lable at ScienceDirect

Clinical Nutrition 34 (2015) 557e564

Contents lists avai

Clinical Nutrition

journal homepage: ht tp: / /www.elsevier .com/locate/c lnu

Review

Measurement of gastric emptying in the critically ill

Palash Kar a, b, *, Karen L. Jones c, d, Michael Horowitz c, d, Marianne J. Chapman a, b, c,Adam M. Deane a, b, c

a Discipline of Acute Care Medicine, University of Adelaide, South Australia, Australiab Intensive Care Unit, Royal Adelaide Hospital, South Australia, Australiac Centre for Research Excellence, University of Adelaide, South Australia, Australiad Discipline of Medicine, University of Adelaide, South Australia, Australia

a r t i c l e i n f o

Article history:Received 25 July 2014Accepted 5 November 2014

Keywords:Gastric emptyingIntensive careCritically illEnteral nutrition

* Corresponding author. Intensive Care Unit, LevelNorth Terrace, Adelaide 5000, South Australia, Austral401 878 997.

E-mail address: p_kar@hotmail.com (P. Kar).

http://dx.doi.org/10.1016/j.clnu.2014.11.0030261-5614/© 2014 Elsevier Ltd and European Society

s u m m a r y

Background & aims: Enteral nutrition is important in critically ill patients and is usually administered viaa nasogastric tube. As gastric emptying is frequently delayed, and this compromises the delivery ofnutrient, it is important that the emptying rate can be quantified.Methods: A comprehensive search of MEDLINE/PubMed, of English articles, from inception to 1 July 2014.References of included manuscripts were also examined for additional studies.Results: A number of methods are available to measure gastric emptying and these broadly can becategorised as direct- or indirect-test and surrogate assessments. Direct tests necessitate visualisation ofthe stomach contents during emptying and are unaffected by liver or kidney metabolism. The mostfrequently used direct modality is scintigraphy, which remains the ‘gold standard’. Indirect tests use amarker that is absorbed in the proximal small intestine, so that measurements of the marker, or itsmetabolite measured in plasma or breath, correlates with gastric emptying. These tests include drug andcarbohydrate absorption and isotope breath tests. Gastric residual volumes (GRVs) are used frequently toquantify gastric emptying during nasogastric feeding, but these measurements may be inaccurate andshould be regarded as a surrogate measurement. While the inherent limitations of GRVs make them lesssuitable for research purposes they are often the only technique that is available for clinicians at thebedside.Conclusions: Each of the available techniques has its strength and limitations. Accordingly, the choice ofgastric emptying test is dictated by the particular requirement(s) and expertise of the investigator orclinician.

© 2014 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

1. Introduction

Enteral nutrition is part of standard care for critically ill patientsand is most commonly delivered via the gastric route [1].Frequently, however, nutritional targets are not achieved via thegastric route, particularly because of critical illness inducedgastrointestinal dysmotility, which leads to slow gastric emptying[2]. While the prevalence and magnitude of delayed gastricemptying in the critically ill are inconsistently reported [3e5],possibly because these variables are dependent on the precision ofthe methodology used to measure gastric emptying, as well as the

4, Royal Adelaide Hospital,ia. Tel.: þ61 8 8222 4624, þ61

for Clinical Nutrition and Metabol

definitions of critical illness and/or delayed gastric emptying, thereis no doubt that markedly delayed gastric emptying occursfrequently [4].

It is now also recognised that the impact of delayed emptying,and hence the potential indications for its measurement, arebroader then had been appreciated. For example, in the critically illpatient, markedly slow gastric emptying may be associated withincreased risk of gastro-oesophageal reflux and pulmonary aspi-ration [6]. The rate of gastric emptying is a major determinant ofpostprandial glycaemia and, in patients treated with insulin, majorchanges in gastric emptying rate may impair the coordination be-tween nutrient absorption and insulin availability predisposing toglycaemic variability [7].

Gastroparesis is a relatively frequent condition in ambulantpopulations with epidemiological data estimating the incidencebetween 2.4 and 9.8 per 100,000 person-years [8]. Delayed gastric

ism. All rights reserved.

Table 1Methods of gastric emptying measurement in the critically ill.

Type Method

Direct ScintigraphyIndirect Drug absorption

Carbohydrate absorption (3-O-Methylglucose; 3-OMG)Isotope breath test

Other UltrasoundMRI

Surrogate Gastric residual volumes (GRVs)

P. Kar et al. / Clinical Nutrition 34 (2015) 557e564558

emptying is particularly prevalent in the critically ill and has beenreported to occur in up to one-half of all mechanically ventilatedpatients, which is similar to the prevalence reported in ambulantpatients with long-standing type 1 or type 2 diabetes [9,10].

In the research setting, the measurement technique must beprecise, whereas in routine clinical situations clinicians may bewilling to trade off accuracy for other factors, such as cost, conve-nience, and invasiveness, so that they can have a ‘point-of-care’ test.Comprehensive evaluation of gastric motility requires the mea-surement of multiple parameters. These include intraluminal vol-ume, flow and pressure, wall motion, and electrical activity. Nosingle technique can measure all of these concurrently. Conse-quently, in research studies a complete assessment of gastricfunction can only be obtained by the simultaneous use of differenttechniques. Additional information specific to measurement ofgastric emptying in the ambulant setting is covered elsewhere[11e14].

Gastric emptying is determined by a number of intraluminal andextraluminal factors, which can be endogenous or exogenous, andmediate their effects via neural and/or hormonal pathways. Theintraluminal factors that modulate gastric emptying include mealcompositione caloric load, volume, temperature and nutrient type,the osmolarity of small intestinal contents and length and region ofsmall intestine exposed to nutriente and systemic (or extraluminalfactors) such as glycaemia, posture, pain, gender and age [14].

The Intensive Care Unit presents some unique challenges whenmeasuring physiological derangements. Patients are often sedatedand attached to multiple lines, tubes and monitoring devices.Perhaps paradoxically, this may facilitate measurement, as theseallow ready access to body fluids such as blood, urine, faeces, andexpired gases. However, as patients are non-ambulatory, it ispreferable that measurements are performed at the bedside,particularly due to the logistical difficulties in transportation ofthese patients [15]. There is also limited space around the patientdue to the frequent presence of essential but bulky equipment, suchas a ventilator and dialysis machines. Techniques must be adapt-able, as studies may be terminated prematurely due to unexpectedclinical requirements. Of particular concern is that the reliability ofa test may be impaired in the critically ill patient when compared toa ‘more controlled’ research setting. This may reflect technicaldifficulties exclusive to the intensive care environment, or physio-logical derangements in the patient caused by their clinical con-dition or treatment. Ideally, all measurement techniques should bevalidated in critically ill patients before their widespread use. Theprimary aim of this narrative review is to provide an overview ofthe measurement of gastric emptying in the critically ill, the evi-dence for each method that may be used to quantify emptying andrecommendations for use in the critical are setting.

2. Methods

We performed a comprehensive search of English languagemanuscripts, on MEDLINE/PubMed, from inception to 1 July 2014.We used both the following MeSH words and combination ofterms: gastric emptying, scintigraphy, absorption method, 3-O-methylglucose, isotope breath tests, octanoic acid, ultrasound,magnetic resonance imaging, gastric residual volumes, criticalillness and intensive care. Additionally, references of extractedmanuscripts were examined for additional studies that had notbeen identified during the initial search. Only studies of humancohorts were included. The methodology of each study was eval-uated and studies that were of greater methodological quality and/or data published more recently were preferred. However data wassourced from older studies when appropriate (due to insufficientrecent data).

2.1. Measurement of gastric emptying in the critically ill

Formal measurement of gastric emptying is rarely performed inthe critically ill other than for research purposes. A number oftechniques are available, but all have potential limitations andparticular difficulties for use in this population. Comparison ofdifferentmethods of measurement of gastric emptying is limited bythe practical inability to perform somemeasurements concurrentlyand differences in parameters acquired.

Changes in gastric volume over time can be quantified bytechniques such as scintigraphy, fluoroscopy, ultrasound and MRI.Techniques such as labelled carbon breath tests and paracetamol orsynthetic glucose absorption, require small intestinal absorption, aswell as gastric emptying of the substrate. In the case of breath tests,metabolism of the substrate must also occur before excretion intoexpired air (Table 1).

Assessment of gastric emptying also requires careful definitionof the variable to be measured. When ambulatory patients arebeing investigated for gastrointestinal symptoms, solid meals aregenerally considered to be a more reliable indicator of abnormalgastric emptying than nutrient liquids. However these are lessapplicable to the critically ill patient where liquid formulae areused for feeding. Because gastric emptying is frequently delayed inthe critically ill and is dependent on the composition of thenutrient (i.e. proportion of carbohydrate, lipid or protein), wefavour the use of lower volumes of liquid nutrient that is repre-sentative of feeds that are clinically used (e.g. 100 ml of 1 kcal/mlcontaining between 50 and 65% carbohydrate, 20e40% lipid,15e30% protein) [16].

In general, measurements of gastric emptying involve aningested marker (in the mechanically-ventilated patient, themarker is infused via a nasogastric tube) that is monitored directly,via imaging, or indirectly, using blood or breath analysis. The ‘ideal’technique for measurement of gastric emptying varies dependingon the circumstances and priorities facing clinicians or researchersat any particular time. In addition, these patients frequently havedelayed gastric emptying and, in some cases it is profoundly slow[2]. For this reason techniques that can measure gastric emptyingover several hours are preferred. Finally, the function of other or-gans (e.g. liver and kidneys) may be impaired, and volume of dis-tribution, metabolism and excretion of a marker may be altered inthese patients [17]. This may influence the outcome of measure-ments obtained from indirect tests [18]. The advantages and dis-advantages of the various measurements of gastric emptying aredetailed below.

2.2. Direct tests of gastric emptying

Direct tests require that the nutrient is ‘visualised’ duringemptying and therefore the technique is unaffected by liver orkidneymetabolism/excretion. Accordingly direct tests are generallymore precise than indirect measurements.

P. Kar et al. / Clinical Nutrition 34 (2015) 557e564 559

2.3. Scintigraphy

Scintigraphy is regarded as the ‘gold standard’ technique tomeasure gastric emptying [19]. A substrate labelled with a radio-active marker is placed in the stomach. This ‘meal’ can be solid orliquid, and the liquid can be nutrient or non-nutrient. In the criti-cally ill, liquid nutrient is the constituent of most interest. Thedisappearance of the radiolabelled meal component from the fieldof view is measured over time with either a fixed or mobile gammacamera, fitted with a large field-of-view collimator.

The choice of radioisotope is crucial to the success of the mea-surements. It should be non-toxic, impermeable to the gastricmucosa, and homogenously distributed throughout the ‘meal’ be-ing evaluated. Markers are available to measure emptying of spe-cific meal components e solid, liquid or fat e but these must bebound tightly to that component. Simultaneous measurement ofgastric emptying of solid and liquid meal components can be per-formed but requires the use of two radioisotopes. For single isotopestudies, the radioisotope selected should ideally have a half-life of4e12 h, which allows adequate time for data acquisition withoutexcessive radiation. Our experience is that the time period for dataacquisition should be substantially longer than that used in health,because the emptying rate is so slow in the critically ill, andextended periods (of up to 6 h) are required to quantify anyemptying.

Radioisotopic data are then analysed using dedicated computersoftware. Using a computer display, a region-of-interest is drawnaround the stomach, excluding the small intestine. This can bedifficult when there is overlying bowel and represents a potentialsource of inaccuracy. Radioactive counts throughout the study arethen measured within the region-of-interest and corrected sub-sequently for isotope decay and tissue attenuation (i.e. depth-related changes in activity), as both may substantially alter re-sults [19]. In health, radioactivity may be corrected for tissueattenuation using factors derived from a lateral image or using thegeometric mean of concurrent anteroposterior images [20]. In thecritically ill, lateral images are difficult or impossible to attain andcorrection factors are frequently not performed. The left anterioroblique (LAO) position is a simple method for correcting for

Fig. 1. a) Normal and delayed gastric emptying of 100 ml of liquid nutrient (2 kcal/ml) measumeal is administered a region of interest (ROI) is drawn around the entire stomach contentsradiolabelled meal progresses from the stomach into the small intestine across each image (win the critically ill patient, with the radiolabelled being almost completed retained in the (p

gamma ray attenuation and should be used in these patients [21].In the case of dual isotope studies, a down-scatter correctionshould also be applied [20]. The maximum counts in the first20 min of the study represents 100% retention of the meal [20].Radioactivity in the region-of-interest is then plotted over time foreach study frame, expressed as a % of the maximum (Figs. 1 and 2aand Supplementary video file 1 and 2). Gastric emptying param-eters may include the radioactivity over time (i.e. Area under thecurve, AUC) or summarised as the time taken for 50% of the mealto empty (T50). Our experience is that complete or near-completestasis occurs in some patients so that more than 50% of the mealremains in the stomach when data acquisition is ceased. As such,the T50 may not be the best parameter in the critically ill, andusing dynamic imaging, the proportion of isotope residing in thestomach at a specific time points (e.g. 60 min post meal) mayprovide more accurate and reproducible data. When emptying ofsolids is measured, a lag phase is observed, corresponding to thetime taken for intragastric redistribution and breakdown to par-ticles 1 mm in size prior to transpyloric emptying. This is notrelevant when measuring the emptying rate of liquid nutrient incritical illness. In addition to total gastric emptying, distribution ofthe meal within the stomach can be assessed. Comparison of re-sults between different centres must be performed with care asmeal properties and acquisition profiles may vary, affecting mea-surements. Normal ranges may therefore vary. Our preference isfor a relatively low calorie nutrient liquid (e.g. 100e200 kcal) withdynamic images collected every 1e3 min with retention deter-mined at 15e30 minutely intervals for between 4 and 6 h perstudy.

Supplementary video related to this article can be found athttp://dx.doi.org/10.1016/j.clnu.2014.11.003.

It should be recognised that scintigraphy measures gastricemptying in terms of the meal volume remaining in the stomachover time relative to that originally instilled into the stomach oringested (per cent of total tracer). The absolute volume of gastriccontents and the rate at which this empties from the stomachcannot be measured with this technique because the volume ofgastric secretions both present initially and produced during thestudy cannot be quantified [22].

red in two critically ill patients (unpublished data) over 5 h. On the first image after the. The images in 1a reveal a relatively ‘normal’ rate of gastric emptying in a patient. Theith image acquisition taken at 1 h intervals). Figure. 1b) shows delayed gastric emptyingroximal) stomach and radioisotope is only sparsely viewed distal to the pylorus.

Fig. 2. a) Retention of gastric contents over time via scintigraphic technique. Data fromtwo critically ill patients after administration of 100 ml of liquid nutrient (2 kcal/ml)labelled 20 MBq 99mTc- calcium phytate (unpublished data). A higher percentage ofcontents retained in the stomach over time indicate a slower rate of gastric emptying(solid circles). In a patient exhibiting ‘normal’ gastric emptying (open circle) a lesserpercentage of the meal is retained over time. b) Cumulative dose of 13CO2 exhaled overtime. This figure displays data from the same two patients after administration of100 ml of liquid nutrient (2 kcal/ml) labelled with 100 mcg of octanoic acid. Theoctanoic acid contains 13C, which is absorbed by the small intestine and undergoeshepatic metabolism to 13CO2, which is then exhaled. An increased amount of exhaled13CO2 indicates faster gastric emptying (open circles). Data from a patient with delayedgastric emptying (solid circles) displays a lesser cumulative dose/hour of CO2 exhaled.Due to the small change in cumulative dose from baseline in patients with very slowgastric emptying, this measure may be less precise when measuring markedly delayedgastric emptying.

P. Kar et al. / Clinical Nutrition 34 (2015) 557e564560

The radiation dose for scintigraphic gastric emptying studies isdependent on the radioisotope used, the dose administered and thegastrointestinal transit time - a typical dose of 20 MBq of99mtechnetium labelled sulphur colloid (which is used to measuregastric emptying) is approximately 0.5 mSv. This is comparable tothe radiation exposure associated with a mammogram.

Table 2Scintigraphy as a measure of gastric emptying in the critically ill.

StudyPrimary investigator (year)

Dose administered(MBq)

Radionuclide used

Ott (1991) 18.5e37 99mTc-diethylene-triacid (DTPA)

Spapen (1995) 20 99mTc-sulphur colloiKao (1998) 18.5 99mTc-phytateNguyen (2008)a 20 99mTc-sulphur colloiChapman (2009)a 20 99mTc-sulphur colloiChapman (2011)a 20 99mTc-sulphur colloi

a Denotes investigators from the same group; T50, scintigraphic half emptying time; R

Several groups, including ours, have used scintigraphy to mea-sure gastric emptying in critically ill patients (Table 2) [3,5,23e26].The precision associated with this technique allows it to determinethe effect of drugs that may have a small, but clinically important,gastrokinetic effect. However, in most centres scintigraphy involvestransporting the patient to the nuclear medicine department,limiting its use, however, mobile cameras, when available, obviatethe need to move a critically ill patient.

2.4. Indirect tests of gastric emptying

Indirect tests use a marker that is not absorbed in the stomach,but is absorbed in the duodenum. Accordingly, the appearance ofthe marker in blood, or a metabolite excreted by the lungs or kid-neys, correlates with the rate of gastric emptying.

2.5. Drug absorption (paracetamol)

In health, plasma concentrations of an ingested drug that is notabsorbed in the stomach, but is rapidly and freely absorbed by thesmall intestine, will reflect the gastric emptying rate [15]. Drugabsorption tests have advantages as they are relatively simple toperform and, at least in the case of paracetamol, clinicians oftenhave access to a laboratory that can quantify plasmaconcentrations.

The paracetamol absorption test has, until recently, been themost frequently used technique to measure gastric emptying in thecritically ill [17,27e36]. However, there have been substantial var-iations in the protocols used by various investigators (Table 3),including differences in the dose and form (tablets or solution) ofparacetamol and the type of meal with which it was administered.There is also a lack of consistency in the calculated parameters usedto report the results. The latter include the concentration at aspecific time point, maximal concentration, time to reach maximalconcentration, area under the curve, and the proportion of areaunder the curve at specific time points. In addition to the lack ofstandardisation, other issues limit the suitability of the paracetamolabsorption test. Paracetamol is an effective and commonly pre-scribed analgesic in the critically ill. While the administration ofparacetamol for the analgesic and/or temperature-lowering effectsmay be clinically desirable, the test requires a baseline paracetamolconcentration of zero. Pharmacokinetics may also be affected dur-ing critical illness, and given the hepatic metabolism of paraceta-mol, its use for research purposes is contraindicated in patientswith marked liver impairment, which occurs frequently in thesepatients [37]. Our opinion is that the paracetamol absorption test islimited and favour alternative techniques.

2.6. Carbohydrate absorption (3-O-methylglucose)

3-O-methylglucose (3-OMG) is a synthetic sugar that is imper-meable to gastric mucosa but absorbed actively in the small in-testine via the same co-transporters as glucose. The small intestinal

Scintigraphy time afteradministration (min)

Outcomes reported

amine-pentaacetic 20e60 min T50

d 120 min T5030 min T50

d 240 min Retd 240 min Retd 240 min T50, Ret

et, intragastric content retention.

Table 3Paracetamol absorption as a measure of gastric emptying in the critically ill.

Primary investigator(year)

Dose administered(mg)

Formulation of paracetamoladministeredS ¼ solubleL ¼ liquid

Plasma sampled at time afteradministration (min)

Outcomes reported

Heyland (1996) 1600 L 30, 60, 90, 120 AUC120

Tarling* (1997) 1000 S 5, 10, 15, 30, 45, 60, 90, 120 AUC60

Goldhill* (1997) 1000 S 5, 10, 15, 30, 45, 60, 90, 120 Cmax, Tmax and AUC60

McArthur (1995) 1000 S 5,10, 15, 20, 30, 45, 60, 75, 90,105, 120, 150, 180

Cmax, Tmax and AUC30 AUC180

Cohen (2000) 1000 L 15, 30, 45, 60 AUC60

Jooste (1999) 1500 S 15, 30, 45, 60, 90, 120 Cmax, AUC120

MacLaren^ (2000) 1000 L 7 samples between 0 and 720 Tmax AUC120

MacLaren^ (2008) 975 L 15, 30, 45, 60, 90, 120, 180, 240, 360 Cmax, Tmax and C60 AUC60

Lucey (2003) 1500 L 15, 30, 45, 60, 90, 120 AUC120

Marino (2003) 1000 S 15, 30, 45, 60, 90, 120 AUC120

Tamion (2003) 1000 S 10, 20, 30, 60, 90, 120, 180 Cmax, Tmax and AUC60 AUC180

A lack of standardisation is apparent between investigators using the paracetamol absorption test to measure gastric emptying in the critically ill. The dose of paracetamoladministered has ranged from 975 to 1600 mg, and some investigators administered the liquid suspension paracetamol, whereas others use dissolved tablets. There is also alack of consistency across studies in terms of the timing of measurements of plasma concentrations and outcomes reported.*,̂ Denotes investigators from the same group; Cmax, peak concentration of paracetamol; Tmax time to peak concentration; AUC area under curve.

P. Kar et al. / Clinical Nutrition 34 (2015) 557e564 561

absorption rates of 3-OMG and glucose are similar, but in contrastto glucose, 3-OMG is metabolically inert (i.e. it is not metabolisedby intestinal mucosa or body tissues) and is excreted unchanged inthe urine [38]. Accordingly, the appearance of 3-OMG in the cir-culation equals the rate of absorption from the small intestine, withthe majority of 3-OMG excreted in the 12 h following a meal [38].Unlike other carbohydrate absorption tests (e.g. D-xylose), abnor-malities in small intestinal absorptive capacity must be substantialprior to any decrease in 3-OMG being detected [38]. These prop-erties make 3-OMG a useful tool to estimate the absorption ofglucose.

Our group has substantial experience with using plasma 3-OMGconcentrations as a marker of gastric emptying and glucose ab-sorption [5,16,39,40]. A particular advantage of the carbohydrateabsorption tests is that energy delivery is a focus of nutritionaltherapy, and tests such as 3-OMG measure the clinically importantvariable (i.e. nutrient absorption). Currently there is no commer-cially available assay to measure 3-OMG concentrations and highperformance liquid chromatography is required. Accordingly thetechnique is expensive and is not widely available. Data aregenerally presented as area under the concentration curve[16,37,39].

Similar to paracetamol absorption, the limitations of the 3-OMGtest include imprecision due to disordered pharmacokinetics in thecritically ill. The volume of distribution of glucose however appearsminimally affected by critical illness [41]. A further limitation is that

Table 4Breath test as a measure of gastric emptying in the critically ill.

StudyPrimary investigator (year)

Dose administered Type of isotope used

Ritz (2001)a 100 mL 13C-octanoic acid

Chapman (2003)a 100 mL 13C-octanoic acid

Ritz (2005)a 100 mL 13C-octanoic acid

Chapman (2005)a 100 mL 13C-octanoic acid

Deane (2010)a 100 mL 13C-octanoic acid

Chapman (2011)a 75 KBq 14C-octanoic acid

a Denotes investigators from the same group; GEC, gastric emptying coefficient; t1/2 g

small intestinal monosaccharide (3-OMG) absorption is impaired inthe critically ill [37], due to mesenteric blood flow and molecularmechanisms [40,42], which are factors distal to the pylorus.Accordingly, drugs may affect small intestinal motility therebyaltering carbohydrate (3-OMG) absorption independent of any ef-fect on gastric emptying [39]. Hence, while the 3-OMG test is auseful measure of a relatively important end-point, the technique isa somewhat imprecise measurement of gastric emptying per se.

2.7. Isotope breath tests

The measurement of gastric emptying with a breath test re-quires the ingestion of a ‘standard meal’, labelled with a stableisotope (e.g. 13C). The isotope is incorporated into a free fatty acid(e.g. octanoic acid or acetate), which is rapidly (and almost entirely)absorbed from the duodenum/small intestine, and then metab-olised in the liver to 13CO2, which is excreted during exhalation[43]. In health gastric emptying is the rate-limiting step in thisprocess, and pulmonary excretion of labelled CO2 is used to indicatethe rate of gastric emptying of the meal. While 13C-acetate ab-sorption and metabolism is independent of the volume and caloricdelivery of the test meal, measurement of the ‘lag’ before gastricemptying commences reflects a postgastric, dose-dependent delayin 13CO2 elimination [44]. 13CO2 is measured using an isotope ratiomass spectrometer. A greater cumulative dose of 13CO2 exhaledover time indicates increased gastric emptying (Fig. 2b). Data are

Breath sampled at time after administration (min) Outcomes reported

Hr 0e1: 5 minutelyHr 1e4: 15 minutely

GEC, t1/2

Hr 0e1: 5 minutelyHr 1e4: 15 minutely

GEC, t1/2

Hr 0e1: 5 minutelyHr 1e4: 15 minutely

GEC, t1/2

Hr 0e1: 5 minutelyHr 1e4: 15 minutely

GEC

Hr 0e1: 5 minutelyHr 1e4: 15 minutely Hr 4e5.5: 30 minutely

GEC

Hr 0e1: 10 minutelyHr 1e4: 15 minutely

GEC, t1/2

astric half-emptying time.

P. Kar et al. / Clinical Nutrition 34 (2015) 557e564562

usually presented as the gastric emptying coefficient (GEC) and/orthe gastric half-emptying time (t1/2). The GEC is a global index forthe gastric emptying rate that accounts for the rate of appearanceand disappearance of tracer in the breath, with the greater thevalue of GEC indicating a more rapid emptying rate. Gastric half-emptying time (t1/2) is the time to 50% of the total 13C recoveredand is also reported. Unlike the scintigraphic T50, which reflects adirectly measured amount (i.e. the time taken for half of theradioactivity to empty from the stomach), the t1/2 is a calculatedmeasure. Moreover, 13CO2 is excreted via non-respiratory pathways(indeed it has been proposed that a three compartment modelexists and a large proportion of 13CO2 is stored) [45,46]. In ouropinion, the t1/2 is less reliable than the GEC in the critically ill, asgastric emptying is markedly delayed inmany patients, so that onlysmall amounts of 13C are recovered during the measurementperiod, and calculations required to estimate t1/2 in these patientsare prone to error.

We have extensive experience with this test (Table 4)[5,16,47e50] and it is a useful measurement of gastric emptying.Similar to other measures of gastric emptying that rely on smallintestinal absorption, there may be factors that impair absorptionwhich can adversely affect the result [19]. Indeed there may befactors within the small intestine that cause substantially differentresults in an individual patient when measuring gastric emptyingusing any of the indirect tests. A further concern is that CO2 pro-duction is variable between patients, with dynamic changes withinpatients during their acute illness. While the test appears to berelatively robust with acceptable intrasubject variability [29,51]and provides a satisfactory correlation with scintigraphy (GEC;r ¼ �0.63 to �0.74; P < 0.0001; t1/2; r ¼ 0.55e0.66; P < 0.001) [25]in critically ill patients with normal to slow gastric emptying therelative imprecision of the test can be balanced by increasing thesample size. This technique is also particularly suited to evaluatingchanges in gastric emptying for example in response to therapy, butis less suitable for comparing gastric emptying between two pa-tient groups. Its use is also limited in patients with markedgastroparesis.

2.8. Other tests of gastric emptying

Other measures of gastric emptying are used less frequently inthe critically ill patient, when compared to the tests describedabove.

2.9. Ultrasound

Conventional real-time ultrasound can be used to study antralcontractility, gastric emptying, transpyloric flow, gastric configu-ration, intragastric distribution of meals, gastric accommodationand strain measurement of the gastric wall [52]. The advantages ofultrasound are that it is non-invasive, involves no radiation, and canbe performed at the bedside.

Gastric emptying can be measured using both 2-dimensional(2D) or 3-dimensional (3D) ultrasound. Using 2D ultrasoundgastric emptying is usually characterised by measurement ofchanges in antral cross sectional area or diameter over time [53].These measurements correlate closely with emptying rates of aliquid or semi-solid ‘meals’, as determined by scintigraphy [54],however, in patients with gastroparesis, a higher proportion of themeal may be retained in the proximal stomach potentiallyincreasing the likelihood of error [55]. 3D ultrasound is an accuratemeasurement of volume (including proximal volume) and thistechnique has been reported to closely match measurements ob-tained with scintigraphy [55].

In our experience, obtaining an acceptable ‘window’ can be verydifficult due to ‘patient factors’ (subcutaneous tissue, interstitialoedema, air in the fundus) and ‘liquid nutrient factors’ (echo-densefeeds may obscure the view of the proximal stomach) (unpublisheddata). A further limitation is that 3D ultrasound requires a ‘breathhold’ and this is technically impossible in most spontaneouslyventilating patients receiving invasive mechanical support unlessparalysing drugs are administered. For these reasons we have beenunable to obtain satisfactory images when piloting this technique(unpublished data). Accordingly, other techniques are currentlypreferred, but due to the potential advantages, further study usingthis technique is warranted.

2.10. magnetic resonance imaging

Magnetic Resonance Imaging (MRI) has many advantages whenmeasuring gastric emptying, such as image acquisition (involves noionising radiation) and analysis (operator independent as inter-pretation does not involve manual tracing). MRI was first describedas a technique to measure emptying liquids but, unlike scintig-raphy, also allowed quantification of gastric secretions. MRI candetect changes in gastric emptying rate caused by pharmaco-therapy [56]. A further advantage of MRI is that emptying of indi-vidual constituents of a liquid meal (i.e. lipid) may be measured inaddition to that of the entire meal. Furthermore, MRI permits directvisualisation of gastric wall motion and morphologicalabnormalities.

The disadvantages of MRI relate to the expense and availabilityof the scanner itself. Moreover, transportation and the periodwithin the MRI scanner for image acquisition may be associatedwith significant adverse events in the critically ill and, accordingly,the use of MRI in this group is impractical.

2.11. Surrogate tests of gastric emptying

Surrogate markers are used in clinical practice and research toestimate gastric emptying.

2.12. Gastric residual volumes (GRVs)

Intermittent measurement of the gastric residual volume (GRV)during the infusion of enteral nutrition is a convenient clinical toolthat is used widely as a surrogate indication of gastric emptying,‘success’ of feeding, and, possibly, the risk of aspiration. Despiteinclusion of GRV measurements in feeding protocols by ICUsworldwide, the utility and significance of GRVs remain controver-sial. In particular, GRVs are dependent on a number of factors suchas the position of the tube, tube characteristics (such as tube typeand number of openings), the operator performing the test, use ofprokinetic drugs and calorie concentration of the liquid nutrient[57,58]. Moreover, GRVs are usually measured 4e6 hourly and thesignificance of a single value may be uncertain. These factors haveled to a lack of consensus for what is an acceptable threshold forGRV and the significance of a single GRV [57]. Our group has re-ported that a GRV � 250 mL in the preceding 24 h was a relativelysensitivemarker of delayed gastric emptying whenmeasured usingscintigraphy [5]. Additionally, patients with a GRV � 250 mL haddelayed gastric emptying subsequently measured by scintigraphy[59]. In general, we favour commencing intragastric feeds at goalrate via a larger bore feeding tube with GRVs aspirated every 6 h.Moreover, while large residual volumes may predict slowemptying, these may also identify patients at greater risk of com-plications such as vomiting or aspiration, but a recent study haschallenged this notion [60]. Further studies are required to confirmthis interesting observation. GRVs continue to be used clinically as

P. Kar et al. / Clinical Nutrition 34 (2015) 557e564 563

this test is an easy, low risk, bedside surrogate measurement toidentify patients at risk of slower gastric emptying. While it stillneeds to be clarified whether an increase in GRVs equates to clin-ically relevant adverse events, such as ventilator associated com-plications, it seems reasonable to use GRVs to select patients thathave the capacity to benefit from gastrokinetic drugs or small in-testinal feeding tubes.

2.13. Strengths and limitations

This review provides a summary of current information relatedto gastric emptying in the critically ill. At our centre we haveconsiderable experience with several of the methodologiesdescribed. While an asset, we recognise that the latter has thepotential to bias our recommendations. Many of these studiesinvolve small numbers of patients studied in highly specialisedcentres with sophisticated methodologies, which may compromisethe generalisability of the outcomes.

2.14. Recommendations

The choice of gastric emptying test depends on the re-quirements of the investigator. For clinicians wanting a ‘crude’ buteasy measurement to identify patients receiving nasogastricfeeding with slow gastric emptying at the bedside, GRV remains auseful, albeit limited and controversial, test and pending furtherdata we continue to use GRVs clinically to determine those patientswho have the potential to benefit from interventions such as gas-trokinetic drugs or small intestinal feeding tubes. We recommend athreshold of GRV � 250 mL to define slow gastric emptying toidentify patients likely to have slow gastric emptying, but recognisethis approach has hitherto not been translated into improvedpatient-centred outcomes.

However, in the research setting, particularly when studyingsmaller cohorts, GRVs are too imprecise. Indirect tests, such ascarbohydrate absorption (3-OMG) or radio-isotope breath tests(13CO2), are minimally invasive and have modest intrasubjectvariability. They are useful when studying the effect of an inter-vention thought to have a potent effect on gastric emptying inlarger cohorts, particularly when researchers use a crossoverdesign. However, when studying the effect of a less potent gas-trokinetic drug and/or using a parallel design study, and a moreprecise measurement are required, direct tests, such as scintig-raphy, are more suitable and remain the gold standard.

Conflict of interest

None.

Acknowledgment

PK is supported by a Royal Adelaide Hospital A.R. ClarksonScholarship. AMD is supported by a National Health and MedicalResearch Council Early Career Fellowship.

References

[1] Martindale RG, McClave SA, Vanek VW, McCarthy M, Roberts P, Taylor B, et al.Guidelines for the provision and assessment of nutrition support therapy inthe adult critically ill patient: Society of Critical Care Medicine and AmericanSociety for Parenteral and Enteral Nutrition: executive summary. Crit CareMed 2009 May;37(5):1757e61. PubMed PMID: 19373044.

[2] Deane AM, Fraser RJ, Chapman MJ. Prokinetic drugs for feed intolerance incritical illness: current and potential therapies. Crit Care Resusc 2009Jun;11(2):132e43. PubMed PMID: 19485878.

[3] Spapen HD, Duinslaeger L, Diltoer M, Gillet R, Bossuyt A, Huyghens LP. Gastricemptying in critically ill patients is accelerated by adding cisapride to a

standard enteral feeding protocol: results of a prospective, randomized,controlled trial. Crit Care Med 1995 Mar;23(3):481e5. PubMed PMID:7874898.

[4] Nguyen NQ, Ng MP, Chapman M, Fraser RJ, Holloway RH. The impact ofadmission diagnosis on gastric emptying in critically ill patients. Crit Care2007;11(1):R16. PubMed PMID: 17288616. Pubmed Central PMCID: 2151889.

[5] Chapman MJ, Besanko LK, Burgstad CM, Fraser RJ, Bellon M, O'Connor S, et al.Gastric emptying of a liquid nutrient meal in the critically ill: relationshipbetween scintigraphic and carbon breath test measurement. Gut 2011Oct;60(10):1336e43. PubMed PMID: 21450697.

[6] Metheny NA, Schallom L, Oliver DA, Clouse RE. Gastric residual volume andaspiration in critically ill patients receiving gastric feedings. Am J Crit Care2008 Nov;17(6):512e9. quiz 20. PubMed PMID: 18978236. Pubmed CentralPMCID: 2627559.

[7] Egi M, Bellomo R, Stachowski E, French CJ, Hart G. Variability of blood glucoseconcentration and short-term mortality in critically ill patients. Anesthesi-ology 2006 Aug;105(2):244e52. PubMed PMID: 16871057.

[8] Jung HK, Choung RS, Locke 3rd GR, Schleck CD, Zinsmeister AR, Szarka LA,et al. The incidence, prevalence, and outcomes of patients with gastroparesisin Olmsted county, Minnesota, from 1996 to 2006. Gastroenterology 2009Apr;136(4):1225e33. PubMed PMID: 19249393. Pubmed Central PMCID:2705939.

[9] Luttikhold J, de Ruijter FM, van Norren K, Diamant M, Witkamp RF, vanLeeuwen PA, et al. Review article: the role of gastrointestinal hormones in thetreatment of delayed gastric emptying in critically ill patients. AlimentPharmacol Ther 2013 Sep;38(6):573e83. PubMed PMID: 23879699.

[10] Horowitz M, Maddox AF, Wishart JM, Harding PE, Chatterton BE, Shearman DJ.Relationships between oesophageal transit and solid and liquid gastricemptying in diabetes mellitus. Eur J Nucl Med 1991;18(4):229e34. PubMedPMID: 2070801.

[11] Lin HC, Prather C, Fisher RS, Meyer JH, Summers RW, Pimentel M, et al.Measurement of gastrointestinal transit. Dig Dis Sci 2005 Jun;50(6):989e1004. PubMed PMID: 15986844.

[12] Schmidt PT, Abrahamsson H, Dolk A, Hausken T, Karling P, Lindberg G, et al.Methods to assess gastric motility and sensation. Scand J Gastroenterol2008;43(11):1285e95. PubMed PMID: 18618332.

[13] Szarka LA, Camilleri M. Methods for measurement of gastric motility. Am JPhysiol Gastrointest Liver Physiol 2009 Mar;296(3):G461e75. PubMed PMID:19147807.

[14] Shin AS, Camilleri M. Diagnostic assessment of diabetic gastroparesis. Dia-betes 2013 Aug;62(8):2667e73. PubMed PMID: 23881199. Pubmed CentralPMCID: 3717858.

[15] Heading RC, Nimmo J, Prescott LF, Tothill P. The dependence of paracetamolabsorption on the rate of gastric emptying. Br J Pharmacol 1973 Feb;47(2):415e21. PubMed PMID: 4722050. Pubmed Central PMCID: 1776534.

[16] Deane AM, Chapman MJ, Fraser RJ, Summers MJ, Zaknic AV, Storey JP, et al.Effects of exogenous glucagon-like peptide-1 on gastric emptying and glucoseabsorption in the critically ill: relationship to glycemia. Crit Care Med 2010May;38(5):1261e9. PubMed PMID: 20228679.

[17] Goldhill DR, Toner CC, Tarling MM, Baxter K, Withington PS, Whelpton R.Double-blind, randomized study of the effect of cisapride on gastric emptyingin critically ill patients. Crit Care Med 1997 Mar;25(3):447e51. PubMed PMID:9118661.

[18] Deane AM, Zaknic AV, Summers MJ, Chapman MJ, Lange K, Ritz MA, et al.Intrasubject variability of gastric emptying in the critically ill using a stableisotope breath test. Clin Nutr 2010 Oct;29(5):682e6. PubMed PMID:20409622.

[19] Camilleri M, Hasler WL, Parkman HP, Quigley EM, Soffer E. Measurement ofgastrointestinal motility in the GI laboratory. Gastroenterology 1998Sep;115(3):747e62. PubMed PMID: 9721173.

[20] Collins PJ, Horowitz M, Cook DJ, Harding PE, Shearman DJ. Gastric emptying innormal subjectsea reproducible technique using a single scintillation cameraand computer system. Gut 1983 Dec;24(12):1117e25. PubMed PMID:6642275. Pubmed Central PMCID: 1420248.

[21] Ford PV, Kennedy RL, Vogel JM. Comparison of left anterior oblique, anteriorand geometric mean methods for determining gastric emptying times. J NuclMed 1992 Jan;33(1):127e30. PubMed PMID: 1730977.

[22] Goetze O, Treier R, Fox M, Steingoetter A, Fried M, Boesiger P, et al. The effectof gastric secretion on gastric physiology and emptying in the fasted and fedstate assessed by magnetic resonance imaging. Neurogastroenterol Motil2009 Jul;21(7). 725ee42. PubMed PMID: 19344341.

[23] Kao CH, ChangLai SP, Chieng PU, Yen TC. Gastric emptying in head-injuredpatients. Am J Gastroenterol 1998 Jul;93(7):1108e12. PubMed PMID:9672339.

[24] Ott L, Young B, Phillips R, McClain C, Adams L, Dempsey R, et al. Altered gastricemptying in the head-injured patient: relationship to feeding intolerance.J Neurosurg 1991 May;74(5):738e42. PubMed PMID: 1901599.

[25] Nguyen NQ, Fraser RJ, Bryant LK, Burgstad C, Chapman MJ, Bellon M, et al. Theimpact of delaying enteral feeding on gastric emptying, plasma cholecysto-kinin, and peptide YY concentrations in critically ill patients. Crit Care Med2008 May;36(5):1469e74. PubMed PMID: 18434906.

[26] Chapman MJ, Fraser RJ, Matthews G, Russo A, Bellon M, Besanko LK, et al.Glucose absorption and gastric emptying in critical illness. Crit Care2009;13(4):R140. PubMed PMID: 19712450. Pubmed Central PMCID:2750198.

P. Kar et al. / Clinical Nutrition 34 (2015) 557e564564

[27] Heyland DK, Tougas G, Cook DJ, Guyatt GH. Cisapride improves gastricemptying in mechanically ventilated, critically ill patients. A randomized,double-blind trial. Am J Respir Crit Care Med 1996 Dec;154(6 Pt 1):1678e83.PubMed PMID: 8970354.

[28] Tarling MM, Toner CC, Withington PS, Baxter MK, Whelpton R, Goldhill DR.A model of gastric emptying using paracetamol absorption in intensive carepatients. Intensive Care Med 1997 Mar;23(3):256e60. PubMed PMID:9083226.

[29] McArthur CJ, Gin T, McLaren IM, Critchley JA, Oh TE. Gastric emptyingfollowing brain injury: effects of choice of sedation and intracranial pressure.Intensive Care Med 1995 Jul;21(7):573e6. PubMed PMID: 7593899.

[30] Jooste CA, Mustoe J, Collee G. Metoclopramide improves gastric motility incritically ill patients. Intensive Care Med 1999 May;25(5):464e8. PubMedPMID: 10401939.

[31] Cohen J, Aharon A, Singer P. The paracetamol absorption test: a useful addi-tion to the enteral nutrition algorithm? Clin Nutr 2000 Aug;19(4):233e6.PubMed PMID: 10952793.

[32] MacLaren R, Kuhl DA, Gervasio JM, Brown RO, Dickerson RN, Livingston TN,et al. Sequential single doses of cisapride, erythromycin, and metoclopramidein critically ill patients intolerant to enteral nutrition: a randomized, placebo-controlled, crossover study. Crit Care Med 2000 Feb;28(2):438e44. PubMedPMID: 10708180.

[33] MacLaren R, Kiser TH, Fish DN, Wischmeyer PE. Erythromycin vs metoclo-pramide for facilitating gastric emptying and tolerance to intragastric nutri-tion in critically ill patients. J Parenter Enter Nutr 2008 JuleAug;32(4):412e9.PubMed PMID: 18596312.

[34] Lucey MA, Patil V, Girling K, Jacques T, O'Leary M. Does neostigmine increasegastric emptying in the critically ill?eresults of a pilot study. Crit Care Resusc2003 Mar;5(1):14e9. PubMed PMID: 16573452.

[35] Marino LV, Kiratu EM, French S, Nathoo N. To determine the effect of meto-clopramide on gastric emptying in severe head injuries: a prospective, ran-domized, controlled clinical trial. Br J Neurosurg 2003 Feb;17(1):24e8.PubMed PMID: 12779198.

[36] Tamion F, Hamelin K, Duflo A, Girault C, Richard JC, Bonmarchand G. Gastricemptying in mechanically ventilated critically ill patients: effect of neuro-muscular blocking agent. Intensive Care Med 2003 Oct;29(10):1717e22.PubMed PMID: 12897996.

[37] Deane AM, Summers MJ, Zaknic AV, Chapman MJ, Di Bartolomeo AE, Bellon M,et al. Glucose absorption and small intestinal transit in critical illness. Crit CareMed 2011 Jun;39(6):1282e8. PubMed PMID: 21336122.

[38] Fordtran JS, Clodi PH, Soergel KH, Ingelfinger FJ. Sugar absorption tests, withspecial reference to 3-0-methyl-d-glucose and d-xylose. Ann Intern Med 1962Dec;57:883e91. PubMed PMID: 13958713.

[39] Deane AM, Wong GL, Horowitz M, Zaknic AV, Summers MJ, Di Bartolomeo AE,et al. Randomized double-blind crossover study to determine the effects oferythromycin on small intestinal nutrient absorption and transit in the criti-cally ill. Am J Clin Nutr 2012 Jun;95(6):1396e402. PubMed PMID: 22572649.

[40] Sim JA, Horowitz M, Summers MJ, Trahair LG, Goud RS, Zaknic AV, et al.Mesenteric blood flow, glucose absorption and blood pressure responses tosmall intestinal glucose in critically ill patients older than 65 years. IntensiveCare Med 2013 Feb;39(2):258e66. PubMed PMID: 23096428.

[41] Ishihara H, Matsui A, Muraoka M, Tanabe T, Tsubo T, Matsuki A. Detection ofcapillary protein leakage by indocyanine green and glucose dilutions in septicpatients. Crit Care Med 2000 Mar;28(3):620e6. PubMed PMID: 10752804.

[42] Deane AM, Rayner CK, Keeshan A, Cvijanovic N, Marino Z, Nguyen NQ, et al.The effects of critical illness on intestinal glucose sensing, transporters, andabsorption. Crit Care Med 2014 Jan;42(1):57e65. PubMed PMID: 23963126.

[43] Sanaka M, Yamamoto T, Kuyama Y. Retention, fixation, and loss of the [13C]label: a review for the understanding of gastric emptying breath tests. Dig DisSci 2008 Jul;53(7):1747e56. PubMed PMID: 18306041.

[44] Goetze O, Fox M, Kwiatek MA, Treier R, Schwizer W, Thumshirn M, et al. Ef-fects of postgastric 13C-acetate processing on measurement of gastricemptying: a systematic investigation in health. Neurogastroenterol Motil2009 Oct;21(10). 1047e1e85. PubMed PMID: 19496943.

[45] Ghoos YF, Maes BD, Geypens BJ, Mys G, Hiele MI, Rutgeerts PJ, et al. Mea-surement of gastric emptying rate of solids by means of a carbon-labeledoctanoic acid breath test. Gastroenterology 1993 Jun;104(6):1640e7.PubMed PMID: 8500721.

[46] Sanaka M, Yamamoto T, Ishii T, Kuyama Y. The Wagner-Nelson method cangenerate an accurate gastric emptying flow curve from CO2 data obtained by a13C-labeled substrate breath test. Digestion 2004;69(2):71e8. PubMed PMID:15031624.

[47] Chapman M, Fraser R, Vozzo R, Bryant L, Tam W, Nguyen N, et al. Antro-pyloro-duodenal motor responses to gastric and duodenal nutrient in criti-cally ill patients. Gut 2005 Oct;54(10):1384e90. PubMed PMID: 15923669.Pubmed Central PMCID: 1774690.

[48] Ritz MA, Fraser R, Edwards N, Di Matteo AC, Chapman M, Butler R, et al.Delayed gastric emptying in ventilated critically ill patients: measurement by13 C-octanoic acid breath test. Crit care Med 2001 Sep;29(9):1744e9. PubMedPMID: 11546976.

[49] Chapman M, Fraser R, de Beaux I, Creed S, Finnis M, Butler R, et al. Cefazolindoes not accelerate gastric emptying in the critically ill. Intensive Care Med2003 Jul;29(7):1169e72. PubMed PMID: 12802484.

[50] Ritz MA, Chapman MJ, Fraser RJ, Finnis ME, Butler RN, Cmielewski P, et al.Erythromycin dose of 70 mg accelerates gastric emptying as effectively as200 mg in the critically ill. Intensive Care Med 2005 Jul;31(7):949e54.PubMed PMID: 15940460.

[51] Ali Abdelhamid Y, Cousins CE, Sim JA, Bellon MS, Nguyen NQ, Horowitz M,et al. Effect of critical illness on triglyceride absorption. J Parenter Enter Nutr2014 Jun 24. PubMed PMID: 24963026. [Epub ahead of print].

[52] Hausken T, Li XN, Goldman B, Leotta D, Odegaard S, Martin RW. Quantificationof gastric emptying and duodenogastric reflux stroke volumes using three-dimensional guided digital color doppler imaging. Eur J Ultrasound 2001Jul;13(3):205e13. PubMed PMID: 11516632.

[53] Marzio L, Giacobbe A, Conoscitore P, Facciorusso D, Frusciante V, Modoni S.Evaluation of the use of ultrasonography in the study of liquid gastricemptying. Am J Gastroenterol 1989 May;84(5):496e500. PubMed PMID:2655434.

[54] Hveem K, Jones KL, Chatterton BE, Horowitz M. Scintigraphic measurement ofgastric emptying and ultrasonographic assessment of antral area: relation toappetite. Gut 1996 Jun;38(6):816e21. PubMed PMID: 8984016. PubmedCentral PMCID: 1383185.

[55] Gentilcore D, Hausken T, Horowitz M, Jones KL. Measurements of gastricemptying of low- and high-nutrient liquids using 3D ultrasonography andscintigraphy in healthy subjects. Neurogastroenterol Motil 2006 Dec;18(12):1062e8. PubMed PMID: 17109689.

[56] Borovicka J, Lehmann R, Kunz P, Fraser R, Kreiss C, Crelier G, et al. Evaluationof gastric emptying and motility in diabetic gastroparesis with magneticresonance imaging: effects of cisapride. Am J Gastroenterol 1999 Oct;94(10):2866e73. PubMed PMID: 10520835.

[57] Montejo JC, Minambres E, Bordeje L, Mesejo A, Acosta J, Heras A, et al. Gastricresidual volume during enteral nutrition in ICU patients: the REGANE study.Intensive Care Med 2010 Aug;36(8):1386e93. PubMed PMID: 20232036.

[58] Deane A, Chapman MJ, Fraser RJ, Bryant LK, Burgstad C, Nguyen NQ. Mecha-nisms underlying feed intolerance in the critically ill: implications for treat-ment. World J Gastroenterol 2007 Aug 7;13(29):3909e17. PubMed PMID:17663503. Pubmed Central PMCID: 4171161.

[59] Nguyen NQ, Bryant LK, Burgstad CM, Chapman M, Deane A, Bellon M, et al.Gastric emptying measurement of liquid nutrients using the (13)C-octanoatebreath test in critically ill patients: a comparison with scintigraphy. IntensiveCare Med 2013 Jul;39(7):1238e46. PubMed PMID: 23471513.

[60] Reignier J, Mercier E, Le Gouge A, Boulain T, Desachy A, Bellec F, et al. Effect ofnot monitoring residual gastric volume on risk of ventilator-associatedpneumonia in adults receiving mechanical ventilation and early enteralfeeding: a randomized controlled trial. JAMA 2013 Jan 16;309(3):249e56.PubMed PMID: 23321763.