Small-bowel and colon capsules are typically ingested whilst the patient is sitting upright and water is given to aid passage of the capsule through the gastro-oesophageal junction (GEJ). Due to the inherent function of the oesophagus to act as efficient conduit into the stomach, it was necessary to develop a modified ingestion protocol for ECE. Normal solid oesophageal transit time is between 4 and 8 s [ 2 ] and, as such, too rapid to allow enough number of images for meaningful interpretation. Table  4.1 details the original ingestion protocol used in clinical trials and early ECE/ESO1 studies. This protocol resulted in highly variable oesophageal transit times between 6 and 1,200 s [ 1 , 3 ] with a mean oesophageal transit time of 189 s. It should be noted that use of a real-time viewer is essential during ECE studies.

Pendlebury et al. [ 4 ] further modified this protocol in an attempt to enhance imaging of the gastric lumen. An extra stage between stage 5 and 6 was added involving the patient lying flat and then rolling onto their left and right sides for 2 min each. The patient was then allowed to sit upright whilst the test completed. This group also employed the use of prokinetics (Metoclopramide or Domperidone) to increase visualisation of the duodenum during ECE studies, however results were mixed.

Unfortunately effective image visualisation of the oesophagus and GEJ remained an issue prompting Gralnek et al. [ 1 , 5 ] to develop a simplified ingestion protocol (SIP). This protocol has the patient ingesting the capsule whilst lying in the right lateral position. Sips of 15 ml of water are given every 30 s from a syringe (or via a straw) until the capsule enters the stomach (Fig.  4.1 b–d).

SIP improved on the original protocol by significantly extending oesophageal transit time (mean 3 m 45 s vs. 0 min 38 s; p  = 0.0001) and improving visualisation of the GEJ but with no change in the number of GEJ frames captured [ 1 ]. De Jonge et al. (2008) [ 6 ] reported that SIP increased complete visualisation rate of the GEJ in a group of patients with Barrett’s and reflux oesophagitis when compared to the original protocol (93 vs. 68 % p  = 0.04) and improved sensitivity in the diagnosis of these disorders. SIP has now been adopted as the protocol of choice for ECE in clinical centres [ 1 , 7 ].

The most common cause of portal hypertension is liver cirrhosis . It is estimated that cirrhosis accounts for more than 25,000 deaths and 373,000 hospital discharges in 1998 [ 28 ]. The rise in portal pressure is associated with the development of collateral circulation (varices) [ 29 ]. Cirrhotic patients develop varices at a rate of 8 % per year [ 29 ]. The risk of variceal rupture/haemorrhage increases with the size of varices [ 29 ]. The average mortality of the first episode of variceal bleeding is ~50 % [ 30 ]. Since primary prophylactic treatment (band ligation or non-selective beta-blockade) in cirrhotic patients with medium/large size varices reduces the risk of bleeding and mortality, current guidelines stress the importance of variceal screening [ 29 ]. The gold standard in the diagnosis of varices is EGD. Endoscopic screening is recommended for every 1–3 years, depending with the size of varices on index endoscopy [ 3 , 30 ].

In 2006, pilot studies from the States and France [ 3 , 31 , 32 ] reported high sensitivity and specificity (81–100 % and 89–100 %, respectively) in the detection of varices. Since then, an increasing evidence base allowed 2 meta-analyses. In the first meta-analysis (2010), Lu et al. included seven studies [ 33 ], all comparing the yield of ECE and EGD in patients diagnosed or suspected of having oesophageal varices . All were performed with the first ESO model (PillCam®ESO1). In a total of 446 patients, the pooled sensitivity and specificity of ECE for detecting oesophageal varices was 85.8 and 80.5 %, respectively. In subgroup analyses, the pooled sensitivity/specificity in patients undergoing oesophageal varices screening and in those under surveillance was 82.7/54.8 % and 87.3/84.7 %, respectively. A year later, Guturu et al. [ 34 , 35 ] presented a meta-analysis of 9 studies, comprising a total of 631 patients, comparing the use of ECE and EGD in screening and surveillance of oesophageal varices. In 619 patients (they were 12 capsule failures), the pooled sensitivity and specificity of PillCam®ESO 1 (as compared to conventional EGD) was 83 and 85 %, respectively. Pooled likelihood ratios (LR) for the detection of oesophageal varices by ECE were also calculated. It is accepted that LR > 5 and < 0.2 give strong diagnostic evidence. The pooled positive likelihood and negative likelihood ratios were 4.09 and 0.25, respectively. Therefore, the authors concluded that ECE can only be an acceptable alternative in certain situations, but it cannot be recommended as replacement option for conventional EGD (Fig.  4.5 ).

The string/tethered PillCam®SB seem to perform similarly well to its ESO counterpart. In a feasibility study [ 36 ] by Ramirez et al., 30 patients with clinical liver cirrhosis were enroled, 19 for surveillance and 11 for screening purposes. The procedure was safe (no strings were disrupted and no capsule was lost). The mean recording time was 5.8 min, the accuracy 96.7 % with only minimal discomfort. The majority (83.3 %) of patients preferred string-capsule endoscopy to EGD. A follow-up study was published in 2012 [ 37 ]; 100 patients (33 for screening and 67 for surveillance) were enroled. The sensitivity and specificity of tethered capsule endoscopy for clinically significant varices was 82 and 90 %, respectively with a PPV of 84 % and NPV of 89 %.

It should not be forgotten that ECE, despite its reduced battery life, has to traverse the rest of the GI tract to be excreted. Therefore, the same contraindication s that apply for small-bowel capsule endoscopy e.g. known and/or suspected critical bowel stenosis, pregnancy and/or prior complex abdominal surgery, apply for ECE. Moreover, ECE is contraindicated in patients with oesophageal achalasia and/or known oesophageal diverticulae [ 38 – 40 ]. Furthermore, the patients with known oropharyngeal dysfunction (due to high risk of aspiration) should be discouraged. Such group includes often elderly patients with oesophageal dysmotility problems. To date, unlike SBCE , there is no recorded case of aspiration of PillCam®ESO but this may be related to the frequency of its use [ 41 ]. Of course, some of the above contraindications are obsolete, in case the tethered capsule is used.

Minimally invasive alternatives for examining the oesophagus and the stomach include the trans-nasal endoscopy with ultrathin scopes and/or single use endoscopes [ 42 , 43 ] (Fig.  4.6 ). Moreover, the advent of a magnetically controlled capsule will theoretically allow an accurate (wireless) examination of the upper digestive tract [ 44 – 47 ]. Swain et al. [ 44 ] modified a capsule endoscope to include neodymium-iron-boron magnets. The capsule’s magnetic switch was replaced with a thermal one that turns on by hot water. One imager was removed from the PillCam® colon and the available space was used to house the magnets. In the first-in-human study, a handheld external magnet used to manipulate this capsule in the oesophagus of a volunteer. The capsule was swallowed and observed in the oesophagus by using a gastroscope. Capsule images were viewed on a real-time viewer. The capsule was manipulated in the oesophagus for 10 min. Furthermore, Hale et al. [ 45 ] showed that examination of the upper gastrointestinal tract is feasible in a porcine model using a magnet and positional change. Using a handheld magnet, Mirocam Navi (Intromedic Ltd), positional changes and a “real-time” viewer hey showed satisfactory marker recognition (Fig.  4.7 ).