Small Bowel Motility Affects Glucose Absorption in a Healthy Man

Studies were performed in seven healthy males (mean age 22 years, range 20–25; BMI 22.3 kg/m², range 20.1–24.6) who had no history of gastrointestinal disease and were not taking any medication. The study was approved by the Ethics Committee of the University Medical Center Utrecht, and informed written consent was obtained from each subject.

A randomized, double-blind, placebo-controlled, crossover study design was used. Subjects were treated with cisapride 10 mg t.i.d. or matching placebo (both manufactured by Janssen-Cilag, Beerse, Belgium) for 3 days before and on the morning of the experiments during which gastrointestinal motility and absorption were studied. There was a time interval of 2 weeks between the two experiments in each subject.

After an overnight fast, the manometric catheter was introduced transnasally and positioned across the pylorus. When an interdigestive phase II was present in the small intestine, a mixed nutrient liquid of 13% protein, 48% carbohydrate, 39% fat, and 1.5 kcal/ml energy content (Nutridrink; Nutricia, Zoetermeer, The Netherlands) was infused intraduodenally for 30 min at a rate of 2 ml/min (3 kcal/min; 90 kcal in total) via the duodenal infusion port of the catheter. At the end of this infusion (at t = 0 min) 7 g of 3-O-methylglucose (OMG) dissolved in 30 ml distilled water was administered intraduodenally over 3 min to assess glucose absorption kinetics. Blood samples were taken at t = 0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 60, 90, 120, and 150 min for measurement of the plasma 3-OMG concentrations.

A 21-channel water-perfused silicone rubber catheter (outside diameter 4 mm, length 160 cm, channel diameter 0.4 mm, perfusion rate 0.2 ml/min) was used (Dentsleeve, Adelaide, Australia). The assembly incorporated 18 manometric sideholes (S1–S18, spaced at 1.5-cm intervals). S1 was located 14.5 cm distally to the midpylorus. A duodenal infusion port was located at 5 cm from the midpylorus.

Continuous measurement of the antroduodenal transmucosal potential difference (TMPD) was carried out via two sideholes spaced at a 7-cm interval, using established criteria (16), which enabled us to monitor and maintain correct catheter position. Pressures from the perfused sideholes were recorded via external transducers (Abbott, Chicago, IL). Digital data were stored in two 12-channel dataloggers (Medical Measurement Systems, Enschede, The Netherlands) with a memory capacity of 4 Mb each, using a sample frequency of 8 Hz for pressure and 1 Hz for TMPD signals.

Interdigestive phases were recognized visually according to established criteria (17). The manometric recordings obtained after completion of the intraduodenal nutrient infusion were analyzed in 10-min blocks. A locally developed computer program was used for calculation of numbers, amplitude, and spatiotemporal characteristics of duodenal pressure waves. Only pressure waves with amplitude of >1.4 kPa were detected. The algorithms used have been described in detail before (18). In short, for each sidehole the number of antegrade propagated pressure waves (APPWs), starting at this recording site and propagating over at least two sites, was calculated. Pressure waves were considered propagated when the propagation velocity was between 0.9 and 16 cm/s (18). APPWs were divided into seven categories according to distance of propagation: over 2 sites (1.5 cm), over 3 sites (3.0 cm), over 4 sites (4.5 cm), over 5 sites (6.0 cm), over 6 sites (7.5 cm), over 7 sites (9.0 cm), or over 8–18 sites (10.5–25.5 cm).

3-OMG is a glucose analog, which is absorbed by the same active intestinal transport mechanism as glucose, but is not metabolized by the liver and is renally cleared. Plasma concentrations of 3-OMG may therefore be used as an index of glucose absorption (19). Plasma glucose was measured according to standard clinical chemical procedures, and plasma 3-OMG was measured according to trimethylsilyl derivative by gas-liquid chromatography, as described by Jansen et al. (20). 3-OMG concentrations were calculated from the areas under the curve (AUCs) of glucose and 3-OMG on the chromatograms in relation to the known plasma glucose concentrations. The absorption fraction (in percent per minute) was calculated from the 3-OMG concentration curve using Wagner-Nelson kinetics (21); maximal absorption was taken as 100% and individual data were calculated accordingly.

Repeated-measures ANOVA was used to test the effect of cisapride on the number and amplitude of pressure waves over the various time periods. The paired two-tailed t test was used to analyze the effect of cisapride on the 3-OMG absorption characteristics. The unpaired t test was used to analyze the effect of cisapride on the duration of the postprandial period, because the reoccurrence of phase III was not necessarily matched in each subject. Data of the two treatments were combined to evaluate relationships between motility (number and amplitude of pressure waves) and 3-OMG absorption (AUC, absorption fraction, and time to the peak concentration, t = max), using the partial correlation coefficient, which controlled for the treatment variable. Relationships were calculated for the time period directly after 3-OMG infusion (t = 0 to t = max), because the increment of the plasma 3-OMG concentration most appropriately reflects the absorptive phase of the curve (which is a net reflection of intestinal absorption and renal clearance). P < 0.05 was considered statistically significant. Data are presented as means ± SEM.

Immediately following intraduodenal 3-OMG administration, the plasma concentration rose sharply, reaching a peak at around 30 min during both treatments (Fig. 1 and Table 1).

Phase III occurred in five of seven subjects during both cisapride and placebo treatment during the recording period of maximally 150 min, with a mean onset after the end of nutrient infusion of 65 min (range 46–84) vs. 79 min (range 58–130) (cisapride versus placebo, NS) (Fig. 2).

The mean amplitude of pressure waves during the postprandial period was higher during cisapride treatment than during placebo (3.8 ± 0.1 vs. 3.1 ± 0.1 kPa, P < 0.05). There were no significant effects of cisapride treatment on the number of pressure waves and APPWs (data not shown) or on the distance of propagation of APPWs (Fig. 3). Most APPWs were propagated over short distances during both treatments; 62 vs. 66% (cisapride versus placebo) traveled over two sites (1.5 cm), 23 vs. 21% over three sites (3 cm).

In the 30-min period following 3-OMG administration, the AUC positively correlated with both the total number of pressure waves (partial correlation coefficient 0.50 for t = 0–10, P = 0.08; 0.53 for t = 0–20 and t = 0–30, both P = 0.06) and APPWs (partial correlation coefficient 0.55 for t = 0–10, P = 0.05; 0.58 for t = 0–20, P < 0.04 [Fig. 4]; 0.49 for t = 0–30, P < 0.05). There was a trend for an inverse relation between t = max, and the number of pressure waves recorded during t = 0–30 (r = −0.38, P = 0.15). Similar correlations were present, although not significant, on data of cisapride or placebo treatment alone. The number of pressure waves or APPWs did not correlate with either the absorption fraction or the peak plasma concentration. The amplitude of pressure waves did not correlate to any of the absorption parameters.

Cisapride did not significantly affect the peak of the plasma 3-OMG concentration, the t = max, the AUC during the period t = 0–30, or the 3-OMG absorption fraction (Table 1).

The main findings of this study were that, in healthy subjects, an increase in number of duodenojejunal pressure waves and APPWs was related to an increase in small intestinal glucose absorption; treatment with cisapride increased the mean amplitude of duodenojejunal pressure waves, but did not affect the number of pressure waves and spatiotemporal organization of APPWs; and cisapride treatment did not affect glucose absorption.

Studies on the interaction between intestinal motility and glucose absorption have yielded conflicting results, one demonstrating increased absorption during more intense contractility (1) and others showing the opposite (2–5). This is probably caused by differences in study designs (e.g. in vivo experiments or use of segmental bowel loops, nature and amounts of infusions, and the way absorption was assessed). In some reports plasma glucose concentrations were used as parameters of absorption (1,4), which could be greatly influenced by hepatic glucose production or metabolization. For this reason, we chose to use the glucose analog 3-OMG, which is not metabolized in the liver (19). In addition, absorption studies in the fed state differ greatly from fasting studies, probably caused by hormonal rather than motility factors (2). Most importantly, small intestinal transit studies cannot be compared with manometric studies as such, because transit appears to be determined by the length of propagated pressure waves, rather than by the number of pressure waves per se (22).

Our study provided further evidence that glucose absorption is positively related to motor activity, but our data do not support the view that glucose absorption is related to small intestinal contractions propagated over long distances. There were no important differences between the correlation coefficients of AUC with pressure waves or APPWs, and the majority (>60%) of propagated sequences traveled over a distance of only 1.5 cm. Thus, short-distance propagated sequences appear to be responsible for the observed relationship between motility and absorption. In other, less-detailed, manometric studies, these sequences would probably have been classified as isolated phasic contractions or would not have been picked up at all. We suggest that the short-distance propagated sequences serve to mix the luminal contents, spreading glucose locally, hence enhancing contact with the mucosal surface and facilitating absorption.

Knowledge of small intestinal glucose absorption in diabetic patients is very limited. Disordered gastric emptying in diabetes has been shown to affect glucose absorption and vice versa, possibly resulting in poor postprandial glycemic control (23). In the present study, the effects of gastric emptying on glucose absorption were overcome by administering both nutrients and 3-OMG intraduodenally. Glucose absorption has been reported to be normal in type 1 diabetes, whereas it may be slightly reduced in type 2 diabetes (6). The magnitude of the contribution of disordered small intestinal motility to postprandial hyperglycemia in type 1 diabetes requires further research.

Several limitations of the present study should also be acknowledged. Firstly, although two experiments were performed in each subject, the total number of subjects was relatively small. Secondly, the carbohydrates in the administered nutrient could have theoretically inhibited intestinal absorption of 3-OMG (19), although presumably to an equal extent in each experiment. A single bolus of 3-OMG was given because this facilitates the kinetic measurements, and a previous study has demonstrated that the chosen amount of 3-OMG produced a dose-response curve with an acceptable analytical accuracy (24).

Our findings regarding the effect of cisapride on duodenojejunal motility appear to be in agreement with most previous reports on the effects of cisapride in healthy subjects (9–11,25). Studies that have measured amplitudes of pressure waves have also shown an increase in amplitude as a result of cisapride treatment (9,11), or an increase in motility index (10), also likely to be a reflection of a higher amplitude of pressure waves. Two postprandial duodenal motility studies found no increase in number of pressure waves (11,13), as in the present study. One study in humans (12) found an increased incidence of jejunal contractions after a single dose of cisapride. It is likely that pharmacokinetic differences were responsible for these discordant findings.

In conclusion, short-traveling pressure waves are related to the rate at which glucose is absorbed in the proximal small intestine, possibly via optimizing mucosal contact of glucose. The prokinetic cisapride enhances the amplitude of duodenojejunal pressure waves, but does not influence the number of pressure waves and intestinal absorption of glucose.

Dr. M. Samsom is a fellow of the Royal Netherlands Academy of Arts and Sciences. The authors thank Dr. J.A.J. Faber of the Center for Biostatistics for his help with the statistical analyses.