Tag Archives: CSF viral PCR

“Something for the weekend?”

I know it is difficult to believe, but patients tend to be just as sick on Saturday and Sunday as they are from Monday to Friday.

I am not sure if the concept of the “weekend” is secondary to religion or the industrial revolution. Regardless of its origins, in healthcare facilities throughout the world, the clinical service that is provided at the end of the week is reduced somewhat. Weekends are good for society, but not so good for patients…

And the weekend is not always just two days. Add on a Public Holiday Monday and the weekend can stretch to three days. Over Christmas, New Year and Easter, the “weekend” is often four days. Four days is a long time if you are sick.

In clinical microbiology laboratories everywhere, molecular testing traditionally never happened at the weekend. The molecular department happily shut up shop on Friday afternoon, and re-opened again on Monday morning. In times gone by, this was due to the rather specialised nature of the testing, and also the relatively slow turnaround times.

Take the following hypothetical scenario, which you may be familiar with…

It is the evening before the Christmas break. A cerebrospinal fluid (CSF) sample comes in to the microbiology laboratory from a patient in ICU with meningoencephalitis. A viral PCR panel is requested to try and ascertain the cause of the patient’s symptoms. However because it is the weekend coming up, followed by two public holidays, it will be 5 (long) days before the PCR is performed.

A viral CSF PCR result can have the following positive effects on patient management:

  • A negative (HSV) result can allow empirical therapies such as acyclovir to be discontinued, or in the case of a positive result, the dose to be optimised.
  • A positive result can prevent further investigations, such as MRI scanning, and  other “exotic” laboratory tests being carried out on the CSF.
  • Can expedite discharge, when diagnosis is known.

The long weekend progresses, and the microbiology department continues to analyse significant volumes of relatively low value specimens. e.g. ear swabs, peri-anal abscess swabs, vaginal swabs, swabs from leg ulcers etc, etc. Yet sitting there patiently in the fridge is that CSF, probably the most important sample in the laboratory, and the one which could have the most immediate and profound effect on patient management. It looks on enviously at all the attention the other samples in the laboratory are getting!

This model doesn’t really cut the mustard anymore… We need to utilise the new (user friendly and rapid turnaround) molecular platforms (Cepheid Genexpert, BD Max as examples), so that we can offer a laboratory service at the weekend which has a genuine clinical impact.

Things are changing however. Molecular testing is slowly being introduced at the weekend in many microbiology laboratories. C. difficile testing and Influenza/RSV PCR are a couple of examples. But this progress takes time, and sometimes it takes more time than it should do.

We need to “disrupt” traditional weekend work at the clinical microbiology laboratory… On the weekend, put those ear and vaginal swabs back in the fridge, and find a way (somehow) to take the CSF out!


Remember “The Purple Cow Giveaway” for new subscribers. 5 new subscribers so far in the draw for 5 books, so the odds of winning are quite good! 4 days left.

“Accepting Rejection”

Apologies for the recent sparcity of posts as I continue my futile efforts at finding a suitable apartment in Paris…

A recent paper in the Journal of Clinical Microbiology caught my interest regarding the use of rejection parameters for deciding whether or not to test CSF for Herpes Simplex Virus (HSV) by PCR, according to the white cell count and other clinical parameters.

Click here for the abstract. (Unfortunately the full article needs subscription to JCM)

More and more laboratories are implementing rejection criteria for testing HSV in CSF. The test is often requested before the cell counts and protein are known (ie on the initial request form), and often the CSF parameters are completely normal. 

The researchers put forward a case for only testing HSV PCR if CSF samples had >10 cells/mmor if the sample was from an immunocompromised patient or a child aged <2 years. 

The researcher’s paper looks convincing. However before we accept such criteria we need to reflect on how serious HSV infection of the CNS is, particularly HSV encephalitis. This is not a diagnosis where as a laboratory we can afford to miss a positive diagnosis. The consequences could potentially be catastrophic. See this article  for some sort of an explanation.

Very very occasionally, as previous similar papers have demonstrated (a pseudo meta-analysis if you like..), HSV encephalitis does occur where the child is older than 2 years, doe not have “classical” immunocompromise and CSF parameters are plum normal.

So while we strive for laboratory efficiency (which I strongly support), we need to be very careful in our interpretation of such papers, maintain close co-operation between the clinician and the laboratory, and be prepared to be flexible in our testing protocols for individual patients.


“Lumbar Puncture: Creating headaches not only for the patient…”


25 years ago, examination of cerebro-spinal fluid (CSF) was a fairly straightforward process. You got a cell count, Gram stain, protein and glucose levels. For the really complex patient you might have added a couple more tests like an Indian Ink or a TB culture.

Not any more….

There are now dozens of tests that can potentially be performed on CSF, most of them molecular, and a particular increase in viral PCR, e.g HSV, EBV, CMV, VZV, HHV6, etc. etc. The acronymical list goes on…

If you work in a clinical microbiology lab, then the chances are you will have come across one of the following scenarios:

  • A CSF sample arrives into the lab with a long list of molecular tests requested. The cell counts, protein and glucose are then completely normal on initial testing.
  • A CSF sample arrives into the lab with a long list of molecular tests requested by a doctor who is in the process of studying for post-graduate exams and wants to show off his/her expertise.
  • A CSF sample arrives into the lab with a long list of molecular tests requested. By the time some of these tests are performed, the patient has long recovered and is sitting at home watching ER with a glass of wine in hand.

CSF testing can now be fairly problematic. Cost issues, poor positive predictive value of individual tests, available CSF volume, distributing bits of CSF to different reference labs, lack of clinical details can all contribute to this headache. Not to mention that the results of these tests often have little or no direct effect on the management of the patient. 

I am convinced that there are huge numbers of molecular tests performed on CSF samples which turn out to be completely unnecessary.

Good communication between the laboratory and clinicians is obviously key, but I also believe that more and more the laboratory needs to become the gatekeeper for complex CSF testing, as opposed to being the submissive, and passive recipient of such requests.


I have added a short powerpoint on toxoplasmosis to the website