Article Text

PDF
PWE-066 Pre-analytical and analytical variability in laboratory practice could affect calprotectin accuracy between uk sites; the results of a national laboratory questionnaire
  1. MJ Brookes1,
  2. R Gama2,
  3. J French3,
  4. C Ford2,
  5. S Whitehead2,
  6. J Folkes2,
  7. B Hayee4,
  8. A Robins3,
  9. R Logan4,
  10. H Steed1
  1. 1Gastroenterology Unit
  2. 2Clinical Chemistry, Royal Wolverhampton NHS Trust, Wolverhampton
  3. 3Birmingham Quality, Birmingham
  4. 4Gastroenterology Deptartment, King–s College Hospital NHS Foundation Trust, London, UK

Abstract

Introduction Faecal calprotectin (f-Cp), a calcium binding protein, is a product of mucosal granulocyte infiltration in Inflammatory Bowel Disease (IBD). In 2003 a national survey identified that 0.5% of UK laboratories offered an assay to measure f-Cp, but variability in practice was demonstrated between these laboratories. In 2013 NICE supported the use of f-Cp in the UK to diagnose IBD. However we have previously reported inter-assay variability, which may limit the efficacy of this test.

Method The aim of this study was to evaluate variability in pre-analytical and analytical evaluation of f-Cp by laboratories across the UK. We designed and undertook a questionnaire evaluation of all laboratories in the UK who currently perform an f-Cp assay and are members of the national quality assessment programme.

Results The questionnaire was sent to 59 laboratories in the UK and responses were received from 69% (n = 41). The following commercial assays were used; Buhlmann ELISA 36.6% (n = 15); Thermo Phadia EliA ELISA 22.0% (n = 9); Quantum Blue 17.1% (n = 7); Immundiagnostik ELISA 12.2% (n = 5). The assay reporting range was limited by the laboratory in 92.7% of cases (n = 38); the upper limit being 300ug/g in 12.2% (n = 5) and 600ug/g in 24.4% (n = 10). 29.3% (n = 12) run the assay manually, whereas the remainder use automated platforms. 82.9% (n = 34) laboratories stored stool specimens before analysis whereas 17.1% extracted calprotectin immediately and stored extract. Those storing stool did so as follows: 55.9% (n = 19) at 4–8ºC; 38.2% (n = 13) at -20˚C; and 2.9% (n = 1) at -40˚C. Stool or extract samples were stored by laboratories for the following durations: less than 1 week by 65.9% (n = 27); 1 week only by 19.5% (n = 8); 2 weeks by 7.3% (n = 3).

Stool extraction was performed either by manual weighing (26.8%; n = 11) or by a variety of extraction devices. For watery samples the assay would not be performed by 12.2% (n = 5) of sites. Samples containing blood were rejected by 12.2% (n = 5) of sites. 39.0% (n = 16) of laboratories would provide an amended clinical report for processed watery specimens, and 42.5% (n = 17) would provide the same if contaminated with blood.

Conclusion Laboratory practices vary widely and there is little standardisation of pre-analytical and analytical stages in measuring f-Cp. In addition the use of different assays by sites across the UK could interfere with the accuracy of calprotectin reporting. These practice variances are likely to further confound f-Cp accuracy which we have previous reported as being assay dependent.

Disclosure of interest M. Brookes Grant/Research Support from: Vifor international, Speaker Bureau of: Warner Chilcott, R. Gama: None Declared, J. French: None Declared, C. Ford: None Declared, S. Whitehead: None Declared, J. Folkes: None Declared, B. Hayee: None Declared, A. Robins: None Declared, R. Logan: None Declared, H. Steed: None Declared.

Statistics from Altmetric.com

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.