Address for Correspondence Pieter Janssen, Translational Research Center for Gastrointestinal Disorders, University of Leuven, Herestraat 49, bus 701, BE-3000 Leuven, Belgium. Tel: +32 (0)16 330147; fax: +32 (0)16 345939; e-mail: email@example.com
Background The stomach relaxes upon food intake and thereby provides a reservoir while keeping the intragastric pressure (IGP) low. We set out to determine whether we could use IGP as a measurement for stomach accommodation during food intake.
Methods In fasted healthy volunteers (n = 7–17) a manometer and an infusion catheter were positioned in the proximal stomach. After a stabilization period a nutrient drink was intragastrically infused at 15, 30 and 60 mL min−1. To investigate the effect of impaired accommodation the effect of NG-monomethyl-l-arginine (L-NMMA) was examined. The volunteers scored satiation until maximum, when the experiment ended. The IGP was presented as a change from baseline (mean ± SEM) and compared with repeated measures anova.
Key Results Independent on the ingestion speed, the IGP decreased initially and gradually increased thereafter. Volunteers scored maximal satiation after 699 ± 62, 809 ± 90 and 997 ± 120 mL nutrient drink infused (15, 30 and 60 mL min−1 respectively; P < 0.01). Maximum IGP decrease was 3.4 ± 0.5 mmHg after 205 ± 28 mL, 5.1 ± 0.7 mmHg after 212 ± 46 mL, and 5.2 ± 0.7 mmHg after 296 ± 28 mL infused volume [15, 30 and 60 mL min−1 respectively; not significant (ns)]. Post hoc analysis showed significant correlations between IGP and satiation score increase. During L-NMMA infusion IGP was significantly increased while subjects drank significantly less (816 ± 91 vs 1032 ± 71 mL; P < 0.005). Interestingly, the correlation between IGP increase and satiation score increase did not differ after L-NMMA treatment.
Conclusions & Inferences The IGP during nutrient drink ingestion provides a minimally invasive alternative to the barostat for the assessment of gastric accommodation. These findings furthermore indicate that IGP is a major determinant of satiation.
In between meals the proximal stomach has a high basal muscle tone mediated by the myoelectrical properties of the fundus1 and by the constant cholinergic input mediated by the vagal nerves.2 Proximal gastric tone decreases upon food intake, a reflex relaxation that is mediated by different (para) sympathetic reflex pathways that have been shown to decrease the contractile cholinergic input and activate the release of nitric oxide (NO).3 This gastric accommodation reflex increases the storage capacity of the stomach by increasing the compliance of the stomach muscles and thus keeps the intragastric pressure (IGP) low during food intake.4,5
The golden standard to determine gastric accommodation is the gastric barostat.6 This technique detects changes in muscle tone by measuring volume changes of an intragastric balloon that is kept at a constant pressure. There are however important drawbacks associated with the technique: the inflated balloon distends the proximal stomach, might exaggerate gastric accommodation, and hampers physiologic responses to food intake.7,8 Moreover, muscle tone can only be measured in a limited stomach region and the invasive nature of this technique, which is perceived as uncomfortable and stressful, excludes its application in routine clinical practice. There is a clear need for an alternative technique to assess gastric accommodation during food intake.9
We recently described a novel technique to measure gastric accommodation during intragastric nutrient drink infusion in rats based on IGP measurements.10 Indeed, IGP is determined by the gastric muscle tone and we demonstrated that increasing or decreasing the gastric tone increased or decreased IGP, respectively. There is also emerging evidence in man that gastric accommodation determines IGP during stomach distension: while in healthy subjects the IGP increase during stomach balloon or air distension is minor or stable and does not increase despite further distension, IGP increase was much more pronounced in patients that underwent vagotomy, patients with Chagas’ disease and patients with functional dyspepsia.11–14 Despite these encouraging reports, IGP during gastric distension has never been advanced to standardization and validation for minimally invasive assessment of gastric accommodation in man.
Our aim was to setup and validate a technique to assess gastric accommodation by measuring the IGP during nutrient drink ingestion. Several experiments were performed: first we have used the well-validated barostat technique to investigate whether IGP during stomach distension can be used to assess gastric accommodation. In a second series of experiments, we setup IGP measurement during nutrient drink ingestion and compared drinking versus intragastric infusion and the effect of ingestion speed. In a last series of experiments, we validated the technique by comparing ingestion of nutrient drink with non-nutritious meals and after NG-monomethyl-l-arginine (L-NMMA) treatment.
All study procedures were approved by the Ethics Committee of the Leuven University Hospital, Belgium. Healthy volunteers (HVs) participated that gave written, informed consent. None of the HVs had symptoms or a history of gastrointestinal disease, other significant diseases, psychological disorders or drug allergies; none were taking any medication or had any drug history. All participants participated after an overnight fast, furthermore were they asked to refrain from alcohol, tea and coffee at least 12 h before participation, and to refrain from smoking cigarettes at least 1 h before the start of the experiment.
Barostat: IGP during stomach distension
Nine HVs (six men, age: 29 ± 3 years, body mass index: 22.8 ± 0.8 kg m−2) participated in the barostat study. Upon arrival in the clinic, a double lumen polyvinyl tube (Salem sump tube 14 Ch.; Sherwood Medical, Petit Rechain, Belgium) with an adherent polyethylene bag (maximal volume 1200 mL; 17 cm maximal diameter) was introduced through the mouth and secured to the subject’s chin with adhesive tape. The polyvinyl tube was then connected to a programmable barostat device (Synectics Visceral Stimulator, Stockholm, Sweden). To unfold the bag, it was slowly inflated with a volume of 300 mL for 2 min with the subject in a recumbent position; hereafter the bag was completely deflated. The subjects were then positioned in a comfortable sitting position with the knees bent (80°) and the trunk upright in a specifically designed bed. After a 30-min stabilization period the subjects were asked to drink 200 mL of either tap water or a nutrient drink (Nutridrink; Nutricia, Bornem, Belgium), 5 min later the intragastric balloon was inflated at a rate of 60 mL min−1. The HVs were randomly allocated to the treatment in a cross-over fashion. Subjects were asked to score satiation and abdominal sensation using a graphic rating scale that combined verbal descriptors on a scale graded 0–5 (1, threshold; 5, maximum satiation or discomfort). The balloon was deflated when a score of 5 was reached.
Manometry: IGP measurement during nutrient drink ingestion
Seven independent experiments were performed during which IGP was assessed: (i) drinking nutrient drink at different speeds (n = 13, seven men, age: 31 ± 2 years, body mass index: 22.4 ± 0.5 kg m−2), (ii) intragastric infusion of nutrient drink at different speeds (n = 17, eight men, age: 30 ± 2 years, body mass index: 22.6 ± 0.6 kg m−2), (iii) comparison between nutrient drink and water (n = 9, five men, age: 28 ± 3 years, body mass index: 22.7 ± 0.8 kg m−2), (iv) comparison between nutrient drink and saline (n = 11, seven men, age: 27 ± 2 years, body mass index: 22.8 ± 0.8 kg m−2), (v) effect of L-NMMA (n = 7, four men, age: 33 ± 4 years, body mass index: 22.3 ± 0.6 kg m−2), (vi) effect of catheter position (n = 5, three men, age: 27 ± 4 years, body mass index: 23.2 ± 1.2 kg m−2; results not shown), (vii) effect of volunteer position (n = 10, five men, age: 30 ± 3 years, body mass index: 22.7 ± 0.8 kg m−2; results not shown).
General protocol Two types of manometry probes were used: to study the effect of ingestion speed (while drinking the nutrient drink or during intragastric infusion) a custom-made 6-channel solid-state manometer was used with 4 cm in between each pressure channel [Probe Model P31-6019DD18; Konigsberg Instruments Inc., (Pasadena, CA, USA) recording software: Polygram for Windows 2.05 (Synectics, Stockhom, Sweden)]. For all other experiments a high-resolution solid-state manometer system was used (36 channels, 1 cm in between each channel, Manoscan 360; Sierra Scientific Instruments, Los Angeles, CA, USA; Manoview analysis software v2.0.1). Upon arrival in the clinic the manometer was positioned through the nose so that at least one sensor was positioned in the lower esophageal sphincter (LES; detected as a clearly elevated pressure zone compared to oral and aboral areas), while IGP was measured using the sensor 4 cm under the LES (Koningsberg catheter) or as the average pressure of the first five pressure channels that were clearly positioned below the LES or the pressure area influenced by the LES [approximately 3–8 cm under the LES high-resolution catheter (see data analysis; Fig. 1)].
In those experiments in which the volunteers had to drink the nutrient drink, only the manometry probe was placed. For the experiments in which the drink was intragastrically infused a second infusion catheter (Flocare; Nutricia, Bornem, Belgium) was positioned in the stomach through the mouth. The tip of the infusion catheter was positioned approximately 5 cm under the LES and its position was verified by fluoroscopy. From preliminary experiments we found that the position of the infusion catheter (with the tip in the proximal or distal stomach) did not affect the IGP during intragastric nutrient drink infusion (60 mL min−1; results not shown).
The catheter(s) was/were fixed to the subjects chin and the HVs were asked to take place in a chair. In preliminary experiments we found that the position of the volunteers (supine or sitting down) did not influence IGP and satiation indicating that gravitational forces do not influence the results. After a stabilization period of at least 15 min, nutrient drink (Nutridrink; Nutricia, Zoetmeer, The Netherlands; 630 kJ, 6 g proteins, 18.4 g carbohydrates, and 5.8 g lipids per 100 mL) was infused directly in the stomach at a constant speed (15, 30 or 60 mL min−1) determined by an automated system using a peristaltic pump. When IGP was studied during drinking, a peristaltic pump was used to automatically fill cups (15, 30 or 60 mL min−1), while the HVs were asked to empty the cups so they could keep up with the filling rate.
Healthy volunteers were asked to score their satiation (explained to them as the opposite of hunger), at 1-min intervals, using a graphic rating scale that combines verbal descriptors on a scale graded of 0–5 (1, threshold; 5, maximum satiation). Intragastric infusion or drinking was stopped as soon as a score of 5 was reached on the satiation scale, hereafter the experiment was terminated.
Nutrient drink vs water and saline The effect of nutrient drink on IGP was compared with tap water and a viscous saline solution. The viscous saline solution was custom-made so that the same osmolarity and viscosity as the nutrient drink was obtained (455 mOsm L−1 and 20 mPas s−1) and contained 13.7 g L−1 NaCl (sodium chloride ≥99% AR grade; Sigma-Aldrich, Bornem, Belgium) and 10 g L−1 (hydroxypropyl)methyl cellulose (viscosity of 2% aqueous solution: 40–60 centipoises; Sigma-Aldrich), dissolved in tap water. The comparison between water and nutrient drink was performed using an infusion speed of 15 mL h−1.
Effect of NG-monomethyl-l-arginine After introduction of the catheters and a 15-min stabilization period, an intravenous (i.v.) infusion of either saline (100 mL h−1) or L-NMMA (8 mg kg−1 h−1) was started and continued for the remainder of the experiment. Allocation of the treatment was organized in a blind randomized cross-over fashion. We have used the same dose of L-NMMA in previous research in which we established a role for NO in gastric accommodation in man.3 Intragastric nutrient drink infusion started 20 min after the i.v. infusion.
Balloon distension Per 50 mL volume increase intraballoon pressure was represented as an increase from baseline pressure (median pressure in the deflated balloon in the 2 min before inflation started), represented as mean ± SEM, and compared with a two-way repeated measures anova followed by a Tukey post-test for multiple comparison (P < 0.05 was considered significant).
IGP The original data was imported from the recording software to excel. We were primarily interested in slow IGP changes that could reflect changes in gastric muscle tone. Therefore, and in order to avoid influence from movement artifacts and artifacts caused by coughing, sneezing, moving, or swallowing a moving median was calculated per channel from the original data (median value over 1 min of original data). Per channel, a baseline value was calculated from the moving median data as the average pressure in the last 5 min of the stabilization period. Data was presented per minute as the difference of the minimum moving median value in that minute and the baseline value. When the high-resolution manometer was used, data was represented as the average value of the five measurement channels that were clearly positioned below the LES as described above. Data was presented as mean ± SEM and compared with a two-way repeated measures anova and a Tukey post-test for multiple comparisons when applicable. From preliminary observations, a remarkable correlation between IGP (apart from the initial decrease) and satiation scores appeared. In order to investigate a possible correlation between IGP (from nadir IGP) and satiation scores we performed a post hoc correlation analysis (P < 0.05 is considered significant).
Barostat: IGP during stomach distension, a measure for gastric accommodation
Balloon pressure increased progressively during inflation (Fig. 2). After the subjects drank water, the pressure in the bag increased significantly more as compared to nutrient drink (P < 0.05). Balloon inflation was stopped when subjects scored discomfort at 671 ± 34 mL (after water ingestion) and 714 ± 53 mL (after nutrient drink ingestion; ns).
Manometry: IGP during nutrient drink ingestion: intragastric infusion vs drinking and effect of ingestion speed
Healthy volunteers scored maximal satiation after 699 ± 62, 809 ± 90 and 997 ± 120 mL nutrient drink infused (15, 30 and 60 mL min−1 respectively; P < 0.01 between 15 and 60 mL min−1). When the subjects drank the nutrient drink, maximum satiation was reached after 810 ± 85, 823 ± 87 and 1018 ± 119 mL (15, 30 and 60 mL min−1 respectively; P < 0.05 between 15 and 60 mL min−1). During infusion or while drinking the IGP initially decreased rapidly, followed by a gradual increase thereafter (Fig. 3). Maximum IGP decrease during nutrient drink infusion was 3.4 ± 0.5 mmHg after 205 ± 28 mL for 15 mL min−1, 5.1 ± 0.7 mmHg after 212 ± 46 mL for 30 mL min−1 and 5.2 ± 0.7 mmHg after 296 ± 28 mL for 60 mL min−1 (ns). Maximum IGP decrease during drinking was 5.2 ± 0.8 mmHg after 272 ± 54 mL for 15 mL min−1, 6.5 ± 1.4 mmHg after 201 ± 31 mL for 30 mL min−1 and 4.8 ± 1.0 mmHg after 231 ± 56 mL for 60 mL min−1 (ns).
Post hoc analysis: correlation between IGP recovery and satiation scores A post hoc analysis was performed on the relation between IGP and satiation during intragastric infusion of nutrient drink at 15, 30 and 60 mL min−1. A significant correlation was found between the IGP increase from nadir and the corresponding satiation scores in each group (P < 0.0001; Fig. 3B). During nutrient drink infusion the correlation coefficients were not significantly different between infusion speeds, while during drinking, the correlation coefficient at 15 mL min−1 was significantly smaller that when volunteers drank 60 mL min−1 (P < 0.05). Both for the infusion and drinking data, the slopes of the linear regressions and the intercepts were not significantly different from each.
Manometry: validation experiments
Nutrient drink vs water and saline Healthy volunteers ingested significantly more water than nutrient drink before maximal satiation was scored (1198 ± 139 vs 956 ± 92 respectively; P < 0.05). Maximum IGP decrease was 1.6 ± 0.4 mmHg after infusion of 220 ± 55 mL water and 4.0 ± 1.0 mmHg after infusion of 273 ± 69 mL nutrient drink (ns; Fig. 4A).
No significant changes were observed when comparing IGP during nutrient and viscous saline infusion. Maximum satiation was scored after 941 ± 97 and 1088 ± 112 mL nutrient and viscous saline infused respectively (ns). Maximum IGP decrease was 5.1 ± 0.5 mmHg after 344 ± 35 mL nutrient drink and 5.8 ± 0.8 mmHg after 535 ± 135 mL viscous saline respectively (ns; Fig. 4B).
Effect of NO synthase inhibition During i.v. infusion of L-NMMA, the baseline IGP significantly increased as compared to saline infusion (Fig. 5A). Moreover, during intragastric nutrient drink infusion the IGP was significantly increased during i.v. L-NMMA infusion. When compared to the average IGP in the 5 min preceding the intragastric infusion, the maximal IGP decrease during intragastric nutrient drink infusion was 2.7 ± 0.9 mmHg during L-NMMA infusion and 3.9 ± 0.8 mmHg during saline infusion (ns).
Volunteers drank less during L-NMMA infusion (816 ± 91 mL vs 1032 ± 71 mL; P < 0.005). Interestingly, when correlating the IGP increase from nadir with the corresponding satiation scores there was no significant difference between the correlation during saline or L-NMMA treatment: the correlation coefficients were 0.79 ± 0.09 and 0.65 ± 0.09 for saline and L-NMMA respectively, the linear regressions had a slope of 1.07 ± 0.27 and 1.33 ± 0.29 units satiation score increase per mmHg increase from nadir for saline and L-NMMA treatment, while the intercepts were 0.10 ± 0.12 and 0.31 ± 0.21 for saline and L-NMMA respectively (ns from 0).
During food intake the muscles of the proximal stomach are known to relax. This gastric accommodation reflex is known to provide a reservoir for the ingested food, while avoiding IGP increase.4,15–17 The barostat is regarded as the golden standard to assess gastric accommodation: an intragastric bag is kept at constant pressure, while changes in bag volume are recorded to assess gastric muscle tone changes.18 However, the technique makes use of an inflated gastric balloon, which makes it particularly unphysiological to assess gastric accommodation during food intake.7,8 Moreover, its invasive nature excludes its application in routine clinical practice.9 We hereby present a series of experiments that were performed to setup and validate IGP measurement during intragastric nutrient drink infusion as an alternative technique to assess gastric accommodation.
In a first experiment we showed, using the well-validated barostat technique, that IGP increase during stomach distension was significantly lower after ingestion of a nutrient drink as compared to ingestion of water. As we previously showed, using a barostat, that ingestion of 200 mL nutrient drink evoked significant gastric relaxation,19 this experiment indicates that gastric accommodation lowers IGP of a distended stomach and that IGP during stomach distension can be used to assess gastric accommodation.
In order to setup a method to assess gastric accommodation by measuring the IGP during food intake we compared, in a second series of experiments, IGP and satiation while drinking nutrient drink or during intragastric infusion of nutrient drink at different speeds. Independent on the speed or on the ingestion method IGP decreased initially to recover thereafter. When comparing the IGP during drinking or intragastric infusion, no significant differences could be observed. However, as we were interested in the relationship between gastric accommodation and satiation we decided to continue with intragastric infusion experiments because results obtained while drinking might be more sensitive to factors that are not directly related to gastric accommodation such as taste (aversion), sensation of the nutrient drink while swallowing, and other subjective factors that might determine satiation or the volume consumed.
When comparing different ingestion speeds, it was observed that the ingested volume at maximal satiation differed. Most authors agree that (semi)-liquid meals are emptied from the stomach in a linear fashion with an emptying rate depending on the caloric content (rates vary from 0.4 to 2.5 kcal min−1).20–24 In our study HVs consumed the nutrient drink at a rate of 22.5, 45 and 90 kcal min−1. In a preliminary study using scintigraphy we examined gastric emptying of the nutrient drink used in this study and found that after initial emptying of a small fraction, the meal further accumulates in the stomach throughout the test and that more than 80% remained intragastrically at the moment that maximum satiation is reached.25 We can therefore assume that gastric emptying plays a minor role in the determination of our results. However intestinal exposure of nutrients is a major determinant of satiation,26 and probably plays a role in the determination of satiation during our experiments, especially during low ingestion speed. Indeed, not only is gastric emptying relatively more important during low ingestion speed, the time of ingestion (and thus intestinal exposure of nutrients during ingestion) was considerably longer for low ingestion speed. This might explain why HVs consumed significantly more when ingestion speed was high and why the correlation coefficients for 15 mL min−1 were significantly smaller as compared to 30 and 60 mL min−1, indicating that correlations became better with increasing ingestion speed. Indeed, from preliminary observations a remarkable correlation between IGP (apart from the initial decrease) and satiation scores appeared. Post hoc analysis of the correlation between IGP and satiation revealed a highly significant correlation between IGP and the corresponding satiation scores, indicating that IGP might be an important determinant of satiation. In light of the correlation between IGP and satiation at 60 mL min−1 and the fact that HV drank significantly more we chose to proceed using an ingestion speed of 60 mL min−1.
In a preliminary set of experiments we compared whether the position of the infusion catheter (in the proximal or distal stomach) would affect our results (n = 5). In some volunteers, with the infusion catheter in the distal stomach, the experiment provoked nausea that lasted for several hours after the end of the experiment. For this reason this experiment was ended before the planned number of volunteers was included. No differences were observed in satiation or the shape of the IGP curve indicating that infusion of the nutrient drink in the distal or proximal stomach did not influence the IGP during ingestion. As this experiment indicated that dumping-like symptoms could be provoked when the infusion catheter was positioned in the distal stomach, we chose to position the catheter in the proximal stomach.
In a final series of experiments we compared non-nutritious versus nutritious meals and the effect of the NO synthase inhibitor L-NMMA. The IGP course during intragastric saline infusion was comparable to the effect during nutrient drink infusion. However, during water infusion no IGP change could be observed. Water very rapidly leaves the stomach (80% of a given volume empties in about 10 min27), hence we expect no or limited distension of the stomach during water infusion, especially because the infusion speed was only 15 mL min−1. Increasing the viscosity and the osmolarity of a meal is known to delay gastric emptying;28,29 infusion of viscous saline therefore most likely distended the stomach also because the infusion speed was 60 mL min−1. The IGP decrease observed during nutrient drink and saline infusion is likely dependent on the gastric muscle tone: with increased muscle tone (during L-NMMA treatment) the IGP decrease was indeed less pronounced, although this difference was not significant. An alternative explanation might be that activation of osmoreceptors in the duodenum induced gastric relaxation.30
After the initial IGP decrease, IGP continuously increased upon further nutrient drink ingestion. This IGP increase is probably a result of the increasing intragastric volume on one hand and the tone of the muscles of the stomach on the other hand. Indeed, after L-NMMA treatment IGP increase was more pronounced for a similar ingested volume. Interestingly, satiation ratings followed IGP increases in these experiments. Gastric distension has been shown to trigger stretch and tension mechanosensitive receptors that in turn relay their information via vagal and splanchnic nerves to the central nervous system.31,32 We hypothesize that increased IGP increases afferent signaling from the stomach, and could trigger increased feelings of satiation.
Several lines of evidence suggest that NO is a key mediator of gastric accommodation during food intake.33,34 We have previously shown in humans that inhibition of NO synthase with 8 mg kg−1 h−1 L-NMMA impairs accommodation to a meal as assessed with a barostat.3 In this study, L-NMMA treatment increased IGP before and during nutrient drink infusion indicating that NO release mediates gastric tone in basal conditions and is indeed released during nutrient drink infusion. During L-NMMA treatment the HVs ingested significantly less nutrient drink. Interestingly, the increase in satiation per mmHg IGP increase was not different after L-NMMA treatment, indicating that IGP might be a determinant of satiation. Our findings are in contrast with a study by Vozzo et al. who found that treatment with 4 mg kg−1 h−1 L-NMMA did not affect caloric intake or sensations of fullness or hunger.35 Despites the difference in the L-NMMA dose used we believe the dissimilar findings might be explained by the differences in food composition (solid meal in the study by Vozzo et al.) and ingestion speed [in the study of Vozzo et al. volunteers had 30 min to eat as much as they could while in our study intragastric infusion lasted on average 14 and 17 min (during L-NMMA and placebo treatment)].
Future studies will have to indicate whether this method can be used to subjectively discriminate patients with impaired gastric accommodation (e.g. functional dyspeptic patients). The healthy volunteers in this study were recruited mainly from a student population and results obtained in this study have to be used with caution if used as a reference for future studies in patients with functional dyspepsia or other digestive disorders that are more common in middle-age females.36
In conclusion, IGP measurement during nutrient drink ingestion is a promising method to assess gastric accommodation and satiation during food intake. Upon nutrient drink ingestion, IGP decreases initially to gradually recover thereafter. The IGP increase is correlated with satiation score increase and indicates that IGP is a determinant of satiation. We propose that IGP monitoring provides a minimally invasive, and more physiological alternative to the barostat for the assessment of gastric accommodation.
Acknowledgments and disclosures
Pieter Janssen and Lukas Van Oudenhove are postdoctoral research fellows of the FWO Flanders. This work was supported by an FWO grant and a Methusalem grant to Jan Tack, MD, PhD.
PJ contributed to conception and design of the experiments, collection, analysis and interpretation of data and drafting the article; SV, HGL and LVO contributed to collection, analysis and interpretation of data; JT contributed to conception and design of the experiments, and revising the manuscript critically for important intellectual content.