Tag Archives: ear swabs

“The Requesting Pyramid”

In many laboratories, clinical details on request forms can be structured into a pyramid shape as below.

Let’s take the example of otitis externa.

A good proportion of request forms will be at the bottom of the pyramid, where there are no clinical details present to suggest that otitis externa is the clinical suspicion (as opposed to otitis media, cellulitis of the pinna, or some other condition). Also included in this category are cases where clinical details have been included but are unrelated to sample type, e.g. an ear swab sent with clinical details of “sore throat”. This scenario happens in all sample types with disturbing frequency… (e.g. mid-stream urine sent for a patient with clinical details of chest pain!)

The next level up in the pyramid is where clinical details are present but are insufficient to justify the sample being sent to the laboratory. For example the clinical details might state “Otitis externa“. However most patients with straightforward otitis externa do not need an ear swab sent to the laboratory. Laboratory culture of an ear swab in clinically suspected otitis externa should be the exception as opposed to the rule…

The top level of the pyramid is where clinical details are not only present, but they also give a sound rationale as to why the laboratory is receiving a sample. e.g. “Recalcitrant otitis externa not responding to topical treatment.” or “Diabetic with painful inner ear and fever, clinical suspicion of malignant otitis externa“.

This requesting pyramid applies to most different sample types and clinical scenarios.

At my laboratory, we are doing our utmost to turn this pyramid on it’s head. We have made significant progress to date. In fact our pyramid is starting to look more like a rectangle.

By the end of the year we hope to have removed the base of the pyramid altogether by adopting a policy of having accompanying clinical details pre-requisite for all microbiology tests. I.e. No clinical details, no test.

And that is the way it should be…


“Hearing aids”


I have never found ear swabs a very satisfactory microbiological sample, the outer ear canal usually being filled with all manner of debris and micro-organisms regardless of whether an infection is present or not. (Try processing a swab from your own ear!)

In addition it is often difficult to tell from the request form whether the swab has been taken from the external part of the ear or right inside the canal.

I doubt whether the results of many ear swabs ever affect the clinical outcome of the patient.

Nevertheless I have a few rules for ear swabs, to help both the scientists and myself make the results generated as sensible as possible.

  • If respiratory organisms (H. influenzae, S.pneumoniae, M. catarrhalis) isolated from a toddler ear swab, it is likely to be an otitis media with a perforation. (there is also the possibility that the organism has got into the ear from the mouth via the finger of the toddler). I tend to do susceptibilities on such isolates on the rationale that a systemic antibiotic may be indicated for recalcitrant cases.
  • If Staphylococcus aureus and Streptococcus pyogenes are isolated together from an “ear swab” then the chances are it is an impetiginous lesion. Therefore I do susceptibilities on the staphylococcus (but not the S. pyogenes, see “Excuses”)
  • Rule of three. If three or more organisms are isolated from an ear swab then you are essentially wasting your time identifying every member of the zoo. What help is this going to be to the clinicians. Would simply report as “Mixed flora” or “Mixed Coliforms.”
  • With the exception of the first two points above I would hardly ever perform susceptibilities on isolates from ear swabs. I am not aware of any evidence that doing this improves patient outcomes. The one exception to this would be where there is clinical details of “mastoiditis”. In this clinical scenario, if there is a pure or dominant organism, or a Pseudomonas aeruginosa, it should be worked up.

Ear swabs are over-worked, over processed and over-interpreted in clinical microbiology laboratories all over the world. 

The results of ear swabs can sometimes help with clarifying the aetiology of the ear symptoms, as described above. However they do little to affect the management of the patient, and we should bear this in mind when processing them…..