0 Recommendations

GeneXpert as trusted testing tool

By Maopa Lewabeci Raikabula | 19 Sep, 2013

Hi, am Maopa from Fiji,
There are 3 DOTs centres in Fiji and all have been provided with GeneXpert. There has not been any MDR TB reported for Fiji. Recently, we ran a geneXpert for a 3+ AFB case with the following result;
1st specimen = MTB Detected High/Rif resistant Positive
2nd specimen = MTB Detected High/Rif resistant negative
3rd specimen = MTB Detected Low/Rif resistant negative
The result gave a confusing status of Xpert result especially with no other confirmation apart from culture and MPT64 done locally. The specimen has then been sent to QMRL, Australia for confirmation and DST.
Can we still trust GeneXpert as a identification tool for MTB and Rifampicin resistant?



SYDNEY CHIKUKWA Replied at 3:03 AM, 20 Sep 2013

Hie Maopa
GeneXpert like any other equipment is not immune to faults, but they are always measures that are put in place to ensure error is detected on time or minimised and this is commonly refered to quality assurance and include quality control and proficient testing, procurement, training, and so forth. When you receieve any instrument you must be trained by competent suppliers how to use it and trouble shoot minor problems. After that you have to validate if your machin is giving the expected results and this you do by comparing your results with the test method you were using before receiving your GeneXpert and this is usually your culture and sensitivity. You can run 14negative samples and 36 positive(4 scanty, 4x 3+positive, 4x 2+positive, 4x 1+positive) and send same specimens for culture and sensitivity and then when your results come you do a correlation calculations and establish your c.v.%. If you are happy with this then every day before you run your tests you run quality controls and if your controls are in, then you will be able to trust your results. In your case i would advise that since you have 2 sites with GeneXpert collect 2 paired fresh specimens and send them to your 2 sites and then compare the results and share with us your findings. Hope you have got some help, but surely the GeneXpert used with well trained staff and quality assurance is very realiable above 90% specificity and sensitivity
Dr. Sydney Chikukwa(Head of Microbiology Oshakati Namibia Institute of Pathology)

Ellen Munemo Replied at 3:42 AM, 20 Sep 2013

Thank you very much for the detailed response you gave. I take it as an advice to all those who are using the geneXpert or any other equipment for diagnosis. Verification of your results before fully implementing a new diagnostic tool is mandatory. As Sydney said you compare with your gold standard. Quality assurance is also very important. Probably you could share with us more detail on what quality measures you had put in place before running that sample. Also look at a larger number of sputum samples before you conclude. As a diagnostic tool, I find the GeneXpert to be really helping us especially in HIV positive patients. As new technologies for diagnosis are being developed, lets also think on quality assurance as a mandatory step in the implementation of the tool.
Ellen Munemo

Alaine Umubyeyi Nyaruhirira Replied at 5:12 AM, 20 Sep 2013

Dear Maopa

Thx for your question and sharing this finding. Really your results is a little bit strange and can really confuse staff on the interpretation and how to orient clinicians. As advices by Sydney and Ellen , QA program is very important not for the test (Xpert) has its own internal control, but for the performance of the users / staff. It will be better to explain more and check if the samples/ specimens are all collected from the same patient (Spot, morning) . Also a confirmation test by a gold standard technology can help to clarify the discordance as you can find in a same sample different strains with different profile ( exceptionally situation , but this happen) The finger print using molecular test like MIRU or Spoligotyping can help to clarify.
I 'm involve in GeneXpert implementation in various countries in Africa , the tool is really helping and particularly in PLHIV patients and children.

Keep us posted on the further finding.
Dr Alaine, MSH , Pretoria , South Africa

Evgenia Geliukh Replied at 5:38 AM, 20 Sep 2013

Hi! I'm Evgenia Geliukh, MD from Ukraine.
Dear colleges!
Does anyone know were there any comparative study of the results of Gene-Xpert with results on solid and liquid media? What should be a percentage of matches?
Thank you

Пятница, 20 сентября 2013, 5:13 -04:00 от Alaine Umubyeyi Nyaruhirira via GHDonline < >:
Отправлено из мобильной Почты Mail.Ru

Mohammed suaudi Hassen Replied at 7:24 AM, 20 Sep 2013


it`s nice to release such amazing concept .so could you bring the soft copy materials about GeneXperts their maintenance and how to Calibrate it .
King regard

Mohammed SUAUDI
Medical Laboratory Science.
Mobile phone: +251911355221
Regional Reference TB Laboratory Expert
Oromia Public Health Research Capacity Building and Quality Assurance Laboratory,
P O Box 688

Alaine Umubyeyi Nyaruhirira Replied at 8:11 AM, 20 Sep 2013

Dear Mohammed

You can find all related tool on GeneXpert at the Web site of Cepheid.
Or send me your private mail I can send you all related materials.

soundiram indira Replied at 8:19 AM, 20 Sep 2013

I am happy to see that people are interested and involved on the use of the xpert as diagnostic tool.
Please note that the last version of training materials are loaded on our ftp. You cannot find them on Cepheid's website.

Feel free to visit it anytime you need some updates.
Please copy and paste the link ftp://hbdc: in EXPLORER, not INTERNET EXPLORER

Thank you very much, and feel free to come back to us should you require any further information,

Alaine Umubyeyi Nyaruhirira Replied at 8:31 AM, 20 Sep 2013

Dear Eugenia

Please find attached some paper related to your question.
Hope it will help.

Mohammed suaudi Hassen Replied at 10:02 AM, 20 Sep 2013

this my email

Evgenia Geliukh Replied at 10:28 AM, 20 Sep 2013

Dear Alaine, there is no attached files. My e-mail:

Sophie Beauvais Replied at 10:30 AM, 20 Sep 2013

Hi All, very interesting discussion, please keep it going in the discussion thread so everyone can benefit even if you end up connecting via email!

A few comments:

> Alaine: GHDonline does not support file attachments by email thus the paper you mention in your previous comment was not posted. Could you sign in on the web and post in new reply?

> The ftp files that Soundiram mentions work fine with Internet Explorer and Chrome - just tried

> Evgenia re: comparative study of the results of Gene-Xpert with results on solid and liquid media - Have you searched GHDonline and looked in our past GeneXpert MTB/RIF panels?

The 1st panel "Challenges in Rolling-out the GeneXpert MTB/RIF Diagnostic Test in Resource-Limited Settings":

The 2nd panel "Scaling-up GeneXpert MTB/RIF" http://www.ghdonline.org/drtb/discussion/expert-panel-dec-10-14-scaling-up-ge...

You should find some valuable info in there.

Best, Sophie

Gaddo Flego Replied at 11:00 AM, 20 Sep 2013

Dear all,
I read your discussion being interested more on the impact of the introduction of innovative technology than in the specific topic of MDR TB (an absolutely major issue, anyway!); in fact I believe that shared information on pratical issues is vital for an informed decision on health technology implementation/disinvestment. I would ask GHDonline to consider to start a discussion on Health Technology Assessment (HTA) in global health delivery.
Thanks to all, Gaddo.

soundiram indira Replied at 11:14 AM, 20 Sep 2013

Please note that the last versions of training materials are loaded on Cepheid's ftp.
The training materials are updated from time to time depending on users' needs and updates.

Feel free to visit it anytime and copy the last versions.
Please copy and paste the link ftp://hbdc:

You will find these training materials available in English, French, and Russian so far.

Thank you very much, and feel free to come back to us should you require any further information,

Edy Nacarapa Replied at 11:48 AM, 20 Sep 2013

how i coulg get in ink ftp://hbdc:

____________________________________Edy Nacarapa M.D Global Health ExpertClinical Director Hospital CarmeloDivision of HIV/TB TreatmentChokwe - MozambiqueAv. TrabalhoP. Box 0035Cell: +258825133571
Cell: +258847435380
Skype: edy.nacarapa

Subject: Re: GeneXpert as trusted testing tool

Alphonse SINDAYIGAYA Replied at 12:05 PM, 20 Sep 2013

please send me geneXpert tool on

Ministry of Health
Postgraduate in Internal Medicine
Phone number:0788842406

De : Alaine Umubyeyi Nyaruhirira via GHDonline <>
À : Alphonse SINDAYIGAYA <>
Envoyé le : Vendredi 20 septembre 2013 14h15
Objet : Re: GeneXpert as trusted testing tool

Alaine Umubyeyi Nyaruhirira replied to a discussion in MDR-TB Treatment & Prevention:
Dear Mohammed
You can find all related tool on GeneXpert at the Web site of Cepheid.
Or send me your private mail I can send you all related materials.
--Visit GHDonline to reply, upload a file, recommend, or share this discussion

Replies to this email are immediately shared with the community.
You are receiving this email because you are a member of GHDonline. Too many emails? Change or turn off your email notification settings by updating your profile.
GHDonline | Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115
Stay connected: Twitter Facebook Blog Google+ LinkedIn

Shubhada Shenai Replied at 3:50 PM, 20 Sep 2013

Please find attached papers related to your question.

In GeneXpert assay, the difference in CT between the first (early CT) and last (later CT) of M. tuberculosis-specific molecular beacon (delta CT Max) is the basis of rpoB mutation and RIF resistance detection. RIF resistance is identified if the delta CT Max is more than 3.5 cycles. Generally, rpoB mutations that completely inhibit probe hybridization (and thus caused one or more molecular beacons to show CT values of more than 38) are defined probe “dropouts,” whereas rpoB mutations that permit partial probe hybridization and produce a measurable delta CT Max of more than 3.5 cycles are considered as “delays.” (Blakemore et al 2010)

False resistance call due to any of the MTB specific probe delay could occur due to various reasons eg. presence of hetero-resistant bacterial population, prolong incubation in Sample Reagent (ref. Banada et al 2010) or even due to very high bacterial load (>10E7). The best way to confirm this is DNA sequencing of PCR amplicons (from Xpert cartridge) showing RIF resistance.

Attached resources:

PETER NWADIKE Replied at 4:13 PM, 21 Sep 2013

Thanks Shubhada for the publication links.

Zea Rahim Replied at 11:20 PM, 21 Sep 2013

Dear Colleague,
I am using Internet Explorer. Not explorer. As a result, I am unable to follow the link: ftp://hbdc: <mailto:>
I shall appreciate receiving the guidance how to access this link.
War regards and best wishes.

PETER NWADIKE Replied at 9:15 AM, 22 Sep 2013

Dear Zea,

You need to copy and paste the link (ftp://hbdc: )

Maopa Lewabeci Raikabula Replied at 6:27 PM, 22 Sep 2013

Thanks alot Dr Chikukwa and all for the comments. As first time users of GeneXpert, I do appreciate the extra knowledge I have gained from this communication model. I will now need to share this with my collegues and ensure implementation of QA with new staff. All users were trained by the supplier and we also had validation done.
We do perform AFB smears first before running a geneXpert according to the NTP criteria. We also perform cultures on solid media but send growth for DST to QMRL. We're now looking forward to purchasing liquid cultures. Would it also be possible to have colonization of both Rifampicin sensitive and resistant MTB?

soundiram indira Replied at 4:06 AM, 23 Sep 2013


As many of you are not able to access the ftp to load the training materials, I have attached some of them on this page. (not all are loaded here-most important ones)
Please note that any updates are available upon request or on the FTP.

If you need any information regarding this program, please visit this link http://www.cepheidcares.com/tb/index.php

If you need any training, please contact us or visit our cepheidcares website

Training videos are also available on the following link http://www.cepheidcares.com/tb/index.php/cepheid-solution/

With best regards,

Indira Soundiram
Customer Care& Training Manager

Attached resources:

SYDNEY CHIKUKWA Replied at 5:18 AM, 23 Sep 2013

Thanks so much
Resistance as you are aware occurs in two stages
two steps: 1. Mutation and 2. Selection
§mutations conferring drug resistance occur spontaneously in individual
organisms at a very low rate
§with high numbers of organisms (lung cavity) small populations resistant
to each medicine will exist
§organisms with resistance mutations to a particular medicine will have a
survival advantage over susceptible organisms when all are exposed to that
§the resistant organisms can outgrow the susceptible organisms
§1 in every 109 (billion) cell divisions
INH 10 to power 6
•Spontaneous development of MDR-TB bacilli is extremely rare:

Yes in a patient developing resistance will definatelty have 2 population
and as continued exposure to resistant antibiotic
continues unabated (Treatment monitoring not done), the resistant strains
become the only dominant bacilli in the lung cavity.
Hope you got some help, this is an extract from out TB treatment guidelines
Dr. Chikukwa

Badri Thapa Replied at 6:08 AM, 23 Sep 2013

Dear Sydney and others,
Few things that I would like to add in the discussion on resistance in MTB;
1. Although less described, efflux pumps in the cell wall of mycobacteria also contributes in pumping out the drug molecules and this mechanism is not related to the stages of the development of resistance during the treatment. Somewhat called intrensic/natural resistance.
2. The spontaneous development of the MDR-TB is actually more common than what you/guideline have given (10 to the power 15; Rif 10 to the power 9 plus INH 10 to the power 6). The mathematical modeling described in the paper by Colijin C et al_PLOS One_2011 says that the chance of development of MDR-TB is 10 to the power 4 to 5. Interesting paper to read. Follow the link
Dr B Thapa

SYDNEY CHIKUKWA Replied at 7:12 AM, 23 Sep 2013

Thanks for the insight

This Community is Archived.

This community is no longer active as of December 2018. Thanks to those who posted here and made this information available to others visiting the site.