Hi Claire, there are some helpful references in the 2012 UK guidelines on gonorrhoea testing (which also discusses site of specimen collection): http://www.bashh.org/documents/4490.pdf and also here: http://www.bashh.org/documents/Guidance%20for%20the%20detection%20of%20gonorrhoea%20in%20England%20Aug%202014.pdf

Thank you Catherine, nice to hear from you! This is helpful, although I'm looking for the specific sensitivity value for culture of NG compared to NAAT.  But I'm looking at some of the references in these documents which may point me to something.  

Dear Clare,

A subject close to my heart! Much of the best work on culture sensitivity was done during the 1970s and 1980s, long before NAATs. In those days there was a major difference between those who believed in epidemiological treatment of gonorrhoea contacts (before or without diagnosis – then the norm in USA, rareish in UK) and those who believed in diagnosis (or exclusion) first. We used to examine female gonorrhoea contacts on three separate occasions with microscopy from urethra, cervix and rectum and culture from these three sites plus oropharynx. Such a regimen would be impossible, unacceptable, today.

At St Thomas’ Hospital, the Professor of microbiology, Ian Phillips, took a special interest in STIs and provided a service second to none (he was the first to publish a case of PPNG and we also reported the first cipro’ resistant strain). He provided his own selective but non-inhibitory medium (VCNT) which the lab’ made up on a daily basis. Samples were plated out directly on to VCNT in clinic and immediately placed in an appropriate CO2 incubator.

We were able to show1 that if three sets of tests were negative, the patient did not have gonorrhoea. Applying these findings to an earlier study2 it was clear that ‘epidemiological treatment’ resulted in many women taking unnecessary antibiotics.

We concluded that sensitivity of culture in women approached 99%. But only if serious trouble was taken! Comparing yesterday’s culture sensitivity with today’s NAATs is not possible. Today much less trouble is taken with sample handling and use of commercial, bought-in, growth medium. However, this is not what you will hear from the Manufacturers of NAATs. Slide 7 in my Dubai talk refers to an Abbot presentation in KL in 2006. When I challenged the presenter Dr Ho to provide references he promised to send them to me. I am still waiting3.

I have just uploaded a talk that I gave in Dubai in 2008 (Diagnostic Dilemmas, second page of my publications) which covers all this with references to some of the other contemporary work – feel free to use as you will and do come back to me if you need clarification.

Incidentally, the 1978 Lancet paper continues to be used (in the BASHH gonorrhoea guidelines, for instance). Good luck!

David Barlow

References:

1. Barlow D and Phillips I (1978) Gonorrhoea in Women - diagnostic, clinical and therapeutic aspects Lancet; i: 761-4

2. Barlow D, Nayyar K, Phillips I and Barrow J (1976) Diagnosis of gonorrhoea in women, British Journal of Venereal Diseases; 52:326-8

3. Barlow D SL1Abstract 2010. (Talk at Annual Academic Sessions, Sri Lankan College of Venereologists)

Hi Dr Barlow,

This is a very interesting discussion, thank you for sharing.  The reason I posed the question is that we're aiming to do culture-based antibiotic susceptibility testing from patients that test positive for NG on a NAAT.  Therefore, we're interested to know what proportion (of the NAAT positive) are likely to be culture positive.  

Thank you again. I will check out your slides.

Best,

Claire

Thank you for more information on your project. You may gather that the sensitivity of your culture tests will depend on how much trouble and care is taken. My slide 57 of the Dubai lecture lists pitfalls. Best to examine your culture sensitivity in urethral samples from men. Most of the problems arise in women for whom the microscopy/culture diagnosis is much less precise.

I suggest your own unit or your local microbiology laboratory does some decent controlled comparisons: for instance, if you have a high prevalence gonorrhoea population, test the laboratory with urethral samples from men - microscopy has a very high sensitivity in symptomatics – and see how it fares. One of our standard quality controls in the old days was to examine this correlation. We could pick up problems with the growth medium or incubator within 48 hours. And did so from time to time.

The other important question revolves around the specificity of your NAATs tests. Reliable (truthful?) information is really hard to come by. If you ask a Chrysler dealer which car to buy you can predict which make will be recommended. In the event you come across a GC NAATs expert who has always paid his own registration, airfare and hotel you might get a sensible answer. Professor Magnus Unemo from Orebro in Sweden is my current favourite authority. Our own experience in the early 2000s with an amplified DNA assay resulted in significant numbers of uninfected women arriving in clinic after ‘positive’ screening by their GPs (see slide 34 on). Even today’s better tests will throw up the occasional false positive – particularly if the test is being used in a low prevalence population.

Finally, if your maths is up to it, there is no better critic of diagnostic tests including NAATs for GC, than Alula Hadgu. An oldish (2005) but still relevant publication from when he was at CDC is:

http://www.pubfacts.com/detail/16135935/Evaluation-of-nucleic-acid-amplification-tests-in-the-absence-of-a-perfect-gold-standard-test-a-revi

So, before you do anything else, why not evaluate your own ‘culture’ with samples from men?

David Barlow

Exciting research, Claire! Out of my head (and I know I spent a lot of time searching), I know I came across a number down to 60% for sensitivity of the culture, but I think it was for rectal samples only. I am eager to read your article, but please account for the fact that, while sensitive, NAAT might not be specific (!), so you may have false positives - even more so if you have a relatively low proportion of disease in your population (which,could potentially apply to you, unless your population is somehow selected, like MSM and has higher prevalence...). So when you will calculate culture sensitivity, please account for: 1. different anatomic regions (growth might be different) and 2. false NAAT positives.

NAATs are, in most instances, highly sensitive and specific for chlamydia and gonorrhea testing on urine, cervical swab or urethral swab specimens. However, some studies show that factors within urine samples could hamper the performance of NAATs.

Check this out for specific site of specimen collection and methods: Article The Laboratory Diagnosis of Neisseria gonorrhoeae

Reference: https://nccid.ca/publications/naat-testing-for-gonorrhea-and-chlamydia-a-review-of-diagnostic-accuracy-cost-effectiveness-and-acceptability/

Ah, like a phoenix it rises again! Further to my two earlier offerings above, I published a summary of the pitfalls in diagnostic tests in 2017*. Alyssa's helpful comment references pre-2010 publications and I believe that NAATs for gonorrhoea have beern improved since that time. However, there remains the problem of the 'Gold Standard' and I (boringly) re-stress that standards of laboratory culture need to be exemplary if NAATs are to be convincingly evaluated.

Incidentally, the very last sentence before the References in the paper below should finish with 9 rather than superscript 9.

*David Barlow Gonorrhoea culture vs NAATs – is there a level playing field? Sri Lanka Journal of Sexual Health and HIV Medicine, 3: 42-5 DOI: http://doi.org/10.4038/joshhm.v3i0.62