Article Text

Download PDFPDF

Original research
One-stop shop for variceal surveillance: integration of unsedated ultrathin endoscopy into the routine clinic visit
  1. Ali Eqbal1,
  2. Tehara Wickremeratne1,
  3. Stephanie Turner1,
  4. Sarah Elizabeth Higgins1,
  5. Andrew Sloss1,
  6. Jonathan Mitchell1,
  7. James O'Beirne1,2
  1. 1 Gastroenterology & Hepatology, Sunshine Coast University Hospital, Birtinya, Queensland, Australia
  2. 2 Susnhine Coast Health Institute, University of the Sunshine Coast, Maroochydore DC, Queensland, Australia
  1. Correspondence to Dr Ali Eqbal, Gastroenterology & Hepatology, Sunshine Coast University Hospital, Sunshine Coast, QLD 4575, Australia; ali.eqbal{at}


Background The endoscopic appearance of oesophageal varices determines the need for prophylaxis. However, as the point prevalence of varices is low (25%), the majority of surveillance endoscopies are unnecessary and costly. Narrow diameter,ultrathin (UT) endoscopes are more tolerable than conventional upper gastrointestinal (UGI) endoscopes and can be used without sedation. We hypothesised that unsedated UT endoscopy for variceal surveillance could be implemented during the routine outpatient clinic visit allowing accurate diagnosis of varices and the timely provision of prophylaxis.

Methods Patients with cirrhosis awaiting surveillance endoscopy were identified. UT endoscopy was scheduled during routine clinic review at the same time as ultrasound surveillance for hepatocellular carcinoma. UGI endoscopy was performed unsedated using the E.G Scan II disposable endoscope. Varices were graded using the modified Paquet classification. Video recordings of procedures were reviewed by blinded assessors and agreement was assessed using the kappa statistic.

Results 40 patients (80% male) underwent unsedated UT endoscopy. All procedures were successful and tolerated well in 98% of cases. Median procedure time was 2 min (IQR 1–3). Varices were found in 37.5% (17.5% grade 1 and 20% grade 2). Patients with grade 2 varices were prescribed non-selective beta blockers at the clinic appointment. Kappa statistic for the finding of any varices was 0.636 (p=0.001) and 0.8–1.0 for diagnosis of grade 2 varices (p<0.0001).

Conclusions Outpatient unsedated ultrathin endoscopy in patients with cirrhosis is accurate, safe and feasible. This integrative care model is convenient, particularly for regional communities, and is likely to result in significant cost savings associated with variceal surveillance.

  • cirrhosis
  • clinical decision making
  • endoscopy
  • oesophageal varices
  • health economics

Data availability statement

Data are available upon reasonable request. Deidentified participant data are stored securely on a password-protected hospital network computer. Requests for data acquisition can be made to JO’B, ORCID identifier 0000-0003-3400-2816. There are no additional statistical analysis plans.

Statistics from

Key messages

What is already known about this subject?

  • The prevalence of oesophageal varices is low, therefore the majority of surveillance endoscopies are unnecessary, costly and inconvenient.

  • Ultrathin endoscopy is highly sensitive and specific for diagnosing oesophageal varices.

  • Unsedated ultrathin endoscopy is a well-tolerated alternative to conventional endoscopy.

What are the new findings?

  • The use of unsedated ultrathin endoscopy reduces costs and patient inconvenience associated with conventional endoscopy.

How might it have an impact on clinical practice in the foreseeable future?

  • Combining routine outpatient clinical review with ultrathin endoscopic variceal surveillance is a pragmatic healthcare initiative which allowis for the timely institution of primary prophylactic strategies and reduces the frequency of hospital visits.


Variceal haemorrhage can be a life-threatening complication of cirrhosis which carries a significant mortality burden. The mortality rate has been estimated between 17% and 28%1 2 at 6 weeks post the index haemorrhage with recurrent episodes occurring in up to 60% of patients left untreated.3 The prevalence of oesophageal varices is estimated in up to 50% of patients with cirrhosis.4 Accurately identifying patients at risk of this morbid complication is paramount in order to implement preventative medical and interventional therapies. Current guidelines advise screening endoscopy at diagnosis, with surveillance intervals varying based on endoscopic findings and the patient’s compensation status.5 The point prevalence of varices requiring prophylaxis is 25%,6 suggesting that the majority of screening endoscopies provide minimal clinical benefit for the patient. Furthermore, postprocedural complication rates related to endoscopy are not insignificant. A prospective, US study found that procedure-related adverse events occurred in up to 1.07% of procedures. The mean costs associated with postprocedural hospital presentations were 1403 USD per ED visit and 10123 USD per hospitalisation.7 Hours absent from work and personal and procedural costs associated with endoscopy further highlight the need for an efficient and inexpensive alternative.

Although the use of the Baveno VI criteria can reduce the need for screening endoscopy by 40%,8 the need for prophylaxis is determined by the endoscopic appearance of varices, hence the requirement for endoscopic assessment still remains. The advent of narrow diameter or ultrathin (<6 mm) endoscopes offers a convenient alternative to conventional endoscopy.9 Moreover, its diagnostic accuracy and tolerability has been well established in the literature. This was affirmed in a recent systematic review performed by Wickremeratne et al which found that the sensitivity and specificity for diagnosis of varices by ultrathin gastroscopy were 98% (95% CI 93% to 99%) and 96% (95% CI 91% to 99%). respectively.10 The attraction of ultrathin endoscopy is further heightened by its shorter procedure time, cost-effectiveness and the use of disposable scopes which eliminate the need for decontamination processes further reducing costs.11 The use of ultrathin endoscopy also minimises the number of individual outpatient visits in the current COVID-19 climate, thereby reducing the risk of potential virus transmission. Such innovative healthcare services which can operate safely and efficiently despite the restrictions posed by COVID-19 are crucial and may be required for a protracted time until disseminated vaccinations become available.

The aim of our study was to evaluate the integration of unsedated ultrathin endoscopy into clinical practise in a regional, tertiary centre during a patient’s routine outpatient review. We hypothesised that unsedated ultrathin endoscopy for the diagnosis of varices could be implemented successfully in the outpatient clinic setting, thereby allowing us to advise further prophylactic strategies in a time-efficient manner while reducing the need for repeated hospital visits and anaesthetic exposure.


This was a retrospective analysis of a clinical innovation of patients with cirrhosis awaiting screening and surveillance endoscopy. The inclusion criteria included compensated patients with cirrhosis between the ages of 18 and 75 years, not receiving non-selective B-blockade (NSBB) and patients who did not have a contraindication to NSBB. The Baveno VI criteria was applied to include those with a Transient Elastography Score of >20 kPA or platelets<150×109/L. Patients were consented for endocopy at an initial outpatient review. Atelephone preoperative assessment was completed the week prior to the proposed procedure date. The procedure date coincided with their next scheduled outpatient review and hepatocellular carcinoma surveillance scan. Baseline patient characteristics collected included the aetiology of the underlying cirrhosis, biochemical parameters (platelet count, albumin, international normalised ratio, bilirubin) and ultrasound surveillance result.

The EG Scan II (IntroMedic, South Korea) disposable gastroscope system was used in all patients. This is a 6 mm diameter endoscope system with an air channel, which allows for insufflation, and a bending angle of 160° up/160° down allowing for retroflexion within the stomach to assess for potential gastric varices (figure 1). Patients were required to fast for 6 hours prior to the procedure. Topical anaesthesia was applied to the oropharynx with lidocaine spray and patients assumed the left lateral decubitus position. The device was inserted orally and both photo and video documentation were captured. The device was advanced to the antrum of the stomach. Varices were categorised according to the modified Paquet classification system. Video recordings of endoscopic procedures were reviewed by two assessors blinded to the endoscopy reports and clinical details and agreement was assessed using the kappa statistic. Agreement was measured both in terms of presence of any varices and high risk (grade 2 varices.) A kappa statistic of 0.41–0.6 was considered moderate agreement, 0.61–0.8 was considered substantial agreement and 0.81–0.99 was considered almost perfect agreement. All procedures were performed in the outpatient clinic room by the bedside (figure 2). The EG Scan II does possess the capacity to be passed transnasally; however, given the majority of our patients were thrombocytopaenic and the risk of epistaxis with transnasal endoscopy, we elected to adopt a transoral approach.

Figure 1

Demonstration of tip deflection of the EG Scan probe.

Figure 2

Setup of handset (A), processor (B) and ultrathin endoscope (C). The handset has controls for image capture, video acquisition and air insufflation.

Patients with >grade 1 varices were commenced on NSBB on the day of the procedure. A telephone review occurred 2 weeks post procedure for assessment of tolerance and consideration of dose escalation. Patients who were intolerant of NSBB were referred on for conventional endoscopic band ligation (EBL) within 4 weeks.

As this was a retrospective analysis of anonymised patient data, the Research Ethics Committee of the Prince Charles Hospital, Brisbane, Australia, deemed the analysis exempt from formal ethical review and waived the need for individual consent (Project ID 59753, 13 December 2019)


Patient characteristics

To date, 40 patients fulfilled our criteria and underwent unsedated ultrathin endoscopy. The median age was 60 years (IQR 57–67). The underlying aetiology of liver disease was alcohol-related liver disease in 29% patients, hepatitis C virus in 51% patients and non-alcoholic fatty liver disease in 20% patients. The median Model for End-Stage Liver Disease (MELD) was 5 (IQR 3.625–7.000). The majority of the procedures were for screening purposes (80%). The median liver stiffness recorded was 26 kPa (IQR 21–44) and the median platelet count was 132×109/L (IQR 104–153) (table 1).

Table 1

Patient characteristics

Endoscopy results

All procedures were successfully completed (table 2). Mean procedure time was 2 min (IQR 1–3). The prevalence of oesophageal varices was 37.5%. Grade 1 varices were detected in seven patients (17.5%). Grade 2 varices were detected in eight patients (20%). Other findings were portal hypertensive gastropathy in nine patients (22.5%) and two patients each were noted to have oesophagitis or a Schatzki ring. Examples of endoscopic images are shown in figure 3. Table 3 details the endoscopic findings according to the patient’s Child -Pugh class.

Table 2

Ultrathin endoscopy findings

Figure 3

Examples of endoscopic images of oesophageal varices (A), oesophagitis (B) and Schatzki ring (C).

Table 3

Child-Pugh grade and varices grade

Patients with grade 2 varices were prescribed propranolol 20 mg two times per day at the time of review with a subsequent telephone review occurring 2 weeks post initiation for assessment of tolerance and uptitration of the dosage. Three patients were intolerant to propranolol. One patient was successfully switched from propranolol to carvedilol (6.25 mg two times per day). Two patients were referred on for conventional EBL which occurred within 4 weeks. There was 100% concordance for the finding of grade 2 varices at the subsequent endoscopy. Only one patient tolerated the procedure poorly due to a transient episode of tachycardia in the context of known atrial fibrillation. No other significant adverse effects occurred. Kappa statistic for the finding of any varices was 0.636 (p=0.001) and 0.8–1.0 for diagnosis of grade 2 varices (p<0.0001). There were no cases of disagreement in the identification of grade 2 varices. There were four instances of disagreement in the identification of small varices which were thought to be present where compared with the original endoscopist findings of no varices. In clinical practice, this would have led to a further endoscopy for each of these patients in 1 year according to guidelines.


In this study, we have shown that the implementation of unsedated outpatient ultrathin endoscopy for the diagnosis of oesophageal varices in patients with cirrhosis is feasible and pragmatic. Moreover, we have shown that ultrathin endoscopy is accurate in the diagnosis of varices and allows for the timely institution of primary prophylaxis, integrating clinical assessment and endoscopic evaluation in a single combined outpatient review. All procedures were successfully completed with only one patient tolerating the procedure poorly due to a transient episode of tachycardia. Although patient tolerance was interpreted by post procedural observations, the impressive tolerability of narrow diameter endoscopes has been demonstrated in previous publications. Trevisani et al 12 reported successful completion of transoral narrow diameter endoscopy in 97.5% of patients (40/41). Furthermore, multivariate analysis showed that the overall tolerance of transnasal and transoral narrow diameter endoscopes were both significantly higher when compared with conventional diameter endoscopy. There was however no significant difference in tolerance between the transnasal and transoral ultrathin endoscopy groups12

The accuracy for determining grade 2 varices was confirmed with the kappa statistic value of 0.8–1.0 (p<0.0001), allowing for the prompt commencement of primary prophylactic strategies. Although the kappa statistic for the finding of any varices was 0.636 (p=0.001), there has been significant variability in the values reported in the literature. A Danish study reported an average k value of 0.39 in a study on interobserver agreement between 22 endoscopists for the presence any varices.13 A French study reported similar values with a k value of 0.4.14

Although further cost analysis studies for this specific indication are required, ultrathin endoscopy is highly likely to offer an inexpensive alternative to conventional endoscopy. The cost of the endoscope, processor and laptop computer totalled 7755 USD, while the disposable probes cost 38 USD each. Ultrathin endoscopy has been shown to offer a cost-saving benefit of approximately 183 USD per procedure and a reduction in total procedure time including recovery time of 1.5 hours compared with conventional endoscopy.15 Other relevant expenses associated with conventional endoscopy are also eliminated through the use of ultrathin technology, including transport costs, time off work and loss of income for the patient associated with longer procedure times and following anaesthetic. Prescribing NSBB at the time of endoscopy in the outpatient setting further reduces the need for repeated hospital visits, making it an appealing option for patients in regional communities.

Capsule endoscopy has been proposed as another accurate and non-invasive method of diagnosing oesophageal varices. Direct cost-effectiveness analysis is however required to determine which modality is more feasible. The sensitivity for diagnosing oesophageal varices with capsule endoscopy is not sufficient to replace endoscopy as the primary modality for variceal surveillance and screening. Meta-analysis reports sensitivity and specificity rates of 83% (95% CI 0.76 to 0.89) and 85% (95% CI 0.75 to 0.91), respectively.16 Capsule endoscopy may have a role in patients with contraindications to conventional endoscopy.

As a limited endoscopy was performed, only patients who were asymptomatic were included in the study. This could result in pathologies in the distal stomach or duodenum being missed, however the indication of this study was to screen primarily for oesophageal and gastric varices. The bidirectional capability of the EG Scan II allowed for retroflexion and assessment of gastric varices.

In our study, we identified grade 2 varices in 20% of patients, which is a relatively high proportion. This could be accounted for by type 1 error, given the relatively low number of patients in our study. Another consideration is that inadequate air insufflation could have resulted in suboptimal distension of the oesophagus. While air insufflation did appear to be adequate, we cannot exclude that inadequate insufflation may have contributed to the high proportion of grade 2 varices diagnosed. Importantly, however, adverse outcomes in terms of missing varices warranting prophylaxis are avoided.

Given the lack of a therapeutic channel and the inability to perform EBL, only patients who were suitable for NSBB were recruited. NSBB intolerance rates of up to 29% have been reported in previous studies.17 Two patients were intolerant to NSBB therapy and were referred on for traditional EBL, further highlighting the practicality of ultrathin endoscopy. Patients with severe cardiorespiratory disease were excluded from the trial due to the inability to suction out air with the ultrathin endoscope, these patients are unlikely to be suitable for NSBB.

Our healthcare model highlights ultrathin endoscopy as a tolerable, cost-effective alternative to conventional endoscopy for patients and is especially suitable for patients in regional communities where logistical considerations can have an impact on access to medical services.

Data availability statement

Data are available upon reasonable request. Deidentified participant data are stored securely on a password-protected hospital network computer. Requests for data acquisition can be made to JO’B, ORCID identifier 0000-0003-3400-2816. There are no additional statistical analysis plans.

Ethics statements

Patient consent for publication



  • Twitter @drobeirne

  • Contributors AE: lead author of the manuscript. TW and ST: provided the blinded analysis of the endoscopy photos and videos. SEH, AS and JM: provided important intellectual content. JO’B: conceived the study and performed the endoscopies. All authors reviewed and approved the final version of the manuscript.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

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.