Full text loading...
In-house versus external testing
/content/chapter/10.22233/9781910443255.chap1
In-house versus external testing
- Author: Graham Bilbrough
- From: BSAVA Manual of Canine and Feline Clinical Pathology
- Item: Chapter 1, pp 1 - 10
- DOI: 10.22233/9781910443255.1
- Copyright: © 2016 British Small Animal Veterinary Association
- Publication Date: March 2016
Abstract
This chapter discusses the logical approach the veterinary surgeon should take when deciding whether to perform diagnostic testing in-house or by submission to a reference laboratory. Topics include where to test, selecting a reference laboratory, bringing it all together: combining ‘in’ and ‘out’ and the veterinary laboratory and safety procedures.
/content/chapter/10.22233/9781910443255.chap1
Figures
/content/figure/10.22233/9781910443255.chap1.ch1fig3
1.3
(a) Before purchasing an analyser for total T4, a practice performed a small comparison study using 13 feline samples analysed in clinic and at the reference laboratory. Some samples were drawn from cats with suspected hyperthyroidism and some from cats receiving medication for confirmed hyperthyroidism. There was excellent correlation (R
2 >0.9). (b) One sample was analysed six times on both analysers, with six separate aliquots being sent to the reference laboratory. The in-clinic assay was much less precise (see Chapter 2 for a detailed discussion of the coefficient of variation). This does not make the in-clinic analyser unacceptable; however, greater care must be taken when determining whether a trend is present. For example, it would be tempting to conclude that a cat receiving medication, in which the reported T4 concentration over time went from 79 to 53 nmol/l, was responding to the therapy. However, this result could be due to the relatively imprecise nature of the assay. © 2016 British Small Animal Veterinary Association
10.22233/9781910443255/fig1_03_thumb.gif
10.22233/9781910443255/fig1_03.png
1.3
(a) Before purchasing an analyser for total T4, a practice performed a small comparison study using 13 feline samples analysed in clinic and at the reference laboratory. Some samples were drawn from cats with suspected hyperthyroidism and some from cats receiving medication for confirmed hyperthyroidism. There was excellent correlation (R
2 >0.9). (b) One sample was analysed six times on both analysers, with six separate aliquots being sent to the reference laboratory. The in-clinic assay was much less precise (see Chapter 2 for a detailed discussion of the coefficient of variation). This does not make the in-clinic analyser unacceptable; however, greater care must be taken when determining whether a trend is present. For example, it would be tempting to conclude that a cat receiving medication, in which the reported T4 concentration over time went from 79 to 53 nmol/l, was responding to the therapy. However, this result could be due to the relatively imprecise nature of the assay.
/content/figure/10.22233/9781910443255.chap1.ch1fig4
1.4
Chemistry report from an in-clinic chemistry analyser. The presence of haemolysis and lipaemia is clearly indicated and the analytes that are severely affected are suppressed. The user of the in-clinic chemistry analyser should be aware of the influence of these interfering substances on their machine. © 2016 British Small Animal Veterinary Association
10.22233/9781910443255/fig1_04_thumb.gif
10.22233/9781910443255/fig1_04.png
1.4
Chemistry report from an in-clinic chemistry analyser. The presence of haemolysis and lipaemia is clearly indicated and the analytes that are severely affected are suppressed. The user of the in-clinic chemistry analyser should be aware of the influence of these interfering substances on their machine.
/content/figure/10.22233/9781910443255.chap1.ch1fig5
1.5
Most in-clinic analysers provide a graphical display, in addition to the numerical data, to provide additional information about the sample. It is important to appreciate these to gain full understanding. In this example of a ‘white blood cell (WBC) dot plot’ from an in-clinic haematology analyser, each dot represents a cell and each ‘cloud’ represents a subtype of white blood cell. The clouds are not cleanly separated, suggesting that a manual differential would be helpful. In this case, an immature population of neutrophils (e.g. ‘bands’) causes the neutrophil cloud (lilac) to extend further along the vertical axis, spreading over the lymphocyte and monocyte populations. © 2016 British Small Animal Veterinary Association
10.22233/9781910443255/fig1_05_thumb.gif
10.22233/9781910443255/fig1_05.png
1.5
Most in-clinic analysers provide a graphical display, in addition to the numerical data, to provide additional information about the sample. It is important to appreciate these to gain full understanding. In this example of a ‘white blood cell (WBC) dot plot’ from an in-clinic haematology analyser, each dot represents a cell and each ‘cloud’ represents a subtype of white blood cell. The clouds are not cleanly separated, suggesting that a manual differential would be helpful. In this case, an immature population of neutrophils (e.g. ‘bands’) causes the neutrophil cloud (lilac) to extend further along the vertical axis, spreading over the lymphocyte and monocyte populations.
/content/figure/10.22233/9781910443255.chap1.ch1fig7
1.7
Integration of reference laboratory and in-clinic results in a combined report allows for convenient review of all of the patient’s data. However, for some parameters, if the testing is not performed consistently (e.g. on the same analyser and with the same sample handling) it may not be appropriate to draw conclusions from the trend in the results. Likewise, the user must understand the expected biological and analytical variation before deciding whether any change is clinically significant (see Chapter 2). © 2016 British Small Animal Veterinary Association
10.22233/9781910443255/fig1_07_thumb.gif
10.22233/9781910443255/fig1_07.png
1.7
Integration of reference laboratory and in-clinic results in a combined report allows for convenient review of all of the patient’s data. However, for some parameters, if the testing is not performed consistently (e.g. on the same analyser and with the same sample handling) it may not be appropriate to draw conclusions from the trend in the results. Likewise, the user must understand the expected biological and analytical variation before deciding whether any change is clinically significant (see Chapter 2).
