The purpose of this study is to identify the prevalence, nature and location of lesions in the GI tract that may contribute to iron deficiency anaemia and compare diagnostic yied of the upper GI magnetic controlled capsule endoscopy with conventional gastroscopy.

Iron deficiency anaemia (IDA) affects 2-5% of men and post-menopausal women (1). It is thought to occur as a consequence of gastrointestinal blood loss in the majority of cases. Studies suggest that gastroscopy identifies a possible cause in 25-58 % and colonoscopy in 25-33 % of cases. Because pathologies in both upper and lower gastrointestinal tract occur simultaneously in up to 26% of cases, current United Kingdom guidelines recommend both gastroscopy and colonoscopy for patients with IDA.

Wireless capsule endoscopy (CE) is a non-invasive form of endoscopy using a swallowable pill camera which produces images which can be viewed as a video. It is used routinely in clinical practice to examine the small bowel and colon and a role in upper GI investigation is emerging. We have experience of upper GI capsule endoscopy in over 100 patients who have declined conventional gastroscopy using a protocol involving positional change to move a capsule around a water-filled stomach (Ching et al., submitted for publication. It is much better tolerated than conventional endoscopy which requires oral or anal intubation, often following the administration of intravenous sedation and analgesia and incurs a small risk of perforation.Pathology in the small bowel was historically considered to account for only 5% of all gastrointestinal causes of anaemia . Consequently, current guidelines recommend small bowel capsule endoscopy only when IDA has recurred after treatment. However, it is accepted that as many as 30 % of patients with IDA undergoing bidirectional endoscopy have no significant abnormality identified, raising the possibility that the cause is located in the small bowel. The studies which identified that only 5% of IDA was due to small bowel pathology used radiological methods of small bowel imaging, before the advent of CE. Meta-analyses now show significantly better diagnostic yields of CE compared to small bowel radiology in patients with IDA (42% and 6%, respectively).

Although mostly performed in patients with recurrent or refractory (as opposed to first presentation of) anaemia, CE studies show a diagnostic yield of small bowel pathology in 66% and a tumour detection rate of as much as 10% . Even in patients of less than 50 years of age, 5% of patients are found to have tumours. Given the uncertainties about which pathologies cause anaemia, the failure to identify a cause using conventional bidirectional endoscopy in 30% and the availability of a highly sensitive, well tolerated small bowel investigative tool, our primary aim is to determine the incidence, nature and location of pathology in the gastrointestinal tract by performing small bowel capsule endoscopy in patients referred for gastroscopy and colonoscopy for the investigation of IDA.

Prior to passage through the pylorus and small bowel, capsules can now be moved around the stomach using a joystick-controlled robot magnet (Ankon Technologies, Shanghai, China). A multicentre study using this device showed a 90% sensitivity in the detection of gastric focal lesions compared to gastroscopy, irrespective of size or location of the lesion. We have also demonstrated that the diagnostic ability of capsules moved around the stomach either using simple patient positional change or external handheld magnets is comparable to gastroscopy. Patient tolerance significantly favoured CE in these studies and no patient suffered adverse effects. The diagnostic yield using magnetically assisted CE (MACE) of the upper gastrointestinal tract will be compared with gastroscopy as a secondary outcome measure in this study of patients with IDA.

Overall, this study aims to report on the prevalence of lesions in entire gastrointestinal tract by endoscopy in patients with IDA. This is novel as there has been no such study reporting pan enteric pathology by endoscopy in unselected patients with iron deficiency since the advent of CE in 2000. This is important because uncertainty about the likelihood that certain upper gastrointestinal and colonic pathologies, such as oesophagitis, gastritis, diverticulosis and colonic polyps, are the cause of IDA, is widely acknowledged. In up to 25% of patients synchronous upper and lower GI pathologies are found on gastroscopy and colonoscopy, but we are unsure of the rates and significance of synchronous small bowel pathologies in those deemed to have a cause found on gastroscopy or colonoscopy. If there are significant synchronous pathologies in the small bowel it may be that small bowel CE should become part of first line investigation. Furthermore, should MACE prove to be sensitive in upper GI pathology detection when compared to gastroscopy, it may have a role in investigating the stomach and small bowel simultaneously.

Inclusion Criteria:

• Male and female patients aged 18 years and over and up to but not exceeding 80 years
• Patients presenting with IDA whom require gastroscopy and colonoscopy as per national guidelines (1)

Exclusion Criteria:

• Patients who have contraindications to gastroscopy or colonoscopy
• Patients under the age of 18 years
• Patients over the age of 80 years
• Active vomiting
• Patients with a permanent pacemaker, implantable cardioverter-defibrillator or REVEAL device
• Patients with any electronic/magnetic/mechanically controlled devices e.g. sacral nerve stimulators, bladder stimulators
• Patients with dysphagia, odynophagia or known swallowing disorder
• Patients with known Zenker's diverticulum
• Patients with suspected bowel obstruction or bowel perforation
• Patients with prior bowel obstruction
• Patients with gastroparesis or known gastric outlet obstruction
• Patients with known Crohn's disease
• Patients who are taking daily non-steroidal anti-inflammatory drugs (excluding prophylactic doses of aspirin) for more than six months
• Patients who have received abdominopelvic radiotherapy treatment
• Patients with a history of GI tract surgery (Billroth I, Billroth II, Oesophagectomy, gastrectomy or bariatric procedure)
• Patients that are pregnant or lactating
• Patients with altered mental status that would limit their ability to swallow
• Patients with allergy to conscious sedation, polyethylene glycol or metoclopramide
• Patients unwilling to swallow the capsule
• Patients with known dementia affecting ability to consent
• Patients who are unable to understand or speak English
• Patients unable to provide written informed consent

There is no plan to share individual participant data with other researchers