1.2 Impact of sample quality on testing and reporting of results

Not all samples received by a laboratory will be of high quality, that is, contaminant free and in sufficient quantity. Prior to starting the testing process, laboratory staff should be aware of potential quality issues relating to specific sample types that could lead to false results.

Blood and CSF are normally sterile; therefore any microorganisms found in samples taken from these parts of the body are likely to be the cause of an infection. However, if there is contamination of the sample during the process of collection, commensal bacteria may be isolated as a contaminant. Similarly, urine samples should also be sterile but are easily contaminated at the time of collection. These contaminants can grow in the sample, making interpretation of the culture results difficult.

It is important to recognise that when samples are taken from most other sites in the body the environment is not sterile. This means that a mix of the normal commensal flora, colonising organisms and contaminants will all be present. For example, normal flora are likely to be found on superficial swabs, genital swabs and in respiratory specimens.

Clinical staff can take several measures to improve the quality of the specimens they send to laboratories, either by reducing contamination and/or increasing the yield of true pathogens. For example, sterile containers and equipment should be used for sample collection, to avoid the introduction of contaminating organisms.

In general, ‘the larger the better’ is a good rule to follow for specimens. Surface swabs just dipped in material make much poorer specimens than a larger, deeper sample because there is less chance of finding the infectious agent amongst the contaminants. A spoonful of faeces is better than a swab dipped in faeces, which is likely to pick up many orders of magnitude more commensals than pathogens. This is likely to make subsequent isolation of the pathogen more difficult.

Some types of sample may require special treatment or have other considerations.

• Sputum samples, for instance are very useful for diagnosing TB and are frequently sent to diagnose other respiratory tract infections. However, the results have to be considered carefully as the sample is likely to be contaminated with upper respiratory tract flora. Identifying all the organisms in a sputum sample will not help the diagnosis and takes up a lot of laboratory time and resources.
• Blood cultures, which are important in diagnosing severe and/or bloodstream infections, are another sample type that is easily contaminated. Even though the blood is taken carefully to minimise contamination and inoculated directly into blood culture media by the person taking the samples, contaminants will still sometimes be found. Prof Koch’s guide to perfect blood cultures [Tip: hold Ctrl and click a link to open it in a new tab. (Hide tip)] (2010), shows how careful decontamination of the skin and bottles, plus aseptic technique can minimise sample contamination.
• Urine samples need to be collected carefully in midstream. The sample should be refrigerated if it cannot be processed immediately. Boric acid can be used as a preservative if this is not possible. Microscopy helps determine whether a urinary tract infection (UTI) is present before culturing by looking for pus cells which indicate inflammation/infection.

Laboratory staff also need to take care not to introduce contaminants into cultures. Good laboratory practice should be followed at all times.

After testing it is important to avoid reporting any additional organisms and instead focus only on the causative pathogen, if found. Correct reporting is needed to guide accurate treatment decisions and for surveillance purposes.