Low faecal haemoglobin concentration potentially rules out significant colorectal disease

Authors


Correspondence to: Prof. Callum G Fraser, Centre for Research into Cancer Prevention and Screening, Ninewells Hospital and Medical School, Dundee DD1 9SY, UK.

E-mail: callum.fraser@nhs.net

Abstract

Aim

The study aimed to determine whether faecal haemoglobin (Hb) concentration can assist in deciding who with lower abdominal symptoms will benefit from endoscopy.

Method

Faecal Hb concentrations were measured on single samples from 280 patients referred for lower gastrointestinal tract endoscopy from primary care in NHS Tayside who completed a faecal immunochemical test (FIT) for Hb and underwent subsequent endoscopy.

Results

Among 739 invited patients, FIT and endoscopy were completed by 280 (median age 63 (18–84) years; 59.6% women), with a median time between FIT and endoscopy of 9 days. Six (2.1%) participants had cancer, 23 (8.2%) had high-risk adenoma (HRA) (more than three adenomas or any > 1 cm), 31 (11.1%) low-risk adenoma (LRA) and 26 (9.3%) inflammatory bowel disease (IBD) as the most serious diagnosis. Those with cancer had a median faecal Hb of > 1000 ng Hb/ml buffer. Those with cancer + HRA + IBD had a median faecal Hb concentration of 75 ng Hb/ml buffer (95% CI 18–204), which was significantly higher than that of all remaining participants without significant colorectal disease (< 0.0001). Using a cut-off faecal Hb concentration of 50 ng Hb/ml buffer, negative predictive values of 100.0%, 94.4%, 93.4% and 93.9% were found for cancer, HRA, LRA and IBD. Patients with reasons for referral other than rectal bleeding and family history did not have high faecal Hb concentrations.

Conclusion

Faecal Hb concentration measurements have considerable potential to contribute to reducing unnecessary endoscopy for the majority of symptomatic patients.

Ancillary