Tag Archives: test requesting

“I told you I was sick”

Patient: “I have just come back from Bali and I have a fever and joint pains.”

Doctor: “Not to worry, I think you might have Dengue. I will do a blood test now, and another one in a month’s time to see if I am right.”

Patient: “Just a month, oh, it can’t be too serious then…”

Doctor: “No, it’s just Dengue, and what’s more I have some good news for you, I have just got some results back from the laboratory, that other illness you had 6 weeks ago turned out to be Legionella after all.”

Patient: “I told you I was sick…”

Even in the molecular age, it is surprising how often acute and convalescent serological testing is still utilised in clinical practice. Acute and convalescent serology is still frequently used for Legionella diagnosis, Dengue and other arboviruses, Lyme disease, Leptospirosis, the list goes on…

There are several potential reasons for this:

  • Requestor is used to ordering acute and convalescent serology-This is what the requestor has been doing for the past 30 years. Why change now?
  • Requestor is not aware that other options are available.- The requesting clinician may not be aware of the other options that are available for that particular condition, e.g PCR, antigen testing, etc.
  • Requestor scared/unaware they are allowed to order more “exotic” tests.- I find this a particular challenge when dealing with requests from the community setting. “I didn’t realise we were allowed to request a PCR test for this condition.”
  • Alternatives perceived to be more expensive.- And they are to a certain extent, particularly for arboviruses, but less so for Legionella. But don’t forget that 2 sets of serology along with the time for labour, interpretation and reporting is not an insignificant cost by any means.
  • Laboratory continues to offer acute and convalescent serology.- The laboratory can be guilty too, of not placing appropriate restrictions on acute and convalescent serology. The lab managers and clinical microbiologists of today are often  from an era where acute and convalescent serology was a mainstay of “diagnosis”.

If you were building a microbiology laboratory from scratch, and compiling a schedule or menu of tests, I would like to think acute and convalescent serology would be very low on the priority list.

There may be Public Health reasons why you might want to (retrospectively) confirm the diagnosis with acute and convalescent serology, but if this is the case, then it should be requested (or approved by) by Public Health doctors.

However for the purposes of acutely diagnosing the patient, by the time you get the result, the patient has either recovered, or not recovered…

Better to carefully restrict access to acute and convalescent serology, and “push” requestors into trying to make a real time diagnosis with more modern tests. The patient will thank you for it…



“Hidden Agendas”


In my pursuit of clinical details for all request forms, I have been asked a few times specifically about urine samples. Why bother adding clinical details for a urine sample that is sent to the lab. Is the request not obvious?

However when you think about it and examine a little more closely there are actually many “less obvious” reasons that a urine sample is sent to the lab. Take the following examples below:

Test of Cure: Once symptoms have settled, this is generally not indicated.

As part of an admission screen: Positive predictive value, even in an elderly patient, must be awful.

Prior to orthopaedic (or other) surgery: With the exception of urological surgery, the evidence is now strongly against this.

Wellness check: Just in case…even if you did culture bacteria, treatment would not be indicated in an asymptomatic patient.

Insurance/Immigration requests: as above.

Looking for proteinuria or glucose: In diabetics. Requestor not interested in culture result.

Looking for haematuria: If lab not aware, likely to increase risk of false negative due to sub-optimal testing.

Cloudy/smelly urine in catheter bag: Sending such a sample to the lab is hardly ever indicated.


So as you can see there are actually many reasons that urine samples sent to the lab. There are probably many more which I have not thought of from the top of my head. Many are inappropriate, others are nothing whatsoever to do with infection.

With regards to patients with “symptoms”, there can be a whole range of clinical scenarios, from acute dysuria and frequency to non-specific fatigue. Each set of symptoms will have a different pre-test probability and will therefore potentially affect how the test is reported and interpreted, particularly for borderline results.

And then there all the supporting clinical details which may affect how the test is processed and reported, e.g. pregnancy, immunocompromise, antibiotic allergies, presence of catheter, presence of urinary stents etc etc..

So if someone says to you there is no point in submitting clinical details for urine specimens, I would strongly object to this for all of the reasons above.

In short there is no excuse for not submitting clinical details, not for urine samples or any other samples. Don’t let anyone tell you otherwise….



“The ABC of Hepatitis serology”


I have been thinking about hepatitis serology recently and more particularly, best practice when trying to diagnose a viral hepatitis using serological testing.

There are several viral causes of hepatitis, such as Hep A, B, C, D & E, Epstein Barr virus, cytomegalovirus,  and HIV. (Toxoplasmosis often included in this group as well, even though not a virus!)

….and that is even before you get started on the more esoteric viral causes of hepatitis.

However all these viruses have varying clinical presentations, different incubation periods and particular risk factors. Some are acute and some are chronic.

I therefore find it a little frustrating when the request form asks for “hepatitis serology” without specifying the particular viruses that require testing, along with the clinical rationale.

As a laboratory profession, I don’t think we do ourselves or our patients any favours by accepting non-specific requests such as “hepatitis serology”, “viral hepatitis screen”, “hepatitis screen” etc etc. It is esssentially encouraging poor practice.

“Hepatitis serology” is not really a test request. It is more of a chapter in a textbook…..