Cost comparison of oral, transnasal and magnet assisted capsule endoscopy in the examination of the upper gastrointestinal tract in patients with dyspepsia

Abstract

Introduction

Conventional oral upper gastrointestinal (GI) endoscopy can obe uncomfortable. By comparison, transnasal endoscopy (TNE) and magnet assisted capsule endoscopy (MACE) have superior tolerability. A cost comparison of competing upper GI endoscopic modalities have yet to be performed.

Methods

We performed a cost comparison study of oral, TNE and MACE by a combination of activity-based costing and averaging of fixed costs over 24 481 upper GI endoscopies performed for dyspepsia over a 10-year period.

Results

On average, 9.4 procedures were performed daily. TNE was cheapest at €125.90 per procedure, costing 30% less than oral endoscopy at €184.10 and threefold cheaper than MACE at €407.10. Flexible endoscope reprocessing cost €53.80. TNE was cheaper than oral endoscopy as sedation was not required. Oral endoscopies have a further rate of infectious complications, estimated to cost €16.20 per oral procedure in inpatient admissions. Oral and TNE equipment are more expensive to purchase and maintain than MACE costing €79 330 and €81 819, respectively compared with MACE at €15 420 per annum. However, capsule endoscopes cost significantly more per procedure at €369.00 than the consumables for flexible endoscopy (per oral €12.30, TNE €5.30).

Conclusions

TNE cost less to perform than conventional per oral endoscopy. The cost of capsule endoscopes will need to be reduced significantly if routine use is to be expected.

What is already known on this topic

• Dyspepsia is a common indication for gastroscopy which comes at great cost.

• Gastroscopy can be poorly tolerated and other minimally invasive technologies like transnasal endoscopy and capsule endoscopy are much better tolerated and preferred by patients

What this study adds

• In a local NHS setting, minimally invasive transnasal endoscopy is 30% cheaper than conventional gastroscopy in the routine examination of patients with dyspepsia.

• Non-invasive magnet assisted capsule endoscopy is at present too costly for routine use.

How this study might affect research, practice or policy

• Adoption of transnasal endoscopy in the routine investigation of the upper GI tract should be encouraged.

• Magnet assisted capsule endoscopy may have a role in the investigation of the upper GI tract in paediatric patients

Introduction

Upper gastrointestinal (GI) endoscopy can be uncomfortable and although the use of sedation may improve patient experience,¹ postprocedure care before and after discharge adds to the complexity of the investigative pathway and patient inconvenience. There is less pharyngeal stimulation with transnasal endoscopy (TNE), which may be better tolerated.² Nonetheless, meaningful diagnoses are made in only one-third of upper GI endoscopies performed for dyspepsia and only 0.4% are malignancy.³

Capsule endoscopy is well tolerated and preferred by patients.^{4 5} Oesophageal capsule endoscopy was first described in 2003 and such capsules, with image sensors on both ends, have a sensitivity of 78% in the detection of Barrett’s oesophagus in meta-analyses.⁶ Examination of the stomach requires luminal distension with water, and a degree of capsule control, with the lumen being capacious and non-uniform in shape. A sequence of simple patient position changes can move a capsule endoscope around the water-filled stomach, but both procedure and subsequent video reporting is time-consuming.⁷

The NaviCam magnet assisted capsule endoscopy (MACE; AnX Robotica, Texas, USA) comprises a robot­-controlled magnet and a capsule endoscope that can be rotated along its axes using a combination of manual joystick manoeuvres and automated programmes. Aided by live views from the capsule endoscope, the capsule is magnetically propelled using a sequence of manoeuvres, which induces the capsule to perform cartwheels over rugal folds (figure 1). MACE had a 90.4% sensitivity in detecting gastric focal lesions by comparison to gastroscopy in 350 patients, with a mean examination time of 26 min.⁵ A mean examination time of 15 min was reported in a later study of 3182 patients with no significant complications.⁸ MACE is non-aerosol generating and has been used to further minimise the risk of COVID-19 transmission by examining patients remotely; the patient follows instructions from an endoscopist in an adjacent control room through audio–visual communication.⁹

Figure 1.

The NaviCam system consists of (A) robot magnet platform with (B) computer console for live view and (C) wireless sensor belt and data recorder. The rotation of the capsule endoscope (D) pitch and (E) yaw, is controlled by moving the left joystick and capsule translocation achieved by movements of the magnet in the (F) cranio caudal (y axis), anteroposterior (z axis) and medio lateral (z axis) directions by controlling the right joystick.

This technology appears to be safe, effective and patient-friendly, and therefore, fulfils many criteria required for a diagnostic test. Investigation in the community, rather than a hospital centre, is a possibility. However, no cost analyses have been performed. The aim of this study is to examine the cost involved in performing conventional per oral upper GI endoscopy, TNE and magnet assisted- capsule endoscopy (MACE) using the NaviCam in patients with dyspepsia.

Methods

The primary outcome is to establish a theoretical cost comparison of endoscopies between the three different modalities based on procurement costs of a local tertiary endoscopy centre (Sheffield, UK). Process maps of diagnostic upper GI endoscopy pathways using conventional per oral, TNE and magnet assisted capsule endoscopy (MACE; AnX Robotica) were developed in consultation between a committee of two endoscopy fellows, two consultant gastroenterologists and two endoscopy nurse specialists based on local patient pathways.

Total costs in Euros (€) were estimated per procedure by a combination of activity-based costing or a ‘bottom up’ aggregation of the costs associated with each procedure, and a ‘top down’ averaging of fixed costs. These are fixed cost that exist irrespective of number of procedures performed per endoscopy session. To allow for comparisons between endoscopic modalities, we estimated the cost of a single dedicated endoscopy room performing only diagnostic upper GI endoscopy for dyspepsia, assuming all examinations were performed sequentially over two sessions a day, 5 days a week, over a 10-year period (2010–2019).

The overall cost per procedure was calculated by considering only costs unique to each endoscopic modality. General overheads and administration, for example, would not be included as it concerns all procedures. The exception was in the consideration of substantial utility costs (electricity and water) in reprocessing an endoscope. Otherwise, equipment, maintenance and consumable costs were based local National Health Service (NHS) trust procurement costs, staff wage based on recent NHS pay scales and costs of inpatient admission for complications related to endoscopic procedures accounted for using the NHS national schedule of reference costs (2018/2019).^{10 11}

Cost of capital equipment was amortised over an estimated life span of 10 years. Costs of staffing and maintaining endoscopy equipment were considered a fixed cost as a minimum number of staffing (as determined by committee) was required to operate a session irrespective of throughput. Finally, endoscopy reprocessing suites cater for a number of specialities (eg, urology) and so unit costs of associated endoscopic reprocessing for flexible endoscopes were considered across all the service users of the local endoscopic reprocessing facility.

It was assumed that oral endoscopy is performed with conscious sedation in 50% (personal communication; K. Siaw, UK National Endoscopy Database), and sedation is not required for transnasal or capsule endoscopies. Rates of 30 days unplanned attendances for infections related to per oral endoscopy in an ambulatory outpatient setting is reported to be 6.9 per 1000 conventional oral endoscopies.¹² After MACE, it may be necessary to perform sampling biopsies, which will require either a further conventional endoscopy. Western studies of endoscopic diagnoses in patients with dyspepsia show Barrett’s oesophagus in 1.5%, gastro-oesophageal cancer in 0.3% and Helicobacter pylori negative gastric intestinal metaplasia in 2%.^{3 13} We assume 2% of dyspeptic patients will require a tissue biopsies for further histological examination.

Results

Average endoscopy capacity

A total of 61 135 outpatient diagnostic gastroscopies were performed of which 24 481 (40%) were performed for dyspepsia between 1 January 2010 and 31 December 2019, averaging to 4.7 upper GI endoscopies for dyspeptic patients per session. Based on these average capacities, the cost of oral, transnasal and MACE are highlighted in table 1. The process map of each endoscopic modality are illustrated in figure 2.

Table 1.

Cost (€) comparison of per oral, transnasal and magnet assisted capsule endoscopy (MACE)

Costs, €	Oral			Transnasal			MACE
	Per procedure *	Per session†	Per annum	Per procedure *	Per session†	Per annum	Per procedure *	Per session†	Per annum
Endoscopy
Equipment	22.90	107.80	56 150	24.00	112.60	58 639	5.10	24.00	12 495
Maintenance	9.50	44.50	23 180	9.40	44.50	23 180	1.20	5.60	2925
	32.40	152.30	79 330	33.40	157.10	81 819	6.30	29.60	15 420
Staff
Procedure room	44.80	210.80		33.40	157.20		31.70	149.20
Recovery	24.60	115.20		–	–		–	–
	69.40	326.00		33.40	157.20		31.70	149.20
Procedure consumables
Endoscopy	9.40			5.30			369.10
Sedation	1.40			–			–
	12.30			5.30			369.10
Complications‡	16.20			–			–
Endoscope reprocessing
Equipment	10.00	–	180 583	10.00	–	180 583	–
Maintenance	8.80	–	158 460	8.80	–	158 460	–
Staff§	10.80	–		10.80	–		–
Consumables, electricity and water	24.20			24.20			–
	53.80			53.80
Total	184.10			125.90			407.10

*Cost per procedure assumes 9.4 upper GI endoscopies performed for dyspepsia per day.

†Cost per endoscopy session assumes 2 sessions per day, 5 days a week (521 sessions) and each session lasting 4 hours.

‡Accounting for relative rate of complication each admission on average costs €2336.

§Assuming 18 000 endoscopes are reprocessed per year requiring two band 3 staff 27 min each to reprocess one endoscope (local audit data 2018/2019).

GI, gastrointestinal .

Figure 2.

Process map of per oral, transnasal and magnetic-assisted capsule endoscopy (MACE).

Fixed costs

The cost of purchasing endoscopy equipment to equip an endoscopy room ready to perform 10 oral or transnasal or MACE procedures per session cost €561 504, €586 397 and €124 956, respectively. Maintenance of equipment cost €23 180 for flexible endoscopes and €2923 for MACE per annum. Assuming 4.7 procedures per session are performed, equipment and maintenance cost €32.40, €33.40 and €6.30 per procedure, respectively.

Labour

In addition to the endoscopist, staffing endoscopy for oral gastroscopy requires a nurse and support worker in both the procedure room and recovery room each. For a transnasal or MACE session, only a nurse or support worker, respectively, is required in the procedure room. Assuming the endoscopist is a nurse practitioner, the cost per 4-hour session for endoscopy staffing is €326.00, €157.20 and €149.20 for peroral, transnasal and MACE, respectively. Therefore, it cost €69.40, €33.40 and €31.70, respectively, in labour per procedure.

Endoscopy consumables

Endoscopy consumables were €12.30, €5.30 and €369.10 for per oral, transnasal and MACE procedures, respectively, assuming 50% of patients undergoing a per oral procedure required conscious sedation, but none with TNE and MACE. The capsule endoscope cost €369.00. The need for oral suction, mouth guard and differences between pre procedure pharyngeal and nasal preparation accounted for the differences in per oral and transnasal consumables.

Endoscope reprocessing

The cost of reprocessing a single flexible endoscope is €53.80 (table 1) including €24.20 in consumables, electricity and water. Locally, a total of 18 000 reprocessing cycles occurred per annum between all GI and non-GI service users. Endoscope reprocessing equipment and maintenance amortised over 10 years cost €180 583 and €158 460 per annum, and therefore, €10.00 and €8.80 per procedure, respectively. Local audit data showed it required two members of staff 27 min each to reprocess one endoscope costing €10.80 in labour per endoscope reprocessed.

Complications

Data from Wang et al suggest the rate of infectious complications 30 days post OGD is 1.08% and 64% require inpatient admission (6.9 inpatient admissions per 1000 elective OGDs). The top three infections accounted for 87% of unplanned admissions 30 days post-OGD and these were (rate per 1000 procedures) respiratory tract infection (1.26), septicaemia (0.65) and GI tract infections (0.35).¹² When patients are admitted to an NHS hospital these complications cost €1849, €6328 and €1764, respectively (NHS schedule of tariff cost codes DZ11V, WJ06C and FD01J).¹⁰ Accounting for proportional rates of complication, each admission on average costs €2336 or €16.20 per oral endoscopy performed. Capsule aspiration is rare (0.1%) and no cases of retention have been reported among a large series of patients undergoing upper GI MACE.^{8 14} There is a risk of epistaxis in 2%–5% of TNE,^{15 16} however, the majority are self-limiting and do not need inpatient admission. Therefore, we assume no complications related to transnasal and MACE require admission.

Overall cost

The cost of each oral, transnasal and MACE was €184.10, €125.90 and €407.10 per procedure (table 1). Assuming 2% of MACE procedures required further oral gastroscopy for biopsies, the overall cost of each MACE is €415.20.

Discussion

We report a theoretical cost comparison between oral, transnasal and MACE. In examining patients with dyspepsia, we show that TNE is the cheapest modality per procedure. It is threefold cheaper per procedure than MACE and a third cheaper than conventional oral endoscopy. Although the initial cost of TNE is dearer than conventional per oral endoscopes by around 3%, it is simpler and cheaper in the long run. The need for sedation and endoscopy recovery in many adds complexity and significant labour cost to oral endoscopy. Furthermore, the cost of complications related to infections may be considered exclusively due to an unprotected airway during a sedated oral procedure with pharyngeal anaesthesia. Capsule endoscopy is simpler still; it obviates the need for endoscope reprocessing, as capsules are for single use only. The cost of the MACE console and data recorders required to start examining patients is fivefold cheaper than a flexible endoscopy before the costs of a decontamination facility is considered. However, the cost of the capsule endoscope, at present, prevents adoption of the technology in the mainstream by comparison to conventional flexible endoscopy.

To our knowledge, this is the first cost comparison of all three upper GI endoscopic modalities. Previous studies have examined only the cost of conventional per oral upper GI endoscopy; in Spain in 2014, the cost of unsedated gastroscopy was estimated at €39 excluding the cost of the endoscopist and endoscope reprocessing.¹⁷ Similar to this study, Crott et al performed a detailed activity-based costing analysis of per oral endoscopies and report that in Canada, procedures range from €70 for unsedated with biopsies to €88 with sedation and biopsies (€94–€118 when adjusted for inflation in 2022).¹⁸ They report the cost (€) of reprocessing an endoscope (inflation adjusted) to be €9 (€12) and by comparison to our study, do not include the endoscopists labour. Ofstead et al performed a detailed micro costing of endoscope reprocessing from the USA in 2017 and showed that reprocessing costs can in fact be more involved than commonly believed.¹⁹ Reprocessing one endoscope took 76 min and between €98 and €241 per endoscope to reprocess. The majority of this cost, however, was between €54 and €110 in endoscope repairs after leak testing failure. In this study and locally, endoscope repairs are covered by maintenance contracts and therefore, including such contracts, we report the cost of reprocessing each endoscope as €54, within the range reported by Ofstead et al .

Costing studies, as performed here, attempt to consider differences in costs unique to each, but do not encompass the full economic cost. For example, the opportunity costs to a patient opting for sedation, or the differences in the environmental impact of conventional flexible endoscopy and MACE, for instance, the carbon footprint of the production and transport of consumables to endoscopy. As part of a detailed itemisation of consumables in this study, we observed a total of 20 disposable items for per oral endoscopy (31 with sedation), 15 for TNE and an additional 16 items for each endoscope reprocessed. In contrast, only a pair of gloves, a biodegradable straw and a capsule endoscope is required for the MACE. A study from an Italian hospital showed that digestive endoscopy was the third highest producer of healthcare waste, amounting to 3 kg of disposable waste per bed per day with the cost of waste management amounted to €2 per kg of waste.²⁰ Further to differences in consumables and labour costs, we found that the cost of water and electricity during each wash cycle was substantial and accounted for 10%–16% of the total cost of each flexible endoscopy at €16 per cycle.

On the other hand, single-use capsule endoscopes may be considered a step backwards in achieving greener endoscopy.²¹ The environmental impact of releasing capsule endoscopes into the environment has yet to be examined. In modern sewage systems, capsule endoscopes are filtered out during initial processing of sewage and taken to landfill. However, the risks of eventual capsule dissolution and the impact of environmental contamination by batteries and electronic components are unknown.²² Furthermore, the carbon footprint of mining precious materials required for electronics and batteries of capsule endoscopes are also largely unknown. When costs of capsule endoscopes allow its use to be scaled up in a meaningful way, the impact of these needs to be examined in more detail. In particular, the idea of recycling capsule materials such as components of electronics and batteries (as oppose to reusing capsules) could address both issues, but would require capsule retrieval.

There are a number of limitations to MACE. MACE may incompletely examine the oesophagus and a further degree of control may be achieved by attaching a detachable string.²³ However, where patients have dysphagia, capsule endoscopy is less useful. On the other hand, after upper GI MACE has been completed, the capsule endoscope can proceed to examine the rest of the small bowel and has been shown to improve diagnostic yield in patients with recurrent anaemia and suspected upper GI bleeding.^{4 24}

This study costed for a nurse practitioner endoscopist. By comparison to the West, in China, upper GI MACE is more common than small bowel capsule investigations and mostly performed by a trained technician, which could reduce NHS staffing cost by a further 45% compared with a nurse practitioner. There will likely be differences between learning curves, and therefore, the cost of training endoscopist in MACE may be cheaper than conventional endoscopy. A fully automated MACE procedure was recently described and may further reduce the cost of MACE by leveraging artificial intelligence.²⁵ Regardless, the cost of capsule endoscopes will need to be reduced significantly before MACE can be adopted in routine practice. At present though, there may be certain situations where MACE may be cost-effective, for example, among paediatric patients where upper GI endoscopy is often more involved and current guidance recommend examination under general anaesthesia or deep sedation.²⁶

Conclusion

TNE was the least costly at €125.90 per procedure, costing 30% less than per oral endoscopy at €184.10 and threefold cheaper than MACE at €407.10. Flexible endoscope equipment was more expensive to purchase and maintain than MACE costing €32.40 and €6.30, respectively, per procedure over a 10-year life cycle. Capsule endoscopes, however, cost significantly more at €369.10 per procedure than the consumables for flexible endoscopy. The cost of capsule endoscopes will need to be reduced significantly if routine use is to be expected.

Data availability statement

Data are available on reasonable request.

Ethics statements

Patient consent for publication

Not applicable.