Article Text

Download PDFPDF
PTU-076 The “getting fit” project in nottingham: a comparison of haemoglobin levels as measured by oc sensor and hm jack in two week wait referrals
  1. CJ Chapman1,
  2. A Banerjea2,
  3. O Ng2,
  4. A Barke1,
  5. A Duffin1,
  6. K Dale1,
  7. J Allen3,
  8. RF Logan1
  1. 1Bowel Cancer Screening Programme
  2. 2Colorectal Service, Nottingham University Hospitals
  3. 3School of Medicine, University of Nottingham, Nottingham, UK


Introduction NICE guidance (NG12), published in 2015, on indications for urgent referral for possible colorectal cancer (CRC) included the use of faecal occult blood tests (FOB). We report here our early experience of using two different FIT kits in the two week wait (2WW) pathway.

Method In September 2016 FIT was incorporated into Nottingham’s straight to test (STT) pathway. FIT kits from 2 different manufacturers were posted to patients referred on the 2WW pathway without rectal bleeding with instructions for use. Patients were asked to complete both kits and sample faecal material from the same bowel motion. All returned kits were analysed according to manufacturer’s protocols (EIKEN OC-Sensor and Alpha HM Jack) and results expressed in µg Hb/g faeces. Statistical analyses were undertaken on GraphPad Prism 3.0. Data on patients who returned kits between 6/9/16 and 10/2/17 are included in this report.

Results So far 348/441 (79%) individuals within the 2WW pathway have returned samples, with 334 individuals returning two analysable samples. Mean time to return was 7 days, with ~95% returned within 14 days. FIT results for OC sensor and HM Jack are shown in the Table. Linear regression analysis of 310 pairs of samples (with values<40 µg/g) revealed that the OC sensor results were generally 1.54x higher than that for the HM Jack.

14 CRCs were identified in the 2WW cohort. Eleven of these cancers had values>150 µg Hb/g faeces in at least 1 test (10 in both tests). The other 3 (21%) had values between 3 and 25µgHb/g in both tests. These CRCs were all right sided cancers in patients with anaemia at the time of referral.

Using a cut off of 10 µg/g for OC sensor, or 7 µg/g for HM Jack, 2 of these cancers would have been missed, depending on the test used. Although rare, levels of faecal Hb were observed to vary as much as 100x within the same bowel motion with 3 individuals having Hb levels<7 ug/g in 1 FIT and >100 µg/g in the other.

Conclusion Our experience shows that FIT sent by post can be introduced into symptomatic pathways with a high and rapid return rate. With either FIT a value of >150 µg/g has a high predictive value for CRC but even at the proposed 10 µg/g threshold with a single FIT 21% would go undetected.

Finally differences in in levels of Hb within the same individual bowel motion may be due to differences in distribution of blood within a single faecal sample, highlighting the importance of repeat testing for individuals whose symptoms persist.

Disclosure of Interest None Declared

  • Faecal immunochemical test for haemoglobin
  • screening

Statistics from

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.