Measurement of gastric accommodation: a reappraisal of conventional and emerging modalities


Address for Correspondence
Daphne Ang, MBBS, MRCP, Department of Gastroenterology, Changi General Hospital, Singapore 529889.
Tel: +65 68503558; fax: +65 67888833; e-mail:


Impaired gastric accommodation is an important cause of functional dyspepsia. Currently available tests that evaluate gastric accommodation provide relevant physiological information, but they pose technical difficulties and their clinical impact remains controversial. Gastric barostat remains the gold standard, but it is an invasive procedure. In recent years, emerging modalities including single photon emission computed tomography (SPECT), three-dimensional ultrasound and magnetic resonance imaging have been developed to measure gastric volumes and hold promise as alternative methods of assessing gastric accommodation non-invasively. Studies are underway to validate these techniques with recent data proving the performance characteristics of SPECT. The non-invasive nutrient drink test measures satiety scores as a surrogate marker of gastric accommodation and remains controversial. More recently, intragastric monitoring has been proposed as yet another non-invasive modality to assess gastric accommodation. Each of these different modalities brings its associated advantages and disadvantages, as is discussed in this review. Ongoing studies to validate these new techniques are in progress and are likely to lead to further progress in neurogastroenterology.


Functional dyspepsia (FD) is a clinical syndrome that is characterized by persistent or recurrent upper abdominal pain or discomfort in the absence of evidence of organic disease that is likely to explain the symptoms.1 The pathophysiology of FD is heterogeneous, and suggested mechanisms include delayed gastric emptying, hypersensitivity to gastric distension, impaired gastric accommodation, Helicobacter pylori infection, altered duodenal response to acid and lipids, altered duodenojejunal motility and central nervous system dysfunction.2

The accommodation reflex is an important mechanism of normal gastric physiology. It is a vagally mediated reflex that occurs postprandially, resulting in a reduction of gastric tone and an increase in compliance in response to food intake, allowing for an increase in proximal gastric volume without a corresponding rise in pressure.3 Under normal physiological conditions, adaptive relaxation of the proximal stomach is not perceived. Several studies have shown that impaired gastric accommodation to a meal is a major pathophysiological mechanism in FD,4–8 resulting in early satiety and weight loss.4

Identifying patients with impaired gastric accommodation is therefore important in understanding the underlying pathophysiology of FD and has been used in evaluating the effects of various pharmacological agents on proximal gastric function in drug therapy trials.

Gastric Barostat

The gastric barostat was the first method to measure gastric accommodation,3,9,10 with excellent correlation and reproducibility in healthy subjects and dyspeptic patients (R = 0.71 and 0.74, respectively).11 To date, it is generally regarded as the gold standard9,11 to investigate changes in proximal stomach function and for which all other techniques need to be compared for validation.12 The barostat consists of a double-lumen polyvinyl tube with an adherent plastic bag with a maximum capacity of 1000–1200 mls.9,11 The polyvinyl tube is connected to a barostat device which allows isobaric or isovolumic expansion of the gastric balloon. By measuring changes in the volume of air in the compliant bag that is maintained at a constant pressure, changes in gastric tone as well as gastric sensitivity to distension can be studied. Barostat studies have shown that the accommodation response is impaired up to 40% of patients with FD.4 In addition, the potential of various drugs to modulate the accommodation response in acute studies involving healthy volunteers and in dyspeptic subjects has also been studied with the barostat.

However, it should be emphasized that the procedure is invasive, time consuming, and uncomfortable for patients, limiting its feasibility in routine clinical practice. In addition, the presence of a gastric balloon has been shown to interfere with normal gastric physiology, as the direct stimulus imposed by the balloon on the proximal stomach wall may alter intragastric distribution of the meal and result in exaggeration of antral relaxation.13 These limitations have served as an impetus to develop various alternative non-invasive diagnostic techniques which assess change in gastric volume as a reflection of gastric accommodation. In recent years, volume-based methods have been proposed to measure gastric capacity.

Single Photon Emission Computed Tomography

The use of SPECT to quantitate gastric accommodation was first proposed and subsequently validated by the Mayo group.14 This technique combines imaging of the gastric wall using intravenously administered 99mTc pertechnetate (taken up by gastric parietal and mucin-secreting cells) with a non-invasive SPECT gamma camera. Three-dimensional reconstruction of data is performed using commercially available software. The image analysis obtained provides a measure of total gastric volume and assesses gastric response to a meal. Compared to the barostat, SPECT detects changes in postprandial gastric volumes and is less suitable for detecting changes in gastric tone and assessing the gastric sensory response. An early validation study by Bouras et al.15 reported a good correlation (R2 = 0.7) of mean volumes of postprandial to fasting volumes measured concurrently by SPECT and barostat, respectively. However, a subsequent validation study16 where subjects underwent SPECT and barostat on separate occasions showed poor correlation between the two techniques with respect to meal induced accommodation. In addition, meal induced accommodation assessed by SPECT did not differ from ingested meal volumes. The authors concluded that in the absence of a distending pressure, gastric volumes determined by SPECT scanning reflect ingested volumes rather than gastric relaxation and this is likely to account for the poor correlation between the two techniques.16

Disadvantages of SPECT include a high exposure to ionizing radiation, and the need to perform the study in a supine position within a short interval after 99mTc pertechnetate isotope injection. However, the less invasive nature of SPECT and the reduced interference with normal gastric physiology are considerable main advantages of SPECT over the barostat.

In this issue of the journal, Breen et al.17 have further validated the performance characteristics of SPECT based on a retrospective analysis of studies performed in subjects who had undergone prior research studies. The intraday and day-to-day variabilities of gastric accommodation were quantitated by determining the coefficient of variation (COV = 100 x SD/mean, expressed in %) for fasting gastric volumes, mean postprandial volumes and accommodation volumes in 433 subjects. The COV in a subgroup of 47 healthy volunteers who had undergone repeated studies was studied to evaluate the variations within subjects.

The results from this study showed similar inter and intra-individual coefficients of variation (COVINTER and COVINTRA) between and within individuals for fasting, postprandial and accommodation volumes, respectively. Additionally, they did not observe any significant differences by gender or patient subgroups. The findings from Breen et al. serve to establish the reliability of measuring gastric volumes by SPECT in the largest number of patients to date. Indeed, in a recent study in 20 healthy volunteers18 who underwent a total of three SPECT studies at least 3 days apart at the same and different times of the day, the intraclass correlation coefficient values of gastric accommodation for the same time, first different time and second different times studied were 0.681, 0.630, and 0.774, respectively, demonstrating good reproducibility.

Single photon emission computed tomography has demonstrated reliability17 and reproducibility18 and appears to be an emerging tool as a non-invasive modality to measure gastric volumes. However, further validation studies are required before it transcends from a research to clinical setting.

Gastric ultrasound

Conventional ultrasound imaging has been used to estimate gastric volumes by multiplying the proximal gastric area that is measured in a sagittal plane (area from the top fundus margin to 7 cm aborad) with the proximal gastric diameter that is measured in a frontal oblique plane.19 A more sophisticated technique of three-dimensional ultrasound (3DUS)20,21 uses specialized software to calculate images of transverse and longitudinal planes on the basis of the information in the sagittal images and the measured position and orientation of the ultrasound probe. The regions of interest (ROIs) are constructed using the inner layer of the gastric wall, measuring gastric volumes. After delineation of the gastric wall in the sagittal plane, the computer calculates ROIs in the other two planes and creates a three-dimensional image of the stomach. The total gastric volume is calculated using this 3D reconstruction. This sophisticated technique is highly operator dependent and anatomical structures and gas interposition may post technical difficulties. Direct comparison of proximal gastric volumes assessed with the 3DUS and gastric barostat showed a good correlation value of 0.55.22

Both two-and three-dimensional ultrasonography are non-invasive and do not pose a radiation burden. However, anatomical structures and gas interposition may pose technical difficulties in this operator dependent technique. It remains an investigational tool and has not been validated against the barostat.

Magnetic resonance imaging

Similar to ultrasound, magnetic resonance imaging is a non-invasive means of measuring fasting and postprandial gastric volumes through three-dimensional reconstruction of the stomach, and does not pose a radiation burden. Magnetic resonance imaging has mainly been used to assess intragastric meal distribution and gastric emptying. The need to perform the study in a supine position and the long procedural time within the confinements of the MRI scanner are its main limitations. It remains a research tool and has thus far not been well-validated for the assessment of gastric accommodation.23,24

Drinking tests

The sophistication and costs involved in the above mentioned diagnostic tools has largely restricted the use of these modern day equipment to a few specialized centers. Whilst ongoing research studies are being conducted to validate these tools, a number of cheaper and non-invasive modalities have been well-described.

The nutrient drinking test was designed to assess the symptom of early satiety and thus serve as a surrogate marker of proximal stomach function, hence predicting impaired gastric accommodation.4,25 This test is based on the assumption that impaired accommodation results in a limited gastric volume capacity and this is reflected in the maximum tolerated ingested volume. Reproducibility and correlation with barostat measurement of gastric accommodation in a small group of patients was demonstrated.4

Although initially designed to serve as a simple test to differentiate FD patients from normal, it has become a source of both knowledge and contradiction.26 Different tests based on water or nutrient drinks, coupled with different drinking rates have led to contradictory results.4,25,27 A study by Gonenne et al.28 showed a poor correlation between maximum tolerated volume in a nutrient-drinking test and SPECT measurement of gastric volumes and thus caution the utility of using maximal tolerated volumes as a marker of gastric volumes.

Intragastric pressure

More recently, Janssens et al.29 studied intragastric pressure (IGP) recordings in healthy volunteers during nutrient drink ingestion. Intragastric pressure recordings decreased initially and subsequently increased gradually after nutrient-drink infusion. Post hoc analysis showed significant correlations between IGP and satiation scores, indicating that IGP is a determinant of satiation. The findings from this study hold promise to IGP recordings as a minimally invasive and more physiological alternative to gastric barostat measurements.


The difficulties encountered with measuring gastric accommodation present an ongoing challenge in neurogastroenterology research. The limitations of the gastric barostat have paved the way for numerous new developments. A summary of the advantages and disadvantages of the various methods that have been studied to date is illustrated in Table 1. Research is underway to further validate the novel volume-based methods including SPECT, ultrasound and gastric MRI in tertiary centers. Clearly the costs and sophistication of such techniques are the main limiting factors. On the other end of the spectrum, simple to perform nutrient drink tests and a more recently described IGP recordings may prove a useful and feasible alternative in routine clinical practice.

Table 1.   Summary of advantages and disadvantages of various techniques used in evaluating gastric accommodation
 Parameter(s) measuredAdvantageDisadvantagesQuality of evidence
  1. +Investigational (No validation studies available).

  2. ++Limited validation studies.

  3. +++Gold standard.

Barostat studiesProximal gastric volume at fixed pressureGold standard
Measures gastric tone, compliance and sensitivity
UltrasoundGastric volumes, gastric emptyingNon-invasive
No radiation
Low cost
User dependent
Technical difficulties
Magnetic resonance imagingTotal and regional gastric volumeNon-invasive
No radiation
Supine position required during study
High costs
Single photon emission computed tomographyTotal and regional gastric volumeNon-invasiveSupine position required during study
Radiation burden
High costs
Nutrient drinking testsGastric sensation and fillingNon-invasive
Easy to perform, low cost
Limited reproducibility data++
Intragastric pressure monitoringGastric tone, satiation scoresSimple to perform, low costLimited data+