Peer reviewed articles

POLICY PLATFORM

Innovative digital technologies for quality

assurance of diagnosis of human African

trypanosomiasis

Epco Hasker 1☯

, Jean Kwete 2☯

, Raquel Inocencio da Luz 1*, Alain Mpanya2,

Nicolas Bebronne 3 , Jacquies Makabuza

2 , Yves Claeys

1 , Jérémie Ilunga

2 , Veerle Lejon

4 ,

Dieudonné Mumba Ngoyi 5 , Philippe Büscher

3 , Marleen Boelaert

1 , Erick Mwamba Miaka

2

1 Institute of Tropical Medicine, Department of Public Health, Antwerp, Belgium, 2 Programme National de

Lutte Contre la Trypanosomiase Humaine Africaine, Kinshasa, the Democratic Republic of the Congo,

3 Institute of Tropical Medicine, Department of Biomedical Sciences, Antwerp, Belgium, 4 Institut de

Recherche pour le Développement, University of Montpellier, Montpellier, France, 5 Institut National de

Recherche Biomédicale, Department of Parasitology, Kinshasa, the Democratic Republic of the Congo

☯ These authors contributed equally to this work. * rinocencio@itg.be

Policy platform

Early diagnosis and treatment is a key and often life-saving strategy in many disease control

programs, including those for human African trypanosomiasis (HAT). Good quality diagnosis

is hence as important as good quality medicine, though quality assurance (QA) systems in the

diagnostics field tend to be less well developed [1]. In recent years, the reported number of

HAT cases has drastically decreased, and the World Health Organisation aims to eliminate

HAT due to Trypanosoma brucei gambiense as a public health problem by 2020. In this context, there is a need to strictly maintain the quality of diagnostic procedures, both for screening and

for confirmation of HAT diagnosis [2]. We discuss the specific challenges of QA systems for

HAT diagnosis, compare them with QA systems for diagnostics in tuberculosis (TB) control,

and highlight possible solutions making use of new technologies.

A minimal QA program comprises of quality control (QC), external quality assessment

(EQA), standard operating procedures (SOP), and competency assessment (CA) of coworkers.

To achieve certification by the International Organization for Standardization (ISO), the ISO

15189:2012 regulation specifies requirements for quality and competence in medical laborato-

ries [3]. QC is an internal system focusing on intrinsic performance of tests used, quality of

reagents and materials used, and inclusion of positive and negative controls. CA probes the

ability of laboratory staff to properly execute their tasks and can be evaluated through internal

quality assurance (IQA) systems, such as reprocessing a number of samples and comparing

results [4]. The central concept of EQA is to monitor the level of performance of clinical labo-

ratories by comparing them to a national or supranational standard. The EQA systems in use

in disease control programs are often—but not always—based on the dispatching of a "profi-

ciency panel," i.e., a set of well-characterized samples with a known result prepared by a refer-

ence laboratory and sent for reanalysis and rereading to the clinical laboratories. One possible

downside of EQA based on testing of proficiency panels is that it monitors the optimal perfor-

mance of a laboratory but not necessarily its routine performance level [5].

TB control programs relying on sputum smear microscopy for diagnosis have for many

years understood the need for permanent QA systems. Both IQA and EQA systems are in use.

PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0006664 September 13, 2018 1 / 6

a1111111111

a1111111111

a1111111111

a1111111111

a1111111111

OPEN ACCESS

Citation: Hasker E, Kwete J, Inocencio da Luz R,

Mpanya A, Bebronne N, Makabuza J, et al. (2018)

Innovative digital technologies for quality

assurance of diagnosis of human African

trypanosomiasis. PLoS Negl Trop Dis 12(9):

e0006664. https://doi.org/10.1371/journal.

pntd.0006664

Editor: Philippe Solano, Institut de recherche pour

le developpement, FRANCE

Published: September 13, 2018

Copyright: © 2018 Hasker et al. This is an open access article distributed under the terms of the

Creative Commons Attribution License, which

permits unrestricted use, distribution, and

reproduction in any medium, provided the original

author and source are credited.

Funding: We gratefully acknowledge financial

support of the Belgian Development cooperation,

the Bill and Melinda Gates Foundation, and the

development of the Android 5 compatible

application (DRIA-2014-306-DiTECT-HAT) is

funded by the EDCTP2 programme, which is

supported under Horizon 2020, the European

Union’s Framework Programme for Research and

Innovation. The funders had no role in study

design, data collection and analysis, decision to

publish, or preparation of the manuscript.

Competing interests: The authors have declared

that no competing interests exist.

The recommended system of EQA for Ziehl–Neelsen staining relies on a re-examination of

slides from each laboratory involved in the program [6]. All slides have to be stored, and peri-

odically, a laboratory supervisor will take a random sample of slides and dispatch it to the ref-

erence lab for rechecking. The results of the expert reader are then compared to those of the

clinical lab. A major asset of this EQA system is that it monitors routine rather than optimum

performance. A Lot Quality Assurance Sampling procedure (LQAS) is used to ensure that,

with minimum effort, all laboratories in the network are monitored with comparable sensitiv-

ity and specificity [7]. The actual number of slides to be sampled depends on the workload of

the peripheral laboratory and the proportion of positive slides reported [8]. Finding false posi-

tive slides or more than the allowed number of false negatives will trigger further investigation.

Such systems have proven their value and ensure the reliability of TB microscopy results in

many countries [9].

The development of a QA system for gambiense–HAT diagnosis has always posed major

challenges. The diagnostic algorithm for HAT involves two steps: first, a screening by an anti-

body detection test such as the card agglutination test for trypanosomiasis (CATT)/T.b. gam- biense or a Rapid Diagnostic Test (RDT) to identify suspects, followed by a set of confirmation tests, usually the direct microscopic examination of lymph node puncture aspirate and/or

Giemsa-stained thick blood film, followed by a concentration method on blood such as the

microhematocrit centrifugation technique (mHCT) or the mini-Anion Exchange Centrifuga-

tion Technique (mAECT) (Fig 1). In these concentration methods, living parasites can be seen

at relatively low magnifications. The live parasites make characteristic movements, and for

many years, the diagnostic procedure has been considered to be virtually 100% specific if per-

formed well. Unfortunately for EQA considerations, preparations of live parasites (such as

mAECT) cannot be stored. This severely limits the options for EQA. Microscopy slides of thin

or thick films stained with Giemsa can be stored, but these are not the preferred ones for

Fig 1. The diagnostic algorithm for HAT in use in DRC and potential implementation of digital quality control systems [18]. DRC, Democratic Republic of the

Congo; HAT, human African trypanosomiasis.

https://doi.org/10.1371/journal.pntd.0006664.g001

PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0006664 September 13, 2018 2 / 6

routine diagnosis of HAT because of their poor sensitivity. The same limitations that apply to

storing slides would apply to proficiency panels being sent out by a reference lab. Because of

these obvious constraints, HAT control programs have so far only focused on IQA.

HAT control programs conduct case detection in two different contexts: 1. active screening

sessions conducted at the community level by mobile units and 2. passive detection in fixed

health centers. In active screening campaigns, IQA is assured by assigning two readers to each

sample and having a third senior reader solve the discrepancies. In passive screening, usually

no IQA is implemented so far.

In the context of HAT elimination initiative, even stricter QA rules apply while increasing

its complexity. The sensitivity of parasitological confirmation tests in HAT has always been an

issue. Even optimal combinations of all available tests do not reach more than 95% sensitivity

in most evaluations [10–12]. This inherent deficit in sensitivity, in a context of declining HAT

incidence, easily leads to eroding laboratory skills. Between October 2013 and April 2014, an

ad hoc EQA assessment with a focus on malaria diagnosis was conducted in the Democratic

Republic of the Congo (DRC), targeting fixed laboratories in the public and private sector.

Based on a standard set of five stained blood slides (thick films and thin films) dispatched to

445 laboratory technicians, it was observed that, overall, less than half of the participants

(45.4%) could correctly report the presence of Trypanosoma. This number went further down to 16.5% when the slide manifested a mixed Plasmodium falciparum/Trypanosoma spp. infec- tion. Recognition of Trypanosoma was significantly better in participants who declared to rou- tinely perform the diagnosis of HAT compared to those who did not perform the diagnosis of

HAT routinely (22% versus 8%), but even in the former category, performance was still poor.

Also, an assessment of routinely processed thick blood films in those laboratories showed that

only 13.6% were properly prepared and stained [13].

In recent years, as HAT is increasingly becoming a rare disease, there may also be a problem

of false positive HAT diagnoses, as observed in the northeast of DRC [14]. This is where digital

innovation may step in and become a game changer. A recent pilot study conducted by the

national HAT control program of the DRC (PNLTHA) and the Institute of Tropical Medicine

(ITM) showed that it was technically possible to record a short video of the moving trypano-

somes in the mAECT collector tube. Using a smartphone application developed by our team,

the microscopy technician was able to record an image of any positive confirmation test. We

fit the smartphone on the trinocular microscope with a custom-designed adapter. This allows

recording a video through the lens of the microscope, making use of the camera of the smart-

phone. In a preliminary evaluation of 114 CATT and/or RDT positive HAT suspects in Ban-

dundu province examined in their villages by a lab technician, two were confirmed as HAT

cases. For both patients, good quality videos of the parasite were recorded, which could be

rechecked at a distance by a senior reader, leaving no doubt about the diagnosis. Blood samples

on filter paper were obtained for all 114 HAT suspects and tested in the laboratory with

immune trypanolysis, a highly sensitive and specific test for HAT antibodies [15]. Two samples

out of 114 tested positive, which corresponded to the samples of the two HAT patients identi-

fied by the lab technician in the village. Though a sample size of two confirmed cases is insuffi-

cient to draw far-reaching conclusions on the performance of the system, the potential offered

by digital tools is obvious. This potential could be exploited in a well-designed EQA system,

adapted to the specific context. In the meantime, we have further developed the video tool by

developing an application that can be used on any Android 5 smartphone or tablet and that

uses an external camera fitted in the tube of the ocular, making the methodology more robust

and obliterating the need for a trinocular microscope (Fig 2). The app ensures that the

recorded videos are automatically saved with the registration number of the patient and with a

timestamp for easy and foolproof linkage to the corresponding patient record. This innovative

PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0006664 September 13, 2018 3 / 6

Fig 2. Digital imaging system for recording videos of microscopic images using an android tablet and external camera mounted

onto the ocular tube of a microscope.

https://doi.org/10.1371/journal.pntd.0006664.g002

PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0006664 September 13, 2018 4 / 6

digital imaging system allowed us to overcome the issue of not being able to store positive T.b. gambiense slides. The same app also addresses EQA of the screening tests (CATT or RDT), as a picture of the screening test can be taken and stored for later rechecking by the supervisor.

Synchronization of the data in the tablet or smartphone with a central database on a network

associated server (NAS) occurs whenever a WiFi connection is made.

At least equally important in EQA is a system to ensure that samples read as negative are

truly negative. Here, TB control programs recognize that errors may occur because a techni-

cian will never be able to examine each and every field in a slide. He will examine 100–200

fields, which are not necessarily the same fields as those examined by the expert in the refer-

ence lab. Thus, it may happen that false negative slides do occur, and depending on the thresh-

old chosen in the LQAS approach, a certain number of false negatives may be accepted [6]. In

HAT, recording videos of negative diagnostic confirmation tests would be meaningless. What

can be done, however, is to collect blood samples on filter paper of all subjects on whom diag-

nostic confirmation tests are done, as was done in the pilot study mentioned earlier. Such filter

papers can be stored at ambient temperature in zipper bags containing silicagel to be sent to a

reference lab on a regular basis. At the reference lab, these samples can be tested with enzyme-

linked immunosorbent assay (ELISA), immune trypanolysis, or a sequence of both tests. As in

the TB EQA system, finding more than the expected number of positives among supposedly

negative samples does not necessarily equate poor performance but should trigger further

investigation, e.g., a visit to the lab concerned by a laboratory supervisor.

Whereas in TB control, microscopy is usually performed in fixed health facilities that are

part of the general healthcare system, as explained above; microscopy for HAT is also per-

formed by specialized mobile teams that are part of a vertical program. In the DRC, there are

currently 35 such teams. The EQA system described above can easily be applied to such mobile

teams as well. It will enable the HAT control program management to focus its limited super-

visory and training capacity on those teams that need it most. As HAT is increasingly becom-

ing a rare disease, it is also important to bear in mind the lessons learned in the early days of

TB control programs. When Styblo tried to fully integrate TB control into primary healthcare,

he found out that microscopy was not always performed in a reliable manner (personal com-

munication, Jaap Veen). Some health workers just did not have enough exposure to acquire

the skills required. Styblo then advised to differentiate the tasks, meaning some health centers

would refer suspect cases and others would provide the full diagnostic workup [16, 17]. A simi-

lar approach will be required for HAT as well because one cannot expect a health worker see-

ing no more than one HAT case per year to maintain adequate skills in HAT microscopy. To

put in place a well-organized EQA system in the facilities that currently conduct microscopy

does not have to cost much more than what is already spent. The Android tablet and external

camera used in our projects together cost approximately US$600, which is not much compared

to the cost of a good microscope. On the other hand, access to mobile phones and internet net-

work is currently still challenging; until network coverage is complete across the country, crea-

tive solutions, such as offline data collection and uploading at a later stage, are feasible.

In conclusion, QA in HAT diagnostics is needed and is technically feasible. New digital

technologies offer possibilities that were, until recently, not there to set up a comprehensive

permanent QA system. The next step will be to test the operational feasibility and effectiveness

of these tools at a larger scale and in routine working conditions.

References 1. Peeling RW. Diagnostics in a digital age: an opportunity to strengthen health systems and improve

health outcomes. Int Health. 2015; 7(6):384–9. https://doi.org/10.1093/inthealth/ihv062 PMID:

26553825.

PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0006664 September 13, 2018 5 / 6

2. Franco JR, Simarro PP, Diarra A, Ruiz-Postigo JA, Jannin JG. The journey towards elimination of gam-

biense human African trypanosomiasis: not far, nor easy. Parasitology. 2014; 141(6):748–60. https://

doi.org/10.1017/S0031182013002102 PMID: 24709291.

3. International Organization for Standardization. ISO 15189:2012 Medical laboratories—Requirements

for quality and competence2012.

4. Scherz V, Durussel C, Greub G. Internal quality assurance in diagnostic microbiology: A simple

approach for insightful data. PLoS ONE. 2017; 12(11):e0187263. https://doi.org/10.1371/journal.pone.

0187263 PMID: 29135992; PubMed Central PMCID: PMCPMC5685576.

5. Sciacovelli L, Secchiero S, Zardo L, Plebani M. The role of the External Quality Assessment. Biochem

Medica. 2010; 20(2):160–4. PubMed PMID: WOS:000279350700007.

6. Ridderhof J, Humes R, Boulahbal F. External Quality Assessment for AFB Smear Microscopy. 2003.

7. Van Rie A, Fitzgerald D, Kabuya G, Van Deun A, Tabala M, Jarret N, et al. Sputum smear microscopy:

evaluation of impact of training, microscope distribution, and use of external quality assessment guide-

lines for resource-poor settings. J Clin Microbiol. 2008; 46(3):897–901. https://doi.org/10.1128/JCM.

01553-07 PMID: 18174302; PubMed Central PMCID: PMCPMC2268372.

8. Otero L, Van Deun A, Agapito J, Ugaz R, Prellwitz G, Gotuzzo E, et al. Quality assessment of smear

microscopy by stratified lot sampling of treatment follow-up slides. Int J Tuberc Lung Dis. 2011; 15

(2):211–6, i. PMID: 21219683.

9. Parsons LM, Somoskovi A, Gutierrez C, Lee E, Paramasivan CN, Abimiku A, et al. Laboratory diagno-

sis of tuberculosis in resource-poor countries: challenges and opportunities. Clin Microbiol Rev. 2011;

24(2):314–50. https://doi.org/10.1128/CMR.00059-10 PMID: 21482728; PubMed Central PMCID:

PMCPMC3122496.

10. Lutumba P, Meheus F, Robays J, Miaka C, Kande V, Buscher P, et al. Cost-effectiveness of algorithms

for confirmation test of human African trypanosomiasis. Emerg Infect Dis. 2007; 13(10):1484–90.

https://doi.org/10.3201/eid1310.060358 PMID: 18257991; PubMed Central PMCID:

PMCPMC2851529.

11. Lutumba P, Robays J, Miaka C, Kande V, Mumba D, Buscher P, et al. [Validity, cost and feasibility of

the mAECT and CTC confirmation tests after diagnosis of African of sleeping sickness]. Trop Med Int

Health. 2006; 11(4):470–8. https://doi.org/10.1111/j.1365-3156.2006.01591.x PMID: 16553930.

12. Mumba Ngoyi D, Ali Ekangu R, Mumvemba Kodi MF, Pyana PP, Balharbi F, Decq M, et al. Perfor-

mance of parasitological and molecular techniques for the diagnosis and surveillance of gambiense

sleeping sickness. PLoS Negl Trop Dis. 2014; 8(6):e2954. https://doi.org/10.1371/journal.pntd.

0002954 PMID: 24921941; PubMed Central PMCID: PMCPMC4055587.

13. Mukadi P, Lejon V, Barbe B, Gillet P, Nyembo C, Lukuka A, et al. Performance of Microscopy for the

Diagnosis of Malaria and Human African Trypanosomiasis by Diagnostic Laboratories in the Demo-

cratic Republic of the Congo: Results of a Nation-Wide External Quality Assessment. PLoS ONE. 2016;

11(1):e0146450. https://doi.org/10.1371/journal.pone.0146450 PMID: 26788725; PubMed Central

PMCID: PMCPMC4720473.

14. Van Nieuwenhove S. Challenges in Diagnosing Human African Trypanosomiasis: Evaluation of MSD

OCG’s HAT project in Dingila, DRC. http://fieldresearch.msf.org/msf/handle/10144/581666., 2015.

15. Van Meirvenne N, Magnus E, Buscher P. Evaluation of variant specific trypanolysis tests for serodiag-

nosis of human infections with Trypanosoma brucei gambiense. Acta Trop. 1995; 60(3):189–99. PMID:

8907397.

16. Styblo K. Overview and epidemiologic assessment of the current global tuberculosis situation with an

emphasis on control in developing countries. Rev Infect Dis. 1989; 11 Suppl 2:S339–46. PMID:

2652250.

17. Narvaiz de Kantor ISJ, K.; Frieden, T.; Laszlo, A.; Luelmo, F.; Norval, P-Y.; Rieder, H.; Valenzuela, P.,

Weyer, K. Laboratory services in tuberculosis control. 1998.

18. PNLTHA MdlsR. Declaration de la politique nationale de lutte contre la trypanosomiase humaine afri-

caine en Republique Democratique du Congo. DRC: PNLTHA, 2015 Novembre 2015. Report No.

PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0006664 September 13, 2018 6 / 6

Copyright of PLoS Neglected Tropical Diseases is the property of Public Library of Science and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.