SEARCH

SEARCH BY CITATION

Keywords:

  • enteral nutrition;
  • manometry;
  • osmolarity;
  • postprandial motor response;
  • small bowel

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosures
  9. Author Contribution
  10. References

Background  Although there is profound knowledge about cyclic fasting motility, the postprandial intestinal motor response is not well investigated. It is intriguing to speculate that nutrient composition alters small bowel motility significantly and, in a clinical setting, may account for adverse gastrointestinal symptoms in enteral nutrition (EN). We aimed to assess the impact of different caloric loads and osmolarities of EN on human jejunal motility.

Methods  Sixteen healthy subjects underwent a series of duodenal infusions of EN solutions, either with iso-osmolar solution with different caloric loads (1.32, 2.64, or 3.96 kcal min−1), or with solutions of different osmolarities with constant caloric loads (300, 600, or 1200 mosmol). Jejunal solid-state manometry was analyzed over 90 min both visually and using dedicated computer software.

Key Results  All tested nutrient solutions were able to trigger conversion to a postprandial jejunal motility pattern after a mean lag phase of 9.4 + 2.3 min (P = NS between different nutrient solutions). Different caloric loads did not result in significant differences in small bowel motility. However, increasing osmolarities caused a significant inhibition of contractile and propagative activity.

Conclusions & Inferences  Small bowel motility under duodenal infusion of nutrient solutions is not influenced by caloric load in a physiological range, whereas high osmolarities inhibit small bowel motility.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosures
  9. Author Contribution
  10. References

There is profound knowledge about cyclic fasting motility.1 This so-called migrating motor complex (MMC) comprises three well-defined phases of motor silence (phase I), irregular contractions (phase II), and the ‘intestinal housekeeper’ (phase III), motility with dense, rhythmic contractions. After ingestion of a meal, a continuous, irregular motility pattern is established (Fig. 1).2

image

Figure 1.  Excerpt from a jejunal manometry recording. Marked are phase I (motorical silence), phase II, and phase III (‘intestinal housekeeper’) of cyclic fasting motility (MMC). The timing of the study protocol is displayed: A short lag phase and an intense postprandial motor response were observed after start of duodenal tube feeding.

Download figure to PowerPoint

However, the influence of the nutrient composition on small bowel motility has been systematically studied almost exclusively in animal experiments.3–6 It is tempting to assume that either the osmolarity or the caloric loads have motility effects.

In a clinical setting, enteral nutrition (EN) by tube feeding is frequently limited by side effects. Adverse effects of EN are mostly of gastroenterological nature including emesis, aspiration, and nosocomial pneumonia, and can occur even when EN is applied in a semirecumbent position or postpyloric tube position.7 Small bowel feeding is frequently accompanied by abdominal discomfort, cramps, and diarrhea. It is intriguing to speculate that these adverse effects are due to altered intestinal motility.

Thus, we aimed to assess the impact of nutrient caloric load and osmolarity on human small bowel motility under standardized conditions in a setting resembling enteral tube feeding.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosures
  9. Author Contribution
  10. References

Subjects

Sixteen healthy volunteers (eight males, eight females) with a mean age of 26 (range 24–30) years without a history of gastrointestinal diseases or abdominal surgery participated in the study. A full previous and current medical history was taken using a questionnaire, ruling out any relevant medical conditions. All subjects were non-smokers, did not take any medication, and usually consumed no or only small amounts of alcohol (<20 g day−1). They were instructed to refrain from any alcohol intake for 2 weeks before the study. Written informed consent was obtained from each subject.

Study design

Three test solutions were investigated in each individual in a randomized order. Eight subjects (four males, four females) received a series of test solutions with different osmolarities, and eight other subjects (four males, four females) received a series of iso-osmolar test solutions with different caloric loads. In each subject, each test solution was studied on a separate day. After an overnight fast, subjects underwent transnasal intubation of the small intestine with a tube assembly for intestinal perfusion and manometry. A fine (2.8 mm OD) flexible polyurethane catheter that incorporated six piezoresistive pressure sensors (Unisensor, Winterthur, Switzerland) located 3, 6, 9, 12, 15, and 18 cm from the tip was used to record motility. A radio-opaque feeding tube (Freka, Fresenius AG, Bad Homburg, Germany; 2.8 mm OD) served for intestinal perfusion and was attached to the recording catheter, so that the perfusion point was situated 30.5 cm from the catheter tip. Using fluoroscopic control, the catheter assembly was positioned with the perfusion point located in the descending portion of the duodenum. The pressure sensors were thus spanning a recording segment of 15 cm aboral the duodenojejunal flexure. Recording protocol is shown in Fig. 1. After an accommodation period of 1 h, one MMC cycle was recorded. Ten minutes after a phase III had passed the recording segment, nutrient perfusion was started. Each solution was perfused for 120 min at a rate of 5 mL min−1, which agrees with postprandial gastric fluid emptying.8 The protocol, which was in accordance with the Declaration of Helsinki, has been approved by the local ethics committee.

Composition of test solutions

All test solutions were prepared freshly from a commercially available nutrient solution (Nutricomp F, B. Braun, Melsungen, Germany) on the day of the experiment and kept under 37 °C during infusion. Nutricomp F contains 59% carbohydrates, 24% lipids, and 17% proteins (nutrient composition in %kcal 55 : 30 : 15). The content of dietary fiber is 1.5%. Three solutions of different osmolarities of 300, 600, and 1200 mosmol were prepared by the addition of sodium chloride to Nutricomp F. Three iso-osmolar (300 mosmol) solutions of different caloric loads (1.32, 2.64, and 3.96 kcal min−1 at 5 mL min−1 infusion rate) were achieved by the addition of sodium chloride and water to Nutricomp F.

Digital recording system

The recording catheter, which was calibrated before each measurement, was connected to a data logger (PMT Megalogger, Göttingen, Germany), which served for data storage.9 From each recording sensor, pressure was sampled at a rate of 3 Hz, which has been reported to permit recording of motility of the human upper small intestine without significant reduction in the contraction amplitudes.10

Data analysis

Both visual and computer-aided data analyses were undertaken.9,11 For visual analysis, printouts of the tracings in the compressed (paper speed of 0.8 cm min−1) and the expanded mode (paper speed of 7.5 cm min−1) were analyzed by two observers on separate occasions without prior knowledge of the test solution. By visual inspection, the so-called lag period of intestinal motility, i.e., the time interval (minutes) between the start of the enteral perfusion and the onset of postprandial motility in the recording segment, was evaluated.12 In a similar way, recordings underwent visual screening for a special motor pattern, the so-called migrating clustered contractions (MCCs).13 An MCC was defined as a rhythmic series of phasic contractions occurring at a frequency of 10–12 per min,14 lasting <2 min, preceded and followed by at least 30 s of absent motor activity, and showing aboral migration through the whole recording segment. The frequency of MCCs (per hour), their duration (seconds), and their aboral migration velocity (centimeters per second) were calculated. For computer-aided analysis, as the lag period ranged from 2 to 20 min, it was decided to use a period of 90 min (from 30 to 120 min after the start of the perfusion) for comparison of the solutions (compare Fig. 1). A computer program (SBMA, 1.32; Roland Widmer, Munich, Germany) developed and validated in our laboratory was used to reject artifacts and to identify individual phasic contractions and analyze their spatial and temporal relationships.11 Briefly, phasic pressure events exceeding an amplitude of 9.7 mmHg, a duration of 2.8 s, and an area under the curve of 18.4 s × mmHg without simultaneous events in other recording channels were considered by the algorithm as a real contraction. From the data files of recognized contractions, the mean values for contraction frequency (per minute), contraction amplitude (millimeters of mercury), duration of contraction (seconds), and a normalized motility index defined as total area under the curve divided by time (seconds times millimeters of mercury divided by minutes) were calculated. To analyze the aboral propagation of individual contractions, the time window technique was used.15 Aboral propagation velocities between 0.6 and 4.5 cm s−1 were permitted to occur.16 If a contraction at a distal recording site appeared within this window, the computer identified it as a propagated contraction and then looked for propagation at the next site. For contractions that propagated in less than 3 cm, the distance between two adjacent pressure sensors were defined as non-propagated or stationary. The number of propagated contractions per minute and their mean aboral propagation distance (centimeters) were calculated. Finally, the analysis periods were divided into 10 time intervals of identical duration (deciles), to analyze whether any parameter of motility showed significant changes over time.

All subjects were questioned about abdominal pain, cramps, or other abdominal symptoms 60 min after start of the EN and 15 min after end of the EN.

Statistical analysis

All results are expressed as means (of the means) ± SEM. Significance was assessed by Student’s t-test for paired observations. Differences were considered significant if P < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosures
  9. Author Contribution
  10. References

In all subjects, the recordings with three different nutrient solutions could be completed according to the protocol. Enteral feeding was well-tolerated with all different nutrient solutions, except two subjects who complained about abdominal fullness after infusion of the solution with 1200 mosmol L−1.

Visual analysis

After the start of the nutrient perfusion, some time elapsed until the postprandial motor response started in the recording segment. The mean duration of the lag phase was 9.4 min (+2.3 min) and did not differ significantly between recordings with 300, 600, and 1200 mosmol nutrient solutions (9.3 + 2.2, 9.2 + 2.4, and 10.1 + 2.1 min) or nutrient solutions with a caloric load of 1.32, 2.64, or 3.96 kcal min−1 (9.1 + 2.6, 8.8 + 2.0, and 9.9 + 2.2 min). In none of the subjects, a phase III activity occurred after the start of the perfusion. Examples of duodenojejunal motility with the three solutions of different osmolarities are shown in Fig. 2. With increasing osmolarity, contractile activity appeared to be less intense. Analysis of clustered activity, which was present in all experiments, is listed in Tables 1 and 2. Rising osmolarity led to a significant decrease (P < 0.02) in the rate of MCCs, when 1200 mosmol L−1 was compared with 300 mosmol L−1. Rising caloric load did not exhibit significant differences between the three solutions.

image

Figure 2.  Examples of duodenojejunal motility with the three solutions of different osmolarities in the same individual. (A) 300 mosmol, (B) 600 mosmol, (C) 1200 mosmol. With increasing osmolarity, motility appeared to be less intense. Displayed are three channels at distances of 6 cms.

Download figure to PowerPoint

Table 1.   Motility parameters after exposure to nutrient solutions of different caloric loads
Caloric load (kcal min−1)1.322.643.96
  1. Means ± SEM; P = ns for all comparisons.

Overall contractile activity
 Contraction frequency (min−1)3.0 ± 0.22.9 ± 0.32.9 ± 0.3
 Contraction amplitude (mmHg)20.9 ± 0.321.3 ± 1.020.6 ± 0.6
 Motility index  (s × mmHg × min−1)117 ± 10116 ± 11110 ± 11
Propagative activity
 Propagated contractions (min−1)1.1 ± 0.11.1 ± 0.11.0 ± 0.2
 Aboral contraction spread (mm)51 ± 252 ± 251 ± 2
 Migrating clustered  contractions (per hour)14 ± 312 ± 211 ± 3
Table 2.   Motility parameters after exposure to nutrient solutions of different osmolarities
Osmolarity (mosmol kg−1)3006001200
  1. Means ± SEM; *P < 0.02 vs 300 mosmol kg−1; P = ns for other comparisons.

Overall contractile activity
 Contraction frequency (min−1)4.0 ± 0.63.7 ± 0.42.7 ± 0.3*
 Contraction amplitude (mmHg)25.2 ± 1.324.0 ± 1.422.3 ± 0.8*
 Motility index  (s × mmHg × min−1)210 ± 27181 ± 23118 ± 16*
Propagative activity
 Propagated contractions (min−1)1.2 ± 0.21.0 ± 0.10.7 ± 0.1*
 Aboral contraction spread (mm)57 ± 157 ± 151 ± 2*
 Migrating clustered  contractions (per hour)19 ± 89 ± 37 ± 2*

Automated analysis

The results of computer-aided data evaluation are listed in Tables 1 and 2 (see also Fig. 3). Contractile activity with the solutions of different caloric loads showed contraction parameters and a spatial and temporal distribution of contractions that were not statistically different when compared with each other. The number of MCCs was similar.

image

Figure 3.  Motility indices of all investigated individuals after duodenal tube feeding with solutions of (A) different caloric loads, (B) different osmolarities.

Download figure to PowerPoint

When EN was applied with solutions of different osmolarities, with increasing osmolarity contraction frequency, amplitude, motility index, number of propagated contractions, aboral contraction spread, and number of MCCs were lower. When recordings with osmolarities of 300 and 1200 mosmol L−1 were compared, all parameters at 1200 mosmol L−1 were significantly (< 0.02) lower. The analysis of the time course of motility revealed that none of the parameters listed in Tables 1 and 2, including migrating clustered activity, exhibited any significant changes over time.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosures
  9. Author Contribution
  10. References

In the present study the effects of enteral nutrient solutions of different osmolarities and caloric loads on jejunal motility were investigated in healthy subjects. Our main findings were as follows:

  •  All tested nutrient solutions were able to trigger a conversion into a postprandial jejunal motility pattern.
  •  Different caloric loads did not result in significant differences in small bowel motility.
  •  Increasing osmolarities caused a significant inhibition of contractile and propagative activity.

Technical aspects

A frequent criticism to manometric and tube feeding studies is a potential influence of fluid volume independently of its makeup, as, e.g., caused by water perfused manometry catheters. However, Behrns et al. showed in canine experiments that differing infusion rates within physiologic ranges from 0 to 15 mL min−1 of non-nutrient, iso-osmolaric solutions limited to duodenum, or jejunoileum did not affect markedly local or distant motor patterns.17 We excluded such potential confounders systematically. Manometry was performed with solid-state catheters; flow rates were kept identical between all tested nutrient preparations.

Often, when EN has been studied, the timing of the initiation of tube feeding within the cyclic MMC has not been taken into account. However, evidently motor response to intraluminal nutrient application is dependent on the phase of MMC. ‘Typical’ conversion to a feeding pattern regularly occurred when EN was started during phase I, not during phase II.14 Thus, we assured a standardized initiation point in phase I, 10 min after phase III had passed the studied segment (see Fig. 1), to overcome such limitations.

Different caloric loads

Studies in dogs showed that enteral nutrient infusion of more than 0.5 kcal min−1 converts cyclical fasting activity into a so-called postprandial pattern, until at rates of approximately 4 kcal min−1 conversion will always be triggered.3,18 Our results demonstrate that in humans, energy loads of 1.32–3.96 kcal min−1, lying within physiologic range of gastric emptying, always initiate such a motor response.

As reported in previous studies, there is a lag period of approximately 10 min between initiation of EN and start of the intestinal motor response, which is now shown to be independent of different caloric loads.6,19

One major finding in our study is that motility parameters did not differ significantly when nutrient solutions with different caloric loads were used within these limits.

In animal and human studies, unabsorbed fat, carbohydrates, or protein in the ileum has been demonstrated to activate the so-called ‘ileal brake’, resulting in reduced jejunal pressure wave activity, a delay of intestinal transport, and thus an increase in intestinal transit time, probably influenced by an elevated release of GLP-1, neurotensin, and peptide tyrosine–tyrosine.20–23 However, in our study, comparison of early and late intervals within the recordings showed no significance, not even at a caloric load of 4 kcal min−1. Possibly, the highest caloric load in this study did not exceed intestinal absorption capacities and thus did not stimulate the ‘ileal brake’.

Similar to our data, in a pilot study by Riachi et al., findings in duodenojejunal manometry with duodenal nutrient infusion at 1 and 2 kcal min−1 have not shown significant quantitative differences in motility parameters.6

Our data allow the conclusion that different caloric loads administered in the descending duodenum do not modulate jejunal motility significantly when physiological caloric amounts and physiologic proportions of nutrient components are maintained stable. Consequently, one may wonder why in a clinical setting, diarrhea is a frequently observed adverse effect of duodenal tube feeding. In this context, it is noteworthy that throughout the 1.5 h of manometry recordings under enteral tube feeding in all investigated individuals and settings, the postprandial motor response remained uninterrupted. This finding is in good accordance with data by Ledeboer et al.4 When comparing tube feeding at 2 kcal min intragastrally vs intraduodenally, they found that conversion to a fasting pattern did not occur during the whole duodenal feeding time of 6 h, whereas during continued gastric infusion phase III reoccurred in over 50% of subjects.4 Small bowel transit time was significantly shortened in the duodenal feeding group. Thus, the prolonged presence of the postprandial motility pattern under continuous duodenal nutrient infusion may explain the prokinetic effect of enteral tube feeding. This is confirmed in dogs by data from Defilippi et al., who found sustained postprandial motor response over 250 min with complete suppression of MMC at caloric loads of 2 and 4 kcal min−1, not 0.5 kcal min−1.3

Moreover, not simply caloric load but composition of the nutrient solution could influence intestinal motor response. Harder et al. studied evacuation of intestinal gas in response to duodenal infusion of different nutrient contents in humans. Postprandial clearance of jejunal even more than ileal gas was hampered significantly by isocaloric duodenal infusion (1 kcal min−1) of isolated lipids or amino acids but not by glucose.24 However, the underlying contractile activity was not subject to this study.

Different osmolarities

We found an osmolarity-dependent inhibitory effect of increasingly hyperosmolar nutrient solutions on jejunal motility parameters. Both frequency of pressure waves and amplitude/motility index were suppressed with increasing osmolarity. We recently showed in combined jejunal manometry/impedance recordings that both single contractions and, even more efficiently, MCCs cause aboral intraluminal bolus transfer.25 Consequently, that the occurrence of MCCs was suppressed with higher osmolarities underlines the inhibitory effects of rising osmolarities.

All these inhibitory effects started immediately with the conversion to a feeding motility pattern and did not change over the investigated infusion time of 90 min. Thus, the perception of osmolarity appears to be a function of upper small intestine and not bound to further aboral segments of the gut.

Concerning the influences of osmolarity on intestinal motility, to our knowledge, data so far were scarce and restricted to animal experimental studies. Lin et al. published a series of studies in dogs, showing (i) that hyperosmolar mannitol infusions into the stomach ranging from 300 to 1200 mosmol slowed gastric emptying and increased duodenal resistance by inducing duodenal spike bursts and (ii) in segmental intestinal infusions, the effect of mannitol was most probably due to osmoreception in the first 10 cm of the duodenum.5 In dogs, Schmidt and Ehrlein et al. found increasing osmolarities up to 1520 mosmol in saline or glucose solutions tolerable and, similarly, in direct proportion with increasing inhibition of jejunal propulsive motility and reduced single contraction parameters.12 Both studies concluded that a stronger effect of hyperosmolar glucose compared with mannitol or saline was probably to be interpreted as nutrient specific and not mediated by osmoreception.

We conclude that small bowel motility under duodenal infusion of nutrient solutions is not influenced by caloric load (at least in a physiological range), whereas high osmolarities inhibit small bowel motility. To elucidate the implications of our observations in clinical practice of EN, further studies under clinical circumstances, especially under prolonged infusion and recording periods, are mandatory.

Author Contribution

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosures
  9. Author Contribution
  10. References

HS performed experiments, analyzed the data, and wrote the manuscript; FG performed experiments and contributed to data analysis; TS contributed to the analysis and discussion and helped to design the study; AP designed the study, analyzed the data, and wrote parts of the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosures
  9. Author Contribution
  10. References
  • 1
    Quigley EM. Gastric and small intestinal motility in health and disease. Gastroenterol Clin North Am 1996; 25: 11345.
  • 2
    Abell TL, Malagelada JR. Electrogastrography. Current assessment and future perspectives. Dig Dis Sci 1988; 33: 98292.
  • 3
    Defilippi C. Canine small bowel motor activity in response to intraduodenal infusion of nutrient mixtures of increasing caloric load in dogs. Dig Dis Sci 2003; 48: 14826.
  • 4
    Ledeboer M, Masclee AA, Coenraad M et al. Antroduodenal motility and small bowel transit during continuous intraduodenal or intragastric administration of enteral nutrition. Eur J Clin Invest 1999; 29: 61523.
  • 5
    Lin HC, Elashoff JD, Gu YG et al. Nutrient feedback inhibition of gastric emptying plays a larger role than osmotically dependent duodenal resistance. Am J Physiol 1993; 265: G6726.
  • 6
    Riachi G, Ducrotte P, Guedon C et al. Duodenojejunal motility after oral and enteral nutrition in humans: a comparative study. JPEN J Parenter Enteral Nutr 1996; 20: 1505.
  • 7
    Heyland DK, Dhaliwal R, Drover JW et al. Canadian clinical practice guidelines for nutrition support in mechanically ventilated, critically ill adult patients. JPEN J Parenter Enteral Nutr 2003; 27: 35573.
  • 8
    Bernier JJ, Adrian I, Vidon N. L’eau et les electrolytes dan l’intestin grêle. In: Bernier JJ, Adrian I, Vidon N, eds. Les aliments ans le tube digestif. Paris: Dorin Editeurs, 1988: 13564.
  • 9
    Schmidt T, Widmer R, Pfeiffer A et al. Effect of the quaternary ammonium compound trospium chloride on 24 hour jejunal motility in healthy subjects. Gut 1994; 35: 2733.
  • 10
    Lindberg G, Iwarzon M, Stal P et al. Digital ambulatory monitoring of small-bowel motility. Scand J Gastroenterol 1990; 25: 21624.
  • 11
    Widmer R, Schmidt T, Pfeiffer A et al. Computerized analysis of ambulatory long-term small-bowel manometry. Scand J Gastroenterol 1994; 29: 107682.
  • 12
    Schmid HR, Ehrlein HJ. Effects of enteral infusion of hypertonic saline and nutrients on canine jejunal motor patterns. Dig Dis Sci 1993; 38: 106272.
  • 13
    Husebye E. The patterns of small bowel motility: physiology and implications in organic disease and functional disorders. Neurogastroenterol Motil 1999; 11: 14161.
  • 14
    Medhus AW, Sandstad O, Bredesen J et al. Stimulation of the small intestine by nutrients in relation to phase of the migrating motor complex. Scand J Gastroenterol 2000; 35: 494500.
  • 15
    Sarna SK, Soergel KH, Harig JM et al. Spatial and temporal patterns of human jejunal contractions. Am J Physiol 1989; 257: G42332.
  • 16
    Ledeboer M, Masclee AA, Biemond I et al. Effect of intragastric or intraduodenal administration of a polymeric diet on gallbladder motility, small-bowel transit time, and hormone release. Am J Gastroenterol 1998; 93: 208996.
  • 17
    Behrns KE, Sarr MG, Hanson RB et al. Effect of enteric nonnutrient infusions on motor patterns in neurally intact and neurally isolated canine jejunum. J Surg Res 1995; 59: 40514.
  • 18
    Ehrlein HJ, Schmid HR, Feinle C. Characteristic motor patterns of phase II and behaviour of phase III in the fed state. J Gastrointest Motil 1992; 4: 31727.
  • 19
    Schmidt T, Eberle R, Pfeiffer A et al. Effect of ethanol on postprandial duodenojejunal motility in humans. Dig Dis Sci 1997; 42: 162833.
  • 20
    Dumoulin V, Moro F, Barcelo A et al. Peptide YY, glucagon-like peptide-1, and neurotensin responses to luminal factors in the isolated vascularly perfused rat ileum. Endocrinology 1998; 139: 37806.
  • 21
    Hammer J, Hammer K, Kletter K. Lipids infused into the jejunum accelerate small intestinal transit but delay ileocolonic transit of solids and liquids. Gut 1998; 43: 1116.
  • 22
    Siegle ML, Schmid HR, Ehrlein HJ. Effects of ileal infusions of nutrients on motor patterns of canine small intestine. Am J Physiol 1990; 259: G7885.
  • 23
    Spiller RC, Trotman IF, Adrian TE et al. Further characterisation of the ‘ileal brake’ reflex in man–effect of ileal infusion of partial digests of fat, protein, and starch on jejunal motility and release of neurotensin, enteroglucagon, and peptide YY. Gut 1988; 29: 104251.
  • 24
    Harder H, Hernando-Harder AC, Franke A et al. Duodenal infusion of different nutrients and the site of gaseous stimulation influence intestinal gas dynamics. Scand J Gastroenterol 2006; 41: 294301.
  • 25
    Seidl H, Gundling F, Pehl C et al. Multiple intraluminal impedance measurement and manometry in the jejunum of healthy subjects: a new tool for diagnosis of small bowel motility. Digestion 2010; 81: 17280.