.

A too-acidic agar pH increases ß-lactam activity, while an alkaline pH favours aminoglycoside and macrolide activity. Agar thickness has an impact on the diffusion gradient of the antibiotic.¹⁸ Nevertheless, a few laboratories controlled these parameters in our context. Maintaining the required agar thickness and pH depends on the medium dispensing technique, which requires a calibrated dispenser and a pH meter; however, only a few laboratories had these.

Fertility and sterility are the most important indicators in medium quality control. The medium must be carefully checked to avoid contamination before culture.¹⁹ All sentinel laboratories checked the culture medium quality indicators to ensure each medium met the microbiological characteristics before use.

Åhman et al. found that many discs exhibited significant quality issues, with 33% (n = 26 of 78) out-of-range readings in the first study and 17% (n = 20 of 120) in the second. Furthermore, discs from certain manufacturers showed unexpected variations in inhibition diameter zones (4 mm – 9 mm) for discs from the same vial.²⁰ In our study, only 67% of our sentinel laboratories monitored the quality indicators of antibiotic discs; this situation could increase the chances of obtaining inaccurate results and, consequently, ineffective antibiotic therapy.²¹ This lack of disc quality control is concerning given the already alarming global resistance situation for some anitbiotics. Indeed, the resistance of E. coli to colistin, currently used as a last-resort antibiotic, has been observed in Asia, Europe, America, Africa, and Oceania (specific percentages for each continent: 66.72%, 25.49%, 5.19%, 2.27%, and 0.32%, respectively).^22,23 The low rate of laboratories complying with these indicators can be attributed to staff mobility due to various assignments, lack of staff involvement, and the fact that, in most laboratories, personnel selection for training is not based on expressed training needs or competence assessments. In fact, in many laboratories, the staff trained in microbiological testing by the Ministry of Health and its partners did not necessarily come from the microbiology department due to an established internal rotation for various continuous training programmes. Staff were selected without necessarily considering the specific field covered by the training. We believe that the Ministry of Health should keep trained staff as long as possible. Furthermore, although participation in training is a source of motivation for staff, it would be good to take into account the specific field covered by the training in the selection of participants.

The reference strains are used to assess the overall performance of the test. They are used for quality control of all steps of culture and the study of bacterial sensitivity to antibiotics.¹⁸ Quality control of antibiotic discs using reference strains should be performed at least every two weeks.¹² Performing checks at short intervals allows for a retrospective analysis of previous results to identify issues in the event of discrepant results, allowing for corrective actions to be taken in case of non-conformity results. Despite this, only 58.83% of the laboratories carrying out the quality controls of the antibiotic discs had defined periodicity of passage, and only 71.43% respected this predefined periodicity. This situation demonstrates that implementing the procedures in the national AST manual is still weak. The Ministry of Health should strengthen awareness, communication, and especially laboratory staff training.