Few things can be as tedious in this world than trying to diagnose bacterial vaginosis (BV) by traditional methods in the microbiology laboratory.

Sitting at a microscope, trawling through endless rows of Gram stains/wet films looking for ‘Clue Cells’ cannot be good for one’s sanity. My own laboratory receives 60,000 vaginal swabs a year so I appreciate just how soul destroying this work can be.

In addition, ask 10 different scientists to look at 50 Gram stains or wet films for BV and you can be sure you will inevitably get a range of results. Outwith the clear positives and clear negatives, there is a large degree of subjectivity in the laboratory diagnosis. This is of course not the fault of examining scientist, just the nature of the beast. (Have you ever wondered why there are virtually no External QC programmes for the diagnosis of bacterial vaginosis? Because it is very much in the too difficult basket…

There must be a better way….

An increasing number of molecular tests are now available for the diagnosis of BV (usually combined with testing for the other causes of “vaginitis” such as Candida sp. and Trichomonas vaginalis).

These molecular tests are designed to pick up organisms/markers of bacterial vaginosis such as significant levels of Gardnerella vaginalis (low level vaginal colonisation is normal for this bacterium), Atopobium vaginae or Megasphaera-1.

The diagnosis of Bacterial Vaginosis has always been very much a combination of clinical and laboratory findings, e.g. Amsel’s criteria  (or Nugent’s score for the purists). Therefore the molecular result may not always reflect the final diagnosis, but at least with molecular methods, it can be produced in a standardised fashion.

Large diagnostic laboratories are now in the process of switching the processing of stool samples over to molecular methods. I don’t think vaginal swabs (for vaginitis related organisms) will be too far behind…

Large volume (and repetitive) bacteriology work which does not require culture for susceptibility testing, lends itself to automation by molecular methodology.

It is just a matter of time…

Michael